diff --git "a/usage_train.jsonl" "b/usage_train.jsonl"
new file mode 100644--- /dev/null
+++ "b/usage_train.jsonl"
@@ -0,0 +1,3908 @@
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:000394:3:1", "occurrence_id": "8c3c16e8e7ff9989ff1e6d62", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  4.7 people, UBOS National Household Survey 2016-2017.|The indicator will track the number of host communities and refugees beneficiaries of access<br>to grid, mini-grid, and off-grid technologies under the project in District hosting refugees.<br>Average household size of 4.7 people, UBOS National Household Survey 2016-2017.|The indicator will track the number of host communities and refugees beneficiaries of access<br>to grid, mini-grid, and off-grid technologies under the project in District hosting refugees.<br>Average household size of 4.7 people, UBOS National Household Survey 2016-2017.|The indicator will track the number of host communities and refugees beneficiaries of access<br>to grid, mini-grid, and off-grid technologies under the project in District hosting refugees.<br>Average household size of 4.7 people, UBOS National Household Survey 2016-2017.|The indicator will track the number of host communities and refugees beneficiaries of access<br>to grid, mini-grid, and off-grid technologies under the project in District hosting refugees.<br>Average household size of 4.7 people, UBOS National Household Survey 2016-2017.|\n|People provided with<br>access to electricity<br>with grid, mini-grid, off-<br>grid under the project,<br>of which refugees<br>(Number)|0.00|Jan/2022|0.00|10-Jul-2024|0.00|22-Jan-2025|267,000.00|Jun/2027|\n|People provided with<br>access to electricity<br>with grid, mini-grid, off-<br>grid under the project,<br>of which refugees<br>(Number)|Comments on<br>achieving targets|\n\nMention: \"UBOS National Household Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Household survey whose 4.7-person household size sizes the project's refugee and host-community beneficiary target.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:006265:19:0", "occurrence_id": "a17d2b5be38c46a09dae6b2d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Upon requisition, the funds are transferred to Kshs account at Central Bank of Kenya. Initial Withdrawal\nshall be based on cash forecast while subsequent replenishment requests shall be through filing of\nwithdrawal applications through the Bank secure client connection. Replenishment and reimbursement\nof withdrawal applications will be accompanied by Statements of Expenditures (SOEs). Program funds\ndeposited in Designated accounts are disbursed to only finance eligible program activities. The assessment\nnotes that there are appropriate designated signatories to the program accounts.\n\n\n9. Funds are transferred to the MoE and TSC county offices local bank accounts at the duly approved\nbank based on approved AIEs/approved activities. Drawing of the funds is strictly for the purpose as\ndetailed in instructions. On spending, the support documents are sent to MoE/TSC headquarters where\nthey reviewed and posted in the IFMIS.\n\n\n_10._ **_Key potential risk and action:_** _a) Financial processes in MoE/TSC county offices are manual and_\n_therefore the potential risk of manipulation b) ineffective monitoring of financial spending at the county_\n_level. There is need to strengthen the financial reporting and monitoring at the county level by automating_\n_operations._\n\n\n**6.3** **Accounting**\n\n\n11. At MoE, JKF and TSC there are well established accounting and financial systems with adequate\nstaffing to ensure adequate program financial records are maintained. Standard Chart of Accounts has\nbeen developed and in use. Also accounting standards and templates have been developed through the\nPublic Sector Accounting Standard Board (PSASB). MoE and TSC financial transactions are captured in the\nIFMIS through budget classification codes. JKF uses an integrated system (Syspro 8) that allows for proper\nrecording of transactions and allocation of expenditures to respective components or cost centers and\nsources of funds. The accounting arrangements including maintaining of financial records and practices\nare captured through finance and accounting manuals. At MoE, the\n\nMention: \"IFMIS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Spending support documents are reviewed and posted into IFMIS, so the system is the record-keeping target rather than a source the passage analyses.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:006265:7:2", "occurrence_id": "6f35ab68cd1d18610e643cc2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Kshs 107= USD|Exchange Kshs 107= USD||||\n\n\n24. **Key potential risks and actions:** a) there are some delays in exchequer releases to MoE\nand from MoE to counties and schools b) Misstated capitation per school due to wrong data\narising from manual intervention rather than fully relying on NEMIS.\n\n\n25. **Proposed actions:** a) The Exchequer in National Treasury should ensure timely release of funds\nand MoE to ensure fast verification and monitoring process for the DLIs achieved. b) Fix/upgrade NEMIS\nto eliminate human intervention. Meanwhile, cleansing and harmonizing of the NEMIS data vs actual\nenrollment in the schools is necessary. (c) MoE to address issues for late release of capitation as well as\ndelays in release of funds for county-based activities.\n\n\n**3.2.2 Accounting and financial reporting.**\n\n\n26. The reforms implemented under PFMR 2013-2018 were successful in establishing a solid\nfoundation for PFM. These are: Government SCOA developed and in use, IFMIS was rolled out to MCDAs,\nPublic Sector Accounting Standards Board formulating accounting standards and templates was\nestablished, and financial reporting improved with timely and consolidated financial statements prepared.\nThe PFMR 2018-2022 has built on the successes of PFMR 2013-18. This has resulted to improvements in\nthe accounting and reporting.\n\n\nPage 8 of 25\n\nMention: \"NEMIS data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"The passage calls for cleansing NEMIS data and reconciling it against actual school enrolment, so the dataset itself is the object of a quality fix.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:006265:9:0", "occurrence_id": "d000cde3589c67fca24fca48", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **3.3.2 Internal audit.**\n\n\n33. On internal audit, the PFM Act 2012 and PFM Act regulation 2015 defines the legal framework for\ninternal audit. The Internal Audit function falls under the Directorate of Accounting Services domiciled at\nnational Treasury. The Internal Audit Department staff are seconded to the MDAs and have dual reporting\nlines, technically to the MDAs Audit Committees/Internal Auditor General and administratively to the\naccounting officers in the MDAs. The Audit Committee of the Ministry was recently appointed. It was\nhowever inactive from end of 2019 due to expiry of their term, but the term was extended this year and\nmade a sitting in June 2022 and continues to operate and to have another meeting in November 2022.\nThe MoE IAD, has 11 auditors who audit all the operations of the ministry. The staff are well qualified and\nare experienced.  Their annual work plan is approved by the audit committee. Reports are issued after\ncompletion of every engagement and management takes action on the internal audit findings which is\nmainly that they will issue a circular to the Counties. In addition to the IAD, MoE has a directorate that\nspecifically deals with schools’ audits.\n\n\n34. **Key risks identified** (i) IAD Limited time to sample many schools per county and sub county (ii)\nlimited capacity in terms of staff given wide audit universe - the whole Ministry operations and at times\nfunding issues for their activities impeding the effectiveness of the internal audit function (iii) Challenges\nauditing through the systems-NEMIS platform. (iv) Period of inactivity of Audit Committee can weaken\nthe oversight (iv) Failure by counties to adhere to laid down regulations (v) Lack of value for money (iv)\n<mark>Lack of authenticity for payment made leading to loss of public funds. (v</mark> i) Enrolment numbers at the\nschool did not tally with NEMIS data which was the basis by which cap\n\nMention: \"NEMIS data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"School enrolment counts were checked against NEMIS data, which had been used as the basis for capitation, exposing a mismatch.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:017096:12:1", "occurrence_id": "8ff1486a98b174a5663087fa", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  19% in\nKiswahili, 15% in Gikuyu and 20% in Dholuo in their last term of Standard 3 could not\nread a single word correctly (Piper, 2010).\n\n**Low learning achievement in Early Grade Mathematics** -Kenya Vision 2030 places\ngreat emphasis on the role of mathematics education. Yet, evidence suggests that\nKenyan children are not achieving minimum levels of numeracy. The Early Grade\nMathematics Assessment (EGMA) study as part of the PRIMR pilot indicates low levels\nof competencies. For example, 30% of students at the beginning of Grade 2 cannot do\nsimple subtraction in the 6 5 counties that the test was administered. In a more\nrepresentative sample, UWEZO survey findings throughout 2009, 2011 and 2012 indicate\nthat less than a third of children enrolled in grade 3 have basic grade 2 level numeracy\nskills.\n\n**Falling learning achievement levels in** **_public_** **schools** -The influx of incoming pupils\nafter the FPE introduces complex changes in Kenya’s primary education system.\nEvidence suggests that the FPE successfully encouraged pupils from poorer socioeconomic groups to enrol in primary schools in general. However, there was no increase\n\n\n**Page | 13ESMF KENYA GPE PRIMARY EDUCATION DEVELOPMENT PROJECT**\n\nMention: \"UWEZO survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"UWEZO learning results across three rounds are analysed to show that most grade 3 children lack grade 2 numeracy.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:017096:43:0", "occurrence_id": "b54652c5c6017cbbf61393d5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: that the quantities of E-wastes has increased since this time therefore making a case for\nand justification to conduct a new inventory.\n\nA mass flow study carried out in 2007 by Kenya ICT Action Network showed that:\n1,513 tons of electronics entered the Kenya market. The consumer in addition to\nreceiving 1,489.4 tons from the formal market,also received 151.3 tons from the second\nhand market. Other sources also indicate that Kenya has reached a mobile penetration\nrate of more than 63% and an internet penetration of more than 18.6%. The number of\ninternet users in (2007) was estimated at 10.2 million.\n\nThe number of ICT companies along the sector increased to more than 4000 companies.\nMoreover, many shops and kiosks also resell ICT related services; especially services and\nproducts relating to the mobile phone. ICT is being extensively used in the education,\nhealth, industrial, trade and communication sectors. Private sector has been installing\nheavy computing equipment and data centers, mainly mobile operators, banks, and\nManufacturing sector companies. From statistics, the consumers are likely to:\n\n\n            - Dispose of 1,210.4 tons in the second-hand market;\n\n            - Dispose of 18.6 tons to collectors or to be refurbished;\n\n            - Dispose of 18.6 tons directly to recyclers.\n\nAccording to National Environment Management Authority, each year the country\ngenerates 3,000 tons of electronic waste. This number raises doubts on the accuracy of\nand prediction or estimation of E-waste due to the fact that if 3,000 tons are generated\nannually, it follows that an almost similar amount of E-waste should be disposed. This\nanomaly points to the critical need to undertake an accurate inventory of E-waste to date\nin Kenya and further design methodologies for estimating or ascertaining E-wastes\ngenerated and disposed on an annual basis.\n\n\n**Page | 44ESMF KENYA GPE PRIMARY EDUCATION DEVELOPMENT PROJECT**\n\nMention: \"mass flow study\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The 2007 mass flow study supplies the e-waste quantities the passage reports and its case for a new inventory.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:021150:15:0", "occurrence_id": "bc50351aae954aec618ff220", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nUGANDA INTERGOVERNMENTAL FISCAL TRANSFERS - ADDITIONAL FINANCING (P172868)\n\n\nthe management information system (MIS). Districts and Municipalities will promote environmental\nmanagement functions initially focusing on the implementation of public investments, including water\nresources management and actions against deforestation. MWE will also develop, together with LGs, MIS\nfor use by districts in the management of water supply and sanitation, which are integrated with national\ndatabases such as the Water Atlas. **At the level of service delivery**, actions will focus on strengthening\nwater supply at the subcounty level. Resources will target sub-counties, where functional water supply\ncoverage is below a specified level and target improving the functionality of existing rural water facilities\ncountrywide.\n\n\n39. **For micro irrigation,** it is proposed to support smallholder farmers in selected LGs to purchase\nand use micro irrigation equipment. This would involve the development of a new system for delivering\nmicro irrigation at the local level, managed by local governments. The latter will include matching grants,\nwhich will be rolled out based on a costed phase-in plan. UgIFT will support the first pilot phase covering\n40 districts with preparations for a planned scale-up country-wide envisaged for the last year of the\nProgram. **At the level central government level,** MAAIF will be responsible for administering grants,\ntechnical oversight and guidance to LGs and for capacity building of staff at central and local level in\nirrigation and irrigated agriculture. **At the district level**, production departments will be responsible for\ncoordinating and monitoring the program. A LG Irrigation Management Information System (MIS) will be\ndeveloped, providing geo-tagged information on infrastructure and services at farmer and district levels.\nThis system will be used to track infrastructure and services being provided and the functioning of farmer\norganizations, with the objective of supporting performance improvement. **At the\n\nMention: \"Water Atlas\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"New district water and sanitation information systems are to be integrated with national databases such as the Water Atlas.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:021150:15:0", "occurrence_id": "774d502cf28efafc57a2da41", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nUGANDA INTERGOVERNMENTAL FISCAL TRANSFERS - ADDITIONAL FINANCING (P172868)\n\n\nthe management information system (MIS). Districts and Municipalities will promote environmental\nmanagement functions initially focusing on the implementation of public investments, including water\nresources management and actions against deforestation. MWE will also develop, together with LGs, MIS\nfor use by districts in the management of water supply and sanitation, which are integrated with national\ndatabases such as the Water Atlas. **At the level of service delivery**, actions will focus on strengthening\nwater supply at the subcounty level. Resources will target sub-counties, where functional water supply\ncoverage is below a specified level and target improving the functionality of existing rural water facilities\ncountrywide.\n\n\n39. **For micro irrigation,** it is proposed to support smallholder farmers in selected LGs to purchase\nand use micro irrigation equipment. This would involve the development of a new system for delivering\nmicro irrigation at the local level, managed by local governments. The latter will include matching grants,\nwhich will be rolled out based on a costed phase-in plan. UgIFT will support the first pilot phase covering\n40 districts with preparations for a planned scale-up country-wide envisaged for the last year of the\nProgram. **At the level central government level,** MAAIF will be responsible for administering grants,\ntechnical oversight and guidance to LGs and for capacity building of staff at central and local level in\nirrigation and irrigated agriculture. **At the district level**, production departments will be responsible for\ncoordinating and monitoring the program. A LG Irrigation Management Information System (MIS) will be\ndeveloped, providing geo-tagged information on infrastructure and services at farmer and district levels.\nThis system will be used to track infrastructure and services being provided and the functioning of farmer\norganizations, with the objective of supporting performance improvement. **At the\n\nMention: \"LG Irrigation Management Information System\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"A planned local-government irrigation MIS will capture geo-tagged infrastructure and service records to track delivery.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:021150:4:2", "occurrence_id": "a5e0bbdf6caee29555a08184", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  observed for the 2005-2009 and 2009-2012 periods respectively. The recent trends not only\ncompare unfavorably with Uganda’s past but also with peer countries. In addition, the regional gap in\nliving conditions within the country persists, with the Northern and Eastern region lagging behind the rest\nof Uganda across many dimensions; despite the progress observed in the North between 2012 and 2016.\n\n\n1 Uganda’s official poverty estimates are based on the Uganda National Household Survey (UNHS), collected by UBOS (Ugandan Bureau of\nStatistics). The latest waves were collected in 2012/13 and 2016/17, and the latest SCD covered the poverty trends up to 2012/13.\n2 This reliance on agriculture income is even more pronounced for the bottom 40 percent of the distribution (UBOS, 2017), which explains why\nthe drought increased the number of rural poor so markedly.\n\n\nMar 17, 2020 Page 5 of 24\n\nMention: \"Uganda National Household Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Official poverty estimates and their regional and temporal trends are drawn from successive UNHS waves.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:021197:17:1", "occurrence_id": "c0cf4c13aed4c21fe4860857", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: rievance mechanisms, and roll-out of the communication<br>strategy. These steps should better ensure that eligible<br>households from among vulnerable and marginalized<br>communities are included in the programme, are aware of<br>the programme and their entitlements under it and can<br>lodge complaints against front-line programme staff.|NSNP PIBS<br>survey|\n|Component<br>Two:<br>_Programme_<br>_Implementation_<br>_in Selected_<br>_Sub-counties_|This component aims at<br>using cash transfers to<br>strengthen the ability of<br>vulnerable households to<br>protect and care for<br>OVC, ensuring that OVC<br>stay within their<br>communities and can be<br>cared for effectively|(+) Improved beneficiary<br>household consumption,<br>enhanced social standing and<br>increased access to social<br>services;<br> <br>(-) Potential tension in some<br>communities because of few<br>households that benefit in a<br>context of widespread need;<br> <br> <br>(-) Possibility that Vulnerable|Increased government resourcing to broaden coverage of<br>the programme. This is a key objective of Component 2<br>of the Additional Financing and the planned expansion of<br>the NSNP;<br> <br>Robust community awareness sessions about the<br>programme objectives, procedures and resource<br>availability;<br> <br> <br>Design of specific procedures to ensure awareness of<br>programme officials and community members on<br>existence of remote communities and need to make effort|NSNP PIBS<br>survey and CT<br>OV\n\nMention: \"NSNP PIBS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The NSNP PIBS survey is listed as the data source for whether eligible vulnerable households are reached and can complain.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:021197:4:1", "occurrence_id": "998dd6bafe397997f0dde758", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  since the 1970s. However, since\n2007, the country has registered growth rates of above 5 percent only once - in 2010, which was\nalso the only year in which the country did not suffer from domestic or international shocks.\nHowever, poverty incidence remains high at 46.6 percent in 2005/06, having declined from 52.2\npercent in 2000. <sup>1</sup> [^1: This reflects the most recent nationally representative household survey. An updated Kenya Integrated Household\nBudget Survey (KIHBS) should be available in 2014.\n2Ministry of State for Planning, National Development, and Vision 2030 (2012), “Kenya Social Protection Sector\nReview,” Nairobi based on KIHBS 2005/06 and the 2009 Census.\n3MGCSD (2011), “Kenya National Social Protection Policy,” Ministry of Gender, Children, and Social\nDevelopment, Nairobi.] Poverty rates are markedly higher in rural areas (49.7 percent) than in urban\nareas (34.4 percent), although residents of informal urban settlements often experience great\ndeprivation. Poverty rates also tend to be higher among vulnerable groups such as children (53.5\npercent), including orphans and vulnerable children (54.1 percent), older people (53.2 percent),\nand people with disabilities (57.4 percent). <sup>2</sup>\n\n4. The Government of Kenya has recently approved the National Social Protection Policy,\nwhich aims to strengthen the delivery of social assistance to poor and vulnerable populations.\nThe National Social Protection Policy (NSPP), which was approved by the Cabinet in May 2012,\noutlines a vision for gradually realizing the right to social protection as articulated in the\nConstitution and, within this, the government’s ambition to enhance social assistance. <sup>3</sup> As a first\nstep in this reform agenda, the government is establishing a National Safety Net Programme\n(NSNP), which aims to strengthen operational systems while expanding\n\nMention: \"Kenya Integrated Household\nBudget Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"A footnote names the updated household budget survey as the forthcoming source for current poverty measurement.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:021197:4:1", "occurrence_id": "d0bfc034ea782e1787a577d8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  since the 1970s. However, since\n2007, the country has registered growth rates of above 5 percent only once - in 2010, which was\nalso the only year in which the country did not suffer from domestic or international shocks.\nHowever, poverty incidence remains high at 46.6 percent in 2005/06, having declined from 52.2\npercent in 2000. <sup>1</sup> [^1: This reflects the most recent nationally representative household survey. An updated Kenya Integrated Household\nBudget Survey (KIHBS) should be available in 2014.\n2Ministry of State for Planning, National Development, and Vision 2030 (2012), “Kenya Social Protection Sector\nReview,” Nairobi based on KIHBS 2005/06 and the 2009 Census.\n3MGCSD (2011), “Kenya National Social Protection Policy,” Ministry of Gender, Children, and Social\nDevelopment, Nairobi.] Poverty rates are markedly higher in rural areas (49.7 percent) than in urban\nareas (34.4 percent), although residents of informal urban settlements often experience great\ndeprivation. Poverty rates also tend to be higher among vulnerable groups such as children (53.5\npercent), including orphans and vulnerable children (54.1 percent), older people (53.2 percent),\nand people with disabilities (57.4 percent). <sup>2</sup>\n\n4. The Government of Kenya has recently approved the National Social Protection Policy,\nwhich aims to strengthen the delivery of social assistance to poor and vulnerable populations.\nThe National Social Protection Policy (NSPP), which was approved by the Cabinet in May 2012,\noutlines a vision for gradually realizing the right to social protection as articulated in the\nConstitution and, within this, the government’s ambition to enhance social assistance. <sup>3</sup> As a first\nstep in this reform agenda, the government is establishing a National Safety Net Programme\n(NSNP), which aims to strengthen operational systems while expanding\n\nMention: \"2009 Census\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The census is cited alongside the household budget survey as a basis of the social protection review's estimates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:000498:8:3", "occurrence_id": "5925c2b95d8a4c1c8c3da521", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  from PIMIS which is still being rolled out.|Corporate result indicator introduced recently. Baseline, actual, and targets to be updated<br>next reporting period using data from PIMIS which is still being rolled out.|Corporate result indicator introduced recently. Baseline, actual, and targets to be updated<br>next reporting period using data from PIMIS which is still being rolled out.|Corporate result indicator introduced recently. Baseline, actual, and targets to be updated<br>next reporting period using data from PIMIS which is still being rolled out.|Corporate result indicator introduced recently. Baseline, actual, and targets to be updated<br>next reporting period using data from PIMIS which is still being rolled out.|Corporate result indicator introduced recently. Baseline, actual, and targets to be updated<br>next reporting period using data from PIMIS which is still being rolled out.|\n|People with enhanced<br>resilience to climate<br>risks – Youth (Number<br>of people) CRI|||1,000,000|12-Jun-2025|1,000,000|10-Dec-2025|2,000,000|Dec/2028|\n|People with enhanced<br>resilience to climate<br>risks – Youth (Number<br>of people) CRI|Comments on<br>achieving targets|Comments on<br>achieving targets|Corporate result indicator introduced recently. Baseline, actual, and targets to be updated<br>next reporting period using data from PIMIS which is still being rolled out.|Corporate result indicator introduced recently. Baseline, actual, and targets to be updated<br>next reporting period using data from PIMIS which is still being rolled out.|Corporate result indicator introduced recently. Baseline, actual, and targets to be updated<br>next reporting period using data from\n\nMention: \"data from PIMIS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Corporate result indicators are to be updated next reporting period from PIMIS, which is still being rolled out.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:000614:19:1", "occurrence_id": "214d1d5b61ff45eaf7d6d199", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  data, as presented in the semi-annual and annual progress reports, along with the Annual Performance\nAssessment (APA), were instrumental for the World Bank and Government Authorities in addressing issues within various\nULGs and prioritizing field visits. These reports formed the foundation for discussions during Implementation Support\nMissions (ISM). M&E findings were systematically communicated to stakeholders and utilized to inform program\nmanagement and decision-making processes. The World Bank and Government teams leveraged this information to\nidentify challenges and implement targeted measures to address them. Furthermore, the APA figures were crucial for the\ndisbursement of funds based on performance results, thereby encouraging continuous performance improvement among\nparticipating ULG.\n**Justification of Overall Rating of Quality of M&E**\n36. **Overall, the quality of Monitoring and Evaluation (M&E) UIIDP was rated as Substantial.** Although the M&E\nframework was well-designed and implemented as planned, there were a few shortcomings. Delays in recruiting the APA\nfirm and the identification of some poor-quality infrastructure projects during site visits by the Bank team were noted.\nDespite these issues, the M&E system provided a solid foundation for tracking progress and informing decision-making\nprocesses.\n\n**B.** **ENVIRONMENTAL, SOCIAL, AND FIDUCIARY COMPLIANCE**\n\n**Environmental and Social**\n37. **Overall, the social and environmental safeguards performance of the Program was rated Moderately**\n**Satisfactory,** with no major issues reported. All 117 participating ULGs established environmental and social systems,\nendorsed the ESMG and RSG, and assigned safeguard focal persons before the UIIDP began. Federal and regional\nagencies, along with steering and technical committees, played crucial roles in coordination, training, oversight, audits,\nand reporting. The UIIDP incentivized effective E&S risk management systems through Disbur\n\nMention: \"APA figures\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Annual Performance Assessment figures were assessed against results and determined disbursement to local governments.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:001297:14:2", "occurrence_id": "32a0dd0cfbc65194f9d7b1da", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  specifically allocated to URRAP roads. In the same timeframe, out of the ETB\n3,482.0 million budgeted for Woreda road maintenance, only 19 percent was disbursed. This meager allocation for\nmaintenance has led to the rapid degradation of Woreda roads, as they receive insufficient attention and resources for\ntheir proper upkeep. This operation aims to close these gaps by utilizing a separate and targeted disbursement linked\nindicator (DLI) for rural road maintenance.\n\n**12.** **GoE has prepared a** Transport Sector’s 10-year perspective development plan **which was officially launched in**\n**2021.** Its main objective is to enhance the accessibility of target populations served by roads to markets and services and\nstrengthen the institutional capabilities of the road sector to deliver and maintain roads in a safe and satisfactory condition.\nTargets are set for transport infrastructure, transport services, safety, logistics, environmental climate changes,\nimplementation, and capacity building initiatives.\n\n**13.** **Improving rural communities' accessibility to agricultural inputs and market opportunities is critical for**\n**physically connecting both producers and consumers to markets.** While food security depends on many factors, this\noperation is prioritizing accessibility of potentially high food production areas to markets. This catalyzes increased\nagricultural production and contributes to reduced post-harvest losses thus improving availability of food and improving\n\n\n9 International Benchmarking of South Africa’s Infrastructure Performance: Infrastructure Development in Sub-Saharan Africa: A Scorecard: Affric\nIRF World Road Statistics 2019, latest year of available data for countries a’s Infrastructure, A Time for Transformation: The Global Competitiveness\nReport (2017/18)\n\n3\n\nMention: \"IRF World Road Statistics\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The road statistics database is credited in a footnote as the latest-year source for international infrastructure benchmarking.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:001487:35:0", "occurrence_id": "c85d0629fbae454bb57f3b1c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nEthiopia: National Quality Infrastructure Development Project (P160279) ICR DOCUMENT\n\n\n_Source: UN Comtrade data._\n\n\n**ANNEX 6. BORROWER, CO-FINANCIER AND OTHER PARTNER/STAKEHOLDER COMMENTS**\n\nReceived on Monday, December 29, 2025\n\n\n_Dear Bisrat,_\n\n\n_My sincere apologies for the late response, as I had unfortunately overlooked your message._\n\n\n_Thank you very much for sharing the draft Implementation Completion Report (ICR). I have reviewed the document carefully,_\n_and I appreciate the effort made in preparing a well structured and comprehensive report._\n\n\n_However, I would like to raise one important observation. In the final report, I could not find adequate reflection or_\n_acknowledgment of the role of the implementing ministry, including myself, as well as the focal persons and heads of the_\n_beneficiary institutions. The report appears to focus primarily on the task force team, while the contributions and_\n_responsibilities of the implementing ministry and beneficiary institutions are not sufficiently captured in one consolidated_\n_section._\n\n\n_I kindly request that this aspect be revisited and appropriately reflected in the final version of the report, to ensure balanced_\n_recognition of all key stakeholders involved in the implementation of the project._\n\n\n_Thank you again for sharing the draft, and I appreciate your continued collaboration._\n\n\n_Kind regards,_\n\n\n_Ato Endalew Mekonnen_\n_Minister of trade and regional integration of Ethiopia_\n\n\nPage 28\n\n\nOfficial Use Only\n\nMention: \"UN Comtrade data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"A trade figure in the annex is credited directly to the UN Comtrade data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:001520:4:0", "occurrence_id": "503de9c90dce14c5eb748077", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Prof. Njuguna Ndung'u\n\n\nand quality assurance of the **CBC** and **CBA** including adapting **CBC** for Special Needs\nEducation **(SNE)** and digitizing **CBC** for selected grades; **(d)** strengthen the supply chain for\nsanitary towels; (e) assist teachers in primary schools to access and utilize the online portal for\nformative learning assessments; **(f)** develop implementation guidelines for school meals and\nnutrition policy; **(g)** monitor the School-Based Teacher Support **(SBTS)** initiative; (h) support\nactivities for out of school children; (i) support the capacity of Kenya National Examinations\nCouncil **(KNEC)** to conduct and disseminate a National Assessment System for Monitoring\nLearner Achievement **(NASMLA)** for grade **3** learners; **(j)** analyze TPAD data in target schools\nfor lessons coverage; and **(k)** conduct assessments to measure teacher proficiency.\n\n\n**2.3.** Carrying out a program of activities designed to strengthen education system capacity, such\nprogram to include: (a) strengthening gender responsive planning and policy development for\nsystem wide impact, including through reviewing teacher recruitment, management, and\ndevelopment processes, and developing an action plan to address gender gaps/inequalities\nidentified; **(b)** training of county directors of education, KICD and **KNEC,** to integrate gender\naspects in annual workplans and budgets including for **CBC** and **CBA;** developing a gender\nequality module to train teacher trainees in target TTCs on gender equality in basic education;\nand establishing multi-agency national technical working group for coordination and\ncollaboration on girl's education; (c) mobilizing coordinated\n\nMention: \"TPAD data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Project funds require analysis of teacher appraisal data in target schools to check lessons coverage.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:001708:10:0", "occurrence_id": "d0aaf89e9974c8249c2cf3f9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **UGANDA** **INVESTING** **IN** FORESTS **AND** PROTECTED **AREAS** FOR **CLIMATE** SMART **DEVELOPMENT**\n**PROJECT;** **IDA** **CREDIT-6592-UG** **AND** GRANT **D596-UG**\nAnnual Financial Statements for the year ended **3** **0th** June **2023**\n\n\n**2.0** **PROJECT** **OBJECTIVES**\n\n\n\nThe main project objective is to improve sustainable management of forests and\n\nprotected areas and increase benefits to communities from forests in target landscapes.\nThe project is impimented under four major components and consists of the following\n\nparts:\n\n\nPart **1:** Improved Management of Protected Areas\n\n\n\n**1.1** (a) Carrying out a program of activities designed to improve the basic management\nof Protected Areas, such a program to include such activities as: (i) grading and\n\nmaintenance of tracks and trails within the Proteced Areas; (ii) planning (including\ncommunity consultations) and demarcation of the Protected Areas' boundaries; (iii)\ninstallation of infrastructure such as fences, walls and gates as well as construction\nof rangers' housing and posts; (iv) undertaking of community-oriented activities to\n\nmanage human-wildlife conflict including providing equipment as neededl (v)\nprocurement of field vehicles, road equipment, communication equipment, field,\nand office equipment; and (vi) support for the National Forest Monitoring System.\n\n\n\n**(b)** Carrying out a program of activities designed to improve wildlife management\nand control of invasive plants in the Protected Areas, sucha a program to include\nsuch activities as: (i) provision of technical assistance and advisory services to\nsupport\n\nMention: \"National Forest Monitoring System\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Project financing includes support for the national forest monitoring system alongside equipment procurement.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:001746:4:0", "occurrence_id": "44753e927342e11cfc297c28", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank** Implementation Status & Results Report\nEthiopia General Education Quality Improvement Program for Equity (P163050)\n\n\n\nComments\n\n\n\nThe Steering committee has been held at national and sub-national level which co-chaired by\nrespective institutions.\n\n\n\n\n\n\n\n\n\n\n|Action Description|1.7 At least one Education Working Group meeting devoted to refugee issues, including both<br>WEO and ARRA, held annually in each main refugee-hosting Woreda|Col3|Col4|Col5|Col6|\n|---|---|---|---|---|---|\n|Source|DLI#|Responsibility|Timing|Timing Value|Status|\n|Technical|NA|WEO/ARRA|Recurrent|Yearly|Completed|\n|Completion<br>Measurement|Meetings held as stipulated|Meetings held as stipulated|Meetings held as stipulated|Meetings held as stipulated|Meetings held as stipulated|\n|Comments|Meetings held as stipulated.|Meetings held as stipulated.|Meetings held as stipulated.|Meetings held as stipulated.|Meetings held as stipulated.|\n\n\n\n\n\n\n\n\n|Action Description|1.8. Include full data for refugee primary and secondary schools, on a par with MoE schools,<br>in Educational Statistics Annual Abstract (ESAA)|Col3|Col4|Col5|Col6|\n|---|---|---|---|---|---|\n|Source|DLI#|Responsibility|Timing|Timing Value|Status|\n|Technical|NA|MoE|Recurrent|Yearly|Completed|\n|Completion<br>Measurement|ESSA published annually including information for refugee schools for all primary and<br>secondary indicators\n\nMention: \"Educational Statistics Annual Abstract\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"A disbursement-linked action requires refugee school data to be included in the annual education statistics abstract on the same footing as ministry schools.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:001786:3:0", "occurrence_id": "a8cf3320cfad2c2bf8cd1e63", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: I advised the Accounting Officer to follow up with the Ministry of Education and\nsports, re-assess the project plan and have the census activity concluded in line\nwith the project objective.\n\n\n**8.2** **Under** **absorption of FAQ** **Grants**\n\n\nThe agreement between the Food and Agriculture Organization **(FAO)** of United\n\nNations and Uganda Bureau of Statistics was signed on the 14 <sup>th</sup> March **2018,** for\nthe support of the Advancement of the Uganda Annual Agricultural Survey **(AAS)**\nbased on the AgriSurvey Methodology **(GCP/GLO/677/USA).** The agreement was\nexpected to start on **28/02/2018** and terminate on **31/12/2019.**\n\n\n\nThe activity agreement stipulated that **UBOS** was to implement the project and\n\n**FAO** would fund the activity for a total amount of UGX.1.233Bn. The project\nimplementation started in FY **18/19** with the release of UGX.659.98Mn. During\nthe financial year **19/20** and 20/21, the Bureau received an additional\nUGX.3.669Bn as per the agreement for the annual agriculture survey. During the\nFY 2021/2022, an additional UGX.2.135Bn was received for the annual\nAgricultural Surveys **by** **UBOS.** However, the Bureau failed to absorb part of the\nfunds and refunded UGX.1.869Bn on the 22/12/2021 to the **FAO** International\n\nRepresentative.\n\n\n\nFailure to absorb the availed funds may be construed as lack of capacity to\nundertake such projects which may affect future funding of the Bureau.\n\n\nThe Accounting Officer explained that the\n\nMention: \"Uganda Annual Agricultural Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"An FAO grant to UBOS advanced the annual agricultural survey using the AgriSurvey methodology.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:002376:3:1", "occurrence_id": "127ab7bb9d416d75698211d7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  seekers as of 2019,\nthe most of any country in Africa, and has a progressive policy framework which provides for free access\nto education and training for refugees.\n\n5. **Despite gains in access, most students in Uganda are not acquiring basic literacy skills, and the situation**\n**has worsened since the onset of the COVID-19 pandemic** . According to _The State of Global Learning_\n_Poverty_, 81.9 percent of children in Uganda are not able to read and understand a simple passage by age\n10. <sup>3</sup> [^3: <u>[https://www.worldbank.org/en/topic/education/publication/state-of-global-learning-poverty](https://www.worldbank.org/en/topic/education/publication/state-of-global-learning-poverty)</u>] This estimate predates COVID-19-era school closures, which were among the longest in the world at\n22 months. A 2022 Early Grade Reading Assessment (EGRA) in six local languages found that the COVID\npandemic had largely erased gains in literacy since 2016. The share of P2 students who were “nonreaders” (i.e., unable to read a single word) was high in all languages, ranging from 66 percent in Luganda\nto 91 percent in Lusoga, and 73 percent were non-readers in English. <sup>4</sup> [^4: USAID. “Learning Recovery in Primary Schools in Uganda.” PowerPoint presentation, 6 December 2022.] Low literacy levels contribute to\nprimary survival and completion rates that are among the lowest in the region: out of the ten pupils who\nenter P1, around six reach P5 and only three reach P7.\n\n\n6. **Learning outcomes are also highly inequitable.** According to the 2018 National Assessment of Progress\nin Education (NAPE), a low-stakes national assessment, there was a 21.4 percentage-point gap in literacy\nproficiency rates between rural and urban districts in P3. Rural-urban gaps grow as\n\nMention: \"2022 Early Grade Reading Assessment\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The 2022 reading assessment across six local languages is used to show that pandemic school closures erased literacy gains made since 2016.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:002376:3:1", "occurrence_id": "e21b8fd0f18cd9fe8ad1d795", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  seekers as of 2019,\nthe most of any country in Africa, and has a progressive policy framework which provides for free access\nto education and training for refugees.\n\n5. **Despite gains in access, most students in Uganda are not acquiring basic literacy skills, and the situation**\n**has worsened since the onset of the COVID-19 pandemic** . According to _The State of Global Learning_\n_Poverty_, 81.9 percent of children in Uganda are not able to read and understand a simple passage by age\n10. <sup>3</sup> [^3: <u>[https://www.worldbank.org/en/topic/education/publication/state-of-global-learning-poverty](https://www.worldbank.org/en/topic/education/publication/state-of-global-learning-poverty)</u>] This estimate predates COVID-19-era school closures, which were among the longest in the world at\n22 months. A 2022 Early Grade Reading Assessment (EGRA) in six local languages found that the COVID\npandemic had largely erased gains in literacy since 2016. The share of P2 students who were “nonreaders” (i.e., unable to read a single word) was high in all languages, ranging from 66 percent in Luganda\nto 91 percent in Lusoga, and 73 percent were non-readers in English. <sup>4</sup> [^4: USAID. “Learning Recovery in Primary Schools in Uganda.” PowerPoint presentation, 6 December 2022.] Low literacy levels contribute to\nprimary survival and completion rates that are among the lowest in the region: out of the ten pupils who\nenter P1, around six reach P5 and only three reach P7.\n\n\n6. **Learning outcomes are also highly inequitable.** According to the 2018 National Assessment of Progress\nin Education (NAPE), a low-stakes national assessment, there was a 21.4 percentage-point gap in literacy\nproficiency rates between rural and urban districts in P3. Rural-urban gaps grow as\n\nMention: \"2018 National Assessment of Progress\nin Education\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The national assessment provides the rural-urban literacy proficiency gap used to show inequitable learning outcomes.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:002529:5:0", "occurrence_id": "fc62aad24570de6c16ce3235", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank** Implementation Status & Results Report\nStrengthen Ethiopia’s Adaptive Safety Net (P172479)\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n|Beneficiaries of cash-<br>based interventions -<br>Female (Number of<br>people) CRI|Comments on<br>achieving targets|Col3|Source FSCO FIC report for December 2025: Total 4,146,799 Minus Special woredas 59,230.|Col5|Col6|Col7|Col8|Col9|\n|---|---|---|---|---|---|---|---|---|\n|Beneficiaries of cash-<br>based interventions -<br>Youth (Number of<br>people) CRI|0|Jun/2020|0|30-Jun-2025|0|31-Mar-2026|0|Jun/2026|\n|Beneficiaries of cash-<br>based interventions -<br>Youth (Number of<br>people) CRI|Comments on<br>achieving  targets|Comments on<br>achieving  targets|The project does not track this indicator, but it cannot be deleted from the Operational<br>Workspace, as it is embedded in another indicator.|The project does not track this indicator, but it cannot be deleted from the Operational<br>Workspace, as it is embedded in another indicator.|The project does not track this indicator, but it cannot be deleted from the Operational<br>Workspace, as it is embedded in another indicator.|The project does not track this indicator, but it cannot be deleted from the Operational<br>Workspace, as it is embedded in another indicator.|The project does not track this indicator, but it cannot be deleted from the Operational<br\n\nMention: \"FSCO FIC report\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Cash transfer beneficiary counts in the results table are taken from the December 2025 financial and inclusion control report.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:002629:3:0", "occurrence_id": "4054f0495c6980ab6ad75395", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank** Implementation Status & Results Report\nEthiopia Digital Foundations Project (P171034)\n\n\nValue 0.00 0.00 0.00 65.00\n\n\nIN01283576\n\nNumber of digital economy jobs created for rural persons (Number, Custom Supplement)\n\n\n<u>Baseline</u> <u>Actual (Previous)</u> <u>Actual (Current)</u> <u>End Target</u>\n\n\nValue 0.00 0.00 0.00 1,300.00\n\n\n**Overall Comments PDO**\n\n\nInternet user statistics have been rebased downwards based on new data from ITU. Other indicators have not changed.\n\n\n**Intermediate Results Indicators by Components**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n5/25/2023 Page 4 of 9\n\nMention: \"new data from ITU\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Internet user statistics were rebased downwards once newer international telecommunication data became available.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:003095:24:1", "occurrence_id": "1d1202f7a323f885777dea9c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **<br>**Original Target**<br>**Formally Revised**<br>**Target**<br>**Actual Achieved at**<br>**Completion**<br>Urban areas that utilize at<br>least 50 percent of the<br>budget intended for their<br>urban investments <br>Number<br>0.00<br>17.00<br> <br>57.00<br> <br>31-Jul-2017<br>31-Dec-2023<br> <br>31-Dec-2023<br> <br>**Comments (achievements against targets):** <br>Target surpassed: 96% of beneficiary municipalities absorbed 100% of UDG allocation under FY 2018/2019 UDG allocation and over 50% of the 2019/2020<br>UDG allocation per the program’s eligible menu. According to APA 3 (last and final APA), 57 municipalities had used 100% of the 2018/2019 UDG allocation<br>and over 50% of the 2019/2020 allocation, per the eligible expenditure menu. Only two municipalities, Kitui and Hola, did not meet this minimum<br>condition. Although no new APA was undertaken, NPCT has provided demand-driven support to Kitui and Hola municipalities to attain this target. <br>|Urban areas that utilize at<br>least 50 percent of the<br>budget intended for their<br>urban investments <br>|Number|0.00|17.00||57.00|57.00|\n|<br>**Indicator Name**<br>**Unit of Measure Baseline**<br>**Original Target**<br>**Formally Revised**<\n\nMention: \"APA 3\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Municipal absorption of the intergovernmental allocation is reported from the final annual performance assessment.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:003157:7:0", "occurrence_id": "6160a274b70c7e1af7abfe5f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nHorn of Africa Gateway Development Project (P161305)\n\n\n**Table 1: Resettlement Compensation of Project Affected Persons: Upgrading the Road Sections Financed**\n\n**by the World Bank**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n|Road Section1|Resettlement<br>Category|Number of<br>PAPs|Resettlement<br>Compensation in<br>US$million|Resettlement Compensation in<br>US$million|\n|---|---|---|---|---|\n|**Road Section1**|**Resettlement**<br>**Category**|**Number of**<br>**PAPs**|**(As per revised RAP**<br>**2019)2**|**(As per NLC validation 2023)3**|\n|Isiolo-Modogashe (200km)|Assets and<br>livelihood|1,511|4.40|5.80|\n|Wajir-Elwak (200km)|Assets and<br>livelihood|955|5.90|7.404|\n|** Total**|** Total**|2,466|**10.30**|**13.20**|\n|Contingency 10% to cover<br>for price adjustment5|Assets and<br>livelihoods||1.10|1.30|\n|**Grand Total**<br>|**Grand Total**<br>|**2,466**|**11.40**|**14.50**|\n\n\n_2. KeNHA used 1US$=KES130 exchange rate in deriving the\n\nMention: \"NLC validation 2023\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Resettlement compensation totals are shown both as planned in the 2019 plan and as revalidated in 2023, raising the compensation estimate.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:003243:6:0", "occurrence_id": "c42fbe679de9530cbd6fbb88", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: -7\n\n**SCHEDULE 1**\n\n\n**Project Description**\n\n\nThe objective of the Project is to: (i) rebuild and improve access to basic services\nand climate-resilient community infrastructure; and (ii) improve access to multi-sectoral\nresponse services for Gender-Based Violence (GBV) survivors; in selected conflictaffected communities in Ethiopia.\n\n\nThe Project consists of the following parts _subject to such modifications thereof as_\n_the Recipient and the Bank may agree upon from time to time_ :\n\n\n**_Part 1: Rebuilding and Improving Access to Basic Services and Climate-resilient_**\n\n**_Community Infrastructure_**\n\n\nImplementation of the following program of activities designed to rebuild and\nimprove access to basic services and climate-resilient community infrastructure in\ntargeted communities, namely:\n\n\n1.1. Provision of mobile essential services such as health, education and WASH for\nconflict-affected communities based on, _inter alia_, findings of the post conflict\nimpact assessment and recovery and reconstruction plan, and the Woreda needs,\nconflict, and capacity assessments.\n\n\n1.2. (a) Carrying out of community-led (including key stakeholders such as displaced\nhouseholds and host communities) consultations and planning exercises with a\nview to identifying community recovery development investments such as\nreconstruction of existing community health, education and WASH facilities or\nconstruction of new climate-resilient facilities or communal livelihood\ninfrastructure (“Community Subprojects”) and preparing associated community\nrecovery development plans; (b) implementation of said Community Subprojects;\nand (c) strengthening the institutional and technical capacities at the federal,\nregional, Woreda and Kebele levels for implementation of said plans and\nCommunity Subprojects, all through the provision of goods, works, consulting\nservices and non-consulting services for the purpose.\n\n\n1.3. Strengthening of institutions at the federal, regional, Woreda, and Kebele levels\nfor\n\nMention: \"post conflict\nimpact assessment and recovery and reconstruction plan\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Mobile essential services for conflict-affected communities are to be provided on the basis of the recovery plan's findings.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:003243:6:0", "occurrence_id": "16d422236fa328b350f49225", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: -7\n\n**SCHEDULE 1**\n\n\n**Project Description**\n\n\nThe objective of the Project is to: (i) rebuild and improve access to basic services\nand climate-resilient community infrastructure; and (ii) improve access to multi-sectoral\nresponse services for Gender-Based Violence (GBV) survivors; in selected conflictaffected communities in Ethiopia.\n\n\nThe Project consists of the following parts _subject to such modifications thereof as_\n_the Recipient and the Bank may agree upon from time to time_ :\n\n\n**_Part 1: Rebuilding and Improving Access to Basic Services and Climate-resilient_**\n\n**_Community Infrastructure_**\n\n\nImplementation of the following program of activities designed to rebuild and\nimprove access to basic services and climate-resilient community infrastructure in\ntargeted communities, namely:\n\n\n1.1. Provision of mobile essential services such as health, education and WASH for\nconflict-affected communities based on, _inter alia_, findings of the post conflict\nimpact assessment and recovery and reconstruction plan, and the Woreda needs,\nconflict, and capacity assessments.\n\n\n1.2. (a) Carrying out of community-led (including key stakeholders such as displaced\nhouseholds and host communities) consultations and planning exercises with a\nview to identifying community recovery development investments such as\nreconstruction of existing community health, education and WASH facilities or\nconstruction of new climate-resilient facilities or communal livelihood\ninfrastructure (“Community Subprojects”) and preparing associated community\nrecovery development plans; (b) implementation of said Community Subprojects;\nand (c) strengthening the institutional and technical capacities at the federal,\nregional, Woreda and Kebele levels for implementation of said plans and\nCommunity Subprojects, all through the provision of goods, works, consulting\nservices and non-consulting services for the purpose.\n\n\n1.3. Strengthening of institutions at the federal, regional, Woreda, and Kebele levels\nfor\n\nMention: \"Woreda needs,\nconflict, and capacity assessments\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"District-level needs, conflict and capacity assessments are named alongside the recovery plan as the basis for locating services.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:003400:3:3", "occurrence_id": "6859533e5a3ec3934b3ee276", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  small enterprises through MNOs is ($70 - $1500) and<br>the average loan size of loans to SMEshrough MFBs is ($1,000 - $5,588).|\n|Of which, women<br>borrowers (Number)|0.00|Nov/2021|200.00|30-Nov-2024|12,584|30-May-2025|109,382.00|Dec/2026|\n|Of which, women<br>borrowers (Number)|Comments on<br>achieving targets|Comments on<br>achieving targets|This indicator is designed to include all women-owned MSMEs that receive financing under<br>the project's Component 1 (with an assumption that 40% of borrowers are women based on<br>MNO data)|This indicator is designed to include all women-owned MSMEs that receive financing under<br>the project's Component 1 (with an assumption that 40% of borrowers are women based on<br>MNO data)|This indicator is designed to include all women-owned MSMEs that receive financing under<br>the project's Component 1 (with an assumption that 40% of borrowers are women based on<br>MNO data)|This indicator is designed to include all women-owned MSMEs that receive financing under<br>the project's Component 1 (with an assumption that 40% of borrowers are women based on<br>MNO data)|This indicator is designed to include all women-owned MSMEs that receive financing under<br>the project's Component 1 (with an assumption that 40% of borrowers are women based on<br>MNO data)|This indicator is designed to include all women-owned MSMEs that receive financing under<br>the project's Component 1 (\n\nMention: \"MNO data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Operator transaction data underpins the assumption that 40 percent of micro and small enterprise borrowers are women.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:003840:24:0", "occurrence_id": "73876bc347732d605d2d40f9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nEthiopia First Sustainable and Inclusive Growth Development Policy Operation (P181591)\n\n\nExpanded tax base and reduced tax expenditures will support higher and more sustainable domestic financing for\ndevelopment.\n\n\n**Prior Action** **5** : To support more efficient, transparent, and sustainable procurement, the Recipient’s Parliament\nhas approved the Federal Public Procurement and Property Administration Proclamation mandating the use of eprocurement and the incorporation of green procurement principles.\n\n\n**Indicative Trigger 4:** To strengthen public procurement, the Recipient has: (a) amended the Federal Public\nProcurement and Property Administration Proclamation to: (i) expand its applicability to public bodies, including\nfederal public entities; and (ii) mandate the disclosure of beneficial ownership; (b) mandated the Federal Public\nProcurement and Property Authority to conduct performance audits; and (c) mandated the use of the e-GP system\nfor all public entities.\n\n\n56. **Rationale:** Public procurement amounts to 70 percent of spending by federal public entities (US$10 billion\na year), with significant opportunities to strengthen efficiency, transparency, and integrity and to support climate\naction by the public and private sectors. Critical challenges identified in the Ethiopia 2021 MAPS Assessment include\na fragmented legal/regulatory framework, the exclusion of SOEs from coverage, insufficient disclosure, overlapping\nfunctions and duties of regulatory bodies, and generally weak accountability, integrity, and transparency. A new egovernment procurement (e-GP) platform brings substantial benefits but is used by only 40 percent of federal public\nentities. Other challenges include the irregularity of procurement audits and weaknesses in complaint mechanisms. <sup>15</sup> [^15: Ethiopia 2021 Methodology for Assessing Procurement Systems (MAPS) Assessment.]\n\n\n57. **Substance** : The new law enshrines applicability to all federal public bodies and defines value for money from\npublic procurement. It also clearly defines competitive bidding, transparency, fairness, and accountability concepts.\nThe new law also\n\nMention: \"Ethiopia 2021 MAPS Assessment\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The assessment's diagnosis of fragmented regulation, SOE exclusion and weak disclosure frames the passage's account of public financial management weaknesses.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:003956:2:2", "occurrence_id": "9cc5414be7e1026941cfc716", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  NRMD M&E System|Administrative data and FY20 results accessible in NRMD M&E System|\n|Comments|The NRM Map database is fully functional and capable of storing spatial and non-spatial<br>data, plans and reports related to the program.|The NRM Map database is fully functional and capable of storing spatial and non-spatial<br>data, plans and reports related to the program.|The NRM Map database is fully functional and capable of storing spatial and non-spatial<br>data, plans and reports related to the program.|The NRM Map database is fully functional and capable of storing spatial and non-spatial<br>data, plans and reports related to the program.|The NRM Map database is fully functional and capable of storing spatial and non-spatial<br>data, plans and reports related to the program.|\n\n\n\n|Action Description|T6: MoU between RLAUD and regional BLAUs on: (i) staffing and facilities, setting conditions<br>for RLAUD support for SLLCs and NRLAIS roll-out; and, (ii) a system of incentives to promote<br>the updating of land transactions in NRLAIS.|Col3|Col4|Col5|Col6|\n|---|---|---|---|---|---|\n|Source|DLI#|Responsibility|Timing|Timing Value|Status|\n|Technical|6|RLAUD, BLAUs|Due Date|30-Sep-2019|Completed|\n\n\nAug 07, 2025 Page 3 of 15\n\nMention: \"NRMD M&E System\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Administrative data and results are to be made accessible in the programme's monitoring and evaluation system.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:004575:4:0", "occurrence_id": "8cc4e8e4a0c81055274c9b48", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank** Implementation Status & Results Report\nSecond Ethiopia Resilient Landscapes and Livelihoods Project (P174385)\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n|Project area showing an<br>increase in NDVI correcting<br>for climate effects<br>(Percentage)|Comments on<br>achieving targets|Col3|The Normalized Difference Vegetation Index (NDVI) measures photosynthetic activity and<br>vegetation cover. Changes in vegetation cover and intensity correlates with improvement in<br>land productivity, increased carbon storage, and greater resilience to climate change due to<br>improved absorptive and adaptive capacity (as per the PDO). Utilizing visible-red and near-<br>infrared spectral bands, NDVI isne measure for detecting vegetation cover and can be used<br>to track changes in vegetation over time. This indicator is meant to add value when used in<br>combination with other indicators, and provides a benchmark for physical achievement<br>under the operation and can be computed using remote-sensed satellite imagery data.<br>Progress under this indicator is tracked by computing, at the pixel-level (using a spatial<br>resolution of 30mx30m), the change in annual average NDVI from baseline, selecting pixels<br>showing an improvement over the baseline after adjusting for external factors (i.e. seasonal<br>or climatic variables). NDVI values are computed using medium resolution imagery (i.e.<br>LandSat 8 or Sentinel-2) and incorporating a masking routine to exclude pixels that can result<br>in unreliable estimates (i.e. containing clouds, shadows, water cover, etc.). The share of the<br>project area showing an improvement in NDVI is evaluated after an appropriate lag (i.e. 1 or<br>2 years) on areas where interventions have taken\n\nMention: \"LandSat 8\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Medium-resolution satellite imagery is used to compute the NDVI improvement indicator, with masking of clouds, shadows and water.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:004575:4:0", "occurrence_id": "a40039cdfb7649deeb883246", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank** Implementation Status & Results Report\nSecond Ethiopia Resilient Landscapes and Livelihoods Project (P174385)\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n|Project area showing an<br>increase in NDVI correcting<br>for climate effects<br>(Percentage)|Comments on<br>achieving targets|Col3|The Normalized Difference Vegetation Index (NDVI) measures photosynthetic activity and<br>vegetation cover. Changes in vegetation cover and intensity correlates with improvement in<br>land productivity, increased carbon storage, and greater resilience to climate change due to<br>improved absorptive and adaptive capacity (as per the PDO). Utilizing visible-red and near-<br>infrared spectral bands, NDVI isne measure for detecting vegetation cover and can be used<br>to track changes in vegetation over time. This indicator is meant to add value when used in<br>combination with other indicators, and provides a benchmark for physical achievement<br>under the operation and can be computed using remote-sensed satellite imagery data.<br>Progress under this indicator is tracked by computing, at the pixel-level (using a spatial<br>resolution of 30mx30m), the change in annual average NDVI from baseline, selecting pixels<br>showing an improvement over the baseline after adjusting for external factors (i.e. seasonal<br>or climatic variables). NDVI values are computed using medium resolution imagery (i.e.<br>LandSat 8 or Sentinel-2) and incorporating a masking routine to exclude pixels that can result<br>in unreliable estimates (i.e. containing clouds, shadows, water cover, etc.). The share of the<br>project area showing an improvement in NDVI is evaluated after an appropriate lag (i.e. 1 or<br>2 years) on areas where interventions have taken\n\nMention: \"Sentinel-2\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Alongside the other medium-resolution source, this imagery supplies the pixels from which NDVI change is measured.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:005010:6:2", "occurrence_id": "fc5664d86abcd4fdad4d2f18", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 59 months, and postnatal care. In\naddition, RBF increased the level of non-wage recurrent financing to health facilities with some\nfacilities receiving more than two times what they used to get per year reversing years of decline in\nfunding. No specific data are provided in the ICR.\n\n   - The preliminary results of the 7th Uganda Demographic and Health Survey (2022) showed\n\nimprovements in the coverage of key RMNCAH outcomes. Since the previous DHS in 2016, deliveries\nin health facilities had increased from 73.0% to 91.0%; visits to antenatal clinics (4 or more visits)\nincreased from 60.0% to 72.0%; childhood mortality declined from 64 to 52 (per 1000 live births);\n\n\nPage 7 of 17\n\nMention: \"Uganda Demographic and Health Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Preliminary results from the latest survey round are compared with 2016 to show rises in facility delivery and antenatal visits.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:005032:14:5", "occurrence_id": "9b16afd4d3767e45cff87fe1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Tongue Reading in Phase 1 schools (by gender)**|**Improvement in Grade 2 learning outcomes in Mother Tongue Reading in Phase 1 schools (by gender)**|**Improvement in Grade 2 learning outcomes in Mother Tongue Reading in Phase 1 schools (by gender)**|**Improvement in Grade 2 learning outcomes in Mother Tongue Reading in Phase 1 schools (by gender)**|**Improvement in Grade 2 learning outcomes in Mother Tongue Reading in Phase 1 schools (by gender)**|**Improvement in Grade 2 learning outcomes in Mother Tongue Reading in Phase 1 schools (by gender)**|\n|Indicator Name|Baseline|Baseline|Actual (Previous)|Actual (Previous)|Actual (Current)|Actual (Current)|Closing Period|Closing Period|\n|Indicator Name|Value|Month/Year|Value|Date|Value|Date|Value|Month/Year|\n|Average % of students in<br>Grade 2 who can read at<br>functional fluency level or<br>above in Mother Tongue<br>(Text)|25|Jun/2018|23|18-Apr-2024|23|01-Aug-2024|25|Jul/2023|\n|Average % of students in<br>Grade 2 who can read at<br>functional fluency level or<br>above in Mother Tongue<br>(Text)|Comments on<br>achieving  targets|Comments on<br>achieving  targets|Results for phase 1 schools<br>Source: EGRA 2023|Results for phase 1 schools<br>Source: EGRA 2023|Results for phase 1 schools<br>Source: EGRA 2023|Results for phase 1 schools<br>Source: EGRA 2023|Results for phase 1 schools<br>Source: EGRA\n\nMention: \"EGRA 2023\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Grade 2 mother-tongue reading results for phase 1 schools are credited to this assessment round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:005623:4:0", "occurrence_id": "91fefbad11c533c9728a76b4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nOne WASH—Consolidated Water Supply, Sanitation, and Hygiene Account Project (One WASH—CWA) (P167794)\n\n\nSectoral and Institutional Context\n\n\n3. **Ethiopia has made significant progress in improving water supply and sanitation (WSS) services in the past**\n**ten years** . Over 52 million people in Ethiopia now live within 1.5 kilometers (km) of an improved drinking water source\ncompared with only 6 million people in 1990. Over the same period, rates of open defecation (OD) fell by 63 percent,\nwhich was the largest decrease observed in the world. <sup>5</sup> [^5: World Bank, _Maintaining the Momentum While Addressing Service Quality and Equity: A Diagnostic of Water Supply, Sanitation,_\n_Hygiene, and Poverty in Ethiopia_ (Washington, DC: World Bank, 2018), 1.] About 67 million people gained access to a latrine, at an average\nrate of 2.6 million people per year. Sanitation and hygiene promotion were integrated into wider health-care delivery\nmechanisms and utilized behavioral change communication (BCC).\n4. **Notwithstanding the progress, much remains to be done.** At present, about 60 million people lack access to\nbasic drinking water, 112 million people lack access to basic sanitation, and about 22 million practice open defecation\n(JMP, 2022). In rural areas, basic drinking water <sup>6</sup> is accessible to only 42% of the population compared to 83% in urban\nareas <sup>7</sup> [^7: As per JMP 2022 report, 39% of urban population have access to safely managed, 76% have the service accessible on premises, 68% have water supply available when\nneeded and 39% free from contamination.] . About 76% and 39% of rural and urban schools, and 45% and 22% of rural and urban health facilities are without\nwater\n\nMention: \"JMP 2022 report\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Urban access to safely managed water and its component conditions are quoted from the joint monitoring programme report.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:006041:36:1", "occurrence_id": "84603fa451f0b101a562165f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  deemed high due to a number of factors such\nas transformers, defective or rotten poles which needed replacement\nin the shortest of time after the commissioning of these lines. For\ninstance, according to UEDCL data for Kabale – Kisoro – Bunagana\nScheme, the annual maintenance cost was at UGX.34.7Bn against\nUGX.518Bn generated revenues. This translates 6.8% of revenue\nwhich is above the 5% threshold requirement.\n\nMention: \"UEDCL data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Utility maintenance cost and revenue figures are compared to show maintenance spending exceeds the five percent revenue threshold.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:006079:4:1", "occurrence_id": "79a52c2f3dd3753ac9285db1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  I and II) 2015-2025. This framework is a\nkey component of both phases of Ethiopia’s Growth and Transformation Plans (GTPI and II),\ndemonstrating the government’s support for prioritization of human development, allocation of\nresources to high-priority primary health care (PHC) interventions, and a robust community\nownership approach built upon an extensive Community Health Workers program- the Health\nExtension Program and the Health Development Army.\n\n**3.** **The government of Ethiopia has embarked on a multi-sectorial approach, especially at**\n**the lower level, to address lagging dimensions of health outcomes and improve the very low**\n**Human Capital Index (HCI).** Over the past decade, the Government of Ethiopia has invested a\nsignificant portion of its resources to ensure that its people have access to basic services such as\nHealth, Water, Sanitation, and Hygiene (WASH) and Education. Ethiopia has made progress in\nrecent years in critical human capital outcomes. It achieved the MDG on child mortality three years\nahead of target and has made progress in child survival rates and primary education enrollment. <sup>1</sup>\nHowever, these improvements in human capital come from a low base of essential health,\neducation, agriculture, and WASH services and Ethiopia ranks among the bottom 21 of the 174\ncountries included in the HCI. The 2020 Human Capital Index (HCI) estimates that a child born in\n\n\n1 National Planning Commission and the United Nations in Ethiopia. October 2015. _Millennium Development_\n_Goals Report 2014 Ethiopia: Assessment of Ethiopia’s Progress towards the MDGs._ Addis Ababa, Ethiopia.\nhttps://www.undp.org/ethiopia/publications/ethiopia-mdg-report-2014\n\n\n5\n\nMention: \"Human Capital Index\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"The low index value is invoked as the reason government pursued a multi-sectoral approach to health outcomes.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:006079:4:1", "occurrence_id": "48aab871fb2283943e4662b1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  I and II) 2015-2025. This framework is a\nkey component of both phases of Ethiopia’s Growth and Transformation Plans (GTPI and II),\ndemonstrating the government’s support for prioritization of human development, allocation of\nresources to high-priority primary health care (PHC) interventions, and a robust community\nownership approach built upon an extensive Community Health Workers program- the Health\nExtension Program and the Health Development Army.\n\n**3.** **The government of Ethiopia has embarked on a multi-sectorial approach, especially at**\n**the lower level, to address lagging dimensions of health outcomes and improve the very low**\n**Human Capital Index (HCI).** Over the past decade, the Government of Ethiopia has invested a\nsignificant portion of its resources to ensure that its people have access to basic services such as\nHealth, Water, Sanitation, and Hygiene (WASH) and Education. Ethiopia has made progress in\nrecent years in critical human capital outcomes. It achieved the MDG on child mortality three years\nahead of target and has made progress in child survival rates and primary education enrollment. <sup>1</sup>\nHowever, these improvements in human capital come from a low base of essential health,\neducation, agriculture, and WASH services and Ethiopia ranks among the bottom 21 of the 174\ncountries included in the HCI. The 2020 Human Capital Index (HCI) estimates that a child born in\n\n\n1 National Planning Commission and the United Nations in Ethiopia. October 2015. _Millennium Development_\n_Goals Report 2014 Ethiopia: Assessment of Ethiopia’s Progress towards the MDGs._ Addis Ababa, Ethiopia.\nhttps://www.undp.org/ethiopia/publications/ethiopia-mdg-report-2014\n\n\n5\n\nMention: \"2020 Human Capital Index\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The index round supplies the country ranking and the child-level human capital estimate behind the low-base argument.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:006890:30:0", "occurrence_id": "0e71e9458d03925bed1975bc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 5. Record national health facilities information using a database system (Master Facility Registry) and\navail full national data.\n6. Conduct national health facilities assessment according to the national standards, label health facilities\n(green, yellow and red) and take actions as necessary\n7. Capacity building and support for regional health regulatory bodies (training, technical and financial\nsupport).\n\nThe Healthcare Waste Management Directive (2005), Proclamation No. 661/2009, Regulation No.\n299/2013, Health facilities standards since 2012 and Health Facilities Promotion Directive (2014) are being\nutilized in guiding the activities of the directorate.\n\nThe health regulatory structures are not uniform from federal to woreda levels. Due to this, the\nimplementation of regulatory laws and standards are not uniform across the regulatory bodies at each level.\n\n\n**Hygiene and Environmental Health Directorate**\n\n\nHygiene and Environmental Health Directorate has four case teams: Sanitation, Water, Food safety and\nhygiene, Institutional WASH and Climate resilient and Social conformity. The Roles and responsibilities\nof the directorate include:\n\n\n  - Environmental Health Impact Assessment,\n\n  - Promoting hygiene and environmental health services,\n\n  - Prepare policies, strategies, guidelines, and manuals, coordinate and support the implementation,\n\n  - Strengthen institutional arrangement and implementation capacity for the program at all level,\n\n  - Mobilize and properly utilize resources for program implementation,\n\n  - Conduct research for evidence based decision,\n\n  - Provide capacity building supports for concerned stakeholders,\n\n  - Strengthen and coordinate multi-sectorial and private sector integration,\n\n  - Monitor and evaluate progresses against targets,\n\n  - Facilitate experience sharing and scaling up of good experiences,\n\n  - Strengthening climate resilient health system,\n\n\nThe Health care waste management Manual (2020), Health Impact Assessment Guideline (2021) and\nWASH in health care facilities guidelines are being utilized in guiding the activities of the directorate.\n\n\n**Women, Youth and Children Affairs Directorate**\n\n\nThe Roles and responsibilities of the directorate include:\n\n\n - Facilitate Gender Mainstreaming in programs and operations,\n\n - Enhance women empowerment at institutional\n\nMention: \"Master Facility Registry\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"A reform milestone requires health facility information to be recorded in the registry and full national data to be made available.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:007039:5:0", "occurrence_id": "1647e4f04486851a86697917", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nResponse - Recovery - Resilience for Conflict-Affected Communities in Ethiopia (P177233)\n\n\nsurface water supply will heighten the vulnerability of populations depending on these sources for daily\nactivities; more intense and frequent storms and flooding may cause storm water flows, which increase the\nlikelihood of water contamination of both surface sources and shallow wells. This is potentially a particularly\nserious adverse impact as people rely heavily on surface water when wells dry up. Increased temperatures\nand intense rainfall are putting greater pressure on the water and sanitation sector, with potential to further\nimpact development gains. <sup>13</sup>\n\n\n7. **Conflict, impacts from climate variability and change, and associated humanitarian crises have led to**\n**large-scale internal displacement, particularly after 2016.** <sup>**14**</sup> Data from the International Organization for\nMigration (IOM) from September 2021 (the latest available data) indicate that there are 4.23 million IDPs in\nEthiopia, constituting one of the largest IDP populations in the world. <sup>15</sup> Just over 50 percent of IDPs are female,\nwhile at least 720,000 children have been displaced. <sup>16, 17</sup> The vast majority of IDPs (85 percent) have been\ndisplaced by conflict, while other main causes of displacement are climate-induced, namely drought (seven\npercent) and seasonal floods (three percent). More minor causes are also climate-induced, namely flash\nfloods, landslides, swampy lands, and strong winds. Notably, there are regional differences between the\nproportion of IDPs displaced by conflict and those displaced because of other reasons, including climaterelated disasters. Approximately 40 percent of the IDPs in the country are in a state of protracted\ndisplacement (displaced for more than five years) and the remaining 60\n\nMention: \"Data from the International Organization for\nMigration\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Latest tracking data give the internally displaced population, its composition and how recent the enumeration is.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:007316:5:0", "occurrence_id": "b4aea27610193e38fd7bc000", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Example: Psychometric Credit Scoring** <sup>**8**</sup>\n\nPsychometric (personality) data from loan applicants are\nused to predict likelihood of loan repayment to reduce the\nrisk of lending to “thin file” clients with little credit history.\nThe psychometric test for WEDP clients was designed\nby an international vendor who provides the quiz and\nproduces a credit score that the WEDP MFI can use to\ninform lending decisions.\n\n\nPotential borrowers take an interactive, tablet-based test\nconsisting of business scenarios, puzzles, and questions.\n\n\n\nThe quiz assesses potential borrowers’ traits such as fluid\nintelligence, business acumen, conscientiousness, riskaversion, and honesty. These traits are then combined\nvia machine learning into a model to predict credit risk. <sup>9</sup>\n\n\nA study with Wasasa Microfinance, located in Oromia in\nEthiopia, was conducted for 357 women entrepreneurs\nwho were registered with WEDP from 2018-2021. A\nrandomized controlled trial impact evaluation randomly\nassigned 131 firms who scored high enough on the\ntest - half into a treatment group that were offered an\nuncollateralized loan and half to a control group that were\nnot offered a loan. The “WEDP-X” loans were individualliability and worth up to US$7,500. Loans were labelled\n“WEDP-X” to distinguish them from the WEDP portfolio\nsince these loans were uncollateralized and screened\nthrough psychometric testing. <sup>10</sup> Findings suggest the\nloans led to a reduction in the likelihood of firm closure\nduring the coronavirus pandemic and conflict in Ethiopia:\n33% of firms closed in the control group and 17% of\nfirms closed in the loan treatment group after 3 years.\nThere were large positive impacts on access to finance\nwhereby assignment to the treatment group doubled the\nlikelihood that a woman had accessed a formal loan from\nany bank or MFI for their business (from 42\n\nMention: \"Psychometric (personality) data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Applicant personality test responses are modelled to predict repayment and so extend credit to clients with thin credit files.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:007333:46:1", "occurrence_id": "16b792f49331adaa16c87110", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  2\n\n\n(sample-based area estimate) is presented below.\n\n\nThe study used Oromia region spatial data from Oromia Planning Commission and Economic\n\n\nDevelopment which has full information of 21 zones with total area of about 37.12 million hectares.\n\n\nBased on this boundary, 92,820 sample points were systematically plotted with 2 by 2 Km grid\n\n\ninterval. All information related to each sample plots land use land cover and change information\n\n\nin this boundary was collected using CEO and analysed using Sample based area estimation to get\n\n\ndeforestation baseline of each 21 zones (although there is a change in number of Zones currently\n\n\nrequiring update) for phase one Emission Reduction Result Based Payment\n\n\n**Regional rate of Deforestation and other Land use change rates baseline**\n\n\nBased on the deforestation Baseline assessment conducted between 2007 and 2017, about\n\n\n8,930,531.7 ha was identified as forest in 2007 and the extent of forest cover in 2017 is reduced to\n\n\n**47 |** P a g e\n\nMention: \"deforestation Baseline assessment\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Forest extent at the start and end of the assessment window is used to set the regional deforestation baseline for result-based payments.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:007599:6:0", "occurrence_id": "1ee8ec990cc6da2657e2b401", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nHealth Sustainable Development Goals Program-for-Results (P123531)\n\n|Percent of pregnant women taking iron and folic<br>acid (IFA)|42.1%|60%|54.1%|\n|---|---|---|---|\n|Percent of Children 6-59 months receiving<br>Vitamin A Supplements|45%|45%*|53%|\n\n\n\n- No updates will be verified by next DHS survey planned for 2021\nData sources: Ethiopia Demographic Health Survey (EDHS) 2016 report; 2019 Mini-DHS; Nutrition Data Verification Joint Review Mission Report,\nDec 2018; 2012 HMIS Data Quality Assessment (DQA) Report; 2014 HMIS DQA and current data from 2018 SARA report.\n\n\n4. With regard to the three IPF sub-components, the Program status is as follows:\n\n\n**(a)** **_Sub-component I: Support for Civil Registration and Vital Statistics (CRVS)_** : The June 2021 implementation\nsupport mission noted significant progress in undertaking the comprehensive, national assessment and in\ndeveloping the five-year CRVS System Improvement Strategy and Costed Action Plan (2022-2026). Advocacy and\nCRVS awareness raising activities have been conducted at federal, regional, woreda and kebele levels to create\npublic awareness on the importance of CRVS. The printing of blank birth and death certificates and registry books\nfor regions is completed and all the printed registration materials have been delivered to regions. However, the\nprinted materials at the central store have not yet been delivered to Tigray region due to the security situation and\nwill be delivered once the situation improves. Trainings have been provided to the different stakeholders who\nactively work on the CRVS system. However, the national Immigration Nationality and Vital Events Agency (INVEA)\nis experiencing delays in getting timely data and\n\nMention: \"Ethiopia Demographic Health Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The survey report is listed first among the verified data sources for the nutrition and maternal-health indicators.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:007599:6:0", "occurrence_id": "0c416ec10e34a33810421fd1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nHealth Sustainable Development Goals Program-for-Results (P123531)\n\n|Percent of pregnant women taking iron and folic<br>acid (IFA)|42.1%|60%|54.1%|\n|---|---|---|---|\n|Percent of Children 6-59 months receiving<br>Vitamin A Supplements|45%|45%*|53%|\n\n\n\n- No updates will be verified by next DHS survey planned for 2021\nData sources: Ethiopia Demographic Health Survey (EDHS) 2016 report; 2019 Mini-DHS; Nutrition Data Verification Joint Review Mission Report,\nDec 2018; 2012 HMIS Data Quality Assessment (DQA) Report; 2014 HMIS DQA and current data from 2018 SARA report.\n\n\n4. With regard to the three IPF sub-components, the Program status is as follows:\n\n\n**(a)** **_Sub-component I: Support for Civil Registration and Vital Statistics (CRVS)_** : The June 2021 implementation\nsupport mission noted significant progress in undertaking the comprehensive, national assessment and in\ndeveloping the five-year CRVS System Improvement Strategy and Costed Action Plan (2022-2026). Advocacy and\nCRVS awareness raising activities have been conducted at federal, regional, woreda and kebele levels to create\npublic awareness on the importance of CRVS. The printing of blank birth and death certificates and registry books\nfor regions is completed and all the printed registration materials have been delivered to regions. However, the\nprinted materials at the central store have not yet been delivered to Tigray region due to the security situation and\nwill be delivered once the situation improves. Trainings have been provided to the different stakeholders who\nactively work on the CRVS system. However, the national Immigration Nationality and Vital Events Agency (INVEA)\nis experiencing delays in getting timely data and\n\nMention: \"2019 Mini-DHS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The abbreviated round is cited among the sources updating the nutrition indicators between full survey rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:007599:6:0", "occurrence_id": "8eee565d1a40a1fe3f6a8e35", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nHealth Sustainable Development Goals Program-for-Results (P123531)\n\n|Percent of pregnant women taking iron and folic<br>acid (IFA)|42.1%|60%|54.1%|\n|---|---|---|---|\n|Percent of Children 6-59 months receiving<br>Vitamin A Supplements|45%|45%*|53%|\n\n\n\n- No updates will be verified by next DHS survey planned for 2021\nData sources: Ethiopia Demographic Health Survey (EDHS) 2016 report; 2019 Mini-DHS; Nutrition Data Verification Joint Review Mission Report,\nDec 2018; 2012 HMIS Data Quality Assessment (DQA) Report; 2014 HMIS DQA and current data from 2018 SARA report.\n\n\n4. With regard to the three IPF sub-components, the Program status is as follows:\n\n\n**(a)** **_Sub-component I: Support for Civil Registration and Vital Statistics (CRVS)_** : The June 2021 implementation\nsupport mission noted significant progress in undertaking the comprehensive, national assessment and in\ndeveloping the five-year CRVS System Improvement Strategy and Costed Action Plan (2022-2026). Advocacy and\nCRVS awareness raising activities have been conducted at federal, regional, woreda and kebele levels to create\npublic awareness on the importance of CRVS. The printing of blank birth and death certificates and registry books\nfor regions is completed and all the printed registration materials have been delivered to regions. However, the\nprinted materials at the central store have not yet been delivered to Tigray region due to the security situation and\nwill be delivered once the situation improves. Trainings have been provided to the different stakeholders who\nactively work on the CRVS system. However, the national Immigration Nationality and Vital Events Agency (INVEA)\nis experiencing delays in getting timely data and\n\nMention: \"HMIS Data Quality Assessment\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Health management information system assessments are cited as the checks on the reliability of the reported routine data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:007599:6:0", "occurrence_id": "ee3d3ff225b4fbdc5376a404", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nHealth Sustainable Development Goals Program-for-Results (P123531)\n\n|Percent of pregnant women taking iron and folic<br>acid (IFA)|42.1%|60%|54.1%|\n|---|---|---|---|\n|Percent of Children 6-59 months receiving<br>Vitamin A Supplements|45%|45%*|53%|\n\n\n\n- No updates will be verified by next DHS survey planned for 2021\nData sources: Ethiopia Demographic Health Survey (EDHS) 2016 report; 2019 Mini-DHS; Nutrition Data Verification Joint Review Mission Report,\nDec 2018; 2012 HMIS Data Quality Assessment (DQA) Report; 2014 HMIS DQA and current data from 2018 SARA report.\n\n\n4. With regard to the three IPF sub-components, the Program status is as follows:\n\n\n**(a)** **_Sub-component I: Support for Civil Registration and Vital Statistics (CRVS)_** : The June 2021 implementation\nsupport mission noted significant progress in undertaking the comprehensive, national assessment and in\ndeveloping the five-year CRVS System Improvement Strategy and Costed Action Plan (2022-2026). Advocacy and\nCRVS awareness raising activities have been conducted at federal, regional, woreda and kebele levels to create\npublic awareness on the importance of CRVS. The printing of blank birth and death certificates and registry books\nfor regions is completed and all the printed registration materials have been delivered to regions. However, the\nprinted materials at the central store have not yet been delivered to Tigray region due to the security situation and\nwill be delivered once the situation improves. Trainings have been provided to the different stakeholders who\nactively work on the CRVS system. However, the national Immigration Nationality and Vital Events Agency (INVEA)\nis experiencing delays in getting timely data and\n\nMention: \"2014 HMIS DQA\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"A later quality assessment of routine reporting is listed alongside the current data used for the indicators.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:007659:0:1", "occurrence_id": "7c16ee5cf65bf563ff844078", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: , despite the\nmultifaceted implementation challenges over the last two years due to COVID 19 pandemic, flood and locust emergencies, and security issues in\nmany of the project Woredas. The implementation progress was assessed based on an independent MTR survey conducted as part of the MTR\nexercise. The project made good progress in all components and subcomponents. The overall achievements and feedbacks obtained from key\nstakeholders further substantiate the relevance, effectiveness, sustainability, and impact of the project. A total of 10,560 CIGs were established over\nthe last three years, achieving 100 percent of life of project (LOP) target. Out of these CIGs, about 10,333 CIGs with a total membership of 150,381\n(40 percent female) have secured an approved business plan by the respective Woreda Development Committees (WDCs). At present, 8,904 CIGs\nare fully engaged in the management and production of their preferred value chains. These CIGs reported the production of 75.6 million litters of\n\n\n4/7/2022 Page 1 of 10\n\nMention: \"independent MTR survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Implementation progress and component achievement were assessed against this independently conducted mid-term survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:007832:10:33", "occurrence_id": "5c20470297eb509782be914f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: > <br> <br> <br>|\n\n\n\nSources: GOE, IMF, World Bank.\n\n\n**11.** **Higher social spending and a gradual reform of regulated prices are softening the impact of the macroeconomic**\n**adjustment on the poor.** To protect the poor from reform impacts, the government increased safety net benefits by 20\npercent, raised wages for the lowest-paid civil servants by up to 300 percent, and provided temporary fuel and fertilizer\nsubsidies as part of a fiscal package costing 1.5 percent of GDP. <sup>5,6</sup> In addition, it exempted monthly electricity consumption\nof up to 200 kwh from VAT and provided a significant cross-subsidy (up to 84 percent) on consumption of up to 100 kwh. <sup>7</sup>\nHowever, World Bank price monitoring surveys from Addis Ababa indicate some impact on poorer households from\nincreased prices of cooking oil, coffee beans, sugar, medicines and charcoal. Improved land tenure security has expanded\naccess to finance, with households obtaining US$74 million in certificate-backed loans in the first 10 months of the fiscal\nyear, 60 percent higher than the year before.\n\n\n**12.** **A new interest-rate-based monetary framework is starting to improve monetary transmission.** In July 2024, the\n\n\n5 Monthly salaries of about 20 percent of civil servants were below the national poverty line before the increase.\n6 Reform monitoring surveys indicate considerable support for such measures. See World Bank (2025). Economic Reforms in\nEthiopia: Findings from Impact Monitoring During October 2024-May 2025.\n7 Nearly two-thirds of the consumption of bottom 40 percent of households falls under this category.\n\n\nPage 4\n\nMention: \"World Bank price monitoring surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Price monitoring in the capital is used to show poorer households were affected by higher food, medicine and fuel prices.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:007832:10:33", "occurrence_id": "681cd8438d3095b166048c4b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: > <br> <br> <br>|\n\n\n\nSources: GOE, IMF, World Bank.\n\n\n**11.** **Higher social spending and a gradual reform of regulated prices are softening the impact of the macroeconomic**\n**adjustment on the poor.** To protect the poor from reform impacts, the government increased safety net benefits by 20\npercent, raised wages for the lowest-paid civil servants by up to 300 percent, and provided temporary fuel and fertilizer\nsubsidies as part of a fiscal package costing 1.5 percent of GDP. <sup>5,6</sup> In addition, it exempted monthly electricity consumption\nof up to 200 kwh from VAT and provided a significant cross-subsidy (up to 84 percent) on consumption of up to 100 kwh. <sup>7</sup>\nHowever, World Bank price monitoring surveys from Addis Ababa indicate some impact on poorer households from\nincreased prices of cooking oil, coffee beans, sugar, medicines and charcoal. Improved land tenure security has expanded\naccess to finance, with households obtaining US$74 million in certificate-backed loans in the first 10 months of the fiscal\nyear, 60 percent higher than the year before.\n\n\n**12.** **A new interest-rate-based monetary framework is starting to improve monetary transmission.** In July 2024, the\n\n\n5 Monthly salaries of about 20 percent of civil servants were below the national poverty line before the increase.\n6 Reform monitoring surveys indicate considerable support for such measures. See World Bank (2025). Economic Reforms in\nEthiopia: Findings from Impact Monitoring During October 2024-May 2025.\n7 Nearly two-thirds of the consumption of bottom 40 percent of households falls under this category.\n\n\nPage 4\n\nMention: \"Reform monitoring surveys\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"A footnote cites impact monitoring findings as evidence of public support for the reform measures.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:007931:1:0", "occurrence_id": "d720ed2b17e22708cde2b418", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The World Bank\n\n\n\nReport No: ISR12522\n\n\n\nAs of November 22, 2013, a total of US$116,205,971.82 had been disbursed constituting 76.8% of total project funding, which shows significant improvement from the last ISR of\nJune 2013, when disbursement was at 67%.\n\nAchievement of the PDO is rated moderately unsatisfactory due to improvements in data collection measures by Government since 2011, arising from the demand for more accuracy\nand reliability of the Education Management Information System (EMIS) data with specific reference to enrollment statistics where reporting by private schools is about 64%\ncompared to almost 100% of public schools. The improvement in data quality has resulted in lower enrollment numbers than previously projected by the MoES. For the years 2011 &\n2012 the Uganda Bureau of Statistics (UBOS) and Ministry of Education (MoES) agreed that a trend analysis for overall enrollment based indicators would be resumed in 2013 after\nstabilization of EMIS data; and 2013 enrollment statistics would serve as a baseline for future years. As such, the GER appears to have dropped but the Project's baseline and targets\ndata were inflated. Further discussions with Government will be held and a supplementary ISR submitted in mid January 2014. The intermediate indicators have however, been met\nbecause they draw from the public schools data where reporting is almost 100%, validated by the annual head count exercise undertaken in all public schools. Four of the 5 triggers\nhave also been met, with a high likelihood that the fifth trigger will be met by the close of the project.\n\nOverall implementation progress rating is still Moderately Unsatisfactory mainly arising from the late roll out of the school based civil works procurement reform instituted by the\nsector. Persistent challenges include:\n1. Delays in civil works completion. Despite the significant gains made in the completion status for civil works in schools since the last ISR (376 of 659 schools are either ready for\nhandover or at defects liability period of\n\nMention: \"Education Management Information System\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Demand for more accurate enrolment reporting in the system exposed low private-school reporting and corrected inflated enrolment.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:007931:1:0", "occurrence_id": "89609ca7b45f42f2be60579c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The World Bank\n\n\n\nReport No: ISR12522\n\n\n\nAs of November 22, 2013, a total of US$116,205,971.82 had been disbursed constituting 76.8% of total project funding, which shows significant improvement from the last ISR of\nJune 2013, when disbursement was at 67%.\n\nAchievement of the PDO is rated moderately unsatisfactory due to improvements in data collection measures by Government since 2011, arising from the demand for more accuracy\nand reliability of the Education Management Information System (EMIS) data with specific reference to enrollment statistics where reporting by private schools is about 64%\ncompared to almost 100% of public schools. The improvement in data quality has resulted in lower enrollment numbers than previously projected by the MoES. For the years 2011 &\n2012 the Uganda Bureau of Statistics (UBOS) and Ministry of Education (MoES) agreed that a trend analysis for overall enrollment based indicators would be resumed in 2013 after\nstabilization of EMIS data; and 2013 enrollment statistics would serve as a baseline for future years. As such, the GER appears to have dropped but the Project's baseline and targets\ndata were inflated. Further discussions with Government will be held and a supplementary ISR submitted in mid January 2014. The intermediate indicators have however, been met\nbecause they draw from the public schools data where reporting is almost 100%, validated by the annual head count exercise undertaken in all public schools. Four of the 5 triggers\nhave also been met, with a high likelihood that the fifth trigger will be met by the close of the project.\n\nOverall implementation progress rating is still Moderately Unsatisfactory mainly arising from the late roll out of the school based civil works procurement reform instituted by the\nsector. Persistent challenges include:\n1. Delays in civil works completion. Despite the significant gains made in the completion status for civil works in schools since the last ISR (376 of 659 schools are either ready for\nhandover or at defects liability period of\n\nMention: \"EMIS data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Trend analysis was postponed until the system's data stabilised, with the 2013 figures adopted as the new baseline.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:008049:4:1", "occurrence_id": "0109e6cf454273fdbabb7012", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  the program works||Percentage|Value|68.00|73.00|90.00|\n|Beneficiaries who recieved all information<br>needed to understand how the program works||Percentage|Date|31-Dec-2008|11-Jun-2012|30-Jun-2015|\n|Beneficiaries who recieved all information<br>needed to understand how the program works||Percentage|Comments|From IFPRI/CSA 2008<br>Descriptive Report and is a<br>simple average.|Source: 2010 PSNP Impact<br>Evaluation. Will be updated in<br>November 2013||\n|Beneficiaries and non-beneficiaries reporting<br>that the targeting processes are fair||Percentage|Value|85.00|65.00|90.00|\n|Beneficiaries and non-beneficiaries reporting<br>that the targeting processes are fair||Percentage|Date|31-Dec-2008|11-Jun-2012|30-Jun-2015|\n|Beneficiaries and non-beneficiaries reporting<br>that the targeting processes are fair||Percentage|Comments|Baseline data from 2008<br>Financial Transparency and<br>Accountability report on the<br>PSNP.|Source: 2010 PSNP Impact<br>Evaluation. Note that this is a<br>different source than that<br>used for the baseline. Will be<br>updated in November 2013||\n|Woredas with contingency plans in place||Number|Value|0.00|255.00|255.00|\n|Woredas with contingency plans in place||Number|Date|01-Jan-2010|30-Nov-2012|30-Jun-2015|\n|Woredas with contingency\n\nMention: \"2010 PSNP Impact\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Actual values for beneficiary information and perceived fairness of targeting are credited to this impact evaluation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:008049:4:1", "occurrence_id": "74caf604a3c4334deebdff61", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  the program works||Percentage|Value|68.00|73.00|90.00|\n|Beneficiaries who recieved all information<br>needed to understand how the program works||Percentage|Date|31-Dec-2008|11-Jun-2012|30-Jun-2015|\n|Beneficiaries who recieved all information<br>needed to understand how the program works||Percentage|Comments|From IFPRI/CSA 2008<br>Descriptive Report and is a<br>simple average.|Source: 2010 PSNP Impact<br>Evaluation. Will be updated in<br>November 2013||\n|Beneficiaries and non-beneficiaries reporting<br>that the targeting processes are fair||Percentage|Value|85.00|65.00|90.00|\n|Beneficiaries and non-beneficiaries reporting<br>that the targeting processes are fair||Percentage|Date|31-Dec-2008|11-Jun-2012|30-Jun-2015|\n|Beneficiaries and non-beneficiaries reporting<br>that the targeting processes are fair||Percentage|Comments|Baseline data from 2008<br>Financial Transparency and<br>Accountability report on the<br>PSNP.|Source: 2010 PSNP Impact<br>Evaluation. Note that this is a<br>different source than that<br>used for the baseline. Will be<br>updated in November 2013||\n|Woredas with contingency plans in place||Number|Value|0.00|255.00|255.00|\n|Woredas with contingency plans in place||Number|Date|01-Jan-2010|30-Nov-2012|30-Jun-2015|\n|Woredas with contingency\n\nMention: \"Baseline data from 2008\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The baseline percentage comes from an earlier transparency and accountability report, flagged as a different source from the evaluation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:008497:2:0", "occurrence_id": "4fea05d5007323f1aacb5426", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: local level along with normal food deliveries, did not work . Similarly, the school attendance program, meant to use\nfood or cash transfers to help maintain school attendance of girl pupils during the drought, was implemented in only a\nfew districts.\nThe project generated direct employment for over 900,000 individuals (who, according to an end-user survey, earned\nan average of 340 Birr (US$40.00) from working on a sub-project). This was no doubt an important source of\neconomic support, in part because it likely allowed the families to maintain (or reduce destocking of) their livestock.\nHowever, although the project created employment, it did not have a dedicated intervention for livelihood\ndevelopment. Therefore, many of the sub-projects with respect to soil and water conservation (described below) are\nlikely to produce more general environmental benefits and strengthen future drought resilience rather than directly\nenhance livelihood prospects.\n(ii) To put in place infrastructure and support mechanisms to mitigate the medium term effects of current and future\ndrought emergencies on vulnerable groups : rated **Substantial** .\nOn the community assets side, there was a large range and number of physical achievements . Achievements in\nwater development include 327 ponds, 127 micro-irrigation systems, 3 pump irrigation systems, 332 springs\n(development and maintenance), 119 hand dug wells, and 22 roof water harvesting structures. Achievements in soil\nand water harvesting included construction of bunds and terraces, check dams, cut -off drains and waterways as well\nas range management and 13.4 million of tree seedlings planted. In addition, this component financed construction\nand maintenance of 2,940 km of roads, and rehabilitated 368 schools and 88 health posts. Consistent with the\ncommunity-driven nature of these sub-projects, it had not been feasible to define ex -ante specific end-of-project\ntarget values; hence, that yardstick is unavailable to measure project success . As explained above\n\nMention: \"end-user survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Average earnings per worker on sub-projects are reported from a survey of programme end users.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:008546:74:1", "occurrence_id": "c4cb56d366728755829bbbf3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: CEmOC) services|20%|44.6%|24.6%||\n|Health facilities without stock-out of<br>medicines|26%|84%|58||\n\n\n\n<u>Source: Data has been taken from the Annual Health Sector Performance Reports, 2010/11 and 2015/16, Ministry</u>\nof Health, The Republic of Uganda\n\n\n69\n\nMention: \"Annual Health Sector Performance Reports\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Service availability and stock-out percentages in the results table are taken from two annual sector performance rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:008658:51:0", "occurrence_id": "fa6bbe320c80d543b47ec87a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **ANNEX 7.** **SUMMARY OF BORROWER'S ICR AND/OR COMMENTS ON DRAFT ICR**\n\n\n**Borrower’s ICR for the IDA Credit: Learning and Innovation Lending (LIL)**\n\nThe Conservation and Sustainable Use of Medicinal Plants Project (CSMPP) was developed and\nimplemented by the IBC in collaboration with stakeholders. The specific objectives of the project’s credit\ncomponent were to support the establishment of FGBs for medicinal plants, strengthen the capacity of\ntraditional healers, develop an IPR policy guideline, develop five phytomedicines for human and\nlivestock diseases, establish a national medicinal plant database, develop techniques for propagating\nmedicinal plants, and carry out a national economic survey on the use of medicinal plants. The IBC was\ndesignated in a government-signed MOU as the project’s lead implementing agency, to work with 12\ncollaborating institutions. IBC established a Project Monitoring and Coordination Unit (PCMU) to\nfacilitate implementation.\n\nThe project’s credit component was financed by IDA with a matching fund allotted by the Government of\nEthiopia. The project was originally to last four years, but a six-month extension was granted to complete\na validation study for one phytomedicine. The original allocation of IDA was US$ 2.5 million, but about\nUS$ 600,000 dollars were cancelled.\n\nBased on the its objectives and key performance indicators (KPIs), the project’s achievements can be\nsummarized as follows:\n\n  - An M&E system manual, reviewed and approved by IDA, was developed and used.\n\n  - A national consultant conducted an institutional and human resource capacity assessment to\nidentify training needs and other inputs. Based on the results, training and procurement plans\nwere prepared and implemented.\n\n  - THAs conducted a number of workshops to raise members’ awareness of the importance of\nconserving medicinal plants, validating traditional medicines, following a code of ethics and\npractices, and other\n\nMention: \"national medicinal plant database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Establishing the database is listed among the credit objectives, alongside propagation techniques and a use survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:008800:4:1", "occurrence_id": "2527138f98bceb2d6b7b5071", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  from entrenched and significant unequal\naccess to education. However, a focus on disadvantaged and vulnerable groups may mean that the\ninequalities between individuals within groups are overlooked. **Vertical inequalities within groups**\n(e.g. groups of adolescent girls) are often larger than **horizontal inequalities** between groups (e.g.\nrural-urban populations). Existing data collection and monitoring systems have to be expanded to\nconsider fully the range of vertical differences within the vulnerable and disadvantaged groups. This\nwill be needed in future, to support better targeting of limited resources and programmes which can\nfully address the diverse range of opportunities and barriers, to quality education, across vulnerable\nand disadvantaged groups and populations.\n\n\n**<u>Enabling Background for GEQIP Phase 2</u>**\nConstitutional rights, policies and programmes in Ethiopia strongly support and promote the equitable\ninclusion of regions and vulnerable groups lagging behind in access to quality education. The GoE\nhas moved on a range of fronts (laws, policies and programmes) to address both the supply and\ndemand side constraints affecting enrolment, repetition and drop-out.\n\n\nGEQIP Phase I included an explicit focus on gender. Attention was paid to the selection of female\nteachers for training courses and all data was disaggregated. EMIS data shows the following national\nachievements with respect to female participation in primary and secondary education since GEQIP\nPhase 1 was launched in 2009: the proportion of female teachers has increased from 33% to 34%;\ngirls primary completion rate has increased from 45% to 52% at grade 8; girls’ drop out has decreased\nfrom 19% to 15%and 17,000 more girls passed the grade 10 examination\n\nGEQIP Phase I included support to Alternative Basic Education Centres which are designed as low\ncost alternatives to formal schools and reach vulnerable and disadvantaged children, especially in the\nemerging regions. ABECs use locally recruited facilitators to deliver the curriculum in a flexible way.\n\nMention: \"EMIS data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"System data provide the female teacher share and girls' completion trends used to judge the gender focus of the first phase.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:008875:1:1", "occurrence_id": "b6ca2761929ef744faf09b8c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ** The State Department for Labour (SDL) has made progress in completing an informal sector survey, updating the Kenyan\nNational Occupational Classification (KNOCS), and setting up a Labour Market Observatory (LMO) that is implementing a roadmap for the Kenya\nLabour Market Information System (KLMIS).\n\n\n**Component 4** . National Youth Policy: The National Youth Policy was endorsed by the Cabinet in November 2019 and is now poised to be\ndisseminated to key stakeholders before moving to the Kenya National Assembly for legislation **.**\n\n\n**Risks**\n\n\n**Systematic Operations Risk-rating Tool**\n\n\nRisk Category Rating at Approval Previous Rating Current Rating\n\n\nPolitical and Governance Substantial Substantial Substantial\n\n\nMacroeconomic Moderate Substantial Substantial\n\n\nSector Strategies and Policies Low Low Low\n\n\nTechnical Design of Project or Program Substantial Substantial Substantial\n\nInstitutional Capacity for Implementation and\nHigh High High\n<u>Sustainability</u>\n\nFiduciary High High High\n\n\nEnvironment and Social Low Moderate Moderate\n\n\nStakeholders Moderate Moderate Moderate\n\n\nOther -- -- -\n\nOverall Substantial Substantial Substantial\n\n\n**Results**\n\n\n**PDO Indicators by Objectives / Outcomes**\n\n\n6/12/2021 Page 2 of 8\n\nMention: \"informal sector survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Completing the survey is reported as progress alongside an occupational classification update and a labour market observatory.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:009082:65:0", "occurrence_id": "32acaa21a0c4a15d10bba8d1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: - Ministry of Finance and Economic development and the Road Fund Office, through\n\ngreater efficiency in the use of resources and reduced demands placed on the national\nbudget for road development and maintenance.\n\n**7** **Impact of Intervention on Mobility**\n\nAn assessment of traffic along the main roads has revealed a rapid increase in volume of traffic.\nThe rate of traffic growth is on average about 9 percent per annum, even where improvement\nworks have not yet been completed under the RSDP. The traffic growth in all rehabilitated and\nupgraded projects under APL I has significantly increased. The following table shows the\ncomparison of volume of traffic before and after the intervention of the road projects under APL\nI program.\n\n### **APL 1 Roads Traffic Growth Rate**\n\n\n\n\n\n\n\n\n\n|No.|Direction|Year|Col4|Cars|Buses|Trucks|Truck<br>&<br>Trailer|Total|\n|---|---|---|---|---|---|---|---|---|\n|**1 **|**Nazareth - Assela**|**2003**|**AADT**|**273**|**408**|**635**|**66**|**1382**|\n|**1 **|**Nazareth - Assela**|**2008**|**AADT**|**651**|**783**|**878**|**262**|**2574**|\n|**1 **|**Nazareth - Assela**|**2008**|**% Growth**|138|92|38|297|86\n\nMention: \"APL 1 Roads Traffic Growth Rate\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Traffic counts before and after each road intervention are tabulated by vehicle class to show growth on the upgraded corridors.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:009114:6:0", "occurrence_id": "139d1ac5a54fccc7275f5902", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Independent Evaluation Group (IEG)** Implementation Completion Report (ICR) Review\nKE-Municipal Program (FY10) (P066488)\n\n\nto reduced repairs and business disruption. People interviewed in an audit of environmental and social\nsafeguards compliance, confirmed that the April May 2017 rains did not flood the usual areas.\n\n   - 11 of the 14 integrated strategic urban development plans have been approved by county executive\ncommittees while the rest were awaiting approval. Three plans were approved by county assemblies.\nCounty governments, like Mombasa, were using the plans to inform future infrastructure projects, based\non evidence obtained from the Beneficiary Assessments. Under the 2012 County Government Act,\ncounties prepare 5-year county development plans. The project trained staff of nearly all 47 counties\nbased on the Handbook and built capacity of county staff who generated the integrated urban\ndevelopment plans.\n\nThe following outcome was partially achieved:\n\n\n   - 69,200 urban residents (target 170,000) gained access to all season roads within a 500-meter range\nunder the project. The government dropped some contracts for roads and drainage following the 2015\nRestructuring primarily because the Mombassa drainage cost overruns reduced available resources(see\nSection 5. Efficiency).\n\nThe following indicators were dropped as a result of the 2015 Restructuring:\n\n\n   - at the national level - strengthening of intergovernmental fiscal frameworks to enhance volume and\npredictability of resources transferred to municipalities based on their performance;\n\n   - implementing a financial management information systems at the urban level;\n\n   - implementing the annual performance assessment of participating municipalities;\n\n   - developing revenue enhancement plans for each of the participating municipalities;\n\n   - supporting the annual audits of municipal financial statements; and\n\n   - strengthened accountability of municipalities.\n\n\n**Revised Rating**\nSubstantial\n\n\nPHEFFICACYTBL\n# **Objective 2**\n\n**Objective**\n\nTo improve service delivery to selected municipalities.\n\n\n**Rationale**\n**OUTPUTS** :\nThe following outputs were substantially achieved:\n\n\n   - 15 km of\n\nMention: \"Beneficiary Assessments\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"County governments used evidence from these assessments to inform future infrastructure investment plans.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:009442:38:3", "occurrence_id": "c4fbe9125cf11e4b73f525c5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |Jul|Au|\n|2002<br>|9.0|9.6|10.2|9.2|9.7|8.7|7.2|7.5|7.9|5.0|3.5|2.8|\n|2003|6.8|9.3|9.7|9.8|9.1|10.7|8.1|9.9|8.5|6.5|5.6|4.4|\n\n\n\n_Source: Bureau of Plan & Finance Tigray Region, Statistical Bulletin 2002 and 2003 EFY_\n\n###### **5.4. Biological Environment**\n\n\n**5.4.1.** **Vegetation and Flora**\n\n\nThe cumulative and intensified human and animal intervention that has been taking place on the\nproject area and its vicinity resulted on the existing highly degraded land. The major human\ninterventions in the project site are clearing of trees for firewood, borrow pits and settlement activities.\n\n\n[26]\n\nMention: \"Statistical Bulletin 2002 and 2003\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The monthly climate series in the table is credited to two regional statistical bulletins.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:009973:2:0", "occurrence_id": "b6ef9f9097238b82d88d8b58", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The World Bank\n\n\n\nReport No: ISR13513\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n|Gross enrollment rate S1-S4|Col2|Percentage|Value|27.00|37.00|41.00|\n|---|---|---|---|---|---|---|\n|Gross enrollment rate S1-S4||Percentage|Date|27-Nov-2008|31-Oct-2013|31-Jul-2014|\n|Gross enrollment rate S1-S4||Percentage|Comments||Results of the re-analysis of<br>the EMIS data for the years<br>revealed that past<br>assessments since 2011 were<br>based on two EMIS regimes.<br>The change in reporting<br>norms of the EMIS led to a<br>declining in enrollment growth<br>affecting the project targets.<br>Variations in baseline values<br>also affected this. The<br>Restructuring Paper for<br>August 2013 put the 2008<br>baseline at 32.95%, against<br>an EMIS target valueof<br>31.1%. Application of<br>appropriate measures to<br>remedy the changes in<br>reporting system of 2011 is<br>thus not possible, and so is<br>restructuring to change<br>targets.||\n|Boys||Percentage<br>Sub Type<br> Breakdown|Value|28.00|38.90|42.00|\n|Boys||Percentage<br>Sub Type<br> Breakdown|Date|27-Nov-2008|31-Oct-2013|31-Jul-2014|\n|Boys||Percentage<br>Sub Type\n\nMention: \"EMIS data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Re-analysis showed enrolment since 2011 mixed two reporting regimes, so the apparent decline was a reporting artefact affecting targets.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:010100:19:0", "occurrence_id": "93a5b1f45089dd0ace160b77", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 32. Adoption of improved fruit trees increased from 5.3 percent to 7.1 percent for target farmers.\nFertilizer use on maize increased from 18.0 percent to 21.0 percent for target farmers indicating a 4.3\npercentage point increase. Adoption of animal feed preservation technologies increased for target farmers\nfrom 56.7 percent to 62.2 percent, while the control farmers had a modest increase from 43.5 percent to\n44.7 percent. Adoption of value addition technology, both for farm produce and livestock products were\nestimated at 8.6 and 6.7 percentage points, respectively. Soil, water and environmental conservation\ntechnologies were adopted in the Central Highlands and the Western Highlands with impact estimates of\n6.4 and 14.6 percent, respectively. Table 3.2 summarizes the above mentioned results.\n\n**<u>Table 3. 2: Summary of Adoption of Various Technologies</u>**\n\n\n\n\n\n\n\n|Technology Adopted|Before Project<br>(KAPP I Baseline<br>Survey)%|Col3|After Project<br>(Technology<br>Adoption Survey)%|Col5|Double<br>Difference|\n|---|---|---|---|---|---|\n|** Technology Adopted**|**Target**|**Control**|**Target**|**Control**|**Control**|\n|Growing improved fruits|5.3|3.9|7.1|4.0|1.7|\n|Use of fertilizer on maize|18.0|18.1|21.0|16.0|4.3|\n|Animal feed preservation|56.7|43.5|62.2|44.7|4.3|\n|Ownership of grade cows|5.0|7\n\nMention: \"KAPP I Baseline\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Before-project adoption shares for each technology come from the baseline survey, for both target and control areas.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:010100:19:0", "occurrence_id": "34977e0a5b7c21fec1f5578c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 32. Adoption of improved fruit trees increased from 5.3 percent to 7.1 percent for target farmers.\nFertilizer use on maize increased from 18.0 percent to 21.0 percent for target farmers indicating a 4.3\npercentage point increase. Adoption of animal feed preservation technologies increased for target farmers\nfrom 56.7 percent to 62.2 percent, while the control farmers had a modest increase from 43.5 percent to\n44.7 percent. Adoption of value addition technology, both for farm produce and livestock products were\nestimated at 8.6 and 6.7 percentage points, respectively. Soil, water and environmental conservation\ntechnologies were adopted in the Central Highlands and the Western Highlands with impact estimates of\n6.4 and 14.6 percent, respectively. Table 3.2 summarizes the above mentioned results.\n\n**<u>Table 3. 2: Summary of Adoption of Various Technologies</u>**\n\n\n\n\n\n\n\n|Technology Adopted|Before Project<br>(KAPP I Baseline<br>Survey)%|Col3|After Project<br>(Technology<br>Adoption Survey)%|Col5|Double<br>Difference|\n|---|---|---|---|---|---|\n|** Technology Adopted**|**Target**|**Control**|**Target**|**Control**|**Control**|\n|Growing improved fruits|5.3|3.9|7.1|4.0|1.7|\n|Use of fertilizer on maize|18.0|18.1|21.0|16.0|4.3|\n|Animal feed preservation|56.7|43.5|62.2|44.7|4.3|\n|Ownership of grade cows|5.0|7\n\nMention: \"Adoption Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"After-project adoption shares in target and control areas come from the technology adoption survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:010190:1:1", "occurrence_id": "38d875647f70bcf9dd8b4b36", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  has actually fallen since 2006. Overall, most of the new jobs created in Kenya\nin recent years are in the informal sector, also known as the Jua Kali, which consists of the selfemployed, unpaid family workers, and those working for wages in small household enterprises. The\nfailure of the formal sector to generate sufficient wage employment to accommodate all new\nentrants to the labor force has led to many youths starting their own businesses.\n\n5. Poor employment and weak productivity growth in services exacerbate some of the employment\nand productivity constraints that already disproportionately affect youth in Kenya. Youth tend to\nlack relevant work experience and even if they are educated, come out of the education system\nlacking the cognitive, technical, socioemotional, and business skills needed to be productive in\ntoday’s labor markets. This is important both for self-employment and for wage employment. In\naddition, the lack of market information about what skills are in demand and where jobs can be\nfound, perpetuate these mismatches. Beyond experience and skills, youth also often lack the social,\nfinancial, and physical capital necessary to get a job or to successfully start or expand a business.\nWomen or disadvantaged youth—those living in conflict-affected areas or coming from poor\nfamilies, for example—face additional constraints related to family and household responsibilities\nor violence, respectively.\n\n6. As a result of all of these forces, youth in Kenya—especially the least educated—have\npoor employment prospects. Youth unemployment rates measured by the 2009 Kenya Population\nand Housing Census were highest for younger members of the youth cohort and those in urban\nareas (table 1). The rate of unemployment in 2009 for all Kenyans between 15 and 64 years of age\nwas 8.6 percent. For those aged 15 to 19 years, it was 15.8 percent, and for those aged 20 to 24\nyears, it was 13.1 percent, whereas unemployment rates for those over 25 years of age approached\nthat of the adult population. Unemployment\n\nMention: \"2009 Kenya Population\nand Housing Census\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Census-derived unemployment rates by age and urban residence support the argument that least-educated youth face the worst prospects.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:011200:13:0", "occurrence_id": "74179955b1685d930d14f3e3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nKenya Inclusive Growth and Fiscal Management DPO 2 (P172321)\n\n\n**Economic Impact of COVID-19**\n\n**2.2** **The COVID-19 global pandemic will have a large negative impact on the Kenyan economy.** The\nCOVID-19 shock is estimated to further reduce growth in 2020 to 1.5 percent with a twin shock both from\nthe external and domestic sources. On the external side, the decline in demand following implementation\nof measures to delay the spread of COVID-19 will hit hard travel, tourism and agricultural exports. Access\nto international markets is virtually shut to frontier issuers such as Kenya, adding vulnerability to its\nexternal payment position. Major stock markets, globally and the Nairobi Securities Exchange, are now in\na bear market, consistent with pricing in a global recession ( **Error! Reference source not found.** ). Dollar\nfinancing costs for Kenya and other emerging markets have risen dramatically ( **Error! Reference source**\n**not found.** ). On the domestic front, the top priority is to implement effective containment and mitigation\nmeasures to contain the spread of corona virus and to save lives. Measures being taken to slow down the\nrate of infection, including home confinement, travel restrictions, closure of schools and entertainment\nspots, suspension of public gathering and conferences, and a nightly curfew, are affecting both production\nand consumption. This revised growth estimate is subject to much uncertainty and potential further\nrevision, depending on how widespread COVID-19 becomes in the country and the intensity and duration\nof social distancing measures that are contributing to a slowdown in economic activity.\n\n\n\n**Figure 1: Stocks pricing in global recession**\n_(Nairobi-20 and DWJ Global, indexed: 27 Mar_\n_2019 = 100)_\n\n\n\n**Figure 2: Higher external financing costs**\n_(EMBIG index of sovereign\n\nMention: \"Nairobi-20 and DWJ Global\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The two equity indices plotted in the figure are credited to these market series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:011366:1:1", "occurrence_id": "c54073c00e139e45c1f5eae2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 37.00<br>Date<br>14-Dec-2010<br>Comment|Value<br>8400.00<br>Date<br>30-Jun-2013<br>Comment<br>POI#2|\n|Indicator|Baseline|Current|End Target|\n|Indicator Name<br>Proportion of sexually active youth 15- 24 who report having had sex<br>with a non-spousal, non-regular partner in the past 12 months.<br>Type<br>Unit of Measure<br> Custom Indicator<br> Text|Value<br>F=30%<br>M=84% (DHS 2003)<br>Date<br>01-Jan-2007<br>Comment|Value<br>F=33%<br>M=83% (DHS 2008)<br>Date<br>14-Dec-2010<br>Comment|Value<br>F=20%<br>M=70%<br>Date<br>30-Jun-2013<br>Comment<br>POI#3<br>Targets based on about 10%<br>reduction every three years and<br>apply if a KAIS or DHS will be done<br>in 2012 or 2013|\n\n\nPage 2 of 8\n\nMention: \"DHS 2003\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The sex-disaggregated baseline value for the youth risk-behaviour indicator is taken from this survey round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:011756:22:1", "occurrence_id": "2165094669a95c22a688ff2c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  levels of disease and destitution characterize the proposed project area. Studies conducted in the area\nreveal linkages between:\n\n     - Poverty and land investments. Poor households invest very little in agricultural inputs;\n\n     - Agriculture, ill-health and poverty;\n\n     - Wetland drainage for irrigation agriculture and land degradation; and\n\n     - Low soil fertility and lack of technical advice.\n\n\nPoverty in Kenya remains a challenge requiring urgent attention. It is a multi-dimensional phenomenon,\nwhich includes inadequacy of income and deprivation of basic needs and rights, and access to productive\nassets as well as to social infrastructure and markets (PRSP, 2003). The quantitative approach of measuring\npoverty defines the poor as those who cannot afford basic food and non-food items. The 1997 Welfare\nMonitoring Survey estimated the absolute poverty line at KShs.1,239 per person per month and KShs.2,648\nrespectively for rural and urban areas. This poverty line is determined and based on the expenditure\nrequired to purchase a food basket that allows minimum nutritional requirements to be met. The minimum\nnutritional requirements are set at 2,250 calories per adult equivalent per day. Based on Welfare Monitoring\nSurvey (WMS III) of 1997, the proportion of the population that cannot purchase the basic basket of goods\nwas estimated to be 53 per cent of the rural and 50 per cent of the urban population.\n\n\nUsing the qualitative approach of Participatory Poverty Assessments (PPAs), people define, view and\nexperience poverty in different ways. In the 2001 PPA reports, people mainly defined poverty as the\ninability to meet their basic needs. Poverty was associated with features such as lack of land,\nunemployment, inability to feed oneself and one’s family, lack of proper housing, poor health and inability\nto educate children and pay medical bills.\n\n\nIn Kenya the poor tend to be clustered into certain social categories such as:\n\n\n   - the landless;\n\n\n   - people with disabilities;\n\n\n   - female headed households;\n\n\n   - households headed by people without formal education;\n\nMention: \"1997 Welfare\nMonitoring Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The survey supplies the absolute poverty line, separately for rural and urban areas, on which the passage's poverty definition rests.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:011756:22:1", "occurrence_id": "f7493502493f57f9d82ec12f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  levels of disease and destitution characterize the proposed project area. Studies conducted in the area\nreveal linkages between:\n\n     - Poverty and land investments. Poor households invest very little in agricultural inputs;\n\n     - Agriculture, ill-health and poverty;\n\n     - Wetland drainage for irrigation agriculture and land degradation; and\n\n     - Low soil fertility and lack of technical advice.\n\n\nPoverty in Kenya remains a challenge requiring urgent attention. It is a multi-dimensional phenomenon,\nwhich includes inadequacy of income and deprivation of basic needs and rights, and access to productive\nassets as well as to social infrastructure and markets (PRSP, 2003). The quantitative approach of measuring\npoverty defines the poor as those who cannot afford basic food and non-food items. The 1997 Welfare\nMonitoring Survey estimated the absolute poverty line at KShs.1,239 per person per month and KShs.2,648\nrespectively for rural and urban areas. This poverty line is determined and based on the expenditure\nrequired to purchase a food basket that allows minimum nutritional requirements to be met. The minimum\nnutritional requirements are set at 2,250 calories per adult equivalent per day. Based on Welfare Monitoring\nSurvey (WMS III) of 1997, the proportion of the population that cannot purchase the basic basket of goods\nwas estimated to be 53 per cent of the rural and 50 per cent of the urban population.\n\n\nUsing the qualitative approach of Participatory Poverty Assessments (PPAs), people define, view and\nexperience poverty in different ways. In the 2001 PPA reports, people mainly defined poverty as the\ninability to meet their basic needs. Poverty was associated with features such as lack of land,\nunemployment, inability to feed oneself and one’s family, lack of proper housing, poor health and inability\nto educate children and pay medical bills.\n\n\nIn Kenya the poor tend to be clustered into certain social categories such as:\n\n\n   - the landless;\n\n\n   - people with disabilities;\n\n\n   - female headed households;\n\n\n   - households headed by people without formal education;\n\nMention: \"Welfare Monitoring\nSurvey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The same round gives the rural and urban shares unable to afford the basic consumption basket.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:012481:11:0", "occurrence_id": "dd3c6ef1dd771c7dffdd6340", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nUganda Secondary Education Expansion Project (P166570)\n\n\n**Figure 7: Enrolment: Government vs Private**\n\n\n\n\n\n\n\n_Source: WB based on data from ESSAPR FY16/17_\n\n\n16. **Some of the private schools were partially financed by government funding under the now**\n**defunct Private-Public-Partnership (PPP) policy.** **The objective of the PPP arrangement was to spur**\n**enrollment at the secondary level by increasing access to private schools by making them more**\n**affordable** . Under the PPP policy, the government transfers a subsidy of 47,000 UGX per student per term\n(US$36.5 per year) to selected private schools which comply with some principles that keep them\naffordable and of reasonable quality. The PPP started with 363 schools in 2007 enrolling approximately\n40,000 students and grew to over 800 schools, enrolling nearly half a million students or one-third of all\nstudents in secondary schools in Uganda. <sup>23</sup> According to a study done by ARK Education (supported by\nDFID), there is little difference in academic performance between Government schools and PPP schools,\nwith PPP schools providing slightly better quality education at a lower cost, as Government schools charge\nsimilar fees to PPP schools <sup>24</sup> in addition to the funding they receive from the national education budget.\nBetter scores are obtained by students in PPP schools than in government <sup>25</sup>, despite that teachers in PPP\nschools are paid around 25 percent less and have less experience and qualifications.\n\n\n17. **The relative success of the PPP arrangement was accompanied by several challenges in its**\n**design and implementation which eventually undermined its effectiveness** . The main issue pertained to\nlack of accountability to government and poor learning outcomes in some private schools. In addition\n\nMention: \"data from ESSAPR\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The comparison of government and private enrolment is drawn from this annual education performance reporting round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:012713:35:3", "occurrence_id": "5779c153b08502f05ec0900f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  subregions by<br>end of project|N/A|<br>500|<br>1,309*|\n\n\n\n15 Deduced from subproject attrition trends (source: NUMU database)\n\n\n16 This is a proportion of the estimated number of CRCM beneficiaries.\n\n\n17 Listenership Survey (2008)\n\n\n27\n\nMention: \"Listenership Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"A footnote credits this survey as the basis for estimating the beneficiary population behind a target.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:013394:56:0", "occurrence_id": "e1d81ee988adba9a85fcca79", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 15. The rationale for providing free secondary access is to **narrow the gap between the**\n**girls and boys rates** of participation. Component 1 of the project was meant to improve equity\nin access to free lower secondary education. Between 2010 and 2014, there has been noticeable\njump in both girls and boys enrolment by 52%, but a slight increase (by 2%) in the share of girls,\nas boys’ enrolments slightly reduced by the same margin between two periods (Figure 3). There\nis sizeable gender gap from senior one through senior four. Wide gaps are also noted in the year\n2010 especially in senior three and four across all the grades; while narrow gaps are evident in\n2014 where about 71% of boys attain senior four compared to 65% of girls. Many of the\nstudents enrolled would likely have enrolled even without expansion of school infrastructure, so\nthe impact of the project would depend on how many additional students enrol and stay in\nschool due to the improved learning environment.\n\n**Figure 3: School participation by grade and gender in Uganda, 2010 & 2014**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nSource: EMIS - Annual School Head Count Exercise that informs capitation grants\n\n**Quality, Efficiency and Effectiveness**\n\n16. **Internal Efficiency:** Improving internal efficiency of lower secondary sub-sector can\nlead to quality improvements and cost savings. Reduction in grade repetition reduces the time\nstudents take to complete the cycle and also reduces on the financial burden on part of\ngovernment and parents. Under this project, efficiency of education could be achieved through\neffective utilization of both school infrastructure and instruction resources under management of\nretooled head teachers. High completion rates trigger the improvements in transition rates.\nHowever, internal efficiency indicators did not change significantly due to data issues, and are\nunlikely to have led to significant cost savings; as such changes usually take long time to happen.\nThere is no project data on survival and dropout rate but Table 3 indicates that by\n\nMention: \"Annual School Head Count Exercise\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Participation by grade and gender is plotted from the exercise that also determines capitation grants.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:013446:29:0", "occurrence_id": "7e7d7f56fa315444f3022a40", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: productivity and competitiveness through the matching grants scheme and by supporting\nstartups through business plan competitions. The Project interventions supported\ntechnology and skills transfer under the matching grant schemes, enhanced the available\nskill supply for the private sector through support to the business school and provided\nsupport for market development, technology transfer, amendments of competition policy\nand WTO accession.\n\n79. Table 2 below shows the linkage between the results achieved and the project\ndevelopment objective. For ease of analysis, the PDO has been split into three intended\noutcomes: (i) increasing participation of the private sector, (ii) improving productivity,\nand (iii) export competitiveness.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n|Table 2  Intended outcome|Intervention|Result achieved|\n|---|---|---|\n|<br>**Intended outcome**<br>|**Intervention**<br>|**Result achieved**<br>|\n|<br>Increasing<br>participation of the<br>private sector<br>|<br>- Privatization of<br>SOEs<br>- Matching Grants<br>Scheme<br>- BIZPLAN<br>|<br>- 44 SOEs were transferred to private<br>ownership and 77percent of SOE total<br>asset value SOE identified in the PAD<br>have gone through divestiture by the end<br>of the project<br>- Volume of exports by beneficiary firms<br>has increased by 29percent, with estimated<br>increase in revenue of US$ 35 million31.<br>- 5,719 new jobs were created by MGS<br>beneficiary firms (of which 60.4percent<br>were women)<br>- PPESA data shows that though redundant<br>employees were laid off during<br>privatizations of SOEs, the privatized<br>enterprises\n\nMention: \"PPESA data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Project records are used to show that privatisation shed redundant staff yet raised output and exports among surviving firms.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:013897:31:1", "occurrence_id": "4446fc939c005863ffc10835", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: (number)|||Comment|Source:<br>Administrative<br>Reports of<br>HSSF.<br>Estimated<br>population in<br>catchment<br>areas of<br>facilities<br>(levels 2 & 3)<br>benefiting<br>from HSSF.|Source:<br>Administrative<br>reports of HSSF.<br>Based on the<br>catchment<br>population<br>covered by the<br>facilities<br>receiving HSSF.|Date for end target<br>changed|\n|Marked for<br>Deletion|Facilities receiving HSSF<br>Funds within 15 days of the<br>beginning of the quarter<br>(percent).||Percentage|Value|0.00|0.00|90.00|\n|Marked for<br>Deletion|Facilities receiving HSSF<br>Funds within 15 days of the<br>beginning of the quarter<br>(percent).|||Date|05-Jul-2010|12-Feb-2013|31-Mar-2015|\n|Marked for<br>Deletion|Facilities receiving HSSF<br>Funds within 15 days of the<br>beginning of the quarter<br>(percent).|||Comment|Source: HMIS<br>Reports of|Source: HSSF<br>Secretariat|No longer relevant<br>as this will be|\n\n\n\n23\n\nMention: \"HMIS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Facility reporting from the health management information system is cited as the source for a service timeliness indicator.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:014663:19:0", "occurrence_id": "f25f26522f0ed50640904a6c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nNairobi Metropolitan Services Improvement Project (P107314)\n\n\ninitiated the implementation of one of these plans, the Nairobi Railway City Redevelopment Plan (425 acres), as\nthe Nairobi Railway Station is the main railway hub of the whole country and, thus, a site of national importance.\nThe SGR Embakasi Development Plan was prepared to cater for the emerging importance of this station as a\ntransport hub, something that was not foreseen in the 2013 Nairobi Integrated Urban Development Plan\n(NIUPLAN). The Eastlands Urban Renewal plan provides strategic guidance in delivering affordable housing and\nrestoring old and dilapidated neighborhoods and fits into the Government's Affordable Housing Program. None\nof the plans have been implemented but their preparation was considered a major step in supporting sustainable\ndevelopment of the metro region particularly in Kajiado, Kiambu, and Machakos Counties which developing\nrapidly with no planning due to proximity to Nairobi.\n\n“ _The project promoted integrated planning and thinking by bringing together a cocktail of experts_\n_including among others; planners, environmentalists, engineers, security experts, and economists”-_\n_County Planner, Nairobi_ .\n\n  - **Physical addressing systems.** The physical addressing systems have both been adopted by the respective counties\nand once implemented, will increase the ease of navigation around the city and other urban areas, positively\ncontributing to ecommerce (in which the youth are heavily engaged), and swifter response time and access for\nemergency services. The systems will also improve the ability of counties to properly identify properties and\nconsequently accurately charge property rates, thus improving county revenue.\n\n  - **Valuation roll.** A valuation roll was prepared for the Nairobi County which is the first updated roll since 1980\nwhen the last valuation roll for the city was prepared. Given the extensive urban development in Nairobi over\nthe last 20 years, and the related increases in land value, the updating of the roll was long overdue and the county\nhas been\n\nMention: \"Valuation roll\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"A newly prepared property valuation roll lets the county identify properties and charge rates accurately.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:014663:19:0", "occurrence_id": "6d49dd1d668b7ca137ccc6eb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nNairobi Metropolitan Services Improvement Project (P107314)\n\n\ninitiated the implementation of one of these plans, the Nairobi Railway City Redevelopment Plan (425 acres), as\nthe Nairobi Railway Station is the main railway hub of the whole country and, thus, a site of national importance.\nThe SGR Embakasi Development Plan was prepared to cater for the emerging importance of this station as a\ntransport hub, something that was not foreseen in the 2013 Nairobi Integrated Urban Development Plan\n(NIUPLAN). The Eastlands Urban Renewal plan provides strategic guidance in delivering affordable housing and\nrestoring old and dilapidated neighborhoods and fits into the Government's Affordable Housing Program. None\nof the plans have been implemented but their preparation was considered a major step in supporting sustainable\ndevelopment of the metro region particularly in Kajiado, Kiambu, and Machakos Counties which developing\nrapidly with no planning due to proximity to Nairobi.\n\n“ _The project promoted integrated planning and thinking by bringing together a cocktail of experts_\n_including among others; planners, environmentalists, engineers, security experts, and economists”-_\n_County Planner, Nairobi_ .\n\n  - **Physical addressing systems.** The physical addressing systems have both been adopted by the respective counties\nand once implemented, will increase the ease of navigation around the city and other urban areas, positively\ncontributing to ecommerce (in which the youth are heavily engaged), and swifter response time and access for\nemergency services. The systems will also improve the ability of counties to properly identify properties and\nconsequently accurately charge property rates, thus improving county revenue.\n\n  - **Valuation roll.** A valuation roll was prepared for the Nairobi County which is the first updated roll since 1980\nwhen the last valuation roll for the city was prepared. Given the extensive urban development in Nairobi over\nthe last 20 years, and the related increases in land value, the updating of the roll was long overdue and the county\nhas been\n\nMention: \"valuation roll\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"The passage stresses that no roll had been updated since 1980 despite two decades of urban growth.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:014768:15:0", "occurrence_id": "8a5a6368a12dfa9bc360b8eb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank** Implementation Status & Results Report\nEthiopia Health MDG Support Operation (P123531)\n\n\nBaseline Actual (Previous) Actual (Current) N/A\n\n\nValue 27.00 33.00 33.00 51.00\n\n\nDate -- 24-Apr-2019 24-Apr-2019 -\n\nData source: Nutrition Data Verification Joint Review Mission, December 2018.\n**Comments**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n6/28/2019 Page 16 of 17\n\nMention: \"Nutrition Data Verification Joint Review Mission\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"The current nutrition indicator value is attributed to a joint verification mission rather than to routine reporting.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:015290:4:1", "occurrence_id": "203d03427b22a39158f88b40", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  in data\ncollection. This indicator will be assessed more carefully through operational research and examined during midline data collection. For now, only\nthe baseline data has been updated.\n\n[1] Measure DHS and Uganda Bureau of Statistics. 2012. Uganda Demographic and Health Survey 2011. Calverton, Maryland USA and\nKampala, Uganda.\n\n\n**Intermediate Results Indicators**\n\n\nPHINDIRITBL\n\n\nChildren 0-24 months participating in monthly growth monitoring and promotion sessions (Number, Custom)\n\n\nBaseline Actual (Previous) Actual (Current) End Target\n\n\nValue 0.00 0.00 7500.00 8000.00\n\n\nDate 25-Sep-2013 15-Sep-2014 25-Nov-2015 09-Jun-2017\n\n\n12/9/2015 Page 5 of 9\n\nPublic Disclosure Copy\n\nMention: \"Uganda Demographic and Health Survey 2011\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The updated baseline for the growth-monitoring indicator is taken from this survey round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:015306:93:0", "occurrence_id": "c44314f3bc68b98ed92a06e1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |Col1|Col2|Col3|Commiphora woodland|Col5|\n|---|---|---|---|---|\n|Tama|3269|Southern<br>Ethiopia|_Acacia-Commiphora_ <br>woodland &_Combretum-_<br>_Terminalia_ woodland &<br>Savanna|Giraffe, Burchell’s Zebra, &<br>Lelwel Hartebeest|\n\n\nList of Controlled Hunting Areas in Ethiopia\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n|Name|Area<br>(Km2)|Region|Form of<br>hunting|Major Trophy Species|\n|---|---|---|---|---|\n|Hanto|480|Oromiya|Concession|Mountain Nyala<br>Menelik’s Bush buck|\n|Arbagugu|225|Oromiya|Concession|Mountain Nyala<br>Menelik’s Bush buck|\n|Munessa Kuke|111|Oromiya|Concession|Mountain Nyala<br>Menelik’s Bush buck|\n|Ababasheba<br>Demero|210|Oromiya|Concession|Mountain Nyala<br>Menelik’s Bush buck<br>Giant Forest Hog|\n|Besmena<br>Odobulu|350|Oromiya|Concession|Mountain Nyala<br>Menelik’s Bush buck<br>Giant Forest Hog|\n|Kebena|300|Afar|Concession|Beisa Oryx<br>Soemmerring’s Gazelle|\n|Blen hertele|1095||Concession|Gerenuk<br>Beisa Oryx<br>Soemmering’s Gazelle|\n|Telalk\n\nMention: \"List of Controlled Hunting Areas in Ethiopia\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The table of areas, sizes, regions and trophy species is reproduced from this official list.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:015421:12:0", "occurrence_id": "75a325197c76e88b157d02b5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nUG Teacher and School Effectiveness Project (P133780)\n\n\n3) Revision of targets for several IRI indicators and replacement of other IRIs. The new targets and revisions to the IRI\nframework are highlighted in Table 3 below.\n\n(v) **Reallocate funds between the disbursement categories.** The following allocations are proposed to reflect the changes\n\ndescribed above.\n\n\n**<u>Table 3. Allocation of Funds between Categories</u>**\n\n|Category|Current Allocation, USD|Proposed Allocation, USD|\n|---|---|---|\n|**Category 1 (Component 1 and 2)**|77,200,000.00|83,417,350.00|\n|**Category 2 (Component 3)**|14,900,000.00|16,582,650.00|\n|**Category 3 (Unallocated)**|7,900,000.00|0.00|\n\n\n\n**Results Framework and Project Development Objectives (PDOs)**\n\n\nAs discussed earlier, project activities will be simplified but the PDO remains the same, “ **to support the Government in**\n**improving teacher and school effectiveness in the public primary education system”** . The activities under the project will\nremain complementary and they will be coordinated with each other for maximum impact. In terms of PDO indicators, three of\nthe four indicators will remain the same with the targets only changing for PDO indicator 3, which focuses on school construction\n(see Table 4). The fourth PDO indicator will change from percent of teachers present in public schools in targeted districts to a\nnew PDO indicator -Percentage of pupils reading 20 or more words per minute in Grades 1-3. The baseline for this indicator\nwas collected in October 2016, during the Early Grade Reading Assessment of Pupils in grades 1\n\nMention: \"Early Grade Reading Assessment\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The new reading-fluency indicator takes its October 2016 baseline from this assessment of grade 1-3 pupils.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:015577:3:1", "occurrence_id": "30f8228f3173ad4716b03ea5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  north (Longitude 1° 41’ S), to Tanzania’s border at Vanga in the south\n(Longitude 4° 40’ S). Kenya’s coast includes thirteen districts, seven of which are\ntargeted by the project: Kilifi, Kwale, Lamu, Malindi, Mombasa, Taita Taveta and Tana\nRiver. The coastal region extends 150 km inland from the seafront, covering an area of\n67,500 km <sup>2</sup> constituting about 11.5% of the total area of the Republic of Kenya. A\nfringing coral reef runs parallel to the coastline, from Vanga to Malindi Bay. Other unique\nfeatures include: (a) the Lamu archipelago with its extensive mangrove forests and\ncultural sites; (b) Mombasa Island; (c) the southern complex of Gazi Bay; (d) Chale\nIsland; (e) Funzi Bay and Wasini Island; (f) Tana and Sabaki rivers that drain into the\nIndian Ocean. Kenya has about 500 km <sup>2</sup> of mangrove forest which are a vital part of the\ncoastal ecosystem, providing economic development, environmental services and social\nand cultural values.\n\nThe Coast province covers an area of 83,603 km² and has a population of 2,487,264\ninhabitants according to the 1999 census. About 62% of the coastal population lives\nbelow the poverty line, making this province the second poorest of Kenya’s eight\nprovinces. Key reasons include the inequitable land tenure regimes and weak\ninstitutionalization and lack of proper integration of urban planning into the development\nagenda.\n\nAlthough the exact number of villages has not yet been determined, the project would\ncover at least 15 villages in 8 coastal districts that have natural resource-based economies\nrelated to the following three geographic areas of the coastal and marine environment: (i)\n\nMention: \"1999 census\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Provincial population and area are given from the census to frame the coastal poverty discussion.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:015756:19:1", "occurrence_id": "908ce84760e220a89d54bc74", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: s) and were able to bid and win contracts for other infrastructure projects. In summary, over the fiveyear program period, a total of 657,494 jobs were created of which 45% were female, comprising 485,209\ntemporary jobs (45% female) and 172,285 permanent jobs (43% female) <sup>10</sup> .\n\n57. **Increased income and Food Security.** Participants in the Beneficiary Assessment indicated that as a\nresult of increased business opportunities and jobs, their households were food secure, they were able to pay\nfor their utilities, had accumulated household assets and had saved and invested in business. 75.6 percent of\nthe respondents said they had invested; 76.2 percent had acquired household assets and 71.4 percent said\nthey were better positioned to pay their children’s fees.\n\n\n58. **Capacity Building.** Youth, women and persons with disabilities that participated in the construction of\ncobblestone roads acquired lifetime skills like chiseling and paving. For some, the knowledge and skills\nobtained laid the foundation for their transition to entrepreneurs or proprietors of MSEs that would\nindependently bid for works.\n\n\n10 Borrower ICR, page 40.\n\n\n12\n\nMention: \"Beneficiary Assessment\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Participant testimony is used to link jobs and business opportunities to household food security, asset building and saving.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:015824:58:0", "occurrence_id": "a3e8f9bc1f5295e4f05ef2b6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |15|Time saved due to improved water<br>availability within 1.5 km for rural<br>and 0.5 m in urban|And time used for what? It is more important that we understand the impacts<br>of the time saved on livelihood, well-being and women’s empowerment.<br>Impacts of Time same can be tracked through case studies, external research<br>and potentially an additional question into the EDHS.|\n|---|---|---|\n|**16**|Percentage increase in enrollment of<br>female students in school; % decrease<br>in dropouts among female students|<br> <br>This is complex and not directly attributable – of course It is related to the<br>provision of WaSH in part – not all reductions in dropout and increases in<br>enrollment can be attributed to WaSH facilities.  Probably GBV (rape and<br>sexual harassment) and early marriage outrides WaSH.|\n\n\n\n48\n\nMention: \"EDHS\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"The reviewer proposes adding a question to the survey so time-saved impacts on women could be tracked.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:015924:17:0", "occurrence_id": "6a0b1f35fc735dbd6604dfb6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |#|Indicator Name|Core|Unit of<br>Measure|Base<br>Line|Revised<br>Target|Cumulative<br>Target<br>Values|June<br>2020<br>Progress|December 2020<br>Progress|Col10|March 2021<br>Progress|Col12|Remark|\n|---|---|---|---|---|---|---|---|---|---|---|---|---|\n|**#**|**Indicator Name**|**Core**|**Unit of**<br>**Measure**|**Base**<br>**Line**|**Revised**<br>**Target**|YR3|YR3|value|%age|value|%age||\n|||||||||||||domestic and 15 % non-domestic customer share<br>and a HH size of 7 and 14 for domestic and non-<br>domestic customers respectively.|\n|7|Participating utilities<br>that have<br>established/ are using<br>NRW data<br>management system<br>(Number)||Number|-|23|14|0|0||0|||\n|7.1|Addis Ababa<br>(Number)|Addis Ababa<br>(Number)|Addis Ababa<br>(Number)|-|1|1|0|0||0|||\n|7.2|Secondary Cities<br>(Number)|Secondary Cities<br>(Number)|Secondary Cities\n\nMention: \"NRW data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"An indicator counts utilities that have established or are using a non-revenue water data management system.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:016402:31:0", "occurrence_id": "d62ffad446a6da7832c23e3b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: the 1980s and 1990s, the Borana adopted farming and increasingly demarcated and privatized formerly\ncommunal lands. On the Kenya Ethiopian border, 200 sq. km of land has been enclosed for crop farming\n(millet and maize).\n\n66. Ethnic identity among the northern pastoralist groups living near the border with Ethiopia and\nSomalia is often fluid and instrumental, people switching between being Oromo (Ethiopians) and Kenyans\nat different times and in different situations.\n\n67. **_Orma:_** This is semi-nomadic pastoralist group that lives in the north-eastern part of Kenya, along\nthe Tana River in Garissa and Tana River counties. The 2009 census lists them as an independent tribe\nwith a population of 66,000. They are descendants of the Galla nation of Ethiopia and northern Kenya,\nwho moved to the Tana Delta in the late 19th century after tribal clashes. They have been entirely Muslim\nfor four generations. Their language is classified as Cushitic and is widely used.\n\n68. Orma herd cattle, sheep and goats. They have their base villages in the delta but move back and\nforth with their cattle from the hinterlands (wet season grazing areas) towards the river (dry season\ngrazing areas).\n\n69. **_Pokot:_** The Pokot live both in Uganda and in Kenya. In Kenya, they are part of the Kalenjin tribal\ngroup and the 2009 census puts their population at 633,000. They live in Baringo and West Pokot\ncounties. They speak Pökoot, language of the Southern Nilotic language family. Like most East African\npastoral societies, the Pokot have an egalitarian political organization. Elders preside over important\ncommunity decisions, festivals and religious ceremonies. Neighborhood councils grouping together elders\nfrom the communities are important for decision-making at local level and regarding communal resources\n(water, pastures).\n\n70. Based on area and cultural differences,\n\nMention: \"2009 census\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The census is cited for the group's population and its listing as a distinct category.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:016404:7:0", "occurrence_id": "aefd10117b76d723882cc4bd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The** **Uganda** **Multi** **Sectoral** **Food** **Security** **and Nutrition Project** **(UIMFSNP)**\n**Financial Statements** **for** **the Period** **ended** **3** **0 th** **June 2020**\n\n\n**1.0.INTRODUCTION**\n\nThe Human Development Index (HDI) **2013** listed Uganda as one of **15**\ndeveloping countries whose income growth benefited the poor. However,\nhuman development remains 'low' and Uganda is ranked **161** out of **186**\ncountries, underscoring the need to improve pro-poor policies in social\nsectors. Under nutrition is a critical element in this ranking because **of** its\ncontributions to productivity losses from poor physical status, increased health\ncare costs, and reduced cognitive functions and learning performance, In\nUganda, **33** percent of under-5 children are stunted, higher than in\nneighboring countries with lower per capita income. Stunting is nearly twice\nas high in rural compared to urban areas **(36** percent compared to **19**\npercent), and there are great regional variations. Micronutrient deficiencies\n\n(particularly vitamin **A** and iron) are also **highly** prevalent across **all** regions.\nRecognizing that nutrition interventions are essential investment in public\nhealth for long term development, the Government of Uganda **(GoU)** has\nbegun the implementation of the 'Uganda Multi-sectoral Food Security and\nNutrition Project' aimed at increasing the production and consumption of\nmicronutrient-rich foods and utilization of community-based nutrition services\nin smallholder households. This project is being implemented collectively **by*\n\nMention: \"Human Development Index\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The index round is used both to credit income growth that reached the poor and to show development remains low at rank 161 of 186.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:016949:41:0", "occurrence_id": "a4519a0370966d2d7ca7c5b3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Table 5 **. Population Size of Dessie Town by Sex Distribution, 2011**\n\n|Male|Col2|Female|Col4|Total|Col6|\n|---|---|---|---|---|---|\n|Number|%|Number|%|Number|%|\n|84,224|48.21|90,481|51.79|174,705|100|\n\n\n\nSource: CSA, Population Size by Sex, Area, and Density by Region, Zone and Woreda, 2011\n\n\nThe settlement pattern of communities around the project area has created significant problem in\n\n\nthe provision of infrastructure and other services. This is mainly due to the unfavorable\n\n\ntopography of the place which is dominated by mountainous and gorge landscape. This\n\n\ncombined with landslide problem that is recurring along ‘Tossa’ mountain ridges has forced\n\n\nsome residents to think resettlement to other better places.\n\n###### **5.2.2. Health service**\n\nThe health problem and disease causes among the city population are mainly communicable diseases\n\n\nwater and/or air born diseases. Major factor contributing to these diseases is the poor waste\n\n\nmanagement and pollution impacts of both the air and water resources of the area. Health facilities\n\n\nare reasonably increasing in the area with improved coverage. Health services are provided both by\n\n\ngovernment, NGOs and private institutions. Table below gives list of health institutions available in\n\n\nthe city.\n\n|No|Type of institution|Government|Private|Total|\n|---|---|---|---|---|\n|1|Hospital|2|3|5|\n|2|Health center|7|0|7|\n|3|Clinic|0|15|15|\n|4|Pharmacy||10|10|\n|5|Rural drug vendor||25|25|\n\n\n\nTable 6. Health Institutions in\n\nMention: \"CSA\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Town population by sex is reproduced from the statistical agency's population-by-area release.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:017362:0:2", "occurrence_id": "dcb0cd42cffeb9b0647f8d14", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: b) building the capacity of communities to support child development<br>programs through community planning and sensitization workshops and training in entrepreneurship to<br>increase incomes of mothers and caregivers.<br>(2) Community Support Grants and Innovation Grants for Child Development(US$10.5m base cost, 29%<br>of total; US$7.72m actual, 19% of total).  The former are grants offered on the basis of matching<br>contributions from communities that cover activities within the menu in the project implementation manual,<br>for communities that develop Community Action Plans, while the latter are grants made to communities to<br>address child related problems through interventions outside the project's menu .<br>(3) National Level Program Support for Child Development(US$5.4m base costs, 15% of total; US$7.64m<br>actual cost, 19% of total), which included support for: monitoring and evaluation; preventive program<br>initiatives on micronutrients; early childhood development (ECD) curriculum development; training of<br>trainers for ECD; and information, education and communications (IEC) and advocacy for children's rights.<br>(4) National Household Survey(not appraised; US$1.18m actual cost, 3% of total), for the Uganda Bureau<br>of Statistics to conduct the 1999/2000 survey.  It is unclear how this component would have contributed to<br>the achievement of the project's objectives or, indeed whether nutritional or health status indicators were<br>collected for children 0-3.<br> **c. Comments on Project Cost, Financing, Borrower Contribution, and Dates**<br> The Implementing Agency was originally supposed to be the Ministry of Finance and Economic Planning .<br>However, effectiveness was delayed\n\nMention: \"National Household Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"A component financed the statistical office to run this survey round, though its contribution to objectives was questioned.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:017532:4:0", "occurrence_id": "540bace535023db159360cce", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Public Disclosure Copy\n\n\n**The World Bank** Implementation Status & Results Report\nTana & Beles Integrated Water Resources Development (P096323)\n\n\nconsultations. consultations. stakeholders.\n\n\nDate 29-May-2008 31-Jul-2013 12-Dec-2013 30-Sep-2014\n\n\n**PHINDPDOTBL**\n\n\n Soil Erosion Monitoring and Evaluation system developed and monitoring network in place (Text, Custom)\n\n\nBaseline Actual (Previous) Actual (Current) End Target\n\n\n\nM&E system\nand network for\nsoil erosion\nmonitoring and\nevaluation\noperational by\nBureaus of\nWater\nResource,\nAmhara region\n(for microwatersheds)\nand TaSBO (for\nsubwatersheds)\n\n\n\nNRM baseline survey\ncompleted. M&E\nnetwork established.\nHydrological gauging\nstations at subwatershed level\nestablished.\nHydrological data has\nbeen regularly collected\nand analyzed.\n\n\n\nValue\n\n\n\nNo soil and water\nmonitoring system is in\nplace in the project area of\nAmhara region\n\n\n\nNRM baseline survey\ncompleted. M&E\nnetwork established.\n\n\n\nDate 29-May-2008 31-Jul-2014 12-Dec-2014 30-Sep-2014\n\n\n**PHINDPDOTBL**\n\n Cumulative area with improved natural resources management practices (soil conservation, afforestation, rain fed agriculture\nimprovement etc.) in the Ribb, Gumera and Jemma sub-watersheds (Hectare(Ha), Custom)\n\n\nBaseline Actual (Previous) Actual (Current) End Target\n\n\nValue 0.00 63703.00 70886.00 83000.00\n\n\nDate 01-May-2008 31-Jul-2014 12-Dec-2014 30-Sep-2013\n\n\n**PHINDPDOTBL**\n\n Initial integrated basin management plans for Tana and Beles sub-basins developed and approved by key stakeholders\n(regional bureaus and authorities, Basin Organizations, civil society organizations a (Number, Custom Supplement)\n\n\nBaseline Actual (Previous) Actual (Current) End\n\nMention: \"NRM baseline survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Completing the baseline is reported alongside establishing the monitoring network and gauging stations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:017532:4:0", "occurrence_id": "115819ac8cdea30be3523309", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Public Disclosure Copy\n\n\n**The World Bank** Implementation Status & Results Report\nTana & Beles Integrated Water Resources Development (P096323)\n\n\nconsultations. consultations. stakeholders.\n\n\nDate 29-May-2008 31-Jul-2013 12-Dec-2013 30-Sep-2014\n\n\n**PHINDPDOTBL**\n\n\n Soil Erosion Monitoring and Evaluation system developed and monitoring network in place (Text, Custom)\n\n\nBaseline Actual (Previous) Actual (Current) End Target\n\n\n\nM&E system\nand network for\nsoil erosion\nmonitoring and\nevaluation\noperational by\nBureaus of\nWater\nResource,\nAmhara region\n(for microwatersheds)\nand TaSBO (for\nsubwatersheds)\n\n\n\nNRM baseline survey\ncompleted. M&E\nnetwork established.\nHydrological gauging\nstations at subwatershed level\nestablished.\nHydrological data has\nbeen regularly collected\nand analyzed.\n\n\n\nValue\n\n\n\nNo soil and water\nmonitoring system is in\nplace in the project area of\nAmhara region\n\n\n\nNRM baseline survey\ncompleted. M&E\nnetwork established.\n\n\n\nDate 29-May-2008 31-Jul-2014 12-Dec-2014 30-Sep-2014\n\n\n**PHINDPDOTBL**\n\n Cumulative area with improved natural resources management practices (soil conservation, afforestation, rain fed agriculture\nimprovement etc.) in the Ribb, Gumera and Jemma sub-watersheds (Hectare(Ha), Custom)\n\n\nBaseline Actual (Previous) Actual (Current) End Target\n\n\nValue 0.00 63703.00 70886.00 83000.00\n\n\nDate 01-May-2008 31-Jul-2014 12-Dec-2014 30-Sep-2013\n\n\n**PHINDPDOTBL**\n\n Initial integrated basin management plans for Tana and Beles sub-basins developed and approved by key stakeholders\n(regional bureaus and authorities, Basin Organizations, civil society organizations a (Number, Custom Supplement)\n\n\nBaseline Actual (Previous) Actual (Current) End\n\nMention: \"Amhara region\n\n\n\nNRM baseline survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"In a region with no prior soil and water monitoring, the baseline is recorded as the first such reference.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:017547:33:0", "occurrence_id": "a2ee177ea650c515985f3b1f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Environmental & Social Impact Assessment Project Report for the Rehabilitation and Improvement of Quarry_\n\n_<u>Roads in Mulolongo Area of the Nairobi Metropolitan Region</u>_\n\n\n**3.6** **Community Organizations/Non-State Actors**\n\n\nMajority of the county’s population is organized into community self-help groups,\nand both producer and marketing cooperative societies. There are a number of\nother non-state actors operating in the county including local and international NonGovernmental Organizations whose objective is to improve the socio-economic\nwell-being of the local people.\n\n\n**3.7** **Gender**\n\n\nGender disparities are minimal in primary and secondary education where\nenrolments are 50 per cent for both boys and girls. Women have been\ndiscriminated against when it comes to access to ownership of property and\nfinances. 80 per cent of women constitute the agricultural workforce but only a\nsmall percentage of them hold title deeds to land. This imposes a great constraint\non their ability to make major land-related investment decisions including obtaining\ncredit using title deeds as collateral.\n\n\n**3.8** **Poverty**\n\n\nThe county has high poverty levels which according to the 2005/2006 Kenya\nIntegrated Household Budget Survey, about 42 per cent of the population live\nbelow the poverty line. The poor are not able to access the basic necessities of life\nsuch as food, shelter and education.\n\n\n**3.9** **HIV and Aids**\n\n\nHIV and AIDS pandemic poses a serious threat to the development of the area as\nthe prevalence rate stands at 3.6 per cent for Machakos County compared to 4.6%\nfor the country as whole. The scourge is on the increase virtually in all the\nconstituencies of the county. AIDS related deaths are common and those mainly\naffected are those within the productive age group of 15-49 years of age. Also, the\nnumber of HIV/AIDS orphans\n\nMention: \"Kenya\nIntegrated Household Budget Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The county poverty share of about 42 per cent below the poverty line is taken from this survey round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:017825:40:1", "occurrence_id": "7c4257eea2b5e1cf62f4a879", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  In order to provide greater\ntransparency and information dissemination, CAAPU’s composition should be expanded to more\nstakeholders, particularly representatives of the construction industries and universities.\n\n**Review urban structural plans to incorporate the new land value capture instruments.**\nMunicipalities should establish a new posture on the use and occupation of land, setting maximum\nurban indexes and coefficients for specific neighborhoods or regions, considering the local\ncharacteristics and needs of the community. When registering the property before a PPU or\ndetailed plan, holders of vacant and underused land should obtain permits from the municipality.\nIn case of noncompliance, they should be required to make social use of the property, under penalty\nof losing possession and returning it to the municipality.\n\n**Establish higher minimum penalties for nonpayment of property tax** . Higher fines could help\nincrease property tax collection. But penalties should be preceded or accompanied by a wide\ncampaign to encourage payment of the overdue property taxes. The current dual-payment system\n(bank and Finance Department) is archaic, prone to errors and fraud. IPRA invoices are issued and\ndelivered, and payment can be made at bank branches or directly to the Department of Finance of\nMaputo or Matola. Although payment at the Department of Finance is directly registered in the\nIPRA cadaster, those at banks are considered paid only when the actual payment slips are\nphysically presented to the Department of Finance (and subsequently recorded in the IPRA\ncadaster). Most taxpayers seem therefore prefer paying the IPRA directly at the Department of\nFinance.\n\n**Set a limited period (1 year) for individuals and firms to pay overdue IPRA before increasing**\n**penalties** . Preceding the amendments, fines and collection procedures, a campaign must be\nimplemented for the settlement of debts in arrears, to encourage payment, before the\nimplementation of new procedures.\n\n\n32\n\nMention: \"IPRA cadaster\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Property tax payments are recorded in the cadaster, but only once payment slips are physically presented, creating a lag.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:017938:4:0", "occurrence_id": "acb2856b96779136e4d97155", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Second, an analysis over the period 2005-2009 was carried out by the ICR team using sub-district (division) \n\u0001\nlevel data from the project-supported household early warning system (EWS), satellite-based rainfall estimates, \nand information from project-supported micro-project investments. The analysis covered 51 treatment and 46 \ncontrol divisions, approximately equally divided between arid (52) and semi-arid (45). The results indicate a \n19.7% difference between treatment and control divisions in terms of the percentage change in the proportion of \npeople assessed as needing food aid. This result was statistically significant at 5%. However, the absolute fall in \nthe share of the population needing food aid in the treatment divisions was only 1.7% on average (from 31.9% to \n30.2%). For the ASALs as a whole (arid and semi-arid), the average fall (weighted by population) was 4% (from \n27.2% to 23.2%).\nOverall, there is evidence indicating a small decrease in the food -aid needs of the population benefiting from the \n\u0001\nproject in the arid districts (11 of the 28 covered) which were targeted by all three components. Also in the arid \ndistricts, there is a statistically significant correlation between this decrease and cumulative project expenditures \nindicating a reasonable degree of attribution . In the semi-arid districts, attribution is less clear.\nWith regard to speed of response, the ILRI impact evaluation indicated that users of the EWS monthly bulletins, \n\u0001\nproduced and disseminated in the project area and at national level on a monthly basis, reduced their average \nresponse time to drought-related emergencies from 7.6 weeks in the year 2000 to 3.5 weeks in 2009. “The \nanalysis is based on a survey by ILRI of [District Steering Group] members in 10 study districts to find mean \nresponse times for users and non-\n\nMention: \"project-supported household early warning system\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Sub-district records from the early warning system are one of three inputs to the treatment-control impact analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:017938:4:0", "occurrence_id": "8f6df2dd662c1049de1023d6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Second, an analysis over the period 2005-2009 was carried out by the ICR team using sub-district (division) \n\u0001\nlevel data from the project-supported household early warning system (EWS), satellite-based rainfall estimates, \nand information from project-supported micro-project investments. The analysis covered 51 treatment and 46 \ncontrol divisions, approximately equally divided between arid (52) and semi-arid (45). The results indicate a \n19.7% difference between treatment and control divisions in terms of the percentage change in the proportion of \npeople assessed as needing food aid. This result was statistically significant at 5%. However, the absolute fall in \nthe share of the population needing food aid in the treatment divisions was only 1.7% on average (from 31.9% to \n30.2%). For the ASALs as a whole (arid and semi-arid), the average fall (weighted by population) was 4% (from \n27.2% to 23.2%).\nOverall, there is evidence indicating a small decrease in the food -aid needs of the population benefiting from the \n\u0001\nproject in the arid districts (11 of the 28 covered) which were targeted by all three components. Also in the arid \ndistricts, there is a statistically significant correlation between this decrease and cumulative project expenditures \nindicating a reasonable degree of attribution . In the semi-arid districts, attribution is less clear.\nWith regard to speed of response, the ILRI impact evaluation indicated that users of the EWS monthly bulletins, \n\u0001\nproduced and disseminated in the project area and at national level on a monthly basis, reduced their average \nresponse time to drought-related emergencies from 7.6 weeks in the year 2000 to 3.5 weeks in 2009. “The \nanalysis is based on a survey by ILRI of [District Steering Group] members in 10 study districts to find mean \nresponse times for users and non-\n\nMention: \"satellite-based rainfall estimates\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Remote rainfall estimates are combined with household alerts to compare outcomes across arid and semi-arid divisions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:018575:57:0", "occurrence_id": "a898a3a33fddddc1caede9cf", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 2009 census with 8,000 and 161,000 individuals respectively.\n\nThe **Somali** include various clans, including the Ajuran, Degodia, Arri (Gurreh, Gari),\nHawiyab, Murile, Ogaden, Wardei, etc., some of whom are listed as independent groups in\nthe 2009 census.\n\n\n**58 |** P a g e\n\nMention: \"2009 census\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Group population figures and their classification as independent groups come from this enumeration.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:019291:6:0", "occurrence_id": "139bb1d36d8873d2c6713a0e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: National Integrity Survey corruption had not been reduced during the recent PRSC series and Transparency\nInternational came to the same conclusion. Uganda ranks 130th out of 180 countries in its 2009 Corruption\nPerception Index which is below many other African countries (ICR, pages 19 and 35) . The efficacy of this\nsub-objective was rated as negligible.\n\n\n**Improved public sector management** **.** Weakness in public sector management in Uganda is consistent with its\nCPIA ratings. Low revenues and inadequate staffing were major problems . During the PRSC series the\npercentage of local government (LG) revenues as a share of their budgets fell from 6 to 3 percent between\n2002/03 and 2008/09, far short of the 9 percent target.  The decline was probably caused by abolition of the\ngraduated tax (flat tax on employed Ugandans over 18).  The PRSCs assisted by designing other revenue\nsources to offset the loss of the tax .  Also, under PRSC 6 the Ministry of Local Government was restructured to\nimprove its capacity and PRSC 7 supported actions to enhance LG staffing fill rates from 35 to 46 percent.\nDespite the LG's other revenue sources, the percentage of transfers from the central government that had\nconditions attached rose to 95 percent of LG budgets compared with a target of 91 percent (ICR, pages 20 and\n21). The efficacy of this sub-objective was rated as modest .\n\n\nThis Review rates the efficacy of the second strategic objective under Pillar 4 (made up of four sub-objectives)\nas modest.\n\n\n**PEAP Pillar 5 - Human Development**\n\n\n**(i) A better educated society - including increased access to and quality of primary, secondary and tertiary**\n**education** . The ICR notes the many challenges facing the education system, gender disparities . stagnation in\ncompletion rates and regional inequalities. The PRSC series supported an increase in the number of teachers,\ntextbooks and\n\nMention: \"National Integrity Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Survey evidence is used to conclude that corruption had not fallen during the reform credit series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:019291:6:0", "occurrence_id": "b3045f7da64ddc6e895915ba", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: National Integrity Survey corruption had not been reduced during the recent PRSC series and Transparency\nInternational came to the same conclusion. Uganda ranks 130th out of 180 countries in its 2009 Corruption\nPerception Index which is below many other African countries (ICR, pages 19 and 35) . The efficacy of this\nsub-objective was rated as negligible.\n\n\n**Improved public sector management** **.** Weakness in public sector management in Uganda is consistent with its\nCPIA ratings. Low revenues and inadequate staffing were major problems . During the PRSC series the\npercentage of local government (LG) revenues as a share of their budgets fell from 6 to 3 percent between\n2002/03 and 2008/09, far short of the 9 percent target.  The decline was probably caused by abolition of the\ngraduated tax (flat tax on employed Ugandans over 18).  The PRSCs assisted by designing other revenue\nsources to offset the loss of the tax .  Also, under PRSC 6 the Ministry of Local Government was restructured to\nimprove its capacity and PRSC 7 supported actions to enhance LG staffing fill rates from 35 to 46 percent.\nDespite the LG's other revenue sources, the percentage of transfers from the central government that had\nconditions attached rose to 95 percent of LG budgets compared with a target of 91 percent (ICR, pages 20 and\n21). The efficacy of this sub-objective was rated as modest .\n\n\nThis Review rates the efficacy of the second strategic objective under Pillar 4 (made up of four sub-objectives)\nas modest.\n\n\n**PEAP Pillar 5 - Human Development**\n\n\n**(i) A better educated society - including increased access to and quality of primary, secondary and tertiary**\n**education** . The ICR notes the many challenges facing the education system, gender disparities . stagnation in\ncompletion rates and regional inequalities. The PRSC series supported an increase in the number of teachers,\ntextbooks and\n\nMention: \"2009 Corruption\nPerception Index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"The country's rank of 130 of 180 is used to corroborate the survey finding against regional peers.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:019593:52:1", "occurrence_id": "ef27559d10aafcc19cd34453", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .8 square kilometers of which 86% of it is dry\nland, 12% is open water and 2% is permanent swamp/wetland. Meanwhile 63% of the land area is\noccupied by nature and wildlife conservation schemes, and other government projects such as\nprisons, mining and irrigation farming leaving only 37% for human settlement. With a population\ngrowth rate projected at 3.6% per annum according to the 2002 population census where its\npopulation was 523,033 people, this alludes to a high population density in the area. The density\nwas 183 persons per square kilometer in 2002 which translates into 450 persons per sq. km in the\narea actually occupied by people. This high density exerts a lot of pressure on available resources\nespecially land which is accompanied by land related problems in the district.\n\n\n**5.1.2** **Ethnicity:**\nThe district is largely inhabited by the Bakonjo who are the majority and dominant group according\nto findings. There were other minority groups identified as the Bagabo, Basongora, Banyabindi,\nBakinguz and IPs ( _Batwa-_ also referred to as Pigmies) who were reported to be very few. These\ngroups were reported to be marginalized within the district because of their low numbers. The\nBasongora occupy parts of 3 out of 29 sub counties namely Lake Katwe, Karusandara and Katwe,\nKabatoro sub counties. The rest of the minority ethnic groups in Kasese were reported to be\nscattered all over the district. There are also people from other tribes within Uganda who have\nsettled in the District. The most commonly spoken language is Lukonzo and other small languages.\n\n\n**5.1.3** **Gender Dynamics in Kasese district:**\nAccording to the 2014 population census, women accounted for **52%** of the district’s total\npopulation. In certain instances, large\n\nMention: \"2002 population census\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Population and growth rate from the census are used to derive the density actually experienced by settlers.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:020040:2:2", "occurrence_id": "59d003b07f31c294c0fbe29c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ---|---|---|---|\n|Issues in NACC#s  qualified report to be<br>addressed by end of March of following year||Yes/No|Value||No|Yes|\n|Issues in NACC#s  qualified report to be<br>addressed by end of March of following year||Yes/No|Date|01-Jan-2007|27-Oct-2011|30-Jun-2013|\n|Issues in NACC#s  qualified report to be<br>addressed by end of March of following year||Yes/No|Comments|N/A This is a new indicator<br>which was not being|Audit completed, but not<br>cleared by KENAO due to<br>CSO component|IOI# 2 Audit conducted by<br>December each year.|\n|The proportion o f KNASP M&E indicators<br>included in the annual M&E report,<br>disseminated and available on time for the<br>JAPR.||Percentage|Value|55.00|89.00|85.00|\n|The proportion o f KNASP M&E indicators<br>included in the annual M&E report,<br>disseminated and available on time for the<br>JAPR.||Percentage|Date|01-Jan-2007|30-Dec-2011|30-Jun-2013|\n|The proportion o f KNASP M&E indicators<br>included in the annual M&E report,<br>disseminated and available on time for the<br>JAPR.||Percentage|Comments||Kenya Global AIDS report|IOI#3|\n|The proportion o f KNASP M&E indicators<br>included in\n\nMention: \"Kenya Global AIDS report\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The report is noted as the reference behind an indicator on timely dissemination of strategy monitoring results.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:020132:10:0", "occurrence_id": "ac4aa2e794fbc3cd1b59a943", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ###### Resettlement Action Plan (RAP) For Dire Dam Rehabilitation Project\n\n**1. INTRODUCTION**\n\n**1.1 Background and Rationale of the project**\n\nAddis Ababa has a population of 2,738,248 with annual growth rate of 2.1 percent\n(Population and housing census of CSA, 2007). Addis Ababa water supply and\nsewerage Authority has a due responsibility to supply safe water for the Addis Ababa\ncity population from different sources including ground and surface water. The year to\nyear water supply demand increment following the population growth as well as fast\ndevelopment in various aspects has made the authority to increase its effort to meet the\ndemand.\n\nTo this effect AAWSA water supply and sanitation and rehabilitation project office has\ndesigned to upgrade the capacity of the Dire Dam reservoir by increasing the Dam\nheight by 1.25 meter that would enable to increase the storage capacity of the reservoir\nfrom 19 MCM to 23.9 MCM.\n\nIn effect the proposed project would enable AAWSA to increase its water supply\ncoverage by 4.9 MCM which could help to reduce the water supply shortage problem\nthat exists in the city by increasing the supply coverage.\n\nThe Dire Dam upraising project will involve construction activities in the area mainly\nusing local resources. As the effect of Dam upraising there is a need to expand the\nsanitary zone around the reservoir which will be protected from human activity as well\nas livestock movement. The sanitary zone expansion will require land that will be\nexpropriated from local farmers who use the land for different purpose including farm\nand grazing.\n\nIn short the proposed project has a considerable positive impact on socio-economic\ndevelopment of the city population in particular and economic development of the\ncountry in general since the city is the seat of various Diplomats, international\norganization and first focus of tourists.\n\nOn other hand the project is anticipated to create some negative impact on a few\nhouseholds who reside in the project area. The major impact of the project is\n\nMention: \"Population and housing census of CSA\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"City population and growth rate are quoted from the census to establish the water demand baseline.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:020579:1:0", "occurrence_id": "078a19f83a13a71be294a91b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Demographic Health Survey (KDHS 2008) began to show some remarkable improvements in infant\nand under-five mortality. Infant mortality declined from by 33 percent to 52 per thousand live births\nand under five mortality decreased by 36 percent to 74 per thousand live births during the period\n2004-2008. This decrease can be attributed in part to successful implementation of vertical health\nprograms especially immunization, malaria, tuberculosis and HIV/AIDS in addition to sustained\neconomic growth and poverty reduction. However, several challenges still remain. The prevalence\nof stunting of over a third of Kenyan children has not changed significantly over the past few\nyears, and the maternal mortality rates remain stubbornly high (around 530 per 100,000 live\nbirths), and achievement of MDG 5 remains out of reach. Geographic and economic health\ninequities continue where nearly 90 percent of the women in Nairobi deliver their babies in a health\nfacility in sharp contrast to only 17 percent in the North-East highlighting the geographic inequities\nwithin Kenya. Health services in Kenya are provided by a wide range of players consisting of\npublic, faith based, private not for profit, and private for profit sectors. About 40 percent of\nservices are estimated to be provided by the private sector, and public health services at primary\ncare facilities are utilized by the poor and particularly\n\nwomen and children in many remote areas rely on services provided by faith based organizations.\nKenya's new constitution (August 2010) provides an overarching framework for a more\ncomprehensive rights-based approach towards health care. The constitution devolves power and\naccountability to the 47 counties and the proposed devolution is considered to be one of the most\nambitious in the world. The devolved governance model promises Kenyans a more equitable\ndevelopment model by creating opportunities for new growth centers. The national government will\nbe providing the broader policy direction.\nProviding universal health coverage received priority attention by the new government, which has\nalready made important policy pronouncements related to the health sector. The President as a part\nof the first 100-day agenda\n\nMention: \"Demographic Health Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Survey evidence is used to show falls in infant and under-five mortality over the review period.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "fcv", "source_key": "fcv_pads_east_africa:020579:1:0", "occurrence_id": "5cfa442a266c87d417bc10ed", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Demographic Health Survey (KDHS 2008) began to show some remarkable improvements in infant\nand under-five mortality. Infant mortality declined from by 33 percent to 52 per thousand live births\nand under five mortality decreased by 36 percent to 74 per thousand live births during the period\n2004-2008. This decrease can be attributed in part to successful implementation of vertical health\nprograms especially immunization, malaria, tuberculosis and HIV/AIDS in addition to sustained\neconomic growth and poverty reduction. However, several challenges still remain. The prevalence\nof stunting of over a third of Kenyan children has not changed significantly over the past few\nyears, and the maternal mortality rates remain stubbornly high (around 530 per 100,000 live\nbirths), and achievement of MDG 5 remains out of reach. Geographic and economic health\ninequities continue where nearly 90 percent of the women in Nairobi deliver their babies in a health\nfacility in sharp contrast to only 17 percent in the North-East highlighting the geographic inequities\nwithin Kenya. Health services in Kenya are provided by a wide range of players consisting of\npublic, faith based, private not for profit, and private for profit sectors. About 40 percent of\nservices are estimated to be provided by the private sector, and public health services at primary\ncare facilities are utilized by the poor and particularly\n\nwomen and children in many remote areas rely on services provided by faith based organizations.\nKenya's new constitution (August 2010) provides an overarching framework for a more\ncomprehensive rights-based approach towards health care. The constitution devolves power and\naccountability to the 47 counties and the proposed devolution is considered to be one of the most\nambitious in the world. The devolved governance model promises Kenyans a more equitable\ndevelopment model by creating opportunities for new growth centers. The national government will\nbe providing the broader policy direction.\nProviding universal health coverage received priority attention by the new government, which has\nalready made important policy pronouncements related to the health sector. The President as a part\nof the first 100-day agenda\n\nMention: \"KDHS 2008\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The specific round supplies the mortality levels per thousand live births.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "jdc_operational", "source_key": "jdc_operational:000028:38:0", "occurrence_id": "5b785b68f69e1c00ca3984b0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nEnhancing Community Resilience and Local Governance Project Phase II (P177093)\n\n\nmembers as mechanics, and trained the water user committees to collect user fees to pay for spare parts\nand mechanics’ services. The project team participates regularly in relevant cluster working groups such\nas protection, WASH, camp coordination and camp management, community communication, and\nengagement to follow the ongoing dialogue on external assistance. The team also engages regularly with\nUNOCHA, United Nations Mission in South Sudan, relevant donors, and CSOs. Through these engagements,\nthe project will continue to coordinate with other programs.\n\n\n67. **The proposed project will coordinate closely with and complement other World Bank-financed**\n**projects.** ECRP-II will have a shared PMU with the planned PFMIS Project. Further, the two projects will\nwork in a complementary manner to lay the foundation for transparent and accountable\nintergovernmental fiscal transfers for local service delivery. ECPR-II and the ongoing South Sudan Safety\nNet Project (SSSNP) (P169274) follow the same community institutional structure of BDCs and PDCs, and\nwhere target sites overlap, the two projects will use the same community structures. ECRP-II and the\nongoing Provision of Essential Health Services Project ( <mark>P168926)</mark> will coordinate so that where ECRP-II\nbuilds/rehabilitates health clinics, the Health Project can provide the software.\n\n\n68. **ECRP-II will continue to coordinate closely with development and humanitarian agencies on**\n**DRM.** As part of the 2020 Flood Damage Needs Assessment, <sup>71</sup> [^71: World Bank/GFDRR. 2021 _. South Sudan Floods 2020. Remote Flood Damage and Needs Assessment,_ September 2021] the project team coordinated with key\nagencies (for example, FAO, World Food Programme, REACH Initiative, IOM, and UNHCR) and donors (for\nexample, GIZ, Swiss Agency for Development and Cooperation, Dutch Embassy, USAID) to obtain data\n\nMention: \"2020 Flood Damage Needs Assessment\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The flood assessment is the occasion on which the project team coordinated response agencies in the affected areas.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "jdc_operational", "source_key": "jdc_operational:000028:25:1", "occurrence_id": "454099831d22150d0167ed2a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: , ECRP-II will scale up to include refugees in the two refugee-hosting\ncounties that were already targeted under ECRP-I <sup>49</sup> and two new flood-prone vulnerable counties to\nsupport Subcomponent 1.2 activities, leading to a total of 12 counties. Five counties have been shortlisted for the flood risk reduction subcomponent based on predetermined selection criteria (see subcomponent 1.2). Based on budgetary availability, additional counties may be included based on\nconsultation with the Government. Alternatively, some may be replaced should security or accessibility\nbecome an issue during implementation. It was confirmed that the project will use the administrative\ndivision of 10 states, 79 counties, and 2,008 _payams_ just as under ECRP-I in the absence of more recent\nofficial boundaries _._ <sup>50</sup> Any discrepancies on the ground will be resolved in consultation with the\nGovernment.\n\n\n48 The 10 counties were originally selected based on a composite vulnerability index that the World Bank team developed and which the\nGovernment agreed to. The index included (a) concentration of returnees, (b) access to basic services, (c) food insecurity, (d) incidents of violence,\n(e) remoteness, and (f) exposure to natural disasters. The long list of the most vulnerable counties was short-listed based on security and\naccessibility.\n49 As refugees are exclusively in refugee camps, ECRP-I only targeted host communities and not the refugees. Under ECRP-II, the project will target\nboth refugees and host communities in the two refugee-hosting counties of Maban and Pariang.\n50 Such practice is in line with how the national budget is being allocated and how other development partners are operating.\n\n\nPage 20 of 73\n\nMention: \"composite vulnerability index\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"County selection was made from a six-component index combining returnee concentration, service access, food insecurity, violence, remoteness and hazard exposure.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002094:14:0", "occurrence_id": "8697be892d2801efa25cfd9b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Managing the workforce is also important in reducing operating expenses and easing\nliquidity pressures. SMEs tend to adopt measures on both the intensive and extensive margins (i.e.\nincreasing the intensity of the response, or trying something different) but are only able to focus\nmore on the former. This is aligned with evidence from other markets: firms in Ghana and Senegal\nfor example mostly reducing hours or pay, while firms in Bangladesh and South Africa more likely\nto lay off employees. <sup>19</sup> [^19: Based on the World Bank’s COV-BPS.] In India, most SMEs are not laying off their workers, but up to 25% of them\n(depending on the size category) are unable to pay salaries (Buteau and Chandrasekhar 2020).\nNaturally, all firms are found to have decreased working hours, given broad-based lockdowns in\nmany countries. Firms that do increase working hours are most likely involved in the essential healthrelated sectors. On the extensive margin, while more firms reduced than increased their permanent\nworkers, larger firms are found to be more capable of utilizing temporary workers to quickly adjust\nthe size of their workforce.\n\n\n3.2 SMEs have limited avenues amid severe financial constraints\n\nA key concern for most firms experiencing a sudden drop in sales is liquidity constraints.\nBoth cash inflows and outflows require careful management, without which businesses become\nhighly susceptible to permanent closure. Firms generally cope with an abrupt economy-wide\ncontraction in liquidity is by utilizing payment facilities on credit terms or ceasing payments\naltogether and facing the risk.\n\nYet, larger firms have generally better access to credit. In limiting immediate cash outflows,\nsmaller firms are found to be less able to engage with their suppliers on credit terms. A general\ndecline in creditworthiness is observed for these firms across all countries, where the declines in\npurchases via credit are larger or at least as bad as those among large firms. For SMEs with greater\ncapacity, changing the credit terms with suppliers has been crucial in protecting their balance sheet,\n\nMention: \"COV-BPS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Cross-country firm responses to the pandemic in hours and layoffs are attributed to this business pulse survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002094:19:0", "occurrence_id": "bb6449a4215afb33bdd874d0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: But which countries are in highest need of external assistance? Countries that have been\naccumulating debt over the last years have limited fiscal space to support the private sector. When\nin addition a large number of firms in an economy are highly leveraged then there is simply little\nspace for either financial institutions, or the government to fill that liquidity gap with more debt.\nCountries in which both these conditions hold are at greatest risk of rapid economic contraction from\nthe shock and thus in greater need of external assistance. With all their limitations, simple measures\nfor both dimensions – the average sovereign debt ratings of the last three years, <sup>26</sup> [^26: The foreign currency long-term sovereign debt ratings has an index between 1 and 21 and can be mapped\nto the classification used by each of the major credit rating agencies.] and the current\nratio of publicly-listed firms – show that upper-middle-income countries like South Africa, Turkey\nand Russia, as well as large lower-middle-income economies like Nigeria and Pakistan are in that\ngroup (see Figure 9). <sup>27</sup> [^27: Using a three-year average of fiscal balance yields similar results (see Figure A4 in Annex).] These indeed are places where development finance is already deployed\nrelatively intensively. Other large economies, such as Brazil, also have limited fiscal space for\nsupport, but its companies have on average more space to take on additional debt.\n\n\n**Figure 9. Upper-middle income and large economies face greater liquidity shortage**\n**and risks: countries’ categorization**\n\n\nSource: Authors’ elaboration. World Bank, S&P Capital IQ database.\n\n\nMany SME-support measures do not necessarily target specific firms but may only reach\nformal ones. In these countries, policies need to be inclusive of both formal and informal sectors,\npossibly using different instruments. While tax relief programs and support through formal financial\ninstitutions may not reach many SMEs in those countries, they can prevent viable firms from sliding\ninto informality or bankruptcy\n\nMention: \"S&P Capital IQ database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The country categorisation of liquidity risk is built from firm balance-sheet data in this commercial database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002094:9:1", "occurrence_id": "6299412d1fe4abfd3be72a3e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .\nSmall firms experience a 10 percentage point greater fall in sales than larger firms, whereas for\nmedium firms the fall is on average 6 percentage points. Similarly, small firms predominantly\nexplain the exposure of SMEs to financial shocks, as one would expect. In contrast, small and\nmedium firms have similar increased exposure to supply shocks and are equally unlikely to\nexperience positive demand shocks or export demand shocks than larger firms.\n\nCOVID has led to massive increases in uncertainty that is likely leading firms to postpone\ninvestments in capital, skills and innovation – with likely scars for the post-COVID recovery.\nMeasuring uncertainty is not straightforward, with scarce data for developing economies. SMEs\ntypically report higher uncertainty over sales than larger firms, in part, due to poorer access to\ninformation and due to the fact that sales is where the impact by firm size is more severe. Evidence\nfor the United Kingdom and the United States finds that while SMEs face around 20%-30% higher\nuncertainty over sales both before and after COVID, firm uncertainty has reached the highest levels\nsince records began and dwarfing the financial crisis (Altig et al., 2020). <sup>10</sup> [^10: The comparison group here is firms with more than 250 employees, but SMEs are defined as having fewer\nthan 100 employees (as in the rest of this paper). Their records began in 1985.] Results from the 13\ncountries surveyed by the World Bank are mixed, with variation in the overall expectations of return\nto normalcy lower among SMEs than large firms in some countries (see Figure 5), suggesting\nsomewhat greater certainty in overall outlook. Large firms might be better able to cope with the\nchallenges, but often have additional uncertainty to take into account such as foreign suppliers or\nthe stock market, which SMEs are less likely to be listed.\n\n\n9 Smaller firms may however be indirectly affected through disruptions to inputs imported by wholesalers.\nThe World Bank Enterprise follow-up surveys do not allow measurement of the import supply shock.\n10\n\nMention: \"World Bank Enterprise follow-up surveys\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"The surveys are cited as unable to measure the import supply shock, so that channel is left out of the estimates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000465:81:1", "occurrence_id": "a6f5ac732c550130adf42638", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: } {305.8} {252.7} {318.8}\n\nIntercept 827.4 4137.0 1034.3 4592.9\n(278.8)*** (337.8)*** (264.5)*** (330.1)***\n\n[300.0]*** [397.1]*** [295.5]*** [401.7]***\n\n{316.8}** {436.8}*** {292.6}*** {488.9}***\n\nNo. of Children -2222.0 -2401.2\n(132.7)*** (154.4)***\n\n[211.3]*** [233.4]***\n\n{366.3}*** {412.4}***\n\n<u>No. Observations</u> <u>2784</u> <u>2784</u> <u>2435</u> <u>2435</u>\n\nNotes: (1) The data used are CFPS 2010. (2) Robust standard errors in parentheses; standard errors clustered\nat county-level in square brackets; standard errors clustered at province-level in braces; *** p<0.01, **\np<0.05, * p<0.1.\n\nMention: \"CFPS 2010\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The table notes state the regressions are estimated on this survey round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000465:5:0", "occurrence_id": "9f59bd40595963d48361a5ca", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: parental gender bias affects both relative and absolute mobility when diminishing returns\n\n\nand/or complementarity are important.\n\n\nA credible empirical analysis of the above ideas, however, depends critically on the\n\n\nquality of the data available. <sup>8</sup> Most of the existing household surveys use coresidency\n\n\ncriteria to define household membership and thus miss children from the sample in a non\n\nrandom fashion. This results in truncation bias in the estimates of widely used measures of\n\n\nintergenerational mobility such as the intergenerational regression coefficient (IGRC). As\n\n\nreported by Emran et al. (2018), the truncation bias due to coresidency can vary signifi\n\ncantly across gender and countries, making it doubly hazardous to rely on the coresident\n\n\nsample for a comparative study of gender bias in China and India. We take advantage\n\n\nof rich household surveys from China and India that do not suffer from coresidency bias\n\n\nfor our analysis. Since the theory yields interesting predictions regarding the effects of\n\n\ngender bias on educational investment, our empirical analysis estimates both the intergen\n\nerational persistence equation (using IHDS 2012 for India, and CFPS 2016 for China) and\n\n\nthe investment equation (using IHDS 2005 and NSS1995 for India, and CFPS 2010 for\n\n\nChina).\n\n\nThe main conclusions from the empirical analysis are as follows. In India, the inter\n\ngenerational mobility equation is _concave_ irrespective of gender and geographic location,\n\n\nrejecting the almost universally used linear specification in the existing literature. The\n\n\nconcavity suggests that the complementarity between financial investment and parental\n\n\neducation emphasized recently by Becker et al. (2018) may not be important in India.\n\n\nThere are strong diminishing returns to both financial investment and parental direct in\n\nputs in education for girls in rural India. Girls face significantly lower relative and absolute\n\n\nmobility when the father is uneducated, but the gender difference becomes negligible when\n\nthe father is college educated. <sup>9</sup> The relative magnitudes of\n\nMention: \"IHDS 2012\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"This round supplies the Indian intergenerational persistence estimates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000465:5:0", "occurrence_id": "dea24e95ce3eb55fb3ea4a96", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: parental gender bias affects both relative and absolute mobility when diminishing returns\n\n\nand/or complementarity are important.\n\n\nA credible empirical analysis of the above ideas, however, depends critically on the\n\n\nquality of the data available. <sup>8</sup> Most of the existing household surveys use coresidency\n\n\ncriteria to define household membership and thus miss children from the sample in a non\n\nrandom fashion. This results in truncation bias in the estimates of widely used measures of\n\n\nintergenerational mobility such as the intergenerational regression coefficient (IGRC). As\n\n\nreported by Emran et al. (2018), the truncation bias due to coresidency can vary signifi\n\ncantly across gender and countries, making it doubly hazardous to rely on the coresident\n\n\nsample for a comparative study of gender bias in China and India. We take advantage\n\n\nof rich household surveys from China and India that do not suffer from coresidency bias\n\n\nfor our analysis. Since the theory yields interesting predictions regarding the effects of\n\n\ngender bias on educational investment, our empirical analysis estimates both the intergen\n\nerational persistence equation (using IHDS 2012 for India, and CFPS 2016 for China) and\n\n\nthe investment equation (using IHDS 2005 and NSS1995 for India, and CFPS 2010 for\n\n\nChina).\n\n\nThe main conclusions from the empirical analysis are as follows. In India, the inter\n\ngenerational mobility equation is _concave_ irrespective of gender and geographic location,\n\n\nrejecting the almost universally used linear specification in the existing literature. The\n\n\nconcavity suggests that the complementarity between financial investment and parental\n\n\neducation emphasized recently by Becker et al. (2018) may not be important in India.\n\n\nThere are strong diminishing returns to both financial investment and parental direct in\n\nputs in education for girls in rural India. Girls face significantly lower relative and absolute\n\n\nmobility when the father is uneducated, but the gender difference becomes negligible when\n\nthe father is college educated. <sup>9</sup> The relative magnitudes of\n\nMention: \"CFPS 2016\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"This round supplies the Chinese intergenerational persistence estimates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000465:5:0", "occurrence_id": "e599fa9cf3f9c9af0a856824", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: parental gender bias affects both relative and absolute mobility when diminishing returns\n\n\nand/or complementarity are important.\n\n\nA credible empirical analysis of the above ideas, however, depends critically on the\n\n\nquality of the data available. <sup>8</sup> Most of the existing household surveys use coresidency\n\n\ncriteria to define household membership and thus miss children from the sample in a non\n\nrandom fashion. This results in truncation bias in the estimates of widely used measures of\n\n\nintergenerational mobility such as the intergenerational regression coefficient (IGRC). As\n\n\nreported by Emran et al. (2018), the truncation bias due to coresidency can vary signifi\n\ncantly across gender and countries, making it doubly hazardous to rely on the coresident\n\n\nsample for a comparative study of gender bias in China and India. We take advantage\n\n\nof rich household surveys from China and India that do not suffer from coresidency bias\n\n\nfor our analysis. Since the theory yields interesting predictions regarding the effects of\n\n\ngender bias on educational investment, our empirical analysis estimates both the intergen\n\nerational persistence equation (using IHDS 2012 for India, and CFPS 2016 for China) and\n\n\nthe investment equation (using IHDS 2005 and NSS1995 for India, and CFPS 2010 for\n\n\nChina).\n\n\nThe main conclusions from the empirical analysis are as follows. In India, the inter\n\ngenerational mobility equation is _concave_ irrespective of gender and geographic location,\n\n\nrejecting the almost universally used linear specification in the existing literature. The\n\n\nconcavity suggests that the complementarity between financial investment and parental\n\n\neducation emphasized recently by Becker et al. (2018) may not be important in India.\n\n\nThere are strong diminishing returns to both financial investment and parental direct in\n\nputs in education for girls in rural India. Girls face significantly lower relative and absolute\n\n\nmobility when the father is uneducated, but the gender difference becomes negligible when\n\nthe father is college educated. <sup>9</sup> The relative magnitudes of\n\nMention: \"IHDS 2005\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The Indian educational investment equation is estimated on this earlier round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000465:5:0", "occurrence_id": "be6a2a2782ad38ed9afb2c1f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: parental gender bias affects both relative and absolute mobility when diminishing returns\n\n\nand/or complementarity are important.\n\n\nA credible empirical analysis of the above ideas, however, depends critically on the\n\n\nquality of the data available. <sup>8</sup> Most of the existing household surveys use coresidency\n\n\ncriteria to define household membership and thus miss children from the sample in a non\n\nrandom fashion. This results in truncation bias in the estimates of widely used measures of\n\n\nintergenerational mobility such as the intergenerational regression coefficient (IGRC). As\n\n\nreported by Emran et al. (2018), the truncation bias due to coresidency can vary signifi\n\ncantly across gender and countries, making it doubly hazardous to rely on the coresident\n\n\nsample for a comparative study of gender bias in China and India. We take advantage\n\n\nof rich household surveys from China and India that do not suffer from coresidency bias\n\n\nfor our analysis. Since the theory yields interesting predictions regarding the effects of\n\n\ngender bias on educational investment, our empirical analysis estimates both the intergen\n\nerational persistence equation (using IHDS 2012 for India, and CFPS 2016 for China) and\n\n\nthe investment equation (using IHDS 2005 and NSS1995 for India, and CFPS 2010 for\n\n\nChina).\n\n\nThe main conclusions from the empirical analysis are as follows. In India, the inter\n\ngenerational mobility equation is _concave_ irrespective of gender and geographic location,\n\n\nrejecting the almost universally used linear specification in the existing literature. The\n\n\nconcavity suggests that the complementarity between financial investment and parental\n\n\neducation emphasized recently by Becker et al. (2018) may not be important in India.\n\n\nThere are strong diminishing returns to both financial investment and parental direct in\n\nputs in education for girls in rural India. Girls face significantly lower relative and absolute\n\n\nmobility when the father is uneducated, but the gender difference becomes negligible when\n\nthe father is college educated. <sup>9</sup> The relative magnitudes of\n\nMention: \"NSS1995\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The Indian investment equation is also estimated on this national sample survey round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000465:5:0", "occurrence_id": "d5d3c5dd0ef5d82d18815394", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: parental gender bias affects both relative and absolute mobility when diminishing returns\n\n\nand/or complementarity are important.\n\n\nA credible empirical analysis of the above ideas, however, depends critically on the\n\n\nquality of the data available. <sup>8</sup> Most of the existing household surveys use coresidency\n\n\ncriteria to define household membership and thus miss children from the sample in a non\n\nrandom fashion. This results in truncation bias in the estimates of widely used measures of\n\n\nintergenerational mobility such as the intergenerational regression coefficient (IGRC). As\n\n\nreported by Emran et al. (2018), the truncation bias due to coresidency can vary signifi\n\ncantly across gender and countries, making it doubly hazardous to rely on the coresident\n\n\nsample for a comparative study of gender bias in China and India. We take advantage\n\n\nof rich household surveys from China and India that do not suffer from coresidency bias\n\n\nfor our analysis. Since the theory yields interesting predictions regarding the effects of\n\n\ngender bias on educational investment, our empirical analysis estimates both the intergen\n\nerational persistence equation (using IHDS 2012 for India, and CFPS 2016 for China) and\n\n\nthe investment equation (using IHDS 2005 and NSS1995 for India, and CFPS 2010 for\n\n\nChina).\n\n\nThe main conclusions from the empirical analysis are as follows. In India, the inter\n\ngenerational mobility equation is _concave_ irrespective of gender and geographic location,\n\n\nrejecting the almost universally used linear specification in the existing literature. The\n\n\nconcavity suggests that the complementarity between financial investment and parental\n\n\neducation emphasized recently by Becker et al. (2018) may not be important in India.\n\n\nThere are strong diminishing returns to both financial investment and parental direct in\n\nputs in education for girls in rural India. Girls face significantly lower relative and absolute\n\n\nmobility when the father is uneducated, but the gender difference becomes negligible when\n\nthe father is college educated. <sup>9</sup> The relative magnitudes of\n\nMention: \"CFPS 2010\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The Chinese investment equation is estimated on this baseline round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002389:11:0", "occurrence_id": "6b9a78da76720e424c81bbb1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Popular_ (CINEP) _Noche y Niebla_ online database of human rights and political violence events,\nfollowing the criteria and definition of the CNMH. The data exclude attacks perpetrated by\ncriminal gangs or by the state’s armed forces, but nonetheless, these attacks in Colombia since the\nmid-1990s have been almost exclusively a strategy used by guerrilla groups.\n\n\nThe variation across time and space of violent attacks against civilians in the country is depicted\nin Figures 1 and 2. In the first, darker shades indicate a higher number of attacks through the period\nof analysis (1999-2016). Except for the south-east of the country, the most remote and low-density\npopulated region, <sup>25</sup> [^25: This region is also mostly dominated by the Amazon rainforest. DHS are representative of 99% of the country and the\nuncovered areas are disperse rural areas in this part of the country.] the occurrence of attacks is scattered around the country. The map also shows\ntheir rarity, as most municipalities did not experience any attack in 16 years. Figure 2 shows the\nevolution of attacks by month of occurrence. There are more occurrences in the mid-2000s and\nless towards the end of that decade, coinciding with the pattern of conflict intensity in the country.\n\n\nIn the municipalities included in DHS (487 across the three survey rounds), there were 49 attacks\nin the period of study occurring in 29 different municipalities; that is, approximately 6% of the\nmunicipalities in the sample experienced at least one attack between February 1999 and March\n2016. The recurrence of attacks in a municipality is low: ten of them experienced more than one\nattack in the period, and Tumaco, in the Pacific south-west, and Bogotá, the capital, had the highest\noccurrence with six and five attacks respectively. Not all events resulted in victims; in 23 attacks\nthere were no fatalities and in 17 there were no wounded people. This reinforces the idea that the\nfear of potential devastation can be larger than the immediate effect on human lives.\n\nMention: \"DHS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The count of municipalities exposed to attacks is expressed over the municipalities covered by the survey's three rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002389:17:0", "occurrence_id": "410bff9d3144c35178d3f4dc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **C.** **Mortality, fertility and migration**\n\n\nThe full effect of violence can be underestimated if the shock increases fetal deaths (survival\nselection). Using the DHS’s pregnancy calendar, it is possible to distinguish between live births\nand terminations <sup>49</sup> [^49: A full pregnancy history distinguishing miscarriage from stillbirths, abortions and ectopic pregnancies is only available in the\nlatest round of the survey.] and analyze the effect on this outcome. Despite known difficulties in capturing\ninformation about fetal mortality, we do not expect this problem to be more pronounced in\nColombia than elsewhere, especially poorer countries where more births and deaths occur outside\nof the health system. Only pregnancies within five years from the survey are used, a standard\nprocedure used to alleviate recall error. The models are estimated as in Equation 1 but without\ncontrolling for gender, multiple birth and skilled delivery, as this information is not available for\nnon-live births. <sup>50</sup> [^50: Since the day of the pregnancy end is not known in the calendar data, the date of conception is not adjusted for the time of the\nmonth and thus can be subject to greater measurement error than the date of conception of the live births sample.] Because terminations tend to happen early in the pregnancy, <sup>51</sup> [^51: In the sample, terminated pregnancies last 2.4 months on average, 91.3% of pregnancies that reach the seventh month are live\nbirths, and 99.6% at the ninth month.] I interact the\nindicators of exposure to violence in the second and third trimesters with a dummy for whether the\npregnancy reached those trimesters, and check the results with the alternative method of using the\nexpected full-length trimesters. Because the number of cases per woman is much larger than for\nlive births, I can further estimate a mother fixed-effects model.\n\n\nThere are 39,500 thousand pregnancies from non-migrant women\n\nMention: \"DHS’s pregnancy calendar\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The calendar module is used to separate live births from terminations so that fetal loss is not missed.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002622:41:0", "occurrence_id": "6e0d41e74c308c3b138b16ea", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Annex: Secured Transaction Regimes for Movables in the EU 8**\n\n|Hungary|Yes|OVERVIEW<br>• Has a modern and efficient regime for secured transactions (EBRD<br>2003 Transition Report).<br>• The Charge Registry is kept at the Hungarian National Chamber of<br>Public Notaries.<br>• The registry is accessible for anybody to inspect.<br>• The registry is centralized and computerized.<br>• The secured transaction system is defined in the Hungarian Civil Code<br>of 1959. Modifications of the code in 1996 amended the secured<br>transaction system and have led to an increased use of security over<br>movables. The law was further updated in September 2001.<br>TYPES OF PLEDGES<br>• The charged asset may be any tangible object, any transferable right<br>or claim.<br>• A charge on rights and receivables may be created by agreement.<br>• A floating charge may be created on the whole of a business associa-<br>tion without legal entity or on a part of it that operates as a separate<br>economic unit without having to define the objects, rights and receiv-<br>ables constituting the asset, by concluding the charge agreement in<br>form of a notary instrument and registering the charge in the charge<br>registry.|\n|---|---|---|\n|**Latvia**|**Yes**|**OVERVIEW**<br>• Commercial pledges are registered in the Register of Commercial<br>Pledges, a database on commercial pledges.<br>• The register is fully accessible to the public.<br>• The system is centralized and electronic.<br>**TYPES OF PL\n\nMention: \"Charge Registry\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The annex describes the registry's custodian, public accessibility and centralised computerisation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002622:41:0", "occurrence_id": "eea5ca8be0d37ada9099dea5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Annex: Secured Transaction Regimes for Movables in the EU 8**\n\n|Hungary|Yes|OVERVIEW<br>• Has a modern and efficient regime for secured transactions (EBRD<br>2003 Transition Report).<br>• The Charge Registry is kept at the Hungarian National Chamber of<br>Public Notaries.<br>• The registry is accessible for anybody to inspect.<br>• The registry is centralized and computerized.<br>• The secured transaction system is defined in the Hungarian Civil Code<br>of 1959. Modifications of the code in 1996 amended the secured<br>transaction system and have led to an increased use of security over<br>movables. The law was further updated in September 2001.<br>TYPES OF PLEDGES<br>• The charged asset may be any tangible object, any transferable right<br>or claim.<br>• A charge on rights and receivables may be created by agreement.<br>• A floating charge may be created on the whole of a business associa-<br>tion without legal entity or on a part of it that operates as a separate<br>economic unit without having to define the objects, rights and receiv-<br>ables constituting the asset, by concluding the charge agreement in<br>form of a notary instrument and registering the charge in the charge<br>registry.|\n|---|---|---|\n|**Latvia**|**Yes**|**OVERVIEW**<br>• Commercial pledges are registered in the Register of Commercial<br>Pledges, a database on commercial pledges.<br>• The register is fully accessible to the public.<br>• The system is centralized and electronic.<br>**TYPES OF PL\n\nMention: \"database on commercial pledges\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Latvia's register of commercial pledges is described as a publicly accessible electronic database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002622:43:0", "occurrence_id": "e636174b815a3e943cb8b8f5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Annex: Secured Transaction Regimes for Movables in the EU 8**\n\n\n\n\n\n\n\n|Slovak<br>Republic|Yes|OVERVIEW<br>• Has a modern and efficient regime for secured transactions (EBRD<br>2003 Transition Report). System considered one of the best in Europe.<br>• Registration is made in the new Charges Register operated by the<br>Slovak Chamber of Notaries.<br>• The register is maintained in a single electronic database. Registration<br>is made instantly via a terminal in a notary’s office.<br>TYPES OF PLEDGES<br>• Under the new law, any legal or physical person can give a charge<br>over its assets (objects, residential or non-residential property, rights<br>or other assets) in favor of any creditor, to secure his own or another<br>person’s debt. The charge can be taken over any single or collective<br>object, right or asset that is capable of being transferred. This may<br>include:<br>• movable assets,<br>• land and buildings,<br>• rights and receivables,<br>• or even a group of assets, which is constantly changing, such as<br>inventory, trade receivables, stocks of raw materials or an entire<br>enterprise.<br>• It may also include assets to be acquired in the future. The fruits<br>of the collateral are included until they become separated unless<br>otherwise agreed.|\n|---|---|---|\n|**Slovenia**|**No**|**OVERVIEW**<br>• This area is currently under reform. There is no public register to allow<br>tracking of priorities and to alert\n\nMention: \"Charges Register\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Registration in a single electronic database operated by the notaries' chamber is described as instant.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003650:8:1", "occurrence_id": "dba4472c9285ae9eb673b786", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  to innovate.\n\nThe third National Survey of Technological Innovation in Firms was conducted in 2001\nby Chile’s Executive Secretary of Technological Innovation Program and the National\nInstitute of Statistics. It contains data for the years 2001 and predictions for 2002. The\nsurvey had 896 respondents representing a weighted total population of 4,932\nestablishments. The present work focuses on the manufacturing sector, with a total of 706\nrespondents and 687 valid observations weighting to a total of 4,570 observations. <sup>13</sup> The\nmain concern regarding this survey is that there is no information regarding the response\nrate. In this way, weighting the data might risk strengthening the underlying bias\n\n\n12 For more information on the properties of MLE, cfr, among others, Johnston and DiNardo (1997).\n13 The nature of the data is mainly qualitative, since the firms are often asked questions with a yes/no\nanswer and only a few variables are of a strictly quantitative nature (these include the data on innovation\nand R&D expenditure). To overcome the problem of the qualitative nature of the data, a metric scale has\nbeen introduced, with 0 denoting a negative answer (meaning either “no”, “none”, or “0”) and where a\nscale of 1 to 4 denotes a positive answer, with 1 representing the least positive answer and 4 the most\npositive. The data were collected at the firm-level. This means that if a firm has more than one\nestablishment the questionnaire was sent to the central management, in order to capture all the innovation\nactivities within the firm. This methodology is in line with the procedures adopted by Eurostat for\ncollecting the Community Innovation Surveys.\n\n\n7\n\nMention: \"National Survey of Technological Innovation in Firms\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The third round provides the innovation activity data for 2001 with predictions for 2002 across 896 firms.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003650:8:1", "occurrence_id": "d9b2431c734d01c8e877cb76", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  to innovate.\n\nThe third National Survey of Technological Innovation in Firms was conducted in 2001\nby Chile’s Executive Secretary of Technological Innovation Program and the National\nInstitute of Statistics. It contains data for the years 2001 and predictions for 2002. The\nsurvey had 896 respondents representing a weighted total population of 4,932\nestablishments. The present work focuses on the manufacturing sector, with a total of 706\nrespondents and 687 valid observations weighting to a total of 4,570 observations. <sup>13</sup> The\nmain concern regarding this survey is that there is no information regarding the response\nrate. In this way, weighting the data might risk strengthening the underlying bias\n\n\n12 For more information on the properties of MLE, cfr, among others, Johnston and DiNardo (1997).\n13 The nature of the data is mainly qualitative, since the firms are often asked questions with a yes/no\nanswer and only a few variables are of a strictly quantitative nature (these include the data on innovation\nand R&D expenditure). To overcome the problem of the qualitative nature of the data, a metric scale has\nbeen introduced, with 0 denoting a negative answer (meaning either “no”, “none”, or “0”) and where a\nscale of 1 to 4 denotes a positive answer, with 1 representing the least positive answer and 4 the most\npositive. The data were collected at the firm-level. This means that if a firm has more than one\nestablishment the questionnaire was sent to the central management, in order to capture all the innovation\nactivities within the firm. This methodology is in line with the procedures adopted by Eurostat for\ncollecting the Community Innovation Surveys.\n\n\n7\n\nMention: \"Community Innovation Surveys\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"The survey's single-questionnaire-per-firm design is justified by reference to how the European instrument is collected.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003650:9:1", "occurrence_id": "baa6cf8a084ca63d1df420f5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: sup>15</sup>\n\n - A new technology for production\n\n - A product that is known to the market but new to the firm\n\n - A product that is new both to the market and to the firm\n\nProduct innovations are aimed at increasing market share or to open a new market space\nfor the firm. Process innovations, on the other side, are developed to improve the existing\nvalue chain of the firm. The improved process should lower the cost of production and\ntherefore make the product more profitable, reducing the break-even point, either by\ncompeting on price or by increasing sales. Though not as radical in their effect as product\ninnovation, process innovations also provide a substantial economic benefit to the firm.\n\nA firm is considered here as innovative in terms of **process** if it has introduced at least\none of the following:\n\n - A new technology that the firm deems important to improve existing processes.\n\n\n14 There is no use of weights in the survey.\n15 In the last 3 years in the case of the Chilean survey.\n\n\n8\n\nMention: \"Chilean survey\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"A footnote notes only that the reference window for novelty is three years in this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005981:25:0", "occurrence_id": "4b7f3bdbf098ed8208313bb8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: [INSERT TABLES 2a AND 2b AROUND HERE]\n\n\nOnce the estimation is complete, we use the estimated parameter values from the equa\n\ntions above to predict the missing values in the bilateral migration matrices for both years,\n\n\ngenders and educations levels.\n\n\nOne critical issue is that of zero observations where two basic options are available.\n\n\nThe first would be to use PPML estimates. It is worth noting, however, that our predicted\n\n\nnumbers are almost never integers and so our estimates should be interpreted as the expected\n\n\nnumbers of migrants. Although PPML predictions cannot produce true zeroes, many of our\n\n\nestimates are extremely close to zero, such that we could postulate having predicted a zero,\n\n\ne.g. if the expected number of migrants generated by the PPML model is smaller than\n\n\none (or indeed below any other arbitrarily small threshold). The second option which we\n\n\nimplement in this paper, instead involves splitting the aggregate figures provided by OPSW\n\n\nby using our PPML regressions to estimate the shares of males and females and the skilled\n\n\nand unskilled. Since we have access to the totals, predicting these shares is no doubt more\n\n\naccurate than attempting to impute unrestricted totals. In following this methodology, we\n\n\nnecessarily take all zero observations from OPSW as opposed to imputing them with our\n\n\nPPML estimation. Indeed, there are a very few cases where OPSW is greater than zero but\n\n\nwhere a PPML estimate would be close is zero, but these observations constitute are less\n\n\nthan one percent of the total.\n\n\nBefore analyzing the predictions and their implications for the global migration patterns\n\n\nby skill level, we evaluate the out-of-sample performance of our predictions. To do so, we\n\n\n24\n\nMention: \"OPSW\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Aggregate migrant totals from this source are split by gender and skill using the paper's own regressions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005981:60:2", "occurrence_id": "08c4df2da8c554e71fdcd8b7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: br>Macedonia (1994-2002)<br>Malaysia (n.a.-2000)<br>Mali (n.a.-1998)<br>Malta (1995-2005)<br>Mauritius (n.a.-2000)<br>Mongolia (n.a.-2000)<br>Morocco (n.a.-2004)<br>Nepal (n.a.-2001)<br>Nicaragua (n.a.-2005)<br>Oman (1990-2000)<br>Panama (n.a.-2000)<br>Paraguay (n.a.-2002)<br>Peru (n.a.-2007)<br>Philippines (1990-2000)<br>Qatar (1990-2000)|IPUMS Internationalb<br>DIOC-E database (OECD)d<br>Labor Force Surveyc<br>IPUMS Internationalb<br>DIOC-E database (OECD)d<br>IPUMS Internationalb<br>IPUMS Internationalb<br>National Statistical Institute<br>DIOC-E database (OECD)d<br>Census and Statistics Dep.<br>IPUMS Internationalb<br>I.N. Estadistica y Censos<br>Institut National de Stat.<br>Central Bureau of Statistics<br>DIOC-E database (OECD)d<br>Cyprus Statistics<br>United Nations CEPALa<br>DIOC-E database (OECD)d<br>DIOC-E database (OECD)d<br>DIOC-E database (OECD)d<br>DIOC-E database (OECD)d<br>D\n\nMention: \"DIOC-E database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"For several countries the table lists this database as the origin of the migrant-stock observations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006956:24:0", "occurrence_id": "372b1ab1ff2114b6cf57a125", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 2: Minerals prices trends\n\n\nSource: authors’ construction (based on USGS data).\n\n\nFigure 3: Average lights Gini trends in Africa, 2001–2012\n\n\n\n\n\n\n\n\n\n23\n\nMention: \"USGS data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The mineral price series plotted in the figure is constructed from this source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006956:8:0", "occurrence_id": "d4dbfcfd24e81375306cd8b1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **3** **Data and district inequality estimation**\n\n\n**3.1** **Data**\n\n\nWe combine an array of data sources to construct a data set with the relevant important\nvariables for estimating the effects of mining activities on district inequality: the National\nOceanic and Atmospheric Administration, National Geophysical Data Center (NOAANGDC) <sup>9</sup> for data on night-time lights intensity; the United States Geological Survey (USGS) <sup>10</sup>\nfor data on the spatial location of mineral deposits and mining activities in Africa and historical\nUS mineral prices; the United States Global Land Cover Facility (GLCF) <sup>11</sup> for the Normalized\nDifference Vegetation Index (NDVI)—a proxy for agricultural productivity. We also use\nTropical Applications of Meteorology using Satellite Data (TAMSAT) <sup>12</sup> to extract data on\nrainfall—a proxy for climatic shocks. Finally, we extract population data from the Gridded\nPopulation of the World (GPW v4). <sup>13</sup> Since our units of analysis are districts, the combined\ndata set constitutes three clustered levels—districts, regions, and countries.\n\n\nNOAA-NGDC data include annual time-series night-time lights intensity data, which are\nglobally recorded across countries daily from 20:30 to 22:00 local time by a satellite orbiting\nthe earth. The use of this data set follows recent literature that attempts to circumvent the\nabsence of reliable and consistent subnational data (for a detailed discussion, see Chen and\nNordhaus 2011; Henderson et al. 2012), particularly in Africa. <sup>14</sup> These data come in three main\nformats: cloud-free, average visible, and stable light composites. We use the stable light\ncomposites, which\n\nMention: \"Normalized\nDifference Vegetation Index\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The vegetation index is taken from the land cover facility as a proxy for agricultural productivity.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006956:8:0", "occurrence_id": "200fdf90d8655f596b93bcab", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **3** **Data and district inequality estimation**\n\n\n**3.1** **Data**\n\n\nWe combine an array of data sources to construct a data set with the relevant important\nvariables for estimating the effects of mining activities on district inequality: the National\nOceanic and Atmospheric Administration, National Geophysical Data Center (NOAANGDC) <sup>9</sup> for data on night-time lights intensity; the United States Geological Survey (USGS) <sup>10</sup>\nfor data on the spatial location of mineral deposits and mining activities in Africa and historical\nUS mineral prices; the United States Global Land Cover Facility (GLCF) <sup>11</sup> for the Normalized\nDifference Vegetation Index (NDVI)—a proxy for agricultural productivity. We also use\nTropical Applications of Meteorology using Satellite Data (TAMSAT) <sup>12</sup> to extract data on\nrainfall—a proxy for climatic shocks. Finally, we extract population data from the Gridded\nPopulation of the World (GPW v4). <sup>13</sup> Since our units of analysis are districts, the combined\ndata set constitutes three clustered levels—districts, regions, and countries.\n\n\nNOAA-NGDC data include annual time-series night-time lights intensity data, which are\nglobally recorded across countries daily from 20:30 to 22:00 local time by a satellite orbiting\nthe earth. The use of this data set follows recent literature that attempts to circumvent the\nabsence of reliable and consistent subnational data (for a detailed discussion, see Chen and\nNordhaus 2011; Henderson et al. 2012), particularly in Africa. <sup>14</sup> These data come in three main\nformats: cloud-free, average visible, and stable light composites. We use the stable light\ncomposites, which\n\nMention: \"Tropical Applications of Meteorology using Satellite Data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Rainfall is extracted from this satellite rainfall product to proxy climatic shocks.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006956:8:0", "occurrence_id": "cd1c393dcede6e57b799b780", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **3** **Data and district inequality estimation**\n\n\n**3.1** **Data**\n\n\nWe combine an array of data sources to construct a data set with the relevant important\nvariables for estimating the effects of mining activities on district inequality: the National\nOceanic and Atmospheric Administration, National Geophysical Data Center (NOAANGDC) <sup>9</sup> for data on night-time lights intensity; the United States Geological Survey (USGS) <sup>10</sup>\nfor data on the spatial location of mineral deposits and mining activities in Africa and historical\nUS mineral prices; the United States Global Land Cover Facility (GLCF) <sup>11</sup> for the Normalized\nDifference Vegetation Index (NDVI)—a proxy for agricultural productivity. We also use\nTropical Applications of Meteorology using Satellite Data (TAMSAT) <sup>12</sup> to extract data on\nrainfall—a proxy for climatic shocks. Finally, we extract population data from the Gridded\nPopulation of the World (GPW v4). <sup>13</sup> Since our units of analysis are districts, the combined\ndata set constitutes three clustered levels—districts, regions, and countries.\n\n\nNOAA-NGDC data include annual time-series night-time lights intensity data, which are\nglobally recorded across countries daily from 20:30 to 22:00 local time by a satellite orbiting\nthe earth. The use of this data set follows recent literature that attempts to circumvent the\nabsence of reliable and consistent subnational data (for a detailed discussion, see Chen and\nNordhaus 2011; Henderson et al. 2012), particularly in Africa. <sup>14</sup> These data come in three main\nformats: cloud-free, average visible, and stable light composites. We use the stable light\ncomposites, which\n\nMention: \"Gridded\nPopulation of the World\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Population is extracted from this gridded product to build the district-level panel.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006956:8:0", "occurrence_id": "ffd5842dbc8376981b940ad2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **3** **Data and district inequality estimation**\n\n\n**3.1** **Data**\n\n\nWe combine an array of data sources to construct a data set with the relevant important\nvariables for estimating the effects of mining activities on district inequality: the National\nOceanic and Atmospheric Administration, National Geophysical Data Center (NOAANGDC) <sup>9</sup> for data on night-time lights intensity; the United States Geological Survey (USGS) <sup>10</sup>\nfor data on the spatial location of mineral deposits and mining activities in Africa and historical\nUS mineral prices; the United States Global Land Cover Facility (GLCF) <sup>11</sup> for the Normalized\nDifference Vegetation Index (NDVI)—a proxy for agricultural productivity. We also use\nTropical Applications of Meteorology using Satellite Data (TAMSAT) <sup>12</sup> to extract data on\nrainfall—a proxy for climatic shocks. Finally, we extract population data from the Gridded\nPopulation of the World (GPW v4). <sup>13</sup> Since our units of analysis are districts, the combined\ndata set constitutes three clustered levels—districts, regions, and countries.\n\n\nNOAA-NGDC data include annual time-series night-time lights intensity data, which are\nglobally recorded across countries daily from 20:30 to 22:00 local time by a satellite orbiting\nthe earth. The use of this data set follows recent literature that attempts to circumvent the\nabsence of reliable and consistent subnational data (for a detailed discussion, see Chen and\nNordhaus 2011; Henderson et al. 2012), particularly in Africa. <sup>14</sup> These data come in three main\nformats: cloud-free, average visible, and stable light composites. We use the stable light\ncomposites, which\n\nMention: \"NOAA-NGDC data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The night-time lights series used for inequality estimation comes from this archive.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006956:8:0", "occurrence_id": "ac4b9183a7c166a2fc293822", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **3** **Data and district inequality estimation**\n\n\n**3.1** **Data**\n\n\nWe combine an array of data sources to construct a data set with the relevant important\nvariables for estimating the effects of mining activities on district inequality: the National\nOceanic and Atmospheric Administration, National Geophysical Data Center (NOAANGDC) <sup>9</sup> for data on night-time lights intensity; the United States Geological Survey (USGS) <sup>10</sup>\nfor data on the spatial location of mineral deposits and mining activities in Africa and historical\nUS mineral prices; the United States Global Land Cover Facility (GLCF) <sup>11</sup> for the Normalized\nDifference Vegetation Index (NDVI)—a proxy for agricultural productivity. We also use\nTropical Applications of Meteorology using Satellite Data (TAMSAT) <sup>12</sup> to extract data on\nrainfall—a proxy for climatic shocks. Finally, we extract population data from the Gridded\nPopulation of the World (GPW v4). <sup>13</sup> Since our units of analysis are districts, the combined\ndata set constitutes three clustered levels—districts, regions, and countries.\n\n\nNOAA-NGDC data include annual time-series night-time lights intensity data, which are\nglobally recorded across countries daily from 20:30 to 22:00 local time by a satellite orbiting\nthe earth. The use of this data set follows recent literature that attempts to circumvent the\nabsence of reliable and consistent subnational data (for a detailed discussion, see Chen and\nNordhaus 2011; Henderson et al. 2012), particularly in Africa. <sup>14</sup> These data come in three main\nformats: cloud-free, average visible, and stable light composites. We use the stable light\ncomposites, which\n\nMention: \"annual time-series night-time lights intensity data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The passage describes the daily satellite observation window so readers can judge the lights measure's comparability.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007100:26:0", "occurrence_id": "f895157098f31fde4dfe9a7e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: that income persistence is lower in the developing world. However, this comparison is difficult to interpret\n\n\ngiven the differences between the two databases. Specifically, a possible interpretation is that our OLS\n\n\nestimates of -, as well as the MSE-minimizing combination of the OLS and bias-corrected OLS estimates,\n\n\nhave greater small-sample downward bias in the PovcalNet sample where the available time series is\n\n\nmuch shorter than in the WID sample.\n\n\nWhen we correlate our estimates of income persistence with per capita GDP across countries, we\n\n\nfind a negative correlation within the WID and PovcalNet samples – higher persistence is associated with\n\n\nlower incomes. The reason for this is not obvious, and it is unlikely to be mechanical. Note that if countries\n\n\nwere identical in all parameters but the income persistence parameter, then lower persistence would be\n\n\nassociated with lower average income (since lower persistence implies both a lower mean and lower\n\n\ncross-sectional variance of log income in the steady state, and therefore lower mean income given our\n\n\nlognormality assumption). This seems to suggest that countries with lower income persistence differ from\n\n\ncountries with higher persistence in other ways that benefit their income growth.\n\n\nOur approach to approximating individual-level distribution dynamics with aggregate data makes\n\n\nit possible for the first time to compare anonymous and non-anonymous growth rates in a wide cross\n\n\nsection of countries. We show that the anonymous growth rate of the bottom 40 percent (corresponding\n\n\nto the World Bank’s definition of ”shared prosperity”) understates the growth rate of those initially in the\n\n\nbottom 40 percent by 3 to 4 percentage points per year. This implies that the anonymous growth rate is\n\n\na poor approximation to the success of those initially in the bottom of the income distribution, and is not\n\n\na good tool for tracking the effects of policy interventions that were targeted to the initially-poor. The\n\n\nexact opposite holds true when tracking growth at the top end of the income distribution. Growth of\n\nMention: \"PovcalNet\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Persistence estimates are compared across this and another database, with shorter series raising small-sample bias.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007100:23:0", "occurrence_id": "7704fdb082877d1c0409a3b7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: individuals at each percentile of the initial income distribution. The bold line is the lognormal version of\n\nthis curve, i.e. Equation (11), but using the estimate of ���� based on the PSID micro data. Interestingly,\n\n\nthis lognormal version of the growth incidence curve tracks the non-parametric smoothed estimate of the\n\n\ngrowth incidence curve fairly closely, suggesting that our assumption of lognormality provides a\n\n\nreasonable approximation to the true income distribution dynamics in this particular setting. As expected,\n\n\nthe non-anonymous growth incidence curves indicate much higher (lower) growth for those initially in the\n\n\nbottom (top) half of the income distribution. Finally, we plot the lower and upper bounds on the growth\n\nincidence curve based on our bounds on ���� calculated using only aggregate data. These bounds include\n\n\nthe two versions of the actual non-anonymous growth incidence curve based on micro data almost\n\n\neverywhere. It is also useful to note that they exclude the anonymous growth incidence curve everywhere\n\n\nexcept for very close to the middle of the income distribution where the anonymous and non-anonymous\n\n\ngrowth incidence curves are very close to each other.\n\n\nWe next consider growth rates of group average incomes, such as the bottom 40 or the top 10\n\n\npercent of the income distribution. The anonymous and non-anonymous growth rates of group average\n\n\nincomes are simply weighted averages of the corresponding anonymous and non-anonymous growth\n\n\nincidence curves, as summarized in the following proposition:\n\n\n_<u>Proposition 4: Let</u>_ 0 ≤�< �≤1 _denote percentiles of the income distribution. The anonymous and non-_\n\n_anonymous growth rate of group average incomes of those between the_ - <sup>��</sup> _and_ - <sup>��</sup> _percentiles of the_\n\n\n_income distribution are:_\n\n\n\n\n\n\n\n(13) ���(�, �) ≈����(�)����\n\nMention: \"PSID micro data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"A lognormal growth incidence curve built from this panel is used to check the parametric approximation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000002:25:1", "occurrence_id": "08fe860d44e391c11c998f68", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Environmental Policies on Patterns of World\n\nTrade: An Empirical Test, _Kyklos_, Vol. 43, No. 2, pp. 191-209.\nUN COMTRADE (2003), Trade Database of the World Bank and UNCTAD, Online Access:\n\nhttp://wits.worldbank.org.\nVan Beers, Cees and Jeroen van den Bergh (1996), An Overview of Methodological Approaches\n\nin the Analysis of Trade and Environment, _Journal of World Trade_, Vol. 30, No. 1,\npp. 143-167\nVan Beers, Cees and Jeroen van den Bergh (1997), An Empirical Multi-Country Analysis of the\n\nImpact of Environmental Regulations on Foreign Trade Flows, _Kyklos_, Vol. 50, No. 1,\npp. 29-46.\nVogel, David (2004), Trade and Environment in the Global Economy: Contrasting European and\n\nAmerican Perspectives, _Prepared for Green Giants? Environmental Policy of the_\n_United States and the European Union_, in Norman Vig and Michael Faure (eds.)\nCambridge, MA: MIT Press (forthcoming).\nWhite, Halbert (1980), A Heteroskedasticity-Consistent Covariance Matrix Estimator and a\n\nDirect Test for Heteroskedasticity, _Econometrica_, Vol. 48, No. 4, pp. 817-838.\nWilson, John, Tsunehiro Otsuki and Mirvat Sewadeh (2002), _Dirty Exports and Environmental_\n\n_Regulation: Do Standards Matter to Trade?_ World Bank Policy Research Working\nPaper 2806.\nWorld Bank (2001), _Global Economic Prospects and Developing Countries 2001_, Washington,\n\nDC: World Bank.\nWorld Bank (2003), World Development Indicators 2003, Data on CD-ROM.\n\n\n26\n\nMention: \"World Development Indicators 2003\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"The indicators volume appears in the reference list as a cited data publication.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000007:7:0", "occurrence_id": "fdfca2615fb7920c0efd01db", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: terms in at least one year during 2002-07. <sup>7</sup> EMDE equity market valuations, as reflected by\nthe MSCI index, more than quadrupled during 2002-07; EMDE bond spreads, as captured by\nthe EMBI, and sovereign CDS spreads in major EMDEs more than halved between January\n2005 and June 2007. Benign financing conditions encouraged strong investment growth (12\npercent a year, on average, during 2002-07) in EMDEs. More than one-quarter of EMDEs\nwitnessed an investment surge in at least one year during 2002-07. <sup>8</sup> Most of these\ninvestment surges were fueled by credit booms (Ohnsorge and Yu 2016). Meanwhile, private\nconsumption growth remained robust (6.3 percent a year, on average, during 2002-07) as\nhousehold incomes grew and employment opportunities expanded.\n\nThe decline in global poverty was faster than expected prior to the 2009 global recession. In\n\n\n\nhousehold incomes grew and employment opportunities expanded.\n\nThe decline in global poverty was faster than expected prior to the 2009 global recession. In\nSeptember 2000, the international community adopted the Millennium Development Goals.\nAmong them was the goal to halve the share of the global population living on less than $1.25\na day between 1990 and 2015. As a result of rapid EMDE growth, the goal was achieved five\nyears earlier than targeted, in 2010. China’s rapid expansion accounted for about three-fifths\nof this decline in global poverty while the remainder mostly reflected progress in Brazil,\nIndia, Indonesia, and Pakistan (World Bank 2016b). The number of low-income countries\n(LICs) declined to 49 in 2007 from 64 in 2001. Rapid LIC growth (4.6 percent a year, on\naverage, during 2001-07), supported by several factors—such as the commodity price boom,\ndebt relief,\n\nMention: \"MSCI index\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Equity market valuations in emerging economies are measured with this index to show they quadrupled before the crisis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000014:22:0", "occurrence_id": "2b69657b31514f2ae6bde275", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: loans reported in the DRS over the period 2014-17. We assume that the marginal growth rate due to BRI\ninvestment is equal to the average of the marginal growth rates estimated through the LTGM and the\nGMM estimates in section IV (the comparison between the growth rates under the two estimates in\nreported in Annex VI). <sup>36</sup> While we do not make any explicit assumption related to the primary and overall\nfiscal deficit, the cumulative increase in the PPG debt over the 5-year projection period due to BRI-related\nborrowing, could be interpreted as the result of an equal increase in primary expenditure of the general\ngovernment and the increase in interest payment due to the additional BRI-related borrowing. <sup>37</sup>\n\n\nDebt financed investment is expected to help sustain or achieve higher growth, but in all 15 LIDCs and 10\nEMs recipient of BRI investment the growth required to stabilize the debt ratio is higher than the\nestimated growth (Figure 13). Out of 43 countries, only 8 EMs are expected to increase growth, as a result\nof BRI, to fully compensate for the estimated increased in BRI debt financing over the period 2019-2023.\nNote that this result holds under both _A1_ and A2, with no countries ending in the undefined area. We\nhave also compared the growth rate needed to stabilize the PPG debt-to-GDP ratio at end-2016 assuming\nthat the PPG debt would increase only by the BRI financing. <sup>38</sup> Also under this exercise, all countries with\n_A2>0_ ( _A1<0)_ have a required growth rate which is greater (smaller) than the estimated average growth\nrate.\n\n\nFigure 13. Growth Gap\n(In percent of GDP, by debt vulnerability)\n\nLIDCs EMs\n\n\nSources: WEO, and LIC DSF DSAs. Sources: WE\n\nMention: \"DRS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Loan volumes over the study window are drawn from this registry to parameterise the growth assumption.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000016:4:1", "occurrence_id": "b37a9c004839765bbf6156dc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  paper.\n\n\nOur starting point is to rely on an exhaustive description of the provisions included in PTAs based on a new\n\n\ndatabase on the content of PTAs that has been collected by the World Bank (Mattoo, Rocha & Ruta 2020).\n\n\nWe use information on all policy areas (except tariffs) encompassing objectives, substantive commitments and\n\n\nenforcement procedures present in legal texts and available annexes of the 278 PTAs in force and notified to\n\n\nthe WTO up to 2018. Examples of policy areas covered in the database include competition policy, State\n\nowned Enterprises (SOEs), subsidies, public procurement, Technical Barriers to Trade (TBTs) and Sanitary\n\n\nand Phytosanitary Standards (SPS), labor rights, or environment.\n\n\nSuch very rich set of information has to be collapsed in broad categories of PTAs in order to be tractable in\n\n\na GE framework. The first step of the analysis is therefore to define statistically significant groupings. We rely\n\n\non a clustering approach to identify groups of trade agreements based on the provisions’ content. In doing so,\n\n\nwe opt for the iterative “kmean++” algorithm developed by Arthur & Vassilvitskii (2007) which ensures greater\n\n\naccuracy by randomizing starting points at each replication. Given the underlying distribution of provisions in\n\n\neach of the 18 policy areas in the data, the silhouette width criterium (which evaluates cluster fit on within\n\ngroup cohesion and between-group separation) recommends 3 clusters. In words, with 3 groups the distance\n\n\nbetween observations within each group of PTAs is minimized and the distance between groups is maximized.\n\n\nThe underlying assumption is that each group of PTAs has a similar content of provisions by policy subject, and\n\n\nthus should impact differently the trade costs between members of the agreement and between the latter and\n\n\n3\n\nMention: \"database on the content of PTAs\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Provision-level coding of preferential trade agreements is taken from this newly collected database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000018:7:0", "occurrence_id": "a782c422a63d64b56d16b495", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 80%\n70%\n60%\n50%\n40%\n30%\n20%\n10%\n0%\n\n\n\n_Figure 2. Globally, Children Are More Likely Than People Ages 15 or More to Be Poor_\n\n\n\n0-14 15 and over\n\n\nProportion of those living in Extreme Poor households\n\nProportion of Total Population\n\n\n\n\n\n_Source:_ Calculations based on data of the GMD.\n\nA gap in poverty rates by sex has emerged and is concentrated in South Asia and Sub-Saharan Africa\n\n\nThe gap in poverty by sex is nuanced by age. Differences in poverty rates by sex are concentrated among\n\nchildren and, to a larger degree, young adults (see figure 1). This global finding masks the regional drivers of\n\nthese differences. Two regions stand out (figure 3, panels a and b): Central and South Asia and Sub-Saharan\n\nAfrica, where 88.4 percent of the world’s poor reside. Poverty gaps by sex among young adults have also\n\nemerged in Latin America and the Caribbean (figure 3, panel c). However, a far smaller share of the world’s\n\npoor reside in this region (4 percent). <sup>8</sup>\n\n\n_Figure 3. Gaps in Poverty by Sex Are Concentrated in Three Regions_\n\n\na. Central and South Asia b. Sub-Saharan Africa\n\n\n\n25\n\n\n20\n\n\n15\n\n\n10\n\n\n5\n\n\n0\n\n\n\n55\n\n50\n\n45\n\n40\n\n35\n\n30\n\n25\n\n20\n\n\n\nWomen Men\n\n\n8 See appendix A, table A.1 for global and regional poverty rates by age and sex.\n\n\n6\n\n\n\nWomen Men\n\nMention: \"data of the GMD\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Child versus adult extreme poverty shares are calculated from this microdata database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000027:18:0", "occurrence_id": "d09dc4aaec2deb3fa5bd0193", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: spending on subsequent operation quality. In Table 9 we find that prior analytical work is indeed linked to\nimproved project outcomes, especially in lower-income IDA settings and possibly even fragile and conflictaffected settings (FCVs). Analytical work completed up to three years before project approval is associated\nwith an increase of 0.12 in IEG outcome ratings (0.15 points for IDA countries); half the impact associated\nwith a one-point improvement in the CPIA. This suggests an important role for World Bank initiated\nanalytic work in country environments that may be relatively “knowledge scarce”, where the information\ngenerated by these analytic efforts informs subsequent project design.\n\n\n**5.** **Summary and discussion of unsettled questions**\n\nThis paper reviews the literature on the correlates of World Bank project outcome ratings and then extends\nprevious analytic work by identifying further staffing inputs into the production process, linking previous\nsector analytic work to individual projects, and extracting a richer set of design quality explanators from a\nsubset of projects. Results from these analytic extensions all point towards the importance of project design\nin ultimate project success, especially in the more sectoral complex settings of middle- and upper-middleincome countries (IBRD). Previous quantitative work has found fairly weak evidence for the importance of\nproject design; however, it appears that this is in part due to the paucity of relevant indicators captured in\nroutine World Bank administrative data. It would be an error to conclude from these earlier studies that\nproject design is not an especially influential determinant of project success.\n\n\nLike much of the project effectiveness literature, the analysis discussed here is descriptive in nature.\nHence, any observed association between project inputs and outcomes may either represent an underlying\ncausal relationship or may be confounded by further unobserved factors. While our understanding of the\nproject cycle leads us to consider these correlations as plausibly capturing the role certain inputs play in\nproject success, there are alternative explanations that render the correlations spurious. For example,\n\nMention: \"routine World Bank administrative data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Earlier findings on project design are discounted because routine administrative records capture too few relevant indicators.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000028:10:1", "occurrence_id": "098fed9d5dd38bf29c5f6b8b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  or\nwithin a State” (International Organization for Migration, 2011).\n\n\nBoth databases are based on the same underlying migration data and share many of the same\nprocessing methods. In both cases, the UN’s Population Division census data is used to compile the\ndatabase. The same country list is employed for both databases, although the UN DESA data contains six\nmore countries than the World Bank’s. In our merged data set, we only count those countries included in\nboth databases. The original data suffers from a substantial amount of missing observations because many\ncountries do not release national census data every 10 years. These may be prohibitively expensive in terms\nof labor intensity, can be abandoned because of exogenous factors, such as civil unrest or conflict, or are\nnever released for political reasons. The authors chose to minimize the number of gaps in the data through\ninterpolations. For the ‘in-between’ years (1970, 1980, 1990 for the 1960-2000 World Bank data and 2000\nand 2010 for the UN DESA data), they do so by assuming a linear trend before and after missing data\npoints. Where data are lacking for the beginning or end decades, they use growth rates in migration, taken\nfrom the UN Total Migrant Stock database (2006), to estimate bilateral migrant stocks. It is important to\nnote, however, that since both databases use interpolations and predictions to fill in for missing values, our\ncompiled bilateral database will also include a number of predicted values. <sup>13</sup> As a result, our estimation\nresults are partially based on using predicted values as independent variables, which leads to increased\nuncertainty on the results.\n\n\n11 During the timeframe covered by these censuses, many regions reshaped their political boundaries, such as the USSR and\nGermany. For this reason, authors define their “master” country list as the most current set of countries.\n\n12 This definition is used where possible. Whenever birthplace information is missing\n\nMention: \"Population Division census data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Both bilateral migration databases are compiled from this underlying census collection.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000028:10:1", "occurrence_id": "0a77a8c1c6dacc3b1a29cea1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  or\nwithin a State” (International Organization for Migration, 2011).\n\n\nBoth databases are based on the same underlying migration data and share many of the same\nprocessing methods. In both cases, the UN’s Population Division census data is used to compile the\ndatabase. The same country list is employed for both databases, although the UN DESA data contains six\nmore countries than the World Bank’s. In our merged data set, we only count those countries included in\nboth databases. The original data suffers from a substantial amount of missing observations because many\ncountries do not release national census data every 10 years. These may be prohibitively expensive in terms\nof labor intensity, can be abandoned because of exogenous factors, such as civil unrest or conflict, or are\nnever released for political reasons. The authors chose to minimize the number of gaps in the data through\ninterpolations. For the ‘in-between’ years (1970, 1980, 1990 for the 1960-2000 World Bank data and 2000\nand 2010 for the UN DESA data), they do so by assuming a linear trend before and after missing data\npoints. Where data are lacking for the beginning or end decades, they use growth rates in migration, taken\nfrom the UN Total Migrant Stock database (2006), to estimate bilateral migrant stocks. It is important to\nnote, however, that since both databases use interpolations and predictions to fill in for missing values, our\ncompiled bilateral database will also include a number of predicted values. <sup>13</sup> As a result, our estimation\nresults are partially based on using predicted values as independent variables, which leads to increased\nuncertainty on the results.\n\n\n11 During the timeframe covered by these censuses, many regions reshaped their political boundaries, such as the USSR and\nGermany. For this reason, authors define their “master” country list as the most current set of countries.\n\n12 This definition is used where possible. Whenever birthplace information is missing\n\nMention: \"UN DESA data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Country coverage is compared across the two databases, with this one holding six more countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000028:10:1", "occurrence_id": "f56a1489ac9b6994e002ca9f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  or\nwithin a State” (International Organization for Migration, 2011).\n\n\nBoth databases are based on the same underlying migration data and share many of the same\nprocessing methods. In both cases, the UN’s Population Division census data is used to compile the\ndatabase. The same country list is employed for both databases, although the UN DESA data contains six\nmore countries than the World Bank’s. In our merged data set, we only count those countries included in\nboth databases. The original data suffers from a substantial amount of missing observations because many\ncountries do not release national census data every 10 years. These may be prohibitively expensive in terms\nof labor intensity, can be abandoned because of exogenous factors, such as civil unrest or conflict, or are\nnever released for political reasons. The authors chose to minimize the number of gaps in the data through\ninterpolations. For the ‘in-between’ years (1970, 1980, 1990 for the 1960-2000 World Bank data and 2000\nand 2010 for the UN DESA data), they do so by assuming a linear trend before and after missing data\npoints. Where data are lacking for the beginning or end decades, they use growth rates in migration, taken\nfrom the UN Total Migrant Stock database (2006), to estimate bilateral migrant stocks. It is important to\nnote, however, that since both databases use interpolations and predictions to fill in for missing values, our\ncompiled bilateral database will also include a number of predicted values. <sup>13</sup> As a result, our estimation\nresults are partially based on using predicted values as independent variables, which leads to increased\nuncertainty on the results.\n\n\n11 During the timeframe covered by these censuses, many regions reshaped their political boundaries, such as the USSR and\nGermany. For this reason, authors define their “master” country list as the most current set of countries.\n\n12 This definition is used where possible. Whenever birthplace information is missing\n\nMention: \"World Bank data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Interpolated inter-census years are identified so readers know which observations are not directly observed.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000028:10:1", "occurrence_id": "9e5cf9b9efa1cba37f17ebf2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  or\nwithin a State” (International Organization for Migration, 2011).\n\n\nBoth databases are based on the same underlying migration data and share many of the same\nprocessing methods. In both cases, the UN’s Population Division census data is used to compile the\ndatabase. The same country list is employed for both databases, although the UN DESA data contains six\nmore countries than the World Bank’s. In our merged data set, we only count those countries included in\nboth databases. The original data suffers from a substantial amount of missing observations because many\ncountries do not release national census data every 10 years. These may be prohibitively expensive in terms\nof labor intensity, can be abandoned because of exogenous factors, such as civil unrest or conflict, or are\nnever released for political reasons. The authors chose to minimize the number of gaps in the data through\ninterpolations. For the ‘in-between’ years (1970, 1980, 1990 for the 1960-2000 World Bank data and 2000\nand 2010 for the UN DESA data), they do so by assuming a linear trend before and after missing data\npoints. Where data are lacking for the beginning or end decades, they use growth rates in migration, taken\nfrom the UN Total Migrant Stock database (2006), to estimate bilateral migrant stocks. It is important to\nnote, however, that since both databases use interpolations and predictions to fill in for missing values, our\ncompiled bilateral database will also include a number of predicted values. <sup>13</sup> As a result, our estimation\nresults are partially based on using predicted values as independent variables, which leads to increased\nuncertainty on the results.\n\n\n11 During the timeframe covered by these censuses, many regions reshaped their political boundaries, such as the USSR and\nGermany. For this reason, authors define their “master” country list as the most current set of countries.\n\n12 This definition is used where possible. Whenever birthplace information is missing\n\nMention: \"UN Total Migrant Stock database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Migration growth rates from this release are used to estimate bilateral stocks where data are missing.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000031:29:0", "occurrence_id": "1e8945a283ff58bc265b2218", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <mark>male nonparticipation. In that experiment, wives’ home production times do not change, but,</mark>\n<mark>interestingly, males increase their time in all labor markets. This reflects how household optimization</mark>\n<mark>processes interact with policies, making spouses reallocate the portfolio activities. In this experiment,</mark>\n<mark>as wives go more to the labor market, husbands increase home production time, and some decide to not</mark>\n<mark>participate in the labor market.</mark>\n\n\n**Table 9: Observed and simulated mean accepted wages**\n\n\nBenchmark Arrival Laid-off Wages Pt Home_P Fertility Search\n\npt_m 1.3 1.3 1.3 1.5 1.3 1.3 1.3 1.3\n\nft_m 0.7 0.7 0.7 0.7 0.7 0.7 0.7 0.7\n\npt_w 1.2 1.2 1.2 1.5 1.1 1.2 1.2 1.2\n\n<u>ft_w</u> <u>0.6</u> <u>0.6</u> <u>0.6</u> <u>0.7</u> <u>0.6</u> <u>0.6</u> <u>0.6</u> <u>0.6</u>\n_Source:_ Author’s calculations based on 2017 UNDP-WB-EC Roma Regional Survey.\n_Note:_ Abbreviations: FT= employed full-time; PT= employed part-time. This table presents the converted values in 2011 USD PPP dollars.\n\n**Figure 6: Mean time in home production by labor market state (%)*\n\nMention: \"UNDP-WB-EC Roma Regional Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Employment shares and converted earnings in the table are calculated from this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000033:5:1", "occurrence_id": "d232b9af6a8ad47cc6b2788b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  and\nUganda National Panel Survey 2019/20. At least one phone number was recorded for all\nhouseholds with access to a phone, including through a contact person outside the household, such\nas a friend or neighbor. For the HFPS, Ethiopia, Malawi, and Uganda attempted to contact all\nhouseholds with available phone numbers, while Burkina Faso, Nigeria and Mali selected a\nrandom sub-sample of phone numbers to call (Table A.1).\n\nIn the absence of universal access to a mobile phone, the HFPS households are likely to be selected\nsamples of the nationally representative FtF survey samples of households.[14] The publicly\navailable HFPS data sets therefore come with recalibrated household survey weights to counteract\npotential selection biases. The recalibration model takes advantage of the rich information on\nhouseholds with and without access to a phone from the pre-COVID-19 FtF surveys [15,16] and\nfollows a methodology proposed in a reference methodological paper on this subject.[17] The\neffectiveness of this weight recalibration in overcoming selection biases has been documented in\nHFPS data from four of the six countries we study, such that household-level estimates based on\nthe HFPS data can be broadly considered representative at the national level.[14]\n\nFurther, the HFPS survey questions on vaccinations were asked only to the household’s main\nrespondent, who had to be 15 or older, as it is impractical to interview all individuals in a household\nin a phone survey. The selection of main respondents was not randomized so that the group of\nrespondents is likely not representative of the general population of adults at the individual level.\nRather, respondents tend to be household heads or their spouses, better educated and slightly older,\nas a recent study documents.[18] An additional recalibration of survey weights for individual-level\n\n\n4\n\nMention: \"HFPS data sets\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Public releases carry recalibrated weights so phone-based samples stay nationally representative.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000033:5:1", "occurrence_id": "ad05d87b36e500b1b12a63f4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  and\nUganda National Panel Survey 2019/20. At least one phone number was recorded for all\nhouseholds with access to a phone, including through a contact person outside the household, such\nas a friend or neighbor. For the HFPS, Ethiopia, Malawi, and Uganda attempted to contact all\nhouseholds with available phone numbers, while Burkina Faso, Nigeria and Mali selected a\nrandom sub-sample of phone numbers to call (Table A.1).\n\nIn the absence of universal access to a mobile phone, the HFPS households are likely to be selected\nsamples of the nationally representative FtF survey samples of households.[14] The publicly\navailable HFPS data sets therefore come with recalibrated household survey weights to counteract\npotential selection biases. The recalibration model takes advantage of the rich information on\nhouseholds with and without access to a phone from the pre-COVID-19 FtF surveys [15,16] and\nfollows a methodology proposed in a reference methodological paper on this subject.[17] The\neffectiveness of this weight recalibration in overcoming selection biases has been documented in\nHFPS data from four of the six countries we study, such that household-level estimates based on\nthe HFPS data can be broadly considered representative at the national level.[14]\n\nFurther, the HFPS survey questions on vaccinations were asked only to the household’s main\nrespondent, who had to be 15 or older, as it is impractical to interview all individuals in a household\nin a phone survey. The selection of main respondents was not randomized so that the group of\nrespondents is likely not representative of the general population of adults at the individual level.\nRather, respondents tend to be household heads or their spouses, better educated and slightly older,\nas a recent study documents.[18] An additional recalibration of survey weights for individual-level\n\n\n4\n\nMention: \"HFPS data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Representativeness after recalibration has been demonstrated in four of the six study countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000037:19:0", "occurrence_id": "8474ca14b7374a092e00572e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: and the data disclose the date and time of each transaction, a stock identifier, order type (buy or sell),\n\n\ntransaction price, number of shares, broker, and investor type (i.e., individuals or institutions). Transaction\n\n\ncosts such as broker fees are not captured by the CSE. <sup>18</sup> [^18: See Pedraza, et al., (2019) and Pedraza (2020) for additional description of the Colombian stock exchange,\nthe transactional and portfolio data.] During our sample period, trades by individual\n\n\ninvestors accounted for over seven million transactions. Importantly for our analysis, the data include the\n\n\nnational identification number for each individual, which can be used to merge the information with the\n\n\nfinancial education initiative. Overall, 36% of the students in the CSE’s education program had at least one\n\n\nstock trade throughout our sample period, either before or after taking the course. These 3,960 students with\n\n\nsome trading history are more active in the stock market than the average Colombian individual investor.\n\n\nFor example, while the average investor in Colombia owns 2.2 different stocks, makes 1.3 stock\n\n\ntransactions per year, and averages 7,200 USD per trade, the program students who actively trade hold 6.2\n\n\nstocks, make 6.6 trades per year, and average 8,781 USD per trade.\n\n\nThird, we collect additional information about students via an electronic survey. The survey had\n\n\nthree parts: (i) socioeconomic information (i.e., age, education, academic history, and earnings); (ii)\n\n\nexperience in financial assets (e.g., whether the individual had any foreign investments or mutual funds,\n\n\nwhich are not captured in the CSE data); and (iii) self-reported social interaction (i.e., whether they took\n\n\nthe course with a friend or a relative and whether they talked to classmates about investment strategies\n\n\nduring the course). The survey was\n\nMention: \"CSE data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Foreign holdings and mutual funds are absent from these transaction records, so a supplementary survey collects them.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000038:11:0", "occurrence_id": "7cc3d97ff40b2b9a2f9b1ce2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 10\n\n\n  - **RCEP:** implementation of the RCEP agreement among its 16 members begins in 2017. Reduction\nof barriers follows Petri et. al. (2012), Petri and Plummer (2016) and ITC (2015) and ITC (2016).\nImport tariffs are gradually reduced starting with a decline by 24% in 2017 up to 89% in 2027.\n\n  - **FTAAP:** implementation of the FTAAP agreement among its members begins in 2021. Reduction\nof barriers follows Petri et. al. (2012), Petri and Plummer (2016) and ITC (2015) and ITC (2016).\nImport tariffs among the member countries are also gradually reduced starting with a decline by\n46% in 2021 up to 94% in 2030.\n\nIn case of RCEP and FTAAP, initial tariffs are based on the GTAP data, while for the countries that are\nalso TPP members we use the TPP tariff schedules from International Trade Centre (2015) and\nInternational Trade Centre (2016). At the same time, initial NTMs and their cuts follow Petri and\nPlummer (2016).\n\nMultilateral tariffs are projected to decline to minimum levels under CPTPP and TPP12 with more\nmoderate reductions in RCEP. Reductions in actionable non-tariff measures (NTMs) follow the\napproach of Petri and Plummer (2016). NTMs for goods are based on estimates by Kee, Nicita, &\nOlarreaga (2008) updated in 2012 and the services barriers are based on estimates by Fontagné,\nMitaritonna, & Signoret (2016). Only three-quarters of measured barriers are considered as actual trade\nbarriers, the rest is assumed to represent quality-increasing regulations (e.g., product safety standards).\nMeanwhile only three-quarters of the remaining NTMs in the case of goods and one-half in the case of\nservices are assumed to be actionable (i.e.,\n\nMention: \"GTAP data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Initial tariff levels for the simulated agreements are taken from this trade database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000040:6:0", "occurrence_id": "0b1e4ffb8c0ec13c816b61ff", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **2.** **Drivers of the recent oil price plunge**\n\n\nThis section reviews recent oil market developments and uses a structural vector autoregression\n\nmodel to decompose the price decline into demand and supply factors.\n\n\n**2.1 Oil market developments during the onset of the COVID-19 pandemic**\n\n\nBy one measure, the European Brent spot price, crude oil prices fell by 85 percent between January\n\n22, 2020 (the date the first recorded human-to-human COVID-19 infection was announced) and\n\ntheir trough of $9 per barrel on April 21, 2020 before recovering in May to less than half their\n\nJanuary average (Figure 1). <sup>3</sup> [^3: Another frequently used measure, the Dated Brent spot price, fell by 72 percent over this period, on par with the 70\npercent decline during the global financial crisis (end-August to late December 2008) and the 76 percent decline during\nend-June 2014-mid-January 2016. In late-April, the West Texas Intermediate oil price (a U.S. oil price benchmark)\ncontract for delivery in May temporarily fell below zero on concerns about near-full U.S. storage capacity; however,\nno physical oil was traded at negative prices.] The oil market has been hit by an unprecedented combination of\n\ndemand and supply shocks. The pandemic, and the restrictions on business and personal activities\n\nimposed to stem its spread, have triggered a global recession, and a steep drop in the demand for\n\noil (World Bank 2020b). Total oil demand fell by 5 percent in the first quarter of 2020 and is\n\nprojected to decline 18 percent in the second quarter of 2020 (IEA 2020). This coincided with a\n\ndelay in early March of OPEC and its partners (OPEC+) to agree to an extension of their\n\nproduction cuts (World Bank 2020a). Meanwhile, petroleum inventories have risen rapidly and\n\nare now close to capacity (IEA 2020).\n\n\n**2.2 Methodology and data**\n\n\nA structural vector autoregression (SVAR\n\nMention: \"European Brent spot price\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The 85 percent collapse is measured on this price series between the first announced infection and the April trough.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000040:6:0", "occurrence_id": "e69f10f223d9600f4262ef51", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **2.** **Drivers of the recent oil price plunge**\n\n\nThis section reviews recent oil market developments and uses a structural vector autoregression\n\nmodel to decompose the price decline into demand and supply factors.\n\n\n**2.1 Oil market developments during the onset of the COVID-19 pandemic**\n\n\nBy one measure, the European Brent spot price, crude oil prices fell by 85 percent between January\n\n22, 2020 (the date the first recorded human-to-human COVID-19 infection was announced) and\n\ntheir trough of $9 per barrel on April 21, 2020 before recovering in May to less than half their\n\nJanuary average (Figure 1). <sup>3</sup> [^3: Another frequently used measure, the Dated Brent spot price, fell by 72 percent over this period, on par with the 70\npercent decline during the global financial crisis (end-August to late December 2008) and the 76 percent decline during\nend-June 2014-mid-January 2016. In late-April, the West Texas Intermediate oil price (a U.S. oil price benchmark)\ncontract for delivery in May temporarily fell below zero on concerns about near-full U.S. storage capacity; however,\nno physical oil was traded at negative prices.] The oil market has been hit by an unprecedented combination of\n\ndemand and supply shocks. The pandemic, and the restrictions on business and personal activities\n\nimposed to stem its spread, have triggered a global recession, and a steep drop in the demand for\n\noil (World Bank 2020b). Total oil demand fell by 5 percent in the first quarter of 2020 and is\n\nprojected to decline 18 percent in the second quarter of 2020 (IEA 2020). This coincided with a\n\ndelay in early March of OPEC and its partners (OPEC+) to agree to an extension of their\n\nproduction cuts (World Bank 2020a). Meanwhile, petroleum inventories have risen rapidly and\n\nare now close to capacity (IEA 2020).\n\n\n**2.2 Methodology and data**\n\n\nA structural vector autoregression (SVAR\n\nMention: \"Dated Brent spot price\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"An alternative benchmark is reported to show the decline is comparable to past crises.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000042:9:0", "occurrence_id": "751423efea7fc99b2c035898", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: of qualification for the trade they had selected. A “fitness score” was then computed, reflecting\n\n\nthese four dimensions.\n\n\n**3.** **<u>Data and sample characteristics</u>**\n\n\nThe main data source is a survey conducted among a subset of applicants to the vocational training\n\n\nprogram (we discuss estimates on the full population of applicants to both the vocational training\n\n\nprogram and the apprenticeship program in section 6.3). The sample selection process for this\n\n\nsurvey was driven by the requirements of a complementary but distinct evaluation that the research\n\n\nteam is conducting, tracking the effect of the training itself on participants’ economic outcomes.\n\n\nIn this other evaluation, we will also study the impact of the information treatment on participants’\n\n\neconomic outcomes. To constitute this sample, we started from the universe of applicants to the\n\n\nprogram and excluded people that did not meet the eligibility criteria, which gives 18,518 eligible\n\n\napplicants. This sample was then stratified between individuals who had received the information\n\n\non trade-specific earnings and individuals who had not received it. This was done to maximize\n\n\npower when later measuring the effect of this information on employment and earnings. Within\n\n\nthese two strata, we grouped applicants by their first-choice trade. We calculated the number (N)\n\n\nof spots available in each trade. For each trade, we then selected 2.5xN eligible applicants who\n\n\nopted for the trade in question as first choice to build our study sample, among which two-thirds\n\n\nwere women and one-third were men, whenever possible, i.e., when enough women had selected\n\n\nthis trade. In cases where there were more than 2.5xN applicants for a trade (or more precisely\n\n\nmore than 2.5xNx2/3 female applicants or 2.5xN/3 male applicants), we ranked individuals by the\n\n\n‘fitness score’ assigned to them by training providers when they submitted their application. We\n\n\nthen allocated\n\nMention: \"survey conducted among a subset of applicants\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Applicant characteristics and the computed fitness score come from this survey of a sample of trainee applicants.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000045:31:0", "occurrence_id": "f0325caf74fa53b0edb173d6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 9. Probability of Having Attended Tertiary Education by Skill Deciles for Adults Aged 25–**\n\n\n**64**\n\n\n_Source_ : Authors' calculations based on Colombia STEP Household Survey (2012).\n\n\n_Note:_ Simulations are based on structural estimations of latent skills factors using Sarzosa and Urzúa (2016). Cognitive skills are captured by an\n\nestimated latent factor using measures of reading proficiency, use of reading on and off work, and language. Socioemotional skills are captured\n\nby an estimated latent factor using measures of the Big Five personality traits (extraversion, openness to experience, emotional stability,\n\nconscientiousness), grit, hostility bias, and decision-making styles. See table 1 for definitions.\n\n\n30\n\nMention: \"Colombia STEP Household Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Tertiary attendance by skill decile is calculated from this skills measurement survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000058:27:11", "occurrence_id": "f3ceca0bb08661f53e6c20e2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: *<br>|LFS<br>|LFS<br>|\n\n\nIntegrated Economic Survey, HIICS: Household Integrated Income and Consumption Survey, LFS: Labor Force Survey, NRVA: National Risk and Vulnerability Assessment, NSS:\nNational Sample Survey, PIHS: Pakistan Integrated Household Survey, PSLM: Pakistan Social and Living Standards Measurement. Access to primary data is indicated in **bold** .\nSurveys placed in between two years were collected in both years, for example, in Afghanistan the NRVA was collected in 2007-08.\n\n\n26\n\nMention: \"Labor Force Survey\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"The footnote only expands the survey acronym used in the data-availability table.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000058:27:11", "occurrence_id": "ae0a0990da977e43d902fd75", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: *<br>|LFS<br>|LFS<br>|\n\n\nIntegrated Economic Survey, HIICS: Household Integrated Income and Consumption Survey, LFS: Labor Force Survey, NRVA: National Risk and Vulnerability Assessment, NSS:\nNational Sample Survey, PIHS: Pakistan Integrated Household Survey, PSLM: Pakistan Social and Living Standards Measurement. Access to primary data is indicated in **bold** .\nSurveys placed in between two years were collected in both years, for example, in Afghanistan the NRVA was collected in 2007-08.\n\n\n26\n\nMention: \"National Sample Survey\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"The acronym appears only in the table's list of survey name expansions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000058:27:11", "occurrence_id": "748e600dc0911918ee4ff5c3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: *<br>|LFS<br>|LFS<br>|\n\n\nIntegrated Economic Survey, HIICS: Household Integrated Income and Consumption Survey, LFS: Labor Force Survey, NRVA: National Risk and Vulnerability Assessment, NSS:\nNational Sample Survey, PIHS: Pakistan Integrated Household Survey, PSLM: Pakistan Social and Living Standards Measurement. Access to primary data is indicated in **bold** .\nSurveys placed in between two years were collected in both years, for example, in Afghanistan the NRVA was collected in 2007-08.\n\n\n26\n\nMention: \"Pakistan Integrated Household Survey\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"Listed only to expand the PIHS acronym in the survey key.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000058:27:11", "occurrence_id": "3638a91324146593df919967", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: *<br>|LFS<br>|LFS<br>|\n\n\nIntegrated Economic Survey, HIICS: Household Integrated Income and Consumption Survey, LFS: Labor Force Survey, NRVA: National Risk and Vulnerability Assessment, NSS:\nNational Sample Survey, PIHS: Pakistan Integrated Household Survey, PSLM: Pakistan Social and Living Standards Measurement. Access to primary data is indicated in **bold** .\nSurveys placed in between two years were collected in both years, for example, in Afghanistan the NRVA was collected in 2007-08.\n\n\n26\n\nMention: \"Pakistan Social and Living Standards Measurement\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"The expansion of PSLM appears in the acronym key beneath the table.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000058:27:11", "occurrence_id": "b44b07c38bdd64c773326b81", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: *<br>|LFS<br>|LFS<br>|\n\n\nIntegrated Economic Survey, HIICS: Household Integrated Income and Consumption Survey, LFS: Labor Force Survey, NRVA: National Risk and Vulnerability Assessment, NSS:\nNational Sample Survey, PIHS: Pakistan Integrated Household Survey, PSLM: Pakistan Social and Living Standards Measurement. Access to primary data is indicated in **bold** .\nSurveys placed in between two years were collected in both years, for example, in Afghanistan the NRVA was collected in 2007-08.\n\n\n26\n\nMention: \"NRVA\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"Cited to explain that surveys placed between two years were fielded in both.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000060:42:0", "occurrence_id": "f6818a33bb8c5b8065527ee6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Appendix A: Subjective Well-Being (SWB), Economic Development, and Inequality**\nFigure A1. SWB by economic development at the country level\n\n\n_Sources:_ Gallup World Poll for SWB (2010-2018) and World Development Indicators for GDP per capita. _Notes:_\nN=153 countries. Sampling weights and two-letter country codes are used. No control variables. For both indicators\nthe average of the 2010-2018 period is taken. Non-linear regression line and 95% confidence interval shown. GDP\nper capita is measured at Purchasing Power Parities (PPP) using constant 2017 international dollars. For Brazil, the\naverage SWB is 6.73 and the GDP per capita is $15,092.\n\n\n_Source:_ Gallup World Poll (2010-2018). _Notes:_ N=156 countries. Sampling weights and two-letter country codes are\nused. No control variables. SWB inequality is measured using the method proposed by Kalmijn and Veenhoven\n(2005). Non-linear regression line and 95% confidence interval shown.\n\n\n41\n\nMention: \"Gallup World Poll\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Country-level subjective well-being averages are taken from this poll for 153 countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000060:42:0", "occurrence_id": "ad068e385c992884895f3df3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Appendix A: Subjective Well-Being (SWB), Economic Development, and Inequality**\nFigure A1. SWB by economic development at the country level\n\n\n_Sources:_ Gallup World Poll for SWB (2010-2018) and World Development Indicators for GDP per capita. _Notes:_\nN=153 countries. Sampling weights and two-letter country codes are used. No control variables. For both indicators\nthe average of the 2010-2018 period is taken. Non-linear regression line and 95% confidence interval shown. GDP\nper capita is measured at Purchasing Power Parities (PPP) using constant 2017 international dollars. For Brazil, the\naverage SWB is 6.73 and the GDP per capita is $15,092.\n\n\n_Source:_ Gallup World Poll (2010-2018). _Notes:_ N=156 countries. Sampling weights and two-letter country codes are\nused. No control variables. SWB inequality is measured using the method proposed by Kalmijn and Veenhoven\n(2005). Non-linear regression line and 95% confidence interval shown.\n\n\n41\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"GDP per capita is taken from this indicator series to pair with well-being.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000070:28:0", "occurrence_id": "df0f3d178558eaa55ca6611a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Appendix Table A1. Results from Logit Regressions of**\n\n**<u>Response Among IHPS Individuals Aged 18-64</u>**\n<u>Variable</u> <u>Marginal Effect</u>\nAge 0.0002\n\n[0.53]\nFemale 0.088***\n\n[9.94]\nEducation:\n\nNone 0.193***\n\n[9.98]\nDid not complete primary 0.174***\n\n[7.75]\nCompleted primary 0.202***\n\n[9.22]\nCompleted secondary 0.173***\n\n[7.67]\n\n\n\n-0.064***\nCurrently married\n\n[-4.65]\n\n0.292***\nIndividual is head or spouse\n\n[11.65]\n\n-0.012***\nMonths member is away\n\n[-6.36]\n\nLog household size -0.046***\n\n\n\n\n[-4.65]\n\n\n\n\n[-6.36]\n\n\n\n\n[-4.19]\nHousehold: dependency ratio of children, and 0.027***\nelderly/adults 15-64 [3.47]\n\n\n\nHousehold: dependency ratio of children, and 0.027***\nelderly/adults 15-64 [3.47]\n\n\nDistrict FE Y\nInterview month FE Y\nEnumerator FE Y\n\nObservations 5,762\n<u>Pseudo R2</u> <u>0.25</u>\n**<u>Notes:</u>**\n<u>Average marginal effects are derived from logit models. Z-statistics in</u>\nbrackets. * (p < 0.1), ** (p < 0.05) and *** (p < 0.01). The logit regression\ndropped 14 observations because one enumerator fixed effect predicted the\noutcome perfectly.\n\n\n\n27\n\nMention: \"IHPS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Logit marginal effects on response behaviour are estimated for working-age individuals in this panel.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000071:11:1", "occurrence_id": "d5d7dc1728e62e317b6f2060", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  individuals built\nup large savings that they then withdrew before the endline.\n\nAppendix Figure A4 also shows the corresponding administrative data on PALFSI loan balances\nover the same time period. This examines a related hypothesis that perhaps individuals receiving\nthe aspirations treatment _increased_ their borrowing immediately after treatment, achieved their\ninvestment goals, and subsequently _decreased_ their borrowing by the time of the endline survey.\nThe data are not consistent with this alternative explanation, with treated individuals having similar\ndebt levels as the control group right after training, before borrowing declines for the treated\ntowards our endline period.\n\nFurther, if treated respondents were able to finance their investments through increased savings\nbetween treatment and endline, we might observe positive impacts on business and non-business\ninvestments at endline. In Appendix Table A4, we show that the treatment did not lead individuals\nto start new businesses. There are no differences in total business value or household assets. In\nsum, there is no evidence for the alternative hypothesis that respondents in the aspirations\ntreatment achieved their savings and investment goals in the time period _before_ the endline survey.\n\n##### **5.2 Impact on expenditures**\n\nMean monthly household income reported by households in our follow-up survey is 11,283 pesos\n($267), which is less than mean monthly household expenditures of 13,124 pesos ($311). Even\nallowing for some underreporting of incomes, this suggests households are spending all they earn.\nBoth training programs emphasized that a key way to increase savings was through reducing\nexpenditure. The knowledge training highlighted the value of forgoing typical temptation goods\n\n\n10\n\nMention: \"administrative data on PALFSI loan balances\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Lender loan balance records are plotted over the same period to test whether borrowing rose after the treatment.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000071:11:1", "occurrence_id": "c238e76bc6ada736897b5699", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  individuals built\nup large savings that they then withdrew before the endline.\n\nAppendix Figure A4 also shows the corresponding administrative data on PALFSI loan balances\nover the same time period. This examines a related hypothesis that perhaps individuals receiving\nthe aspirations treatment _increased_ their borrowing immediately after treatment, achieved their\ninvestment goals, and subsequently _decreased_ their borrowing by the time of the endline survey.\nThe data are not consistent with this alternative explanation, with treated individuals having similar\ndebt levels as the control group right after training, before borrowing declines for the treated\ntowards our endline period.\n\nFurther, if treated respondents were able to finance their investments through increased savings\nbetween treatment and endline, we might observe positive impacts on business and non-business\ninvestments at endline. In Appendix Table A4, we show that the treatment did not lead individuals\nto start new businesses. There are no differences in total business value or household assets. In\nsum, there is no evidence for the alternative hypothesis that respondents in the aspirations\ntreatment achieved their savings and investment goals in the time period _before_ the endline survey.\n\n##### **5.2 Impact on expenditures**\n\nMean monthly household income reported by households in our follow-up survey is 11,283 pesos\n($267), which is less than mean monthly household expenditures of 13,124 pesos ($311). Even\nallowing for some underreporting of incomes, this suggests households are spending all they earn.\nBoth training programs emphasized that a key way to increase savings was through reducing\nexpenditure. The knowledge training highlighted the value of forgoing typical temptation goods\n\n\n10\n\nMention: \"follow-up survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Reported monthly income and expenditure come from this endline round, showing households spend all they earn.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000075:5:0", "occurrence_id": "6ce7be3de1a68ec511d4c5f6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 4\n\n\nindividual, however, is impractical, time consuming, and cumbersome, which may help explain\n\n\nwhy households share meals in practice. When sharing, diets would need to be sufficiently\n\n\nnutrient-dense to meet all nutrient needs of all individuals; they would need to be equally or\n\n\nmore nutrient-dense than individualized diets and have lower upper limits on certain nutrients\n\n\n(e.g., copper, zinc) to ensure that the most sensitive member would not exceed their limit\n\n\n(Schneider et al. 2021).\n\n\nAt the extreme, perfect sharing requires a food basket dense enough in each nutrient to\n\n\nprovide a sufficient amount such that every member will have their nutrient requirements\n\n\nsatisfied from the total quantity of the family meal meeting their energy need (Schneider et al.\n\n\n2021). Costing this shared diet provides an upper bound on the cost of adequate diets for the\n\n\nfamily. It must include more nutrient dense foods (i.e., have higher total nutrient density) to meet\n\n\nthat shared set of needs than the summation of individuals procuring a diet meeting only their\n\n\nindividual requirements, hence being higher in cost. But it also must optimize the combination of\n\n\nfoods to meet nutrient density constraints with smaller ranges between the minimum requirement\n\n\nand upper limit, hence potentially becoming less feasible.\n\n\nThe central contributions of this paper are twofold. The first is methodological. It extends the\n\n\nleast-cost diets framework from individuals to households and demonstrates that widely\n\n\navailable nationally representative household survey data on household composition and food\n\n\nconsumption can be linked with frequently collected sub-national monthly food prices and local\n\n\nfood composition data to estimate a bounded range of the cost of nutrient adequate diets that\n\n\nmeet the needs of all household members. This enables policy makers to assess the extent to\n\n\nwhich markets can supply their population with an adequate diet at an affordable cost. The range\n\n\nbetween the bounds offers a more useful policy indicator than either bound on its own, as both\n\nMention: \"local\n\n\nfood composition data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Composition data are linked with household consumption and local prices to bound the cost of a nutrient-adequate diet.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000077:57:0", "occurrence_id": "968bbd36f904a1cbe431330f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Appendix Figure 4: Labor share,**\n\n\n**Source** : Authors’ calculations based on PWT10 data. Outlier Central African\nRepublic was dropped.\n\n\n\n**Appendix Figure 5: Forecast distribution of oil prices,**\n\n\nfrom the estimated OLS process.\n**Data source** unconditional mean is : Real Annual Prices from the World Bank Commodity Price Data, 𝑃𝑃 <sup>�</sup> 𝑒𝑒 <sup>𝜎𝜎2/2</sup> . The forecast distribution is based on 10,000 draws\nThe Pink Sheet. We applied the HP-filter on the 75 <sup>th</sup> percentile.\n\n\n\n**Appendix Figure 6 Simulations for Angola: GDP growth in the LTGM versus**\n**non-oil GDP growth in the LTGM-NR**\nAnnual growth rate, Percent\n\n\n56\n\nMention: \"PWT10 data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The labour share series is calculated from this version of the Penn World Table.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000077:57:0", "occurrence_id": "16c31d8b64d772e1137e8c6a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Appendix Figure 4: Labor share,**\n\n\n**Source** : Authors’ calculations based on PWT10 data. Outlier Central African\nRepublic was dropped.\n\n\n\n**Appendix Figure 5: Forecast distribution of oil prices,**\n\n\nfrom the estimated OLS process.\n**Data source** unconditional mean is : Real Annual Prices from the World Bank Commodity Price Data, 𝑃𝑃 <sup>�</sup> 𝑒𝑒 <sup>𝜎𝜎2/2</sup> . The forecast distribution is based on 10,000 draws\nThe Pink Sheet. We applied the HP-filter on the 75 <sup>th</sup> percentile.\n\n\n\n**Appendix Figure 6 Simulations for Angola: GDP growth in the LTGM versus**\n**non-oil GDP growth in the LTGM-NR**\nAnnual growth rate, Percent\n\n\n56\n\nMention: \"World Bank Commodity Price Data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Real annual oil prices supply the unconditional mean for the forecast distribution.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000082:71:1", "occurrence_id": "4099e4de76b6315102c17a18", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  _старт_ _работе_ _новых_ _железных_ _дорог_ _Жезказган_ - _Бейнеу_ _и_ _Аркалык_ - _Шубарколь_ . August 22, 2014.\n\n\nTemirZholy.kz/press-tsentr. \"President commences work on new Jezkazgan-Beyneu and Arkalyk-Shubarkol railways.\" In Russian..\n\n19. _Протяженность_ _железнодорожных_ _путей_ _Казахстана_ _на_ _начало_ _года_ _составила_ 15,3 _тыс_ . _км_ . _Ж_ / _д_ - _линии_, _проходящие_ _по_ _территории_ _страны_,\n\n_занимают_ 14,5 _тыс_ . _км_, _еще_ 275 _км_ _проходят_ _по_ _территориям_ _других_ _государств_ . s.l. : EnergyProm, July 12, 2016. EnergyProm.kz. \"Total length of\n\n\nrailways in Kazakhstan at the beginning of the year was 15.3 thousand kilometers. Railways that traverse the country's territory cover 14.5 thousand kilometers, while 257\n\nkm pass through other countries.\" In Russian. Cites Kazakhstan Ministry of Transport web portal as data source..\n\n20. Chu, Daye. Moscow-Kazan high-speed railway ‘may be built in stages’ for financial reasons . s.l. : Global Times, October 22, 2017. GlobalTimes.cn. Quotes Russian\n\nRailways (RZD).\n\nMention: \"Kazakhstan Ministry of Transport web portal\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"A footnote records this portal as the source for the reported railway length.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000090:28:0", "occurrence_id": "5736a33cae8f1bbefbc5cece", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Source: GeoTerra Image, Building Based Land Use data\nset, 2014, and City of Cape Town\n\n### Figure 4: Neighborhood View\n\n\n26\n\nMention: \"Building Based Land Use data\nset\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The neighbourhood view figure is produced from this building-level land cover product.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000093:35:0", "occurrence_id": "b2b174588682df7ca9b3b59e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure A2: Capital over Total Assets (%) – top 20% banks and bottom 80% (Weighted by total**\n**assets)**\n\n\n\n**Panel A: EU and Central Europe & Baltic**\n\n**Countries**\n\n\n\n**Panel B: Russia and Turkey**\n\n\n\n\n\n\n\n\n\n\n\n**Panel C: Western Balkans and Eastern Europe** **Panel D: South Caucasus and Central Asia**\n\n\n\n\n\n\n\n\n\n\n\nSource: Archived data from Bureau van Dijk’s Bankscope and BankFocus.\nNote: **EU** stands for Western Europe, Southern Europe, Northern Europe; **CE & BC** for Central Europe & Baltic Countries; **WB** for Western\nBalkans; **SC** for South Caucasus; **CA** for Central Asia; **EE** for Eastern Europe.\n\n\n33\n\nMention: \"Bankscope\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Bank-level capital ratios across regions are computed from archived balance sheet data in this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000093:35:0", "occurrence_id": "077694a1f8b6a713a92836b5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure A2: Capital over Total Assets (%) – top 20% banks and bottom 80% (Weighted by total**\n**assets)**\n\n\n\n**Panel A: EU and Central Europe & Baltic**\n\n**Countries**\n\n\n\n**Panel B: Russia and Turkey**\n\n\n\n\n\n\n\n\n\n\n\n**Panel C: Western Balkans and Eastern Europe** **Panel D: South Caucasus and Central Asia**\n\n\n\n\n\n\n\n\n\n\n\nSource: Archived data from Bureau van Dijk’s Bankscope and BankFocus.\nNote: **EU** stands for Western Europe, Southern Europe, Northern Europe; **CE & BC** for Central Europe & Baltic Countries; **WB** for Western\nBalkans; **SC** for South Caucasus; **CA** for Central Asia; **EE** for Eastern Europe.\n\n\n33\n\nMention: \"BankFocus\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The companion database is credited alongside Bankscope for the same bank panel.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000100:38:0", "occurrence_id": "ae8fee9043f6ac5e7f146eb3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 6. Distribution of Nonfarm Enterprises, by Formality Status and Region**\n\n\na. Enterprises (number, in thousands)\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nb. Enterprises (in percent)\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n_Source_ : Authors’ estimates based on data from Phase I of the IBES.\n_Note_ : The 10 regions of Ghana are as defined prior to the 2019 expansion to the 16 regions that are in effect today.\nRegions are arranged according to regional average household consumption per adult equivalent in 2005/06 (based\non 2005/06 GLSS data), from the lowest to the highest (left to right). The sample is restricted to nonfarm enterprises.\nInformal enterprise = an enterprise that is not registered with the Registrar General’s Department (RGD) and does\nnot professionally maintain financial accounting records. In panel b, the distribution of enterprises across\ninformal/formal statuses and regions sum to 100 percent (rounding error notwithstanding).\n\n\n37\n\nMention: \"data from Phase I of the IBES\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Enterprise counts by formality and region are estimated from this business survey round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000100:38:0", "occurrence_id": "40e52beb7dd47858408a200a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 6. Distribution of Nonfarm Enterprises, by Formality Status and Region**\n\n\na. Enterprises (number, in thousands)\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nb. Enterprises (in percent)\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n_Source_ : Authors’ estimates based on data from Phase I of the IBES.\n_Note_ : The 10 regions of Ghana are as defined prior to the 2019 expansion to the 16 regions that are in effect today.\nRegions are arranged according to regional average household consumption per adult equivalent in 2005/06 (based\non 2005/06 GLSS data), from the lowest to the highest (left to right). The sample is restricted to nonfarm enterprises.\nInformal enterprise = an enterprise that is not registered with the Registrar General’s Department (RGD) and does\nnot professionally maintain financial accounting records. In panel b, the distribution of enterprises across\ninformal/formal statuses and regions sum to 100 percent (rounding error notwithstanding).\n\n\n37\n\nMention: \"GLSS data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Regional consumption per adult is used only to order the regions in the figure.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000101:11:0", "occurrence_id": "1ffd9b9cea8b9e8a998444bd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: suppliers, conditions in the property (collateral) markets, and changing country context. Further,\n\n\nwe assume that:\n\n𝑋� �𝑎�𝑋���, 𝐵3, ε1��, (2)\n\n𝑌�= 𝑏�𝑌���, ε2�), (3)\n\n𝑈�= 𝑐�𝑋�, ε3�), (4)\n\n𝑍�= 𝑑�𝑋�, ε4�), (5)\n\n𝑆�= 𝑏�𝑋�, ε5�), (6)\n\n𝑊�= 𝑒�𝑊���, ε6�) (7)\n\n\nwhere 𝐵3 is the Basel III implementation shock, and ε1�…ε6� are shocks to bank conditions, firm\n\nrisk characteristics, property market conditions, conditions of buyers and suppliers, non-bank\n\n\ncredit institutions, and country context, respectively.\n\n\nSubstituting equations (2) to (7) into equation (1), the reduced-form solution takes the following\n\n\nform:\n\n𝐴2𝐹 <sup>���</sup> �𝑔�𝑋���, 𝑌���, 𝑊���, 𝐵3, ε1� … ε6�� (8)\n\n\nIt is used as a basis for the formulation of our reduced-form linear regression in section 6. Although\n\n\nthe simple model in equations (1) to (8) is not derived from behavioral micro-foundations, it\n\n\nillustrates how the thinking framework in figure 1 could be translated into a formal model. In this\n\n\nway, it also helps motivate the regression model specifications that we present in section 6.\n\n\n**4.** **Data**\n\n\nWe use firm-level data from the World Bank’s Enterprise Surveys as our primary source and match\n\n\nthem with several other databases (World Bank’s Global Financial Development Database,\n\n\nBankScope/Fitch, and BIS Financial Stability Institute’s report on the implementation of Basel\n\n\nreforms). The World Bank Enterprise Surveys, which are conducted periodically, are a survey of\n\nMention: \"Enterprise Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Firm-level observations used in the regressions are drawn from this recurring survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000101:11:0", "occurrence_id": "f79a1f25c1506558f71c6005", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: suppliers, conditions in the property (collateral) markets, and changing country context. Further,\n\n\nwe assume that:\n\n𝑋� �𝑎�𝑋���, 𝐵3, ε1��, (2)\n\n𝑌�= 𝑏�𝑌���, ε2�), (3)\n\n𝑈�= 𝑐�𝑋�, ε3�), (4)\n\n𝑍�= 𝑑�𝑋�, ε4�), (5)\n\n𝑆�= 𝑏�𝑋�, ε5�), (6)\n\n𝑊�= 𝑒�𝑊���, ε6�) (7)\n\n\nwhere 𝐵3 is the Basel III implementation shock, and ε1�…ε6� are shocks to bank conditions, firm\n\nrisk characteristics, property market conditions, conditions of buyers and suppliers, non-bank\n\n\ncredit institutions, and country context, respectively.\n\n\nSubstituting equations (2) to (7) into equation (1), the reduced-form solution takes the following\n\n\nform:\n\n𝐴2𝐹 <sup>���</sup> �𝑔�𝑋���, 𝑌���, 𝑊���, 𝐵3, ε1� … ε6�� (8)\n\n\nIt is used as a basis for the formulation of our reduced-form linear regression in section 6. Although\n\n\nthe simple model in equations (1) to (8) is not derived from behavioral micro-foundations, it\n\n\nillustrates how the thinking framework in figure 1 could be translated into a formal model. In this\n\n\nway, it also helps motivate the regression model specifications that we present in section 6.\n\n\n**4.** **Data**\n\n\nWe use firm-level data from the World Bank’s Enterprise Surveys as our primary source and match\n\n\nthem with several other databases (World Bank’s Global Financial Development Database,\n\n\nBankScope/Fitch, and BIS Financial Stability Institute’s report on the implementation of Basel\n\n\nreforms). The World Bank Enterprise Surveys, which are conducted periodically, are a survey of\n\nMention: \"Global Financial Development Database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Country financial-system characteristics are matched onto the firm data from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000101:11:0", "occurrence_id": "c9a9dddf76211b7c4f1bc213", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: suppliers, conditions in the property (collateral) markets, and changing country context. Further,\n\n\nwe assume that:\n\n𝑋� �𝑎�𝑋���, 𝐵3, ε1��, (2)\n\n𝑌�= 𝑏�𝑌���, ε2�), (3)\n\n𝑈�= 𝑐�𝑋�, ε3�), (4)\n\n𝑍�= 𝑑�𝑋�, ε4�), (5)\n\n𝑆�= 𝑏�𝑋�, ε5�), (6)\n\n𝑊�= 𝑒�𝑊���, ε6�) (7)\n\n\nwhere 𝐵3 is the Basel III implementation shock, and ε1�…ε6� are shocks to bank conditions, firm\n\nrisk characteristics, property market conditions, conditions of buyers and suppliers, non-bank\n\n\ncredit institutions, and country context, respectively.\n\n\nSubstituting equations (2) to (7) into equation (1), the reduced-form solution takes the following\n\n\nform:\n\n𝐴2𝐹 <sup>���</sup> �𝑔�𝑋���, 𝑌���, 𝑊���, 𝐵3, ε1� … ε6�� (8)\n\n\nIt is used as a basis for the formulation of our reduced-form linear regression in section 6. Although\n\n\nthe simple model in equations (1) to (8) is not derived from behavioral micro-foundations, it\n\n\nillustrates how the thinking framework in figure 1 could be translated into a formal model. In this\n\n\nway, it also helps motivate the regression model specifications that we present in section 6.\n\n\n**4.** **Data**\n\n\nWe use firm-level data from the World Bank’s Enterprise Surveys as our primary source and match\n\n\nthem with several other databases (World Bank’s Global Financial Development Database,\n\n\nBankScope/Fitch, and BIS Financial Stability Institute’s report on the implementation of Basel\n\n\nreforms). The World Bank Enterprise Surveys, which are conducted periodically, are a survey of\n\nMention: \"BankScope/Fitch\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Bank financials are merged in from this commercial source to characterise credit supply.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000102:19:0", "occurrence_id": "6901e0f5741505b29b131bf0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: in the sense that they had never paid the tax, this sample differs in that they had paid tax at\n\n\nleast once, but had overdue taxes for one year or more. Their prior compliance suggests their\n\n\nability to overcome barriers to compliance and thus makes them a more practical focus for\n\n\ntargeted enforcement action. Similar to experiment one, this sample also excluded property\n\n\nowners with the highest tax liabilities (above US$1,000) as the LRA prioritized them for\n\n\nenforcement given their revenue potential. In practice, taxpayers below this threshold often\n\n\neffectively fall out of the radar of the tax authority and do not receive significant enforcement\n\n\nattention.\n\n\nTo locate these delinquent properties, a verification team was dispatched with the name\n\n\nof each owner and the property address on file. The task of the team was to go to the\n\n\nbroad location indicated on the property registration record (e.g., “18th Street” or “Catholic\n\n\nHospital community”) and ask around in the community to identify the property. Once the\n\n\nproperty was verified, the team was to capture the GPS coordinates of the property and a\n\n\nphotograph using a form on a mobile device. Of the 1,382 properties the team attempted to\n\n\nlocate, 728 were identified.\n\n\nProperty owners were randomized into two groups. The control group received a notice\n\n\nrequesting the payment of delinquent tax liability. The notice was comparable to the com\n\nbined identification and penalty notice in experiment one: it was personalized to include the\n\n\nname of the property owner and a photograph of the property and also stated the penalties\n\n\nfor noncompliance. The treatment group received a similar notice (the _enforcement_ _notice_ )\n\n\nwith one addition: it included a sentence informing property owners that their property had\n\n\nbeen included in the current batch of properties designated for intensive follow-up and, if\n\n\nthey did not respond by the deadline, that the LRA would proceed to the next stage of\n\n\nenforcement actions. Figure A2 provides an image of the notice.\n\n\nThe randomization was stratified by two\n\nMention: \"property registration record\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Owner names and addresses on file were used to locate delinquent properties for field verification.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000103:6:0", "occurrence_id": "64418ee512cb3f12ba470039", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: thus directly applicable to subscribers with a sufficient level of mobile phone usage (performance is similar\n\n\nwith lighter usage). Within this population, the data include outcomes from a distribution of individuals\n\n\nwho the telecom might conceivably wish to transition to a postpaid plan. As a result, we can evaluate the\n\n\nperformance of any screening rule, including ones that use credit bureau information, phone indicators, or\n\n\nboth. Second, our sample includes both banked and unbanked consumers, which allows us to both\n\n\nbenchmark our performance against credit bureau models, and also evaluate whether performance is\n\n\ncomparable for individuals without bureau records. The data include each applicant’s mobile phone\n\n\ntransaction history prior to the extension of credit, and whether the credit was repaid on time. We predict\n\n\nwho among these individuals ended up repaying, based on how they used their mobile phones prior to the\n\n\nswitch, in a retrospective analysis. Our data include call and SMS metadata, but not mobile money or top\n\n\nup information. Performance is expected to improve with richer data and larger samples, observed over\n\n\nlonger time periods.\n\n\nAfter developing our method, this paper presents three main findings.\n\n\nFirst, the method has the potential to achieve useful predictive accuracy, even with standard\n\n\nmachine learning models. Performance is assessed in two steps. First, in a sample of formally banked (but\n\n\nthin file) consumers, the method actually outperforms credit bureau models, which perform relatively\n\n\npoorly (we consider the industry standard measure Area Under the ROC Curve or AUC, which ranges from\n\n\n0.61-0.76 for our models versus 0.51-0.57 for bureau models). Second, the method performs similarly well\n\n\nfor unbanked consumers, who cannot be scored with traditional methods (AUC 0.63-0.77). Our models\n\n\nperform within the (wide) range of published estimates of traditional credit scoring in the literature (AUC\n\n\n0.50-0.79). Individuals in the highest quintile of risk by our most conservative measure are 2.\n\nMention: \"credit bureau information\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Bureau-based screening rules are the benchmark the phone-based predictors are evaluated against.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000103:6:0", "occurrence_id": "36a5f68082e9e0ba95735b15", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: thus directly applicable to subscribers with a sufficient level of mobile phone usage (performance is similar\n\n\nwith lighter usage). Within this population, the data include outcomes from a distribution of individuals\n\n\nwho the telecom might conceivably wish to transition to a postpaid plan. As a result, we can evaluate the\n\n\nperformance of any screening rule, including ones that use credit bureau information, phone indicators, or\n\n\nboth. Second, our sample includes both banked and unbanked consumers, which allows us to both\n\n\nbenchmark our performance against credit bureau models, and also evaluate whether performance is\n\n\ncomparable for individuals without bureau records. The data include each applicant’s mobile phone\n\n\ntransaction history prior to the extension of credit, and whether the credit was repaid on time. We predict\n\n\nwho among these individuals ended up repaying, based on how they used their mobile phones prior to the\n\n\nswitch, in a retrospective analysis. Our data include call and SMS metadata, but not mobile money or top\n\n\nup information. Performance is expected to improve with richer data and larger samples, observed over\n\n\nlonger time periods.\n\n\nAfter developing our method, this paper presents three main findings.\n\n\nFirst, the method has the potential to achieve useful predictive accuracy, even with standard\n\n\nmachine learning models. Performance is assessed in two steps. First, in a sample of formally banked (but\n\n\nthin file) consumers, the method actually outperforms credit bureau models, which perform relatively\n\n\npoorly (we consider the industry standard measure Area Under the ROC Curve or AUC, which ranges from\n\n\n0.61-0.76 for our models versus 0.51-0.57 for bureau models). Second, the method performs similarly well\n\n\nfor unbanked consumers, who cannot be scored with traditional methods (AUC 0.63-0.77). Our models\n\n\nperform within the (wide) range of published estimates of traditional credit scoring in the literature (AUC\n\n\n0.50-0.79). Individuals in the highest quintile of risk by our most conservative measure are 2.\n\nMention: \"mobile phone\n\n\ntransaction history\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Pre-credit phone usage is the predictor set used to forecast whether borrowers repaid on time.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000103:6:0", "occurrence_id": "fe21c6ca880c2db26ff60266", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: thus directly applicable to subscribers with a sufficient level of mobile phone usage (performance is similar\n\n\nwith lighter usage). Within this population, the data include outcomes from a distribution of individuals\n\n\nwho the telecom might conceivably wish to transition to a postpaid plan. As a result, we can evaluate the\n\n\nperformance of any screening rule, including ones that use credit bureau information, phone indicators, or\n\n\nboth. Second, our sample includes both banked and unbanked consumers, which allows us to both\n\n\nbenchmark our performance against credit bureau models, and also evaluate whether performance is\n\n\ncomparable for individuals without bureau records. The data include each applicant’s mobile phone\n\n\ntransaction history prior to the extension of credit, and whether the credit was repaid on time. We predict\n\n\nwho among these individuals ended up repaying, based on how they used their mobile phones prior to the\n\n\nswitch, in a retrospective analysis. Our data include call and SMS metadata, but not mobile money or top\n\n\nup information. Performance is expected to improve with richer data and larger samples, observed over\n\n\nlonger time periods.\n\n\nAfter developing our method, this paper presents three main findings.\n\n\nFirst, the method has the potential to achieve useful predictive accuracy, even with standard\n\n\nmachine learning models. Performance is assessed in two steps. First, in a sample of formally banked (but\n\n\nthin file) consumers, the method actually outperforms credit bureau models, which perform relatively\n\n\npoorly (we consider the industry standard measure Area Under the ROC Curve or AUC, which ranges from\n\n\n0.61-0.76 for our models versus 0.51-0.57 for bureau models). Second, the method performs similarly well\n\n\nfor unbanked consumers, who cannot be scored with traditional methods (AUC 0.63-0.77). Our models\n\n\nperform within the (wide) range of published estimates of traditional credit scoring in the literature (AUC\n\n\n0.50-0.79). Individuals in the highest quintile of risk by our most conservative measure are 2.\n\nMention: \"call and SMS metadata\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The available features are call and message metadata, explicitly excluding mobile money and top-ups.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000104:19:0", "occurrence_id": "f46ccce67041ce220fd186fb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: main caregivers, before reporting results on health center utilization. While children’s illnesses could be\n\ninterpreted as a final outcome, we report it along with other intermediary outcomes in the health domain\n\nbecause health center utilization is measured conditionally on reported illnesses.\n\n\nS _timulation practices (Intermediary outcome domain #3)_ . The BCC component encouraged psycho-social\n\nstimulation through language and play. We build two indices based on items capturing the family care\n\nenvironment in MICS surveys. The first index captures the frequency at which children have interactions\n\nwith adults through activities such as reading, storytelling, going out together, playing, naming, counting\n\nor drawing. The second index captures the number of types of toys children play with, including home\nmade toys, manufactured toys, household objects, animals, toys that can be pulled, and balls.\n\n\n_Child protection practices (Intermediary outcome domain #4)_ . The BCC component promoted positive\n\ndisciplining to manage conflict and avoid harsh punishments. It also encouraged birth registration. We\n\nfirst build an index aggregating positive disciplining behavior (explain or redirect) minus negative\n\ndisciplining behavior (forbid, shake, yell, spank, spank with an object, berate, slap, hit on the hands, and\n\nhit with an object). Second, we measure whether children have a birth certificate.\n\n\nUltimately, we are interested in testing whether the promotion of behavioral changes in parenting\n\npractices contribute to improve human capital outcomes among young children. We consider final (or\n\ndownstream) outcomes in three domains: anthropometrics, cognitive development and socio-emotional\n\ndevelopment. We again measure indices for each set of outcomes. <sup>24</sup> [^24: The indices are standardized for the control group. Since the distribution of anthropometric measures have long tails, we trim\nthe top and bottom 2% to rule out outlier effects. We perform robustness checks for alternative treatment of outliers in the\nannex.]\n\n\n_Anthropometrics (Final outcome domain #1)_ . We measure height-for-age (used to identify stunting),\n\nweight-for-height (used\n\nMention: \"MICS surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Two child stimulation indices are constructed from the family care environment items in these surveys.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000106:37:0", "occurrence_id": "d9f44a68628bb645efc41757", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Table 6: Reasons for preferring food or cash payments, 2018**\n\n\n**Reasons for preferring at least part of the PSNP payment in food** **Percent**\n\n\n<mark>With cash payment we cannot buy equal amount of grain</mark> 59.0\n\n\n<mark>Cash can be more easily misused than food, because food can only be consumed in small</mark>\n21.8\n<mark>quantities over time while cash is easily spent in one go</mark>\n\n\n<mark>If the payment is in food, we spend less money and/or time to collect the payment</mark> 7.2\n\n\nThere are typically less delays food payment deliveries compared to cash payments 3.9\n\n\nCompared to cash payments, there is less pressure to the share the food payment with\n3.3\nneighbors, relatives or friends\n\n\nOther 4.8\n\n\n_Source_ : PSNP‐4 midline evaluation survey, 2018.\n\n\n_Note:_ Sample size is 2,434 households.\n\n\n**Reasons for preferring at least part of the PSNP payment in cash** **Percent**\n\n\n<mark>With cash we can buy more grain</mark> 51.1\n\n\n<mark>With cash we can buy other foods that are not part of the PSNP food payment package</mark> 22.3\n\n\n<mark>With cash we can buy non‐food items, save the money for future or invest it</mark> 10.1\n\n\n<mark>There are typically less delays cash payment deliveries compared to food payments</mark> 3.6\n\n\n<mark>If the payment is in cash, we spend less money and/or time to collect the payment</mark> 3.3\nCompared to food payments, there is less pressure to the share the cash payment with\n2.1\nneighbors, relatives or friends\n\nWe have difficulties to store the food payments; part\n\nMention: \"PSNP‐4 midline evaluation survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Stated reasons for preferring food or cash transfers are tabulated from this midline round of 2,434 households.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000108:15:3", "occurrence_id": "eb2d5471e0465f09a1604fd4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: >2546.14</u> <u>59.40</u> <u>61.08</u> <u>59.48</u>\n\nDependent variable is the log of total aid disbursements measured in constant USD excluding emergency aid, food aid, and debt relief (OECD Table 2a). All covariates are in logs, except for democracy.\nDemocracy measure taken from Acemoglu, Naidu, Restrepo, and Robinson (forthcoming) and updated\nin accordance with their methodology. GDP and population data is taken from the WDI database.\nAll regressions use recipient-year observations and include a constant term and period dummy variables, which are not reported. Recipient level cluster-robust standard error reported in parenthesis.\nSignificance levels : _∗_ : 10 percent _∗∗_ : 5 percent _∗∗∗_ : 1 percent\n\n\nTable 2 shows that there is a strong relationship between the lagged CPIA measure and current aid\n\n\nafter controlling for the lagged aid level, population, and income per capita. <sup>4</sup> The table shows OLS\n\n\n4Our aid data are from the OECD, Table 2a, that provides aid data in the donor-recipient-year dimensions.\nWe adjust our aid measure and focus on development aid ignoring emergency and food aid and debt forgiveness\nby following the procedure described in Annen and Kosempel (2009).\n\n\n14\n\nMention: \"GDP and population data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Control covariates in the aid regressions are these macroeconomic series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000108:15:3", "occurrence_id": "1d6ec62f2ade5960ba767802", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: >2546.14</u> <u>59.40</u> <u>61.08</u> <u>59.48</u>\n\nDependent variable is the log of total aid disbursements measured in constant USD excluding emergency aid, food aid, and debt relief (OECD Table 2a). All covariates are in logs, except for democracy.\nDemocracy measure taken from Acemoglu, Naidu, Restrepo, and Robinson (forthcoming) and updated\nin accordance with their methodology. GDP and population data is taken from the WDI database.\nAll regressions use recipient-year observations and include a constant term and period dummy variables, which are not reported. Recipient level cluster-robust standard error reported in parenthesis.\nSignificance levels : _∗_ : 10 percent _∗∗_ : 5 percent _∗∗∗_ : 1 percent\n\n\nTable 2 shows that there is a strong relationship between the lagged CPIA measure and current aid\n\n\nafter controlling for the lagged aid level, population, and income per capita. <sup>4</sup> The table shows OLS\n\n\n4Our aid data are from the OECD, Table 2a, that provides aid data in the donor-recipient-year dimensions.\nWe adjust our aid measure and focus on development aid ignoring emergency and food aid and debt forgiveness\nby following the procedure described in Annen and Kosempel (2009).\n\n\n14\n\nMention: \"WDI database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"This database is named as the origin of the covariates used across recipient-year observations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000111:9:0", "occurrence_id": "81e9685a1707a2a5c39f1a7c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <u>Governance. The main measure of institutional quality is a standardized index that summarizes 6 other</u>\n(also standardized) indices from the Worldwide Governance Indicators, produced by the World Bank\n(see Kaufmann, Kraay and Mastruzzi, 2010 for methodological details). The indices, namely Control\nof Corruption, Government Effectiveness, Political Stability and Absence of Violence/Terrorism, Rule\nof Law, Regulatory Quality and Voice and Accountability, aim to capture different dimensions of\ninstitutions. To construct the governance index, we standardize the sum of the 6 standardized indices\nabove. The data are available for 174 countries from 1996 to 2016.\n\n\nThe indices are built based on surveys conducted with a number of respondents worldwide, as well\nas expert assessments. Following the original definitions in Kaufmann, Kraay and Mastruzzi (2010),\nControl of Corruption reflects perceptions of the “extent to which public power is exercised for private\ngain, including both petty and grand forms of corruption, as well as “capture” of the state by elites and\nprivate interests.” Government Effectiveness measures the “perception of quality of public services, the\nquality of the civil service and the degree of its independence from political pressures, the quality of\npolicy formulation and implementation, and the credibility of the government's commitment to such\npolicies”. Political Stability and Absence of Violence/Terrorism summarizes the perceived “likelihood\nof political instability and/or politically-motivated violence, including terrorism”. Rule of Law reflects\n“the extent to which agents have confidence in and abide by the rules of society, and in particular the\nquality of contract enforcement, property rights, the police, and the courts, as well as the likelihood of\ncrime and violence”. Regulatory Quality captures the “ability of the government to formulate and\nimplement sound policies and regulations that permit and promote private sector development”. And\nlastly, Voice and Accountability summarizes the “extent to which a country's citizens are able\n\nMention: \"Worldwide Governance Indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Institutional quality is measured by a standardised index aggregating six of these component indices.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000112:8:0", "occurrence_id": "37d72acc182d11df43609cf5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ### **III. Implementation**\n\nThe primary sampling strategy was very standard. A nominal sample of 2,800 households was to be selected\nin two stages, with 200 Census sectors -hereafter referred to primary sampling units, or PSUs- in the first\nstage and 14 households per PSU in the second stage. However, the design departed from the standard in\nthe way it addressed stratification in both phases.\n###### III.1 First sampling stage\n\nThe 200 primary sampling units (PSUs) were stratified into three categories depending on the Venezuelan\nmigrant density, defined as the ratio between the number of Venezuelan cellphones in the PSU (as per\nTelefonica's prediction) and the total population of the PSU (as per INEC's population census), and allocated\ninto these strata as follows:\n\n`o` _100 High-density PSUs_ : where the density was more than 0.15.;\n\n`o` _80 Medium-density PSUs_ : where the density was between 0.05 and 0.15.; and\n\n`o` _20 Low-density PSUs_ : where the density was less than 0.05.\n\nWithin each stratum, the sample was selected with probability proportional to the number of households\nreported by the 2010 Census.\n###### III.2 Second sampling stage\n\nAll households in each of the selected sectors were listed and stratified into three categories considering\nnationality and demographic composition. In each stratum, households were selected as follow:\n\n`o` _Seven_ non-Venezuelan households (i.e. Ecuadoran, Peruvian, Colombian or any other nationality except\n\nVenezuelan).\n\n`o` Seven Venezuelan households without children aged less than 5 years. <sup>13</sup>\n\n`o` All Venezuelan households with children aged less than 5 years.\n\n\nWithin each stratum, the sample was selected by systematic equal-probability sampling.\n\n\nThe second stage\n\nMention: \"INEC's population census\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Census population is the denominator for migrant density used to stratify primary sampling units.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000112:8:0", "occurrence_id": "39c1d01265a569798384a41a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ### **III. Implementation**\n\nThe primary sampling strategy was very standard. A nominal sample of 2,800 households was to be selected\nin two stages, with 200 Census sectors -hereafter referred to primary sampling units, or PSUs- in the first\nstage and 14 households per PSU in the second stage. However, the design departed from the standard in\nthe way it addressed stratification in both phases.\n###### III.1 First sampling stage\n\nThe 200 primary sampling units (PSUs) were stratified into three categories depending on the Venezuelan\nmigrant density, defined as the ratio between the number of Venezuelan cellphones in the PSU (as per\nTelefonica's prediction) and the total population of the PSU (as per INEC's population census), and allocated\ninto these strata as follows:\n\n`o` _100 High-density PSUs_ : where the density was more than 0.15.;\n\n`o` _80 Medium-density PSUs_ : where the density was between 0.05 and 0.15.; and\n\n`o` _20 Low-density PSUs_ : where the density was less than 0.05.\n\nWithin each stratum, the sample was selected with probability proportional to the number of households\nreported by the 2010 Census.\n###### III.2 Second sampling stage\n\nAll households in each of the selected sectors were listed and stratified into three categories considering\nnationality and demographic composition. In each stratum, households were selected as follow:\n\n`o` _Seven_ non-Venezuelan households (i.e. Ecuadoran, Peruvian, Colombian or any other nationality except\n\nVenezuelan).\n\n`o` Seven Venezuelan households without children aged less than 5 years. <sup>13</sup>\n\n`o` All Venezuelan households with children aged less than 5 years.\n\n\nWithin each stratum, the sample was selected by systematic equal-probability sampling.\n\n\nThe second stage\n\nMention: \"2010 Census\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Household counts determine the probability-proportional-to-size selection within strata.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000126:10:0", "occurrence_id": "4dcf593a8aeaa6daab602a44", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: for an average of 8.3% of the total industrial value added. Such a small proportion precludes the NBS\n\nsurvey teams from effective checks on the rest of the statistics from the local statistical agencies.\n\n\nFurthermore, data on industrial firms suffer from limited information disclosure, further\n\nundermining monitoring from the public. As of now, only the aggregate industrial value-added data\n\nare publicly available at the province and the prefecture level, and these data cannot even be\n\ndecomposed into the one- or two-digit subsectors. <sup>17</sup> Indeed, the largest discrepancy between the total\n\nnational value added and the sum of the corresponding provincial values is found in the industrial\n\nsector. Finally, neither the census nor other data of the NBS are capable of adequately adjusting these\n\nskewed statistics. For example, while the 2008 Benchmark Revisions following the census reduced\n\nthe overall value-added discrepancy as well as discrepancies in the tertiary sector, the data adjustments\n\nonly eliminated a small proportion of the discrepancy in the industrial sector.\n\n\n**2.2.2** **Setting growth target**\n\n\nChinese governments at all levels make their economic plans in advance, setting key economic targets,\n\nincluding GDP growth rates, industry value added, fiscal revenues and expenditures. The drafting of\n\nplans involves a top-down process: In December, the Central Economic Working Conference, jointly\n\nheld by the Central Committee of Communist Party and the State Council, reviews economic\n\nperformance in the current year, and then sets the national growth target for the next year. Provincial\n\nand prefecture governments then set their own targets on the basis of the national targets, taking into\n\naccount the local conditions (Lyu et al. 2018). <sup>18</sup> These annual plans are released in the government\n\nwork reports, and presented in the local People’s Congress at the beginning of each year.\n\n\n**2.3** **Evidence of manipulation at the\n\nMention: \"data of the NBS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Neither the census nor other official series can correct the skewed provincial value-added statistics.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000127:43:0", "occurrence_id": "03ca321fac6c06cbc90620a2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Map 9 : Employment densities for each census tract (Radio censal) unit within the Buenos Aires region in 2012 (data source:**\n**projections based on Censo Nacional Economic 04/05 (CNE 04/05) and Encuesta Permanente de Hogares, 2011 (EPH, 2011))**\n\n\nMap 9 shows that job dispersal exists along the major public transport lines, the railway network\n(represented as black lines), but is nearly non-existent elsewhere. This suggests that the radial public\ntransportation network plays a strong role in the location of employment centers which is characteristic\nof dominantly mono-centric urban forms.\n\n\n42\n\nMention: \"Censo Nacional Economic 04/05\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Employment densities by census tract are projected from this economic census combined with a household survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000127:43:0", "occurrence_id": "824634baf79d7adf1a37d1f8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Map 9 : Employment densities for each census tract (Radio censal) unit within the Buenos Aires region in 2012 (data source:**\n**projections based on Censo Nacional Economic 04/05 (CNE 04/05) and Encuesta Permanente de Hogares, 2011 (EPH, 2011))**\n\n\nMap 9 shows that job dispersal exists along the major public transport lines, the railway network\n(represented as black lines), but is nearly non-existent elsewhere. This suggests that the radial public\ntransportation network plays a strong role in the location of employment centers which is characteristic\nof dominantly mono-centric urban forms.\n\n\n42\n\nMention: \"Encuesta Permanente de Hogares\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The household survey supplies the more recent employment information used in the projections.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000131:7:1", "occurrence_id": "0710d1977d7b572a76636877", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ij_ _∗_ _AFj_ (2)\n\n\n\nwhere _Shareij_ is the fraction of individual _i_ ’s time spent on activity _j_ . _AFj_\nis an activity factor for activity _j_ that measures the intensity of activity _j_\nrelative to rest (equal to one). Individual _AL_ are thus weighted averages\nof activity factors where the weights vary across individuals based on their\ntime-use.\n\nThe most detailed activity data for India comes from the Time-Use Survey\n(TUS) conducted in 1998-1999 by the National Sample Survey Organization\nfor six states (Gujarat, Haryana, Madhya Pradesh, Meghalaya, Orissa, and\n\n\n4Validation studies by (Shetty et al. 1986) and (Ferro-Luzzi et al. 1997) find that\nthese types of prediction equations, which are typically estimated on European descended\npopulations, have a reasonably good fit for subjects in India.\n\n\n6\n\nMention: \"Time-Use Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Activity weights come from this detailed survey of time use across six Indian states.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000132:4:1", "occurrence_id": "494b1f0f8279d25fee832351", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Nigeria and also describe that GHS-panel data used in the analysis.\nSection III lays out the methodology used to examine these two linkages followed by a\npresentation and discussion of the results in Section IV. Section V concludes.\n\n\n**II.** **BACKGROUND AND DATA**\n\n\nThe agricultural sector in Nigeria consists mostly of smallholder farmers who operate at the\nsubsistence level. Despite concerns about the sector’s performance, it remains an important\nsource of rural employment and income. There are many constraints facing the agricultural\nsector in Nigeria, and one of the main constraints is the low rates of input use, particularly\nfertilizer which is also commonly of poor quality due to adulteration <sup>1</sup> . In their review of\nconstraints to agriculture productivity in Nigeria, Philip et al. (2009) attribute the low levels of\ninput use to high transport costs, poor distribution channels, absence of private sector\nparticipation, significant import risk, and inconsistent government policies. Access to\nagricultural credit and efficient coordination of agricultural extension services can have\nbeneficial spillover effects on the adoption of farm technologies, and these are also found to be\nlacking in the Nigerian agricultural sector (Philip et al., 2009). Poor access to input and output\nmarkets (largely due to poor rural infrastructure), land degradation, and low investment in\nagricultural research have also been identified as significant constraints to productivity in the\n\n\n1 See Liverpool-Tessie et al. (2010) for an assessment of fertilizer quality regulation in Nigeria\n\n\n3\n\nMention: \"GHS-panel data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The analysis is conducted on this panel of Nigerian households.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000145:31:1", "occurrence_id": "d1ab3a5c4c14e28e80c9a015", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  the US. Overall, these findings suggest that higher network resilience against natural shocks is also likely\nto result in a reduction of baseline fragility of the network, and thus higher service quality for users.\nMoreover, we find robust empirical evidence of adaptation effects. This implies that at given levels of\nbaseline fragility, grid operators can exploit experience to design measures specifically effective against\nnatural hazards.\n\n\n##### References\n\nAllcott, H., Collard-Wexler, A. and O’Connell, S. D. (2016) ‘How Do Electricity Shortages Affect Industry?\n\n\n\nEvidence from India’, _American Economic Review_, 106(3), pp. 587–624. Available at:\nhttp://www.aeaweb.org/articles?id=10.1257/aer.20140389.\nBangladesh Bureau of Statistics (2010) ‘Report of the Household Income & Expenditure 2010’,\n\n\n\n_Bangladesh Bureau of Statistics. Ministry of Planning_, (1), pp. 1–5. doi: 10.1007/s13398-014-01737.2.\nBartos, M. _et al._ (2016) ‘Impacts of rising air temperatures on electric transmission ampacity and peak\n\nelectricity load in the United States’, _Environmental Research Letters_ . IOP Publishing, 11(11). doi:\n10.1088/1748-9326/11/11/114008.\nDewan, A. _et al._ (2017) ‘Spatial and temporal analysis of a 17-year lightning climatology over Bangladesh\n\nwith LIS data’, _Theoretical and Applied Climatology_ . Theoretical and Applied Climatology, pp. 1–16.\ndoi: 10.1007/s00704-017-2278-3.\nEuropean Commission (2018) _Study on the quality of electricity market data of transmission system_\n\n_operators,\n\nMention: \"LIS data\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"Cited in the references for a lightning climatology study using this satellite sensor.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000147:43:0", "occurrence_id": "044b85457f195506b8c75ea2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **FIGURE 5. CAPITAL MARKET FINANCING AND INDUSTRY PRICES**\nThis figure shows the amount raised per year in capital markets and the annual commodity price index for the mining industry as a whole\n(Panel A) and for a subgroup of mining commodities (Panels B-F). Prices are normalized to 100 in year 2010. The price index for the overall\nmining industry is a weighted average of 18 commodity prices. The weight is calculated based on the commodity's global import share over\nthe period 2014-16. Issuance volumes are reported in millions of constant 2011 U.S. dollars.\n\n\n**A. Mining** **B. Coal Copper**\n\n\n\n400,000\n\n350,000\n\n300,000\n\n250,000\n\n200,000\n\n150,000\n\n100,000\n\n50,000\n\n0\n\n\n30,000\n\n\n25,000\n\n\n20,000\n\n\n15,000\n\n\n10,000\n\n\n5,000\n\n\n0\n\n\n25,000\n\n\n20,000\n\n\n15,000\n\n\n10,000\n\n\n5,000\n\n\n0\n\n\n\n140\n\n\n120\n\n\n100\n\n\n80\n\n\n60\n\n\n40\n\n\n20\n\n\n0\n\n\n\n140\n\n\n120\n\n\n100\n\n\n80\n\n\n60\n\n\n40\n\n\n20\n\n\n0\n\n\n\n70,000\n\n\n60,000\n\n\n50,000\n\n\n40,000\n\n\n30,000\n\n\n20,000\n\n\n10,000\n\n\n0\n\n\n\n**C. Copper** **D. Ferroalloy Ores**\n\n\n\n140\n\n\n120\n\n\n100\n\n\n80\n\n\n60\n\n\n40\n\n\n20\n\n\n0\n\n\n\n8,000\n\n\n6,000\n\n\n4,000\n\n\n2,000\n\n\n0\n\n\n\n**E. Iron** **F. Zinc and Lead**\n\n\n\n140\n\n\n120\n\n\n100\n\n\n80\n\n\n60\n\n\n40\n\n\n20\n\n\n0\n\n\n180\n160\n140\n120\n100\n80\n60\n40\n20\n0\n\n\n160\n\n140\n\n120\n\n100\n\n80\n\n60\n\n40\n\n20\n\n0\n\n\n\n2,500\n\n\n2,000\n\n\n1,500\n\n\n1,000\n\n\n500\n\n\n0\n\n\n\nIssuance in Capital Markets Commodity Price Index\n\nMention: \"annual commodity price index\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Capital market issuance is plotted against a weighted index of 18 commodity prices normalised to 2010.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000147:43:0", "occurrence_id": "a9cedc8e34b4dda0b854a1ac", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **FIGURE 5. CAPITAL MARKET FINANCING AND INDUSTRY PRICES**\nThis figure shows the amount raised per year in capital markets and the annual commodity price index for the mining industry as a whole\n(Panel A) and for a subgroup of mining commodities (Panels B-F). Prices are normalized to 100 in year 2010. The price index for the overall\nmining industry is a weighted average of 18 commodity prices. The weight is calculated based on the commodity's global import share over\nthe period 2014-16. Issuance volumes are reported in millions of constant 2011 U.S. dollars.\n\n\n**A. Mining** **B. Coal Copper**\n\n\n\n400,000\n\n350,000\n\n300,000\n\n250,000\n\n200,000\n\n150,000\n\n100,000\n\n50,000\n\n0\n\n\n30,000\n\n\n25,000\n\n\n20,000\n\n\n15,000\n\n\n10,000\n\n\n5,000\n\n\n0\n\n\n25,000\n\n\n20,000\n\n\n15,000\n\n\n10,000\n\n\n5,000\n\n\n0\n\n\n\n140\n\n\n120\n\n\n100\n\n\n80\n\n\n60\n\n\n40\n\n\n20\n\n\n0\n\n\n\n140\n\n\n120\n\n\n100\n\n\n80\n\n\n60\n\n\n40\n\n\n20\n\n\n0\n\n\n\n70,000\n\n\n60,000\n\n\n50,000\n\n\n40,000\n\n\n30,000\n\n\n20,000\n\n\n10,000\n\n\n0\n\n\n\n**C. Copper** **D. Ferroalloy Ores**\n\n\n\n140\n\n\n120\n\n\n100\n\n\n80\n\n\n60\n\n\n40\n\n\n20\n\n\n0\n\n\n\n8,000\n\n\n6,000\n\n\n4,000\n\n\n2,000\n\n\n0\n\n\n\n**E. Iron** **F. Zinc and Lead**\n\n\n\n140\n\n\n120\n\n\n100\n\n\n80\n\n\n60\n\n\n40\n\n\n20\n\n\n0\n\n\n180\n160\n140\n120\n100\n80\n60\n40\n20\n0\n\n\n160\n\n140\n\n120\n\n100\n\n80\n\n60\n\n40\n\n20\n\n0\n\n\n\n2,500\n\n\n2,000\n\n\n1,500\n\n\n1,000\n\n\n500\n\n\n0\n\n\n\nIssuance in Capital Markets Commodity Price Index\n\nMention: \"Issuance in Capital Markets Commodity Price Index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The legend identifies the two series being compared across the commodity panels.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000151:8:1", "occurrence_id": "1f655cc30e46dc2443b8368c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _ per child conditional on\nthe child’s school attendance and health checks. Families with income per capita above\nthe extreme poverty threshold but below a higher poverty threshold are only eligible for\nthe conditional variable benefits. Benefits are paid to families on a monthly basis from the\nfederal budget, typically into a bank account at _Caixa_, the main state bank in Brazil.\n\nThe relevant income per capita definition is the one recorded in _Cadastro Único_, which\nis a federal registry that is continuously updated. _Cadastro Único_ was created at the same\ntime as PBF and was the result of a similar consolidation effort. Yet, its purpose is to serve\nas basis for other social programs as well. For this reason, it aims to include families with\nincome per capita below one-half of the minimum wage (R$255 in 2010), which is much\nhigher than the two poverty thresholds for PBF eligibility (R$70 and R$140 in 2010). <sup>16</sup>\n\nThe information in _Cadastro_ _Único_ is based on a standardized survey that asks families\nabout income, housing, and assets, among other characteristics. Families can apply to\ntheir municipality to take the survey and be registered in _Cadastro_ _Único_ . Municipalities\ncan also actively identify poor families and survey them, which was the main channel\n\n\n16For reference, the exchange rate in 2010 was about R$2=US$1.\n\n\n7\n\nMention: \"_Cadastro Único_\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The income per capita used for eligibility is the continuously updated value recorded in this federal registry.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000153:19:0", "occurrence_id": "33ad3d28c6016eb6807851d7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: indicator for the extent of regulatory burden on the private firms obtained from the Doing Business\n\n\nproject (World Bank); Business Freedom index obtained from Heritage Foundation’s Economic\n\n\nFreedom in the World that aims to capture (lack of) government interference with the private\n\n\nsector. Note that higher values of the Corruption, DTP and Business Freedom index imply a better\n\n\nbusiness environment (less corruption, less burdensome regulations for the private firms, and more\n\n\nfreedom from government interference).\n\n\n4.2 _Base regression results_\n\n\nOur baseline regression results are provided in Table 6. These results focus on the linear\n\n\nrelationship (without any interaction terms) between labor productivity and informal competition.\n\n\nResults with the interaction terms included are discussed in the next section.\n\n\nRegression results in Table 6 show a large and statistically significant (at the 1 percent\n\n\nlevel) negative relationship between labor productivity and informal competition. This holds with\n\n\nor without the various controls and for all the specifications shown in Table 6. Excluding controls,\n\n\nthe estimated coefficient value of the informal competition variable equals -0.767 (column 1). The\n\n\nestimated value implies that the labor productivity of firms experiencing the highest level of\n\n\ninformal competition (equal to 1) is lower by 54 percent than the labor productivity of firms that\n\n\nface the lowest level of informal competition (equal to 0). Controlling for country fixed effect\n\n\nraises the estimated coefficient value of informal competition variable from -0.767 above to -0.178\n\n\n(column 2). The sharp drop in absolute terms suggests that a large part of the negative labor\n\n\nproductivity and informal competition relationship comes from across country rather than within\n\n\ncountry effects. Controlling for the various firm characteristics increases the estimated coefficient\n\n\nvalue of informal competition (in absolute terms) from -0.178 above to -0.268 (column 5). The\n\n\n18\n\nMention: \"Doing Business\n\n\nproject\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The regulatory burden measure applied to private firms is taken from this project.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000155:11:1", "occurrence_id": "691be3dbd16ec708358a4db6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  the average GPS-based plot area is 0.39 hectare, female-managed plots are, on\naverage, 12 percent smaller than their male-managed counterparts. The use of inorganic fertilizer\nis lower on female-managed plots, whether measured by incidence, average unconditional\namount per hectare, or average conditional amount per hectare. These trends may signal gender\ndifferences in FISP fertilizer voucher distribution and redemption outcomes. In fact, the IHS3\ndata indicates that even though female-headed households are just as likely to receive a fertilizer\nvoucher as their male-headed counterparts, conditional on receipt, the _average number_ of\nfertilizer vouchers that are received among female-headed households (1.56) is lower than the\nanalogous statistic for male-headed households (1.63), and the difference is statistically\n\n\nleast one missing value among the independent variables of interest. Finally, top and bottom 1 percent of the\ndistribution of the log of gross value of output per hectare are trimmed, corresponding to 336 plots. These exclusions\nleave us with the final analysis sample of 16,372 rainy season plots.\n26 The plot-level gross value of output in Malawi Kwacha (MK) is calculated by first multiplying the kilogramequivalent quantity of production for each crop on a given plot by the median crop sales value per kilogram within\nthe corresponding EA, and then aggregating across values of crop production. The median crop sales value per\nkilogram is computed within the corresponding EA only if at least 10 values are available from the survey data.\nOtherwise, the median crop sales value per kilogram is computed at a higher level, in the order of traditional\nauthority, district, region, and country. Our outcome variable is computed by normalizing plot-level gross value of\noutput with GPS-based cultivated plot area.\n27 The overwhelming majority of the owned plots (83 percent) are acquired through inheritance. Another 12 percent\nis reported to have been granted by local leaders. The remaining are acquired as bride price\n\nMention: \"IHS3\ndata\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"This household survey shows female-headed households receive vouchers as often but fewer per household.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000156:10:1", "occurrence_id": "f6e6ab0a0aff2cdbd5df0f70", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \n<mark>currency, and banking sector risk scores of a given country.</mark> <sup>12</sup> [^12: See https://www.eiu.com/n/solutions/country-risk-model/.\n13. Anasashvili et al. (2020); see\n<u>[https://openknowledge.worldbank.org/bitstream/handle/10986/33657/K880541.pdf?sequence=1&isAllowed=y.](https://openknowledge.worldbank.org/bitstream/handle/10986/33657/K880541.pdf?sequence=1&isAllowed=y)</u>] <mark>W</mark> e included this variable in the\nanalysis to assess the macroenvironment in which central banks operate. Alternatively, we use\nthe number of currency crises and the current account balance as a percentage of GDP as another\nmeans of describing the macroenvironment in which the central bank operates.\n\n**_Reserve Adequacy Variables._** Finally, a central bank's risk tolerance and investment policy\n(number of eligible assets and currencies, portfolio duration, allocation to nontraditional asset\nclasses) are linked to reserve adequacy. All else being equal, a central bank with higher reserve\nadequacy typically has a higher risk tolerance and can deploy a broader range of asset classes,\ncurrencies, and longer duration. Therefore, we use reserve adequacy measures to control for\ndifferences in reserve adequacy across central banks. We include two reserve adequacy metrics\nas control variables: reserves to GDP and to M2. Separately, we also control for GDP per capita\nand other macroeconomic variables, such as short-term external debt stocks as a percentage of\ntotal reserves.\n\n**4.2** **RAMP Survey Variables**\n\nRAMP's global surveys on reserve management practices are the source for information on\ncentral banks' governance arrangements, investment policies, and measures of risk taking. RAMP\nconducted two surveys in 2017 and 2019 with responses from\n\nMention: \"RAMP Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Central bank governance, investment policy and risk-taking measures come from these global surveys.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000158:16:0", "occurrence_id": "f170fabd314e0148fb69b3a0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Source: World Development Indicators, World Bank\n\n\n<u>Source: Mexican Secretariat of Labor and Social Welfare (STPS).</u>\nNotes: Difference between male and female labor force participation rates.\n\n\nNotes: Seasonally adjusted.\n\n\n\n15\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"One of the two figure panels is sourced from this indicator database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000159:16:0", "occurrence_id": "6e69187421c8c13885741b2a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 12: Leadership approval ratings & local death count**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nSource: Gallup World Poll and Iraq Body Count (IBC). <sup>3</sup>\n\n\n**5. Quality of Service Delivery**\n\n\nThe Gallup World Poll includes questions on local public services. Among adults in Iraq, 71 percent are\n\nsatisfied with services in the city or area where they live as of 2019, up from 59 percent in 2010 (figure\n\n13). This trend holds across all ethno-religious groups. For each of the public services featured in the\n\nsurvey, women are significantly less likely than men to report satisfaction, and the poor are significantly\n\nless likely to be satisfied than the rich. Some of the largest income and gender gaps are reported for health\n\ncare and quality of air.\n\n\n3 Gallup World Poll data are only available for 13 of the 19 regions. Halabja was pooled with data from Sulaymaniya.\n\n15\n\nMention: \"Gallup World Poll\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Approval ratings and satisfaction with local services are taken from this poll.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000159:16:0", "occurrence_id": "ba5d15648c3b9087a8283fa7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 12: Leadership approval ratings & local death count**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nSource: Gallup World Poll and Iraq Body Count (IBC). <sup>3</sup>\n\n\n**5. Quality of Service Delivery**\n\n\nThe Gallup World Poll includes questions on local public services. Among adults in Iraq, 71 percent are\n\nsatisfied with services in the city or area where they live as of 2019, up from 59 percent in 2010 (figure\n\n13). This trend holds across all ethno-religious groups. For each of the public services featured in the\n\nsurvey, women are significantly less likely than men to report satisfaction, and the poor are significantly\n\nless likely to be satisfied than the rich. Some of the largest income and gender gaps are reported for health\n\ncare and quality of air.\n\n\n3 Gallup World Poll data are only available for 13 of the 19 regions. Halabja was pooled with data from Sulaymaniya.\n\n15\n\nMention: \"Iraq Body Count\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Local death counts plotted against approval come from this casualty recording project.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000159:16:0", "occurrence_id": "288fd0a7e7236dce38c8b5d5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 12: Leadership approval ratings & local death count**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nSource: Gallup World Poll and Iraq Body Count (IBC). <sup>3</sup>\n\n\n**5. Quality of Service Delivery**\n\n\nThe Gallup World Poll includes questions on local public services. Among adults in Iraq, 71 percent are\n\nsatisfied with services in the city or area where they live as of 2019, up from 59 percent in 2010 (figure\n\n13). This trend holds across all ethno-religious groups. For each of the public services featured in the\n\nsurvey, women are significantly less likely than men to report satisfaction, and the poor are significantly\n\nless likely to be satisfied than the rich. Some of the largest income and gender gaps are reported for health\n\ncare and quality of air.\n\n\n3 Gallup World Poll data are only available for 13 of the 19 regions. Halabja was pooled with data from Sulaymaniya.\n\n15\n\nMention: \"Gallup World Poll data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Coverage is limited to 13 of 19 regions, so the regional analysis is incomplete.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000159:16:0", "occurrence_id": "b5cea6b1de9df820a70b08e9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 12: Leadership approval ratings & local death count**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nSource: Gallup World Poll and Iraq Body Count (IBC). <sup>3</sup>\n\n\n**5. Quality of Service Delivery**\n\n\nThe Gallup World Poll includes questions on local public services. Among adults in Iraq, 71 percent are\n\nsatisfied with services in the city or area where they live as of 2019, up from 59 percent in 2010 (figure\n\n13). This trend holds across all ethno-religious groups. For each of the public services featured in the\n\nsurvey, women are significantly less likely than men to report satisfaction, and the poor are significantly\n\nless likely to be satisfied than the rich. Some of the largest income and gender gaps are reported for health\n\ncare and quality of air.\n\n\n3 Gallup World Poll data are only available for 13 of the 19 regions. Halabja was pooled with data from Sulaymaniya.\n\n15\n\nMention: \"data from Sulaymaniya\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Halabja was pooled with the neighbouring region because no separate observations exist.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000160:15:0", "occurrence_id": "f35181f76f0ca4027c195d77", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Figure 4: PM2.5 Diurnal profiles for selected monitoring locations_\n\n\n**Legend: Based on the US-EPA Air Quality Index (AQI)**\n\n\n**Weekly Variations (7-day period)**\nDaily PM2.5 mass concentration over a 7-day period ( **figure 5** ) ranges from **~** 30 to 107µg/m <sup>3</sup> . However,\npollution levels in all monitoring locations and geographical clusters show no apparent correlation across\nall areas i.e. do not conform to particular patterns with different locations having different days of peak\npollution. Kireka, Kawempe and Bunamwaya are exceptions ( **figure 5a** ) with peak pollution levels\nobserved in all cases on Wednesday, Thursday, Friday and Saturday throughout the 6-month period.\nTemporal peaks for Kireka and Kawempe were over 105µg/m <sup>3</sup> and 80µg/m <sup>3</sup> on Wednesday and Thursday,\nwhile Bunamwaya followed suit with a similar pattern from Tuesday to Sunday with highest pollution\nlevels recorded on Wednesday (~55µg/m <sup>3</sup> ). However, all geographical clusters showed haphazard trends\nin the weekly pollution levels but clusters outside Central maintained lower temporal profiles (highest level\nbeing Sunday about 63µg/m <sup>3</sup> experienced on Sunday in Jinja (Buwenge). On the contrary, Entebbe 4b and\nBusia 4c experienced the lowest pollution levels (below 35µg/m <sup>3</sup> ) on Sunday, Monday (Entebbe) and\nSaturday (Busia) ( **figure 5b and 5c** ).\n\nHaphazard pollution patterns\n\nMention: \"US-EPA Air Quality Index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"The figure legend states that measured concentrations are classified using this index.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000164:36:3", "occurrence_id": "b3b5c48f657005b8e45bb6e8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 1.183)\n\nThis table reports the marginal effects from the bivariate probit regression and 2SLS of dummy employment variables\non child care attendance and the control variables of women. For the dependent variables of wages and working hours,\nthe regressions are 2SLS.\nThe observations in these regressions are women of children age 1-5.\nHeteroskedasticity-robust standard errors in parentheses. Standard errors are corrected for sampling weights and cluster\ncorrelation at the commune level.\n*** p<0.01, ** p<0.05, * p<0.1.\nSource: Estimation from VHLSS 2010, 2012, 2014 and 2016.\n\n\n\n35\n\nMention: \"VHLSS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Marginal effects are estimated on four pooled rounds of this household living standards survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000167:7:0", "occurrence_id": "d1f4de054e0395929756282e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: found that point-of-use water treatment is among the most effective WASH interventions to reduce\n\n\ndiarrheal disease incidence (Fewtrell et al 2005; Clasen et al 2007; Wolf et al 2018). Further, a meta\nanalysis examined 14 studies on various WASH interventions (improved microbiological quality of\n\n\ndrinking water, water quantity and supply, sanitation coverage, handwashing, and child feces disposal)\n\nand found that interventions with solar disinfection of water, provision of soap, and improvement of\n\n\nwater quality) had some benefit on the linear growth of children under 5 (Dangour et al 2013).\n\nHowever, two recently published RCTs evaluating the effectiveness of chlorination tablets in rural\n\n\nBangladesh and Kenya found that the water quality intervention had no impact on diarrheal disease\n\nincidence or anthropometric measures (Luby et al 2018; Null et al 2018). The conclusions of these\n\n\nstudies do not necessarily imply that there is no link between contaminated water and nutrition, but\n\nmore likely that the interventions did not effectively limit exposure to feces in a sustainable manner.\n\n\nFurther research is needed in understanding the relationship between pathways and quantities of fecal\n\nexposure and nutritional status of children.\n\n#### 3. Methodology\n\n###### **3.1   Data**\n\n\nWe use cross-sectional data from the 2012-13 MICS survey for Bangladesh, a nationally\n\n\nrepresentative survey with various indicators on household characteristics across urban and rural\n\nregions of Bangladesh. The 2012-2013 MICS is the first and latest national representative survey that\n\n\nprovides water quality data on E. coli contamination in drinking water sources in Bangladesh. Globally,\n\nthere are very few other data sets that examine microbial water quality indicators on a national scale.\n\n\nWe particularly examine indicators on E. coli contamination in drinking water and childhood stunting.\n\nThe independent variables include E. coli contamination at point-of-source (e.g. water quality results\n\n\nfrom the household’s main reported source of drinking water) and\n\nMention: \"2012-13 MICS survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The analysis uses this nationally representative cross-section of household characteristics.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000167:7:0", "occurrence_id": "aa242515cdebabc837c0ebb1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: found that point-of-use water treatment is among the most effective WASH interventions to reduce\n\n\ndiarrheal disease incidence (Fewtrell et al 2005; Clasen et al 2007; Wolf et al 2018). Further, a meta\nanalysis examined 14 studies on various WASH interventions (improved microbiological quality of\n\n\ndrinking water, water quantity and supply, sanitation coverage, handwashing, and child feces disposal)\n\nand found that interventions with solar disinfection of water, provision of soap, and improvement of\n\n\nwater quality) had some benefit on the linear growth of children under 5 (Dangour et al 2013).\n\nHowever, two recently published RCTs evaluating the effectiveness of chlorination tablets in rural\n\n\nBangladesh and Kenya found that the water quality intervention had no impact on diarrheal disease\n\nincidence or anthropometric measures (Luby et al 2018; Null et al 2018). The conclusions of these\n\n\nstudies do not necessarily imply that there is no link between contaminated water and nutrition, but\n\nmore likely that the interventions did not effectively limit exposure to feces in a sustainable manner.\n\n\nFurther research is needed in understanding the relationship between pathways and quantities of fecal\n\nexposure and nutritional status of children.\n\n#### 3. Methodology\n\n###### **3.1   Data**\n\n\nWe use cross-sectional data from the 2012-13 MICS survey for Bangladesh, a nationally\n\n\nrepresentative survey with various indicators on household characteristics across urban and rural\n\nregions of Bangladesh. The 2012-2013 MICS is the first and latest national representative survey that\n\n\nprovides water quality data on E. coli contamination in drinking water sources in Bangladesh. Globally,\n\nthere are very few other data sets that examine microbial water quality indicators on a national scale.\n\n\nWe particularly examine indicators on E. coli contamination in drinking water and childhood stunting.\n\nThe independent variables include E. coli contamination at point-of-source (e.g. water quality results\n\n\nfrom the household’s main reported source of drinking water) and\n\nMention: \"2012-2013 MICS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"It is the only national survey providing E. coli contamination measurements for drinking water sources.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000169:20:0", "occurrence_id": "53af2aa2dcfbe2f92e53e478", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: China will also have to continue to invest in its human capital, as the quality of its labor force\ncatches up to the OECD countries (Figure 25). Over 30 years, China has made primary\neducation universal and expanded access to education at all levels. To obtain greater benefits\nfrom technology diffusion from abroad and to nurture frontier discovery, China now has to\nfocus on investing in the skills of its labor force (Eaton and Kortum, 1996; Seck, 2012).\n\n\n\n**Figure 24 China’s TFP relative to the**\n**global technology frontier, 2017**\n\n(both axes in log scale)\n\n\nSource: PWT 9.1; authors’ calculations.\nNote: OECD countries in orange.\n\n\n\n**Figure 25 Human capital index, 2017**\n\n(x-axis in log scale)\n\n\n\nSource: PWT 9.1; authors’ calculations.\nNote: OECD countries in orange.\n\n\n\n\n\nOver the past four decades, China has made such great strides in physical capital deepening\nthat its government capital stock per worker has already reached OECD levels (Figure 26). <sup>20</sup> [^20: Government investment in the IMF data set accounts for capital spending by central and subnational\ngovernments, so that comparable data are available for a large number of countries. The total infrastructure stock\nfor China is possibly underestimated, as infrastructure investment by state-owned enterprises (SOEs) is included\nin the private capital stock. For details refer to\n[https://www.imf.org/external/np/fad/publicinvestment/data/info.pdf.](https://www.imf.org/external/np/fad/publicinvestment/data/info.pdf)]\nThis suggests that there may be limited economic gains from future high growth in government\ninvestment. But there is scope for rapid growth of private capital. Starting from a very low base\nin the 1970s, private sector capital increased at a remarkable pace but remains considerably\nbelow advanced country levels\n\nMention: \"IMF data set\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Government capital spending across central and subnational tiers is taken from this comparable dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000174:31:0", "occurrence_id": "52d91515e1d6031bb53b5721", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Figure 8 - Sectoral breakdown of credit portfolios by transition-sensitive industries (share of total credit portfolios)_\n\n\n**Source** : Own elaboration.\n**Note** : *Estimated shares for heavy industry. In the cases of Bolivia and the Dominican Republic the credit outstanding information\ndoes not allow us to identify the transition-sensitive sectors within manufacturing according to Table C1 of the Annex. We took\nthe credit outstanding share for the whole manufacturing sector and subtracted the proportion of not transition-sensitive sectors\nbased on the GDP sectoral distribution of each country. See Annex B for additional classification details.\n\n\n_Figure 9 – Interest coverage ratio (ICR), latest available data (2018 or 2019)_\n\n\n_Panel A – Share of firms                        Panel B – Share of debt_\n\n\n**Note** : Sample covers 38,227 firms with total outstanding debt of US$ 2,112 billion. See Table 4 for sample details.\n**Source** : Own elaboration based on Orbis data.\n\n\n30\n\nMention: \"credit outstanding information\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"For two countries the credit records cannot resolve manufacturing subsectors, so shares are estimated by subtraction.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000174:31:0", "occurrence_id": "a79755e4f51c20d179b5e5f1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Figure 8 - Sectoral breakdown of credit portfolios by transition-sensitive industries (share of total credit portfolios)_\n\n\n**Source** : Own elaboration.\n**Note** : *Estimated shares for heavy industry. In the cases of Bolivia and the Dominican Republic the credit outstanding information\ndoes not allow us to identify the transition-sensitive sectors within manufacturing according to Table C1 of the Annex. We took\nthe credit outstanding share for the whole manufacturing sector and subtracted the proportion of not transition-sensitive sectors\nbased on the GDP sectoral distribution of each country. See Annex B for additional classification details.\n\n\n_Figure 9 – Interest coverage ratio (ICR), latest available data (2018 or 2019)_\n\n\n_Panel A – Share of firms                        Panel B – Share of debt_\n\n\n**Note** : Sample covers 38,227 firms with total outstanding debt of US$ 2,112 billion. See Table 4 for sample details.\n**Source** : Own elaboration based on Orbis data.\n\n\n30\n\nMention: \"Orbis data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Firm-level debt for 38,227 companies is drawn from this commercial database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000175:5:1", "occurrence_id": "0d2f3068dc2951ba9127119e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: , vulnerability, resilience, and livelihoods are complex.\nClimate change can increase poverty directly by reducing agricultural productivity and\nproduction, and by hindering asset accumulation and return on assets. Indirectly, climate\nchange affects poverty through output prices, labor productivity and the availability of offfarm employment opportunities. This paper focuses primarily on the direct livelihood effects.\nA <mark>s in other mainly agrarian SSA countries, the incidence of climate shocks such as droughts,</mark>\nseasonal and flash <mark>floods, and e</mark> xtreme temperatures and dry spells is increasing in Zambia.\n<mark>These weather extremes</mark> have significant carryover side effects on rural farmers who depend\non rainfed agriculture for their livelihoods.\n\n\nClimate shocks are likely to become more intense and more frequent. Ahmed et al. (2009)\npredict that by 2080, climate change will increase the intensity of dry spells and subsequently\nworsen the incidence of poverty among smallholder farmers by 4.6 percentage points in all\nagro-ecological regions in Zambia. Average temperatures are predicted to rise by between\n\n\n5 From which the national poverty rate is derived.\n6 At the time of writing, the 2015 LCMS is the data set that provides the most recent estimate of national,\nurban/rural, and provincial poverty rates in Zambia.\n\n\n4\n\nMention: \"2015 LCMS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"This is the most recent source of national, urban-rural and provincial poverty rates used as the baseline.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000178:5:1", "occurrence_id": "70a4f9a2523a9d88f0562c23", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  and insights from existing literature describing the risk landscape in Vietnam\n(Rentschler et. al 2020, chapter 3). In this section, the data employed to model each hazard are described.\n\n2.1. Flooding\n\n\nThis analysis uses one dataset modelling fluvial and pluvial flooding and a second one modelling coastal\nflooding. For sake of simplicity the combination of fluvial and pluvial flooding is addressed as riverine\nflooding in the main report. Fluvial flooding occurs when excessive rainfall over an extended period of\ntime causes a river to exceed its capacity. A pluvial, or surface water flood, is caused when heavy rainfall\ncreates a flood event independent of an overflowing water body. A coastal flood occurs in areas that lie\non the coast of a sea, ocean, or other large body of open water, and is typically the result of extreme tidal\nconditions caused by severe weather (World Meteorological Organization, 2011).\n\n\nThe first dataset combines information on flood extent and flood depth from pluvial and fluvial flooding\ninto one flood map. It was developed by the company Fathom, formerly known as SSBN. The map covers\nall of Vietnam at a resolution of about 90 by 90 meters. Flood events with return periods of 5, 20, 50, 100,\n250, and 500 years are modelled. A 100-year event, for example, is expected to occur once every 100\nyears, on average (figure 2.1.a). As with most large-scale flood maps, the effects of man-made flood\nprotection structures like dikes are not incorporated. The dataset is in raster format (Bangalore et al. 2017).\n\n\nExisting coastal flood maps covering Vietnam, most prominently the Global Tide and Surge Reanalysis\n(GTSR) data set (Muis et al. 2016), are expected to underestimate coastal flood risk in Vietnam. Resulting\nfrom coarse resolution in modelling elevation, bathymetry and meteorological forcing, the dataset\nunderestimates extreme sea levels. In\n\nMention: \"Global Tide and Surge Reanalysis\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Coarse elevation and forcing resolution mean this flood map is judged to understate coastal risk.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000182:14:1", "occurrence_id": "7475815987d5fecea5e3ae37", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 7 **|\n|Low female labor force participation|4|4|**8 **|4|**8 **|\n|Wide regional differences and lack of convergence|3|4|7|3|6|\n|**Dynamic Firms**||||||\n|Low performance in technology absorption and innovation|4|2|6|4|**8 **|\n|Weak corporate governance|4|2|6|2|6|\n|Weak competition policy and its enforcement|4|2|6|2|6|\n|Being stuck in small, mixed cropping agriculture|3|4|7|4|**7 **|\n|**Public Assets and Resources**||||||\n|Congested|4|3|7|2|6|\n|Declining availability of water|3|3|6|3|6|\n|Energy consumptions closely linked with GDP growth|2|3|5|2|5|\n|Inefficient land management|3|3|6|3|6|\n\n\nSource: Adapted from World Bank (2016). Note: Scores in columns (2) and (3) are obtained from World Bank (2016) based on\nstaff estimates and Turkey experts’ survey. The scores in columns (4) equal the sum of the scores in columns (2) and (3). The\nscores in column (5) are ours and the ones in column (6) equal the sum of the scores in columns (2) and (5).\n\n\nThe Turkish case is illustrated in Figure 1. As suggested by global data in the WEF on gender gaps,\n\nTurkey has made progress in closing the education gender gap (the female-male gaps in enrollment in\n\nsecondary education and tertiary education are 99% and 87%, <sup>17</sup\n\nMention: \"Turkey experts’ survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Constraint scores reproduced in the table come from staff estimates combined with this expert survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000182:14:1", "occurrence_id": "06ea54241e8e2ec8df1f537c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 7 **|\n|Low female labor force participation|4|4|**8 **|4|**8 **|\n|Wide regional differences and lack of convergence|3|4|7|3|6|\n|**Dynamic Firms**||||||\n|Low performance in technology absorption and innovation|4|2|6|4|**8 **|\n|Weak corporate governance|4|2|6|2|6|\n|Weak competition policy and its enforcement|4|2|6|2|6|\n|Being stuck in small, mixed cropping agriculture|3|4|7|4|**7 **|\n|**Public Assets and Resources**||||||\n|Congested|4|3|7|2|6|\n|Declining availability of water|3|3|6|3|6|\n|Energy consumptions closely linked with GDP growth|2|3|5|2|5|\n|Inefficient land management|3|3|6|3|6|\n\n\nSource: Adapted from World Bank (2016). Note: Scores in columns (2) and (3) are obtained from World Bank (2016) based on\nstaff estimates and Turkey experts’ survey. The scores in columns (4) equal the sum of the scores in columns (2) and (3). The\nscores in column (5) are ours and the ones in column (6) equal the sum of the scores in columns (2) and (5).\n\n\nThe Turkish case is illustrated in Figure 1. As suggested by global data in the WEF on gender gaps,\n\nTurkey has made progress in closing the education gender gap (the female-male gaps in enrollment in\n\nsecondary education and tertiary education are 99% and 87%, <sup>17</sup\n\nMention: \"WEF on gender gaps\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Enrollment gaps in secondary and tertiary education are quoted from this global gender-gap comparison.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000163:1:0", "occurrence_id": "d36db438c79e7b707dc0a092", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Policy Research Working Paper 9396\n\n##### **Abstract**\n\nThis paper uses Multiple Indicator Cluster Surveys data\n\nfrom the Republic of Congo and São Tomé and Príncipe\nto study the relationships between child stature, mother’s years of education, and indicators of early childhood\ndevelopment. The relationships are contrasted between two\nempirical approaches: the conventional approach whereby\ncontrol variables are selected in an ad-hoc manner and\nthe double machine-learning approach that employs datadriven methods to select controls from a much wider set\nof variables. Overall, the findings based on the preferable\ndouble machine-learning approach differ across the two\ncountries depending on the measures of early childhood\n\n\n\ndevelopment and child stature (height-for-age Z-score and\nstunting) used in the analysis. Double machine-learning\nestimates for the Republic of Congo suggest that heightfor-age Z-score and stunting have a direct causal effect on\nearly childhood development. In contrast, for São Tomé\nand Príncipe, no relationship is found. Thus, country-specific policy advice based on the relationships observed from\ndata in other countries may be quite risky, if not misleading. Double machine-learning provides a practical and\nfeasible approach to reducing threats to internal validity\nto derive robust inferences based on observational data for\nevidence-based policy advice.\n\n\n\nThis paper is a product of the Poverty and Equity Global Practice. It is part of a larger effort by the World Bank to\n\nprovide open access to its research and make a contribution to development policy discussions around the world. Policy\nResearch Working Papers are also posted on the Web at http://www.worldbank.org/prwp. The authors may be contacted\nat eskoufias@worldbank.org.\n\n\n_The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development_\n_issues. An objective of the series is to get the findings out quickly, even if\n\nMention: \"Multiple Indicator Cluster Surveys data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Child stature, maternal education and early childhood development indicators are analysed on these surveys.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000184:17:0", "occurrence_id": "a7f2641102098107f0c315bc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: border delays related to border compliance come from the “trading across borders” section in the\nWorld Bank’s Doing Business Database. <sup>15</sup> [^15: For any border, we use the data on “Border Compliance” and the total delay is assumed to be the sum of\nexport time from the exporting country and the import time from the importing country. We do not include\ndocumentary compliance, as it does not relate to travel time. All data are available at:\n<u>[http://www.doingbusiness.org/data/exploretopics/trading-across-borders.](http://www.doingbusiness.org/data/exploretopics/trading-across-borders)</u>] Border compliance data capture the time associated\nwith compliance with the customs clearance and mandatory inspections regulations. To account\nfor the very high share of maritime shipping in international trade due to price differentials, the\nalgorithm opts for maritime if the shipping time is lower than four times the shipping time incurred\nusing rail links. In the improved scenario of the network simulation, when a project involves\nbuilding a new port or upgrading an old port, the associated “processing time” is assumed to\ndecrease to 50 percent of the port delay in the region or to the lowest worldwide processing time,\nwhichever is higher.\n\n\nFinally, the population weighted time distance between country-pairs is transformed into\nad-valorem equivalents using estimates from Hummels and Schaur (2013) on the “daily value of\ntime” at the sector level. These estimates are added to transport costs and data on tariffs from\nGTAP to obtain country pair-sector values of trade costs. Table 2 presents the results for two\nscenarios, referred to as the “lower-bound” and the “upper-bound”. The “upper-bound” scenario\nallows for changes in transportation mode due to the new infrastructure while the “lower-bound”\nscenario assumes that switching mode of transportation is difficult -allowing for modal changes\nlower than 5 percent with respect to the pre-BRI modes of transport. The decrease\n\nMention: \"Doing Business Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Border crossing times are taken from the trading across borders section of this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000184:17:0", "occurrence_id": "4da3818c86286f035f0a36a2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: border delays related to border compliance come from the “trading across borders” section in the\nWorld Bank’s Doing Business Database. <sup>15</sup> [^15: For any border, we use the data on “Border Compliance” and the total delay is assumed to be the sum of\nexport time from the exporting country and the import time from the importing country. We do not include\ndocumentary compliance, as it does not relate to travel time. All data are available at:\n<u>[http://www.doingbusiness.org/data/exploretopics/trading-across-borders.](http://www.doingbusiness.org/data/exploretopics/trading-across-borders)</u>] Border compliance data capture the time associated\nwith compliance with the customs clearance and mandatory inspections regulations. To account\nfor the very high share of maritime shipping in international trade due to price differentials, the\nalgorithm opts for maritime if the shipping time is lower than four times the shipping time incurred\nusing rail links. In the improved scenario of the network simulation, when a project involves\nbuilding a new port or upgrading an old port, the associated “processing time” is assumed to\ndecrease to 50 percent of the port delay in the region or to the lowest worldwide processing time,\nwhichever is higher.\n\n\nFinally, the population weighted time distance between country-pairs is transformed into\nad-valorem equivalents using estimates from Hummels and Schaur (2013) on the “daily value of\ntime” at the sector level. These estimates are added to transport costs and data on tariffs from\nGTAP to obtain country pair-sector values of trade costs. Table 2 presents the results for two\nscenarios, referred to as the “lower-bound” and the “upper-bound”. The “upper-bound” scenario\nallows for changes in transportation mode due to the new infrastructure while the “lower-bound”\nscenario assumes that switching mode of transportation is difficult -allowing for modal changes\nlower than 5 percent with respect to the pre-BRI modes of transport. The decrease\n\nMention: \"Border compliance data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Customs clearance and inspection times are summed across the two countries to form total border delay.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000184:17:0", "occurrence_id": "3b495cf220ddbf2933e7ba29", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: border delays related to border compliance come from the “trading across borders” section in the\nWorld Bank’s Doing Business Database. <sup>15</sup> [^15: For any border, we use the data on “Border Compliance” and the total delay is assumed to be the sum of\nexport time from the exporting country and the import time from the importing country. We do not include\ndocumentary compliance, as it does not relate to travel time. All data are available at:\n<u>[http://www.doingbusiness.org/data/exploretopics/trading-across-borders.](http://www.doingbusiness.org/data/exploretopics/trading-across-borders)</u>] Border compliance data capture the time associated\nwith compliance with the customs clearance and mandatory inspections regulations. To account\nfor the very high share of maritime shipping in international trade due to price differentials, the\nalgorithm opts for maritime if the shipping time is lower than four times the shipping time incurred\nusing rail links. In the improved scenario of the network simulation, when a project involves\nbuilding a new port or upgrading an old port, the associated “processing time” is assumed to\ndecrease to 50 percent of the port delay in the region or to the lowest worldwide processing time,\nwhichever is higher.\n\n\nFinally, the population weighted time distance between country-pairs is transformed into\nad-valorem equivalents using estimates from Hummels and Schaur (2013) on the “daily value of\ntime” at the sector level. These estimates are added to transport costs and data on tariffs from\nGTAP to obtain country pair-sector values of trade costs. Table 2 presents the results for two\nscenarios, referred to as the “lower-bound” and the “upper-bound”. The “upper-bound” scenario\nallows for changes in transportation mode due to the new infrastructure while the “lower-bound”\nscenario assumes that switching mode of transportation is difficult -allowing for modal changes\nlower than 5 percent with respect to the pre-BRI modes of transport. The decrease\n\nMention: \"data on tariffs from\nGTAP\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Tariff rates are combined with time-cost estimates to produce country-pair sector trade costs.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000185:16:0", "occurrence_id": "0cbbc27f19386f44cba6c2d0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Sindh Government’s Annual School Census (ASC) (2010/11), we ranked districts by size as\n\n\nmeasured by the number of schools and villages in each district. In consultation with the\n\n\nGovernment of Sindh, we then chose one large, one medium-sized and one small district, ensuring\n\n\nrepresentativeness of the selected districts along two key education measures. <sup>8</sup> The selected\n\n\ndistricts were Matiari, Mirpurkhas and Sanghar.\n\n\nIn the absence of an updated population census, the ASC was used to set the population frame for\n\n\nselecting villages. <sup>9</sup> Using probability-proportional-to-size (PPS) sampling, we selected a sample\n\n\nof 550 schools drawn from 377 villages. The administrative data appeared to be unreliable\n\n\nregarding the exact location of schools and the inclusion of schools that are permanently closed.\n\n\nTo address this, we mapped 300 of the 377 villages in a village-census exercise. The maximum of\n\n\n300 villages was reached using a randomized ranking procedure. <sup>10</sup> The process through which\n\n\ntreatment schools were selected from the village mapping exercise is outlined in **<u>Figure 2</u>** .\n\n\nThe school sampling strategy for the baseline covered public primary schools that were open on\n\n\nthe day of the visit or closed for a period of less than one year prior to the day of the visit. We\n\n\nsampled all such schools, identified by the local community, in the main settlement of our sampled\n\n\n8 Matiari was ranked the third smallest district, Mirpurkhas was ranked 12th, and Sanghar was ranked 18th, according\nto the number of schools and villages in each district. In terms of education indicators, Mirpurkhas had one of the\nlowest levels of education outcomes, Matiari was in the middle, while Sanghar was among the highest. They were\ndeemed relatively safe by\n\nMention: \"Annual School Census\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Districts were ranked by school and village counts from this census to select the sample.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000185:16:0", "occurrence_id": "4fe9ee776d76bf86fa1884f3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Sindh Government’s Annual School Census (ASC) (2010/11), we ranked districts by size as\n\n\nmeasured by the number of schools and villages in each district. In consultation with the\n\n\nGovernment of Sindh, we then chose one large, one medium-sized and one small district, ensuring\n\n\nrepresentativeness of the selected districts along two key education measures. <sup>8</sup> The selected\n\n\ndistricts were Matiari, Mirpurkhas and Sanghar.\n\n\nIn the absence of an updated population census, the ASC was used to set the population frame for\n\n\nselecting villages. <sup>9</sup> Using probability-proportional-to-size (PPS) sampling, we selected a sample\n\n\nof 550 schools drawn from 377 villages. The administrative data appeared to be unreliable\n\n\nregarding the exact location of schools and the inclusion of schools that are permanently closed.\n\n\nTo address this, we mapped 300 of the 377 villages in a village-census exercise. The maximum of\n\n\n300 villages was reached using a randomized ranking procedure. <sup>10</sup> The process through which\n\n\ntreatment schools were selected from the village mapping exercise is outlined in **<u>Figure 2</u>** .\n\n\nThe school sampling strategy for the baseline covered public primary schools that were open on\n\n\nthe day of the visit or closed for a period of less than one year prior to the day of the visit. We\n\n\nsampled all such schools, identified by the local community, in the main settlement of our sampled\n\n\n8 Matiari was ranked the third smallest district, Mirpurkhas was ranked 12th, and Sanghar was ranked 18th, according\nto the number of schools and villages in each district. In terms of education indicators, Mirpurkhas had one of the\nlowest levels of education outcomes, Matiari was in the middle, while Sanghar was among the highest. They were\ndeemed relatively safe by\n\nMention: \"ASC\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"With no updated population census, this school census supplied the frame for village selection.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000186:14:1", "occurrence_id": "9ab824741d47c59eba3a82d6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  countries\nwith a more competitive environment. We thus run a robustness test where we add an interaction term\nof labor productivity and a COVID-19 restrictions index to the specification in table 3. Here, COVID-19\nrestrictions are measured by the number of days the Oxford COVID-19 Government Response Tracker’s\nStringency Index—which captures the strictness of “lockdown-style” policies that primarily restrict\npeople’s behavior along nine dimensions ranging from 0 to 100, with higher values indicating more\nrestrictions <sup>8</sup> —was above 70 preceding the fieldwork start day of the ES COVID-19 Follow-up Surveys\nmodule in each country. Table A.7 shows the results. The sample is smaller than in table 3 because the\nCOVID-19 restrictions index is not available for Armenia, Montenegro, and North Macedonia. Most of\nthe main results in table A.7 are similar to those in table 3, with the exception that the interaction of\nlabor productivity and the BTI market organization measure is not statistically significantly correlated\nwith the probability of increasing remote work. Thus, overall, the relationship between competition and\nreallocation from less to more productive firms during the crisis is not driven by differences in COVID-19\nrestrictions across countries.\n\n**5.3 Endogeneity of Government Assistance and Productivity**\n\nA potential source of endogeneity when estimating the relationship between government assistance\nand productivity in table 5 is that more productive firms may have been less likely to receive\ngovernment support than less productive firms since the former experienced smaller drops in sales and\nemployment. The ES data do not include detailed information on when firms received government\n\n\n8 Details on how the index is calculated are available at https://github.com/OxCGRT/covid-policytracker/blob/master/documentation/index_methodology.md.\n\n\n13\n\nMention: \"Oxford COVID-19 Government Response Tracker\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Days above a stringency threshold before fieldwork are counted to build the robustness interaction.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000186:14:1", "occurrence_id": "a2e9105f04f9bad6443c3722", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  countries\nwith a more competitive environment. We thus run a robustness test where we add an interaction term\nof labor productivity and a COVID-19 restrictions index to the specification in table 3. Here, COVID-19\nrestrictions are measured by the number of days the Oxford COVID-19 Government Response Tracker’s\nStringency Index—which captures the strictness of “lockdown-style” policies that primarily restrict\npeople’s behavior along nine dimensions ranging from 0 to 100, with higher values indicating more\nrestrictions <sup>8</sup> —was above 70 preceding the fieldwork start day of the ES COVID-19 Follow-up Surveys\nmodule in each country. Table A.7 shows the results. The sample is smaller than in table 3 because the\nCOVID-19 restrictions index is not available for Armenia, Montenegro, and North Macedonia. Most of\nthe main results in table A.7 are similar to those in table 3, with the exception that the interaction of\nlabor productivity and the BTI market organization measure is not statistically significantly correlated\nwith the probability of increasing remote work. Thus, overall, the relationship between competition and\nreallocation from less to more productive firms during the crisis is not driven by differences in COVID-19\nrestrictions across countries.\n\n**5.3 Endogeneity of Government Assistance and Productivity**\n\nA potential source of endogeneity when estimating the relationship between government assistance\nand productivity in table 5 is that more productive firms may have been less likely to receive\ngovernment support than less productive firms since the former experienced smaller drops in sales and\nemployment. The ES data do not include detailed information on when firms received government\n\n\n8 Details on how the index is calculated are available at https://github.com/OxCGRT/covid-policytracker/blob/master/documentation/index_methodology.md.\n\n\n13\n\nMention: \"ES COVID-19 Follow-up Surveys\nmodule\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Fieldwork dates from this module define when the restriction index is measured for each country.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000186:14:1", "occurrence_id": "5fbbbb9e8d7adf9892836a72", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  countries\nwith a more competitive environment. We thus run a robustness test where we add an interaction term\nof labor productivity and a COVID-19 restrictions index to the specification in table 3. Here, COVID-19\nrestrictions are measured by the number of days the Oxford COVID-19 Government Response Tracker’s\nStringency Index—which captures the strictness of “lockdown-style” policies that primarily restrict\npeople’s behavior along nine dimensions ranging from 0 to 100, with higher values indicating more\nrestrictions <sup>8</sup> —was above 70 preceding the fieldwork start day of the ES COVID-19 Follow-up Surveys\nmodule in each country. Table A.7 shows the results. The sample is smaller than in table 3 because the\nCOVID-19 restrictions index is not available for Armenia, Montenegro, and North Macedonia. Most of\nthe main results in table A.7 are similar to those in table 3, with the exception that the interaction of\nlabor productivity and the BTI market organization measure is not statistically significantly correlated\nwith the probability of increasing remote work. Thus, overall, the relationship between competition and\nreallocation from less to more productive firms during the crisis is not driven by differences in COVID-19\nrestrictions across countries.\n\n**5.3 Endogeneity of Government Assistance and Productivity**\n\nA potential source of endogeneity when estimating the relationship between government assistance\nand productivity in table 5 is that more productive firms may have been less likely to receive\ngovernment support than less productive firms since the former experienced smaller drops in sales and\nemployment. The ES data do not include detailed information on when firms received government\n\n\n8 Details on how the index is calculated are available at https://github.com/OxCGRT/covid-policytracker/blob/master/documentation/index_methodology.md.\n\n\n13\n\nMention: \"ES data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"The survey lacks dates for government support receipt, limiting causal interpretation of the productivity link.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000187:15:1", "occurrence_id": "79bb3b238bf76cb85269d317", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .8 24.8 42.1 60.4 68.8\n<u><mark>Relative incidence</mark></u> <u><mark>0.28</mark></u> <u><mark>0.09</mark></u> <u><mark>0.22</mark></u> <u><mark>0.25</mark></u> <u><mark>0.25</mark></u> <u><mark>0.28</mark></u> <u><mark>0.20</mark></u> <u><mark>0.25</mark></u> <u><mark>0.36</mark></u> <u><mark>0.41</mark></u> <u><mark>0.25</mark></u>\n_<u>Source:</u>_ <u>Elaboration based on data derived from the WELCOM Stata Tool and the 2015/16 HCES; see CSA 2018.</u>\n_Note:_ Consumption values are in birr of 2015.\n\nDespite the positive but unequalizing welfare effects of greater competition on new users, an\naccounting of only the welfare effects on new users shows that the relative welfare gain among\nhouseholds in the bottom decile of the consumption distribution is 6.9 percent, compared with an\naverage of 4.9 percent among households in the richest decile (figure 3).\n\n**Figure 3. Relative Welfare Gains of Greater Competition in Telecommunication, New Users**\n\n\n\n8.0%\n7.0%\n6.0%\n5.0%\n4.0%\n3\n\nMention: \"2015/16 HCES\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Welfare incidences are derived from this household expenditure and consumption survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000190:10:0", "occurrence_id": "997186480024625b622d5ce3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: coordinate efforts to improve security (which means coordinating with security operations) with\nprojects to improve service delivery. <sup>3</sup> [^3: The case of the Arab Republic of Egypt shows how difficult this is in practice. Fiscal reforms were\naccompanied by a World Bank project for social security. However, a government change from Morsi damaged\nthe project because the new government tried to establish control and, according to the national data, poverty\nhas increased. Ironically, sharing fiscal resources with the population is also a way of building fiscal control.]\n\n#### Data\n\nSummary statistics for our data sources are provided in Appendix Table A1.\n\n\nPoverty Data\nThroughout this report, we employ three poverty data sets to explore the relationship between\npoverty and conflict. The first data set is an unbalanced panel of the poverty headcount ratio at $1.90\na day (2011 PPP) at the country level (PovcalNet) from the World Bank. The second data set is the\nGlobal Subnational Atlas of Poverty (GSAP). It provides a cross-section of international poverty\nmeasures circa 2013 at the subnational level. The third data set we rely on is a short panel from\nUganda in 2012 and 2016. This panel consists of around 1,500 sub-counties in the country.\n\n\nConflict Data\nTwo conflict databases are used in the study. The Georeferenced Event Dataset (GED) Global version\n19.1 from the Uppsala Conflict Data Programme (UCDP), henceforth called UCDP, and the Armed\nConflict Location & Event Data Project (ACLED) are databases of geolocated conflict events around the\nworld with a number of fatalities. The two data sets are different in terms of coverage and data\ncollecting process. The UCDP has larger coverage with more countries so it delivers a broader\nillustration of the global situation between 1989-2018 when we implement the cross-country study.\nBut UCDP’s different inclusion criteria, such that a conflict dyad is only coded once it crosses the 25\nbattle-related deaths threshold in a given year, makes it understate fatalities\n\nMention: \"PovcalNet\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Country-level poverty headcounts at the international line form an unbalanced panel from this source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000190:10:0", "occurrence_id": "b39277148329b4fa686a1185", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: coordinate efforts to improve security (which means coordinating with security operations) with\nprojects to improve service delivery. <sup>3</sup> [^3: The case of the Arab Republic of Egypt shows how difficult this is in practice. Fiscal reforms were\naccompanied by a World Bank project for social security. However, a government change from Morsi damaged\nthe project because the new government tried to establish control and, according to the national data, poverty\nhas increased. Ironically, sharing fiscal resources with the population is also a way of building fiscal control.]\n\n#### Data\n\nSummary statistics for our data sources are provided in Appendix Table A1.\n\n\nPoverty Data\nThroughout this report, we employ three poverty data sets to explore the relationship between\npoverty and conflict. The first data set is an unbalanced panel of the poverty headcount ratio at $1.90\na day (2011 PPP) at the country level (PovcalNet) from the World Bank. The second data set is the\nGlobal Subnational Atlas of Poverty (GSAP). It provides a cross-section of international poverty\nmeasures circa 2013 at the subnational level. The third data set we rely on is a short panel from\nUganda in 2012 and 2016. This panel consists of around 1,500 sub-counties in the country.\n\n\nConflict Data\nTwo conflict databases are used in the study. The Georeferenced Event Dataset (GED) Global version\n19.1 from the Uppsala Conflict Data Programme (UCDP), henceforth called UCDP, and the Armed\nConflict Location & Event Data Project (ACLED) are databases of geolocated conflict events around the\nworld with a number of fatalities. The two data sets are different in terms of coverage and data\ncollecting process. The UCDP has larger coverage with more countries so it delivers a broader\nillustration of the global situation between 1989-2018 when we implement the cross-country study.\nBut UCDP’s different inclusion criteria, such that a conflict dyad is only coded once it crosses the 25\nbattle-related deaths threshold in a given year, makes it understate fatalities\n\nMention: \"Global Subnational Atlas of Poverty\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"This atlas supplies subnational poverty measures as a cross-section.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000190:10:0", "occurrence_id": "c91b73c6b112b799b3034ec7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: coordinate efforts to improve security (which means coordinating with security operations) with\nprojects to improve service delivery. <sup>3</sup> [^3: The case of the Arab Republic of Egypt shows how difficult this is in practice. Fiscal reforms were\naccompanied by a World Bank project for social security. However, a government change from Morsi damaged\nthe project because the new government tried to establish control and, according to the national data, poverty\nhas increased. Ironically, sharing fiscal resources with the population is also a way of building fiscal control.]\n\n#### Data\n\nSummary statistics for our data sources are provided in Appendix Table A1.\n\n\nPoverty Data\nThroughout this report, we employ three poverty data sets to explore the relationship between\npoverty and conflict. The first data set is an unbalanced panel of the poverty headcount ratio at $1.90\na day (2011 PPP) at the country level (PovcalNet) from the World Bank. The second data set is the\nGlobal Subnational Atlas of Poverty (GSAP). It provides a cross-section of international poverty\nmeasures circa 2013 at the subnational level. The third data set we rely on is a short panel from\nUganda in 2012 and 2016. This panel consists of around 1,500 sub-counties in the country.\n\n\nConflict Data\nTwo conflict databases are used in the study. The Georeferenced Event Dataset (GED) Global version\n19.1 from the Uppsala Conflict Data Programme (UCDP), henceforth called UCDP, and the Armed\nConflict Location & Event Data Project (ACLED) are databases of geolocated conflict events around the\nworld with a number of fatalities. The two data sets are different in terms of coverage and data\ncollecting process. The UCDP has larger coverage with more countries so it delivers a broader\nillustration of the global situation between 1989-2018 when we implement the cross-country study.\nBut UCDP’s different inclusion criteria, such that a conflict dyad is only coded once it crosses the 25\nbattle-related deaths threshold in a given year, makes it understate fatalities\n\nMention: \"Georeferenced Event Dataset\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Geolocated conflict events with fatalities are drawn from this dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000190:10:0", "occurrence_id": "ef95e70f99cb05ae5f78f4f2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: coordinate efforts to improve security (which means coordinating with security operations) with\nprojects to improve service delivery. <sup>3</sup> [^3: The case of the Arab Republic of Egypt shows how difficult this is in practice. Fiscal reforms were\naccompanied by a World Bank project for social security. However, a government change from Morsi damaged\nthe project because the new government tried to establish control and, according to the national data, poverty\nhas increased. Ironically, sharing fiscal resources with the population is also a way of building fiscal control.]\n\n#### Data\n\nSummary statistics for our data sources are provided in Appendix Table A1.\n\n\nPoverty Data\nThroughout this report, we employ three poverty data sets to explore the relationship between\npoverty and conflict. The first data set is an unbalanced panel of the poverty headcount ratio at $1.90\na day (2011 PPP) at the country level (PovcalNet) from the World Bank. The second data set is the\nGlobal Subnational Atlas of Poverty (GSAP). It provides a cross-section of international poverty\nmeasures circa 2013 at the subnational level. The third data set we rely on is a short panel from\nUganda in 2012 and 2016. This panel consists of around 1,500 sub-counties in the country.\n\n\nConflict Data\nTwo conflict databases are used in the study. The Georeferenced Event Dataset (GED) Global version\n19.1 from the Uppsala Conflict Data Programme (UCDP), henceforth called UCDP, and the Armed\nConflict Location & Event Data Project (ACLED) are databases of geolocated conflict events around the\nworld with a number of fatalities. The two data sets are different in terms of coverage and data\ncollecting process. The UCDP has larger coverage with more countries so it delivers a broader\nillustration of the global situation between 1989-2018 when we implement the cross-country study.\nBut UCDP’s different inclusion criteria, such that a conflict dyad is only coded once it crosses the 25\nbattle-related deaths threshold in a given year, makes it understate fatalities\n\nMention: \"Armed\nConflict Location & Event Data Project\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The second conflict database is compared with the first on coverage and collection process.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000191:14:1", "occurrence_id": "c7bdd6b8fa904bf67f9a32e9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  the sample.\n\n\n\n_Control variables_\n\n\n\n_Control variables_\n\n\ncomes from the NATO ISAF placemat for each Regional Military Command (NATO 2018). Placemats report\n\nInformation on the presence of international and Afghan combat troops – 𝑇𝑇 <sup>𝑊𝑊</sup> and 𝑇𝑇 <sup>𝐴𝐴</sup> respectively –\n\n\n\ncomes from the NATO ISAF placemat for each Regional Military Command (NATO 2018). Placemats report\n\nprovince-level information on the number of ISAF troops by the contributing nations and on the\n\nlocalization of Provincial Reconstruction Teams (PRTs). Placemat issues also include information on the\n\nnumber of Afghan National Army (ANA) troops and on ISAF and ANA major units. Although information\n\non PRTs is systematically reported, information on ISAF major units at the provincial level is only available\n\n\n\nfrom 2011 onwards. Combat troops are measured per 1,000 inhabitants in the province.\n\n\n\nfrom 2011 onwards. Combat troops are measured per 1,000 inhabitants in the province.\n\n\nAssistance Database (DAD), the main publicly available source of information on development assistance\n\nForeign aid 𝐷𝐷 is proxied by development assistance commitments. The data come from the Development\n\n\n\nAssistance Database (DAD), the main publicly available source of information on development assistance\n\nin Afghanistan. Data are available by donor and sector, at the provincial level. They can be disaggregated\n\n\n13\n\nMention: \"NATO ISAF placemat\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Province-level troop numbers by contributing nation and the location of reconstruction teams come from these reports.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000191:14:1", "occurrence_id": "cfc108441f8d7318f8dc3c98", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  the sample.\n\n\n\n_Control variables_\n\n\n\n_Control variables_\n\n\ncomes from the NATO ISAF placemat for each Regional Military Command (NATO 2018). Placemats report\n\nInformation on the presence of international and Afghan combat troops – 𝑇𝑇 <sup>𝑊𝑊</sup> and 𝑇𝑇 <sup>𝐴𝐴</sup> respectively –\n\n\n\ncomes from the NATO ISAF placemat for each Regional Military Command (NATO 2018). Placemats report\n\nprovince-level information on the number of ISAF troops by the contributing nations and on the\n\nlocalization of Provincial Reconstruction Teams (PRTs). Placemat issues also include information on the\n\nnumber of Afghan National Army (ANA) troops and on ISAF and ANA major units. Although information\n\non PRTs is systematically reported, information on ISAF major units at the provincial level is only available\n\n\n\nfrom 2011 onwards. Combat troops are measured per 1,000 inhabitants in the province.\n\n\n\nfrom 2011 onwards. Combat troops are measured per 1,000 inhabitants in the province.\n\n\nAssistance Database (DAD), the main publicly available source of information on development assistance\n\nForeign aid 𝐷𝐷 is proxied by development assistance commitments. The data come from the Development\n\n\n\nAssistance Database (DAD), the main publicly available source of information on development assistance\n\nin Afghanistan. Data are available by donor and sector, at the provincial level. They can be disaggregated\n\n\n13\n\nMention: \"Assistance Database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It is described as the main public source on development assistance commitments.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000191:14:1", "occurrence_id": "e4c74704d65fe603656f4f2a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  the sample.\n\n\n\n_Control variables_\n\n\n\n_Control variables_\n\n\ncomes from the NATO ISAF placemat for each Regional Military Command (NATO 2018). Placemats report\n\nInformation on the presence of international and Afghan combat troops – 𝑇𝑇 <sup>𝑊𝑊</sup> and 𝑇𝑇 <sup>𝐴𝐴</sup> respectively –\n\n\n\ncomes from the NATO ISAF placemat for each Regional Military Command (NATO 2018). Placemats report\n\nprovince-level information on the number of ISAF troops by the contributing nations and on the\n\nlocalization of Provincial Reconstruction Teams (PRTs). Placemat issues also include information on the\n\nnumber of Afghan National Army (ANA) troops and on ISAF and ANA major units. Although information\n\non PRTs is systematically reported, information on ISAF major units at the provincial level is only available\n\n\n\nfrom 2011 onwards. Combat troops are measured per 1,000 inhabitants in the province.\n\n\n\nfrom 2011 onwards. Combat troops are measured per 1,000 inhabitants in the province.\n\n\nAssistance Database (DAD), the main publicly available source of information on development assistance\n\nForeign aid 𝐷𝐷 is proxied by development assistance commitments. The data come from the Development\n\n\n\nAssistance Database (DAD), the main publicly available source of information on development assistance\n\nin Afghanistan. Data are available by donor and sector, at the provincial level. They can be disaggregated\n\n\n13\n\nMention: \"Development\n\n\n\nAssistance Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Donor and sector aid commitments at provincial level are used to proxy foreign aid.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000194:5:0", "occurrence_id": "66470a4a51dfa08382dd9a84", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: For the sample starting in January 2010 to January 2020, the average number of protests per month\nper country is 21. In 34 percent of country-month pairs (2,910 of 8,549), there is no occurrence of\nprotest. Panel A of Appendix Figure A1 presents the histogram of the entire distribution of\navailable occurrence of protests from ACLED (Appendix Table A2 for summary statistics of actual\nprotests). The distribution supports the use of the Zero-Inflated Poisson estimator, which we\ndiscuss in the next section.\n\nWe construct an alternative measure of protests capturing the intensity of media chatter around\nprotests. The rationale for the news-based measure of protests is first and foremost that it allows\nto significantly expand the scope of the data on protests over space and time. This measure\naccounts for the saliency of specific protests. For instance, a protest sparked by a popular figure is\nlikely reported and discussed more extensively in the news media. The news data are from Dow\nJones FACTIVA, which is a global repository of over one billion articles published across the\nworld and reported in 28 different languages. For each language, the following words are\ntranslated: (i) “street” and (ii) “protests” or “protest”. The search for each language is constructed\nsuch that the word “street” and at least one relating to “protest” appear. For example, in English,\nthe precise search query is: _[(“street”) AND (“protest” OR “protests”)]_ .\n\nThis search is performed for all countries and all languages in the Dow Jones database, over the\nperiod from January 1, 2000 to December 31, 2019. In addition, we gather the number of total\narticles in the Dow Jones FACTIVA for each country and each month. A measure of normalized\narticle counts is then constructed by taking the ratio between the number of articles mentioning\n“street protests” and the total\n\nMention: \"Dow\nJones FACTIVA\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"News articles across many languages are searched in this repository to count protest coverage.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000194:5:0", "occurrence_id": "b9b2718e66b7d44a4290b452", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: For the sample starting in January 2010 to January 2020, the average number of protests per month\nper country is 21. In 34 percent of country-month pairs (2,910 of 8,549), there is no occurrence of\nprotest. Panel A of Appendix Figure A1 presents the histogram of the entire distribution of\navailable occurrence of protests from ACLED (Appendix Table A2 for summary statistics of actual\nprotests). The distribution supports the use of the Zero-Inflated Poisson estimator, which we\ndiscuss in the next section.\n\nWe construct an alternative measure of protests capturing the intensity of media chatter around\nprotests. The rationale for the news-based measure of protests is first and foremost that it allows\nto significantly expand the scope of the data on protests over space and time. This measure\naccounts for the saliency of specific protests. For instance, a protest sparked by a popular figure is\nlikely reported and discussed more extensively in the news media. The news data are from Dow\nJones FACTIVA, which is a global repository of over one billion articles published across the\nworld and reported in 28 different languages. For each language, the following words are\ntranslated: (i) “street” and (ii) “protests” or “protest”. The search for each language is constructed\nsuch that the word “street” and at least one relating to “protest” appear. For example, in English,\nthe precise search query is: _[(“street”) AND (“protest” OR “protests”)]_ .\n\nThis search is performed for all countries and all languages in the Dow Jones database, over the\nperiod from January 1, 2000 to December 31, 2019. In addition, we gather the number of total\narticles in the Dow Jones FACTIVA for each country and each month. A measure of normalized\narticle counts is then constructed by taking the ratio between the number of articles mentioning\n“street protests” and the total\n\nMention: \"Dow Jones database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The translated search query is run across all country-language combinations over a twenty-year window.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000198:22:1", "occurrence_id": "1bca56584ca144e7a75d39ad", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  **x**\n*Lao **x** **x**\nMexico **x** **x**\nNicaragua **x** **x** **x**\nNigeria **x** **x** **x**\n\n - Paraguay **x** **x** **x**\n\n - Peru **x** **x** **x**\n\n - Poland **x** **x** **x**\n\n - Romania **x** **x** **x**\n\n - Somalia **x** **x** **x**\n\n - Sudan **x** **x** **x**\nSyria **x** **x** **x**\nUruguay **x** **x**\n\n - Venezuela **x** **x** **x**\nVietnam **x** **x** **x**\nW.Samoa **x** **x**\n\n- Yemen AR **x** **x**\nZambia **x** **x**\n\nTotal 34 **31** **32**\n\n\nSource: Extracted from IMF, Annual Report on Exchange Arrangements and Exchange Restrictions\n\n(Washington **DC:1989)**\n\n- Principal, secondary, tertiary rates, and their applicability to types of transactions are found in **IFS.**\n\nMention: \"IFS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Official interest rate structures and their applicability are looked up in this statistics yearbook.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000200:10:0", "occurrence_id": "5b6d2e5309e7f2bf38123f8d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: grants and guarantees extended by national agencies of up to 134 bilateral creditor countries and by 56\n\ndifferent international organizations (see Appendix III for a full list). Whenever available we collected\n\ndata on the creditor and the borrower, both commitment and disbursement amounts, the purpose of the\n\nloan as well as the interest rates and repayment terms. In other cases, our data is more aggregate. At the\n\nminimum, however, our database contains creditor and borrower country as well as commitment\n\namounts at the annual level allowing us to identify source and direction of official lending and thus the\n\nbilateral patterns of international official flows. We now sketch the main sources used in constructing\n\nthe database.\n\n\n_International treaty series:_ Our main source for official lending during the 19 <sup>th</sup> century and\n\nleading up to World War II are international treaty collections. In doing so, we make use of the fact that\n\nthe conduct of foreign policy became increasingly formalized in the late 18 <sup>th</sup> century, so that\n\ninternational loans between two states were now codified in inter-governmental treaties (Keene 2012).\n\nOur sources show that, over the course of the 150 years from 1790 to World War II, sovereigns\n\nnegotiated thousands of bilateral treaties, in which they regulated their political and commercial\n\ninteractions. To identify bilateral lending, we systematically search the available treaty collections and\n\nrelated databases for bilateral financial agreements that arrange the extension of loans and grants or that\n\nextend guarantees on privately issued international loans or bonds.\n\n_National budget accounts:_ For the main bilateral creditor countries, we supplement the\n\ninformation obtained from international treaties with national budget accounts and plans. In particular,\n\nwe comb through the national budgets of the United Kingdom starting in 1790, France from 1790,\n\nGermany from 1870 and the US starting with its first official, foreign loan in 1917. In addition to\n\nproviding an important consistency check, these sources allow us to capture within-empire (country)\n\nMention: \"international treaty collections\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"State-to-state loans codified in treaties are the main source for historical official lending.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000200:10:0", "occurrence_id": "82d2df256873628a491dcba8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: grants and guarantees extended by national agencies of up to 134 bilateral creditor countries and by 56\n\ndifferent international organizations (see Appendix III for a full list). Whenever available we collected\n\ndata on the creditor and the borrower, both commitment and disbursement amounts, the purpose of the\n\nloan as well as the interest rates and repayment terms. In other cases, our data is more aggregate. At the\n\nminimum, however, our database contains creditor and borrower country as well as commitment\n\namounts at the annual level allowing us to identify source and direction of official lending and thus the\n\nbilateral patterns of international official flows. We now sketch the main sources used in constructing\n\nthe database.\n\n\n_International treaty series:_ Our main source for official lending during the 19 <sup>th</sup> century and\n\nleading up to World War II are international treaty collections. In doing so, we make use of the fact that\n\nthe conduct of foreign policy became increasingly formalized in the late 18 <sup>th</sup> century, so that\n\ninternational loans between two states were now codified in inter-governmental treaties (Keene 2012).\n\nOur sources show that, over the course of the 150 years from 1790 to World War II, sovereigns\n\nnegotiated thousands of bilateral treaties, in which they regulated their political and commercial\n\ninteractions. To identify bilateral lending, we systematically search the available treaty collections and\n\nrelated databases for bilateral financial agreements that arrange the extension of loans and grants or that\n\nextend guarantees on privately issued international loans or bonds.\n\n_National budget accounts:_ For the main bilateral creditor countries, we supplement the\n\ninformation obtained from international treaties with national budget accounts and plans. In particular,\n\nwe comb through the national budgets of the United Kingdom starting in 1790, France from 1790,\n\nGermany from 1870 and the US starting with its first official, foreign loan in 1917. In addition to\n\nproviding an important consistency check, these sources allow us to capture within-empire (country)\n\nMention: \"national budget accounts\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Creditor governments' budgets and plans supplement treaties, starting in the late eighteenth century.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000202:27:0", "occurrence_id": "1af381c916b735744938ccd1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 12. Average markup differences by county relative to Bucharest**\n\n\n_Source:_ World Bank staff calculations based on SBS data set.\n_Note:_ Numbers reflect estimated coefficients (multiplied by 100) for county fixed effects—relative to Bucharest, the\nomitted category—in a fixed-effect regression model controlling for firm covariates plus year and NACE 4-digit sector\nfixed effects, as displayed in equation (11). All bars reflect statistically significant results. See table A1.5 in annex 1 for\nthe complete set of results.\n\n\n\n**A variance decomposition analysis suggests that ownership is the most relevant**\n**characteristic to explain total variance of markup performance across firms in Romania,**\n**followed by capital intensity and export status.** Drawing from estimations of equation (11),\nit is possible to evaluate the relative importance of each variable to explain the variance of the\nmarkup. Letting _d_ <sup>_Exp_</sup> represent any of the explanatory variables of equation (11), the\ncontribution to the variance of the markup of that variable can be defined as\n\n𝛿 **𝑉** 𝑉 𝑑 <sup>**𝐸**</sup> <sup>𝐸</sup> 𝛿 **𝐶** 𝐶 𝑑 <sup>**𝐸**</sup> <sup>𝐸</sup> 𝜇 𝛿 𝑑 <sup>**𝐸**</sup> <sup>𝐸</sup>\n\n**𝐶** **𝐸**\n\n\n\n𝐶 **𝐶** 𝐶𝐶𝑡𝑡𝑑𝑑𝐸 **𝐸** 𝐸𝐸 =\n\n\n\n<u>𝛿3𝛿</u> <sup><u>2</u></sup> <u>𝑉<\n\nMention: \"SBS data set\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"County markup differences are estimated from firm fixed effects on this business survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000204:6:0", "occurrence_id": "a5b74f6adc65b0f051bda61a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: (2006) find that larger countries may find it tougher to control corruption. In contrast, using\n\n\nTransparency International’s Corruption Perceptions Index for 1998, Root (1999) finds that in a\n\n\ncross-section of 60 countries, higher population is significantly associated with less corruption.\n\n\nRoot attributes the findings to economies of scale in governance. Similarly, Anoruo and Braha\n\n\n(2005) find that in Africa, population growth and corruption are negatively correlated. Knack and\n\n\nAzfar (2003) demonstrate that the corruption measures typically used in the literature suffer from\n\n\nsample selection bias and correcting for this bias causes the relationship between country size and\n\n\ncorruption to become much weaker or disappear. Similarly, when using total land area as a measure\n\n\nof country size, Goel and Nelson (2010) also find that country size has no significant impact on\n\n\ncorruption.\n\n\nThe broader literature on the determinants of corruption highlights several factors\n\n\n(discussed in detail in section 2). Treisman (2000) finds that states of government, colonial history,\n\n\ntrade openness, religious affiliation, and exposure to democracy have a significant impact on\n\n\ncorruption. At the same time, greater ethnic fractionalization or diversity in the country has also\n\n\nbeen found to contribute positively to corruption (Dincer 2008, Glaeser and Saks 2004, La Porta\n\n\net al. 1999). If members of one ethnicity are elected to a public position, they are more likely to\n\n\nmaintain the position even if they display corrupt behavior. Goel and Nelson (1998) and\n\n\nsubsequently, Kotera et al. (2012) report that larger government is associated with higher\n\n\ncorruption. They find that in larger government, there appears to be lesser individual accountability\n\n\nand greater layers of bureaucracy which eventually leads to more corruption. Fisman and Gatti\n\n\n(2002) also find that fiscal decentralization in government expenditure is strongly and significantly\n\n\nassociated with lower corruption. This was supported by a later study that found a negative\n\n\n5\n\nMention: \"Corruption Perceptions Index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"A prior study's cross-country result on population size and corruption is based on this index.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000206:30:0", "occurrence_id": "8e569ba6b40da29b81502ff4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Cohen, Wesley M. and Steven Klepper (1996), “A Reprise of Size and R&D,” _Economic Journal_\n106 (437): 925-951.\n\nColeman, Susan (2000), “Access to Capital and Terms of Credit: A Comparison of Men- and\nWomen-owned Small Businesses,” _Journal of Small Business Management_ 38(3): 37-52.\n\nd'Agostino, Giorgio, John P. Dunne, Luca Pieroni (2016), “Corruption and Growth in Africa,”\n_European Journal of Political Economy_ 43: 71-88.\n\nDe Rosa, Donato, Nishaal Gooroochurn and Holger Görg (2010), “Corruption and Productivity:\nFirm-level Evidence from the BEEPS Survey,” World Bank Policy Research Working Paper 5348,\nThe World Bank, Washington DC.\n\nDe Soto, Hernando (1989), _The Other Path_ **,** New York: Harper and Row.\n\nDiwan, Ishac, Philip Keefer and Marc Schiffbauer (2015), “Pyramid Capitalism: Political\nConnections, Regulation, and Firm Productivity in Egypt,” Policy Research Working Paper 7354,\nThe World Bank, Washington DC.\n\nDjankov, Simeon, Rafael La Porta, Florencio Lopez-de-Silanes and Andrei Shleifer (2002), “The\nRegulation of Entry,” _Quarterly Journal of Economics_ 117(1): 1-37.\n\n<mark>Djankov, Simeon, Rafael La Porta, Florencio Lopez-de-Silanes and Andrei Shleifer (2003a),</mark>\n<mark>“Courts,”</mark> _<mark>Quarterly Journal of Economics</mark>_ <mark>118 (2): 453-517.</mark>\n\nDjankov, Simeon, Edward Gla\n\nMention: \"BEEPS Survey\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"The survey appears in a reference title for firm-level evidence on corruption and productivity.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000211:27:9", "occurrence_id": "d97da9d6ef91c4ef88d8a6ec", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Data obtained via the\nCOVID-19 demography of deaths\ndatabase. National Institute for\nDemographic Studies (INED)\n(distributor)). Data include deaths\noccurred in hospitals or elsewhere,\n<u>by 10-year age groups and sex.</u>\n\n\n\n**Spain** 20,536 Yes May\n21,\n2020\n\n\n\nHealth\nInstitute\nCarlos III\n\n\n\n26\n\nMention: \"COVID-19 demography of deaths\ndatabase\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Hospital and non-hospital deaths by age group and sex are obtained from this distributed database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000212:19:0", "occurrence_id": "05efbee0c6a4b62a763f745e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The following is the set of 64 countries and economies with official reported deaths used in the\n\n\nanalysis: Afghanistan; Albania; Algeria; Argentina; Australia; Austria; Bangladesh; Belgium;\n\n\nBrazil; Canada; Chad; Chile; Colombia; Costa Rica; Czechia; Denmark; El Salvador; England\n\n\nand Wales; Eswatini; Finland; France; Germany; Greece; Haiti; Hong Kong SAR, China;\n\n\nHungary; Iceland; India; Indonesia; Israel; Italy; Jamaica; Japan; Jordan; Malawi; Mexico;\n\n\nMoldova; Nepal; Netherlands; New Zealand; Nigeria; Northern Ireland; Norway; Paraguay;\n\n\nPeru; Philippines; Portugal; Romania; Scotland; Slovak Republic; Slovenia; Somalia; South\n\n\nAfrica; Republic of Korea; Spain; Sweden; Switzerland; Taiwan, China; Togo; Turkey; Ukraine;\n\n\nUruguay; United States; Vietnam.\n\n\n**Excess all-cause deaths from vital statistics records**, from the following sources:\n\n\n[Short term mortality fluctuations (STMF) harmonized data series from the Human Mortality](https://www.mortality.org/)\n\n\n<u>Database</u>\n\n\n  - For this analysis; the STMF input files are used. These input files cover 38 countries and\n\n\neconomies: Australia; Austria; Belgium; Bulgaria; Canada; Switzerland; Chile; Czech\n\n\nRepublic; Germany; Denmark; Spain; Estonia; Finland; France; Northern Ireland;\n\n\nScotland; England and Wales; Greece; Croatia; Hungary; Iceland; Israel; Italy; Republic\n\n\nof Korea; Lithuania; Luxembourg; Latvia; Netherlands; Norway; New Zealand; Poland;\n\n\nPortugal; Russian Federation; Slovak Republic; Slovenia; Sweden; Taiwan, China;\n\n\nUnited States.\n\n\n  - Russia does not have data for 2020.\n\n\n  - Chile and Germany do not have data for 2015.\n\n\nColombia vital statistics tabulations ( _Defunciones no fetales_ [) were compiled by DANE. 2019 and](https:\n\nMention: \"vital statistics records\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Excess all-cause mortality is measured from civil registration records rather than reported case data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000212:19:0", "occurrence_id": "23381ec7710e41c26b712e1a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The following is the set of 64 countries and economies with official reported deaths used in the\n\n\nanalysis: Afghanistan; Albania; Algeria; Argentina; Australia; Austria; Bangladesh; Belgium;\n\n\nBrazil; Canada; Chad; Chile; Colombia; Costa Rica; Czechia; Denmark; El Salvador; England\n\n\nand Wales; Eswatini; Finland; France; Germany; Greece; Haiti; Hong Kong SAR, China;\n\n\nHungary; Iceland; India; Indonesia; Israel; Italy; Jamaica; Japan; Jordan; Malawi; Mexico;\n\n\nMoldova; Nepal; Netherlands; New Zealand; Nigeria; Northern Ireland; Norway; Paraguay;\n\n\nPeru; Philippines; Portugal; Romania; Scotland; Slovak Republic; Slovenia; Somalia; South\n\n\nAfrica; Republic of Korea; Spain; Sweden; Switzerland; Taiwan, China; Togo; Turkey; Ukraine;\n\n\nUruguay; United States; Vietnam.\n\n\n**Excess all-cause deaths from vital statistics records**, from the following sources:\n\n\n[Short term mortality fluctuations (STMF) harmonized data series from the Human Mortality](https://www.mortality.org/)\n\n\n<u>Database</u>\n\n\n  - For this analysis; the STMF input files are used. These input files cover 38 countries and\n\n\neconomies: Australia; Austria; Belgium; Bulgaria; Canada; Switzerland; Chile; Czech\n\n\nRepublic; Germany; Denmark; Spain; Estonia; Finland; France; Northern Ireland;\n\n\nScotland; England and Wales; Greece; Croatia; Hungary; Iceland; Israel; Italy; Republic\n\n\nof Korea; Lithuania; Luxembourg; Latvia; Netherlands; Norway; New Zealand; Poland;\n\n\nPortugal; Russian Federation; Slovak Republic; Slovenia; Sweden; Taiwan, China;\n\n\nUnited States.\n\n\n  - Russia does not have data for 2020.\n\n\n  - Chile and Germany do not have data for 2015.\n\n\nColombia vital statistics tabulations ( _Defunciones no fetales_ [) were compiled by DANE. 2019 and](https:\n\nMention: \"Short term mortality fluctuations\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Harmonized weekly death series from this collection provide the comparison period values.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000212:19:0", "occurrence_id": "3eb0e4643340e8debff2adf8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The following is the set of 64 countries and economies with official reported deaths used in the\n\n\nanalysis: Afghanistan; Albania; Algeria; Argentina; Australia; Austria; Bangladesh; Belgium;\n\n\nBrazil; Canada; Chad; Chile; Colombia; Costa Rica; Czechia; Denmark; El Salvador; England\n\n\nand Wales; Eswatini; Finland; France; Germany; Greece; Haiti; Hong Kong SAR, China;\n\n\nHungary; Iceland; India; Indonesia; Israel; Italy; Jamaica; Japan; Jordan; Malawi; Mexico;\n\n\nMoldova; Nepal; Netherlands; New Zealand; Nigeria; Northern Ireland; Norway; Paraguay;\n\n\nPeru; Philippines; Portugal; Romania; Scotland; Slovak Republic; Slovenia; Somalia; South\n\n\nAfrica; Republic of Korea; Spain; Sweden; Switzerland; Taiwan, China; Togo; Turkey; Ukraine;\n\n\nUruguay; United States; Vietnam.\n\n\n**Excess all-cause deaths from vital statistics records**, from the following sources:\n\n\n[Short term mortality fluctuations (STMF) harmonized data series from the Human Mortality](https://www.mortality.org/)\n\n\n<u>Database</u>\n\n\n  - For this analysis; the STMF input files are used. These input files cover 38 countries and\n\n\neconomies: Australia; Austria; Belgium; Bulgaria; Canada; Switzerland; Chile; Czech\n\n\nRepublic; Germany; Denmark; Spain; Estonia; Finland; France; Northern Ireland;\n\n\nScotland; England and Wales; Greece; Croatia; Hungary; Iceland; Israel; Italy; Republic\n\n\nof Korea; Lithuania; Luxembourg; Latvia; Netherlands; Norway; New Zealand; Poland;\n\n\nPortugal; Russian Federation; Slovak Republic; Slovenia; Sweden; Taiwan, China;\n\n\nUnited States.\n\n\n  - Russia does not have data for 2020.\n\n\n  - Chile and Germany do not have data for 2015.\n\n\nColombia vital statistics tabulations ( _Defunciones no fetales_ [) were compiled by DANE. 2019 and](https:\n\nMention: \"Human Mortality\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"This database is the distributor of the harmonized mortality series used above.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000212:19:0", "occurrence_id": "6fd8387bc1713f8cd9859f99", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The following is the set of 64 countries and economies with official reported deaths used in the\n\n\nanalysis: Afghanistan; Albania; Algeria; Argentina; Australia; Austria; Bangladesh; Belgium;\n\n\nBrazil; Canada; Chad; Chile; Colombia; Costa Rica; Czechia; Denmark; El Salvador; England\n\n\nand Wales; Eswatini; Finland; France; Germany; Greece; Haiti; Hong Kong SAR, China;\n\n\nHungary; Iceland; India; Indonesia; Israel; Italy; Jamaica; Japan; Jordan; Malawi; Mexico;\n\n\nMoldova; Nepal; Netherlands; New Zealand; Nigeria; Northern Ireland; Norway; Paraguay;\n\n\nPeru; Philippines; Portugal; Romania; Scotland; Slovak Republic; Slovenia; Somalia; South\n\n\nAfrica; Republic of Korea; Spain; Sweden; Switzerland; Taiwan, China; Togo; Turkey; Ukraine;\n\n\nUruguay; United States; Vietnam.\n\n\n**Excess all-cause deaths from vital statistics records**, from the following sources:\n\n\n[Short term mortality fluctuations (STMF) harmonized data series from the Human Mortality](https://www.mortality.org/)\n\n\n<u>Database</u>\n\n\n  - For this analysis; the STMF input files are used. These input files cover 38 countries and\n\n\neconomies: Australia; Austria; Belgium; Bulgaria; Canada; Switzerland; Chile; Czech\n\n\nRepublic; Germany; Denmark; Spain; Estonia; Finland; France; Northern Ireland;\n\n\nScotland; England and Wales; Greece; Croatia; Hungary; Iceland; Israel; Italy; Republic\n\n\nof Korea; Lithuania; Luxembourg; Latvia; Netherlands; Norway; New Zealand; Poland;\n\n\nPortugal; Russian Federation; Slovak Republic; Slovenia; Sweden; Taiwan, China;\n\n\nUnited States.\n\n\n  - Russia does not have data for 2020.\n\n\n  - Chile and Germany do not have data for 2015.\n\n\nColombia vital statistics tabulations ( _Defunciones no fetales_ [) were compiled by DANE. 2019 and](https:\n\nMention: \"Colombia vital statistics tabulations\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"National death tabulations compiled by the statistical office supply the country's mortality baseline.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000213:23:0", "occurrence_id": "d4d13693c26877400d278dea", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure A6: Percentage of firms experiencing decline in liquidity and cash-flow.\n\n\nSources: World Bank’s Enterprise Surveys COVID-19 Follow-up Surveys.\n\n\nFigure A7: Vulnerability to exit.\n\n\nSources: World Bank’s Enterprise Surveys COVID-19 Follow-up Surveys.\n\n\n22\n\nMention: \"Enterprise Surveys COVID-19 Follow-up Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Firm-level shares reporting liquidity decline and exit vulnerability are drawn from these follow-up rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000214:19:0", "occurrence_id": "7fc09ac684e3bfbd875cb8bc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: COVID-19. Before COVID-19, children living in urban areas and under the age of 15 are\n\n\nmore likely to attend school than children living in rural areas that have the same age.\n\n\nFrom the age of 15 onward, no difference in school attendance is observed between urban\n\n\nand rural children. After schools reopened, i.e., during COVID-19, no difference is ob\n\nserved between urban and rural children, although before the age of 15, urban residents\n\n\nare more still likely to attend school than rural residents.\n\n\nOverall, Figure 2 suggests that COVID-19 decreased school attendance among all age\n\n\ngroups, but even more so, among those whose education is no longer compulsory (age\n\ngroup 15-18). Figure 2 also suggests that COVID-19 did not discriminate between the\n\n\ngenders or between rural and urban children.\n\n\n**2.2.6** **Learning** **Activities** **and** **Contact** **with** **Teachers** **during** **the** **School-Closure** **Pe-**\n\n\n**riod**\n\n\nIn the analysis of the heterogeneous effects, we are interested in whether the impact of\n\n\ntemporary schools’ closure on attendance differs by the opportunity of being engaged in\n\n\nany learning activities or in contact with their teachers during the school closure period\n\n\nin Nigeria. We hypothesize that people who were engaged in any learning activities\n\n\nduring the lockdown are more likely to return to school than those who were not. Indeed,\n\n\nlosing contact with the education system over the whole period of schools’ disruption\n\n\nis expected to lower students’ motivation to schooling by making other options (like\n\n\nemployment and marriage) more attracting alternatives.\n\n\n**_(i) Learning Activities_**\n\n\nTo capture learning activities during the school closure period, we exploit the first\n\n\nfive rounds of the COVID-19 NLP\n\nMention: \"COVID-19 NLP\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Learning activities during school closures are observed across the first five rounds of this longitudinal panel.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000217:11:0", "occurrence_id": "d9f30203519e9787d6f79f72", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: supply of other food products such as meat, cheese, milk, and fish. Other affected countries such as Brazil,\nEcuador, Malaysia and Indonesia also represent a significant share of world supply of respectively citrus\nfruit (48.7), bananas (30.4), palm oil (31.8), and coconut oil (46.6) (Appendix Table 1).\n\n\n**Figure 4. World imports from top-50 most affected by Covid-19**\n\n\n\nMeat\n\nFresh Fruits - Vegetables\n\nVegetable and animal oils\n\nSugar\n\nCereals\n\nFish\n\nLegumes and pulses\n\nDairy and eggs\n\nNuts and derived products\n\nOil-bearing\n\nStimulant crops\n\n\n\n\n\n\n\nWorld imports from 50 worst affected COVID Rest of the world\n\n\n_Source_ **_:_** <u>[Espitia, Rocha, Ruta (2020). “Database on COVID-19 trade flows and policies”.](https://www.worldbank.org/en/data/interactive/2020/04/02/database-on-coronavirus-covid-19-trade-flows-and-policies)</u>\n\nThe negative impact of disruptions in food supply as a result of Covid-19 will be potentially higher for\nimport food dependent countries, <sup>11</sup> with high levels of concentration of imports coming from the top-50\nmost affected countries. Import food dependent developing countries would be particularly vulnerable to\nsupply shocks in key staples such as cereals, meat and fresh fruits, where more than 74 percent of imports\non average come from the top-3 exporters that are most affected by Covid-19 (Table 2). <sup>12</sup> [^12: See Appendix Table 2 for results at the product level.] Import food\ndependent developing countries such as Botswana, Mexico and Jamaica import more than 95 percent of\nrice, wheat and corn from the top-3 exporters most affected by Covid-19. Least developed countries such\n\nMention: \"Database on COVID-19 trade flows and policies\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"World imports by agri-food product group are taken from this pandemic trade and policy database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000220:3:1", "occurrence_id": "57dd18039ecb281bbe6a9314", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  to twelve months with a global average of 3.3 months.\n<u>(Gentilini et.al. 2020).</u>\n\nIn Zambia, the first two cases of COVID-19 were recorded on March 18, 2020 and the spread of\nconfirmed cases was slow in the months that followed. However, swift and deep disruptions to\ninternational trade and global supply chains resulted in an immediate negative economic impact\n<u>(WFP, 2020). Zambia’s economy is expected to have contracted by 3.5 percent in 2020, in contrast</u>\nto a projected growth of 3.6% before the pandemic (IMF, June 2020). This comes at a time when\nthe country also faces a serious macroeconomic crisis because of rising inflation, a high fiscal\ndeficit, a depreciating kwacha, and pressing external debt obligations.\n\nHigh and chronic poverty have been features of Zambian society over the last decade, with the\nnational poverty headcount rate holding steady at around 54 percent since 2010 (Zamstats LCMS\n<u>Report, 2010, 2015). With a Gini index of 57.1, Zambia is also among the most unequal countries</u>\nin the world, ranking alongside South Africa and Namibia in the region. <sup>3</sup> Developments since 2015\n\n\n2 The World Bank defines extreme poverty as the state of living on or less than $1.90 per day in 2011 PPP terms.\n3 World Bank World Development Indicators.\n\n2\n\nMention: \"World Bank World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"A footnote cites this indicator source for the poverty and development comparisons since 2015.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000229:42:3", "occurrence_id": "1fcfac02c32fa4cdc8d26b54", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  if month is March 2020 or later and zero<br>otherwise<br>|Authors’ calculations<br>|\n|Independent<br>Variables for<br>annual cross-<br>sectional<br>regressions|Economic<br>development|GDP per capita|Log value of GDP per capita (PPP, constant 2017<br>international USD)|World Bank World<br>Development Indicators<br>|\n|Independent<br>Variables for<br>annual cross-<br>sectional<br>regressions|<br>Financial<br>development|Domestic Credit|<br> Domestic bank credit to private sector (% of GDP)<br>|<br> World Development<br>Indicators|\n|Independent<br>Variables for<br>annual cross-<br>sectional<br>regressions|<br>Financial<br>development|Financial Account<br>Ownership|Account ownership at a financial institution or with<br>a mobile-money-service provider (% of population<br>ages 15+)|<br>World Bank World<br>Development Indicators<br>|\n|Independent<br>Variables for<br>annual cross-<br>sectional<br>regressions|Remittances|Remittances|<br> Personal remittances, received (% of GDP)<br>|World Bank World<br>Development Indicators|\n|Independent<br>Variables for<br>annual cross-<br>sectional<br>regressions|Economic<br>Freedoms|Monetary Freedom<br>Index|Monetary freedom combines a measure of inflation<br>with an assessment of various government activities<br>that distort prices. (100 – free, 0 - repressed)<br>|<br\n\nMention: \"Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The variable table lists this source for GDP per capita and domestic credit controls.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000234:21:1", "occurrence_id": "a1611635757dfbfc65d51458", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \nJ. Chen, Z. Cheng, K. Gong, and J. Li. Riding out the covid-19 storm: How government policies\n\naffect smes in china. _Available at SSRN 3660232_, 2020.\n\n\nR. Chetty, J. N. Friedman, N. Hendren, M. Stepner, and T. O. I. Team. How did covid-19 and\n\n\nstabilization policies affect spending and employment? a new real-time economic tracker based\n\n\non private sector data. Working Paper 27431, National Bureau of Economic Research, June 2020.\n\n\nX. Cirera, M. Cruz, E. Davies, A. Grover, L. Iacovone, J. Cordova, D. Medvedev, F. Maduko, G. Nayyar,\n\n\nS. Reyes, and J. Torres. Policies to Support Busineses through the Covid-19 Shock: a Firm-Level\n\n\nPerspective. Technical report, 2020.\n\n\nJ. Crespo Cuaresma, J. Hlouskova, and M. Obersteiner. Natural disasters as creative destruction?\n\nevidence from developing countries. _Economic Inquiry_, 46(2):214–226, 2008.\n\n\nM. Cruz, J. Hill, L. Iacovone, E. Lopez-Cordova, P. Calice, and G. Teima. Assessing the impact and\n\n\npolicy responses in support of private-sector firms in the context of the COVID-19 pandemic.\n\n\nTechnical report, 2020.\n\n\nW. Cui, J. Hicks, and M. Norton. How well-targeted are payroll tax cuts as a response to covid-19?\n\n\nevidence from china. Technical report, SSRN, September 2020.\n\n\nR. Dai, J. Hu, and X. Z. Zhang. The impact of coronavirus on china’s smes: Findings from the\n\n\nenterprise survey for innovation and entrepreneurship in china.\n\nMention: \"enterprise survey for innovation and entrepreneurship in china\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"Referenced as the survey behind another study's findings on coronavirus effects on small firms.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000238:5:1", "occurrence_id": "3960940afb3228e3fb9eb70c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  in order to assess which\nsource is responsible for the larger welfare loss. Next, when we seek to summarize the inter-country\ndistribution of welfare losses, we suggest a plausible range for α: between five and twenty years.\n\nThe rest of this short paper is organized as follows. The next section describes how we compute the\nnumbers of years of life lost (LY) and additional poverty years (PY) for each country, given the available\ndata, and presents the estimates for 145 countries. Building on those ingredients, Section 3 summarizes\nthe evidence on the relative importance of poverty and mortality in lowering welfare around the world.\nIt also investigates the global distribution of those losses under plausible values for the key normative\nparameter, both for a constant absolute poverty line and under a more relativist approach to poverty\nidentification. Section 4 concludes.\n\n\n**2.** **Estimating additional mortality and poverty with imperfect data**\n\nThe two key data sources for our mortality estimates are the UN Population Division database and the\nGlobal Burden of Disease Database (Dicker et al., 2018). Combining data from these two sources and using\nsmoothing procedures described in the Appendix to DFMS, we obtain estimates for the number of people\nof age _a_ living in country _j_, _Naj_, for 0 ≤𝑎𝑎≤99. <sup>8</sup> [^8: Those aged 100 or more are counted as being 99.] Using these country-specific population pyramids, we can\n19-induced mortality rates _maj_, for all countries in our sample, then our estimate of years of life lost due\nalso construct estimates for age-specific residual life expectancies, 𝜆𝜆𝑎 **𝑎** . If we observed age-specific COVIDto the pandemic in each country would be given by the sum across all ages of the number of deaths from\nthat age group ( 𝑁𝑁𝑎 **𝑎** 𝑚𝑚𝑎 **𝑎** ), times the average residual life-expectancy\n\nMention: \"UN Population Division database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Age-specific population counts by country are taken from this database to build mortality estimates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000238:5:1", "occurrence_id": "64f0961ec136560d9cd70c9b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  in order to assess which\nsource is responsible for the larger welfare loss. Next, when we seek to summarize the inter-country\ndistribution of welfare losses, we suggest a plausible range for α: between five and twenty years.\n\nThe rest of this short paper is organized as follows. The next section describes how we compute the\nnumbers of years of life lost (LY) and additional poverty years (PY) for each country, given the available\ndata, and presents the estimates for 145 countries. Building on those ingredients, Section 3 summarizes\nthe evidence on the relative importance of poverty and mortality in lowering welfare around the world.\nIt also investigates the global distribution of those losses under plausible values for the key normative\nparameter, both for a constant absolute poverty line and under a more relativist approach to poverty\nidentification. Section 4 concludes.\n\n\n**2.** **Estimating additional mortality and poverty with imperfect data**\n\nThe two key data sources for our mortality estimates are the UN Population Division database and the\nGlobal Burden of Disease Database (Dicker et al., 2018). Combining data from these two sources and using\nsmoothing procedures described in the Appendix to DFMS, we obtain estimates for the number of people\nof age _a_ living in country _j_, _Naj_, for 0 ≤𝑎𝑎≤99. <sup>8</sup> [^8: Those aged 100 or more are counted as being 99.] Using these country-specific population pyramids, we can\n19-induced mortality rates _maj_, for all countries in our sample, then our estimate of years of life lost due\nalso construct estimates for age-specific residual life expectancies, 𝜆𝜆𝑎 **𝑎** . If we observed age-specific COVIDto the pandemic in each country would be given by the sum across all ages of the number of deaths from\nthat age group ( 𝑁𝑁𝑎 **𝑎** 𝑚𝑚𝑎 **𝑎** ), times the average residual life-expectancy\n\nMention: \"Global Burden of Disease Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Cause-of-death information is combined with population data to estimate excess mortality.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000242:4:0", "occurrence_id": "01591d167819263508b39e2b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: build-up of external and fiscal buffers, and the implementation of sound debt management\nstrategies.\n\n\n**Figure 1:** Public Debt\n\n\n\n**(a)** IBRD and IDA: Public debt 2010-2020\n\n(percent GDP)\n\n\n\n**(b)** Composition of PPG debt by income\n\ngroup (in percent)\n\n\n\n**_Source:_** **_WEO, WB International Debt Statistics; and WB staff calculations_**\n\n\n**Figure 2:** External Public Debt Maturity and Average Interest Rates\n\n(In years and percentage points)\n\n\n\n**(a)** Average Maturity for External Official\n\nDebt (Years)\n\n\n\n**(b)** Average Interest Rates on External\n\nOfficial Debt (%)\n\n\n\n**_Source:_** **_WB International Debt Statistics._** **_Excluding high-income countries; and WB staff calculations_**\n\n\nThis narrative paints a broad picture of a fictional balloon for a group of countries. The\nexternal environment is also changing rapidly, especially with increased risk of tightening\nmonetary policy in developed economies, in general, and in the United States, in particular.\nSeveral countries have already defaulted, although risks of a systemic crisis seem limited\nnow. The operator of the balloon is also changing with some country leaders (see Dovis\net al. (2016) for a theory of debt on cycles of populism vs. austerity) pushing a populist\nnarrative that could spook market participants.\n\nWhile it is not surprising that countries default, it is surprising that lessons from the past\nare not internalized, thus building pressure for action, including better debt management\nand crisis planning. A relatively persistent number of defaults over time illustrates that hard\n\n\n2\n\nMention: \"WB International Debt Statistics\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Public debt levels and composition by income group are taken from this debt series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000247:23:0", "occurrence_id": "52f093ff342b45ff1faa663b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Importantly, at longer time windows, as the number of trades identified as vehicle trades\n\n\ndecreases, the share of trades that are cross-border trades increases. This observation is driven by two\n\n\nseparate dynamics:\n\n\n1. As the difference between Figure 4a and Figure 4b below, the average (median) time that passes\n\n\nbetween two trades that are identified as two legs of one vehicle trade differs significantly\n\n\nbetween domestic and international transfers: 44 (13) vs 83 (45) minutes.\n\n\n2. With an average (median) trade-size of 139 (45) USD of cross-border vehicle trades, compared\n\n\nto 115 (28) USD for domestic trades, cross-border trade sizes are significantly larger. <sup>28</sup>\n\n\n**_Figure 4 - Time between two trade legs of vehicle trades._**\n\n\n(a) Domestic Vehicle Trades (b) Cross-Border Vehicle Trades\n\n\nSource: LocalBitcoins.com API, Authors’ Calculations\n\n\n28 These statistics are based on the identified trades using a five-hour window. Whenever we present ratios or\nshares of vehicle trades, we moreover consider any trade whose individual hypothesis test leads to a rejection of\nthe match not being a vehicle trade with a 95% confidence level. Under the assumption that false-discoveries are\nhomogenously distributed across our sample, the false-discoveries and numerators and denominators cancel out,\nso that the false-discovery-control is not required.\n\n21\n\nMention: \"LocalBitcoins.com API\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Trade-level timestamps behind the vehicle trade timing figure come from this exchange interface.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000248:5:0", "occurrence_id": "f835f815e1254a5490e80d39", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Several recent papers make significant progress in addressing these issues. In terms of\n\n\nclassification, most papers had simply used an indicator variable measuring the presence of labor\n\n\nclauses, regardless of the content of the clauses. Using such an indicator, Kamata (2014) finds no\n\n\nstatistically significant effects of RTAs with or without labor clauses. Using the “Labor\n\n\nProvisions in Trade Agreements” (LABPTA) dataset created by Raess and Sari (2018), Carrere\n\n\net al. (2017) extend this work by differentiating labor clauses as those with and without\n\n\nenforcement and labor clauses designed to foster deep cooperation. Using a sample covering the\n\n\n1995-2004 period, restricted to countries with positive trade flows, and by differentiating trade\n\n\nflows from low-income to high-income countries, they find that enforcement has no effect on\n\n\ntrade flows, but clauses with “deep” cooperation are associated with more trade.\n\n\nExtending the analysis to include a legal-based classification of labor clauses, including\n\n\ndifferent forms of enforcement, LeClercq et al. (2020) find that labor provisions, defined in their\n\n\npaper as a combination of standards and enforcement clauses, are associated with lower trade\n\n\nflows but they are accompanied by clauses the promote trade, so that the overall effect of labor\n\n\nclauses on trade is positive.\n\n\nThis paper offers three extensions to the current literature. The main question asked by\n\n\nthis paper is whether the application of new techniques and incorporating other key\n\n\ncharacteristics of trade agreements affects the estimate of the relationship between labor clauses\n\n\nand trade flows. The first contribution is to apply the Pseudo-Poisson Maximum Likelihood\n\n\n(PPML) estimation with country-pair and importer and exporter time effects to the estimate the\n\n\nrelationship between trade flows and labor clauses. This approach is widely accepted in the\n\n\ngravity model literature, but has not been applied in most papers assessing the labor clause-trade\n\n\nflow relationship.\n\n\nThe second contribution is to use the depth of the Handbook of Deep Trade Agreements\n\n\n(HDTA)\n\nMention: \"Labor\n\n\nProvisions in Trade Agreements\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Earlier studies use this dataset to distinguish labour clauses by enforcement and cooperation type.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000249:10:0", "occurrence_id": "35a747c0b3994db112a28e9e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Ingo_ _Borchert_ _and_ _Mattia_ _Di_ _Ubaldo_\n\n#### **2 Related literature**\n\n\nServices are an increasingly important part of a country’s production and trade, as re\n\nflected by the growing attention to services in domestic regulation, as well as multilateral\n\n\nand bilateral policy instruments.\n\n\nFrom a general equilibrium perspective, entering a services PTA is likely to generate\n\n\nwelfare gains akin to those arising from entering PTAs liberalizing trade in goods, as\n\n\nthe PTA is expected to lower services input prices, final goods prices, and allow more\n\n\nconsumption than would otherwise be possible. Further, as barriers to services trade typ\n\nically consists of regulatory measures which act behind a country’s border, their removal\n\n\ndoes not come at the expense of a loss in tariff revenues, making services PTAs even more\n\n\ndesirable (Egger and Shingal, 2020).\n\n\nThe economic literature on how PTAs affect services trade is a great deal less abun\n\ndant than work on goods trade. Partly the reason is data availability: it is only with the\n\n\npublication of the OECD database on bilateral services trade in the early 2000s that a\n\n\nliterature attempting to estimate determinants of bilateral services trade rapidly emerged,\n\n\nfinding generally a positive and significant impact of PTAs (Ceglowski, 2006; Kimura and\n\n\nLee, 2006; Walsh, 2008; Marchetti, 2011; Shingal, 2010; Egger et al., 2012). <sup>3</sup>\n\n\nAnother reason why analyses on PTAs and services are scant, is that most of the\n\n\nPTAs explicitly featuring services chapters have been negotiated in the last two decades.\n\n\nFor this reason, early work investigated the impact of goods PTAs, rather than services\n\n\nPTAs. <sup>4</sup> More recently, work focusing explicitly on the impact of services PTAs confirms\n\n\n3Authors\n\nMention: \"OECD database on bilateral services trade\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its publication in the early 2000s is credited with enabling the empirical literature on bilateral services trade.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000250:11:1", "occurrence_id": "3c834df14d48c0e03bc8d609", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  _t_ is\n\n\n10 Despite the international reporting practices set by the IMF, countries tend to i) deviate from the\n10 percent ownership threshold, ii) not use uniform industrial classification, or iii) not report all types\nof investment properly (short-term intra-company loans, re-invested earnings are often missing).\n\n\n11 The data does not cover merge and acquisition or other equity and non-equity investments.\nTypically, projects are captured at the announcement stage for capital-intensive projects and at the\nopening stage of services operations with limited capital investment required. Capital-intensive\nprojects take on average 2 years to become operational.\n12 Recent academic research using fDi Markets data includes: (Crescenzi, Pietrobelli and Rabellotti\n2013), Paniagua and Sapena 2014), (Amoroso, Dosso and Moncada-Paternò-Castello 2015), and\n(Antonietti, Bronzini and Cainelli 2015).\n\n\n10\n\nMention: \"fDi Markets data\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"Listed only as the source used by several recent studies on project and operation stages.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000256:12:0", "occurrence_id": "0caca0a6d0deffff7a392e39", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Table 4: Composition of economic sectors in Kinshasa**\n\n\nAgriculture and fishing 5.9\n\nMining and quarrying 0.0\n\nManufacturing 7.8\n\nElectricity, gas, and water supply 0.4\n\nConstruction 6.3\n\nCommerce 41.7\n\nTransportation, storage, and communication 6.0\n\nFinancial, insurance, and real estate 1.4\n\nPublic administration 6.9\n\n<u>Other services</u> <u>23.8</u>\n_Source_ <u>: Kinshasa Survey 2018.</u>\n\n\nInformal enterprises are characterized by few years of operation, reliance on household capital,\nand self-employment (that is, no employee other than the owner). According to the Kinshasa\nSurvey, 50 percent of informal enterprises have been in operation for fewer than 3 years (Figure\n5), indicating their brief existence. The average years of operation is 6.1 years. These informal\nenterprises mostly rely on household savings (62 percent) or income from family property (15\npercent). The use of bank credit is negligible—97 percent of informal entrepreneurs did not apply\nfor loans. In addition, most of them are microenterprises, comprising only the owner (79 percent).\nOnly 4 percent of informal enterprises hire 4 or more employees (including the owner him/herself).\n\n\n**Figure 5: Years of operation of informal enterprises in urban Kinshasa**\n\n\n_Note:_ ‘p90’ indicates 90 percentile. For example, 90 percent of informal enterprises\noperated for 16 years or less.\n\n\nPart-time jobs are concentrated in self-employment jobs. One may expect part-time jobs (hours\nworked less than 25 hours per week) to be a common characteristic of self-employment jobs, yet\nthis is not necessarily the case. In 2018, the share of part-time jobs in self-employment (51 percent)\n\n\n11\n\nMention: \"Kinshasa Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Sector shares of the city's economy and the characteristics of informal enterprises come from this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000258:41:1", "occurrence_id": "3ed9ad3ba8bad539b8b1f026", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  (i.e. outflows)**\n1. Withdrawals from the Fund occur automatically when actual oil revenues fall short from\nbudgeted oil revenues by 10 percent or more.\n2. Oil revenue shortfalls up to 10 percent of budgeted oil revenues will be accommodated\nthrough expenditure adjustment.\n3. Any shortfall beyond 10 percent of budgeted oil revenues will be compensated subject to\navailability of resources in the Fund.\n4. The Stabilization Fund may only be used to finance expenditure budgeted in a given fiscal\nyear. The Fund may not be utilized for the satisfaction of any sovereign or commercial debt\nof the government, and no legal or beneficial interest in the Fund may be created.\n**iii.** **Formula to budget oil revenues**\n\nBudgeted oil revenues are estimated using conservative assumptions: Future oil prices are\nestimated with a US$3/barrel discount on the forecast published in the World Economic Outlook\nby the IMF. The forecasted volume of production is set 10 % below the projected production\nvolumes estimated by petroleum companies operating in Chad.\n\n**Illustrated behavior of the mechanism**\n\nThe baseline scenario assumes a constant oil price of oil of US$60. Figures B1 and B2 below illustrate how\nthe oil revenue management mechanism for stabilization works under three alternative scenarios:\n\n  - Scenario 1 corresponds to a fall in the price of oil of US$10 for one year.\n\n  - Scenario 2 corresponds to a US$20 fall for one year.\n\n  - Scenario 3 entails a US$10 increase in the price of oil.\n\n\n40\n\nMention: \"World Economic Outlook\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its price forecast, discounted by three dollars a barrel, sets budgeted oil revenues.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000259:44:0", "occurrence_id": "4ebc95a643cee7ec20775d84", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ###### 1.2 Data for Cross-Country Cross-Sector Analysis\n\nFor country-sector speci…cations, we estimate the impact of country-sector determinants in\nyear t-3 on country-sector GVC participation in year t, e¤ectively covering the period 19992015. Summary statistics for the EORA GVC participation shares and all the determinants\nat the country-sector level are provided in Table A5 and correlations among all determinants\nat country-sector level are provided in Table A6.\n\nWe interact each country-year determinant with its corresponding sectoral intensity of\nuse across manufacturing sub-sectors to obtain measures that vary at the country-sectoryear level. For labor endowments we use the country-level share of high-skilled workers\nfollowing the ILO’s de…nition of low skill (skill levels 3 and 4) based on the International\nStandard Classi…cation of Occupations. We interact the country skilled labor endowment\nwith a benchmark measure of skill intensity in the US de…ned as the ratio of skilled workers\nto unskilled workers by sector taken from the NBER-CES Manufacturing Industry Database.\nFor capital endowments, we use the ratio of real capital stock to total employment at the\ncountry level from Penn World Tables which we interact with a benchmark measure of capital\nintensity in the US by sector taken from the NBER-CES Manufacturing Industry Database.\n\nTo capture natural resources endowments, we interact the natural resources to GDP ratio\nfrom WDI with an indicator for natural resource intensive sectors based on Braun (2003).\nFor country institutional quality we use the rule of law index from the World Governance\nIndicators database which is interacted with the measure of contract intensity proposed by\nNunn (2007). The measure of contract intensity is determined by the share of specialized\nand customized intermediate inputs used in the production of the …nal good for each sector\nbased on a U.S. input-output table. <sup\n\nMention: \"NBER-CES Manufacturing Industry Database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"US sectoral skill and capital intensity benchmarks are taken from this industry database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000259:44:0", "occurrence_id": "4e9043dd5c9c94beeb0c94c2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ###### 1.2 Data for Cross-Country Cross-Sector Analysis\n\nFor country-sector speci…cations, we estimate the impact of country-sector determinants in\nyear t-3 on country-sector GVC participation in year t, e¤ectively covering the period 19992015. Summary statistics for the EORA GVC participation shares and all the determinants\nat the country-sector level are provided in Table A5 and correlations among all determinants\nat country-sector level are provided in Table A6.\n\nWe interact each country-year determinant with its corresponding sectoral intensity of\nuse across manufacturing sub-sectors to obtain measures that vary at the country-sectoryear level. For labor endowments we use the country-level share of high-skilled workers\nfollowing the ILO’s de…nition of low skill (skill levels 3 and 4) based on the International\nStandard Classi…cation of Occupations. We interact the country skilled labor endowment\nwith a benchmark measure of skill intensity in the US de…ned as the ratio of skilled workers\nto unskilled workers by sector taken from the NBER-CES Manufacturing Industry Database.\nFor capital endowments, we use the ratio of real capital stock to total employment at the\ncountry level from Penn World Tables which we interact with a benchmark measure of capital\nintensity in the US by sector taken from the NBER-CES Manufacturing Industry Database.\n\nTo capture natural resources endowments, we interact the natural resources to GDP ratio\nfrom WDI with an indicator for natural resource intensive sectors based on Braun (2003).\nFor country institutional quality we use the rule of law index from the World Governance\nIndicators database which is interacted with the measure of contract intensity proposed by\nNunn (2007). The measure of contract intensity is determined by the share of specialized\nand customized intermediate inputs used in the production of the …nal good for each sector\nbased on a U.S. input-output table. <sup\n\nMention: \"Penn World Tables\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Country capital stock relative to employment is taken from this table as the capital endowment.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000259:44:0", "occurrence_id": "7daaeaf7f1b5e004247d92f1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ###### 1.2 Data for Cross-Country Cross-Sector Analysis\n\nFor country-sector speci…cations, we estimate the impact of country-sector determinants in\nyear t-3 on country-sector GVC participation in year t, e¤ectively covering the period 19992015. Summary statistics for the EORA GVC participation shares and all the determinants\nat the country-sector level are provided in Table A5 and correlations among all determinants\nat country-sector level are provided in Table A6.\n\nWe interact each country-year determinant with its corresponding sectoral intensity of\nuse across manufacturing sub-sectors to obtain measures that vary at the country-sectoryear level. For labor endowments we use the country-level share of high-skilled workers\nfollowing the ILO’s de…nition of low skill (skill levels 3 and 4) based on the International\nStandard Classi…cation of Occupations. We interact the country skilled labor endowment\nwith a benchmark measure of skill intensity in the US de…ned as the ratio of skilled workers\nto unskilled workers by sector taken from the NBER-CES Manufacturing Industry Database.\nFor capital endowments, we use the ratio of real capital stock to total employment at the\ncountry level from Penn World Tables which we interact with a benchmark measure of capital\nintensity in the US by sector taken from the NBER-CES Manufacturing Industry Database.\n\nTo capture natural resources endowments, we interact the natural resources to GDP ratio\nfrom WDI with an indicator for natural resource intensive sectors based on Braun (2003).\nFor country institutional quality we use the rule of law index from the World Governance\nIndicators database which is interacted with the measure of contract intensity proposed by\nNunn (2007). The measure of contract intensity is determined by the share of specialized\nand customized intermediate inputs used in the production of the …nal good for each sector\nbased on a U.S. input-output table. <sup\n\nMention: \"WDI\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Natural resources as a share of GDP is drawn from this indicator series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000259:44:0", "occurrence_id": "ffec23734c335bd247ba9794", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ###### 1.2 Data for Cross-Country Cross-Sector Analysis\n\nFor country-sector speci…cations, we estimate the impact of country-sector determinants in\nyear t-3 on country-sector GVC participation in year t, e¤ectively covering the period 19992015. Summary statistics for the EORA GVC participation shares and all the determinants\nat the country-sector level are provided in Table A5 and correlations among all determinants\nat country-sector level are provided in Table A6.\n\nWe interact each country-year determinant with its corresponding sectoral intensity of\nuse across manufacturing sub-sectors to obtain measures that vary at the country-sectoryear level. For labor endowments we use the country-level share of high-skilled workers\nfollowing the ILO’s de…nition of low skill (skill levels 3 and 4) based on the International\nStandard Classi…cation of Occupations. We interact the country skilled labor endowment\nwith a benchmark measure of skill intensity in the US de…ned as the ratio of skilled workers\nto unskilled workers by sector taken from the NBER-CES Manufacturing Industry Database.\nFor capital endowments, we use the ratio of real capital stock to total employment at the\ncountry level from Penn World Tables which we interact with a benchmark measure of capital\nintensity in the US by sector taken from the NBER-CES Manufacturing Industry Database.\n\nTo capture natural resources endowments, we interact the natural resources to GDP ratio\nfrom WDI with an indicator for natural resource intensive sectors based on Braun (2003).\nFor country institutional quality we use the rule of law index from the World Governance\nIndicators database which is interacted with the measure of contract intensity proposed by\nNunn (2007). The measure of contract intensity is determined by the share of specialized\nand customized intermediate inputs used in the production of the …nal good for each sector\nbased on a U.S. input-output table. <sup\n\nMention: \"World Governance\nIndicators database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The rule of law index used for institutional quality comes from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000260:31:0", "occurrence_id": "88e4b783451e21e3cb18a3b0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 140\n\n\n120\n\n\n100\n\n\n80\n\n\n60\n\n\n40\n\n\n20\n\n\n0\n\n\nUSUK NON-USUK\n\n\n**Figure 1: Global private equity investments, 1997-2017.** This figure plots the time series\nof total global private equity investments over the last two decades. The U.S. and U.K.\n(USUK, darker shading) are plotted separately from all other countries (NON-USUK,\nlighter shading). Data, provided by Burgiss, are summed across deal-level equity\ninvestments classified by location of the corporate headquarters.\n\n\n30\n\nMention: \"Data, provided by Burgiss\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Deal-level equity investments are summed by headquarters location to plot global private equity activity.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000261:19:0", "occurrence_id": "8f76f8c126e58311e692c29c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: that the SSA region lags behind the global average in terms of human capital endowment. The\n\n\nsample average of net migration as a share of total population is roughly 2.6% while net\n\n\nremittances as a share of GDP is less than 1%, on average, for SSA countries. For the skill\n\n\nvariables, _Table 1_ shows that on average skilled labor content in exports is lower than the unskilled\n\n\nlabor content. This suggests that although skilled labor is more valued in exports than unskilled\n\n\nlabor, the participation of SSA countries in manufacturing GVCs is concentrated in low-skilled\n\n\nmanufacturing products.\n\n\n**Figure 2: Trend in GVC participation (at sector level) for SSA countries**\n\n\nSource: _Authors’ using UNTAD-MERO data_\n\n\nTo compare manufacturing GVC with other sectors, _Figure 2_ presents GVC participation rates for\n\n\nthe primary, service and manufacturing sectors. It shows that SSA economies are more integrated\n\n\ninto backward manufacturing GVC compared to the agriculture and service sectors. However,\n\n\nbackward participation has been declining for all these three sectors since 2011. On the other hand,\n\n\nforward participation in manufacturing GVC is the lowest compared to the primary and service\n\n\nsectors. This suggests that SSA countries supply less intermediate manufacturing inputs to the\n\n\nexports of other countries than they supply service and primary inputs in the exports of other\n\n\ncountries. Moreover, forward participation in manufacturing GVC exhibits a declining trend since\n\n\n2000 while forward participation in agriculture and service GVCs started declining in 2008 during\n\n\n18\n\nMention: \"UNTAD-MERO data_\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Sector-level GVC participation rates for sub-Saharan economies are computed from this regional input-output source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000263:3:0", "occurrence_id": "72a8c3b7546d67861f711d38", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **I.** **Introduction**\n\n\nUnderstanding the determinants of social distancing is central to addressing both the medical and economic\n\naspects of COVID-19. <sup>4</sup> On the one hand, reducing interactions among people is critical to reducing the\n\npropagation and a variety of non-pharmaceutical interventions (NPIs), such as closure of nonessential\n\nbusinesses, stay-at-home orders, or school closings, have been put in place to this end, with some success. <sup>5</sup>\n\nWhile there is controversy around whether this should be the goal in developing countries as well (Barnett\nHowell and Mobarak 2020, Loayza 2020), there is also concern about whether such measures would work:\n\ngovernment capabilities to enforce may be weaker, and resistance may be higher since the trade-off with\n\nlivelihood is harsher. At the other extreme of the cycle - where the debate is when to loosen NPIs as it is in\n\nseveral advanced countries – preliminary evidence from Wuhan suggests that when opened, mobility and\n\neconomic activity may not respond quickly. <sup>6</sup> Similarly, recent polls suggesting that 58% of Americans are\n\nconcerned that restrictions will be lifted too soon raise the question of how much of an impact opening will\n\nhave in practice and hence the shape of the recovery, whether V or U. <sup>7</sup>\n\n\nThis paper uses Google mobility data to explore which factors are proving important during the 2020 COVID\n19 outbreak in the United States and globally.  In all but the poorest countries, it confirms that NPIs can be\n\neffective, but that voluntary de-mobilization on the part of the population is much more important,-driven by\n\nfear or perhaps a sense of social responsibility. This suggests that much social distancing will happen regardless\n\nof the presence of restrictions and that NPIs may often function more like\n\nMention: \"Google mobility data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Mobility movement trends are used to test what drives social distancing across countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000264:6:0", "occurrence_id": "5d716a4aa829a3feef609d4b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Figure 0.2 From \"Twin Surpluses\" To \"Twin Deficits\" for Country Groups in MENA Region_\n\n**Figure 2.2 From \"Twin Surpluses\" To \"Twin Deficits\" for Country Groups in MENA Region**\n\n\n\n**1. Fiscal** **balance**\n_(Percentage of GDP)_\n\n15\n\n\n10\n\n\n5\n\n\n0\n\n\n-5\n\n\n-10\n\n\n-15\n\n\n\n\n\n\n\n2011 2012 2013 2014 2015 2016 2017 2018 2019\n\n\n\n**2. Current account**\n_(Percentage of GDP)_\n\n\n30\n\n\n25\n\n\n20\n\n\n15\n\n\n10\n\n\n5\n\n\n0\n\n\n-5\n\n\n-10\n\n\n-15\n\n\n\n2011 2012 2013 2014 2015 2016 2017 2018 2019\n\n\n\nSources: International Monetary Fund, World Economic Outlook Database October 2017; and World Bank Staff calculations.\nNote: Current account balance of country groups are average current account balance across member countries\nw eighted by gross domestic products. GCC: Gulf Cooperation Council countries. OEC: Oil exporting countries other than\nGCC countries. OIC: oil importing countries. Forecast Starting 2017.\n\n\nEven if many MENA countries, especially oil exporters, appear to have relatively low levels of debt,\nthere are important contingent liabilities related to state-owned enterprises—such as non-performing\nloans—that suggest an escalation of public debt. With little ability to increase spending to embark on\nthe kind of transformational agenda that is required to address the socio-economic challenges they\nface, MENA countries will have to mobilize private sector financing.\n\n\n**Infrastructure needs and challenges**\n\n\nThe infrastructure needs in the region are large. They are estimated to be around US$100 billion\nannually. Most of the needs are in electricity generation and transportation, followed by water and\nsanitation, and information and communication technologies.\n\n\nInfrastructure traditionally has been financed, and managed, by governments with little private sector\ninvolvement. Because budgets are tighter, there will be underinvestment in maintenance of publicly\nowned and operated infrastructures. Interestingly, this is what led to the privatization of infrastructure\nin the United Kingdom\n\nMention: \"World Economic Outlook Database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"GDP-weighted current account balances for country groups are taken from this vintage of the fund's database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000274:16:2", "occurrence_id": "ed7421568a2052f91ebe6d1f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: െ0.377 <br>𝜂ൌെ0.328 <br>𝜂ൌെ0.279 <br>𝑝ଵൌ 1.306 <br>𝑝ଵൌ 1.275 <br>𝑝ଵൌ 1.229|𝐹𝑖𝑟𝑚𝑠ൌ ∞ <br>𝜂∗ൌ െ0.1808 <br>𝑝଴ൌ 1.0000|\n\n\n_Source:_ World Bank estimates using ENIGH 2014; Hausman and Ros 2013.\n\n_Note:_ Changes in elasticity occur due to the simulation of a large price drop in the item of interest, resulting in a quantity\n\ndemanded of larger magnitude and decreasing the absolute value of the elasticity.\n\n\nTable 3 shows illustrates the WELCOM prediction that market concentration increases prices in the\nmobile communication market in Mexico by about 34.4 percent with respect to a scenario of perfect\n\n\n24 In this section the term telecommunications and communications will be used to refer to mobile communications\nservices.\n25 The rule that WELCOM follows is to multiply the original number of firms by 1.5 in the first step and then, for the step _n_, multiply\nthe initial number of firms by _n_, when _n_ <u>> 2.</u>\n\n\n15\n\nMention: \"ENIGH 2014\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Estimated demand parameters are derived from this household expenditure survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000278:13:0", "occurrence_id": "0ea59598b2a38d16f9dff89d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 9:** Size of Revenue and Effectiveness of Taxes at Consumable income\n\n_Source:_ Authors’ calculations using CEQ Master Workbooks (MWBs). Marginal contributions of taxes to poverty reduction are\nmissing for direct and indirect taxes in Armenia and indirect taxes for Georgia.\n\n_Notes_ : Countries are ranked in the ascending order of the share of taxes as percent of the GDP.\n\nThe distributional impact of taxes is defined as the marginal change in Gini index/poverty headcount for consumable income when\na tax is removed, expressed as a percentage of consumable income Gini or poverty headcount. A negative value indicates that a tax\nis inequality/poverty increasing, and otherwise.\n\n\n4.3 Government Spending\n\nFigure 10 shows the share of direct public transfers (including contributory pensions and other cash or nearcash benefits) and indirect transfers (public education and health care programs and indirect subsidies) in\nthe GDP. As a result of low revenue in Georgia, Tajikistan, Armenia and Kyrgyzstan, spending on transfers\nare also low compared to the rest of FSU countries (ranging from 11 to 16% of the GDP). Russia’s spending\non transfers (19.5% of the GDP) is lower than it could have been given the total government expenditure.\nThe share of government spending allocated to public transfers in Moldova, Ukraine and Belarus is\nequivalent to the average of advanced economies at around 25 percent of GDP in the early 2010s (IMF\n2014).\n\nSpending on contributory pensions and in-kind transfers dominates spending on transfers in all the countries\nincluded in the study. Only two countries out of eight, Armenia and Ukraine, allocate a significant share of\ntheir transfer spending (39 and 37%, respectively) to cash benefits. In case of Tajikistan, spending on cash\nbenefits accounts for 1% of the total spending on social transfers.\n\n\n12\n\nMention: \"CEQ Master Workbooks\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Tax revenue and its marginal contribution to poverty reduction are calculated from these harmonized microdata workbooks.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000281:13:0", "occurrence_id": "6fb3633c7c3d1f9d90037e9d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: extend the initial stocks forward <sup>and backward</sup> <sup>over time. 12</sup> <sup>The</sup> <sup>required information</sup> <sup>is drawn</sup>\n\nfrom a number of standard sources: <sup>initial</sup> <sup>stocks of domestic</sup> <sup>capital are taken from the Penn</sup>\n\n\n\nWorld Tables, and combined with <sup>_flow_</sup> <sup>data</sup> <sup>on gross</sup> <sup>domestic investment</sup> <sup>to build up capital</sup>\n\n\n\nstock series. For foreign holdings of <sup>domestic</sup> <sup>equity and domestic holdings of foreign equity, we</sup>\n\n\n\nrely primarily on data <sup>on stocks</sup> <sup>and flows of direct and portfolio equity investment</sup> <sup>reported</sup> <sup>in</sup>\n\n\n\nthe IMF's Balance of Payments Statistics Yearbook. <sup>Finally, stocks of borrowing and</sup> <sup>lending</sup>\n\n\n\nare obtained **by** combining <sup>stock data on the debt of developing</sup> <sup>countries</sup> <sup>reported</sup> <sup>in the World</sup>\n\n\n\nBank's Global Development <sup>Finance</sup> <sup>with data on</sup> <sup>debt stocks</sup\n\nMention: \"Balance of Payments Statistics Yearbook\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Stocks and flows of direct and portfolio equity investment are reported in this yearbook.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000282:12:1", "occurrence_id": "a584059273f47946ca12b4cd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: let. The satellite data were obtained from the African Flood and Drought Monitor\n\n(AFDM) <sup>17</sup> and include daily rainfall information for the period 1970-2015 at a 0.25 degree\n\nresolution. From this information, we constructed the monthly historical average level of rainfall\n\nfor each of the 512 village-hamlet units for the period that precedes the intervention (1970\n\n15 The analysis of the impact of cash transfers on resilience was not planned when the RCT was designed. However,\nthis question became central both as a research and policy issue during implementation of the Niger safety net\nproject, in particular when the Sahel Adaptive Social Protection Program was set up and provided additional\nfinancing to the project with a core objective to improve household resilience.\n16 The follow-up sample was stratified based on the proxy means test score “predicted” from the baseline survey.\nWe dropped from the sample 377 households with high predicted PMT score (hence low probability of being\nselected as beneficiary) (see Premand and Barry, 2020).\n17 <u>[https://platform.princetonclimate.com/PCA_Platform/.](https://platform.princetonclimate.com/PCA_Platform/)</u>\n\n\n11\n\nMention: \"African Flood and Drought Monitor\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Daily satellite rainfall at coarse resolution is aggregated into village-level monthly historical averages.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000282:12:1", "occurrence_id": "d9a4a10cb7a39860ecc9cd96", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: let. The satellite data were obtained from the African Flood and Drought Monitor\n\n(AFDM) <sup>17</sup> and include daily rainfall information for the period 1970-2015 at a 0.25 degree\n\nresolution. From this information, we constructed the monthly historical average level of rainfall\n\nfor each of the 512 village-hamlet units for the period that precedes the intervention (1970\n\n15 The analysis of the impact of cash transfers on resilience was not planned when the RCT was designed. However,\nthis question became central both as a research and policy issue during implementation of the Niger safety net\nproject, in particular when the Sahel Adaptive Social Protection Program was set up and provided additional\nfinancing to the project with a core objective to improve household resilience.\n16 The follow-up sample was stratified based on the proxy means test score “predicted” from the baseline survey.\nWe dropped from the sample 377 households with high predicted PMT score (hence low probability of being\nselected as beneficiary) (see Premand and Barry, 2020).\n17 <u>[https://platform.princetonclimate.com/PCA_Platform/.](https://platform.princetonclimate.com/PCA_Platform/)</u>\n\n\n11\n\nMention: \"baseline survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Predicted proxy means test scores from it stratified the follow-up sample and justified exclusions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000283:13:0", "occurrence_id": "1bcec6e622c0836405b13165", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: temperature played a significant role for the spread of COVID-19 in 122 cities in China. Holtmann\net al. (2020) also find that weather conditions affect the initial stages of the spreading of\nCoronavirus in 47 countries, with higher temperatures leading to slower spreading. While\nambient temperatures may affect the COVID-19 survival in the environment, it is also possible\nthat a population confined to indoor activities is more susceptible to the infection. We thus control\nfor the minimum temperature ( _Temp_ ) recorded each day in each country and expect that higher\n\n\n**Given that most of the CCMs entail the curtailing of individual liberties and the**\n\ntemperatures may induce governments to delay the adoption of CCMs (𝛽𝛽8 < 0).\n\n**pandemic has a global nature, political regimes and external pressures or**\n**demonstration effects from outside may also influence the adoption of CCMs** . More\ndemocratic regimes may find difficulties in restricting individual liberties affecting mobility and\nsocial contact, while autocratic regimes tend to be more able to apply tighter restrictions to\ncontain the pandemic. We control for a measure of democracy levels obtained from The\n\ndelay the adoption of CCMs, authorities have faced international pressure and learned from the\nexperience of other countries in dealing with the pandemic. In particular, the adoption of Economist (2020) database and expect 𝛽𝛽9 < 0 _._ In addition to internal pressures to implement or\nstringency measures in Western European countries, which were among the first to suffer from\nmassive contagion, influenced the speed at which other countries implemented CCMs. We use a\nmeasure of the stringency of CCMs in 26 European economies as a demonstration effect for the\nadoption of CCMs in the world and expect 𝛽𝛽10 - 0. <sup>4</sup>\n\n**4.** Econometric Results\n\n\n_a._ _Adoption of CCMs_\n\n\n**The econometric results of our estimations for\n\nMention: \"Economist (2020) database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It supplies the measure of democracy levels used as an explanatory variable.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000285:43:0", "occurrence_id": "05b483344439e80071468c94", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 7: Nominal Annual Wage Growth, by Cell Exposure to the Increase in the Minimum**\n**Wage**\n\n\n40%\n\n\n35%\n\n\n30%\n\n\n25%\n\n\n20%\n\n\n15%\n\n\n10%\n\n\n5%\n\n\n0%\n\n\nLow exposure Hıgh exposure\n\n\n_Source:_ Turkish Enterprise Information System.\n_Note:_ Exposure to the minimum wage increase is measured in the fourth quarter of 2015 as the proportional rise in\nthe wage bill required to bring all workers in the cell up to the new 2016 minimum wage. Cells are considered high\nexposure if the cell-level measure of exposure to the minimum wage is above the sample median, but low exposure\notherwise.\n\n\n42\n\nMention: \"Turkish Enterprise Information System\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Wage growth by minimum-wage exposure cell is measured from this employer administrative system.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000287:33:2", "occurrence_id": "ff8b93b0d63dcf298381207e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: br>Kose et al. (2017)|\n|<br>Debt/Reserves (short term)<br>|<br>Short term external debt stocks, % of reserves (External and private sector debt)<br>|<br>Short term external debt stocks, % of reserves (External and private sector debt)<br>|\n|<br>Government size<br>|<br>General government final consumption expenditure, % of GDP.<br>|World Development<br>Indicators (WDI)|\n|<br>External debt<br>|<br>Total external debt stocks to gross national income (GNI). Sum of public, publicly guaranteed, and private nonguaranteed long-<br>term debt, use of IMF credit, and short-term debt.<br>|<br>Total external debt stocks to gross national income (GNI). Sum of public, publicly guaranteed, and private nonguaranteed long-<br>term debt, use of IMF credit, and short-term debt.<br>|\n|Education<br>|<br>Education refers to the number of grades (years) in secondary school.<br>|<br>Education refers to the number of grades (years) in secondary school.<br>|\n|<br>Mineral rents<br>|<br>The difference between the value of production for a stock of minerals at world prices and their total costs of production.<br>Minerals included in the calculation are tin, gold, lead, zinc, iron, copper, nickel, silver, bauxite, and phosphate.<br>|<br>The difference between the value of production for a stock of minerals at world prices and their total costs of production.<br>Minerals included in the calculation are tin, gold, lead, zinc\n\nMention: \"Indicators (WDI)\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Government size and external debt variables in the variable table are sourced from this indicator series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000288:15:0", "occurrence_id": "72c7559758ae630e44ba5318", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 2: Correlation between the relative intensity of communication versus manual tasks**\n\n**and the share of teleworkable jobs in each of ISCO three-digit occupations in 2018**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nSource: Own elaboration based on EU-LFS data and O*NET surveys.\nNote: The size of the circles represents the relative size of each ISCO three-digit occupation in employment. The\ncorrelation uses the relative size of each occupation as weights.\n\n###### 3.3 Migrant-Native Gaps in Income and Health Risk Exposure to COVID-19\n\n\nBefore illustrating the econometric analysis, we present several descriptive statistics of our sample\n\nof working age individuals (ages 25-64) from the EU Labour Force Survey. As of 2018, there were\n\n270 million working-age people residing in Western European countries. Almost one out of five\n\n(18.7%) of them were foreign-born (Table 1). These immigrants originally come from various\n\nhigh-income Western European countries (3.1%), new EU member states such as Poland and\n\nRomania (3.6%), other European countries that are not EU members (3.5%), and other locations\n\noutside of Europe (8.1%). Migrant workers are slightly younger than native workers on average\n\n(mean age is 43 versus 45.4) and have a slightly lower share of men than the native workforce\n\n(48.4% versus 50.2%). Migrants also have slightly lower education levels, with 31.2% having\n\n\n14\n\nMention: \"EU-LFS data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Occupational employment shares are taken from this labour force survey to weight the correlation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000288:15:0", "occurrence_id": "e07b37cb3aaf6f5054a0fed7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 2: Correlation between the relative intensity of communication versus manual tasks**\n\n**and the share of teleworkable jobs in each of ISCO three-digit occupations in 2018**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nSource: Own elaboration based on EU-LFS data and O*NET surveys.\nNote: The size of the circles represents the relative size of each ISCO three-digit occupation in employment. The\ncorrelation uses the relative size of each occupation as weights.\n\n###### 3.3 Migrant-Native Gaps in Income and Health Risk Exposure to COVID-19\n\n\nBefore illustrating the econometric analysis, we present several descriptive statistics of our sample\n\nof working age individuals (ages 25-64) from the EU Labour Force Survey. As of 2018, there were\n\n270 million working-age people residing in Western European countries. Almost one out of five\n\n(18.7%) of them were foreign-born (Table 1). These immigrants originally come from various\n\nhigh-income Western European countries (3.1%), new EU member states such as Poland and\n\nRomania (3.6%), other European countries that are not EU members (3.5%), and other locations\n\noutside of Europe (8.1%). Migrant workers are slightly younger than native workers on average\n\n(mean age is 43 versus 45.4) and have a slightly lower share of men than the native workforce\n\n(48.4% versus 50.2%). Migrants also have slightly lower education levels, with 31.2% having\n\n\n14\n\nMention: \"O*NET surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Task intensity measures are matched to ISCO occupations from this source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000288:15:0", "occurrence_id": "59e5d3c18e7f3450179b1ea7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 2: Correlation between the relative intensity of communication versus manual tasks**\n\n**and the share of teleworkable jobs in each of ISCO three-digit occupations in 2018**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nSource: Own elaboration based on EU-LFS data and O*NET surveys.\nNote: The size of the circles represents the relative size of each ISCO three-digit occupation in employment. The\ncorrelation uses the relative size of each occupation as weights.\n\n###### 3.3 Migrant-Native Gaps in Income and Health Risk Exposure to COVID-19\n\n\nBefore illustrating the econometric analysis, we present several descriptive statistics of our sample\n\nof working age individuals (ages 25-64) from the EU Labour Force Survey. As of 2018, there were\n\n270 million working-age people residing in Western European countries. Almost one out of five\n\n(18.7%) of them were foreign-born (Table 1). These immigrants originally come from various\n\nhigh-income Western European countries (3.1%), new EU member states such as Poland and\n\nRomania (3.6%), other European countries that are not EU members (3.5%), and other locations\n\noutside of Europe (8.1%). Migrant workers are slightly younger than native workers on average\n\n(mean age is 43 versus 45.4) and have a slightly lower share of men than the native workforce\n\n(48.4% versus 50.2%). Migrants also have slightly lower education levels, with 31.2% having\n\n\n14\n\nMention: \"EU Labour Force Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Descriptive statistics on working-age residents and the foreign-born share are computed on this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000294:38:0", "occurrence_id": "40768138623f9c59e3539a35", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure C4: Ethnicities as a proportion of total cases (by role in the case) and the total population in Kenya\n\n\nProportions of the total population are derived from the 2019 census. Specific ethnicities are intentionally masked.\n\n\n37\n\nMention: \"2019 census\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Population shares by ethnicity are taken from this enumeration to compare against case shares.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000299:7:0", "occurrence_id": "914f6b0b8ed191bf27cb8fb4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: intensive margin (percentage decline in sales). Figure 1 illustrates the point graphically using\n\n\nfirm’s own financial condition while figure 2 does the same using a proxy measure of the firm’s\n\n\nfinancial condition (cell average) mentioned above and used in the regressions below. According\n\n\nto our final baseline specification, the probability of sales decline during the pandemic for a firm\n\n\nwith the best access to finance is lower by 25.2 percentage points compared to a firm with the\n\n\npoorest access to finance; and the corresponding percentage decline in sales (intensive margin) is\n\n\nsmaller by 14.3 percentage points. These differences are statistically significant (at the 5 percent\n\n\nlevel or less) and also robust to the tests mentioned above.\n\n\n**2. Data and Main Variables**\n\n\n2.1 _Data description_\n\n\nThe analyses in this paper are based mainly on two sources of data: the World Bank’s Enterprise\n\n\nSurveys (ES) and the World Bank’s Enterprise Surveys Covid-19 follow-up (COV-ES). The ES\n\n\nare firm-level surveys representative of a country’s private sector. A standard sampling procedure\n\n\nbased on location, sector, and size stratifications ensures representativeness both at the country\n\n\nlevel and at the sectoral level, while the use of a common questionnaire guarantees the\n\n\ncomparability of the estimates across countries. <sup>1</sup> [^1: For the full note on the ES methodology, refer to: https://www.enterprisesurveys.org/en/methodology.] Informal firms, businesses operating in the\n\n\nagriculture and extractive sector, and establishments with fewer than five employees are excluded\n\n\nfrom the sampling universe. The ES included in this paper were conducted between 2016 and 2020\n\n\nand all of them cover the pre-pandemic period. After the pandemic was declared, the same firms\n\n\ninterviewed in the ES were re-contacted with the aim of measuring the effects of the pandemic\n\n\n(COV-ES). These interviews were administered over the phone. Out of the total of 18,677 firms\n\n\n[1 For the full\n\nMention: \"Enterprise\n\n\nSurveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Firm-level data representative of national private sectors form the main analytical sample.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000299:7:0", "occurrence_id": "3fae92e0908d7812dbf7f34a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: intensive margin (percentage decline in sales). Figure 1 illustrates the point graphically using\n\n\nfirm’s own financial condition while figure 2 does the same using a proxy measure of the firm’s\n\n\nfinancial condition (cell average) mentioned above and used in the regressions below. According\n\n\nto our final baseline specification, the probability of sales decline during the pandemic for a firm\n\n\nwith the best access to finance is lower by 25.2 percentage points compared to a firm with the\n\n\npoorest access to finance; and the corresponding percentage decline in sales (intensive margin) is\n\n\nsmaller by 14.3 percentage points. These differences are statistically significant (at the 5 percent\n\n\nlevel or less) and also robust to the tests mentioned above.\n\n\n**2. Data and Main Variables**\n\n\n2.1 _Data description_\n\n\nThe analyses in this paper are based mainly on two sources of data: the World Bank’s Enterprise\n\n\nSurveys (ES) and the World Bank’s Enterprise Surveys Covid-19 follow-up (COV-ES). The ES\n\n\nare firm-level surveys representative of a country’s private sector. A standard sampling procedure\n\n\nbased on location, sector, and size stratifications ensures representativeness both at the country\n\n\nlevel and at the sectoral level, while the use of a common questionnaire guarantees the\n\n\ncomparability of the estimates across countries. <sup>1</sup> [^1: For the full note on the ES methodology, refer to: https://www.enterprisesurveys.org/en/methodology.] Informal firms, businesses operating in the\n\n\nagriculture and extractive sector, and establishments with fewer than five employees are excluded\n\n\nfrom the sampling universe. The ES included in this paper were conducted between 2016 and 2020\n\n\nand all of them cover the pre-pandemic period. After the pandemic was declared, the same firms\n\n\ninterviewed in the ES were re-contacted with the aim of measuring the effects of the pandemic\n\n\n(COV-ES). These interviews were administered over the phone. Out of the total of 18,677 firms\n\n\n[1 For the full\n\nMention: \"Enterprise Surveys Covid-19 follow-up\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The pandemic follow-up round supplies the shock outcomes analysed alongside the base survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000300:10:1", "occurrence_id": "2469264c83110296ec9c2a65", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  type of customs fraud. This is\n\nconsistent with formal fraud records which classify 67.2 percent of all fraud as underreporting of\n\nvalue, 27.4 percent as underreporting of quantities, and the remainder as misclassification (4.9\n\npercent) and misreporting country of origin (0.5 percent). <sup>15</sup> However, only 38 percent of inspectors\n\nsaid they knew which firms were likely to commit customs fraud, suggesting that asymmetric\n\n\n10 In 2017, 15,000 candidates applied for a 300-position recruitment program launched by Madagascar customs. The\nbest-ranked graduates from the national school of administration usually select customs as their preferred\nadministration post.\n11 Madagascar ranked 158th out of 180 countries in Transparency International’s 2019 Corruption Perception Index.\n12 According to a very senior civil servant, informal payments to some inspectors are north of 200,000 USD each year.\n13 Cantens et al. (2010) show similar environments with a culture of impunity across customs agencies in African\ncountries.\n14 Due to a close collaboration with the World Bank, the Director General of Customs made the response to our survey\nbe mandatory for all inspectors.\n15 The importance of undervaluation as the main form of tariff evasion in contrast to product and especially country\nof origin misclassification is also consistent with the very simple import tariff structure imposed by Madagascar with\nonly four ad-valorem tariff rates (0, 5, 10, and 20) and no specific tariffs except on petroleum and gas (WTO, 2015).\n\n\n9\n\nMention: \"2019 Corruption Perception Index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The country's rank among 180 countries is quoted to characterise the governance environment.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000301:12:0", "occurrence_id": "78dd950259c75fed101c57fa", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: datasets available to measure CBI. For robustness, we also use the data collected by Garriga (2016),\n\n\nwhich yields similar results.\n\n\nTo capture inequality dynamics, we rely on data from the World Bank’s PovcalNet database.\n\n\nBesides providing an extensive coverage of countries over a comparably long timespan, a key ad\n\nvantage of this dataset is that includes information on income dynamics at the decile level, which\n\n\nallows us to study income dynamics in a more fine-grained manner. Our headline results rely on\n\n\nthe share of total income earned by each decile, ordered from the lowest- (bottom 10%) to the\n\n\nhighest-earners (top 10%). Our conjecture is that the latter should benefit from CBI, whereas the\n\n\nformer should be hurt. Together, this would indicate an increase in inequality.\n\n\nIn addition, we test the validity of the channels through which CBI operates. We discuss these\n\n\nvariables in Section 5. Suffice here to say that we examine a range of variables that capture financial,\n\n\nlabor market, and social policy deregulation (outputs). In addition, we also investigate the e↵ect\n\n\nof CBI on a range of financial and other economic indicators (outcomes) that should be a↵ected if\n\n\nour argument is correct.\n\n\nOur baseline models are parsimonious by design. We are wary of including post-treatment\n\n\ncontrol variables that would bias our results (Angrist and Pischke, 2008; Montgomery, Nyhan,\n\n\nand Torres, 2016). Likewise, we are concerned by estimates that depend on complicated modeling\n\n\nstrategy (Leamer, 1983). Therefore, we prefer a simple model that uses within-country variation and\n\n\neliminates time shocks. In the models below, we also control for GDP and population (both logged)\n\n\nto account for di↵erences in baseline economic wealth and size. In the appendix, we include several\n\n\nmodels with a fuller set of adjustments for potential confounders (see Section 4.\n\nMention: \"data collected by Garriga\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"An alternative conflict dataset is used as a robustness check and yields similar results.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000301:12:0", "occurrence_id": "812bf615ff3d5fc8b07b3b98", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: datasets available to measure CBI. For robustness, we also use the data collected by Garriga (2016),\n\n\nwhich yields similar results.\n\n\nTo capture inequality dynamics, we rely on data from the World Bank’s PovcalNet database.\n\n\nBesides providing an extensive coverage of countries over a comparably long timespan, a key ad\n\nvantage of this dataset is that includes information on income dynamics at the decile level, which\n\n\nallows us to study income dynamics in a more fine-grained manner. Our headline results rely on\n\n\nthe share of total income earned by each decile, ordered from the lowest- (bottom 10%) to the\n\n\nhighest-earners (top 10%). Our conjecture is that the latter should benefit from CBI, whereas the\n\n\nformer should be hurt. Together, this would indicate an increase in inequality.\n\n\nIn addition, we test the validity of the channels through which CBI operates. We discuss these\n\n\nvariables in Section 5. Suffice here to say that we examine a range of variables that capture financial,\n\n\nlabor market, and social policy deregulation (outputs). In addition, we also investigate the e↵ect\n\n\nof CBI on a range of financial and other economic indicators (outcomes) that should be a↵ected if\n\n\nour argument is correct.\n\n\nOur baseline models are parsimonious by design. We are wary of including post-treatment\n\n\ncontrol variables that would bias our results (Angrist and Pischke, 2008; Montgomery, Nyhan,\n\n\nand Torres, 2016). Likewise, we are concerned by estimates that depend on complicated modeling\n\n\nstrategy (Leamer, 1983). Therefore, we prefer a simple model that uses within-country variation and\n\n\neliminates time shocks. In the models below, we also control for GDP and population (both logged)\n\n\nto account for di↵erences in baseline economic wealth and size. In the appendix, we include several\n\n\nmodels with a fuller set of adjustments for potential confounders (see Section 4.\n\nMention: \"PovcalNet database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Decile-level income dynamics over a long timespan are drawn from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000302:45:1", "occurrence_id": "bd6fba7ae8dc6cbb669ab9d3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \nPower outages (hours) 21.30 69.92 0 720 20,295\nFinance obstacle (0-4 scale) 1.32 1.25 0 4 20,295\nOverdraft Y:1 N:0 0.43 0.49 0 1 20,295\nInflation (CPI, annual %) 7.24 5.03 -1.99 53.23 20,295\n<u>Primary Enrollment (Gross rate)</u> <u>102.60</u> <u>10.93</u> <u>33.96</u> <u>145.86</u> <u>20,295</u>\n<u>Source: Author’s own calculations based on Enterprise Surveys data and World Development</u>\nIndicators (WDI, World Bank).\n\n\n44\n\nMention: \"Enterprise Surveys data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Firm-level constraints such as outages and finance obstacles are taken from this survey for the summary statistics.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000304:52:0", "occurrence_id": "3c4a36fab9a83488d716980a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |Col1|Source: Enterprise Surveys.<br>www.enterprisesurveys.org|\n|---|---|\n|Country fixed effects<br>|<br>Dummy variables (one for each country)<br>indicating the country where the firm is<br>located.<br>Source: Enterprise Surveys.<br>www.enterprisesurveys.org <br>|\n|GDP per capita (logs) <br>|<br>Log of GDP per capita (PPP adjusted and at<br>constant 2011 International Dollars). Values of<br>the variable used are for the fiscal year in<br>which the Enterprise Survey was conducted. In<br>cases where the Enterprise Survey spanned<br>more than one calendar year, the GDP per<br>capita value is the weighted average over the<br>calendar year values covered by Enterprise<br>Surveys, where the weights used are the<br>proportion of firms surveyed in the calendar<br>year.<br>Source: World Development Indicators, World<br>Bank.<br>|\n|Growth rate of GDP per capita<br>|<br>Annual percentage growth rate of real GDP per<br>capita. Values of the variable used are for the<br>fiscal year in which the Enterprise Survey was<br>conducted. In cases where the Enterprise<br>Survey spanned more than one calendar year,<br>the GDP per capita growth rate value is the<br>weighted average over the calendar year values<br>covered by Enterprise Surveys, where the<br>weights used are the proportion of firms<br>surveyed in the calendar year.<br>Source: World Development Indicators, World<br>Bank.<br>|\n|Courts: Obstacle Severity (CI avg.)<br\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"GDP per capita levels and growth rates in the variable table are taken from this series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000304:52:0", "occurrence_id": "6f7347372dc29ddd7e811201", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |Col1|Source: Enterprise Surveys.<br>www.enterprisesurveys.org|\n|---|---|\n|Country fixed effects<br>|<br>Dummy variables (one for each country)<br>indicating the country where the firm is<br>located.<br>Source: Enterprise Surveys.<br>www.enterprisesurveys.org <br>|\n|GDP per capita (logs) <br>|<br>Log of GDP per capita (PPP adjusted and at<br>constant 2011 International Dollars). Values of<br>the variable used are for the fiscal year in<br>which the Enterprise Survey was conducted. In<br>cases where the Enterprise Survey spanned<br>more than one calendar year, the GDP per<br>capita value is the weighted average over the<br>calendar year values covered by Enterprise<br>Surveys, where the weights used are the<br>proportion of firms surveyed in the calendar<br>year.<br>Source: World Development Indicators, World<br>Bank.<br>|\n|Growth rate of GDP per capita<br>|<br>Annual percentage growth rate of real GDP per<br>capita. Values of the variable used are for the<br>fiscal year in which the Enterprise Survey was<br>conducted. In cases where the Enterprise<br>Survey spanned more than one calendar year,<br>the GDP per capita growth rate value is the<br>weighted average over the calendar year values<br>covered by Enterprise Surveys, where the<br>weights used are the proportion of firms<br>surveyed in the calendar year.<br>Source: World Development Indicators, World<br>Bank.<br>|\n|Courts: Obstacle Severity (CI avg.)<br\n\nMention: \"Enterprise Surveys\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Survey fieldwork timing and the share of firms interviewed determine how macro values are averaged.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000305:25:0", "occurrence_id": "8fad1f180a85b564b9022006", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: connections are neither significantly less prevalent nor significantly less valuable in the manufacturing\n\nsector than in other sectors.\n\n\n**6 Conclusion**\n\nUsing Indonesian manufacturing plant-level survey data spanning Indonesia’s democratic transition, we\n\nhave shown that Suharto’s resignation substantially eroded the premium on being connected to him. His\n\nfall also had a modest but significant pro-competitive impact on industries in which his family had\n\nextensive business interests relative to industries that did not. Democratization thus improved competition.\n\n\nThe contrast between the erosion of the firm-level premium on being connected and weak pro\ncompetitive impacts at the industry level is plausibly in part due to the nature of privileges having been\n\ntargeted to specific firms rather than entire industries. Second, connected firms remained large; the\n\ncapacity, capital, know-how, and relationships they had accumulated during the Suharto era enabled many\n\npreviously connected firms to remain competitive even after the privileges that propelled their initial\n\ngrowth had been removed. More fundamentally, the nature of state–business relationships did not alter\n\ndramatically.\n\nNotwithstanding the persistence of cronyism, our results point toward the potential of democratic\n\ninstitutions to promote competition and curb state capture. Yet Indonesia’s experience simultaneously\n\nserves as a reminder that democratization alone does not suffice to sustain competitive markets.\n\n\n24\n\nMention: \"Indonesian manufacturing plant-level survey data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Changes in the return to political connection across the democratic transition are estimated on these plants.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000312:9:1", "occurrence_id": "4cb524eebb5b369dd3d0e35a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  was slower in industries with heavier regulation. Also, Cacciatore and Fiori (2016)\nshow that business regulation affects firm entry costs, workers’ unemployment benefits and\nfiring restrictions. They use panel VAR estimates for OECD countries to demonstrate the\nrecessionary short-run and expansionary long-run effects of reforms. The authors, thus, prove\na strong effect of market deregulation on business cycle fluctuations’ efficiency.\n\n\nIn the context of construction permitting, another area measured by _Doing Business_, Poel et al.\n(2014) find that higher administrative burdens adversely affect economic growth. Using a fixed\neffect regression analysis of data from 182 economies, the authors show a clear correlation\nbetween the number of days and procedures required to obtain a construction permit and\neconomic growth. Plus, a similar relationship exists between economic growth and days and\nprocedures to start a business as well as to pay taxes. Likewise, in the area of getting electricity,\nthe analysis of Abotsi (2016), using the Enterprise Surveys data for Africa, demonstrates that\nthe number of power outages experienced in a given month negatively impacts firms’\nproduction efficiency. Several other studies emphasize the importance of effective tax\nregulation on firm formalization. Rocha et al. (2018) analyzing firm-level administrative and\nindividual panel data in Brazil show that the costs of remaining formal in the case of taxes –\nand not entry costs – pose a serious obstacle to formalization. Specifically, reducing taxes after\nthe registration costs have already been removed lowers firm informality rates, while\nregistration costs reduction per se has no evident effect on informality. At the same time, lower\ntaxes do not necessarily impact creation of new formal firms.\n\n\nAnalyzing regulation of property registration, Berkowitz et al. (2015) show that a property law\nenactment, which grants more rights to creditors over the assets that are used for loans to private\nfirms leads to a substantial increase in firm value. And the impact is stronger on firms that are\nless\n\nMention: \"Enterprise Surveys data for Africa\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"A prior study uses this regional extract to link power outages to production efficiency.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000317:11:0", "occurrence_id": "bf38c8d1aeb34a4f9ebfd60c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: that limited use of instructional time is a contributor to poor student learning and high levels of student\ndisruption can lead to teacher burnout and decreased effectiveness (Bruns and others 2015; Ptero-Lopez\n2009). The majority of classroom observations have been conducted in regions outside MENA;\nhowever, a recent example from the West Bank and Gaza may illustrate the importance of teachers\nstaying on task (box 1).\n\n\nBox 1: Classroom Observations in the West Bank and Gaza\n\n\nThe Stallings’ classroom observation tool established a good practice benchmark of 85 percent of\ntotal class time to be used for instruction. In schools across the West Bank and Gaza, territories\nwhere students undoubtedly are faced with personal, economic and political stresses, student\nperformance is on average high compared to other MENA countries. The Stallings observations in\nclassrooms find that teachers are spending 90 percent of their time on instruction, 9 percent on\nclassroom-management and only 1 percent time-off-task (Yarrow and others 2014). While this is\nnot indicative of what could be occurring in classrooms with high levels of disciplinary issues, it is\nsuggestive that even in contexts of conflict and fragility, effective teaching practices can ensure\nhigh quality of learning. It is also indicative, that students with a higher ability for self-regulation\nare better able to learn.\n\n\n**Table 4. Marginal Impact of School Safety on Mathematics Performance, TIMSS data 2015**\n\n\nMENA countries Other countries\n\n<u>Estimated</u> <u>sd error</u> <u>Estimated</u> <u>sd error</u>\nSocioeconomic background (individual) 13.1 0.3 21.8 0.4\nSocioeconomic background (school) 56.0 2.7 55.7 2.1\nSchool safety 6.5 0.8 8.1 1.0\nN. obs. 80,218 150,382\n<u>R</u>\n\nMention: \"TIMSS data 2015\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Estimated effects of school safety on mathematics scores are computed separately for MENA and other countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000319:16:0", "occurrence_id": "d73556b4edf665b93eee3960", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 2009. Similar to the analysis performed for Jordan, we define the sample as women and men\n\n\naged 15 or more in 2006 and up to 64 in 2009. The sample includes 4,261 women and 3,641\n\n\nmen who were present in both waves of the survey. We define an indicator of household\n\n\ninternet adoption or continuation between 2006 and 2009 as the explanatory variable of\n\n\ninterest and the change in labor force participation as the outcome variable. Table 9 shows\n\n\nthat women’s labor force participation increased from 49 to 52 percent between 2006 and\n\n\n2009, while men’s rate improved from 78 to 84 percent. Internet access at the household\n\n\nlevel moved from approximately 20 to 37 percent.\n\n\nIn Chile, the public fixed telephone network has been the most significant component of\n\n\nthe internet access infrastructure at the beginning of the 2000s (Cominetti C. (2002)). Our\n\n\nhypothesis is that access to a fixed telephone line at that time is a good predictor of the\n\n\ndistance to the system’s backbone in future year, and thereby to internet access. We propose\n\n\nto use the share of households in each province that had a fixed telephone line in 2002 as\n\n\nan instrumental variable for the indicator of internet adoption or continuation between 2006\n\n\nand 2009. We used information from the national 2002 Census which shows that this share\n\n\nranged between 0.13 and 0.75.\n\n\nWe estimate OLS and IV models using individual and household observable character\n\nistics in 2006 as control variables. We include the same set of regressors as in the previous\n\n\nmodels for Jordan –age, educational level, marital status, indicator of urban area, household\n\n\nsize, and per capita household income. Models include region fixed effects and standard\n\n\nerrors are clustered at the province level. Our main results appear in Table 10. Panel A\n\n\npresents the second stage estimates, while Panel B summarizes first stage results. The share\n\n\nof household that in each province had a fixed telephone line in 2002 is a good predictor of\n\n\ninternet\n\nMention: \"national 2002 Census\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Province-level fixed telephone ownership in 2002 is used as the instrument for internet adoption.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000323:6:1", "occurrence_id": "6a03054f5397ea615b956366", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  considered the RMNCH-N care-seeking, perceptions and health outcomes for\nthe most-recent childbirth and child health episode from each DHS survey.\n\n**2.2 Variables**\n\nNineteen outcome variables (Table 2) related to service utilization, quality of care and nutrition status were\nselected from the DHS survey. Demand-side explanatory variables (or potential determinants) selected from\nthe DHS were socio-economic and demographic characteristics and location of service seekers (Table 3).\nSupply-side explanatory variables chosen from the BFHS and the DHS include – type of provider, quality\nof care, availability of services (infrastructure, drugs, equipment and skilled staff) and readiness for care\nprovision (Table 4). For the BFHS data, where possible, similar variables were combined and aggregated\nto create indices (e.g., medical equipment index and drug availability index) to reduce the variance of\nsimilar variables based on the categorization or criteria of indexing provided by the BFHS survey (more\ndetails can be found in BFHS reports). For utilization of reproductive and maternal care and quality of\nANC, the unit of analysis was an individual woman. The unit of analysis for child related indicators was a\nchild under-5. All study variables were selected based on the contextual literature and compatibility for a\ncomparative analysis across time points. <sup>16,19–22</sup>\n\n**Table 2: Outcome variables**\n\n\n\n\n\n\n\n\n\n\n|Domain|Indicator|\n|---|---|\n|<br>Utilization|<br>**Antenatal care (ANC)**<br>• <br>4+ ANC visits<br>• <br>ANC from medically trained provider<br>• <br>ANC in public sector<br>|\n|<br>Utilization|<br> <br>**Delivery**<br>• <br>Skilled\n\nMention: \"DHS survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Nineteen utilisation, quality and nutrition outcomes plus demand-side determinants are selected from this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000323:6:1", "occurrence_id": "6448b3085c22fc3f00e69f57", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  considered the RMNCH-N care-seeking, perceptions and health outcomes for\nthe most-recent childbirth and child health episode from each DHS survey.\n\n**2.2 Variables**\n\nNineteen outcome variables (Table 2) related to service utilization, quality of care and nutrition status were\nselected from the DHS survey. Demand-side explanatory variables (or potential determinants) selected from\nthe DHS were socio-economic and demographic characteristics and location of service seekers (Table 3).\nSupply-side explanatory variables chosen from the BFHS and the DHS include – type of provider, quality\nof care, availability of services (infrastructure, drugs, equipment and skilled staff) and readiness for care\nprovision (Table 4). For the BFHS data, where possible, similar variables were combined and aggregated\nto create indices (e.g., medical equipment index and drug availability index) to reduce the variance of\nsimilar variables based on the categorization or criteria of indexing provided by the BFHS survey (more\ndetails can be found in BFHS reports). For utilization of reproductive and maternal care and quality of\nANC, the unit of analysis was an individual woman. The unit of analysis for child related indicators was a\nchild under-5. All study variables were selected based on the contextual literature and compatibility for a\ncomparative analysis across time points. <sup>16,19–22</sup>\n\n**Table 2: Outcome variables**\n\n\n\n\n\n\n\n\n\n\n|Domain|Indicator|\n|---|---|\n|<br>Utilization|<br>**Antenatal care (ANC)**<br>• <br>4+ ANC visits<br>• <br>ANC from medically trained provider<br>• <br>ANC in public sector<br>|\n|<br>Utilization|<br> <br>**Delivery**<br>• <br>Skilled\n\nMention: \"BFHS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Supply-side variables such as provider type, infrastructure and staff skill come from this facility survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000323:6:1", "occurrence_id": "98d63e5e479477d3833337ce", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  considered the RMNCH-N care-seeking, perceptions and health outcomes for\nthe most-recent childbirth and child health episode from each DHS survey.\n\n**2.2 Variables**\n\nNineteen outcome variables (Table 2) related to service utilization, quality of care and nutrition status were\nselected from the DHS survey. Demand-side explanatory variables (or potential determinants) selected from\nthe DHS were socio-economic and demographic characteristics and location of service seekers (Table 3).\nSupply-side explanatory variables chosen from the BFHS and the DHS include – type of provider, quality\nof care, availability of services (infrastructure, drugs, equipment and skilled staff) and readiness for care\nprovision (Table 4). For the BFHS data, where possible, similar variables were combined and aggregated\nto create indices (e.g., medical equipment index and drug availability index) to reduce the variance of\nsimilar variables based on the categorization or criteria of indexing provided by the BFHS survey (more\ndetails can be found in BFHS reports). For utilization of reproductive and maternal care and quality of\nANC, the unit of analysis was an individual woman. The unit of analysis for child related indicators was a\nchild under-5. All study variables were selected based on the contextual literature and compatibility for a\ncomparative analysis across time points. <sup>16,19–22</sup>\n\n**Table 2: Outcome variables**\n\n\n\n\n\n\n\n\n\n\n|Domain|Indicator|\n|---|---|\n|<br>Utilization|<br>**Antenatal care (ANC)**<br>• <br>4+ ANC visits<br>• <br>ANC from medically trained provider<br>• <br>ANC in public sector<br>|\n|<br>Utilization|<br> <br>**Delivery**<br>• <br>Skilled\n\nMention: \"BFHS data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"Similar facility items are combined into equipment and drug availability indices to reduce variance.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000323:6:1", "occurrence_id": "38a072b2eed92b50d2acf871", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  considered the RMNCH-N care-seeking, perceptions and health outcomes for\nthe most-recent childbirth and child health episode from each DHS survey.\n\n**2.2 Variables**\n\nNineteen outcome variables (Table 2) related to service utilization, quality of care and nutrition status were\nselected from the DHS survey. Demand-side explanatory variables (or potential determinants) selected from\nthe DHS were socio-economic and demographic characteristics and location of service seekers (Table 3).\nSupply-side explanatory variables chosen from the BFHS and the DHS include – type of provider, quality\nof care, availability of services (infrastructure, drugs, equipment and skilled staff) and readiness for care\nprovision (Table 4). For the BFHS data, where possible, similar variables were combined and aggregated\nto create indices (e.g., medical equipment index and drug availability index) to reduce the variance of\nsimilar variables based on the categorization or criteria of indexing provided by the BFHS survey (more\ndetails can be found in BFHS reports). For utilization of reproductive and maternal care and quality of\nANC, the unit of analysis was an individual woman. The unit of analysis for child related indicators was a\nchild under-5. All study variables were selected based on the contextual literature and compatibility for a\ncomparative analysis across time points. <sup>16,19–22</sup>\n\n**Table 2: Outcome variables**\n\n\n\n\n\n\n\n\n\n\n|Domain|Indicator|\n|---|---|\n|<br>Utilization|<br>**Antenatal care (ANC)**<br>• <br>4+ ANC visits<br>• <br>ANC from medically trained provider<br>• <br>ANC in public sector<br>|\n|<br>Utilization|<br> <br>**Delivery**<br>• <br>Skilled\n\nMention: \"BFHS survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"The survey's own reports supply the categorisation rules used for indexing.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000325:7:1", "occurrence_id": "20719137ed5bea56c5f10edc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  (GHS Panel), the Tanzania\nNational Panel Survey (TZNPS), and the Uganda National Panel Survey (UNPS).\n4 The 2010 survey round in Nigeria, and the survey rounds in Niger were not considered because the distinction\nbetween wage and self-employment cannot be made. In the case of Tanzania, the 2008 survey round was not\nconsidered because the definition of wage employment was not comparable to the definition used in the later rounds.\n5 Household membership is defined as having spent at least six months (three in case of Tanzania) in the household\nduring the past 12 months.\n6 To correct for potential attrition bias, we run a logit model with a binary dependent variable taking the value of one\nif the individual appears in the sample at least twice. We control for variables at individual-level (age, gender,\neducation level, current school enrollment, marital status, relation to household head and employment status) and at\nthe household-level (household size, dependency ratio, wealth index (a factor analysis-based index of household\ndwelling attributes and durable asset ownership) and participation in farm activities). Baseline enumeration area fixed\neffects are also included in the specification. We predict the probability of non-attrition for each individual and\ncalculate the attrition weight as the inverse of this probability. To make the sample nationally representative, we use\nthe sampling weights calculated by the national statistical offices and multiply it with the attrition weight. We\nwinsorize the weights at the top 1 percent level to minimize the effects of potential outliers.\n\n\n6\n\nMention: \"GHS Panel\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"One of the four country panels used, though its 2010 round is excluded for classification reasons.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000328:34:0", "occurrence_id": "af348d207333e518fb4c1a86", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 4:** **Event Study Economic Index**\n\n\n\n\n\n\n\nNotes: Event study coefficients are plotted in the figure for economic index. Event time is defined as the age when\nthe midwife program started in the community of residence. The age of exposure is calculated using birth year and\nthe year the midwife started working in the community. Age greater than 9 or above is the omitted category.\nSource: Authors’ analysis based on The Indonesian Family Life Survey.\n\n\n33\n\nMention: \"Indonesian Family Life Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Event study coefficients on the economic index are estimated from this household panel.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000334:7:1", "occurrence_id": "8695909e32d8cb513c0dd71a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  the Moody’s ratings\non our estimated ratings, the estimated and actual ratings have a pairwise correlation of 0.8 with\n89 percent of the actual ratings within 4 notches of the best-fit line. There is no sign of any\ndirectional bias in the estimates (Figure 3).\n\n\n\n**Table 1: Mapping of numerical ratings**\n**corresponding to published Moody’s**\n\n\n\n**Figure 3: Actual and estimated Moody’s ratings**\n\n\n\n\n\n\n\nHaving constructed our dependent variable (estimated ratings), we turned next to the data\ncapturing the quality of economic governance, policies, and institutions. For this, we used the\nWorld Bank’s Country Policy and Institutional Assessment (CPIA) scores. These assess how well\ncountries’ policy and institutional frameworks function in support of growth, poverty reduction,\n\n\n6\n\nMention: \"Country Policy and Institutional Assessment\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Governance, policy and institutional quality is captured by these scores.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000335:6:0", "occurrence_id": "73185965b8ff231142a690de", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: investments by foreign companies. Third, from a practical viewpoint, data are available on these three\ndimensions at the NUTS-3 level, but data are missing on some other dimensions. European official\nstatistical sources provide information on a large set of variables for NUTS-1 and NUTS-2 regions,\nbut only a reduced amount of information is available at the NUTS-3 level.\n\n**Unfortunately, it was not possible to include data series on firm size.** We would like to\nestimate the role of micro (fewer than 10 employees), small (10 to 49 employees), medium (50 to 249\nemployees), and large (250 employees or more) firms in regional economic growth. Such analysis\ncould help reveal the relative importance of initiatives on SMEs in lagging regions, versus initiatives\nto boost competitiveness of large firms as drivers of growth. Unfortunately, these data series –\navailable from the _Regio_ database provided by Eurostat – present two key shortcomings. First,\nregional data are based on a too broad classification of firms with respect to their size, as data are\navailable for zero-employment firms, firms with 1 to 9 employees, and firms with 10 or more\nemployees. This represents a key problem, as small, medium, and large firms are grouped together in\na unique size class, such that it is impossible to capture the role of SMEs versus large firms. Second,\nbusiness demography data that are available only from the year 2008 – while our analysis starts in\nthe year 2003 – do not cover nine EU Member States, and present severe missing values or\ninconsistent estimates over time for the remaining countries. <sup>12</sup> Since the lack of availability is not\nsystematic, we would risk bias in the analysis if we included only the available data.\n\n**We use data for the 2003 to 2017 period, from Eurostat, the Organisation for Economic**\n**Co-operation and Development (OECD),\n\nMention: \"_Regio_ database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Regional business statistics are available but too coarse by firm size and limited to a short period.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000338:8:0", "occurrence_id": "4a4bae81ca99ec359c0d20e8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 7\n\n\n\n7\n\n~~characteristics of households from the same caste but not having agriculture as their main activity,~~\n_UnonAGRc_ <u>. Given the social structure of India, assuming that unobserved characteristics are correlated</u>\nwithin castes across occupations seems reasonable. This assumption is discussed in more detail in section\n5. <sup>5</sup> [^5: It is also worth noting that _δ_ <sup>ˆ</sup> _o_ and _δ_ <sup>ˆ</sup> _nonAGRc_ provide insights into the bias in the estimates of the other parameters of the]\n\n\n\n**3.** **Data and Descriptive Statistics**\n\n\n\nData and Variables Definition\nThe data used to conduct this study are the ARIS-REDS data from the National Council of Applied Economic\nResearch (NCAER). Since 1971, the NCAER has been conducting surveys on a _sample_ of households in 232\nvillages in the 17 major states of India. Most of the results come from the round of 2006, which has the\nadvantage of providing a _listing_ of every household in the surveyed villages with basic information on\nhouseholds and heads of household such as their demographic characteristics (gender, age, education\nlevel, size of the household) and their main occupation. The average level of education   of members from\nthe same caste can therefore be computed using all the heads of the households from the same caste in the\nvillage and not only the heads of the households that have been sampled, which makes the measure very\nprecise. The information on education from the listing is also useful to deal with measurement error in\neducation of the head of household, as is shown in section 4. The round of 1982 is used for robustness\nchecks.\n\nIn the round of 2006, detailed information about agricultural inputs, such as cultivated area or labor\nuse, is not provided in the listing, so the analysis is restricted to households that cultivate land in the\n\n\n\neducation of the head of household, as is shown in section 4. The round\n\nMention: \"ARIS-REDS data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Village-level household surveys across major states since 1971 provide the study's data, mostly the 2006 round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000342:40:1", "occurrence_id": "39710791caad093bfa2ff7cf", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 01<br>-0.02<br>(0.03)<br>(0.06)<br> <br> <br>0.03<br>-0.01<br>(0.03)<br>(0.06)<br> <br> <br>-1.83<br>4.3<br>(1.36)<br>(2.71)|\n|Controls<br> <br>Municipality fixed<br>effects<br> Time fixed effect|<br> <br>Yes<br>Yes<br>Yes<br>Yes|<br> <br>Yes<br>Yes<br>Yes<br>Yes|\n\n\n_Notes:_ Each entry in panels A–B, rows a–c, columns 2–5 represents a coefficient from a different regression. All the\n\n\nregression include municipality fixed effects, year fixed effects, a vector of maternal background, and a vector of\n\n\nmunicipality characteristics. All the regressions control for cluster by municipality, and standard error are consistent\n\n\nfor heteroskedasticity and error correlation within municipalities over time.\n\n\n*** _p_ < .01 ** _p_ < .05 * _p_ < .1 _Source:_ Underlying data are from the Chilean socio-economic household\n\n\nsurvey (CASEN) and administrative data from the Ministry of Education for 1990–2011.\n\n\n39\n\nMention: \"Chilean socio-economic household\n\n\nsurvey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Household outcomes are combined with education ministry records in the municipality-clustered regressions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000346:59:0", "occurrence_id": "a757b30807218469d9c81e0d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ##### Table 1a: Financial loan data for Indian manufacturing plants\n\n1989 1994 2000 2005 2010\n\n\nA. Raw count of data observations after basic pruning\n\n\nTotal 127,786 197,992 244,339 116,175 135,270\nOrganized sector 33,356 42,649 25,571 35,101 33,077\nUrban areas 22,899 29,024 15,296 20,379 19,323\nRural areas 10,457 13,625 10,275 14,722 13,754\nUnorganized sector 94,430 155,343 218,768 81,074 102,193\nUrban areas 40,584 53,262 131,652 40,062 49,818\nRural areas 53,846 102,081 87,116 41,012 52,375\n\n\nB. Estimated number of plants in India using sample weights from Panel A\n\n\nTotal 13,350,304 12,120,050 16,978,168 16,938,433 17,797,958\nOrganized sector 75,442 93,035 99,101 104,581 118,150\nUrban areas 54,675 64,921 62,835 63,044 72,750\nRural areas 20,767 28,114 36,266 41,537 45,400\nUnorganized sector 13,274,862 12,027,015 16,879,067 16,833,852 17,679,808\nUrban areas 2,823,783 2,967,126 5,048,512 4,901,061 7,396,214\nRural areas 10,451,079 9,059,888 11,830,554 11,932,791 10,283,594\n\n\nC. Share of raw data observations from Panel A that report positive levels\n\nof financial loans\n\n\nTotal 0.34 0.28 0.19 0.38 0.28\nOrganized sector 0.90 0.90 0.90 0.90 0.89\nUrban areas 0.92 0.91 0.91\n\nMention: \"Financial loan data for Indian manufacturing plants\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Observation counts after pruning are reported by year, sector and urbanicity for this loan extract.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000348:27:0", "occurrence_id": "e284bfa49c0dae874021f051", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: https://databank.worldbank.org/data/source/world-development-indicators,world-developmentindicators,world-development-indicators#.\n\nWorld Bank. 2017. _Implementation Completion and Results Report for the Nepal: Vocational Education_\n\n_Project (P104015)_ . Washington, DC: World Bank.\n\n\n26\n\nMention: \"world-development-indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"A databank link to this indicator source is given as the data provenance for the entry.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000348:27:0", "occurrence_id": "70d453d7ea0ef1cb807c1540", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: https://databank.worldbank.org/data/source/world-development-indicators,world-developmentindicators,world-development-indicators#.\n\nWorld Bank. 2017. _Implementation Completion and Results Report for the Nepal: Vocational Education_\n\n_Project (P104015)_ . Washington, DC: World Bank.\n\n\n26\n\nMention: \"world-developmentindicators\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"A repeated tag in the same databank URL fragment names the same source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000349:21:1", "occurrence_id": "964209be21888b5d8a06ddf0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  (HS) 1992 nomenclature, for\nmore than 200 countries from 1995 to 2016. <sup>17</sup> An advantage of using BACI rather than the\nunderlying information from UN Comtrade is that the same trade flow, which can be reported\ndifferently by the exporter and importer, has been reconciled in order to have a single statistic\non each directional bilateral relationship. BACI only reports positive trade flows and we\nbalance the dataset along three dimensions (exporter, product, and time) by including zerovalued trade flows.\n\nWe measure changes in market access in two ways: whether the exporter-product pair is\nunder a preferential trade agreement (discrete measure) and the magnitude of the preferences\ngranted (continuous measure). To construct the latter, we use information on ad-valorem tariff\nrates applicable under each preferential scheme—GSP, EBA and GSP+ for imports into the\nEU and AGOA for imports into the United States—for all beneficiary countries. These data\nare obtained from WITS, a database maintained by the World Bank which provides access to\nseveral international measures. The original source of tariffs rates in WITS is UNCTAD\nTRAINS. In order to calculate the preferential tariff margin, defined as the difference between\npreferential and non-preferential rates, we also include the MFN tariff rate for all products.\nThe WITS database contains an identifier for groups of countries to which a particular tariff\n\n\n17 Original data are provided by the United Nations Statistical Division (COMTRADE database). BACI is constructed using\na procedure which reconciles the declaration of importers and exporters as explained in Gaulier and Zignago (2010).\n\n20\n\nMention: \"WITS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Preferential tariff rates under GSP, EBA, GSP+ and AGOA are obtained from this trade database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000349:21:1", "occurrence_id": "3d138b4759f203f01c3d686c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  (HS) 1992 nomenclature, for\nmore than 200 countries from 1995 to 2016. <sup>17</sup> An advantage of using BACI rather than the\nunderlying information from UN Comtrade is that the same trade flow, which can be reported\ndifferently by the exporter and importer, has been reconciled in order to have a single statistic\non each directional bilateral relationship. BACI only reports positive trade flows and we\nbalance the dataset along three dimensions (exporter, product, and time) by including zerovalued trade flows.\n\nWe measure changes in market access in two ways: whether the exporter-product pair is\nunder a preferential trade agreement (discrete measure) and the magnitude of the preferences\ngranted (continuous measure). To construct the latter, we use information on ad-valorem tariff\nrates applicable under each preferential scheme—GSP, EBA and GSP+ for imports into the\nEU and AGOA for imports into the United States—for all beneficiary countries. These data\nare obtained from WITS, a database maintained by the World Bank which provides access to\nseveral international measures. The original source of tariffs rates in WITS is UNCTAD\nTRAINS. In order to calculate the preferential tariff margin, defined as the difference between\npreferential and non-preferential rates, we also include the MFN tariff rate for all products.\nThe WITS database contains an identifier for groups of countries to which a particular tariff\n\n\n17 Original data are provided by the United Nations Statistical Division (COMTRADE database). BACI is constructed using\na procedure which reconciles the declaration of importers and exporters as explained in Gaulier and Zignago (2010).\n\n20\n\nMention: \"WITS database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Its country-group identifiers allow MFN and preferential rates to be matched to beneficiary groups.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000349:21:1", "occurrence_id": "2f69d5c9cc612c62499503a4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  (HS) 1992 nomenclature, for\nmore than 200 countries from 1995 to 2016. <sup>17</sup> An advantage of using BACI rather than the\nunderlying information from UN Comtrade is that the same trade flow, which can be reported\ndifferently by the exporter and importer, has been reconciled in order to have a single statistic\non each directional bilateral relationship. BACI only reports positive trade flows and we\nbalance the dataset along three dimensions (exporter, product, and time) by including zerovalued trade flows.\n\nWe measure changes in market access in two ways: whether the exporter-product pair is\nunder a preferential trade agreement (discrete measure) and the magnitude of the preferences\ngranted (continuous measure). To construct the latter, we use information on ad-valorem tariff\nrates applicable under each preferential scheme—GSP, EBA and GSP+ for imports into the\nEU and AGOA for imports into the United States—for all beneficiary countries. These data\nare obtained from WITS, a database maintained by the World Bank which provides access to\nseveral international measures. The original source of tariffs rates in WITS is UNCTAD\nTRAINS. In order to calculate the preferential tariff margin, defined as the difference between\npreferential and non-preferential rates, we also include the MFN tariff rate for all products.\nThe WITS database contains an identifier for groups of countries to which a particular tariff\n\n\n17 Original data are provided by the United Nations Statistical Division (COMTRADE database). BACI is constructed using\na procedure which reconciles the declaration of importers and exporters as explained in Gaulier and Zignago (2010).\n\n20\n\nMention: \"COMTRADE database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It is the original source of the bilateral trade flows from which the BACI dataset is reconciled.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000350:14:0", "occurrence_id": "20faecd61c1e00aed8792dad", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: school level, teachers complete the questionnaires delivered to them. These questionnaires\n\n\nare collected by district government officials and returned to the provincial education\n\n\ndepartment, where the information is digitized.\n\n\nUsing the annual censuses, we construct a school-level panel. <sup>13</sup> [^13: While government schools are assigned unique identification (ID) numbers, school ID protocols were adjusted\nby the provincial and district education departments in certain years and assigned school IDs were not strictly\nadhered to during data collection, entry, and processing. Consequently, to construct an accurate school-level panel,\nwe worked closely with provincial education department officials with intimate working knowledge of the Annual\nSchool Census data and ground-level knowledge of schools.] On the basis of this\n\n\npanel, we construct the outcome measure—the absolute change in school-level girls’\n\n\nenrollment in secondary grades for six subperiods of different lengths originating from\n\n\nbaseline (2003): 2003–05, 2003–07, 2003–09, 2003–11, 2003–13, and 2003–15.\n\n\n_Empirical strategy_\n\n\nWe estimate program effects on the basis of a sharp RD design. One of two outcomes is\n\n\npossible for school 𝑠, conditional on the district literacy rate 𝑋�. Schools in districts with a\n\n\nliteracy rate below 40 percent ( _X_ _D_  40) are covered by the program. These schools have a\n\n#### potential girls’ secondary school enrollment outcome of Y s  1 . Schools in districts with a\n\n\nliteracy rate equal to or above 40 percent ( _X_ _D_  40) are not covered by the program. These\n\n\nschools have a potential outcome of _YS_ (0) . In theory, the average observed outcomes, given\n\n#### X D, are two functions:  YS (1)| XD  if X D  40 and  YS (0)| X D  if X D  40 . The sharp\n\n##### RD program effect at the cutoff,  RD, reflects the difference\n\nMention: \"Annual\nSchool Census data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"Officials' knowledge of these census records was used to repair an unreliable school panel before measuring enrollment gains.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000356:19:1", "occurrence_id": "9e696f1576b52df784548085", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  2000; Djankov et al., 2008).\n\n\n**4.5** **Publicly** **available** **data**\n\n\nUp to this point, we have conducted the analysis with a restricted dataset from the BIS that\n\n\nallows us to break down each country’s total foreign deposits, which is public information, into\n\n\ndeposits in havens and deposits in non-havens, which is not publicly available. To enhance\n\n\ntransparency and to facilitate work by other researchers on aid and foreign deposits, we show\n\n\nthat results similar to our main results can be obtained with a publicly available dataset from\n\n\nthe BIS. This recently released data includes quarterly data on cross-border deposits at the\n\n\nbilateral level for a selected group of banking centers.\n\n\nTable 6 summarizes the publicly available information. In our main sample of 22 highly aid\n\ndependent countries (Column 1), the average of total foreign deposits taken across all quarters\n\n\nin the sample period 1999-2010 stands at $199 million (corresponding to the sum of Columns\n\n\n2 and 3 in Table 1). With the public dataset, 29% of these deposits can be assigned to six\n\n\n18Not only does the point estimate on aid increase as we raise the threshold, the ratio of haven deposits to\nGDP also increases, which implies a higher leakage rate for a given point estimate.\n\n\n17\n\nMention: \"restricted dataset from the BIS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"It splits total foreign deposits into haven and non-haven components, which are not public.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000364:13:0", "occurrence_id": "0988638fbc46ed7c7579c6e3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 2. IPCA price level by State (January 2018 = 1)**\n\n\n1.05\n\n\n1.03\n\n\n1.01\n\n\n0.99\n\n\n0.97\n\n\n0.95\n\n\n0.93\n\n\nRO AC AM RR PA AP TO MA PI\n\nCE RN PB PE AL SE BA MG ES\n\nRJ SP PR SC RS MS MT GO DF\n\n\n\n_Source:_ Own compulation using IBGE data.\n_Note:_ Three-month moving average shown.\n\nWe use household-specific Paasche food price indices to adjust average cost per calorie for spatial price\ndifferences:\n\n𝑝\n\n\n\n(3)\n\n\n\n−1\n\n𝑝 <sup>**𝑃**</sup> <sup>𝑃</sup> <sup>𝑖</sup>\n\n\n\n𝑃𝑃𝐹𝐹ℎ\n\n<sup>**𝑃**</sup> **𝑃**\n\n\n\n𝐼𝐼\n\n𝐹𝐹ℎ = ��𝑤𝑤ℎ,𝑖𝑖\n\n𝑖𝑖=1\n\n<sup>**𝑃**</sup> **𝑃**\n\n\n\n𝑖𝑖=1\n\n<sup>**𝑃**</sup> **𝑃**\n\n\n\n𝐼𝐼\n\n\n\n<u>𝑝𝑝</u> <sup><u>0,𝑖</u></sup>\n\n~~𝑝~~ 𝑝 <sup>~~𝑃~~</sup> <sup>**𝑃**</sup> <sup>~~𝑃~~</sup> <sup>𝑃</sup> <sup>~~,𝑖~~</sup> <sup>𝑖</sup> <sup>~~�~~</sup>\n\n<sup>**𝑃**</sup> **𝑃**\n\n\n\n<u>𝑝𝑝</u>\n\nMention: \"IBGE data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"State-level consumer price indices are compiled from this statistics office to build household price deflators.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000367:18:0", "occurrence_id": "6dc7afd67c9980e0caca002f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **_Capital-output ratio_**\n\nThe consequence of these very different sectoral trends has been that the overall capital-output stock\nof government assets (schools, hospitals and offices _etc_ .) has only grown slightly faster than GDP, with\nthe ratio of the capital stock to overall GDP remaining low at between 0.2 and 0.3 (Figure 8, Panel A).\nThe increase in the overall capital-output ratio has been driven by the significant rise in the capitaloutput ratio in the infrastructure sector, predominately related to assets for transport and urban\ndevelopment.\n\nThe housing sector has been the other source of the rise in the overall capital-output ratio. Prior to\nthe reform of the housing sector at the turn of the century, investment in housing had been neglected.\nHousing investment picked up markedly a decade ago. Given the rapid increase in GDP after the\ncountry’s entry into the WTO, the pickup in growth of housing stock did not, at first, generate an\nincrease in the ratio of the housing stock to GDP. Housing investment rose to a new high in 2007,\nfueled by expectations that rapid price rises would continue. This coupled with a slackening in the\ngrowth of GDP has resulted in a steady increase in the ratio of the housing stock to GDP from 2007\nonwards (Figure 8, Panel A). The business sector capital-output ratio has increased somewhat whether\nmeasured relative to total GDP (Figure 8, Panel A) or relative to GDP in the business sector (Figure 8,\nPanel B).\n\n**Figure 8 Capital-output ratio by economic sector**\n\nConstant prices\n\nA. Capital stock by sector as ratio of total GDP B. Capital stock by sector as ratio of sectoral GDP\n\n\n\n\n\n\n\n\n\n\n\n_Source:_ Author estimates using NBS data\n\n\nThe above estimates include an estimate for the stock of R&D capital. The stock of this type of capital\nis, however, quite small. Using the revisions to the estimates GFC\n\nMention: \"NBS data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Sectoral capital stock relative to output is estimated by the author from these national accounts.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000370:4:1", "occurrence_id": "b671148753980524ddc48390", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  this imputation method. Second, we exploit a richer and more diverse</mark>\n\n<mark>set of data than the Jordan study, which includes r</mark> egistration data, census-type targeting data, and\n\na household consumption survey. We also rigorously examine the imputation method against\n\ndifferent poverty lines, including the food poverty line, the national poverty line, the international\n\npoverty line, and various other simulated values. Finally <mark>, in addition to the poverty imputation</mark>\n\n<mark>exercise, we test the performance of the proposed method for targeting purposes. In particular, we</mark>\n\n<mark>compare the targeting performance of our method with the targeting method currently used in Chad</mark>\n\n<mark>to administer cash assistance to refugees as well as with the global experience.</mark>\n\n\n<mark>The estimation results indicate that the limited set of variables available in the UNHCR registration</mark>\n\n<mark>data predict household consumption (welfare) reasonably well. Estimates from the three sets of</mark>\n\n<mark>data available for the analysis produce similar welfare figures. Th</mark> e <mark>current targeting strategy in</mark>\n\n<mark>Chad, which is used jointly by the N</mark> ational Commission on the Welcoming and Resettlement of\n\nRefugees <mark>(CNARR), UNHCR, and World Food Programme (WFP), is accurate in predicting</mark>\n\n<mark>household welfare. However, our results suggest that this targeting strategy could be further</mark>\n\n<mark>improved by reducing the inclusion and exclusion errors. If these encouraging results are replicated</mark>\n\n\n3\n\nMention: \"UNHCR registration\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Welfare is predicted from the limited variables captured during registration to test the targeting method.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000372:26:2", "occurrence_id": "ee62714c56ab1b99ae36d2b1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Sample Survey Office (NSSO) Surveys.\n\nModel 1: Final Model\n\nModel 2: District dummies*Time Trend\n\nModel 3: Expenditures at the Extensive Margin\n\nModel 4: Constant Coefficient Model\n\n\nTable 8: Preferred Model vs. Typical Line-Up Method Predictions for 2014-15\n\n\n**<u>Model</u>** **<u>Line-up</u>**\nNational 14.6 11.0\nUrban 10.0 7.1\n<u>Rural</u> <u>16.8</u> <u>12.9</u>\n\n\nSource: India National Sample Surveys\n\n(NSSO) Surveys.\n\n\n25\n\nMention: \"India National Sample Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Model predictions of poverty for 2014-15 are benchmarked against these survey estimates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000378:12:0", "occurrence_id": "753272b3711f2fd6ae2dad12", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **4.** **International allocation of pension funds**\n\n\nPension assets are a large component of global capital markets and have grown in size over\n\n\nthe last two decades (Figure 1). Although total pension assets reached $48.1 trillion at the end of\n\n\n2019, the growth has not been constant over time and it has been mostly uneven across countries.\n\n\nFor example, while pension contributions peaked in 2005, these declined after the 2008 Global\n\n\nFinancial Crisis (GFC) and have not reached pre-crisis levels since (Figure 2). Following the drop\n\n\nin contributions, and exacerbated by lower portfolio valuations during the GFC, pension assets\n\n\nexperienced a decline in size relative to global GDP between 2008 and 2011. After 2012, global\n\n\npension assets experienced modest growth, rising from 25% to 30% of GDP at the end of 2019.\n\n\nAccording to the OECD Global Pension Statistics, during the last 20 years, pension funds\n\n\nacross the world have significantly increased their share in foreign holdings. In 2002, pension\n\n\nfunds had 16% of their portfolio allocated to international investments. Foreign exposure rose\n\n\nconsistently in the next two decades, reaching 36% of total assets by 2019 (Figure 3). Although\n\n\nthe level and pace at which pension funds have increased their foreign assets vary across regions,\n\n\nthe data suggests that over time, pension funds systematically gain exposure to international assets\n\n\n(see Figure 4). For example, while pension funds in ECA and MENA countries had almost no\n\n\nforeign exposure in 2002, the share of foreign assets grew to 11% and 20% respectively by 2019.\n\n\nWe find similar patterns in Latin America, Western Europe and East Asia and Pacific regions.\n\n\n**Figure 1.** Pension assets **Figure 2.** Pension funds’ contributions\n\nMention: \"OECD Global Pension Statistics\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Foreign asset shares of pension funds over two decades are reported in these statistics.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000381:28:0", "occurrence_id": "d44116438ff3b7c9c314e313", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ##### **Appendix C. Countries with negative subsidies**\n\nTable C.1 lists the countries (most of them developed) that present negative subsidies in our estimation.\nThese so-called negative subsidies do not represent actual transfers from consumers (or the state) to\nutilities. Instead, they most probably result from (1) water loss or staffing levels below those assumed to\nbe efficient in the Chilean model, (2) assumptions in the methodology (which relied on data from Chile\nthat cannot necessarily be applied to other contexts) or (3) inaccurate data in the Benchmarking Network\nfor Water and Sanitation Utilities (IBNET) database.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n|Country|W % covered<br>by IBNET|WW % covered<br>by IBNET|Water Subsidy|Wastewater<br>Subsidy|Total Net Subsidy|\n|---|---|---|---|---|---|\n|Australia<br>|<br>81.6<br>|<br>79.5<br>|-2586.9<br>|<br>-2646.3<br>|-5233.2<br>|\n|<br>Czech Republic<br>|<br>58.4<br>|<br>56.5<br>|<br>-43.9<br>|<br>-219.9<br>|<br>-263.9<br>|\n|<br>Denmark<br>|<br>7.5<br>|<br>10.8<br>|<br>-11.8<br>|<br>-72.9<br>|<br>-84.7<br>|\n|<br>Norway<br>|<br\n\nMention: \"data from Chile\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Efficient water loss and staffing benchmarks are assumed from the Chilean model, which may not transfer elsewhere.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000381:28:0", "occurrence_id": "deabe83a09e0458455972b85", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ##### **Appendix C. Countries with negative subsidies**\n\nTable C.1 lists the countries (most of them developed) that present negative subsidies in our estimation.\nThese so-called negative subsidies do not represent actual transfers from consumers (or the state) to\nutilities. Instead, they most probably result from (1) water loss or staffing levels below those assumed to\nbe efficient in the Chilean model, (2) assumptions in the methodology (which relied on data from Chile\nthat cannot necessarily be applied to other contexts) or (3) inaccurate data in the Benchmarking Network\nfor Water and Sanitation Utilities (IBNET) database.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n|Country|W % covered<br>by IBNET|WW % covered<br>by IBNET|Water Subsidy|Wastewater<br>Subsidy|Total Net Subsidy|\n|---|---|---|---|---|---|\n|Australia<br>|<br>81.6<br>|<br>79.5<br>|-2586.9<br>|<br>-2646.3<br>|-5233.2<br>|\n|<br>Czech Republic<br>|<br>58.4<br>|<br>56.5<br>|<br>-43.9<br>|<br>-219.9<br>|<br>-263.9<br>|\n|<br>Denmark<br>|<br>7.5<br>|<br>10.8<br>|<br>-11.8<br>|<br>-72.9<br>|<br>-84.7<br>|\n|<br>Norway<br>|<br\n\nMention: \"Benchmarking Network\nfor Water and Sanitation Utilities\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Utility coverage shares come from this database, though inaccurate entries may produce negative subsidies.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000358:33:0", "occurrence_id": "9678890e440ca74ce05e4eee", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 7. Appendix\n\n\nFigure A shows the (rural) employment rate, disaggregated by sex, in Ethiopia and Malawi\naccording to three types of household surveys reviewed in this paper (DHS, LSMS‐ISA, LFS) as\nwell as the most recent census. It illustrates that survey design matters. The rural employment\nrate in Ethiopia (panel 1) differs considerably depending on the type of survey used to estimate\nit, whereas the employment rate in Malawi is relatively similar across surveys (panel 2). Besides\ndifferences in levels, trends also diverge across surveys. The Ethiopia LSMS‐ISA surveys, for\ninstance, suggest a decreasing rural employment rate, whereas DHS points to the other\ndirection. What explains the differences across surveys is hard to tell, but candidate\nexplanations include differences in (1) the definition of employment, (2) questionnaire design\nincluding the structure of the questionnaire and the phrasing of questions, and (3)\nimplementation in the field such as the number and timing of visits and the training of\nenumerators.\n\n\nFigure A: Different surveys, different results?\n\n1. Ethiopia’s rural employment rate 2. Malawi’s employment rate\n(population aged 15 to 49) (population aged 15 to 49)\n\n\nSource: Based on 2015/16 Ethiopia and 2016/17 Malawi LSMS‐ISA surveys complemented with aggregated statistics from\nDHS, LFS and census data. More detailed results are available upon request.\nNote: We attempted to make the population as comparable as possible across surveys. We calculated the ‘rural’ employment\nrate in Ethiopia (rather than the ‘total’ employment rate as in Malawi) because the first wave of the Ethiopia LSMS‐ISA survey\ndid not cover large towns. For both Ethiopia and Malawi, the employment rate was estimated for the population aged 15 to\n49 because the DHS do not interview household members over 50 years of age.\n\n\n32\n\nMention: \"DHS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its employment estimates are compared across instruments and diverge in trend from others.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000358:33:0", "occurrence_id": "9bc2606361900a8e168af51c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 7. Appendix\n\n\nFigure A shows the (rural) employment rate, disaggregated by sex, in Ethiopia and Malawi\naccording to three types of household surveys reviewed in this paper (DHS, LSMS‐ISA, LFS) as\nwell as the most recent census. It illustrates that survey design matters. The rural employment\nrate in Ethiopia (panel 1) differs considerably depending on the type of survey used to estimate\nit, whereas the employment rate in Malawi is relatively similar across surveys (panel 2). Besides\ndifferences in levels, trends also diverge across surveys. The Ethiopia LSMS‐ISA surveys, for\ninstance, suggest a decreasing rural employment rate, whereas DHS points to the other\ndirection. What explains the differences across surveys is hard to tell, but candidate\nexplanations include differences in (1) the definition of employment, (2) questionnaire design\nincluding the structure of the questionnaire and the phrasing of questions, and (3)\nimplementation in the field such as the number and timing of visits and the training of\nenumerators.\n\n\nFigure A: Different surveys, different results?\n\n1. Ethiopia’s rural employment rate 2. Malawi’s employment rate\n(population aged 15 to 49) (population aged 15 to 49)\n\n\nSource: Based on 2015/16 Ethiopia and 2016/17 Malawi LSMS‐ISA surveys complemented with aggregated statistics from\nDHS, LFS and census data. More detailed results are available upon request.\nNote: We attempted to make the population as comparable as possible across surveys. We calculated the ‘rural’ employment\nrate in Ethiopia (rather than the ‘total’ employment rate as in Malawi) because the first wave of the Ethiopia LSMS‐ISA survey\ndid not cover large towns. For both Ethiopia and Malawi, the employment rate was estimated for the population aged 15 to\n49 because the DHS do not interview household members over 50 years of age.\n\n\n32\n\nMention: \"LSMS‐ISA\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"One of the three household instruments whose rural employment rates are compared.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000358:33:0", "occurrence_id": "afce677fd5c1645672beab0c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 7. Appendix\n\n\nFigure A shows the (rural) employment rate, disaggregated by sex, in Ethiopia and Malawi\naccording to three types of household surveys reviewed in this paper (DHS, LSMS‐ISA, LFS) as\nwell as the most recent census. It illustrates that survey design matters. The rural employment\nrate in Ethiopia (panel 1) differs considerably depending on the type of survey used to estimate\nit, whereas the employment rate in Malawi is relatively similar across surveys (panel 2). Besides\ndifferences in levels, trends also diverge across surveys. The Ethiopia LSMS‐ISA surveys, for\ninstance, suggest a decreasing rural employment rate, whereas DHS points to the other\ndirection. What explains the differences across surveys is hard to tell, but candidate\nexplanations include differences in (1) the definition of employment, (2) questionnaire design\nincluding the structure of the questionnaire and the phrasing of questions, and (3)\nimplementation in the field such as the number and timing of visits and the training of\nenumerators.\n\n\nFigure A: Different surveys, different results?\n\n1. Ethiopia’s rural employment rate 2. Malawi’s employment rate\n(population aged 15 to 49) (population aged 15 to 49)\n\n\nSource: Based on 2015/16 Ethiopia and 2016/17 Malawi LSMS‐ISA surveys complemented with aggregated statistics from\nDHS, LFS and census data. More detailed results are available upon request.\nNote: We attempted to make the population as comparable as possible across surveys. We calculated the ‘rural’ employment\nrate in Ethiopia (rather than the ‘total’ employment rate as in Malawi) because the first wave of the Ethiopia LSMS‐ISA survey\ndid not cover large towns. For both Ethiopia and Malawi, the employment rate was estimated for the population aged 15 to\n49 because the DHS do not interview household members over 50 years of age.\n\n\n32\n\nMention: \"LFS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its labour-specific questions produce employment rates differing from the general household surveys.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000358:33:0", "occurrence_id": "4b36e11fd0a106ebd1ec6aca", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 7. Appendix\n\n\nFigure A shows the (rural) employment rate, disaggregated by sex, in Ethiopia and Malawi\naccording to three types of household surveys reviewed in this paper (DHS, LSMS‐ISA, LFS) as\nwell as the most recent census. It illustrates that survey design matters. The rural employment\nrate in Ethiopia (panel 1) differs considerably depending on the type of survey used to estimate\nit, whereas the employment rate in Malawi is relatively similar across surveys (panel 2). Besides\ndifferences in levels, trends also diverge across surveys. The Ethiopia LSMS‐ISA surveys, for\ninstance, suggest a decreasing rural employment rate, whereas DHS points to the other\ndirection. What explains the differences across surveys is hard to tell, but candidate\nexplanations include differences in (1) the definition of employment, (2) questionnaire design\nincluding the structure of the questionnaire and the phrasing of questions, and (3)\nimplementation in the field such as the number and timing of visits and the training of\nenumerators.\n\n\nFigure A: Different surveys, different results?\n\n1. Ethiopia’s rural employment rate 2. Malawi’s employment rate\n(population aged 15 to 49) (population aged 15 to 49)\n\n\nSource: Based on 2015/16 Ethiopia and 2016/17 Malawi LSMS‐ISA surveys complemented with aggregated statistics from\nDHS, LFS and census data. More detailed results are available upon request.\nNote: We attempted to make the population as comparable as possible across surveys. We calculated the ‘rural’ employment\nrate in Ethiopia (rather than the ‘total’ employment rate as in Malawi) because the first wave of the Ethiopia LSMS‐ISA survey\ndid not cover large towns. For both Ethiopia and Malawi, the employment rate was estimated for the population aged 15 to\n49 because the DHS do not interview household members over 50 years of age.\n\n\n32\n\nMention: \"Ethiopia LSMS‐ISA surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"These rounds imply a falling rural employment rate, opposite to the direction other surveys show.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000358:33:0", "occurrence_id": "c5af0ab2384e82977d3e94b7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 7. Appendix\n\n\nFigure A shows the (rural) employment rate, disaggregated by sex, in Ethiopia and Malawi\naccording to three types of household surveys reviewed in this paper (DHS, LSMS‐ISA, LFS) as\nwell as the most recent census. It illustrates that survey design matters. The rural employment\nrate in Ethiopia (panel 1) differs considerably depending on the type of survey used to estimate\nit, whereas the employment rate in Malawi is relatively similar across surveys (panel 2). Besides\ndifferences in levels, trends also diverge across surveys. The Ethiopia LSMS‐ISA surveys, for\ninstance, suggest a decreasing rural employment rate, whereas DHS points to the other\ndirection. What explains the differences across surveys is hard to tell, but candidate\nexplanations include differences in (1) the definition of employment, (2) questionnaire design\nincluding the structure of the questionnaire and the phrasing of questions, and (3)\nimplementation in the field such as the number and timing of visits and the training of\nenumerators.\n\n\nFigure A: Different surveys, different results?\n\n1. Ethiopia’s rural employment rate 2. Malawi’s employment rate\n(population aged 15 to 49) (population aged 15 to 49)\n\n\nSource: Based on 2015/16 Ethiopia and 2016/17 Malawi LSMS‐ISA surveys complemented with aggregated statistics from\nDHS, LFS and census data. More detailed results are available upon request.\nNote: We attempted to make the population as comparable as possible across surveys. We calculated the ‘rural’ employment\nrate in Ethiopia (rather than the ‘total’ employment rate as in Malawi) because the first wave of the Ethiopia LSMS‐ISA survey\ndid not cover large towns. For both Ethiopia and Malawi, the employment rate was estimated for the population aged 15 to\n49 because the DHS do not interview household members over 50 years of age.\n\n\n32\n\nMention: \"LSMS‐ISA surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The figure's base observations come from these two country rounds, supplemented by other sources.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000358:33:0", "occurrence_id": "9638ab2731155d7861fa4c42", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 7. Appendix\n\n\nFigure A shows the (rural) employment rate, disaggregated by sex, in Ethiopia and Malawi\naccording to three types of household surveys reviewed in this paper (DHS, LSMS‐ISA, LFS) as\nwell as the most recent census. It illustrates that survey design matters. The rural employment\nrate in Ethiopia (panel 1) differs considerably depending on the type of survey used to estimate\nit, whereas the employment rate in Malawi is relatively similar across surveys (panel 2). Besides\ndifferences in levels, trends also diverge across surveys. The Ethiopia LSMS‐ISA surveys, for\ninstance, suggest a decreasing rural employment rate, whereas DHS points to the other\ndirection. What explains the differences across surveys is hard to tell, but candidate\nexplanations include differences in (1) the definition of employment, (2) questionnaire design\nincluding the structure of the questionnaire and the phrasing of questions, and (3)\nimplementation in the field such as the number and timing of visits and the training of\nenumerators.\n\n\nFigure A: Different surveys, different results?\n\n1. Ethiopia’s rural employment rate 2. Malawi’s employment rate\n(population aged 15 to 49) (population aged 15 to 49)\n\n\nSource: Based on 2015/16 Ethiopia and 2016/17 Malawi LSMS‐ISA surveys complemented with aggregated statistics from\nDHS, LFS and census data. More detailed results are available upon request.\nNote: We attempted to make the population as comparable as possible across surveys. We calculated the ‘rural’ employment\nrate in Ethiopia (rather than the ‘total’ employment rate as in Malawi) because the first wave of the Ethiopia LSMS‐ISA survey\ndid not cover large towns. For both Ethiopia and Malawi, the employment rate was estimated for the population aged 15 to\n49 because the DHS do not interview household members over 50 years of age.\n\n\n32\n\nMention: \"Ethiopia LSMS‐ISA survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Because its first wave omitted large towns, a rural rather than total rate is reported for Ethiopia.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000384:11:0", "occurrence_id": "268565f2e1144e5c2c6fb80c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: terns of participation in the two export finance support schemes we evaluate.\n\n\nWe use two data sets in this paper: (i) customs data, and (ii) information provided by\n\n\nSBP on the firms participating in EFS and LTFF. Customs data provided by the Federal\n\n\nBoard of Revenue contain the universe of export and import transactions for firms in Pakistan\n\n\nover the period 2015-2017. These data have information on the value of firms’ exports and\n\n\nimports by product at the HS 8-digit level denominated in Pakistani rupees as well as the\n\n\ncountry of origin and destination of trade flows. Throughout this period there are 20,052\n\n\nfirms reporting at least one positive export transaction in at least one of 2,844 HS 8-digit\n\n\nproducts sold to 202 countries. The data on the usage of export finance support schemes\n\n\ninclude information on which firms made use of the EFS and LTFF financing facilities and\n\n\nthe value of the loans they requested from commercial banks in each fiscal year. <sup>6</sup> The data\n\n\nare linked at the firm level using National Tax Numbers (NTN).\n\n\nTable 1 provides a first pass at the export sector characteristics in our data. The number\n\n\nof active exporters remains stable throughout our period of study, with around 14,500 firms\n\n\nexporting each year on average. On average, Pakistani firms export five HS 8-digit products\n\n\nand serve four foreign markets. Export sales per firm are highly skewed towards high values,\n\n\nwhich is reflected in the large difference between the mean and median exports per exporter.\n\n\nThese figures are closely in line with those reported by Fernandes et al. (2016) for developing\n\n\ncountries of a similar size.\n\n\n6We only observe the total value of loans obtained by a given firm through each scheme. Therefore,\nour data do not allow us to distinguish if loans are provided for some export transactions and not others.\nThe data for EFS do not provide information about which variant(s) (Type I or\n\nMention: \"customs data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"One of two datasets, providing the universe of firm export and import transactions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000384:11:0", "occurrence_id": "ce36414a2105db4e365c22ac", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: terns of participation in the two export finance support schemes we evaluate.\n\n\nWe use two data sets in this paper: (i) customs data, and (ii) information provided by\n\n\nSBP on the firms participating in EFS and LTFF. Customs data provided by the Federal\n\n\nBoard of Revenue contain the universe of export and import transactions for firms in Pakistan\n\n\nover the period 2015-2017. These data have information on the value of firms’ exports and\n\n\nimports by product at the HS 8-digit level denominated in Pakistani rupees as well as the\n\n\ncountry of origin and destination of trade flows. Throughout this period there are 20,052\n\n\nfirms reporting at least one positive export transaction in at least one of 2,844 HS 8-digit\n\n\nproducts sold to 202 countries. The data on the usage of export finance support schemes\n\n\ninclude information on which firms made use of the EFS and LTFF financing facilities and\n\n\nthe value of the loans they requested from commercial banks in each fiscal year. <sup>6</sup> The data\n\n\nare linked at the firm level using National Tax Numbers (NTN).\n\n\nTable 1 provides a first pass at the export sector characteristics in our data. The number\n\n\nof active exporters remains stable throughout our period of study, with around 14,500 firms\n\n\nexporting each year on average. On average, Pakistani firms export five HS 8-digit products\n\n\nand serve four foreign markets. Export sales per firm are highly skewed towards high values,\n\n\nwhich is reflected in the large difference between the mean and median exports per exporter.\n\n\nThese figures are closely in line with those reported by Fernandes et al. (2016) for developing\n\n\ncountries of a similar size.\n\n\n6We only observe the total value of loans obtained by a given firm through each scheme. Therefore,\nour data do not allow us to distinguish if loans are provided for some export transactions and not others.\nThe data for EFS do not provide information about which variant(s) (Type I or\n\nMention: \"information provided by\n\n\nSBP\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Scheme participation records identify which firms used the two export finance facilities.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000384:11:0", "occurrence_id": "b0e6747bbb9b2ecaeca9c236", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: terns of participation in the two export finance support schemes we evaluate.\n\n\nWe use two data sets in this paper: (i) customs data, and (ii) information provided by\n\n\nSBP on the firms participating in EFS and LTFF. Customs data provided by the Federal\n\n\nBoard of Revenue contain the universe of export and import transactions for firms in Pakistan\n\n\nover the period 2015-2017. These data have information on the value of firms’ exports and\n\n\nimports by product at the HS 8-digit level denominated in Pakistani rupees as well as the\n\n\ncountry of origin and destination of trade flows. Throughout this period there are 20,052\n\n\nfirms reporting at least one positive export transaction in at least one of 2,844 HS 8-digit\n\n\nproducts sold to 202 countries. The data on the usage of export finance support schemes\n\n\ninclude information on which firms made use of the EFS and LTFF financing facilities and\n\n\nthe value of the loans they requested from commercial banks in each fiscal year. <sup>6</sup> The data\n\n\nare linked at the firm level using National Tax Numbers (NTN).\n\n\nTable 1 provides a first pass at the export sector characteristics in our data. The number\n\n\nof active exporters remains stable throughout our period of study, with around 14,500 firms\n\n\nexporting each year on average. On average, Pakistani firms export five HS 8-digit products\n\n\nand serve four foreign markets. Export sales per firm are highly skewed towards high values,\n\n\nwhich is reflected in the large difference between the mean and median exports per exporter.\n\n\nThese figures are closely in line with those reported by Fernandes et al. (2016) for developing\n\n\ncountries of a similar size.\n\n\n6We only observe the total value of loans obtained by a given firm through each scheme. Therefore,\nour data do not allow us to distinguish if loans are provided for some export transactions and not others.\nThe data for EFS do not provide information about which variant(s) (Type I or\n\nMention: \"Customs data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Export values by product and destination are used to build the firm-level trade panel.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000384:11:0", "occurrence_id": "4ec64d5c51152677148a616e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: terns of participation in the two export finance support schemes we evaluate.\n\n\nWe use two data sets in this paper: (i) customs data, and (ii) information provided by\n\n\nSBP on the firms participating in EFS and LTFF. Customs data provided by the Federal\n\n\nBoard of Revenue contain the universe of export and import transactions for firms in Pakistan\n\n\nover the period 2015-2017. These data have information on the value of firms’ exports and\n\n\nimports by product at the HS 8-digit level denominated in Pakistani rupees as well as the\n\n\ncountry of origin and destination of trade flows. Throughout this period there are 20,052\n\n\nfirms reporting at least one positive export transaction in at least one of 2,844 HS 8-digit\n\n\nproducts sold to 202 countries. The data on the usage of export finance support schemes\n\n\ninclude information on which firms made use of the EFS and LTFF financing facilities and\n\n\nthe value of the loans they requested from commercial banks in each fiscal year. <sup>6</sup> The data\n\n\nare linked at the firm level using National Tax Numbers (NTN).\n\n\nTable 1 provides a first pass at the export sector characteristics in our data. The number\n\n\nof active exporters remains stable throughout our period of study, with around 14,500 firms\n\n\nexporting each year on average. On average, Pakistani firms export five HS 8-digit products\n\n\nand serve four foreign markets. Export sales per firm are highly skewed towards high values,\n\n\nwhich is reflected in the large difference between the mean and median exports per exporter.\n\n\nThese figures are closely in line with those reported by Fernandes et al. (2016) for developing\n\n\ncountries of a similar size.\n\n\n6We only observe the total value of loans obtained by a given firm through each scheme. Therefore,\nour data do not allow us to distinguish if loans are provided for some export transactions and not others.\nThe data for EFS do not provide information about which variant(s) (Type I or\n\nMention: \"data on the usage of export finance support schemes\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Requested loan values per fiscal year are linked to customs records at firm level.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000386:15:0", "occurrence_id": "98132692ab6e1503d60d05bb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 7: Years of education, by gender and location**\n\n\nNote: 1990s cohort refer to individuals born between 1990 and 1995. Average mean years of education by decade-cohort\nof individual’s year of birth; individuals in 1990 cohort may not have fully completed education at the time of survey (2015),\nand hence these individuals could attain higher mean years of education.\n\n##### 4.2.Defining Trade Policy Scenarios 21 4.2.1. Baseline\n\n\n**As mentioned above, an ex-ante evaluation requires the creation of a counterfactual simulation that**\n**will serve as a business-as-usual scenario with no policy change.** The model relies on the GTAP version 9\ndatabase benchmarked to 2011. The CGE model runs up to 2028, replicating the key historical\nmacroeconomic aggregates from the World Bank World Development Indicators (World Bank, n.d.).\nProductivity growth in the baseline scenario is calibrated to achieve the historical and forecasted GDP\ngrowth rates up to 2021. Productivity growth is adjusted thereafter to be consistent with historical trends. <sup>22</sup> [^22: These productivity assumptions are not altered in in the counterfactual scenarios.]\nThe baseline scenario also incorporates tariff reductions under existing FTAs. These are based on the data\nset provided by the International Trade Center, including all TPP members’ FTA commitments up to 2027\n(International Trade Centre 2016). We also assume that Sri Lanka enjoys the Generalized System of\nPreferences (GSP) preferences on its exports to EU markets.\n\n**The baseline contemplates continuation of recent trends using standard closure rules for**\n**macroeconomic accounting.** The scenarios incorporate three closure rules. First, government expenditures\nare held constant as a share of GDP, fiscal balance is exogenous while direct taxes adjust to cover any\n\n\n21 This section was largely was developed by Guillermo Arenas (Economist, World Bank Group) using the Tariff Reform Impact\nSimulations Tool.\n\n22 These\n\nMention: \"GTAP version 9\ndatabase\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The general equilibrium model is calibrated to this benchmark database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000386:15:0", "occurrence_id": "4acb6da589b970bec1aa99c9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 7: Years of education, by gender and location**\n\n\nNote: 1990s cohort refer to individuals born between 1990 and 1995. Average mean years of education by decade-cohort\nof individual’s year of birth; individuals in 1990 cohort may not have fully completed education at the time of survey (2015),\nand hence these individuals could attain higher mean years of education.\n\n##### 4.2.Defining Trade Policy Scenarios 21 4.2.1. Baseline\n\n\n**As mentioned above, an ex-ante evaluation requires the creation of a counterfactual simulation that**\n**will serve as a business-as-usual scenario with no policy change.** The model relies on the GTAP version 9\ndatabase benchmarked to 2011. The CGE model runs up to 2028, replicating the key historical\nmacroeconomic aggregates from the World Bank World Development Indicators (World Bank, n.d.).\nProductivity growth in the baseline scenario is calibrated to achieve the historical and forecasted GDP\ngrowth rates up to 2021. Productivity growth is adjusted thereafter to be consistent with historical trends. <sup>22</sup> [^22: These productivity assumptions are not altered in in the counterfactual scenarios.]\nThe baseline scenario also incorporates tariff reductions under existing FTAs. These are based on the data\nset provided by the International Trade Center, including all TPP members’ FTA commitments up to 2027\n(International Trade Centre 2016). We also assume that Sri Lanka enjoys the Generalized System of\nPreferences (GSP) preferences on its exports to EU markets.\n\n**The baseline contemplates continuation of recent trends using standard closure rules for**\n**macroeconomic accounting.** The scenarios incorporate three closure rules. First, government expenditures\nare held constant as a share of GDP, fiscal balance is exogenous while direct taxes adjust to cover any\n\n\n21 This section was largely was developed by Guillermo Arenas (Economist, World Bank Group) using the Tariff Reform Impact\nSimulations Tool.\n\n22 These\n\nMention: \"World Bank World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Historical macroeconomic aggregates replicated in the baseline come from this series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000387:15:0", "occurrence_id": "7da945d5189a9a007042183f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Temporary employment and part-time work were relatively common in Kosovo.** Most employees in\nKosovo had temporary contracts (74.8 percent in 2018) and only 4.5 percent had a part-time job. Firms were\nconcentrated in the commerce and services sectors. The share of public sector employment constituted 41.6\npercent and the share of private sector employment was 58.4 percent.\n\n\nRecent trends in wages\n\n\n**There is a public-sector wage premium, and real wages in the public sector increased faster than in**\n**the private sector; gender, age and education wage disparities were also common.** In 2018, the average\nnet monthly wage in Kosovo was approximately 393 euros, constituting 457.3 euros in the public sector and\njust 344 euros in the private sector. Nevertheless, because the average wage may be overestimated or\nunderestimated due to the presence of outliers, the median wage tends to be a more useful comparison. Median\nwages were lower than mean averages, set at 378 euros in 2018. Median wages grew faster in the public sector\nthan in the private sector, as shown in Figure 7 Panel a and can be expected to grow even faster with approval\nof the new wage law for public sector employees in early 2019. <sup>17</sup> [^17: Average and median wage are estimated using Kosovo LFS and wages are deflated using CPI with 2015 as a base year.\n[Series for CPI are obtained from Kosovo Agency of Statistics estimates (http://ask.rks-gov.net/en/kosovo-agency-of-](http://ask.rks-gov.net/en/kosovo-agency-of-statistics/add-news/harmonized-index-of-consumer-prices-hicp-april-2018)\n<u>[statistics/add-news/harmonized-index-of-consumer-prices-hicp-april-2018](http://ask.rks-gov.net/en/kosovo-agency-of-statistics/add-news/harmonized-index-of-consumer-prices-hicp-april-2018)</u> ).] When using median wages, there is still a\nsizable public-sector premium, especially among workers with higher educational attainment (Figure 7 Panel\nb)\n\nMention: \"Kosovo LFS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Average and median wages are estimated from this survey and deflated to a 2015 base.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000388:10:0", "occurrence_id": "4bbc85931c6fb6d6946d0bf2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: creating the best possible mask for our period of analysis. And for 2015, the two approaches result\n\n\nin very similar light data. Clusters are identified by removing outlier observations, averaging\n\n\ncells over time, and clustering areas based on their nighttime light intensity. In practice, this\n\n\napproach amounts to setting to zero cells that are distant from homogenous bright cores. We\n\n\npresent results based on data cleaned with the second method, but our results are robust to both\n\n\ncleaning methods. For this paper, cleaned monthly data are aggregated to the district and city\n\n\nlevels and standardized by area. Nighttime light is measured in Nanowatts/cm2 /steradian.\n\n\n**3.3** **Other** **variables**\n\n\nData on quarterly economic activity measured as gross value added (GVA) is from the Central\n\n\nStatistics Office. Temperature data is collected from the Average Daily Temperature Archive\n\n\nfrom the University of Dayton (2020) and based on recordings of the National Climatic Data\n\n\nCenter. We generate an India aggregate by weighting the daily temperatures recorded in Chennai,\n\nDelhi, Kolkata, and Mumbai by population. <sup>11</sup> Data on Indian holidays is from different online\n\n\nsources. The information about registered COVID-19 infections at the state and district level\n\n\nis from Covindia (2020), the data on previous short-run inmigration and outmigration at the\n\n\nstate and district level from the Development Data Lab (2020), and the share of employment in\n\n\nmanufacturing and services at the state and district level is from the South Asia Spatial Database\n\n\n(Li, Rama, Galdo, and Pinto 2015).\n\n##### **4 Electricity, nighttime lights, and economic activity**\n\n\nIntuitively, electricity consumption and economic activity are closely related since most economic\n\n\nactivity needs electricity. Plenty of studies analyze the relationship of the two over longer periods,\n\n\noften with a focus on the direction of causality. Chen, Kuo\n\nMention: \"Average Daily Temperature Archive\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Daily temperatures for four cities are population-weighted into a national temperature series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000388:10:0", "occurrence_id": "df89348fb88433fca01de0a0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: creating the best possible mask for our period of analysis. And for 2015, the two approaches result\n\n\nin very similar light data. Clusters are identified by removing outlier observations, averaging\n\n\ncells over time, and clustering areas based on their nighttime light intensity. In practice, this\n\n\napproach amounts to setting to zero cells that are distant from homogenous bright cores. We\n\n\npresent results based on data cleaned with the second method, but our results are robust to both\n\n\ncleaning methods. For this paper, cleaned monthly data are aggregated to the district and city\n\n\nlevels and standardized by area. Nighttime light is measured in Nanowatts/cm2 /steradian.\n\n\n**3.3** **Other** **variables**\n\n\nData on quarterly economic activity measured as gross value added (GVA) is from the Central\n\n\nStatistics Office. Temperature data is collected from the Average Daily Temperature Archive\n\n\nfrom the University of Dayton (2020) and based on recordings of the National Climatic Data\n\n\nCenter. We generate an India aggregate by weighting the daily temperatures recorded in Chennai,\n\nDelhi, Kolkata, and Mumbai by population. <sup>11</sup> Data on Indian holidays is from different online\n\n\nsources. The information about registered COVID-19 infections at the state and district level\n\n\nis from Covindia (2020), the data on previous short-run inmigration and outmigration at the\n\n\nstate and district level from the Development Data Lab (2020), and the share of employment in\n\n\nmanufacturing and services at the state and district level is from the South Asia Spatial Database\n\n\n(Li, Rama, Galdo, and Pinto 2015).\n\n##### **4 Electricity, nighttime lights, and economic activity**\n\n\nIntuitively, electricity consumption and economic activity are closely related since most economic\n\n\nactivity needs electricity. Plenty of studies analyze the relationship of the two over longer periods,\n\n\noften with a focus on the direction of causality. Chen, Kuo\n\nMention: \"National Climatic Data\n\n\nCenter\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its station recordings are the underlying observations behind the temperature archive.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000388:10:0", "occurrence_id": "66e6271d9e42880cc58b1ba3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: creating the best possible mask for our period of analysis. And for 2015, the two approaches result\n\n\nin very similar light data. Clusters are identified by removing outlier observations, averaging\n\n\ncells over time, and clustering areas based on their nighttime light intensity. In practice, this\n\n\napproach amounts to setting to zero cells that are distant from homogenous bright cores. We\n\n\npresent results based on data cleaned with the second method, but our results are robust to both\n\n\ncleaning methods. For this paper, cleaned monthly data are aggregated to the district and city\n\n\nlevels and standardized by area. Nighttime light is measured in Nanowatts/cm2 /steradian.\n\n\n**3.3** **Other** **variables**\n\n\nData on quarterly economic activity measured as gross value added (GVA) is from the Central\n\n\nStatistics Office. Temperature data is collected from the Average Daily Temperature Archive\n\n\nfrom the University of Dayton (2020) and based on recordings of the National Climatic Data\n\n\nCenter. We generate an India aggregate by weighting the daily temperatures recorded in Chennai,\n\nDelhi, Kolkata, and Mumbai by population. <sup>11</sup> Data on Indian holidays is from different online\n\n\nsources. The information about registered COVID-19 infections at the state and district level\n\n\nis from Covindia (2020), the data on previous short-run inmigration and outmigration at the\n\n\nstate and district level from the Development Data Lab (2020), and the share of employment in\n\n\nmanufacturing and services at the state and district level is from the South Asia Spatial Database\n\n\n(Li, Rama, Galdo, and Pinto 2015).\n\n##### **4 Electricity, nighttime lights, and economic activity**\n\n\nIntuitively, electricity consumption and economic activity are closely related since most economic\n\n\nactivity needs electricity. Plenty of studies analyze the relationship of the two over longer periods,\n\n\noften with a focus on the direction of causality. Chen, Kuo\n\nMention: \"South Asia Spatial Database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"State and district employment shares in manufacturing and services come from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000392:1:0", "occurrence_id": "99524c6276cce387176f3a14", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Policy Research Working Paper 9344\n\n##### **Abstract**\n\nThe expansionary fiscal contraction (EFC) hypothesis states\n\nthat fiscal austerity can increase output or consumption\nwhen a country is under heavy debt burdens because it\nsends positive signal about the country’s solvency situation and long-term economic wellbeing. Empirical tests\nof this hypothesis have suffered from identification concerns due to data sources and empirical methodology.\nUsing a sample of OECD countries between 1978 and\n2014, this paper combines new IMF narrative data and the\nproxy structural Vector Auto-regression (SVAR) method\n\n\n\nto examine whether fiscal austerities can be expansionary\nwhen debt levels are high. Fiscal austerities are measured\nas 1) narrative fiscal shocks and 2) structural shocks from\na proxy SVAR. Additionally, this paper uses a model-based\napproach to determine the cutoff debt level beyond which\nEFC is expected to be observed. This paper finds empirical evidence in support of the EFC hypothesis for OECD\ncountries: results for output are driven by changes in tax\nrates and are robust to how one defines a high-debt regime\nand how one measures austerity.\n\n\n\nThis paper is a product of the Finance, Competitiveness and Innovation Global Practice. It is part of a larger effort by the\nWorld Bank to provide open access to its research and make a contribution to development policy discussions around the\n\nworld. Policy Research Working Papers are also posted on the Web at http://www.worldbank.org/prwp. The author may\nbe contacted at onie@worldbank.org.\n\n\n_The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development_\n_issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the_\n_names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are\n\nMention: \"IMF narrative data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Narrative fiscal shocks identified from this source are used to test the expansionary contraction hypothesis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000396:9:0", "occurrence_id": "02fb8d131852661b37f920c4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: All the analyses for cities in the report were run with the following identical variables:\n\n\n  - 4km/h walking speed for pedestrians accessing transit\n\n  - 7AM to 9AM on a weekday – peak travel time\n\n  - Within 60 minutes travel time between departure and arrival\n\nWhere GTFS feeds with specific stops were unavailable, 400 meters was assumed as\nthe distance between stops.\n\n\nData produced by WhereIsMyTransport <sup>7</sup> [^7: https://www.whereismytransport.com/about/] were used for five cities – Lusaka, Kampala, Harare,\nDouala and Cape Town. WhereIsMyTtransport is a South African company which collects and\ndigitizes data from formal and informal public transport networks to produce GTFS feeds. Data\ncollected include information on fares, stops, routes, frequencies and travel times and speeds.\nThis mapping is carried out with local partners and identification of all available modes of\ntransport within the city and local teams of data collectors. Routes and data are validated in the\nfield by the local team during daily feedback sessions with data collectors.\n\nThe transport data for Nairobi are based on the Digital Matatus Project. This was carried out in\n2012-2014 by the DigitalMatatus consortium comprising the Universities of Nairobi,\nColumbia, MIT and Groupshot, to produce a map of the Matatu minibus network in Nairobi.\nUniversity of Nairobi students collected information on bus routes, schedules and stops with\nGPS technology based on their phones. Students and the Civic Data Design lab at MIT\nanalyzed the data, and created a public transit map and a GTFS feed (Williams et al. 2015;\nKlopp et al. 2015).\n\n\nThe transport data for Dar es Salaam were a combination of sources, collected and aggregated\nby the World Bank office. The transport system includes the DalaDala routes, mapped as an\ninitiative of the Ramani Huria team, the Tanzania Open Data Initiative (TODI) team from the\nWorld Bank, and Ally. <sup>8</sup>\n\nMention: \"GTFS feeds\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Scheduled transit stops are taken from these feeds, with a 400 metre spacing assumed when they are missing.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000396:9:0", "occurrence_id": "f4b35f604bcb843ba0c4ee5d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: All the analyses for cities in the report were run with the following identical variables:\n\n\n  - 4km/h walking speed for pedestrians accessing transit\n\n  - 7AM to 9AM on a weekday – peak travel time\n\n  - Within 60 minutes travel time between departure and arrival\n\nWhere GTFS feeds with specific stops were unavailable, 400 meters was assumed as\nthe distance between stops.\n\n\nData produced by WhereIsMyTransport <sup>7</sup> [^7: https://www.whereismytransport.com/about/] were used for five cities – Lusaka, Kampala, Harare,\nDouala and Cape Town. WhereIsMyTtransport is a South African company which collects and\ndigitizes data from formal and informal public transport networks to produce GTFS feeds. Data\ncollected include information on fares, stops, routes, frequencies and travel times and speeds.\nThis mapping is carried out with local partners and identification of all available modes of\ntransport within the city and local teams of data collectors. Routes and data are validated in the\nfield by the local team during daily feedback sessions with data collectors.\n\nThe transport data for Nairobi are based on the Digital Matatus Project. This was carried out in\n2012-2014 by the DigitalMatatus consortium comprising the Universities of Nairobi,\nColumbia, MIT and Groupshot, to produce a map of the Matatu minibus network in Nairobi.\nUniversity of Nairobi students collected information on bus routes, schedules and stops with\nGPS technology based on their phones. Students and the Civic Data Design lab at MIT\nanalyzed the data, and created a public transit map and a GTFS feed (Williams et al. 2015;\nKlopp et al. 2015).\n\n\nThe transport data for Dar es Salaam were a combination of sources, collected and aggregated\nby the World Bank office. The transport system includes the DalaDala routes, mapped as an\ninitiative of the Ramani Huria team, the Tanzania Open Data Initiative (TODI) team from the\nWorld Bank, and Ally. <sup>8</sup>\n\nMention: \"Data produced by WhereIsMyTransport\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Digitised formal and informal routes for five cities come from this commercial provider.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000396:9:0", "occurrence_id": "e730ea5fffabfa7d7707dcc8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: All the analyses for cities in the report were run with the following identical variables:\n\n\n  - 4km/h walking speed for pedestrians accessing transit\n\n  - 7AM to 9AM on a weekday – peak travel time\n\n  - Within 60 minutes travel time between departure and arrival\n\nWhere GTFS feeds with specific stops were unavailable, 400 meters was assumed as\nthe distance between stops.\n\n\nData produced by WhereIsMyTransport <sup>7</sup> [^7: https://www.whereismytransport.com/about/] were used for five cities – Lusaka, Kampala, Harare,\nDouala and Cape Town. WhereIsMyTtransport is a South African company which collects and\ndigitizes data from formal and informal public transport networks to produce GTFS feeds. Data\ncollected include information on fares, stops, routes, frequencies and travel times and speeds.\nThis mapping is carried out with local partners and identification of all available modes of\ntransport within the city and local teams of data collectors. Routes and data are validated in the\nfield by the local team during daily feedback sessions with data collectors.\n\nThe transport data for Nairobi are based on the Digital Matatus Project. This was carried out in\n2012-2014 by the DigitalMatatus consortium comprising the Universities of Nairobi,\nColumbia, MIT and Groupshot, to produce a map of the Matatu minibus network in Nairobi.\nUniversity of Nairobi students collected information on bus routes, schedules and stops with\nGPS technology based on their phones. Students and the Civic Data Design lab at MIT\nanalyzed the data, and created a public transit map and a GTFS feed (Williams et al. 2015;\nKlopp et al. 2015).\n\n\nThe transport data for Dar es Salaam were a combination of sources, collected and aggregated\nby the World Bank office. The transport system includes the DalaDala routes, mapped as an\ninitiative of the Ramani Huria team, the Tanzania Open Data Initiative (TODI) team from the\nWorld Bank, and Ally. <sup>8</sup>\n\nMention: \"Digital Matatus Project\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Minibus routes, schedules and stops collected by students underpin the Nairobi analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000396:9:0", "occurrence_id": "1f43fc6a0b620d1f21604023", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: All the analyses for cities in the report were run with the following identical variables:\n\n\n  - 4km/h walking speed for pedestrians accessing transit\n\n  - 7AM to 9AM on a weekday – peak travel time\n\n  - Within 60 minutes travel time between departure and arrival\n\nWhere GTFS feeds with specific stops were unavailable, 400 meters was assumed as\nthe distance between stops.\n\n\nData produced by WhereIsMyTransport <sup>7</sup> [^7: https://www.whereismytransport.com/about/] were used for five cities – Lusaka, Kampala, Harare,\nDouala and Cape Town. WhereIsMyTtransport is a South African company which collects and\ndigitizes data from formal and informal public transport networks to produce GTFS feeds. Data\ncollected include information on fares, stops, routes, frequencies and travel times and speeds.\nThis mapping is carried out with local partners and identification of all available modes of\ntransport within the city and local teams of data collectors. Routes and data are validated in the\nfield by the local team during daily feedback sessions with data collectors.\n\nThe transport data for Nairobi are based on the Digital Matatus Project. This was carried out in\n2012-2014 by the DigitalMatatus consortium comprising the Universities of Nairobi,\nColumbia, MIT and Groupshot, to produce a map of the Matatu minibus network in Nairobi.\nUniversity of Nairobi students collected information on bus routes, schedules and stops with\nGPS technology based on their phones. Students and the Civic Data Design lab at MIT\nanalyzed the data, and created a public transit map and a GTFS feed (Williams et al. 2015;\nKlopp et al. 2015).\n\n\nThe transport data for Dar es Salaam were a combination of sources, collected and aggregated\nby the World Bank office. The transport system includes the DalaDala routes, mapped as an\ninitiative of the Ramani Huria team, the Tanzania Open Data Initiative (TODI) team from the\nWorld Bank, and Ally. <sup>8</sup>\n\nMention: \"GTFS feed\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The project's collected observations were published in this standard transit format.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000396:9:0", "occurrence_id": "bbc8c96857aba22b75576923", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: All the analyses for cities in the report were run with the following identical variables:\n\n\n  - 4km/h walking speed for pedestrians accessing transit\n\n  - 7AM to 9AM on a weekday – peak travel time\n\n  - Within 60 minutes travel time between departure and arrival\n\nWhere GTFS feeds with specific stops were unavailable, 400 meters was assumed as\nthe distance between stops.\n\n\nData produced by WhereIsMyTransport <sup>7</sup> [^7: https://www.whereismytransport.com/about/] were used for five cities – Lusaka, Kampala, Harare,\nDouala and Cape Town. WhereIsMyTtransport is a South African company which collects and\ndigitizes data from formal and informal public transport networks to produce GTFS feeds. Data\ncollected include information on fares, stops, routes, frequencies and travel times and speeds.\nThis mapping is carried out with local partners and identification of all available modes of\ntransport within the city and local teams of data collectors. Routes and data are validated in the\nfield by the local team during daily feedback sessions with data collectors.\n\nThe transport data for Nairobi are based on the Digital Matatus Project. This was carried out in\n2012-2014 by the DigitalMatatus consortium comprising the Universities of Nairobi,\nColumbia, MIT and Groupshot, to produce a map of the Matatu minibus network in Nairobi.\nUniversity of Nairobi students collected information on bus routes, schedules and stops with\nGPS technology based on their phones. Students and the Civic Data Design lab at MIT\nanalyzed the data, and created a public transit map and a GTFS feed (Williams et al. 2015;\nKlopp et al. 2015).\n\n\nThe transport data for Dar es Salaam were a combination of sources, collected and aggregated\nby the World Bank office. The transport system includes the DalaDala routes, mapped as an\ninitiative of the Ramani Huria team, the Tanzania Open Data Initiative (TODI) team from the\nWorld Bank, and Ally. <sup>8</sup>\n\nMention: \"transport data for Dar es Salaam\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"Several community mapping initiatives were combined and aggregated for this city.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000398:20:7", "occurrence_id": "d814678e99c940cc7f81a4b2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .0750) (0.0520) (0.0582) (0.0475)\n\n\n_Policy Response_\nGovernment Response (GR) -0.0526\n_(Oxford tracker)_ (0.0316)\nStringency Index -0.0505* -0.0502* -0.0539* -0.0531* -0.0556* -0.0509* -0.0571** -0.0788** -0.0570** -0.0686** -0.0470*\n_(Oxford tracker)_ (0.0291) (0.0271) (0.0280) (0.0276) (0.0300) (0.0265) (0.0266) (0.0302) (0.0265) (0.0296) (0.0274)\n\n\n_Regional Dummies_\nDummy: Sub-Saharan 2.287* 1.967 2.463* 2.208* 0.412 1.435 1.857 1.274 3.520** 3.104** 2.477* 2.029\nAfrica (SSA) (1.261) (1.377) (1.291) (1.256) (1.154) (1.403) (1.144) (1.154) (1.363) (1.285) (1.357) (1.497)\n\n\nObservations 88 79 88 88 93 81 93 93 78 89 79 88\n<u>R-squared</u> <u>0.39</u> <u>0.483</u> <u>0.440</u> <u>0.397</u> <u>0.358</u> <\n\nMention: \"Oxford tracker\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Government response and stringency variables in the regressions are attributed to this tracker.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000401:6:1", "occurrence_id": "2c8289bc085a2258e42da99e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: , DGA). The database covers the entire\n\n\nuniverse of transactions between 1994 and 2018 and contains unique and time-consistent firm\n\n\nidentifiers that allows us to merge export and import transactions at the firm level. We collapse\n\n\nexports and imports at the firm-year level, preserving the following variables: exported/imported\n\n\n4\n\nMention: \"DGA\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Its universe of transactions with consistent firm identifiers allows exports and imports to be merged.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000406:3:0", "occurrence_id": "e1189663b88229a90eaf7c9f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **1.** **Introduction**\n\n\nImproving the productivity of smallholder farmers is widely considered one of the most effective\navenues for reducing poverty and food insecurity, and thus has been a longstanding goal in many\nAfrican countries (Byerlee et al., 2007). The evidence concerning (i) agriculture contributing up\nto 69 percent of rural household income in Africa (Davis et al., 2017), and (ii) higher rates of\nexpected poverty reduction associated with agricultural vis‐à‐vis nonagricultural growth (Dorosh\nand Thurlow, 2016) helps sustain the policy focus on achieving this goal at the national level.\nSimilarly, at the international level, doubling productivity and incomes of smallholders have been\nidentified as a key target within the Sustainable Development Goal (SDG) #2 of Ending Hunger.\n\n\nAccurate measures of production and productivity are, therefore, essential to (i) tracking\nprogress towards the relevant SDG targets; answering fundamental questions on the role of\nagriculture in household and individual welfare (Darko et al., 2018); and understanding which\nproduction factors have the most important role in determining productivity. Ongoing debates\nabout the relationship between (land) productivity and (i) fertilizer use (Harou et al., 2017), (ii)\nplot/farm size (Bevis and Barrett, 2017; Desiere and Jolliffe, 2018; Gourlay et al., 2017), or (iii)\nsoil quality (Berazneva et al., 2018) reflect the substantial knowledge gaps and the need to\nimprove the accuracy and precision of recommendations for raising productivity in particular\nlocations.\n\n\nThe most common way to assess outcomes related to economic productivity of smallholder\nfarmers, including land productivity (e.g. crop yields), is by using information collected through\nin‐person interviews for household and farm surveys. For example, the household surveys\nsupported by the World Bank Living Standards Measurement Study – Integrated Surveys on\nAgriculture (\n\nMention: \"Integrated Surveys on\nAgriculture\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Cited as an example of household and farm surveys that collect productivity through in-person interviews.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000412:5:1", "occurrence_id": "91ffe19be70cbd3dfc567162", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ,\nvolumetric, load-based and time-of-use — have been fully incorporated (Box 1) (Briceno-Garmendia and\nShkaratan, 2011). This requires assumptions to be made regarding the average monthly electricity bill. For\nresidential customers, the average electricity consumption is derived from the IEA energy balances by\ntaking the total residential electricity consumption and dividing by the number of connected households.\nSince there are limited data on the distribution of electricity consumption within a given country, median\nelectricity consumption may lie below the lifeline tariff level in more cases than implied by the analysis.\nAs a result, the results might overestimate the average electricity tariff faced by households in certain\ndeveloping countries. For non-residential customers, a “representative” consumption level is adopted based\non expert consultations and the characteristics of each sector. Commercial, agricultural and public sector\nconsumption is assumed to be 1,000 kilowatt-hours per month, while industrial consumption is set at 5,000\nkilowatt-hours per month.\n\nSimilarly, assumptions were made about the representative load level for each customer group, as well as\nthe pattern of peak versus off-peak consumption in cases where time-of-use tariffs exist. Following internal\nconsultations, the commercial load factor was set to 11 kilovolts, while the industrial load factor was set to\n33 kilovolts. In cases where customer groups were not explicitly defined by the tariff schedule, these load\nfactors were used for categorization.\n\nIn order to apply time-of-use to a single average nominal tariff for each customer group, a weighted average\nvolumetric time-of-use charge was computed for each customer class based on the following time zones:\npeak times were defined as 6pm through 10pm for commercial customers, and as 6pm through 9pm for\nindustrial users. Day time was defined as 8am through 6pm, and nighttime as 10pm through 8am for all\nusers. Where a choice between linear tariffs and time\n\nMention: \"IEA energy balances\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Total residential consumption divided by connected households gives the average bill assumption.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000418:37:1", "occurrence_id": "9b0c8b4dacc241e5eb4d5e0c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 137) 0 1\nMuslims 7480 0.816 (0.387) 0 1\nCatholics 7480 0.081 (0.273) 0 1\nOrthodox 7480 0.067 (0.251) 0 1\nCoastal region 7480 0.264 (0.440) 0 1\nCentral region 7480 0.282 (0.450) 0 1\nMountain region 7480 0.291 (0.454) 0 1\nTirana region 7480 0.161 (0.367) 0 1\n\n\n_Notes:_ The data used are the 2002 LSMS, the 2005 LSMS, the 2008-09 DHS and the 2012 LSMS\n\n\n35\n\nMention: \"2002 LSMS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"One of four survey waves pooled to build the regression sample.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000418:37:1", "occurrence_id": "2153aebce8aba706ed4639d9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 137) 0 1\nMuslims 7480 0.816 (0.387) 0 1\nCatholics 7480 0.081 (0.273) 0 1\nOrthodox 7480 0.067 (0.251) 0 1\nCoastal region 7480 0.264 (0.440) 0 1\nCentral region 7480 0.282 (0.450) 0 1\nMountain region 7480 0.291 (0.454) 0 1\nTirana region 7480 0.161 (0.367) 0 1\n\n\n_Notes:_ The data used are the 2002 LSMS, the 2005 LSMS, the 2008-09 DHS and the 2012 LSMS\n\n\n35\n\nMention: \"2005 LSMS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"A second living standards wave contributing household observations to the pooled sample.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000418:37:1", "occurrence_id": "700f4bf449ca81cbd2b69777", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 137) 0 1\nMuslims 7480 0.816 (0.387) 0 1\nCatholics 7480 0.081 (0.273) 0 1\nOrthodox 7480 0.067 (0.251) 0 1\nCoastal region 7480 0.264 (0.440) 0 1\nCentral region 7480 0.282 (0.450) 0 1\nMountain region 7480 0.291 (0.454) 0 1\nTirana region 7480 0.161 (0.367) 0 1\n\n\n_Notes:_ The data used are the 2002 LSMS, the 2005 LSMS, the 2008-09 DHS and the 2012 LSMS\n\n\n35\n\nMention: \"2008-09 DHS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The demographic and health round is combined with the living standards surveys.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000418:37:1", "occurrence_id": "67d895f5d2caaa279d0b0871", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 137) 0 1\nMuslims 7480 0.816 (0.387) 0 1\nCatholics 7480 0.081 (0.273) 0 1\nOrthodox 7480 0.067 (0.251) 0 1\nCoastal region 7480 0.264 (0.440) 0 1\nCentral region 7480 0.282 (0.450) 0 1\nMountain region 7480 0.291 (0.454) 0 1\nTirana region 7480 0.161 (0.367) 0 1\n\n\n_Notes:_ The data used are the 2002 LSMS, the 2005 LSMS, the 2008-09 DHS and the 2012 LSMS\n\n\n35\n\nMention: \"2012 LSMS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The most recent living standards round in the pooled dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000420:36:1", "occurrence_id": "f38ca59e5bd0363865227213", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .015 0.025 0.011\n\n(0.017) (0.016) (0.017)\n\n2013 * Two or more 0.048*** 0.041** 0.048***\n\n(0.018) (0.017) (0.019)\n\n\nN 55,539 47,645 52,278\n\nControls for women’s characteristics <sup>a</sup> Yes Yes Yes\n\n<u>Controls for husband’s characteristics</u> <sup>b</sup> <u>No</u> <u>No</u> <u>Yes</u>\n\n<u>Note: The table reports average marginal effects from probit regressions run on an annual birth panel for the period 2003-2013</u>\ncreated with the DHS 2010 and DHS 2015 data. Each row reports the coefficients for year dummies interacted with the indicator\nfor having at least two children. All columns include year fixed effects as well as a dummy for having at least two children. The\nomitted category is 2008. Standard errors clustered at the mother level in parentheses below each estimate. Columns (1) and (3) are\nrun on the entire sample of ever married women, while column (2) restricts the sample to those who have at least one child.\na Control variables as in Table 3, column (1) are included but not reported.\nb Control variables as in Table 3, column (3) are included but not reported.\n*** Significant at 1%; ** significant at 5%; * significant at 10%.\n\nMention: \"DHS 2010\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"Birth histories from this round are used to construct the annual birth panel.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000420:36:1", "occurrence_id": "357e956da474203ceae29f8b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .015 0.025 0.011\n\n(0.017) (0.016) (0.017)\n\n2013 * Two or more 0.048*** 0.041** 0.048***\n\n(0.018) (0.017) (0.019)\n\n\nN 55,539 47,645 52,278\n\nControls for women’s characteristics <sup>a</sup> Yes Yes Yes\n\n<u>Controls for husband’s characteristics</u> <sup>b</sup> <u>No</u> <u>No</u> <u>Yes</u>\n\n<u>Note: The table reports average marginal effects from probit regressions run on an annual birth panel for the period 2003-2013</u>\ncreated with the DHS 2010 and DHS 2015 data. Each row reports the coefficients for year dummies interacted with the indicator\nfor having at least two children. All columns include year fixed effects as well as a dummy for having at least two children. The\nomitted category is 2008. Standard errors clustered at the mother level in parentheses below each estimate. Columns (1) and (3) are\nrun on the entire sample of ever married women, while column (2) restricts the sample to those who have at least one child.\na Control variables as in Table 3, column (1) are included but not reported.\nb Control variables as in Table 3, column (3) are included but not reported.\n*** Significant at 1%; ** significant at 5%; * significant at 10%.\n\nMention: \"DHS 2015 data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"The later round extends the birth panel through 2013.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000422:6:0", "occurrence_id": "4f70c85a746f5cea3b016bf0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: which found that certain socio-demographic characteristics such as education, gender, and income\n\nlevel had a role to play in determining the level of financial inclusion of Sri Lankans. Fintech can\n\nalso play a role in improving financial inclusion for rural communities, as seen in many parts of\n\nthe world. A study by Colombage (2010) explored the relevance of fintech in the Sri Lankan\n\ncontext.\n\n\nHowever, none of the aforementioned studies specifically focused on the implications of financial\n\ninclusion in contributing to inclusive growth in Sri Lanka. Analysis of the status of financial\n\ninclusion with a view towards creating a more level playing field and increasing opportunities for\n\nproductive engagement needs to consider socio-demographic factors as well. This paper provides\n\na comparative analysis of the state of financial inclusion in Sri Lanka, taking into consideration\n\nsocio-demographic factors. It analyzes the determinants of financial inclusion using data from the\n\n2018 National Financial Inclusion Survey for Sri Lanka. It concludes by highlighting some\n\nimplications for financial inclusion and inclusive growth in Sri Lanka.\n\n\n**1.1.** **A Brief History of Financial Inclusion Initiatives in Sri Lanka**\n\n\nThe arrival of the British in 1828, and the establishment of the commercial plantation sector, saw\n\nthe start of formal financial services in Sri Lanka (Seelanantha and Wickremasinghe, 2009). A\n\nparallel informal financial sector also emerged, composed primarily of moneylending and\n\npawnbroking. As pawnbroking became popular, laws were enacted to protect customers <sup>6</sup> [^6: The first Ordinance for Pawnbroking was enacted in 1942.] and\n\nimplement anti-usury policies (Karunagoda, 2007). The 20 <sup>th</sup> century saw the rise of the cooperative\n\nmovement as an organized mechanism to serve the farming community’s financing needs. Sri\n\nLanka’s central banking system was formally established in 1950 following independence from\n\nthe British, where the Central Bank of Sri Lanka’s (CBSL) primary objective initially was on\n\nmonetary policy.\n\nMention: \"2018 National Financial Inclusion Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Determinants of financial inclusion are estimated on this national survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000423:14:1", "occurrence_id": "fd27aec7de93f3f9efe72560", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  in person dropped from 60 percent to 23 percent.\n\n###### Use of Digitalized Accounts Is Also Rising\n\nAccount ownership is key to financial inclusion. But secure and convenient ways to use accounts are\nessential if users are to reap the full benefits of account ownership. The Global Findex database also\nsheds light on whether and how people use their accounts for payments.\n\n\n_Digital Payments_\nGlobally, 52 percent of adults (more than three‐quarters of account owners) reported having made or\nreceived at least one digital payment using their account the previous year. In high‐income economies,\nthe share of account owners using their accounts for digital payments was 97 percent. In the euro area\nthe share of adults who made or received digital payments increased from 87 percent in 2014 to 92\npercent in 2017, and the share making digital payments increased from 92 percent to 97 percent. In\ndeveloping ECA economies, the share of adults making or receiving digital payments rose from 46\npercent to 60 percent between 2014 and 2017, suggesting that the share of account owners doing so\nincreased from 80 percent to 92 percent.\n\n\nECA countries with significant growth in the use digital payments include Belarus, Bosnia and\nHerzegovina, Bulgaria, Croatia, Georgia, Kazakhstan, the Republic of North Macedonia, Montenegro,\nRussia, Turkey, and Ukraine (figure 6). Several countries witnessed an increase of more than 15\npercentage points in the share of adults using digital payments during 2014–17. Uzbekistan, where\nfewer people used digital payments in 2017 than in 2014, seems to be an outlier (table 3).\n\n\n13\n\nMention: \"Findex database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Reported digital payment behaviour of adults worldwide is drawn from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000424:2:0", "occurrence_id": "6ab96396da7420aa26d81a2a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: #### **Financial Risk Management in Agriculture: Analyzing Data from a New** **Module of the Global Findex Database**\n\nLeora Klapper, Dorothe Singer, Saniya Ansar, and Jake Hess <sup>1</sup>\n\n\n**JEL Classification:** D14; G21, G22, O16; Q14\n\n\n**Keywords:** Financial Inclusion, Sub-Saharan Africa, Agriculture, Risk, Insurance, Household Finance\n\n\n1 Leora Klapper (lklapper@worldbank.org), Dorothe Singer (dsinger@worldbank.org), Saniya Ansar\n(sansar1@worldbank.org), and Jake Hess (jhess1@worldbank.org) are with the World Bank. This paper’s findings,\ninterpretations and conclusions are entirely those of the authors and do not necessarily represent the views of the\nWorld Bank, their Executive Directors, or the countries they represent. We are grateful to the Bill & Melinda Gates\nFoundation for providing financial support making the collection and dissemination of the 2017 Global Findex data\npossible and to the AXA Foundation for financial support to develop the module on financial risk management in\nagriculture.\n\nMention: \"Global Findex Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"A new financial risk management module of this database supplies the study's data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000424:2:0", "occurrence_id": "03496bc659f6d9a7b7d35ce9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: #### **Financial Risk Management in Agriculture: Analyzing Data from a New** **Module of the Global Findex Database**\n\nLeora Klapper, Dorothe Singer, Saniya Ansar, and Jake Hess <sup>1</sup>\n\n\n**JEL Classification:** D14; G21, G22, O16; Q14\n\n\n**Keywords:** Financial Inclusion, Sub-Saharan Africa, Agriculture, Risk, Insurance, Household Finance\n\n\n1 Leora Klapper (lklapper@worldbank.org), Dorothe Singer (dsinger@worldbank.org), Saniya Ansar\n(sansar1@worldbank.org), and Jake Hess (jhess1@worldbank.org) are with the World Bank. This paper’s findings,\ninterpretations and conclusions are entirely those of the authors and do not necessarily represent the views of the\nWorld Bank, their Executive Directors, or the countries they represent. We are grateful to the Bill & Melinda Gates\nFoundation for providing financial support making the collection and dissemination of the 2017 Global Findex data\npossible and to the AXA Foundation for financial support to develop the module on financial risk management in\nagriculture.\n\nMention: \"2017 Global Findex data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Funders supported the collection and dissemination of this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000425:24:0", "occurrence_id": "8b087e51b907ed498c682b5c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Appendix**\n\n\n**Table A1: Sample of countries**\n\nIn this table we report the countries included in the analysis, the income group according the 2019 World Bank classification,\nthe number of firms included in the last Enterprise Survey available, the total number of firms in the full sample, and the\nnumber of surveys available since 2008.\n\n\n\n**Last survey**\n**Country** **Income group (WB 2019)**\n\n**<u>available</u>**\n\n\n\n**Last survey** **Number of**\n\n**Full sample**\n**<u>available</u>** **<u>surveys</u>**\n\n\n\n**Country** **Income group (WB 2019)** **Full sample**\n\n**<u>available</u>** **<u>surveys</u>**\n\n<u>India</u> <u>Lower middle income</u> <u>1,789</u> <u>1,789</u> <u>1</u>\nChina Upper middle income 893 893 1\nMexico Upper middle income 585 591 2\nNigeria Lower middle income 548 548 1\nChile High income 436 438 2\nPhilippines Lower middle income 360 360 1\nColombia Upper middle income 325 758 3\nVietnam Lower middle income 323 323 1\nBangladesh Lower middle income 315 315 1\nPeru Upper middle income 284 800 3\nArgentina Upper middle income 247 761 3\nMyanmar Lower middle income 202 331 2\nKenya Lower middle income 183 359 2\nTunisia Lower middle income 178 178 1\nUkraine Lower middle income 169 279 2\nCosta Rica Upper middle income 164 164 1\nGhana Lower middle income 146 146 1\nEl Salvador Lower middle income 141 261 3\nTanzania Low income 133 133 1\nCzech Republic High income 126 184 2\nIndonesia Lower middle income 123 123 1\nMozambique Low income\n\nMention: \"Enterprise Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The latest available round determines the firms and countries entering the sample.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000426:9:0", "occurrence_id": "f2edcc7a1a888d068e60777a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 2: Days of Cash in Hand across Industries\n\n\n\n\n\n\n\n\n\n\n\n**Note** : This figure shows the number of days of operating expenses covered by cash held across U.S. listed firms.\n\n\nThe figure shows 2000-2016 averages by industry.\n\n\n**Source:** Authors’ calculations based on Compustat data.\n\n\nThe resilience of the corporate sector is also tightly linked to the duration of the pandemic shock. The exogenous health emergency is transitory. Eventually a vaccine or an\neffective treatment will be developed, or the disease will run its course, and the source\nof the crisis will basically disappear. Nonetheless, the timeline for these exits from the\npandemic are still unknown and hard to predict. Hence, the duration of the pandemic\ncrisis remains a crucial source of uncertainty. The longer the pandemic lasts, the tougher\nit would be for firms to withstand the economic downturn, survive, and recover. To the\nextent that the shock does not persist for too long, a significant proportion of firms are\nlikely to remain viable. That is, their net worth will still be positive as the economy\nreturns to pre-pandemic levels. However, not all firms can survive a prolonged crisis, with\nextensive lockdowns and other containment measures. While industries themselves will\nweather the shock and survive (e.g., the restaurant or airline industries will not disappear), the same cannot be said about individual firms. Those that do survive might be\nseverely impacted, possibly losing lines of business or customers, and may in fact need to\ntransform themselves (or close) to thrive in the post-pandemic world.\n\n\nThe heavy costs that the COVID-19 outbreak has imposed on the world economy have\nbeen born not only by firms themselves, but rather by all parties. In fact, the pandemic\ncrisis has disturbed a wide range of economic relationships. Shock-hit firms have suffered\na collapse in revenues. Shareholders of firms negatively affected by the pandemic have\n\n\n7\n\nMention: \"Compustat data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Days of cash in hand are calculated from listed firms' financial statements in this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000429:13:1", "occurrence_id": "c264a1d55bed6078aae751db", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  shows that there are both firms entering and exiting\nthe sample each year.\n\n\n**<u>Table 1: Number of factory assessments per factory</u>**\n\n\n**<u>Number of factory assessments</u>** **<u>Number of firms</u>**\n1 124\n2 84\n3 76\n4 63\n5 29\n6 28\n7 33\n<u>8</u> <u>24</u>\nSource: Author’s calculations using data from ILO‐IFC Better Work Vietnam.\n\n\n**<u>Table 2: Number of assessed factories per year</u>**\n\n\n**<u>Year</u>** **<u>Number of firms</u>**\n2010 65\n2011 116\n2012 136\n2013 139\n2014 177\n2015 219\n2016 253\n<u>2017</u> <u>286</u>\n\n\n12\n\nMention: \"data from ILO‐IFC Better Work Vietnam\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Factory assessment counts by year are tabulated from this compliance programme's records.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000434:20:1", "occurrence_id": "7f79dde58aef565873dc157f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  *\nRoad density 0.045 (0.003) *** 0.063 (0.004) *** 0.063 (0.004) ***\nTransport cost to port -0.005 (0.003) * -0.014 (0.003) *** -0.016 (0.003) ***\nln Population -0.225 (0.084) *** 0.556 (0.093) *** 0.604 (0.092) ***\nln Population density -0.417 (0.048) *** -0.826 (0.076) *** -0.856 (0.074) ***\nln Local wage 0.300 (0.081) *** 0.305 (0.078) ***\n<u>Secondary school enrollment</u> <u>0.012 (0.001) ***</u> <u>0.011 (0.001) ***</u>\nObs. 43808 43808 43808\nWald chi2 426.15 643.33 676.14\n<u>Pseudo R-squared</u> <u>0.066</u> <u>0.095</u> <u>0.099</u>\n\n\n7 According to the 2007 BEEP data, about 17 percent of the firms surveyed import materials and equipment\nfrom abroad. About 5 percent of the firms export their products.\n\nMention: \"2007 BEEP data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"A footnote reports the share of surveyed firms that import inputs or export output.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000435:49:0", "occurrence_id": "51e724f499e9980f7da63d01", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **D.** **Questionnaire**\n\n#### **COVID-19 – BUSINESS PULSE SURVEY (COV-BPS)**\n\n###### **The World Bank** **2020 – 2 nd Wave**\n\n\n|QUESTIONNAIRE|Col2|Col3|\n|---|---|---|\n|Firm ID|**id**|Unique identifier|\n|Date and time of the interview (start)|**cov0**|Date and time|\n\n\n\n\n\n\n\n\n\n\n|COV0. General characteristics/Screener|Col2|Col3|\n|---|---|---|\n|Question|Variable|Answer|\n|What type of product or service (describe)<br>represents this establishment’s largest share of<br>current sales? <br>|**cov0b** <br>|Text<br>(Product with largest share of annual<br>sales)|\n|How many paid workers (full-time and part-time)<br>does this establishment have at the moment?<br>INSTRUCTION: Include all full time and part time paid<br>workers.|**cov0c**<br> <br>|_____ Total number of workers<br> <br>|\n|What is the total share of female workers at the<br>moment?|**cov0d**|_____% Women<br>|\n\n\n\n\n\n\n\n\n\n|COV1. Operation status: Impact of COVID|Col2|Col3|\n|---|---|---|\n|Question|Variable|Answer|\n|What is the current status of this establishment?<br> <br> <br>\n\nMention: \"COVID-19 – BUSINESS PULSE SURVEY\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The appendix reproduces this wave's questionnaire and variable identifiers.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000436:9:1", "occurrence_id": "62dcc8703d7ee69c4f55ff1f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  show that the mean firm operating in the DRC has been in\n\nbusiness for nine years and employs approximately six people. Around 4 percent of establishments in the\n\nDRC are part of a multi-establishment firm, 2 percent engage in exporting activity and 3 percent are foreign\n\nowned. Other firm characteristics of note include total female employment, as well as female participation\n\nin management and ownership: the average firm employs only one woman (although the standard deviation\n\nis high), while approximately 23 percent of firms have a top female manager and 27 percent have some\n\ndegree of female participation in ownership. The data show that women have greater representation in the\n\ncapital city (Kinshasa), where the average firm employs two women, 28 percent of firms have a female top\n\nmanager and 32 percent have female participation in ownership. <sup>13</sup>\n\n\nThe DRC survey contains the standard ES questions on firm characteristics, performance, and business\n\nenvironment. The survey also contains several country-specific questions included to ascertain women’s\n\nautonomy, equality of economic opportunity, and gender discrimination in the DRC. These specific\n\nquestions were built based on the legal gender discriminations identified by the WBL data and discussed in\n\nthe previous section. <sup>14</sup>\n\n\n13 A breakdown of Table 3 by firm location is not provided here but is available upon request.\n14 The specific questions are presented in Table A1 in the Annex; in each case the question was only posed where\nrelevant, as explained in the second column of Table A1.\n\n8\n\nMention: \"WBL data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Legal gender discriminations identified by this index shaped the survey questions added for the country.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000437:5:1", "occurrence_id": "8f33c6658e32ca9404220dae", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  proceeds in three sections. We discuss the multiple data sources used in Section\n\n\n2. We then outline the empirical strategy and present the results in Subsection 3.1 and 3.2. We\n\n\nconclude in Section 4.\n\n##### **2 Data**\n\n\nWe base our analysis on information from three main data sources. First, the Business Environment\n\n\nand Enterprise Performance Surveys (BEEPSs) provide information about firms’ characteristics.\n\n\nSecond, the World Development Indicators (WDI) and Global Development Finance (GDF) contain\n\n\nthe macroeconomic variables to correct for inflation and exchange rate fluctuations, and capture\n\n\nthe macroeconomic conditions faced by firms. Third, the National Accounts Estimates of Main\n\n\nAggregates and Trade Policy Information System (TPIS) data allow us to estimate demand shocks.\n\n\nThe World Bank jointly with the European Bank for Reconstruction and Development admin\n\nistered the BEEP Surveys for four years (2002, 2005, 2009 and 2013) in 27 Eastern European and\n\nCentral Asia countries. <sup>3</sup> The firms interviewed were selected to form a representative sample of the\n\n\n1In a study of Spanish manufacturing firms, Campa and Shaver (2002) suggest that firms may decide to export to\nforeign markets to smooth their cash flow and thus exploit the imperfect correlation between the destination country\nand the Spanish business cycles.\n2Pavcnik (2002) concludes that Chilean manufacturing plants in import-competing sectors underwent large productivity improvements during the massive trade liberalization of the late 1970s. Similarly, examining a period of significant\nchanges in Colombian trade policy across industries between 1977 and 1991, Fernandes (2007) finds that tariff liberalization boosts plant productivity and the effect appears to be particularly strong if the firm is larger and operating\nin a less competitive industry.\n3Specifically the surveys are fielded in Albania, Armenia, Azerbaijan, Belarus, Bosnia and Herzegovina, Bulgaria, Croa\n\n4\n\nMention: \"Business Environment\n\n\nand Enterprise Performance Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Firm characteristics are taken from these business environment surveys.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000437:5:1", "occurrence_id": "d12d02e51328d36d0693ace8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  proceeds in three sections. We discuss the multiple data sources used in Section\n\n\n2. We then outline the empirical strategy and present the results in Subsection 3.1 and 3.2. We\n\n\nconclude in Section 4.\n\n##### **2 Data**\n\n\nWe base our analysis on information from three main data sources. First, the Business Environment\n\n\nand Enterprise Performance Surveys (BEEPSs) provide information about firms’ characteristics.\n\n\nSecond, the World Development Indicators (WDI) and Global Development Finance (GDF) contain\n\n\nthe macroeconomic variables to correct for inflation and exchange rate fluctuations, and capture\n\n\nthe macroeconomic conditions faced by firms. Third, the National Accounts Estimates of Main\n\n\nAggregates and Trade Policy Information System (TPIS) data allow us to estimate demand shocks.\n\n\nThe World Bank jointly with the European Bank for Reconstruction and Development admin\n\nistered the BEEP Surveys for four years (2002, 2005, 2009 and 2013) in 27 Eastern European and\n\nCentral Asia countries. <sup>3</sup> The firms interviewed were selected to form a representative sample of the\n\n\n1In a study of Spanish manufacturing firms, Campa and Shaver (2002) suggest that firms may decide to export to\nforeign markets to smooth their cash flow and thus exploit the imperfect correlation between the destination country\nand the Spanish business cycles.\n2Pavcnik (2002) concludes that Chilean manufacturing plants in import-competing sectors underwent large productivity improvements during the massive trade liberalization of the late 1970s. Similarly, examining a period of significant\nchanges in Colombian trade policy across industries between 1977 and 1991, Fernandes (2007) finds that tariff liberalization boosts plant productivity and the effect appears to be particularly strong if the firm is larger and operating\nin a less competitive industry.\n3Specifically the surveys are fielded in Albania, Armenia, Azerbaijan, Belarus, Bosnia and Herzegovina, Bulgaria, Croa\n\n4\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Macroeconomic variables used to deflate values come from this indicator series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000437:5:1", "occurrence_id": "8214ad0d0eb61b7d01e7c205", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  proceeds in three sections. We discuss the multiple data sources used in Section\n\n\n2. We then outline the empirical strategy and present the results in Subsection 3.1 and 3.2. We\n\n\nconclude in Section 4.\n\n##### **2 Data**\n\n\nWe base our analysis on information from three main data sources. First, the Business Environment\n\n\nand Enterprise Performance Surveys (BEEPSs) provide information about firms’ characteristics.\n\n\nSecond, the World Development Indicators (WDI) and Global Development Finance (GDF) contain\n\n\nthe macroeconomic variables to correct for inflation and exchange rate fluctuations, and capture\n\n\nthe macroeconomic conditions faced by firms. Third, the National Accounts Estimates of Main\n\n\nAggregates and Trade Policy Information System (TPIS) data allow us to estimate demand shocks.\n\n\nThe World Bank jointly with the European Bank for Reconstruction and Development admin\n\nistered the BEEP Surveys for four years (2002, 2005, 2009 and 2013) in 27 Eastern European and\n\nCentral Asia countries. <sup>3</sup> The firms interviewed were selected to form a representative sample of the\n\n\n1In a study of Spanish manufacturing firms, Campa and Shaver (2002) suggest that firms may decide to export to\nforeign markets to smooth their cash flow and thus exploit the imperfect correlation between the destination country\nand the Spanish business cycles.\n2Pavcnik (2002) concludes that Chilean manufacturing plants in import-competing sectors underwent large productivity improvements during the massive trade liberalization of the late 1970s. Similarly, examining a period of significant\nchanges in Colombian trade policy across industries between 1977 and 1991, Fernandes (2007) finds that tariff liberalization boosts plant productivity and the effect appears to be particularly strong if the firm is larger and operating\nin a less competitive industry.\n3Specifically the surveys are fielded in Albania, Armenia, Azerbaijan, Belarus, Bosnia and Herzegovina, Bulgaria, Croa\n\n4\n\nMention: \"Global Development Finance\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Together with the other indicator series it captures the macro conditions firms face.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000437:5:1", "occurrence_id": "56e7438d4193c51cbc9e47fc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  proceeds in three sections. We discuss the multiple data sources used in Section\n\n\n2. We then outline the empirical strategy and present the results in Subsection 3.1 and 3.2. We\n\n\nconclude in Section 4.\n\n##### **2 Data**\n\n\nWe base our analysis on information from three main data sources. First, the Business Environment\n\n\nand Enterprise Performance Surveys (BEEPSs) provide information about firms’ characteristics.\n\n\nSecond, the World Development Indicators (WDI) and Global Development Finance (GDF) contain\n\n\nthe macroeconomic variables to correct for inflation and exchange rate fluctuations, and capture\n\n\nthe macroeconomic conditions faced by firms. Third, the National Accounts Estimates of Main\n\n\nAggregates and Trade Policy Information System (TPIS) data allow us to estimate demand shocks.\n\n\nThe World Bank jointly with the European Bank for Reconstruction and Development admin\n\nistered the BEEP Surveys for four years (2002, 2005, 2009 and 2013) in 27 Eastern European and\n\nCentral Asia countries. <sup>3</sup> The firms interviewed were selected to form a representative sample of the\n\n\n1In a study of Spanish manufacturing firms, Campa and Shaver (2002) suggest that firms may decide to export to\nforeign markets to smooth their cash flow and thus exploit the imperfect correlation between the destination country\nand the Spanish business cycles.\n2Pavcnik (2002) concludes that Chilean manufacturing plants in import-competing sectors underwent large productivity improvements during the massive trade liberalization of the late 1970s. Similarly, examining a period of significant\nchanges in Colombian trade policy across industries between 1977 and 1991, Fernandes (2007) finds that tariff liberalization boosts plant productivity and the effect appears to be particularly strong if the firm is larger and operating\nin a less competitive industry.\n3Specifically the surveys are fielded in Albania, Armenia, Azerbaijan, Belarus, Bosnia and Herzegovina, Bulgaria, Croa\n\n4\n\nMention: \"Trade Policy Information System\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Combined with national accounts, this system's data allow demand shocks to be estimated.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000437:5:1", "occurrence_id": "1e92b3f4061f6aa2fe837276", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  proceeds in three sections. We discuss the multiple data sources used in Section\n\n\n2. We then outline the empirical strategy and present the results in Subsection 3.1 and 3.2. We\n\n\nconclude in Section 4.\n\n##### **2 Data**\n\n\nWe base our analysis on information from three main data sources. First, the Business Environment\n\n\nand Enterprise Performance Surveys (BEEPSs) provide information about firms’ characteristics.\n\n\nSecond, the World Development Indicators (WDI) and Global Development Finance (GDF) contain\n\n\nthe macroeconomic variables to correct for inflation and exchange rate fluctuations, and capture\n\n\nthe macroeconomic conditions faced by firms. Third, the National Accounts Estimates of Main\n\n\nAggregates and Trade Policy Information System (TPIS) data allow us to estimate demand shocks.\n\n\nThe World Bank jointly with the European Bank for Reconstruction and Development admin\n\nistered the BEEP Surveys for four years (2002, 2005, 2009 and 2013) in 27 Eastern European and\n\nCentral Asia countries. <sup>3</sup> The firms interviewed were selected to form a representative sample of the\n\n\n1In a study of Spanish manufacturing firms, Campa and Shaver (2002) suggest that firms may decide to export to\nforeign markets to smooth their cash flow and thus exploit the imperfect correlation between the destination country\nand the Spanish business cycles.\n2Pavcnik (2002) concludes that Chilean manufacturing plants in import-competing sectors underwent large productivity improvements during the massive trade liberalization of the late 1970s. Similarly, examining a period of significant\nchanges in Colombian trade policy across industries between 1977 and 1991, Fernandes (2007) finds that tariff liberalization boosts plant productivity and the effect appears to be particularly strong if the firm is larger and operating\nin a less competitive industry.\n3Specifically the surveys are fielded in Albania, Armenia, Azerbaijan, Belarus, Bosnia and Herzegovina, Bulgaria, Croa\n\n4\n\nMention: \"BEEP Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Four rounds administered with the regional bank form a representative firm sample across the region.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000438:10:1", "occurrence_id": "31d60c488ab85428a18c08b3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  from 13,975 to 5,565. <sup>28</sup> Furthermore, this is even\nmore complicated for multiproduct plants where the methodology approximates TFPQ by considering only\nthe major product of the firm. <sup>29</sup> In an analysis where firm growth is measured by change in overall\nemployment, it would be misleading to consider plant TFPQ of the major product of the plant. <sup>30</sup>\n\n**_Distances to Cities/Urban Agglomerations_**\n\nThis work uses GIS software to compute distances between every town in the LMMIS data and five major\nEthiopian cities that are considered to be centers of broader agglomerations. The five cities/urban\nagglomerations are Addis Ababa, Awassa, Bahir Dar, Dire Dawa and Mekele. Besides the agglomeration\ncenter itself, the following distance bands are defined from the nearest agglomeration center à la Ghani et\nal. (2016): 0-50 kilometer (km), 50-100 km, 100-150 km, and the residual category of towns that are located\nover 50 km away.\n\n**_Measuring Spatial Clustering_**\n\nThe Ellison-Glaeser index (EGI) measures the extent of spatial concentration in a given industry. It takes\nnegative values when the industry is more dispersed than what random assignment would lead to, while\npositive values where industries show greater levels of concentration. <sup>31</sup> The EGI for industry _s_ at time _t_ is\ndefined as:\n\n\n27 The project also experimented with alternative definitions of TFP and found the results to be robust to such changes. Another\nproblem related to TFP estimation is measurement errors; these are arguably all the more pervasive in the context of developing\ncountries.\n28 Examples of such products include bricks of clay, cement blocks, nails, sugar, bread and wheat\n\nMention: \"LMMIS data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Town locations in this enterprise list are used to compute GIS distances to five urban centres.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000439:28:1", "occurrence_id": "8f820254ce2289c335686b95", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  <mark>3). Indirect</mark>\n<mark>taxes have the highest negative marginal contributions to inequality reduction (they actually increase inequality in</mark>\n<mark>Moldova more than in other countries (F</mark> igure 1 <mark>3). In terms of progressivity, most interventions are similar in</mark>\n<mark>Moldova, Belarus, Romania, Russia and Ukraine, however the indirect taxes are most regressive in Moldova (F</mark> igure\n14).\n\n**Figure 13. Marginal contributions to inequality reduction in Moldova and peers, percentage points.**\n\n14\n\n\n12\n\n\n10\n\n\n8\n\n\n6\n\n\n4\n\n\n2\n\n\n0\n\n\n-2\n\n\n\n-4\n\n\n\nSSC Direct Taxes Pensions Direct Transfers Indirect Taxes Indirect\nSubsidies\n\n\n\nIn-kind Health In-kind\nEducation\n\n\n\nBelarus Moldova Romania Russia Ukraine\n\nNote: Some numbers on this chart are estimated using backward calculations from the original source and might be not fully precise.\nSource: authors’ calculations using HLCS-2016 data for Ukraine and respective CEQ analysis for other countries: Belarus – Bornukova et\nal., 2017; Poland - <mark>Goraus and Inchauste, 2016; Russia – Popova et al., 2018.</mark>\n\n\n27\n\nMention: \"HLCS-2016 data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Fiscal incidence for Ukraine is calculated on this household survey, with other countries using comparable analyses.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000442:18:1", "occurrence_id": "952ed75eeb0d09d2d50b39b8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: * 𝐿𝐿𝑖𝑖, (4)\n\nwhere 𝑃 **𝑃** 𝑃𝑃𝑖𝑖 is the probability of early termination for project _i_, 𝐸 **𝐸** 𝐸𝐸𝑖𝑖 is the exposure of the\ngiven termination.\ngovernment from project 𝑖𝑖 at termination, and 𝐿𝐿𝐺𝐺𝑇𝑇𝑖𝑖 is the loss of the government from project 𝑖𝑖\n\n**𝑃** **𝐸** **𝐿**\n**𝐸**\n\n\n\n**𝑃** **𝐸**\ngiven termination. 𝑃 **𝑃** 𝑃𝑃𝑖𝑖, 𝐸 **𝐸** 𝐸𝐸𝑖𝑖, and 𝐿 **𝐿** 𝐿𝐿𝑖𝑖 are obtained separately. The probability of early termination of a PPP is\n\npredicted using the econometric estimations presented in section 3. 𝐸 **𝐸** 𝐸𝐸𝑖𝑖 is estimated based on\ncountries in case of termination of PPP contracts and the recovery rates of defaulted loans to\nthe debt and equity invested in the project. 𝐿 **𝐿** 𝑇𝑇𝑖𝑖 is determined based on different practices by\ninfrastructure projects.\n\n\n\n**_5.1_** **_Sample Description_**\nThis paper is part of a research project that analyzes the fiscal risks from SOEs and PPPs in the\nelectricity and transport sectors. The project focuses on a sample of 19 countries, selected based\non the availability of detailed financial information of SOEs. Two of the 19 countries are not\nincluded in the PPI database. Therefore, the sample of countries used here are the remaining 17\ncountries: Albania, Argentina, Bhutan, Brazil, Bulgaria, Burundi, Ethiopia, Georgia, Ghana,\nIndonesia, Kenya, Kosovo, Peru, Romania, the Solomon Islands, South Africa, and Ukraine. As the\nbroader research agenda focuses on the electricity and transport sectors, the rest of this paper\nalso confines itself to these two sectors.\n\n\nTable 3 summarizes the active portfolio of the sample countries in terms of portfolio size, number\n\nMention: \"PPI database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"It supplies the financial data on private participation projects and determines which countries remain in the sample.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000446:10:0", "occurrence_id": "b750cf323dcabb06b92252b5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: mean price of wheat increased from 8.20/kg in wave 1 to 10.48/kg in wave 3; the mean price\n\n\nfor maize increased from 5.33/kg in wave 1 to 6.40/kg in wave 3. For purposes of our analysis,\n\n\nwe compute a yearly average price for each of the three crops at the zone level. <sup>12</sup> <sup>13</sup>\n\n\n**4.** **Empirical Approach**\n\n\nTo assess how food prices affect child undernutrition, we estimate the following equation:\n\n\n𝑌�� �𝛽� �𝛽�𝑋�� �𝛽�𝑋�� �𝛽�𝑋�� �𝛽�𝑃�� �𝛾� �𝜀�� (1)\n\n\nWhere 𝑌�� is the outcome indicator for child _i_ at time _t_ in terms of the z score for height for age\n\nand weight for height; <sup>14</sup> 𝑋�� is a vector of individual characteristics; 𝑋�� is a vector of household\n\ncharacteristics; 𝑋�� is a vector for community characteristics such as road access and access to\n\nmarkets. Our variable of particular interest, 𝑃��, is the price for each crop at the district/zone\n\nlevel within each region in Ethiopia while 𝛾� are region fixed effects <sup>15</sup> and 𝜀�� is the error term _._\n\n\nFor stunting, which measures persistent long-term child undernutrition, the impact of changes\n\n\nin prices is likely to take time to be realized. We therefore lag the crop price by one period\n\nbefore merging again with wave 2 and wave 3 of the LSMS data set and include 𝑃���� in the\n\nmodel. For wasting, which captures more rapid weight loss, we use the contemporaneous price.\n\n\nThe vector of control variables is drawn from previous studies that seek to explain child\n\n\nundernutrition rates (see Cintron, Seff, and Baird (\n\nMention: \"LSMS data set\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"Lagged crop prices are merged into the second and third waves before estimating nutrition effects.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000449:39:0", "occurrence_id": "0ce07602226c96fbadb07cba", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ## **Appendix**\n\n##### **A Data Appendix**\n\n**A.1** **Survey** **Questions** **in** **the** **SOEP**\n\n\nTable A.1 summarizes the construction of the outcome variables based on the survey questions from\nthe SOEP. In each of the years in Column (4), respondents are asked the survey question in Column\n(2) and presented the mutually exclusive answer options in Column (3). We use the answer options\nto create binary indicators, whereby for all questions except 5)), a value of one indicates an answer\nconsistent with greater social cohesion. With the question about concerns 5), a value of one indicates\nthat people are very concerned about crime. The classification of the binary indicators is shown in\nparentheses in Column (3).\n\nMention: \"SOEP\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Outcome variables are constructed from the wording of this panel's survey questions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000451:12:0", "occurrence_id": "feaca67fca4bdc96277a9691", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Rural-to-urban migration and the resulting increase in urbanization has been studied extensively in the\n\nrelevant literature. The classic model of rural-to-urban migration from Harris and Todaro (1970) starts\n\nfrom the assumption that expected income differentials between rural and urban areas determine a\n\nlong-term equilibrium in which migration eventually equalizes agricultural wages and the expected\n\nurban wage rate. Empirically, this relationship has been demonstrated in a wide range of studies and\n\ncountry contexts, but with important nuances with respect to migrant characteristics, social networks,\n\ndifferences in the cost-of-living, and the quality of local amenities. A growing literature referred to as\n\n“new economic geography” popularized by Paul Krugman (Krugman, 1991; Krugman, 1998)\n\ncomplements the conclusions of the classic Harris and Todaro model by emphasizing the economies\n\nof scale and agglomeration benefits that larger cities tend to generate (Overman and Venables, 2010).\n\nAlso consistent with this literature, Ferré, Ferreira, and Lanjouw (2012) find a strong inverse\n\nrelationship between poverty and city size using small-area estimation techniques. Small-area estimates\n\nreported in Seitz (2019b) suggest the relationship holds true for the subregion of Central Asia, as well\n\nas specifically in Uzbekistan (Figure 3), where poverty rates are substantially lower in the most densely\n\npopulated areas.\n\n\n**Figure 3: Small Area Estimates of Mean Consumption Per Capita in Districts of Uzbekistan**\n**(2011 $PPP)**\n\n\n_Notes: Small area estimates are derived from survey responses in the L2CU Baseline. Author’s Calculations. For more details,_\n\n_see Seitz (2019b)_\n\n\n11\n\nMention: \"L2CU Baseline\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"District-level consumption maps are produced as small area estimates from responses in this baseline.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000452:12:0", "occurrence_id": "59a3e05c9151de1d4b6ac6e7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Ethiopian civil war not be correlated with year of birth and pre-reform levels of schooling in the same way\n\n\nas the FPE reform. Potential bias generated by contemporaneous changes in educational investments, the\n\n\nimpact of the civil war, and the 2000 law banning marriage for those under 18 are explored in more detail\n\n\nin Section 6.4. Additionally, it would be problematic if women and families relocated at the time of the\n\n\nreform’s implementation in such a way that higher-ability students sorted into areas with higher predicted\n\n\nintensity of the reform. However, this type of sorting is unlikely to occur in the studied setting. First, data\n\n\nfrom the 2016 Living Standards Measurement Study show that 86 percent of respondents in the relevant\n\n\ncohort range live in their region of birth. Furthermore, the intensity measure is explicitly designed to predict\n\n\na greater impact of the reform in areas with lower initial levels of schooling. A violation of this assumption\n\n\nwould entail the unlikely scenario that higher-ability students’ families were moving to areas that were worse\n\n\noff at the time of reform implementation, even though they could have received the same reduction in fees\n\n\nin their original education zone.\n\n##### **6 Results**\n\n\n**6.1** **Effect** **of** **FPE** **on** **Years** **of** **Schooling** **and** **Fertility**\n\n\nTo begin the analysis, the OLS relationship documents the general correlation seen in the data. This is\n\n\ndone by estimating equation (9) using the reported years of schooling from the data, not the predicted level\n\n\nfrom the first stage. A negative relationship between fertility and schooling has been well documented in\n\n\nthe literature (Ainsworth et al., 1996; Lam and Duryea, 1999; Schultz, 1994, 1997). The OLS estimates\n\n\nare shown in column 1 of\n\nMention: \"2016 Living Standards Measurement Study\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"It shows most of the relevant cohort still live in their region of birth, arguing against sorting bias.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000453:5:1", "occurrence_id": "0b4730b25ece450c05ba7043", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  or microfinance (Agarwal et al. 2018,\n\nKarlan and Zinman, Azevedo et al. 2019), while our focus is on SME lending.\n\nSecond, our paper relates to the literature on the consequences of information\n\nsharing through credit registries on bank competition and credit supply (eg. Jappelli and\n\nPagano 2002, Djankov et al. 2007, Hertzberg et al. 2011, Liberman 2016, Dobbie et al.\n\n2016, Foley et al. 2018, Sutherland 2018). As noted above, our principal contribution to\n\nthis literature is to document that information spillovers between lenders can also occur\n\nthrough loan approvals rather than loan issuance. Because loan approvals are not encoded\n\nin the Peruvian credit registry, our results imply that information spillovers can occur even\n\nbefore information is encoded in the credit registry, and before the initial lender has derived\n\nany benefit from screening new borrowers. And our results imply that information\n\n4\n\nMention: \"Peruvian credit registry\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Because approvals are absent from this registry, spillovers must occur before information is recorded.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000453:5:1", "occurrence_id": "4d34e1bbf5c2264a2b18044d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  or microfinance (Agarwal et al. 2018,\n\nKarlan and Zinman, Azevedo et al. 2019), while our focus is on SME lending.\n\nSecond, our paper relates to the literature on the consequences of information\n\nsharing through credit registries on bank competition and credit supply (eg. Jappelli and\n\nPagano 2002, Djankov et al. 2007, Hertzberg et al. 2011, Liberman 2016, Dobbie et al.\n\n2016, Foley et al. 2018, Sutherland 2018). As noted above, our principal contribution to\n\nthis literature is to document that information spillovers between lenders can also occur\n\nthrough loan approvals rather than loan issuance. Because loan approvals are not encoded\n\nin the Peruvian credit registry, our results imply that information spillovers can occur even\n\nbefore information is encoded in the credit registry, and before the initial lender has derived\n\nany benefit from screening new borrowers. And our results imply that information\n\n4\n\nMention: \"credit registry\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"The general point is that spillovers precede the encoding of information in such registries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000455:6:2", "occurrence_id": "c34ec1066407a22565e8dd94", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  output per capita is decomposed in several consecutive\nsteps. _First_, growth in output per capita is decomposed into changes in the aggregate output per worker, changes\nin the aggregate employment rate and changes in the share of the working age population among the entire\npopulation. _Second_, changes in the aggregate employment rate are further decomposed into changes in the\nemployment rate by sector. _Third_, changes in the aggregate output per worker are decomposed into changes\nlinked to variations in output per worker within sectors (within-sector productivity growth) and changes linked\n\n\n3 For reasons of data availability, this study uses output (or value-added) instead of GDP. In contrast to GDP, output does not include\ntaxes and subsidies; differences between GDP and output are usually very small. The sum of sectoral outputs from DOSM is used to\ncompute aggregate output.\n\n\n5\n\nMention: \"DOSM\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Sectoral output figures are summed to obtain aggregate output for the decomposition.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000456:6:1", "occurrence_id": "f1898fcc7b6233ec164ea98a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  area measurement\nthan self-reporting. Although the issues with GPS measurement error should not be overlooked and\nare discussed again in the conclusion of the article, here GPS measurement is considered closer to\nthe true land area and we refer to deviations from the GPS area as self-reported measurement error.\n\n_LSMS-ISA data_\n\nThe LSMS-ISA data sets create an opportunity for analyzing the different methods of land data\ncollection since the studies employ GPS measurement while continuing to collect self-reported land\nareas. Data are currently available for four LSMS-ISA countries: Malawi (2010/11), Uganda\n(2009/10), Tanzania (2010/11) and Niger (2011). For agricultural households in each of the\ncountries, the plot manager or head of household is first asked to estimate the plot areas and then the\nplots are measured using GPS, so as not to influence the farmer’s reported figure.\n\nDepending on the survey, farmers may be required to report in a single unit (as in Uganda,\nTanzania and Niger) or they may have the choice to report in a number of units (as in Malawi). In\neither scenario, the estimate is subject to previously noted errors. GPS devices allow for\nmeasurement in a variety of units, but are typically recorded in acres in LSMS-ISA projects and\nrounded to the nearest hundredth.\n\nThe exclusion of certain plots from GPS measurement was mandated by the survey manual for\nthe sake of time and budget. In Malawi, for example, enumerators were instructed to measure all\nplots within two hours of walking from the household. In Tanzania the guidelines were to measure\nall plots within one hour (via any mode of transportation) and Uganda required all plots within the\nenumeration area to be measured (Government of Malawi, NSO, 2010; World Bank Group, n.d.;\nUBOS, 2009). Table 1 provides basic information on the surveys considered in this study and how\nland\n\nMention: \"LSMS-ISA data sets\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Because they collect GPS and self-reported plot areas together, they allow the two methods to be compared.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000458:23:2", "occurrence_id": "24e8e214460393310b5f14e6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  are discussed subsequently in the paper.\n115 Twenty-five countries are included in the study, including: Benin, Botswana, Burkina Faso, Cameroon, Côte d’Ivoire,\nDemocratic Republic of Congo, Ethiopia, Ghana, Kenya, Malawi, Mali, Mauritius, Mozambique, Namibia, Nigeria, Rwanda,\nSenegal, Sierra Leone, South Africa, Swaziland, Tanzania, Togo, Uganda, Zambia, and Zimbabwe.\n116 While this number of data points allows for sufficiently significant and robust results, the n-size will be subsequently increased\nas more data become available in mid-/late-2019.\n117 For instance, comparing countries in Sub-Saharan Africa with OECD countries, Latin America, or East Asia would not be as\ninstructive as the latter are much closer to the frontier in key service delivery areas and would not have the same variation.\n118 Fortunately, the data available support this method, since annual OGD evaluations are released in January of each year, and\naccountability and service delivery indicators are released in October, thus allowing a nine-month lag.\n\n\n22\n\nMention: \"annual OGD evaluations\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Its January release timing creates the nine-month lag used in the identification.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000458:23:2", "occurrence_id": "031fba02723f49bc7e1b5a8d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  are discussed subsequently in the paper.\n115 Twenty-five countries are included in the study, including: Benin, Botswana, Burkina Faso, Cameroon, Côte d’Ivoire,\nDemocratic Republic of Congo, Ethiopia, Ghana, Kenya, Malawi, Mali, Mauritius, Mozambique, Namibia, Nigeria, Rwanda,\nSenegal, Sierra Leone, South Africa, Swaziland, Tanzania, Togo, Uganda, Zambia, and Zimbabwe.\n116 While this number of data points allows for sufficiently significant and robust results, the n-size will be subsequently increased\nas more data become available in mid-/late-2019.\n117 For instance, comparing countries in Sub-Saharan Africa with OECD countries, Latin America, or East Asia would not be as\ninstructive as the latter are much closer to the frontier in key service delivery areas and would not have the same variation.\n118 Fortunately, the data available support this method, since annual OGD evaluations are released in January of each year, and\naccountability and service delivery indicators are released in October, thus allowing a nine-month lag.\n\n\n22\n\nMention: \"accountability and service delivery indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"These outcome indicators, released in October, are the dependent variables.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000464:22:3", "occurrence_id": "39cc4e97f108383fc77c1aff", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  in agriculture, the structure of households, and the gender\nagricultural productivity gap result in higher tax burdens for women than for men. A more progressive perhectare tax schedule with exemptions for smallholders from paying agricultural income tax would reduce the\ntax burdens for women, but it would continue to be regressive.\n\nWe also found a substantial difference between self-reported tax payments and tax liabilities imputed based on\nland area and the tax schedules. The discrepancies in reported tax payments and estimated tax liabilities point\nto the importance of administrative tax data and land registries to complement survey data so as to ensure that\n\n\n\n21\n\nMention: \"administrative tax data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Imputed liabilities diverge from self-reported payments, showing surveys alone are insufficient.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000464:22:3", "occurrence_id": "255e6ce2f12e779cd1370e9f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  in agriculture, the structure of households, and the gender\nagricultural productivity gap result in higher tax burdens for women than for men. A more progressive perhectare tax schedule with exemptions for smallholders from paying agricultural income tax would reduce the\ntax burdens for women, but it would continue to be regressive.\n\nWe also found a substantial difference between self-reported tax payments and tax liabilities imputed based on\nland area and the tax schedules. The discrepancies in reported tax payments and estimated tax liabilities point\nto the importance of administrative tax data and land registries to complement survey data so as to ensure that\n\n\n\n21\n\nMention: \"land registries\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Registry records are proposed alongside tax records to verify plot areas and liabilities.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000467:7:1", "occurrence_id": "5f249c0eeb645f300cc231f3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: . The nation-wide survey of bank loan officers commissioned\n\n\nby the World Bank suggests that women-owned enterprises in Turkey are more likely to suffer\n\n\nfrom bias in the credit market compared with men-owned enterprises, independent of their\n\n\ncredit risk. Alleviating this bias can help ease access to finance constraints for women-owned\n\n\nbusinesses and close the gender business performance gap both in Turkey and more broadly in\n\n\ncountries across the world which face similar gender gaps in SME lending.\n\n\n1 An important caveat from the Enterprise Survey data is that firms managed but not owned by women do not\nappear to face similar constraints. There is no significant difference between male and female-managed firms\nwhen it comes to having a savings account, an overdraft facility or a line of credit with a financial institution.\nThe number of loans requested and the loan approval rate are also similar, as well as the number of days waited\nto receive the loan and the collateral requested (ES, 2015).\n\n\n6\n\nMention: \"nation-wide survey of bank loan officers\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Lender judgements indicate women-owned firms face credit bias independent of risk.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000467:7:1", "occurrence_id": "dab13f72de1d9fa90a6066e4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: . The nation-wide survey of bank loan officers commissioned\n\n\nby the World Bank suggests that women-owned enterprises in Turkey are more likely to suffer\n\n\nfrom bias in the credit market compared with men-owned enterprises, independent of their\n\n\ncredit risk. Alleviating this bias can help ease access to finance constraints for women-owned\n\n\nbusinesses and close the gender business performance gap both in Turkey and more broadly in\n\n\ncountries across the world which face similar gender gaps in SME lending.\n\n\n1 An important caveat from the Enterprise Survey data is that firms managed but not owned by women do not\nappear to face similar constraints. There is no significant difference between male and female-managed firms\nwhen it comes to having a savings account, an overdraft facility or a line of credit with a financial institution.\nThe number of loans requested and the loan approval rate are also similar, as well as the number of days waited\nto receive the loan and the collateral requested (ES, 2015).\n\n\n6\n\nMention: \"Enterprise Survey data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"The data show constraints attach to ownership rather than management, qualifying the conclusion.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000475:3:0", "occurrence_id": "6efdb1d1a4cea78bb19d4bcc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ### 1. Introduction\n\nGlobally, the negative socioeconomic impacts of COVID-19 are expected to affect women more than men.\nFor example, on average, the share of women is larger in the informal sector (ILO 2020), and given that\nthe pandemic affected the informal sector more, the female working population is more likely to be\neconomically affected. Beyond economic impacts, COVID-19 can disproportionately affect women’s\nhealth, education, and voice and agency through different channels, including increased unpaid care work,\nreallocation of health care resources, and increased gender-based violence during lockdown (UN Women\n2020). On the other hand, worldwide male COVID-19 patients are more likely to be directly affected by\nthe virus, needing more intensive treatment or succumbing to mortality (Peckham et al. 2020).\n\nWhile the gender-differentiated transmission channels and impacts of COVID-19 are widely recognized,\nand there is new, sprawling evidence on the gendered impact of COVID-19, few studies quantitatively\nassess the possible or actual economic impacts on women through labor distribution channels and less so\nfor women in fragile and conflict-prone countries. This paper aims to assess the possible and actual\neconomic impacts of COVID-19 on women and female-headed households in Chad using Computable\nGeneral Equilibrium (CGE) model-based simulations and evidence from a household phone survey\nconducted during the pandemic.\n\nChad has a large informal economy and ranks 147 of 153 countries in the global gender gap index. The\nfirst cases of COVID-19 in Chad were reported in March 2020, and since then, as of March 2021, fewer\nthan 5,000 cases of COVID-19 have been reported. There was also a compound health risk with the\noutbreak of malaria and chikungunya in 2020. The risks of the virus spreading remain high with the advent\nof new variants. Even though COVID-19 cases may not seem to be high in\n\nMention: \"household phone survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Evidence on pandemic impacts for women comes from this survey fielded during the crisis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000475:3:0", "occurrence_id": "6a4b51cf3af301a046f7a30c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ### 1. Introduction\n\nGlobally, the negative socioeconomic impacts of COVID-19 are expected to affect women more than men.\nFor example, on average, the share of women is larger in the informal sector (ILO 2020), and given that\nthe pandemic affected the informal sector more, the female working population is more likely to be\neconomically affected. Beyond economic impacts, COVID-19 can disproportionately affect women’s\nhealth, education, and voice and agency through different channels, including increased unpaid care work,\nreallocation of health care resources, and increased gender-based violence during lockdown (UN Women\n2020). On the other hand, worldwide male COVID-19 patients are more likely to be directly affected by\nthe virus, needing more intensive treatment or succumbing to mortality (Peckham et al. 2020).\n\nWhile the gender-differentiated transmission channels and impacts of COVID-19 are widely recognized,\nand there is new, sprawling evidence on the gendered impact of COVID-19, few studies quantitatively\nassess the possible or actual economic impacts on women through labor distribution channels and less so\nfor women in fragile and conflict-prone countries. This paper aims to assess the possible and actual\neconomic impacts of COVID-19 on women and female-headed households in Chad using Computable\nGeneral Equilibrium (CGE) model-based simulations and evidence from a household phone survey\nconducted during the pandemic.\n\nChad has a large informal economy and ranks 147 of 153 countries in the global gender gap index. The\nfirst cases of COVID-19 in Chad were reported in March 2020, and since then, as of March 2021, fewer\nthan 5,000 cases of COVID-19 have been reported. There was also a compound health risk with the\noutbreak of malaria and chikungunya in 2020. The risks of the virus spreading remain high with the advent\nof new variants. Even though COVID-19 cases may not seem to be high in\n\nMention: \"global gender gap index\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"The country's rank is quoted to characterise the starting gender gap.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000478:4:0", "occurrence_id": "4d7563c24686ad35a537adde", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: households, which can be either couples or individuals, make choices over time with regard\n\n\nto labor supply, private savings and retirement in an environment with uncertainty about\n\n\nfuture wages, returns on pension savings, fertility, divorce or widowhood and own survival.\n\n\nMen and women choose not only whether to work or not but also whether to work in the\n\n\nformal sector where pension contributions are mandatory or in the informal sector. The\n\n\ndistinction between formal and informal sector work is important in Chile (and most coun\n\ntries across the world) because many individuals work in the informal sector and reach old\n\n\nage having contributed little to the system. In addition, women choose whether to work\n\n\npart-time or full-time. The model is estimated on longitudinal microdata from the _Encuesta_\n\n\n_de_ _Protecci´on_ _Social_ _(EPS)_ merged with administrative data on pension funds. <sup>4</sup>\n\n\nWe use the estimated model to study how labor supply and savings behavior changes\n\n\nwith the introduction of the 2008 pension reform in comparison to the previous pension\n\n\nsystem behavior and in comparison to alternative pension system designs. We estimate\n\n\nmodel parameters by the method of simulated moments using pre-reform data from the\n\n\n2004 and 2006 EPS Surveys. The estimation results reveal substantial heterogeneity in how\n\n\nindividuals value pecuniary and non-pecuniary dimensions of formal and informal work.\n\n\nSome individuals have strong comparative advantages to working in the formal or informal\n\n\nsectors, which dampens behavioral responses to pension rule changes.\n\n\nIncorporating the pension rules explicitly in our model allows us to construct counterfac\n\ntual simulations in which the reform did not happen, keeping everything else constant. In\n\n\nthis way, we perform an ex ante prediction of the short-run behavioral impacts of the pension\n\n\nreform and compare our predictions to some evidence from other causal impact studies of\n\n\nthe reform’s impacts. Consistently with other studies, we find that the more generous safety\n\n\nnet induces negative wealth\n\nMention: \"administrative data on pension funds\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Contribution records are merged with the household panel to estimate the lifecycle model.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000478:4:0", "occurrence_id": "db6199e831dddf8abfb67c85", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: households, which can be either couples or individuals, make choices over time with regard\n\n\nto labor supply, private savings and retirement in an environment with uncertainty about\n\n\nfuture wages, returns on pension savings, fertility, divorce or widowhood and own survival.\n\n\nMen and women choose not only whether to work or not but also whether to work in the\n\n\nformal sector where pension contributions are mandatory or in the informal sector. The\n\n\ndistinction between formal and informal sector work is important in Chile (and most coun\n\ntries across the world) because many individuals work in the informal sector and reach old\n\n\nage having contributed little to the system. In addition, women choose whether to work\n\n\npart-time or full-time. The model is estimated on longitudinal microdata from the _Encuesta_\n\n\n_de_ _Protecci´on_ _Social_ _(EPS)_ merged with administrative data on pension funds. <sup>4</sup>\n\n\nWe use the estimated model to study how labor supply and savings behavior changes\n\n\nwith the introduction of the 2008 pension reform in comparison to the previous pension\n\n\nsystem behavior and in comparison to alternative pension system designs. We estimate\n\n\nmodel parameters by the method of simulated moments using pre-reform data from the\n\n\n2004 and 2006 EPS Surveys. The estimation results reveal substantial heterogeneity in how\n\n\nindividuals value pecuniary and non-pecuniary dimensions of formal and informal work.\n\n\nSome individuals have strong comparative advantages to working in the formal or informal\n\n\nsectors, which dampens behavioral responses to pension rule changes.\n\n\nIncorporating the pension rules explicitly in our model allows us to construct counterfac\n\ntual simulations in which the reform did not happen, keeping everything else constant. In\n\n\nthis way, we perform an ex ante prediction of the short-run behavioral impacts of the pension\n\n\nreform and compare our predictions to some evidence from other causal impact studies of\n\n\nthe reform’s impacts. Consistently with other studies, we find that the more generous safety\n\n\nnet induces negative wealth\n\nMention: \"EPS Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Pre-reform waves provide the moments used to estimate preferences over formal and informal work.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000479:52:1", "occurrence_id": "7fbf20f6e4c73987e8c487f7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \nObservations 71,866 71,866 38,330\nNumber of districts 1,286 1,286 531\n<u>Mean</u> <u>dep.</u> <u>var</u> <u>0.970</u> <u>0.970</u> <u>0.970</u>\n\n\nDistrict FE YES YES YES\nProvince*Year FE YES YES YES\n<u>Covariates</u> <u>YES</u> <u>YES</u> <u>YES</u>\n\n\n_Notes:_ *** p _<_ 0.01, ** p _<_ 0.05, - p _<_ 0.1. The dependent variable is a dummy indicating whether the child is\ncurrently attending primary school. The independent variables measures the presence of a WJC within a 1km\nEuclidean buffer of the child’s cluster of residence and presence of a WJC center in the child’s district. Robust\nstandard errors (in parentheses) are clustered at the district level. The sample of primary school level includes\nchildren between the ages of 6 and 11. Covariates include age, gender, household’s head years of education,\nnumber of children in the household aged 0-18, number of children in the household aged 0-5, number of female\nadults, number of male adults, rural residence dummy, district fixed effect and province-by-year fixed effect.\nSource: Peru DHS 2006-2014.\n\n\n50\n\nMention: \"Peru DHS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"District-level regressions on household composition and outcomes are estimated on these rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000482:17:0", "occurrence_id": "def956e4bb78308f4257f268", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: crop, livestock, forest, and fish allocation far outperforms the method of using rural population to\ndistribute the agricultural GDP, at least in the case of Brazil.\n\n#### References\n\n\nS. Angel, P. Lamson‐Hall, B. Guerra, Y. Liu, N. Galarza, and A. M. Blei (2018). Our Not‐So‐Urban World,\nWorking Paper No. 42 The Marron Institute of Urban Management, New York University.\n\n\nArvis, J. F., & Shepherd, B. (2013). The Poisson quasi‐maximum likelihood estimator: a solution to the\n‘adding up’ problem in gravity models. _Applied Economics Letters_, _20_ (6), 515‐519.\n\n\nArvis, J.F; Blankespoor, B. Downscaling Economic Variables: An Entropy Estimation. Mimeo.\n\n\nBlanchard, P.B.; Blankespoor, B. Rivera‐Fuentes, J.; Gunasekera, R.; Ishizawa O. and Jimenez L.F. Spatial\nDisaggregation of Gross Domestic Product for applications in Disaster Risk Management. Mimeo.\n\n\nBosker, M.; Park, J.; Roberts, M. (2018). Definition Matters : Metropolitan Areas and Agglomeration\nEconomies in a Large Developing Country (English). Policy Research working paper; no. WPS 8641.\nWashington, D.C. : World Bank Group.\n\n\nCenter for International Earth Science Information Network ‐ CIESIN ‐ Columbia University. (2017).\nGridded Population of the World, Version 4 (GPWv4): Population Count Adjusted to Match 2015\nRevision of UN WPP Country Totals, Revision 10. Palisades, NY: NASA Socioeconomic Data and\nApplications Center (SEDAC). https://doi.org/10.7927/H4JQ0XZW.\n\n\nChen, X., &\n\nMention: \"Gridded Population of the World\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Gridded population counts are used to allocate agricultural output across space.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000482:17:0", "occurrence_id": "ce17276bc6b969058b2620a5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: crop, livestock, forest, and fish allocation far outperforms the method of using rural population to\ndistribute the agricultural GDP, at least in the case of Brazil.\n\n#### References\n\n\nS. Angel, P. Lamson‐Hall, B. Guerra, Y. Liu, N. Galarza, and A. M. Blei (2018). Our Not‐So‐Urban World,\nWorking Paper No. 42 The Marron Institute of Urban Management, New York University.\n\n\nArvis, J. F., & Shepherd, B. (2013). The Poisson quasi‐maximum likelihood estimator: a solution to the\n‘adding up’ problem in gravity models. _Applied Economics Letters_, _20_ (6), 515‐519.\n\n\nArvis, J.F; Blankespoor, B. Downscaling Economic Variables: An Entropy Estimation. Mimeo.\n\n\nBlanchard, P.B.; Blankespoor, B. Rivera‐Fuentes, J.; Gunasekera, R.; Ishizawa O. and Jimenez L.F. Spatial\nDisaggregation of Gross Domestic Product for applications in Disaster Risk Management. Mimeo.\n\n\nBosker, M.; Park, J.; Roberts, M. (2018). Definition Matters : Metropolitan Areas and Agglomeration\nEconomies in a Large Developing Country (English). Policy Research working paper; no. WPS 8641.\nWashington, D.C. : World Bank Group.\n\n\nCenter for International Earth Science Information Network ‐ CIESIN ‐ Columbia University. (2017).\nGridded Population of the World, Version 4 (GPWv4): Population Count Adjusted to Match 2015\nRevision of UN WPP Country Totals, Revision 10. Palisades, NY: NASA Socioeconomic Data and\nApplications Center (SEDAC). https://doi.org/10.7927/H4JQ0XZW.\n\n\nChen, X., &\n\nMention: \"UN WPP Country Totals\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"The gridded counts have been adjusted to match these national population totals.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000483:9:0", "occurrence_id": "c5e463066b83968b972ca8ba", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: trend was quickly reversed by the massive shock brought about by Covid-19 and the policy\n\nresponse (credit incentives) to mitigate the effects of the pandemic. All in all, the inflation rate has\n\nremained in double-digit territory since 2017.\n\n##### **4. Data and Methodology**\n\n\n**_Data_**\n\n\nThe analysis relies on two main sources of data to construct the expenditure-decile adjusted\n\ninflation rates. The first source is the Turkish Household Budget Survey (HBS), a nationally\n\nrepresentative household survey administered by the Turkish Statistical Institute (TUIK) on a\n\nyearly basis. The second source is the Survey on Income and Living Conditions (SILC), also\n\ncollected annually, which is used specifically to examine the evolution of real income and real\n\nminimum wages using the inflation experience of each decile. <sup>6</sup> [^6: For this part of the analysis, income deciles are created in the SILC data. They are only slightly different from the consumption\ndeciles generated using the HBS data due to a strong correlation between income data in the SILC and consumption data in the\nHBS.]\n\n\nThe HBS collects detailed information on household living conditions, expenditures and\n\nincomes. While the exact sample size varies from year to year, since 2009 TUIK has been\n\nsurveying around 1,300 households monthly for a total of around 15,600 per year for each of the\n\nsurveys. Households are asked to report their monthly expenditures corresponding to the month\n\npreceding the survey date. The expenditure data is grouped into 12 main categories according the\n\nClassification of Individual Consumption According to Purpose (COICOP) <sup>7</sup> [^7: COICOP is a reference classification published by the United Nations Statistics Division.] as follows: 1) food\n\nand non-alcoholic beverages, 2) alcoholic beverages and tobacco, 3) clothing and footwear, 4)\n\nhousing, water, electricity, gas and other fuels, 5) furnishings, household equipment and routine\n\nmaintenance of the dwelling, 6) health,\n\nMention: \"Turkish Household Budget Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Detailed expenditure shares by household are used to build decile-adjusted inflation rates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000483:9:0", "occurrence_id": "1030d7eac1c0b61385e5fd6d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: trend was quickly reversed by the massive shock brought about by Covid-19 and the policy\n\nresponse (credit incentives) to mitigate the effects of the pandemic. All in all, the inflation rate has\n\nremained in double-digit territory since 2017.\n\n##### **4. Data and Methodology**\n\n\n**_Data_**\n\n\nThe analysis relies on two main sources of data to construct the expenditure-decile adjusted\n\ninflation rates. The first source is the Turkish Household Budget Survey (HBS), a nationally\n\nrepresentative household survey administered by the Turkish Statistical Institute (TUIK) on a\n\nyearly basis. The second source is the Survey on Income and Living Conditions (SILC), also\n\ncollected annually, which is used specifically to examine the evolution of real income and real\n\nminimum wages using the inflation experience of each decile. <sup>6</sup> [^6: For this part of the analysis, income deciles are created in the SILC data. They are only slightly different from the consumption\ndeciles generated using the HBS data due to a strong correlation between income data in the SILC and consumption data in the\nHBS.]\n\n\nThe HBS collects detailed information on household living conditions, expenditures and\n\nincomes. While the exact sample size varies from year to year, since 2009 TUIK has been\n\nsurveying around 1,300 households monthly for a total of around 15,600 per year for each of the\n\nsurveys. Households are asked to report their monthly expenditures corresponding to the month\n\npreceding the survey date. The expenditure data is grouped into 12 main categories according the\n\nClassification of Individual Consumption According to Purpose (COICOP) <sup>7</sup> [^7: COICOP is a reference classification published by the United Nations Statistics Division.] as follows: 1) food\n\nand non-alcoholic beverages, 2) alcoholic beverages and tobacco, 3) clothing and footwear, 4)\n\nhousing, water, electricity, gas and other fuels, 5) furnishings, household equipment and routine\n\nmaintenance of the dwelling, 6) health,\n\nMention: \"Survey on Income and Living Conditions\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Income deciles from this survey are used to trace real income and real minimum wage evolution.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000483:9:0", "occurrence_id": "773e06d651711b6366494426", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: trend was quickly reversed by the massive shock brought about by Covid-19 and the policy\n\nresponse (credit incentives) to mitigate the effects of the pandemic. All in all, the inflation rate has\n\nremained in double-digit territory since 2017.\n\n##### **4. Data and Methodology**\n\n\n**_Data_**\n\n\nThe analysis relies on two main sources of data to construct the expenditure-decile adjusted\n\ninflation rates. The first source is the Turkish Household Budget Survey (HBS), a nationally\n\nrepresentative household survey administered by the Turkish Statistical Institute (TUIK) on a\n\nyearly basis. The second source is the Survey on Income and Living Conditions (SILC), also\n\ncollected annually, which is used specifically to examine the evolution of real income and real\n\nminimum wages using the inflation experience of each decile. <sup>6</sup> [^6: For this part of the analysis, income deciles are created in the SILC data. They are only slightly different from the consumption\ndeciles generated using the HBS data due to a strong correlation between income data in the SILC and consumption data in the\nHBS.]\n\n\nThe HBS collects detailed information on household living conditions, expenditures and\n\nincomes. While the exact sample size varies from year to year, since 2009 TUIK has been\n\nsurveying around 1,300 households monthly for a total of around 15,600 per year for each of the\n\nsurveys. Households are asked to report their monthly expenditures corresponding to the month\n\npreceding the survey date. The expenditure data is grouped into 12 main categories according the\n\nClassification of Individual Consumption According to Purpose (COICOP) <sup>7</sup> [^7: COICOP is a reference classification published by the United Nations Statistics Division.] as follows: 1) food\n\nand non-alcoholic beverages, 2) alcoholic beverages and tobacco, 3) clothing and footwear, 4)\n\nhousing, water, electricity, gas and other fuels, 5) furnishings, household equipment and routine\n\nmaintenance of the dwelling, 6) health,\n\nMention: \"HBS data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Consumption deciles from this survey are noted to differ only slightly from income deciles.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000484:11:2", "occurrence_id": "de5737b82b559ca2953d0b57", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: </u>\n\n**<u><mark>Average # of Adults Interviewed</mark></u>** **<u><mark>1.89</mark></u>**\n\nWe calculate weights to correct for non-response in the IHPS by running a logistic regression of\nindividual response status among adults eligible for individual interviews. The results from the\nlogistic regression are presented in the Appendix Table A1. <sup>20</sup> Subsequently, we (1) take the\n\n\n19 Among the eligible men who did not respond, 44 percent were heads of household and 35 percent were children of\nthe household head, and among the non-responding women, only 6 percent were household heads, 32 percent were\nspouses of the head, and 37 percent were children.\n20 Van den Broeck and Kilic (2019) use a similar approach to correct for attrition bias in panel data samples, in a study\nof labor market dynamics in Sub-Saharan Africa. Our right-hand-side predictors of “response” include (i) fixed effects\nfor districts, interview months and enumerators, (ii) individual covariates, including age; a dichotomous variable\nidentifying females; a series of dichotomous variables on educational attainment; dichotomous variables identifying\nwhether the individual is currently married, and separately, whether he/she is head/spouse of head; and individual’s\n\n\n10\n\nMention: \"IHPS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Non-response weights are built from a logistic model of adult response status within this panel.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000486:34:0", "occurrence_id": "fae57d17dc7eb071256f317d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: We emphasize that our results also hold when assuming that utility is _strictly_\nincreasing in relative income, instead of _weakly_ increasing. Indeed, Theorem 1\nwould be unchanged if the subset _U_ <sup>_σ_</sup> <sup>¯</sup> excludes _u_ <sup>0</sup> . <sup>41</sup>\n\n#### **6 Empirical Application**\n\n\nIn this section we apply our proposed index, the hierarchical headcount ratio\n( _HHS_ ), to measure global poverty. The goal of this exercise is to illustrate how\nthe switch of index can affect the assessment of poverty. We compare the\nevaluation of poverty according to the hierarchical headcount ratio both with the\nsocietal headcount ratio ( _HS_ ) using the same global poverty line, and with the\nabsolute headcount ratio ( _HA_ ), arguably the most well-known poverty measure.\nWe compare poverty at the country level as well as the evolution and\ndistribution of global poverty. Besides the theoretical arguments for moving\naway from the societal headcount ratio, ultimately, such a change of index would\nbe justified only if it significantly affects the measurement of global poverty.\n\n###### **6.1 Data and Parameters**\n\n\nOur source of data is PovcalNet, <sup>42</sup> the on-line tool for poverty measurement\ndeveloped by the World Bank, which offers income or consumption data from\nmore than 1500 household surveys across more than 160 countries. In order to\nallow for cross-country comparisons, the World Bank converts income or\nconsumption using the 2011 PPP exchange rates for household consumption\nfrom the International Comparison Program. Moreover, for the purpose of\nperforming multi-country aggregations the World Bank defines reference years\naligning survey estimates from different years. To analyze poverty at the global\nor regional level, we only\n\nMention: \"PovcalNet\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Income and consumption distributions from over 1,500 surveys in more than 160 countries are drawn from this tool.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000486:34:0", "occurrence_id": "437378c788ca975b098c6d3a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: We emphasize that our results also hold when assuming that utility is _strictly_\nincreasing in relative income, instead of _weakly_ increasing. Indeed, Theorem 1\nwould be unchanged if the subset _U_ <sup>_σ_</sup> <sup>¯</sup> excludes _u_ <sup>0</sup> . <sup>41</sup>\n\n#### **6 Empirical Application**\n\n\nIn this section we apply our proposed index, the hierarchical headcount ratio\n( _HHS_ ), to measure global poverty. The goal of this exercise is to illustrate how\nthe switch of index can affect the assessment of poverty. We compare the\nevaluation of poverty according to the hierarchical headcount ratio both with the\nsocietal headcount ratio ( _HS_ ) using the same global poverty line, and with the\nabsolute headcount ratio ( _HA_ ), arguably the most well-known poverty measure.\nWe compare poverty at the country level as well as the evolution and\ndistribution of global poverty. Besides the theoretical arguments for moving\naway from the societal headcount ratio, ultimately, such a change of index would\nbe justified only if it significantly affects the measurement of global poverty.\n\n###### **6.1 Data and Parameters**\n\n\nOur source of data is PovcalNet, <sup>42</sup> the on-line tool for poverty measurement\ndeveloped by the World Bank, which offers income or consumption data from\nmore than 1500 household surveys across more than 160 countries. In order to\nallow for cross-country comparisons, the World Bank converts income or\nconsumption using the 2011 PPP exchange rates for household consumption\nfrom the International Comparison Program. Moreover, for the purpose of\nperforming multi-country aggregations the World Bank defines reference years\naligning survey estimates from different years. To analyze poverty at the global\nor regional level, we only\n\nMention: \"International Comparison Program\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its purchasing power exchange rates convert local incomes so countries can be compared.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000487:51:0", "occurrence_id": "080f002f8e332d2202405d94", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: In other words, to convert PovcalNet values to 2011 PPPs, we convert the 2005 PPP USD values\n\n\nto local currency units for 2005, then apply the CPI (available in PovcalNet) between 2005 and\n\n\n2011 to convert LCU values from 2005 to 2011 prices, and then finally apply the 2011 PPP\n\n\nconversion factors to obtain values in 2011 PPP USD. <sup>35</sup> [^35: When the CPI series in PovcalNet between 2005 and 2011 is incomplete, we use the CPI from WDI to\nextend the series. A previous version of this paper relied solely on the CPIs available in WDI.]\n\n\n**_C. Removing rural-urban adjustments for India, China and Indonesia_**\n\n\nIncome distributions derived in 2005 PPPs from PovcalNet contain implicit adjustments for\n\n\nseveral countries, including China, India and Indonesia, as described in Chen and Ravallion\n\n\n(2010). The PPP conversion factors used for these three countries are described in Table A1.2.\n\n\n**Table A1.2: PovcalNet 2005 PPP conversion factors for urban bias adjustment**\n\n\n\n**<u>Adjusted</u>**\n\n**Urban PPP**\n\n\n\n**<u>Adjusted</u>**\n\n**Rural PPP**\n\n\n\n**<u>Ratio Adjusted</u>**\n\n**Urban/Official**\n\n\n\n**<u>Ratio Adjusted</u>**\n\n**Rural/Official**\n\n\n\n**Country**\n\n\n\n**<u>Official</u>**\n\n**PPP**\n\n\n\nChina 4.09 4.09 2.98 1.00 0.73\n\nIndia 15.6 17.24 11.4 1.11 0.73\n\nIndonesia 4192 4795.9 3399.7 1.14 0.81\n\n\nFor these three countries, we undo the adjustments when applying the 2011 PPPs by adding an\n\n\nadditional factor of the ratio of the adjusted to official PPP (two last columns of Table A1.2) to\n\n\nequation\n\nMention: \"CPIs available in WDI\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"These price series extend incomplete consumer price series when converting between purchasing power vintages.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000491:8:0", "occurrence_id": "66bc5ddeff0f7d7ac9245b0f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: provides bilateral gross and value added trade flows for 40 countries and 35 sectors during\n\nthe period 1995-2011. Tables A-1 and A-2 in the Appendix list the countries and sectors\n\nincluded in the WIOD dataset which comprises the country and sector coverage of our\n\nestimation sample.\n\n\nOur outcome variables consist of indicators capturing relevant dimensions of the trade\n\npolicy regime involving _i_ and _j_ as potential signatories of one or more PTAs active at\n\ntime _t_ . We start our analysis using a synthetic measure of the depth of PTAs sourced\n\nfrom the DESTA database. This corresponds to the continuous indicator constructed\n\nby Dür et al. (2014) through latent trait analysis of almost 50 specific variables that\n\nare theoretically related to the depth of an agreement. We rescale the original indicator\n\nto have its minimum value over the distribution of all agreements recorded in DESTA\n\nequal to 0. Due to the country and time coverage of our econometric exercise (see below)\n\nthe shallowest agreement considered in our estimation sample has a value of the rescaled\n\nDESTA indicator that is strictly bigger than 0. For each country pair _ij_ at each point\n\nin time _t_, we define Depth _ijt_ as the max between 0 and the value of the rescaled DESTA\n\ncontinuous indicator of the most recent PTAs whose signatories include countries _i_ and _j_\n\nand that is active at time _t_ .\n\n\nThe resulting set of specifications used to study the effects of trade and GVCs on PTA\n\ndepth is given by\n\n\n(1)\nDepth _ijt_ = _βXijtz_ + _γitz_ + _γjtz_ + _γijz_ + _εijtz_\n\nfor each variable _Xijtz_ among Total exports _ijtz_ ; Domestic value _ijtz_ ; Foreign value - VS _ijtz_ }.\nThe terms _γitz_, _γ\n\nMention: \"WIOD dataset\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Bilateral gross and value added trade flows across countries and sectors define the estimation sample.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000491:8:0", "occurrence_id": "b25fb3e17f85f36cf8571f6a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: provides bilateral gross and value added trade flows for 40 countries and 35 sectors during\n\nthe period 1995-2011. Tables A-1 and A-2 in the Appendix list the countries and sectors\n\nincluded in the WIOD dataset which comprises the country and sector coverage of our\n\nestimation sample.\n\n\nOur outcome variables consist of indicators capturing relevant dimensions of the trade\n\npolicy regime involving _i_ and _j_ as potential signatories of one or more PTAs active at\n\ntime _t_ . We start our analysis using a synthetic measure of the depth of PTAs sourced\n\nfrom the DESTA database. This corresponds to the continuous indicator constructed\n\nby Dür et al. (2014) through latent trait analysis of almost 50 specific variables that\n\nare theoretically related to the depth of an agreement. We rescale the original indicator\n\nto have its minimum value over the distribution of all agreements recorded in DESTA\n\nequal to 0. Due to the country and time coverage of our econometric exercise (see below)\n\nthe shallowest agreement considered in our estimation sample has a value of the rescaled\n\nDESTA indicator that is strictly bigger than 0. For each country pair _ij_ at each point\n\nin time _t_, we define Depth _ijt_ as the max between 0 and the value of the rescaled DESTA\n\ncontinuous indicator of the most recent PTAs whose signatories include countries _i_ and _j_\n\nand that is active at time _t_ .\n\n\nThe resulting set of specifications used to study the effects of trade and GVCs on PTA\n\ndepth is given by\n\n\n(1)\nDepth _ijt_ = _βXijtz_ + _γitz_ + _γjtz_ + _γijz_ + _εijtz_\n\nfor each variable _Xijtz_ among Total exports _ijtz_ ; Domestic value _ijtz_ ; Foreign value - VS _ijtz_ }.\nThe terms _γitz_, _γ\n\nMention: \"DESTA database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"A latent-trait measure of agreement depth is rescaled from this database to characterise trade regimes.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000498:5:2", "occurrence_id": "6e9c191a0fa2f02a0f1d2967", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: sup> ; and an average of 5.0% over 2011‐2016 in the IMF WEO, April 2011\nedition <sup>4</sup> - see Graph 1(b). In the following subsection we look at pre‐conflict trends and projections\n\n\n1 WDI data for Syria are publicly available only up to 2007. We supplement these data for the last three years up to 2010\nwith data on GDP growth from the World Bank’s internal macroeconomic model (MFMod), November 2017 vintage, which\nis consistent with the IMF WEO October 2018 data.\n2 WDI: GDP is at market prices, 2010 USD. IMF WEO: GDP is at market prices, 2000 local currency.\n3 Source: http://documents.worldbank.org/curated/en/293851468162862428/Global‐economic‐prospects‐January‐2011‐\n<u>navigating‐strong‐currents. GDP growth of 5.0% for 2010 is an estimate. No GEP projections for Syria are available beyond</u>\nthe January 2011 publication. GDP is at market prices, 2005 USD.\n4 Source: https://www.imf.org/external/pubs/ft/weo/2011/01/weodata/index.aspx.\n\n\n4\n\nMention: \"WDI data for Syria\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Public coverage stops in 2007, so later years must be supplemented from an internal model.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000498:5:2", "occurrence_id": "fb57aeb08a5d749892a74a97", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: sup> ; and an average of 5.0% over 2011‐2016 in the IMF WEO, April 2011\nedition <sup>4</sup> - see Graph 1(b). In the following subsection we look at pre‐conflict trends and projections\n\n\n1 WDI data for Syria are publicly available only up to 2007. We supplement these data for the last three years up to 2010\nwith data on GDP growth from the World Bank’s internal macroeconomic model (MFMod), November 2017 vintage, which\nis consistent with the IMF WEO October 2018 data.\n2 WDI: GDP is at market prices, 2010 USD. IMF WEO: GDP is at market prices, 2000 local currency.\n3 Source: http://documents.worldbank.org/curated/en/293851468162862428/Global‐economic‐prospects‐January‐2011‐\n<u>navigating‐strong‐currents. GDP growth of 5.0% for 2010 is an estimate. No GEP projections for Syria are available beyond</u>\nthe January 2011 publication. GDP is at market prices, 2005 USD.\n4 Source: https://www.imf.org/external/pubs/ft/weo/2011/01/weodata/index.aspx.\n\n\n4\n\nMention: \"IMF WEO October 2018 data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"The internal macroeconomic model's estimates are checked for consistency with this vintage.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000500:12:1", "occurrence_id": "43fb0e59683223a8f14d9ae6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  calculations, World Development Indicators.\nA. New LICs include countries that were not LICs in 2001: Syria and South Sudan. South Sudan gained independence in 2011.\nC. Sample includes 59 2001 LICs. GNI per capita according to the World Bank Atlas method. GNI per capita for 1999 was published in August\n2000 and reflects the original data used for country income classification in the 2001 World Bank fiscal year, while GNI per capita for 2018\nwas published in July 2019 and reflects the data used for the 2020 World Bank fiscal year. Exceptions are Liberia and Myanmar, for which GNI\nper capita in 2002 is used as a proxy for 1999. Rapid-growing LICs with low starting points are defined as LICs that had per capita incomes\nbelow one-third of the $755 LIC threshold in 2001, and these incomes have increased to above two-thirds of the $1025 LIC threshold in 2020.\nD. Sample includes 13 LICs that became MICs due to data limitations. Year turned MIC reflects the World Bank fiscal year.\nE.-F. WGI index scores are standard normal units that range between -2.5 and 2.5, with zero mean. A negative score implies government\neffectiveness or rule-of-law below the global average. Sample includes all 32 LICs that became MICs. Year turned MIC reflects the World Bank\nfiscal year.\n\n**_New LICs: 2016-20_**\n\n\nCountries that have reached MIC status often face structural challenges that constrain their\nprospects for continued strong growth, while those MICs with incomes near the LIC threshold risk\nfalling back into the LIC bracket (EBRD 2019). Three of today’s LICs fell back from middleincome levels since 2016: the Syrian Arab Republic, Tajikistan, and Yemen. While the Syrian\nArab Republic was classified as a LIC for the first time, Tajikistan, and Yemen relapsed into LIC\nstatus after some of the growth\n\nMention: \"GNI per capita for 1999\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"This vintage determined the income classification applied at the start of the sample.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000500:12:1", "occurrence_id": "b4d45f2bb9d713a0ef31ad18", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  calculations, World Development Indicators.\nA. New LICs include countries that were not LICs in 2001: Syria and South Sudan. South Sudan gained independence in 2011.\nC. Sample includes 59 2001 LICs. GNI per capita according to the World Bank Atlas method. GNI per capita for 1999 was published in August\n2000 and reflects the original data used for country income classification in the 2001 World Bank fiscal year, while GNI per capita for 2018\nwas published in July 2019 and reflects the data used for the 2020 World Bank fiscal year. Exceptions are Liberia and Myanmar, for which GNI\nper capita in 2002 is used as a proxy for 1999. Rapid-growing LICs with low starting points are defined as LICs that had per capita incomes\nbelow one-third of the $755 LIC threshold in 2001, and these incomes have increased to above two-thirds of the $1025 LIC threshold in 2020.\nD. Sample includes 13 LICs that became MICs due to data limitations. Year turned MIC reflects the World Bank fiscal year.\nE.-F. WGI index scores are standard normal units that range between -2.5 and 2.5, with zero mean. A negative score implies government\neffectiveness or rule-of-law below the global average. Sample includes all 32 LICs that became MICs. Year turned MIC reflects the World Bank\nfiscal year.\n\n**_New LICs: 2016-20_**\n\n\nCountries that have reached MIC status often face structural challenges that constrain their\nprospects for continued strong growth, while those MICs with incomes near the LIC threshold risk\nfalling back into the LIC bracket (EBRD 2019). Three of today’s LICs fell back from middleincome levels since 2016: the Syrian Arab Republic, Tajikistan, and Yemen. While the Syrian\nArab Republic was classified as a LIC for the first time, Tajikistan, and Yemen relapsed into LIC\nstatus after some of the growth\n\nMention: \"GNI per capita for 2018\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The end-of-period classification uses this July 2019 publication.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000507:38:1", "occurrence_id": "c2d5cf3c26dbb4b3e6137cd2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  and the vignettes; and in column (5) is the standardized index of the seven items that are reported by the caregiver\nfrom the internalizing/externalizing behaviors sections of the Parent/Caregiver Report Survey. The controls include age in years, girls and\nboys cared for by the respondent, and the educational level of the respondent for primary, high school, or tertiary education—the\nomitted category. Robust standard errors. *** p _<_ 0.01, ** p _<_ 0.05, * p _<_ 0.1.\n\n\n37\n\nMention: \"Parent/Caregiver Report Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"A standardised internalising and externalising behaviour index is built from this instrument's items.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000509:14:0", "occurrence_id": "7660d3642b0998f18084a90d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: divided by the number of employees, expressed in constant US dollars. We also experimented\n\nwith using total factor productivity (TFP) as a measure of productivity, which is often viewed\n\nas a cleaner measure of efficiency. However, the Enterprise Survey does not have data on\n\ncapital stock for a large share of firms (including all those in service sectors). To preserve\n\nsample size, and with the knowledge that the results based on TFP and labor productivity tend\n\nto be similar, we rely on labor productivity to measure firm productivity. The specification we\n\n\n\nuse is as follows:\n\nY𝑖 **𝑖** 𝑖 **𝑖** 𝑖𝑖 = ES𝑐 **𝑐** α1 + DB𝑘𝑘α2 + GC𝑘𝑘α3 + X𝑖𝑖β1 + W𝑘 **𝑘** β2 + θ𝑗𝑗 + η𝑡𝑡 + ϵ𝑗 **𝑗** 𝑗 **𝑗** 𝑗𝑗. (1)\n\nHere, Y𝑗 **𝑗** 𝑗 **𝑗** 𝑗𝑗 is firm performance. When the outcome is employment growth, the unit is in\n\n**𝑐**\n\n\n\nuse is as follows:\n\n**𝑖** **𝑖** **𝑐** **𝑘** **𝑗** **𝑗**\n\n\n\nY𝑖 **𝑖** 𝑖 **𝑖** 𝑖𝑖 = ES𝑐 **𝑐** α1 + DB𝑘𝑘α2 + GC𝑘𝑘α3 + X𝑖𝑖β1 + W𝑘 **𝑘** β2 + θ𝑗𝑗 + η𝑡𝑡 + ϵ𝑗 **𝑗** 𝑗 **𝑗** 𝑗𝑗. (1)\n\npercentage points (that is, one means one percentage point). Subscript _i_ stands for firm; _c_ for\n\nHere, Y𝑗 **𝑗** 𝑗 **𝑗** 𝑗𝑗\n\nMention: \"Enterprise Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"It lacks capital stock for many firms including services, so labour productivity is used instead of total factor productivity.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000516:9:0", "occurrence_id": "b8715546bba3a09f0b40b884", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Source:_ MEC/INEP/Censo Escolar (2015) for Brazil, DINIEE (2016) for Argentina, and INEGI (2017) for Mexico.\n_Note:_ Dropout rate is defined as the percentage of students who were enrolled in school at year _t_ and do not enroll in year _t+1_ .\n\nWhile free public secondary and tertiary education is available in the three countries analyzed,\nrelatively few youth ages 18-25 are enrolled in school, and many are simultaneously working.\nFigures 2(a) through 2(c) report youth activity status (enrolled in school, employed, enrolled and\nemployed, neither employed nor enrolled) by age for Argentina, Brazil and Mexico. In each\ncountry, there is a notable shift away from schooling and towards employment as age increases\nfrom 15 to 25. As shown by the school enrollment rates above, there is considerable crosscountry heterogeneity in the distribution of school enrollment across age, with Argentinian\nstudents staying in school until older. In Brazil and Argentina, there is a larger overlap between\nschool and work, as many employed young adults remain enrolled in school. This is less\ncommon in Mexico.\n\nAfter the end of secondary school, the share of youth who are neither in school nor working (the\nso-called _ninis_ ) remains relatively stable at around 20-22% after age 18 in Argentina and Brazil\nand about 30% in Mexico. <sup>9</sup> [^9: See De Hoyos et al. (2016) for a regional study of _ninis_ in Latin America.] While many _ninis_ are out of the labor force, high youth\nunemployment rates in Argentina and Brazil account for some of the _ninis_ . <sup>10</sup> [^10: In the three countries a person is considered unemployed if that person reported to be not employed but available\nto work and actively looking for a job during the week prior to the interview.] The youth\nunemployment rate, as with the overall unemployment rate, is substantially lower in Mexico\n(\n\nMention: \"Censo Escolar\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Brazilian enrolment records underpin the school dropout rate shown for that country.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000516:9:0", "occurrence_id": "e5ef4a206dc09951336550a5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Source:_ MEC/INEP/Censo Escolar (2015) for Brazil, DINIEE (2016) for Argentina, and INEGI (2017) for Mexico.\n_Note:_ Dropout rate is defined as the percentage of students who were enrolled in school at year _t_ and do not enroll in year _t+1_ .\n\nWhile free public secondary and tertiary education is available in the three countries analyzed,\nrelatively few youth ages 18-25 are enrolled in school, and many are simultaneously working.\nFigures 2(a) through 2(c) report youth activity status (enrolled in school, employed, enrolled and\nemployed, neither employed nor enrolled) by age for Argentina, Brazil and Mexico. In each\ncountry, there is a notable shift away from schooling and towards employment as age increases\nfrom 15 to 25. As shown by the school enrollment rates above, there is considerable crosscountry heterogeneity in the distribution of school enrollment across age, with Argentinian\nstudents staying in school until older. In Brazil and Argentina, there is a larger overlap between\nschool and work, as many employed young adults remain enrolled in school. This is less\ncommon in Mexico.\n\nAfter the end of secondary school, the share of youth who are neither in school nor working (the\nso-called _ninis_ ) remains relatively stable at around 20-22% after age 18 in Argentina and Brazil\nand about 30% in Mexico. <sup>9</sup> [^9: See De Hoyos et al. (2016) for a regional study of _ninis_ in Latin America.] While many _ninis_ are out of the labor force, high youth\nunemployment rates in Argentina and Brazil account for some of the _ninis_ . <sup>10</sup> [^10: In the three countries a person is considered unemployed if that person reported to be not employed but available\nto work and actively looking for a job during the week prior to the interview.] The youth\nunemployment rate, as with the overall unemployment rate, is substantially lower in Mexico\n(\n\nMention: \"DINIEE\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Argentine education statistics are taken from this information directory.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000511:26:1", "occurrence_id": "4bb940e424758a35f3a3d989", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Explaining Gender\n\n\nDifferentials in Agricultural Production in Nigeria.\" World Bank Policy Research Working Paper,\n\n\n(6809).\n\n\nOseni, Gbemisola and Paul Winters. 2009. \"Rural Nonfarm Activities and Agricultural Crop\n\n\nProduction in Nigeria.\" Agricultural Economics, 40(2), 189-201.\n\n\nPfeiffer, Lisa; Alejandro López-Feldman and J Edward Taylor. 2009. \"Is Off-Farm Income Reform\n\ning the Farm? Evidence from Mexico.\" Agricultural Economics, 40(2), 125-38.\n\n\nRadchenko, Natalia. 2014. \"Heterogeneity in Informal Salaried Employment: Evidence from the\n\n\nEgyptian Labor Market Survey.\" World Development, 62, 169-88.\n\n\nReardon, T. (1997). Using evidence of household income diversification to inform study of the rural\n\n\nnonfarm labor market in Africa. World Development, 25(5), 735-747. doi:http://dx.doi.org/10.1016/S0305\n\n750X(96)00137-4\n\n\nReardon, Thomas; Julio Berdegué; Christopher B Barrett and Kostas Stamoulis. 2007. \"Household\n\n\nIncome Diversification into Rural Nonfarm Activities.\" Transforming the rural non-farm economy:\n\n\nopportunities and threats in the developing world, 115-41.\n\n\nReuters, “Nigeria Floods Kill 137, Displace Thousands,” 2012.\n\n\n25\n\nMention: \"Egyptian Labor Market Survey\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"Appears in a reference title as the evidence base for a study of informal salaried employment.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000517:6:0", "occurrence_id": "ff83a3fe0d921ca6ea3b289b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: equal to the 2010 GDP per capita from the World Development Indicators. It should be\n\n\nnoted that this is only an approximation to more proper international comparisons (see e.g.\n\n\nDeaton and Dupriez, 2011) and that the HIT database is not the World Bank Group’s official\n\n\npoverty data, which can be found in PovcalNet. <sup>2</sup>\n\n\nFigure 4 pools all the data and shows how the aggregate spending categories identified\n\n\nin our data vary with (the log of) household income per capita (in constant 2010 USD).\n\n\nTo start, households spend a large share of their income on food and agricultural products.\n\n\nAcross all countries, the average household in our sample spends 44.7% of its income on\n\n\nfood items, 17.4% on manufacturing goods, and 15.1% on services. Another 16.9% of\n\n\nexpenditure is accounted for by goods households have produced themselves, which highlights\n\n\nthe importance of dealing with home consumption in the analysis. This also implies that the\n\n\ntotal expenditure share in agriculture is 61.6%. While there is huge heterogeneity in spending\n\n\npatterns both across and within countries, the graph shows that, as households get richer, the\n\n\nshare of income spent on food decreases, especially for the richest households. Spending on\n\n\nmanufacturing goods, services, and other goods first declines with income but then increases\n\n\nsharply. The opposite happens with home consumption. The implication is that the tariff\n\n\nburden on different households will vary as a result of these different consumption patterns.\n\n\nAgriculture is the most important source of income, accounting for 38.5% of total\n\n\nhousehold income on average. This total comprises a 20.9% share of production for household\n\n\nconsumption and 17.6% of sales of agricultural products in the market. Labor income\n\n\nrepresents 29.6% of household income, on average, and non-farm enterprises account for\n\n\nanother 12.7%. Figure 5 shows how these aggregate\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"GDP per capita from this series anchors the income scaling used in the figures.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000517:6:0", "occurrence_id": "0290c3989c71c487d758d08b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: equal to the 2010 GDP per capita from the World Development Indicators. It should be\n\n\nnoted that this is only an approximation to more proper international comparisons (see e.g.\n\n\nDeaton and Dupriez, 2011) and that the HIT database is not the World Bank Group’s official\n\n\npoverty data, which can be found in PovcalNet. <sup>2</sup>\n\n\nFigure 4 pools all the data and shows how the aggregate spending categories identified\n\n\nin our data vary with (the log of) household income per capita (in constant 2010 USD).\n\n\nTo start, households spend a large share of their income on food and agricultural products.\n\n\nAcross all countries, the average household in our sample spends 44.7% of its income on\n\n\nfood items, 17.4% on manufacturing goods, and 15.1% on services. Another 16.9% of\n\n\nexpenditure is accounted for by goods households have produced themselves, which highlights\n\n\nthe importance of dealing with home consumption in the analysis. This also implies that the\n\n\ntotal expenditure share in agriculture is 61.6%. While there is huge heterogeneity in spending\n\n\npatterns both across and within countries, the graph shows that, as households get richer, the\n\n\nshare of income spent on food decreases, especially for the richest households. Spending on\n\n\nmanufacturing goods, services, and other goods first declines with income but then increases\n\n\nsharply. The opposite happens with home consumption. The implication is that the tariff\n\n\nburden on different households will vary as a result of these different consumption patterns.\n\n\nAgriculture is the most important source of income, accounting for 38.5% of total\n\n\nhousehold income on average. This total comprises a 20.9% share of production for household\n\n\nconsumption and 17.6% of sales of agricultural products in the market. Labor income\n\n\nrepresents 29.6% of household income, on average, and non-farm enterprises account for\n\n\nanother 12.7%. Figure 5 shows how these aggregate\n\nMention: \"HIT database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Household spending categories are analysed on this dataset, which is explicitly not official poverty data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000517:6:0", "occurrence_id": "1b1e6c4956066fa525ceb4e6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: equal to the 2010 GDP per capita from the World Development Indicators. It should be\n\n\nnoted that this is only an approximation to more proper international comparisons (see e.g.\n\n\nDeaton and Dupriez, 2011) and that the HIT database is not the World Bank Group’s official\n\n\npoverty data, which can be found in PovcalNet. <sup>2</sup>\n\n\nFigure 4 pools all the data and shows how the aggregate spending categories identified\n\n\nin our data vary with (the log of) household income per capita (in constant 2010 USD).\n\n\nTo start, households spend a large share of their income on food and agricultural products.\n\n\nAcross all countries, the average household in our sample spends 44.7% of its income on\n\n\nfood items, 17.4% on manufacturing goods, and 15.1% on services. Another 16.9% of\n\n\nexpenditure is accounted for by goods households have produced themselves, which highlights\n\n\nthe importance of dealing with home consumption in the analysis. This also implies that the\n\n\ntotal expenditure share in agriculture is 61.6%. While there is huge heterogeneity in spending\n\n\npatterns both across and within countries, the graph shows that, as households get richer, the\n\n\nshare of income spent on food decreases, especially for the richest households. Spending on\n\n\nmanufacturing goods, services, and other goods first declines with income but then increases\n\n\nsharply. The opposite happens with home consumption. The implication is that the tariff\n\n\nburden on different households will vary as a result of these different consumption patterns.\n\n\nAgriculture is the most important source of income, accounting for 38.5% of total\n\n\nhousehold income on average. This total comprises a 20.9% share of production for household\n\n\nconsumption and 17.6% of sales of agricultural products in the market. Labor income\n\n\nrepresents 29.6% of household income, on average, and non-farm enterprises account for\n\n\nanother 12.7%. Figure 5 shows how these aggregate\n\nMention: \"PovcalNet\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"Named only as the location of the official poverty data this dataset is not.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000520:46:0", "occurrence_id": "d8d2c1c7d89522c93b74e259", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure A3. Education gaps in rate of work stoppage by country\n\n\nSource: Authors’ calculations based on HFPS.\nNotes: Dark (light) colors indicate that the difference between groups is (not) statistically significant at 5% level or\n\nless.\n\n\nFigure A4. Location gaps in rate of work stoppage by country\n\n\nSource: Authors’ calculations based on HFPS.\nNotes: Dark (light) colors indicate that the difference between groups is (not) statistically significant at 5% level or\n\nless.\n\n\n**3.** **Understanding the gender gap in the rate of work stoppage**\n\n\nThe between-group difference in the rate of work stoppage was larger when grouping by gender.\n\nSpecifically, 34% of women stopped working since the pandemic started while the rate for men\n\nwas 27%. Three possible reasons behind this result are the differential incidence of childcare\n\n\n45\n\nMention: \"HFPS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Education gaps in work stoppage rates by country are calculated from this household pulse survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000521:13:0", "occurrence_id": "b0c1863961b9b8d1b9992bec", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Table 1. Mobile Internet Connectivity Data for Individuals, 2018-19**\n\n\n\n**<u>Individuals connected and not connected to internet through their cell phones</u>**\n\n\n\n**<u>Senegal</u>**\n\n\n\n**<u>Distribution Groups</u>** **<u>Connected (%)</u>** **<u>Not Connected (%)</u>**\n**<u>Income Threshold</u>**\n\n\n\nAbove 52 48\nBelow 21 79\nUrban 49 51\nRural 17 83\nMen 40 60\nWomen 30 70\n**Age**\n\n15 to 24 years old 37 63\n25 to 40 years old 45 54\n41 and older 21 79\n**Education**\nLess than primary 20 80\nPrimary and secondary 48 52\nTertiary 86 14\nRead and write French 54 46\nDo not read and write\n\nFrench and illiterate 19 81\n**Sector**\nAgriculture 13 87\nIndustry 45 54\nServices 46 54\nUnemployed/Inactive 31 68\n**Household assets**\nComputer ownership 65 35\nT.V. ownership 47 53\nTablet ownership 62 38\n**Complementary infrastructure**\nAccess to electricity 47 53\n\n\n\n**<u>Total population</u>** <u>34</u> <u>66</u>\n\n_Source_ : EHCVM 2018/2019 survey.\n_Note:_ Only individuals 15+. Only includes individuals that accessed the internet through their mobile phones. Each tabulation is carried out at the\nindividual level adjusting for household weights. The agriculture sector includes jobs in crops, fisheries and animal breeding. The industry sector includes\nextractive and other industries, and public works/construction jobs. The services sector includes commerce, restaurants/hotels, transportation,\ncommunication, education, health, other and personal services jobs. Secondary education is defined as individuals with less than tertiary but\n\nMention: \"Mobile Internet Connectivity Data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Connected and unconnected individuals by asset and employment group are tabulated from this data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000521:13:0", "occurrence_id": "0e849c0c682f8461d6ac1b46", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Table 1. Mobile Internet Connectivity Data for Individuals, 2018-19**\n\n\n\n**<u>Individuals connected and not connected to internet through their cell phones</u>**\n\n\n\n**<u>Senegal</u>**\n\n\n\n**<u>Distribution Groups</u>** **<u>Connected (%)</u>** **<u>Not Connected (%)</u>**\n**<u>Income Threshold</u>**\n\n\n\nAbove 52 48\nBelow 21 79\nUrban 49 51\nRural 17 83\nMen 40 60\nWomen 30 70\n**Age**\n\n15 to 24 years old 37 63\n25 to 40 years old 45 54\n41 and older 21 79\n**Education**\nLess than primary 20 80\nPrimary and secondary 48 52\nTertiary 86 14\nRead and write French 54 46\nDo not read and write\n\nFrench and illiterate 19 81\n**Sector**\nAgriculture 13 87\nIndustry 45 54\nServices 46 54\nUnemployed/Inactive 31 68\n**Household assets**\nComputer ownership 65 35\nT.V. ownership 47 53\nTablet ownership 62 38\n**Complementary infrastructure**\nAccess to electricity 47 53\n\n\n\n**<u>Total population</u>** <u>34</u> <u>66</u>\n\n_Source_ : EHCVM 2018/2019 survey.\n_Note:_ Only individuals 15+. Only includes individuals that accessed the internet through their mobile phones. Each tabulation is carried out at the\nindividual level adjusting for household weights. The agriculture sector includes jobs in crops, fisheries and animal breeding. The industry sector includes\nextractive and other industries, and public works/construction jobs. The services sector includes commerce, restaurants/hotels, transportation,\ncommunication, education, health, other and personal services jobs. Secondary education is defined as individuals with less than tertiary but\n\nMention: \"EHCVM 2018/2019 survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Household-weighted individual tabulations come from this agricultural household survey round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000522:12:0", "occurrence_id": "8dccca7a55f36519d340f539", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 2. Difference between adolescents and adults (adolescent coefficient, adults omitted**\n**reference category) in gender role attitudes, justification of domestic violence, decision-**\n\n\nSource: Authors’ calculations based on JLMPS 2016\nNotes: Bars indicate 95% confidence intervals. Domestic violence, decision making, and\nmobility are available for women only. Models include controls, see online supplemental\nappendix Table A12 and Table A13.\n\n\nWe next turn to the question of whether gender role attitudes differ by sex; we cannot test\nthis for justification of domestic violence or gendered behaviors since they are only available for\nwomen. We examine gender differences in gender role attitudes among Syrian and Jordanian\nadolescents, as well as among Syrian and Jordanian adults in Figure 3. There are significant\ndifferences for all groups. Coefficients range from 0.26 to 0.38, meaning female adolescents or\nadults have more equitable gender role attitudes than their male counterparts by approximately\none third of a standard deviation.\n\n\n11\n\nMention: \"JLMPS 2016\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Adolescent-adult differences in gender role attitudes are estimated from this youth and household panel.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000528:6:1", "occurrence_id": "ff9348236783dd7d923334be", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ; Salehyan and Gleditsch 2006). The surge’s\nupsetting of ethnic balances and the governing coalition explain Jordan’s subsequent economic reforms.\n\nThe last and best feature of this sudden natural experiment is that the Kuwaiti-Palestinians had full\nlegal, political, and economic rights immediately upon entering Jordan, a situation unique in the Arab\n\n\n\ngroups in the host country (Buhaug and Gleditsch 2008; Salehyan and Gleditsch 2006). The surge’s\nupsetting of ethnic balances and the governing coalition explain Jordan’s subsequent economic reforms.\n\nThe last and best feature of this sudden natural experiment is that the Kuwaiti-Palestinians had full\nlegal, political, and economic rights immediately upon entering Jordan, a situation unique in the Arab\nworld (Zureik 1994).\n\nThree features of this exogenous shock make it less useful as a natural experiment. The first is that the\nKuwaiti-Palestinians were overwhelmingly Sunni Muslims, just like Jordan’s population. The second is\n\n\n\nlegal, political, and economic rights immediately upon entering Jordan, a situation unique in the Arab\nworld (Zureik 1994).\n\nThree features of this exogenous shock make it less useful as a natural experiment. The first is that the\nKuwaiti-Palestinians were overwhelmingly Sunni Muslims, just like Jordan’s population. The second is\nthat Jordan did not have a democracy like other nations that accepted large numbers of immigrants.\nJordanians voted in the 1989 elections and expected future elections, but their impact on policy through\nvoting was limited. Third, Jordan did not have a welfare state as large as those in the developed world\n(Brynen 1992).\n\n**4.** This study uses **Data** Gwartney, Lawson, and Hall (2017) Economic Freedom of the World Annual Report (EFW)\n\nto measure changes in Jordan’s economic institutions. The EFW index is a reasonable proxy for economic\ninstitutions, incorporating 42 variables collected\n\nMention: \"Economic Freedom of the World Annual Report\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its composite of 42 variables is the proxy for changes in the country's economic institutions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000531:6:2", "occurrence_id": "cc8a62db2e61a0dbe2531453", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  from\na survey in Côte d’Ivoire from 2015. From 2015 to 2016 IMF data suggest that GDP per capita in Côte\nd’Ivoire grew by 5.6%. Figure 1 explores how this growth can be distributed if inequality stays unchanged\nor if the Gini increases or decreases by 1% (ignoring for the moment that only part of this growth is passed\nthrough to the consumption observed in surveys). The initial Gini in Côte d’Ivoire was 41.5, meaning that\na 1% drop (Lowering the Gini by 1% does not have to impose a large cost on the top of the distribution. Because of 𝛼𝛼= −0.01 ) would bring the Gini to 41.1, while a 1% increase ( 𝛼𝛼 =0.01) would bring it to 41.9.\n\n\n\nLowering the Gini by 1% does not have to impose a large cost on the top of the distribution. Because of\nthe large income share of the top of the distribution, the reduction in the growth rate of the wealthiest\nindividuals necessary to ensure that the bottom grows substantially faster than the mean is relatively\nsmall. For example, in the case of Côte d’Ivoire, a convex growth incidence such that the Gini decreases\nby 1% means that households at the 10 <sup>th</sup> percentile grow 2.5 pp faster than the mean, yet only reduces\nthe growth at the 90 <sup>th</sup> percentile by 0.5 pp.\n\n\n\n5\n\nMention: \"survey in Côte d’Ivoire\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Growth incidence is simulated by distributing observed growth across the survey's income distribution.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000534:9:0", "occurrence_id": "ea549e95d428e9a9fe407198", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _highly concentrated in a few producer countries or the main world supplier is China were_\n\n\n_more vulnerable to the pandemic._\n\n\nThe COVID-19 shock may have adversely affected productivity due to reduced mobility and\n\n\nthe disruption in in-person production activities but less so for more complex and knowledge\n\nintensive products. This is because, for such products, productivity relies more on knowledge\n\n\ncollaborations that were effectively shifted to virtual exchanges. The exponential growth of virtual\n\n\nonline interactions for research and innovation collaborations - critical for productivity - and the\n\n\nwider use of virtual platforms have proven to be highly effective as is illustrated by the surprisingly\n\n\nquick responses to the pandemic, namely the rapid development of vaccines (Paunov and Planes\n\nSatorra, 2021). Our third hypothesis to test in the empirical work is as follows:\n\n\n_H3: The productivity of complex and knowledge-intensive production processes may increase_\n\n\n_relative to that of less complex and knowledge-intensive processes since the former benefited_\n\n\n_more from remote work interactions. Hence less complex and knowledge- intensive products_\n\n\n_were more vulnerable to the pandemic._\n\n\n**2.** **Data**\n\n\n**High-frequency bilateral product level trade data**\n\n\nOur analysis relies on several sources of data. For trade outcomes, we use monthly data on\n\n\nimport flows at the product level by the major markets of the 27 EU countries, Japan, and the US\n\n\nfrom January 2018 to December 2020. The specific sources of data are, respectively, Eurostat\n\n\nmonthly trade flows for EU countries, the Ministry of Finance for Japan, and the United States\n\n\nInternational Trade Commission for the US. These data are exploited to examine the evolution of\n\n\nexport flows\n\n\n8\n\nMention: \"Eurostat\n\n\nmonthly trade flows\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Product-level monthly import flows for EU markets are taken from this source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000535:56:0", "occurrence_id": "30d86321782ccd6b6fc76ffd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: of each detailed occupation mapped to that occupational group to total hours of work in the\nprevious week. In order to be able to compare the skill content across occupational groups, we\nstandardize the skill measures using the mean and standard deviation taken across the six\noccupational groups. These normalized skill measures are the input we use to generate skill\nmeasures for each province-industry-year in the Sakernas survey using the following calculation:\n\n\n𝑆���� ��𝜔���� ∗𝑠 <sup>��</sup>\n\n                                    \n<mark>where</mark> <mark>𝜔</mark> ������ <mark>is the share of occupational group</mark> _<mark>i</mark>_ <mark>in province</mark> _<mark>p</mark>_ <mark>, industry</mark> _<mark>j</mark>_ <mark>, and year</mark> _<mark>t</mark>_ <mark>total</mark>\n\n<mark>employment obtained from Sakernas, and 𝑠</mark> <sup>��</sup> <mark>is the value of the normalized skill measure</mark> _<mark>k</mark>_ <mark>for</mark>\n<mark>occupational group</mark> _<mark>i</mark>_ <mark>from the Philippines STEP survey. When constructing skills measures by</mark>\n<mark>demographical groups –women, men, high-skilled workers and low-skilled workers, we transform</mark>\n<mark>previous expression using the shares of each occupation in each province, industry and year total</mark>\n<mark>employment for each specific demographical group:</mark>\n\n𝑆����� ��𝜔����� ∗𝑠��\n\n                                    \nwhere <mark>𝜔<\n\nMention: \"Sakernas survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Normalised occupational skill scores are projected onto province-industry-year cells in this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000535:56:0", "occurrence_id": "d85c34b61b8060608cb970eb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: of each detailed occupation mapped to that occupational group to total hours of work in the\nprevious week. In order to be able to compare the skill content across occupational groups, we\nstandardize the skill measures using the mean and standard deviation taken across the six\noccupational groups. These normalized skill measures are the input we use to generate skill\nmeasures for each province-industry-year in the Sakernas survey using the following calculation:\n\n\n𝑆���� ��𝜔���� ∗𝑠 <sup>��</sup>\n\n                                    \n<mark>where</mark> <mark>𝜔</mark> ���� <mark>is the share of occupational group</mark> _<mark>i</mark>_ <mark>in province</mark> _<mark>p</mark>_ <mark>, industry</mark> _<mark>j</mark>_ <mark>, and year</mark> _<mark>t</mark>_ <mark>total</mark>\n\n<mark>employment obtained from Sakernas, and 𝑠</mark> <sup>��</sup> <mark>is the value of the normalized skill measure</mark> _<mark>k</mark>_ <mark>for</mark>\n<mark>occupational group</mark> _<mark>i</mark>_ <mark>from the Philippines STEP survey. When constructing skills measures by</mark>\n<mark>demographical groups –women, men, high-skilled workers and low-skilled workers, we transform</mark>\n<mark>previous expression using the shares of each occupation in each province, industry and year total</mark>\n<mark>employment for each specific demographical group:</mark>\n\n𝑆����� ��𝜔����� ∗𝑠��\n\n                                    \nwhere <mark>𝜔<\n\nMention: \"Sakernas\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Total employment by province, industry and year is taken from this source as the weighting base.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000535:56:0", "occurrence_id": "4a54804862048ef8c6ee93b8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: of each detailed occupation mapped to that occupational group to total hours of work in the\nprevious week. In order to be able to compare the skill content across occupational groups, we\nstandardize the skill measures using the mean and standard deviation taken across the six\noccupational groups. These normalized skill measures are the input we use to generate skill\nmeasures for each province-industry-year in the Sakernas survey using the following calculation:\n\n\n𝑆���� ��𝜔���� ∗𝑠 <sup>��</sup>\n\n                                    \n<mark>where</mark> <mark>𝜔</mark> ���� <mark>is the share of occupational group</mark> _<mark>i</mark>_ <mark>in province</mark> _<mark>p</mark>_ <mark>, industry</mark> _<mark>j</mark>_ <mark>, and year</mark> _<mark>t</mark>_ <mark>total</mark>\n\n<mark>employment obtained from Sakernas, and 𝑠</mark> <sup>��</sup> <mark>is the value of the normalized skill measure</mark> _<mark>k</mark>_ <mark>for</mark>\n<mark>occupational group</mark> _<mark>i</mark>_ <mark>from the Philippines STEP survey. When constructing skills measures by</mark>\n<mark>demographical groups –women, men, high-skilled workers and low-skilled workers, we transform</mark>\n<mark>previous expression using the shares of each occupation in each province, industry and year total</mark>\n<mark>employment for each specific demographical group:</mark>\n\n𝑆����� ��𝜔����� ∗𝑠��\n\n                                    \nwhere <mark>𝜔<\n\nMention: \"Philippines STEP survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its standardised occupational skill scores are transferred to the local occupational structure.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000536:4:1", "occurrence_id": "9b00de980b3ee516cc148fce", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: S-ISA data spanning 3 waves from 2009 to 2012, to quantify the\n\neffect of the refugee presence on households’ welfare. We consider refugees from various\n\nsource countries and residing close to local communities (clusters in LSMS). We construct a\n\nrefugee index which weights the number of refugees in the closest refugee settlements by the\n\ninversed distance from those settlements to the clusters. In order to limit endogeneity concerns,\n\nwe instrument this variable of interest with a shift-share instrumental variable which is based\n\non the distance of the refugee settlements to the closest border crossing points for each source\n\ncountry.\n\n\nOur findings with regards to household consumption are similar to those found in Kenya (Alix\nGarcia et al., 2018), Rwanda (Alloush et al., 2017), and Tanzania (Maystadt & Verwimp,\n\n2014). Our results indicate that rural households, living close to the refugee settlements benefit\n\nfrom the presence of refugees. Similarly, Alix-Garcia and Saah (2009) find that rural\n\n\n1 Uganda National Household Survey carried out in the three waves of 2002–03, 2005–06 and 2009–10.\n\n\n3\n\nMention: \"S-ISA data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Three waves are used to estimate how nearby refugee presence affects host household welfare.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000537:43:0", "occurrence_id": "ae731e3707cf16142644be0b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ##### **Appendix C: Additional Charts and Tables**\n\n\n\n\n\n\n\n\n\n\n\n_Sources:_ Bharat Biotech, CanSino, Duke Global Health Innovation Center, UN Population Prospects, Wouters et\n\n\nal. (2021), World Bank Income Classifications.\n\n\n_Notes:_ The population of Russia, China and India is 2.97 billion, and these are classified as middle-income countries\n\n\nby the World Bank.\n\n\n41\n\nMention: \"UN Population Prospects\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Population totals used to scale vaccine capacity are drawn from these projections.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000541:3:1", "occurrence_id": "67fbb758d46e564b14cd09c2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  worsening the conflict (e.g., Berman et al. 2013).\n\n\nThis article investigates the change in humanitarian assistance following a particularly in\nteresting event- the announcement of a food emergency in Yemen in 2017. The United Na\ntions announced the 2017 Integrated Food Security and Phase Classification (IPC) in March\n\nof 2017, which is a classification system that identifies food emergencies. Although the 2017\n\nIPC announcement did not officially declare a famine, the announcement detailed a dire food\n\nemergency facing Yemen, and humanitarian agencies began describing the food emergency with\nmore urgency (e.g., OCHA 2016, OCHA 2017). <sup>1</sup>\n\n\nIn order to empirically estimate the impacts of the 2017 IPC announcement, this analysis\n\nutilizes a novel mobile phone survey conducted by the World Food Programme (WFP) that pro\nvides high-frequency data on food assistance and food access that are regionally disaggregated.\n\nThe survey is one of the most complete data sources describing the evolution of food assistance\n\nduring the course of a humanitarian disaster and offers a unique opportunity to investigate the\n\ntargeting of humanitarian assistance. Although the survey misses households without access to\n\nmobile phones who potentially are more in need of assistance, the relatively high penetration\n\nof mobile phones allows an analysis of targeting for the majority of the population.\n\n\nThe 2017 IPC announcement led to an almost immediate and significant increase in the share\n\nof respondents receiving food assistance in the country that further increased in the year-and-a\n\n1Food insecurity was already widespread in the early stages of the war (e.g., Tandon 2019), and the empirical\nanalysis demonstrates that they was not a large change in food security in the country leading up to the\nannouncement of the 2017 IPC. Section 2 presents possible reasons why the 2017 IPC announcement might have\nhad a larger impact than other IPC announcements.\n\n\n2\n\nMention: \"mobile phone survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"High-frequency regionally disaggregated data on food assistance and access identify the effects of the food security announcement.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000541:3:1", "occurrence_id": "4faf79944bf7d17851fa0795", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  worsening the conflict (e.g., Berman et al. 2013).\n\n\nThis article investigates the change in humanitarian assistance following a particularly in\nteresting event- the announcement of a food emergency in Yemen in 2017. The United Na\ntions announced the 2017 Integrated Food Security and Phase Classification (IPC) in March\n\nof 2017, which is a classification system that identifies food emergencies. Although the 2017\n\nIPC announcement did not officially declare a famine, the announcement detailed a dire food\n\nemergency facing Yemen, and humanitarian agencies began describing the food emergency with\nmore urgency (e.g., OCHA 2016, OCHA 2017). <sup>1</sup>\n\n\nIn order to empirically estimate the impacts of the 2017 IPC announcement, this analysis\n\nutilizes a novel mobile phone survey conducted by the World Food Programme (WFP) that pro\nvides high-frequency data on food assistance and food access that are regionally disaggregated.\n\nThe survey is one of the most complete data sources describing the evolution of food assistance\n\nduring the course of a humanitarian disaster and offers a unique opportunity to investigate the\n\ntargeting of humanitarian assistance. Although the survey misses households without access to\n\nmobile phones who potentially are more in need of assistance, the relatively high penetration\n\nof mobile phones allows an analysis of targeting for the majority of the population.\n\n\nThe 2017 IPC announcement led to an almost immediate and significant increase in the share\n\nof respondents receiving food assistance in the country that further increased in the year-and-a\n\n1Food insecurity was already widespread in the early stages of the war (e.g., Tandon 2019), and the empirical\nanalysis demonstrates that they was not a large change in food security in the country leading up to the\nannouncement of the 2017 IPC. Section 2 presents possible reasons why the 2017 IPC announcement might have\nhad a larger impact than other IPC announcements.\n\n\n2\n\nMention: \"World Food Programme\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Named as the agency that fielded the survey, establishing its provenance.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000542:31:1", "occurrence_id": "e5ac002f5011cbf99e161a59", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ¯ _re_ = 2.8%.\n\nWe set the schooling transition probabilities _πs_ ( _·_ ) to match the schooling\n\ndistribution in developed and developing economies (I2D2). <sup>49</sup> We assume agents may\n\n\nonly accumulate up to 25 years of schooling as some countries only record schooling\n\n\nup to 25. Finally, we set the employment probabilities _πe_ ( _·_ ) using participation, youth\n\n\nunemployment and adult unemployment rate averages reported by World Bank (2018a).\n\n\nSee Table 5 for a summary of the values of the parameters we used. See\n\n\nWeb Appx. Section A.6 for details of the parameterization of the accounting framework.\n\n##### **6.3. Main Quantitative Findings**\n\n\nA typical developing economy has 24% of the GDP per capita of a typical developed\n\n\neconomy (World Bank, 2018a). In the simulations, we find a value of 48 _._ 5%. The gap\n\n\nbetween 48 _._ 5 and 100 is 68% of the gap between 24 and 100. Thus, in our accounting\n\n\nmodel, differences in education, experience and the returns thereof, plus demographic\n\nand labor market differences, account for two thirds of the income gap. <sup>50</sup> Demographic\n\n\nand labor market differences are small, however, with the exception of labor market\n\n\nparticipation. Assume instead that demographic and labor market factors are the same\n\n\nas in the developed economy, allowing only differences in the stock of schooling and the\n\n\nreturns to human capital. Then, we find that relative _pcGDP_ is 53 _._ 5 percent, accounting\n\n\n48See Web Appx. Fig. 8(a) for the mortality distributions.\n49See Web Appx. Fig. 8(b) for the distribution of schooling we use, based on the I2D2 data (1990-2016)\n\nMention: \"I2D2\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Schooling transition probabilities are calibrated to the schooling distribution in this dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000543:14:1", "occurrence_id": "614a4cf1d9e3349eea8b07e4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \nmodern day ‘slavery.’ Toman and Trebesch (2010) provide some of the only strong empirical\nevidence confirming that a lack of information is related to rates of human trafficking. Using\noriginal household survey data from Belarus, Bulgaria, Moldova, Romania, and Ukraine, Toman\nand Trebesch (2010, pp 180) show that, “Households in regions with higher awareness of the\nphenomenon of human trafficking are less likely to have a trafficked family member. The same is\ntrue for households which use the TV as main source of information on social and political\nissues.” <sup>13</sup> Bettio and Nandi (2010) come to a similar conclusion in their analysis of IOM data on\nsex trafficking victims. They find that, “nationals of Belarus and Ukraine consider trafficking in\nhuman beings to be less important than do the nationals of the three other large source countries\nin the survey—Bulgaria, Romania and Moldavia. Lower awareness of, and sensitivity to, human\ntrafficking often means a higher tolerance of rights violations; or it may simply foster ignorance\nabout such violations and thus reduce the pressure on traffickers” (Bettio and Nandi 2010, pp 32).\n\n\n11 Cho (2013) instrumentalizes social globalization with the KOF index. It includes information flows,\npersonal contacts, and cultural sharing across countries.\n12 These proxies are ‘reliability of the police’ and rates of ‘bribe seeking’ among police.\n13 Toman and Trebesch (2010) operationalize information in two ways: “First, we code a dummy for\nhouseholds that use the TV, as opposed to other media and social contacts, to get informed about social and\npolitical issues. […] Second, we include a regional risk-awareness measure by taking the share of\nrespondents who stated that they had heard of the phenomenon of human trafficking before.”\n\n\n13\n\nMention: \"IOM data on\nsex trafficking victims\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"A prior study uses these victim data to compare awareness of trafficking across source countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000544:16:0", "occurrence_id": "e587a7f938668d71a6caa182", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Second, perceptions do not seem fixed, confirming the original intuition of Zaller (1992). In fact,\n\nfor some countries the shifts in perceptions are quite remarkable. For example, Poland and Portugal. <sup>19</sup> [^19: Note that we do not use panel data, so the shift in share may simply be due to a cohort effect, i.e. people from\nyounger cohorts may have different opinions and that may explain (part) of the shift observed in the figure.]\n\n\nFinally, there seems to be some correlation between the evolution of objective inequality and subjective\n\n\nperceptions (more on this below).\n\n\nThe ISSP surveys of 1992, 1999, and 2009 also provide data on the evolution of demand for\n\n\nredistribution. ‘Net’ demand for redistribution is defined as the difference between the share of individuals\n\n\nwho strongly agree with the statement: “it is the responsibility of the government to reduce income\n\n\ndifferences between people with high incomes and those with low incomes” and those who strongly\n\n\ndisagree with it. A negative value indicates that more individuals disagree with the statement than agreeing\n\n\nwith it, while a positive value indicates the opposite.\n\n\nFigure 3 plots the evolution of demand for redistribution over time and across countries. As in the\n\n\ncase of perceptions, some clear differences between countries are highlighted: European countries, both in\n\n\nthe East and the West, have a stronger demand for redistribution than the rest of the world, particularly\n\n\nwhen compared to the United States, which is the only country in the sample that has a negative net demand\n\n\nfor redistribution. This is not surprising given the differences in preferences between European and U.S.\n\n\ncitizens well documented in the literature. Within Europe, Eastern European countries show a higher\n\n\ndemand than most Western and Southern countries. Over time, demand for redistribution has also moved\n\n\ndifferently in the various countries, increasing from 1999 to 2009 in some countries (e.g., Hungary, Poland,\n\n\nand France), and decreasing in\n\nMention: \"ISSP surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Three waves provide the measured demand for redistribution over time.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000545:6:2", "occurrence_id": "443d6e89aa3b30a61f65de50", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: /tiva` for the OECD TiVA database, `wiod.org` for the\nWorld Input-Output Database and `worldmrio.com` for the Eora Global Supply Chain Database.\nPlease, also remember to cite the reference to the ICIO database you are analyzing through `icio` .\n\n5Multiple years have to be loaded sequentially and then the results should be appended.\n\n\n5\n\nMention: \"OECD TiVA database\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"Listed as one of the trade-in-value-added sources users should cite when running the tool.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000545:6:2", "occurrence_id": "b9d0dcf35523f8a0bde6d47c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: /tiva` for the OECD TiVA database, `wiod.org` for the\nWorld Input-Output Database and `worldmrio.com` for the Eora Global Supply Chain Database.\nPlease, also remember to cite the reference to the ICIO database you are analyzing through `icio` .\n\n5Multiple years have to be loaded sequentially and then the results should be appended.\n\n\n5\n\nMention: \"World Input-Output Database\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"Named with its web address among the input-output sources the software can load.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000545:6:2", "occurrence_id": "40d8ba2cad69bec15e6200eb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: /tiva` for the OECD TiVA database, `wiod.org` for the\nWorld Input-Output Database and `worldmrio.com` for the Eora Global Supply Chain Database.\nPlease, also remember to cite the reference to the ICIO database you are analyzing through `icio` .\n\n5Multiple years have to be loaded sequentially and then the results should be appended.\n\n\n5\n\nMention: \"Eora Global Supply Chain Database\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"Another supported global supply chain source listed for citation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000545:6:2", "occurrence_id": "15236f60633f3660dc67b58f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: /tiva` for the OECD TiVA database, `wiod.org` for the\nWorld Input-Output Database and `worldmrio.com` for the Eora Global Supply Chain Database.\nPlease, also remember to cite the reference to the ICIO database you are analyzing through `icio` .\n\n5Multiple years have to be loaded sequentially and then the results should be appended.\n\n\n5\n\nMention: \"ICIO database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Users are instructed to cite whichever of these databases they analyse.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000548:34:2", "occurrence_id": "1fb980c6a78b14c0b50080fd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  km inland. We thus adopt a\nto directly determine 𝐹𝐹, Vickery et al. (2009) note that in open water the reduction factor is\nreduction factor that linearly decreases within this range as we consider points _i_ further inland\nfrom the coast. To determine 𝐵𝐵 we employ Holland’s (2008) approximation method, whereas\nhurricane track data is the HURDAT Best Track Data, which provides six hourly data on all\nwe use the parametric model estimated by Xiao et al. (2009) to estimate 𝑅𝑅𝑚 **𝑚** 𝑚𝑚 . Our source for\ntropical cyclones, including the position of the eye and the maximum wind speed of the storm.\n\nwe take the centroid of the barangay’s PSG coordinates.\nThese tracks are linearly interpolated to hourly positions. Finally, as set of points, 𝑗𝑗= 1, … 𝐽𝐽,\n\n\n\n33\n\nMention: \"HURDAT Best Track Data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Six-hourly storm positions and maximum wind speeds are used to parameterise the hurricane exposure model.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000552:17:0", "occurrence_id": "a2e022c26e3ad3a41f5a025e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ments to competition law defines the market affected at the 4-digit ISIC level and the year\nfor which the correction of competition was enforced. Most cases did not specify a particular region, which suggests that the anti-competitive behavior was not limited to a narrow\nlocation, so each case is assumed to have national impact. Moreover, the limited number\nof observations per region in the firm database reduced the possibility of studying the few\ncases that had a region-specific scope. When a resolution applied to a sub-region rather\nthan the whole national market, our coefficients would be picking up a diluted, underestimated effect that would represent a lower bound of what the actual effect is.\n\nWe assume that competition enforcement is exogenous to firm performance. Osiptel\nand Indecopi are independent and autonomous bodies that receive cases initiated externally by complaint from any physical or moral persons and internally based on monitoring\nand control organs. This ensures that competition enforcement is applied nationwide regardless of firm or regional idiosyncrasies. Violation of competition law needs to be legally\nconfirmed, following pre-defined requirements, for a case to be accepted. Rulings can be\nappealed to independent courts. All these elements support the assumption that interventions are exogenous.\n\nThe information about _de_ _jure_ and _de_ _facto_ regulations was collected from the digital\nlegislative archive from OSIPTEL and INDECOPI, following the criteria that the degree or\nresolution had to be relevant (i.e. pertain to at least one of the ICT branches of hardware,\nsoftware, telecom or ICT infrastructure), timely (i.e. had to be published and implemented\nin the 2007-2017 period) and had to be related to competition (e.g., encouraging firm\nentry, reducing misallocation and administrative barriers that hinder innovation).\n\n##### **4.2 Empirical Approach**\n\n\nThe methodology described above requires that productivity and markups be identified\nconsistently if the\n\nMention: \"firm database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Too few regional observations mean region-specific cases cannot be studied separately.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000552:17:0", "occurrence_id": "6ad6cbf0fb53b43d25aac2cd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ments to competition law defines the market affected at the 4-digit ISIC level and the year\nfor which the correction of competition was enforced. Most cases did not specify a particular region, which suggests that the anti-competitive behavior was not limited to a narrow\nlocation, so each case is assumed to have national impact. Moreover, the limited number\nof observations per region in the firm database reduced the possibility of studying the few\ncases that had a region-specific scope. When a resolution applied to a sub-region rather\nthan the whole national market, our coefficients would be picking up a diluted, underestimated effect that would represent a lower bound of what the actual effect is.\n\nWe assume that competition enforcement is exogenous to firm performance. Osiptel\nand Indecopi are independent and autonomous bodies that receive cases initiated externally by complaint from any physical or moral persons and internally based on monitoring\nand control organs. This ensures that competition enforcement is applied nationwide regardless of firm or regional idiosyncrasies. Violation of competition law needs to be legally\nconfirmed, following pre-defined requirements, for a case to be accepted. Rulings can be\nappealed to independent courts. All these elements support the assumption that interventions are exogenous.\n\nThe information about _de_ _jure_ and _de_ _facto_ regulations was collected from the digital\nlegislative archive from OSIPTEL and INDECOPI, following the criteria that the degree or\nresolution had to be relevant (i.e. pertain to at least one of the ICT branches of hardware,\nsoftware, telecom or ICT infrastructure), timely (i.e. had to be published and implemented\nin the 2007-2017 period) and had to be related to competition (e.g., encouraging firm\nentry, reducing misallocation and administrative barriers that hinder innovation).\n\n##### **4.2 Empirical Approach**\n\n\nThe methodology described above requires that productivity and markups be identified\nconsistently if the\n\nMention: \"digital\nlegislative archive\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"De jure and de facto regulations in the communications sector are collected from these regulators' archives.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000555:26:0", "occurrence_id": "9ad9870796a5511a730e312b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 1. Budget balance and gross government debt over the last two global crises**\n\n\n**A.** **Budget balance** **B.** **Gross government debt**\n\n\nS _ources:_ IMF World Economic Outlook; World Bank.\n_Notes:_ EMDEs = emerging market and developing economies _._ Country weights based on PPP GDP.\n_A._ Weighted averages of budget balance (as percent of GDP) for 35 advanced economies and 148 EMDEs. Data from\n\n2000 to 2021.\n_B._ Weighted averages of gross government debt (as percent of GDP) for 30 advanced economies and 144 EMDEs.\n\nData from 2000 to 2021.\n\n\n**Figure 2. Commodity prices**\n\n\n_Source:_ World Bank.\n_Note:_ Last observation is December 2022.\n\n\n25\n\nMention: \"IMF World Economic Outlook\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Budget balances and gross government debt for advanced and emerging economies come from this source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000561:5:1", "occurrence_id": "8533a4e965494bfb0b9d0ac9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  counteract the negative impact brought\n\n\nby mobility restrictions and a change in consumption patterns (see Watanabe and Omori (2020)).\n\nIn most of the paper, unless otherwise noted, the analysis restricts the sample to firms with 5 or\n\nmore employees, which was a threshold used in most countries for which data is available. <sup>3</sup>\n\n\nThree main indicators regarding the digital response by firms were collected. First, whether the\n\n\nfirm has started or increased the use of digital platforms in response to the pandemic. Second, if the\n\nfirm invested in the implementation or use of digital solutions; and third, the share of sales done\n\n\nthrough online channels. Throughout the paper, analyses explore how firms differed in their digital\n\n\nresponse towards the pandemic, and the possible implications of such responses.\n\n\nFor the analysis, the paper will focus on the 57 countries <sup>4</sup> with data on digital changes. In total,\n\n\nthere is information on 65,130 firms, where 25,255 are panel firms. This feature of the data allows to\n\nanalyze how digital changes happened since the start of the pandemic <sup>5</sup> . The first wave of surveys\n\n\n3The WBES uses the threshold of 5+ employees.\n4Countries include: Argentina, Bangladesh, Benin, Brazil (Sao Paulo), Bulgaria, Cambodia, Chile, Croatia, Cyprus,\nCzechia, the Arab Republic of Egypt, El Salvador, Estonia, Georgia, Greece, Guatemala, Honduras, Hungary, India,\nIndonesia, Italy, Jordan, Kenya, Kosovo, Kyrgyzstan, Latvia, Lesotho, Lithuania, Madagascar, Malaysia, Malta, Moldova,\nMongolia, Morocco, Mozambique, Nepal, Nicaragua, Pakistan, Paraguay, Poland, Portugal, Romania, Senegal, Sierra\nLeone, the Slovak Republic, Slovenia, Somalia, South Africa, Sri Lanka, Tajikistan, Tanzania, Tunisia, Türkiye, Uzbekistan,\nVietnam, Zambia,\n\nMention: \"WBES\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Its panel of firms with a five-employee threshold allows digital changes since the pandemic to be tracked.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000566:9:0", "occurrence_id": "c619c27d1996243114c89c90", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: periodic breaks. Table A-5 lists the resulting set of indicators and their sources.\n\n\nWe retain 14 distinct environmental indicators, covering different types of pollution,\n\n\nhabitat preservation, and forest protection. Three come from the World Bank World\n\n\nDevelopment Indicators: the ratio of CO2 emissions in kg to GDP (in PPP $), forest area\n\n\n(% of land area) and SO2 emissions. Eleven additional outcome measures are sourced\n\n\nfrom the EPI data. These include two measures related to Climate Change mitigation:\n\n\nblack carbon intensity and greenhouse gas emissions (measured in terms of average an\n\nnual rate of increase); three indicators related to Environmental Health (PM2.5 exposure,\n\n\nNOx Intensity, and Ozone Exposure, all measured using the number of age-standardized\n\n\ndisability-adjusted life-years lost per 100,000 persons due to exposure to ground-level\n\n\nozone pollution); and six measures of Ecosystem Vitality - a protected areas index (the\n\n\nproportion of biologically scaled environmental diversity included in a country’s terrestrial\n\n\nprotected areas, with higher values reflecting better performance), the species habitat in\n\ndex (the proportion of suitable habitat within a country that remains intact for each\n\n\nspecies in that country relative to a baseline set in the year 2001), the sustainable nitro\n\ngen management index measuring the efficiency in application of nitrogen fertilizer with\n\n\nmaximum crop yields (higher values indicating worse performance), a species protection\n\n\nindex measuring the species-level ecological representativeness of each country’s protected\n\n\narea network, tree cover loss (five-year moving average of the percentage of forest lost\n\n\nrelative to forest cover in 2000), and wetland loss (quantified using a five-year moving\n\n\naverage of the percentage of gross losses in wetland areas relative to a 1992 baseline). For\n\n\nEPI indices, a higher value indicates (sometimes counter-intuitively) better performance\n\n\nwith respect to the outcome of interest. <sup>8</sup>\n\n\nFor labor protection we use two outcome variables:\n\nMention: \"World Bank World\n\n\nDevelopment Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Three of the fourteen environmental indicators, including emissions intensity and forest cover, come from this series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000566:9:0", "occurrence_id": "5b19badb931cdff17581eae2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: periodic breaks. Table A-5 lists the resulting set of indicators and their sources.\n\n\nWe retain 14 distinct environmental indicators, covering different types of pollution,\n\n\nhabitat preservation, and forest protection. Three come from the World Bank World\n\n\nDevelopment Indicators: the ratio of CO2 emissions in kg to GDP (in PPP $), forest area\n\n\n(% of land area) and SO2 emissions. Eleven additional outcome measures are sourced\n\n\nfrom the EPI data. These include two measures related to Climate Change mitigation:\n\n\nblack carbon intensity and greenhouse gas emissions (measured in terms of average an\n\nnual rate of increase); three indicators related to Environmental Health (PM2.5 exposure,\n\n\nNOx Intensity, and Ozone Exposure, all measured using the number of age-standardized\n\n\ndisability-adjusted life-years lost per 100,000 persons due to exposure to ground-level\n\n\nozone pollution); and six measures of Ecosystem Vitality - a protected areas index (the\n\n\nproportion of biologically scaled environmental diversity included in a country’s terrestrial\n\n\nprotected areas, with higher values reflecting better performance), the species habitat in\n\ndex (the proportion of suitable habitat within a country that remains intact for each\n\n\nspecies in that country relative to a baseline set in the year 2001), the sustainable nitro\n\ngen management index measuring the efficiency in application of nitrogen fertilizer with\n\n\nmaximum crop yields (higher values indicating worse performance), a species protection\n\n\nindex measuring the species-level ecological representativeness of each country’s protected\n\n\narea network, tree cover loss (five-year moving average of the percentage of forest lost\n\n\nrelative to forest cover in 2000), and wetland loss (quantified using a five-year moving\n\n\naverage of the percentage of gross losses in wetland areas relative to a 1992 baseline). For\n\n\nEPI indices, a higher value indicates (sometimes counter-intuitively) better performance\n\n\nwith respect to the outcome of interest. <sup>8</sup>\n\n\nFor labor protection we use two outcome variables:\n\nMention: \"EPI data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Eleven of the environmental outcome measures, covering mitigation and health, are taken from this performance index.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000569:13:0", "occurrence_id": "67569eb046959cfbb9070203", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 2016 (Table 2). Furthermore, by 2019, Figure 7 shows that these high-wealth individuals\ncontinued to report significantly larger assets, with their declared holdings nearly tripling\ncompared to 2015.\n\nOne concern in this analysis revolves around the possibility that disclosing assets\nmight lead to a reshuffling of individuals across different percentile groups over time.\nTo address this potential concern, we utilize data from Argentina’s income tax returns,\nwhere individuals must report assets held in both the current and previous tax years.\nThis approach lets us compare asset changes while keeping individuals’ income rankings\nconstant. <sup>7</sup> [^7: For example, the 2016 income tax tabulation, which ranks individuals based on their income, includes data\non assets held in 2015 and 2016 for each income bracket.\n8Argentina’s wealth tabulations rank individuals based on total assets. To recover the share of foreign assets\n(and the effective tax rate) for each fractile, we cumulate the amount of foreign assets (and wealth taxes)\nby total assets, interpolate the cumulative function for each fractile with a smooth cubic spline function,\nand differentiate the interpolated function. Reassuringly, a linear interpolation delivers virtually the same\nresults.] The results are depicted in Figure 8, highlighting that the most significant increases in reported assets are primarily observed among individuals in the top 0.1% of the\nincome distribution. In this group, reported assets more than doubled, with these individuals reporting roughly one-third more assets in 2016 than in 2015. Interestingly, the scale\nof the asset increase reported by Argentina’s richest individuals aligns with what has been\nobserved in other regions such as Colombia (Londoño-Vélez and Ávila-Mahecha, 2021),\nScandinavia (Alstadsæter et al., 2019), and the Netherlands (Leenders et al., 2023).\n\nLastly, Figure A.13 compares high-net-worth individuals’ likelihood of reporting a\nforeign asset, and the share of foreign assets declared. This information is presented separately by the top fractile groups, decomposing the top 1\n\nMention: \"data from Argentina’s income tax returns\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Assets reported for consecutive tax years let asset changes be compared while holding income rank fixed.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000570:18:0", "occurrence_id": "26f1056f6bb2e50131d0e1be", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: other hand, in Gezira, North Darfu, and Blue Nile, the percentage of stunted children increased by\n\n13 percent, 12 percent, and 10 percent, respectively, between 2010 and 2014.\n\n\n**Figure 2: Prevalence of Stunting among Under-Five-Year-Old Children in Sudan: National, Rural, and Urban**\n**as well as Changes across States**\n\n\n**(a) National, Rural, and Urban** **(b) Changes across States**\n\n\n\n\n\n\n\n_Source_ : Authors’ calculation using MICS 2010 and 2014 data.\n\n\nStunting is more prevalent among children from poorer households. Using a composite indicator of\n\n\nhousehold wealth, it can be observed that stunting is more prevalent among households in the bottom\n\n\nof the wealth distribution and remained high in 2014. More than 40 percent of children in the poorest,\n\n\npoor, and middle wealth quintiles were stunted in 2014—more than 20 percent of these children were\n\nseverely stunted. Although the prevalence of stunting increased across the wealth distribution between\n\n\n2010 and 2014, children from households at the bottom of the distribution remain disproportionately\n\n\naffected. These statistics highlight the centrality of household poverty in the prevalence of stunting.\n\n\nUnderstanding the extent to which poverty influences undernutrition, particularly its role in driving\n\ninequalities in access to the underlying determinants of nutrition, is a crucial first step in designing and\n\n\nimplementing needed reforms. This paper aims to make a contribution in this regard- by highlighting\n\n\nthe role of multisectoral approaches. There also appears to be significant heterogeneity in the\n\n\nprevalence of stunting across households based on the education level of mothers. In both 2010 and\n\n2014, moderate and severe stunting were more prevalent among children whose mothers had no\n\n\nformal education compared to those whose mothers had some primary, secondary, or higher\n\n\neducation. For instance, in 2014, the percentage of moderately and severely stunted children whose\n\n\n17\n\nMention: \"MICS 2010 and 2014 data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Stunting prevalence nationally and by state is calculated from these two rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000571:45:0", "occurrence_id": "6d51432c4264d3e7a44442c2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 10: The marginal effect of education and location on employment\n\n\nSource: DHS rounds (see, for details, Table A2 in the annex). The marginal effects (measured as percentage points on the vertical axes) are\nestimated by regression the outcome variable on location (rural versus urban) or higher education (secondary and tertiary education versus no\nor primary education), its interactions with year of birth, age, and state fixed effects. Standard errors are clustered by the state. We focus on the\nslope of the marginal effects. Moving from left to right in the graph allows to compare the difference in the marginal effects between old and\nmore recent cohorts who are categorized into urban versus rural resident or higher versus lower education level.\n\n\n44\n\nMention: \"DHS rounds\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Marginal effects of education and rural location on employment are estimated across these rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000572:5:1", "occurrence_id": "3a84110ade0cd9fd858c98a5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  that the shift toward informal work and away from agricultural work\n\n\nis larger within ban states and during the 2018-2019 survey round, especially during the post-harvest\n\n\nseason. Finally, to investigate possible welfare implications of the measured labor responses after exposure\n\n\nto herder-related violence, we also estimate additional “knock-on” or lagged effects that materialize during\n\n\nthe post-harvest period of the 2018-2019 GHS survey and are associated with exposure to herder-related\n\n\nviolence in the previous planting season. We find that households exposed to herder-related violence and\n\n\nwho have also been previously exposed to this violence (i) shift their consumption toward food and away\n\n\nfrom other expenditures, (ii) are more likely to sell raw crops to relatives/friends or via the farmgate,\n\n\nand (iii) are more likely to operate a non-farm enterprise.\n\n\nOur paper is closely related to two sets of existing literature. The first set studies the role of the\n\n\ninformal sector as a safety net in response to adverse shocks (Gunther and Launov, 2012; Epstein and\n\n\nShapiro, 2017; Adhvaryu _et_ _al._, 2019; Colombo _et_ _al._, 2019). For example, Adhvaryu _et_ _al._ (2019) study\n\n\nthe effect of price shocks in the global coffee market on informal work in Tanzania. The authors document\n\n\nthat households cope with declines in the global price for coffee by increasing informal household enterprise\n\n\nownership. Our paper documents the labor-related effects of a different type of shock, specifically exposure\n\n\nto heightened violence, exposure to which may reasonably be expected to alter individual labor allocation.\n\n\nThe investigation of the specific effects of Nigeria’s farmer-herder conflict links our paper to a second\n\n\nset of literature, which examines exposure to conflict in Nigeria specifically. This violence has been shown\n\n\nto influence agricultural output (Adelaja and George, 2019), food security, (\n\nMention: \"GHS survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Post-harvest consumption and labour responses to prior-season violence are measured in this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000573:49:0", "occurrence_id": "24f9059eb69fb415fb08b894", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: #### Appendix C: Validation of the wave model\n\n\n\nThe wave transformation model is validated against field wave measurements collected\nat a salt marsh fringe by Vuik et al. (2016). Wave propagation through the salt marshes is\n\n\n\nat a salt marsh fringe by Vuik et al. (2016). Wave propagation through the salt marshes is\n\n\noffshore boundary. The bathymetry varies between an offshore water depth of 2 m, and a\n\nmodeled for a significant wave height of 𝐻𝐻𝑚𝑚0 = 0.6 m and a peak period of 𝑇𝑇𝑝𝑝 = 3.5 s at the\n\n\n\noffshore boundary. The bathymetry varies between an offshore water depth of 2 m, and a\nnearshore water depth of 0.5 m. The salt marshes extend over a length of 55 m and have\nvarying vegetation properties across the profile, but uniform properties over the vertical\ncoordinate. The vegetation properties are summarized in Table C1. A drag coefficient of\n\n\n\ncoordinate. The vegetation properties are summarized in Table C1. A drag coefficient of\n\n\n(2016).\n\n𝑐𝑐𝐷𝐷,𝑤𝑤 = 0.4 is used in the simulations, which is equal to the value calibrated by Vuik et al.\n\n\n\n(2016).\n\n\n\n_Table C1. Salt marsh properties used for the validation from Vuik et al. (2016), consisting of the mean vegetation height hv,_\n_the mean stem density Nv, and the mean stem width bv. The location of the sections across the profile is indicated in Vuik et al._\n\n_(2016)._\n\n|Section|h (m)<br>v, mean|N (plants/m2)<br>v|b (m)<br>v|\n|---|---|---|---|\n|<br>S1-S2|<br>0,2|944|<br>0,030|\n\nMention: \"field wave measurements\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Instrumented waves at a salt marsh fringe are used to test the wave transformation model.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000576:37:0", "occurrence_id": "b919cc80a444af9eb1f5d581", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure** **2:** **The** **Investigative** **Process** . Diagram developed by the authors based on COFECE (2017). Case dismissal, case not\npresented, case closure, and resolution are terminology used to code cases in the database of cases. This paper studies outcomes of\ncases that have a investigation by the Commission, and therefore a description of the market. Investigations end either in case closure\nor a trial procedure with sanction and/or commitment.\n\nMention: \"database of cases\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Case terminology and outcomes are coded from this competition authority's case records.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000577:34:0", "occurrence_id": "d14b89c4eba980e0f57fd06d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Western countries” (Buttorff and Welborne 2015, 1). <sup>16</sup> [^16: The study by Buttorff and Welborne primarily used data from the Arab Barometer’s 2006–2007 survey.] That said, there seems to be some gender\ndifferences in the use of _wasta_ depending on the country context. In a study comparing several Arab\ncountries, men and women appeared to use roughly comparable amounts of _wasta_ in Algeria, Morocco,\nand Lebanon, while men were using considerably more _wasta_ than women in Jordan, Palestine, and\nYemen (Buttorff and Welborne 2015).\n\n\nIn Sudan, the use of _wasta_ to shape current institutions can be traced back to colonial times. Mann\nexplains how the British colonial administration used the _wasta_ of Sudan’s sectarian and tribal leaders to\ndevelop Sudanese institutions modelled after British institutions. For example, following independence,\nGordon Memorial College became the University of Khartoum, and the children of the sectarian and tribal\nleaders, and their trusted circles, had an easier time securing positions in Gordon Memorial College and\nthen posts in the government. Similarly, Sudan’s post-1983 leaders used the _wasta_ of regional leaders in\nIslamic banking to Islamize Sudan’s banking system. Those whose personal connections brought them into\nthe fold of these banks, like the youth of the Colonial era who were educated in Gordon Memorial College,\nbecame the country’s future leaders in business and politics.\n\n\n_Wasta_ continues to play an important role in both accessing and rising within the Sudanese labor market.\nMore than half, 56 percent, of the Arab Barometer (2018) respondents believe that it is often impossible\nto get a job without _wasta_ . According to Mann’s fieldwork in Sudan, _wasta_ plays a greater role among the\ncurrent generation than it did among previous generations. <sup>17</sup> Mann writes, “[ _Wasta_ ] has become so\nimportant in\n\nMention: \"Arab Barometer\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"A footnote records that the cited study relied primarily on this survey round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000578:16:0", "occurrence_id": "f4217a2e8d28c68535e4c738", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Table 1: Descriptive Statistics\n\n\n**Observations** **Mean** **S.D.** **Min** **Max**\n\n\n**Panel** **A:** **Election** **Outcomes** (1) (2) (3) (4) (5)\n\n\nTotal no. of candidates 9,200 10.36 7.26 1 301\nEffective number of parties 9,197 2.92 0.97 1.05 12.50\nNo. of female candidates 9,200 0.49 0.94 0 49\nNo. of male candidates 9,200 9.87 6.83 1 252\nNo. of independent candidates 9,200 5.30 6.22 0 298\nNo. of parties 9,200 5.03 2.61 0 24\nConstituencies with _≥_ one female candidate 9,200 0.34 - 0 1\nFemale winner 9,200 0.05 - 0 1\nAligned winner 9,200 0.57 - 0 1\nWinner from incumbent party 9,200 0.30 - 0 1\n\n\n**Panel** **B:** **Candidate** **Characteristics** (1) (2) (3) (4) (5)\n\n\nCompleted primary schooling 33,895 0.93 - 0 1\nCompleted secondary schooling 33,895 0.78 - 0 1\nAccused in criminal case 35,033 0.16 - 0 1\nAccused in serious criminal case 35,033 0.08 - 0 1\nTotal declared assets (INR in millions) 35,033 20.38 - 0 10,205\n\n\n_Notes:_ Panel A shows descriptive statistics for variables from the SHRUG dataset on district _×_ year-level state\nelection outcomes during 1974 to 2018. Panel B shows descriptive statistics from affidavit data provided by\nthe Association of Democratic Reforms on candidate characteristics for\n\nMention: \"SHRUG dataset\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"District-by-year state election outcomes are drawn from this dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000578:16:0", "occurrence_id": "7bb5fff64865abf28c60ed73", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Table 1: Descriptive Statistics\n\n\n**Observations** **Mean** **S.D.** **Min** **Max**\n\n\n**Panel** **A:** **Election** **Outcomes** (1) (2) (3) (4) (5)\n\n\nTotal no. of candidates 9,200 10.36 7.26 1 301\nEffective number of parties 9,197 2.92 0.97 1.05 12.50\nNo. of female candidates 9,200 0.49 0.94 0 49\nNo. of male candidates 9,200 9.87 6.83 1 252\nNo. of independent candidates 9,200 5.30 6.22 0 298\nNo. of parties 9,200 5.03 2.61 0 24\nConstituencies with _≥_ one female candidate 9,200 0.34 - 0 1\nFemale winner 9,200 0.05 - 0 1\nAligned winner 9,200 0.57 - 0 1\nWinner from incumbent party 9,200 0.30 - 0 1\n\n\n**Panel** **B:** **Candidate** **Characteristics** (1) (2) (3) (4) (5)\n\n\nCompleted primary schooling 33,895 0.93 - 0 1\nCompleted secondary schooling 33,895 0.78 - 0 1\nAccused in criminal case 35,033 0.16 - 0 1\nAccused in serious criminal case 35,033 0.08 - 0 1\nTotal declared assets (INR in millions) 35,033 20.38 - 0 10,205\n\n\n_Notes:_ Panel A shows descriptive statistics for variables from the SHRUG dataset on district _×_ year-level state\nelection outcomes during 1974 to 2018. Panel B shows descriptive statistics from affidavit data provided by\nthe Association of Democratic Reforms on candidate characteristics for\n\nMention: \"affidavit data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Candidate characteristics including criminal accusations and declared assets are taken from these filings.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000584:73:1", "occurrence_id": "9e846f2e889ff68443344a1b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Total subsidy amount</u>\n\n1,053,699,411,307\n1,503,939\n2017\n\n2,534,649,735,000\n<u>2021</u> 1,776,137\n\n_Source_ : DPS, Pension Solidarity Fund, GEIH, ENPH. *Simulated values.\n\nC. Jóvenes en Acción (“Youth in Action”)\n\n\nThe amount of the conditional transfer to young people in the program depends on where they\nstudy (SENA or HEI), as well as their academic performance. The cost of Jóvenes en Acción for\n2020 was COP586,742,148,000 and benefited more than 494,000 young people.\n\nHowever, the GEIH underestimates the beneficiaries of the Jóvenes en Acción program (155,903),\nso potential young people who could receive the benefit were identified. They are between 16\nand 28 years old, declare in the survey that they receive government aid, and are in poverty\naccording to the Sisbén IV level. This adjustment leads to the identification of 337,000 additional\nyoung people, which represents 68 percent of the population actually recorded in the\nadministrative data. Understanding this limitation, the average per capita value of the ordinary\ntransfers was imputed, so that the household survey yields COP586 billion in ordinary transfers\nfrom the program. This, in contrast to the outliers reported in the survey, would have resulted in\nan overestimation of the total amount allocated.\n\n\n**<u>Table A16. Jóvenes en Acción transfer program. Beneficiaries and subsidy amount</u>**\n\n<u><mark>Panel A: Household surveys</mark></u>\n\n\n\nReported in survey\nBeneficiaries\n\n<u>(COP)</u>\n\n\n72\n\n\n\nAssigned in survey\n\n<u>(COP)</u>\n\nMention: \"GEIH\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"One of the sources behind the subsidy amounts tabulated for the pension solidarity programme.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000584:73:1", "occurrence_id": "5d91ee13a904a98621da88a7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Total subsidy amount</u>\n\n1,053,699,411,307\n1,503,939\n2017\n\n2,534,649,735,000\n<u>2021</u> 1,776,137\n\n_Source_ : DPS, Pension Solidarity Fund, GEIH, ENPH. *Simulated values.\n\nC. Jóvenes en Acción (“Youth in Action”)\n\n\nThe amount of the conditional transfer to young people in the program depends on where they\nstudy (SENA or HEI), as well as their academic performance. The cost of Jóvenes en Acción for\n2020 was COP586,742,148,000 and benefited more than 494,000 young people.\n\nHowever, the GEIH underestimates the beneficiaries of the Jóvenes en Acción program (155,903),\nso potential young people who could receive the benefit were identified. They are between 16\nand 28 years old, declare in the survey that they receive government aid, and are in poverty\naccording to the Sisbén IV level. This adjustment leads to the identification of 337,000 additional\nyoung people, which represents 68 percent of the population actually recorded in the\nadministrative data. Understanding this limitation, the average per capita value of the ordinary\ntransfers was imputed, so that the household survey yields COP586 billion in ordinary transfers\nfrom the program. This, in contrast to the outliers reported in the survey, would have resulted in\nan overestimation of the total amount allocated.\n\n\n**<u>Table A16. Jóvenes en Acción transfer program. Beneficiaries and subsidy amount</u>**\n\n<u><mark>Panel A: Household surveys</mark></u>\n\n\n\nReported in survey\nBeneficiaries\n\n<u>(COP)</u>\n\n\n72\n\n\n\nAssigned in survey\n\n<u>(COP)</u>\n\nMention: \"Sisbén IV\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Its poverty classification identifies potential young beneficiaries not captured elsewhere.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000584:73:1", "occurrence_id": "6a57d0cb69b4a2802cfd9b36", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Total subsidy amount</u>\n\n1,053,699,411,307\n1,503,939\n2017\n\n2,534,649,735,000\n<u>2021</u> 1,776,137\n\n_Source_ : DPS, Pension Solidarity Fund, GEIH, ENPH. *Simulated values.\n\nC. Jóvenes en Acción (“Youth in Action”)\n\n\nThe amount of the conditional transfer to young people in the program depends on where they\nstudy (SENA or HEI), as well as their academic performance. The cost of Jóvenes en Acción for\n2020 was COP586,742,148,000 and benefited more than 494,000 young people.\n\nHowever, the GEIH underestimates the beneficiaries of the Jóvenes en Acción program (155,903),\nso potential young people who could receive the benefit were identified. They are between 16\nand 28 years old, declare in the survey that they receive government aid, and are in poverty\naccording to the Sisbén IV level. This adjustment leads to the identification of 337,000 additional\nyoung people, which represents 68 percent of the population actually recorded in the\nadministrative data. Understanding this limitation, the average per capita value of the ordinary\ntransfers was imputed, so that the household survey yields COP586 billion in ordinary transfers\nfrom the program. This, in contrast to the outliers reported in the survey, would have resulted in\nan overestimation of the total amount allocated.\n\n\n**<u>Table A16. Jóvenes en Acción transfer program. Beneficiaries and subsidy amount</u>**\n\n<u><mark>Panel A: Household surveys</mark></u>\n\n\n\nReported in survey\nBeneficiaries\n\n<u>(COP)</u>\n\n\n72\n\n\n\nAssigned in survey\n\n<u>(COP)</u>\n\nMention: \"administrative data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"The reconstructed population covers only about two thirds of those actually recorded administratively.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000584:73:1", "occurrence_id": "c7e39595f9ba3d7aa9052008", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Total subsidy amount</u>\n\n1,053,699,411,307\n1,503,939\n2017\n\n2,534,649,735,000\n<u>2021</u> 1,776,137\n\n_Source_ : DPS, Pension Solidarity Fund, GEIH, ENPH. *Simulated values.\n\nC. Jóvenes en Acción (“Youth in Action”)\n\n\nThe amount of the conditional transfer to young people in the program depends on where they\nstudy (SENA or HEI), as well as their academic performance. The cost of Jóvenes en Acción for\n2020 was COP586,742,148,000 and benefited more than 494,000 young people.\n\nHowever, the GEIH underestimates the beneficiaries of the Jóvenes en Acción program (155,903),\nso potential young people who could receive the benefit were identified. They are between 16\nand 28 years old, declare in the survey that they receive government aid, and are in poverty\naccording to the Sisbén IV level. This adjustment leads to the identification of 337,000 additional\nyoung people, which represents 68 percent of the population actually recorded in the\nadministrative data. Understanding this limitation, the average per capita value of the ordinary\ntransfers was imputed, so that the household survey yields COP586 billion in ordinary transfers\nfrom the program. This, in contrast to the outliers reported in the survey, would have resulted in\nan overestimation of the total amount allocated.\n\n\n**<u>Table A16. Jóvenes en Acción transfer program. Beneficiaries and subsidy amount</u>**\n\n<u><mark>Panel A: Household surveys</mark></u>\n\n\n\nReported in survey\nBeneficiaries\n\n<u>(COP)</u>\n\n\n72\n\n\n\nAssigned in survey\n\n<u>(COP)</u>\n\nMention: \"household survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Reported transfer outliers are replaced with imputed averages so the survey yields a plausible total.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000585:10:0", "occurrence_id": "10d7bbabc191ee928242db8d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Figure IV.2: Female decision-making index in sampled districts in 2014_\n\n\n_Figure IV.3: Female decision-making index in sampled districts in 2021_\n\n\nb. Conflict event data: ACLED\n\n\nData on conflict come from the Armed Conflict and Location Event Database (ACLED) which codes\nthe location and date of a wide range of conflict-related events. We use this geolocated\ninformation to assign conflict data to subnational regions at the ADM 2 level (i.e., district). While\nthere have been some violent events before the start of the current conflict, we select the year(s)\n\n\n9\n\nMention: \"Armed Conflict and Location Event Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Geolocated conflict events are assigned to second-level administrative districts.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000592:45:3", "occurrence_id": "2dcc460ded8d63be78015ade", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  time trend. Marginal effects are derived from estimation of the probability of a worker having one of three\ndurations of job tenure. Estimation of the probability of a worker having less than one year of job tenure is conducted\non a sample of respondents older than 20. The probability of a worker having 5–10 years of tenure is estimated on a\nsample of respondents 25 and older. The probability of having tenure of 10 or more years is estimated on a sample of\nworkers 30 and older. EU-LFS 1995-2020. Bootstrapped standard errors. <sup>***</sup> indicates that the coefficient is significant\nat 1% level, <sup>**</sup> - at 5% level, <sup>*</sup> - at 10% level.\n\n\n44\n\nMention: \"EU-LFS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Job tenure probabilities are estimated on age-specific samples drawn from this pooled survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000593:18:0", "occurrence_id": "003777c7d7c971a2bd4537e1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 12 - Average quality of Corporate ESG Responsibility, by size and region** (composite\nindicator, z-score)\n\n\nZ-score calculation is based on the Enterprise Survey sample of 41 countries. Corporate ESG _weighted_ : 40% for E, 25% for S, 35%\nfor G. CEE: Central and Eastern Europe; WB: Western Balkans; CA: Central Asia; EN: Eastern Neighborhood; MENA: Middle\nEast and North Africa. Source: Authors’ calculations based on Enterprise Survey data.\n\n\nThere are several factors that may explain the large differences in corporate ESG responsibility\ncomposite indicator scores across firms within a given country. Besides the firm’s sector, these\ninclude other firm characteristics such as size, age and ownership, as well as external factors such\nas customer or regulatory pressure, and experiencing environmental and climate change-related\nproblems.\n\n\nLooking at firms in MENA, customer pressure is the most important determinant of the quality\nof corporate ESG responsibility practices. Indeed, estimates from a firm-level regression (Figure\n11) indicate that firms whose customers require environmental certifications or adherence to\ncertain environmental standards as a condition of doing business with them on average have better\ncorporate ESG responsibility practices than firms whose customers do not require this. This is\nclosely followed by whether a firm is an exporter. Several other determinants also play a role, such\nas being a large, old firm, having experienced monetary losses due to pollution by others or due to\nextreme weather events, or having at least 25% foreign ownership. Such results regarding the\ndeterminants of the “Corporate ESG Responsibility” composite indicator for firms in MENA\ncountries are similar to those of the entire sample of emerging economies in the ES. The only\nrelevant differences appear being a listed firm (as listed companies have better ESG practices\namong emerging markets but not in MENA) and being subject to the energy tax (firms subject to\nenergy tax have better ESG performance in emerging markets but not in MENA).\n\n\n17\n\nMention: \"Enterprise Survey data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"A weighted environmental, social and governance composite is built from firm responses in this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000596:14:0", "occurrence_id": "c4610e9e78097102c491b15e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: AgGDP:\n\n\n_PreAgGDPi_ = _Cropi_ + _Livestocki_ + _Forestryi_ + _Fishingi_ + _Huntingi_ (8)\n\n\nwhere we assume hunting occurs in areas with equal probability.\n\n\nTheoretically, the sum of these components should be close to the official values obtained\n\nfrom the World Development Indicators. We make sure that the official AgGDP values are\n\nguaranteed to be no less than the sum of all five components of agricultural GDP. Therefore,\n\nwe first sum up all prior estimations of AgGDP.\n\n\n_AgGDPx_ = Σ _i∈xPreAgGDPi_ (9)\n\n\nThen, we rescale the prior _AgGDP_ to be consistent with the official AgGDP value:\n\n\n\n_PriorAgGDPi_ = <sup>_<u>PreAgGDPiAgGDPx</u>_</sup>\n\nΣ _iPreAgGDPx_\n\n\nThen we calculate the prior for Si as a probability by normalizing PriorAgGDP:\n\n\n_<u>PriorAgGDPi,x</u>_\n_PreAlloci_ =\nΣ _i∈XPriorAgGDPi_\n\n\n\n(10)\n\n\n(11)\n\n\n\nFinally, we formulate a cross entropy model in the following mathematical optimization\n\nframework:\n\n\n_MIN_ _CE_ ( _Si_ ) = Σ _iSilog_ ( _Si_ ) _−_ Σ _iSilog_ ( _PreAlloci_ ) (12)\n\n\nSubject to the following three conditions:\n\n\nΣ _iSi_ = 1 (13)\n\n\nΣ _i∈k_ (Σ _AgGDP_ ) _Si_ = _SubAgGDPk_ _∀k_ ) (14)\n\n\n0 _≤_ _Si_ _≤_ 1 _∀i_ (15)\n\n\nwhere _i_ : i=1,2,3,. . .\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Official agricultural output values are used as a floor for the sum of the five modelled components.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000597:17:0", "occurrence_id": "705a3ce3affd667a47495dca", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: the interaction terms remain positively and significantly correlated with financial inclusion in\n\n\nall regressions. Hence, our findings are unlikely to be confounded by other cultural factors.\n\n\n**5.2. Alternative dependent variable**\n\n\nBeyond the simple ownership of bank accounts, Demirgüç-Kunt and Klapper (2013)\n\n\nhighlight the importance of considering their usage. The authors explain that 10 percent of\n\n\naccount owners in developing economies maintained an inactive account, i.e., “they make\n\n\nneither withdrawals from nor deposits into their account”. Moreover, Aterido et al. (2011) show\n\n\nthat the use of financial services also varies significantly between men and women by\n\n\nconsidering nine Sub-Saharan countries. As suggested by Duflo (2012), women are relatively\n\n\nless inclined than men to use the account if it was easy for their husbands to get the money out.\n\n\nThus, one may argue that having an account does allow for women’s empowerment, whereas\n\n\nusing the account to achieve development goals may be more relevant for accessing economic\n\n\nopportunities. This is particularly relevant in India, where despite the massive growth in\n\n\nownership, 43 percent of accounts remain inactive, the highest rate in the world according to\n\n\nthe 2021 Global Findex report.\n\n\nThus, we consider an alternative measure of financial inclusion to capture the frequency\n\n\nof account use. We focus on withdrawals as suggested by Allen et al. (2016), because such\n\n\noperations are actively initiated by account owners while deposits might be initiated by others\n\n\n(for example, employers or governments). Account holders answer the following question: “In\n\n\na typical month, about how many times is money taken out of your personal account(s)?”\n\n\nRespondents are asked (categorically) if they conducted: ( _i_ ) 0 withdrawals, ( _ii_ ) 1–2\n\n\nwithdrawals, ( _iii_ ) 3–5 withdrawals, or ( _iv_ ) 6 or more withdrawals on average per month.\n\nMention: \"Findex report\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The finding that 43 percent of accounts remain inactive is quoted from this report.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000599:13:1", "occurrence_id": "21b29849eeccaab537a3a959", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: very substantial, equaling</mark>\n\n14.5 percentage points, and is statistically significant at the 1% level. In column 2 of the same\n\ntable, we look at the proportion of individuals having worked as usual in their wage job in the\n\nprevious week (including remotely). Workers in FCV areas were on average about 10 percentage\n\npoints more likely to work at their usual jobs, although the difference is not precisely estimated\n\nand therefore not statistically significant. In contrast, there is almost no difference between FCV\n\nand non-FCV areas when looking at food insecurity, as measured by adults going for an entire day\n\nwithout eating. This finding might seem surprising at first sight given our results regarding income\n\nchanges. Nevertheless, a plausible explanation for this is that people in FCV areas, due to the\n\neconomic conditions typically prevailing there, have developed effective mechanisms over the\n\nyears to safeguard their level of food security in the face of transitory shocks. Alternatively, this\n\nfinding may simply be due to the greater provision of humanitarian aid to FCV areas, but the\n\ninformation available in HFPS does not permit us to further probe into this possibility.\n\n\n12\n\nMention: \"HFPS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its information is insufficient to distinguish coping strategies from humanitarian aid effects.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000601:5:0", "occurrence_id": "7de1442158c77ef8a5df73c5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: on theory suggesting that there is preference for a large potential award, even if the chances of\nwinning are low, compared to a guaranteed small award. <sup>43,44</sup>\n\nBaseline data were collected from August 2018-February 2019. Data collection was paused\nfrom March-October 2020 due to COVID-19; endline data collection was completed from June\n2021-January 2022. Surveys were conducted by research assistants from Innovations for Poverty\nAction Tanzania in Kiswahili on tablets. All surveys were administered in private spaces at a clinic\nand included an assessment of past 3-month physical and sexual violence.\n\n_Outcomes_\n\nIn this pre-specified <sup>45</sup> secondary analysis of RESPECT II data, we estimated the effect of\nthe lottery intervention on GBV, expressed as the proportion of FSW who experienced physical\nand/or sexual violence perpetrated by a spouse, boyfriend, or other sexual partner in the past three\nmonths. This outcome was measured as a binary variable and was assessed through an endline\nsurvey at 36 months.\n\nWe also investigated the effect of the lottery intervention on two distinct forms of GBV:\n1) IPV, defined as physical and/or sexual violence perpetrated by a boyfriend or spouse; and 2)\nphysical and/or sexual violence perpetrated by a non-partner (i.e., other sexual partners including,\nbut not limited to clients). These forms of GBV are not mutually exclusive and are both\ncomponents of the GBV indicator described above.\n\n_Sample size and power_\n\nThe sample size for this study was pre-determined based on the RESPECT II trial, which\nenrolled 2,206 participants at baseline to achieve 80% power to detect a 30-percentage point\ndifference in the primary trial outcome of combined HIV/HSV2 incidence. In this analysis, after\naccounting for anticipated endline attrition and estimated baseline reports of past 3-month\n\nMention: \"RESPECT II data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"A pre-specified secondary analysis estimates the lottery's effect on violence against women from this trial's survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000603:8:0", "occurrence_id": "3eaac87fc250811f54dc881f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: It is important to note that, while the results of this study provide some evidence on the efficacy of\n\n\ngasoline and diesel price changes, many other avenues for reducing air pollution are available. Non\n\nprice measures, such as technology mandates and catalyzer regulations, are underused and could\n\n\nimprove air quality without diminishing consumption levels (UNEP 2021). And while 64 percent of\n\n\ncountries have ambient air quality standards in legislative instruments, their distribution is lopsided\n\n\ntoward developed countries. For example, while all EU countries have such regulation, only 40 percent\n\n\nof Commonwealth countries do. Therefore, encouraging the implementation of air quality\n\n\nlegislation—such as mandatory filters and fuel mileage requirements for cars—could improve urban\n\n\nair quality. A single price adjustment alone will not curb air pollution; rather, a combination of\n\n\nmeasures need to work in tandem. This paper documents the role prices could play.\n\n###### 3. Data and methods 3.1. Data\n\n\nTo evaluate the relationship between fossil fuel prices and air pollution, we combine three datasets\n\n\nto construct a panel dataset with PM2.5 readings, GDP per capita, and fossil fuel prices and\n\n\nconsumption for 133 countries.\n\n\nFirst, we take surface PM2.5 concentrations in μg/m³ from van Donkeelar et al. (2021), who estimate\n\n\nmean annual air pollution levels globally using satellite imagery from 1998 to 2019. PM2.5 measures\n\n\nanthropogenic fine particulate pollution, such as exhaust emissions from fuel combustion, and is\n\n\nadjusted to remove naturally occurring particles, such as desert dust and sea salt. As it is possible to\n\n\nchoose among a host of locations within a country that could affect readings, we homogenize our data\n\n\nto the center of each country’s capital city.\n\n\nSecond, we merge GDP per capita values, measured in 2017 international $ adjusted for purchasing\n\n\npower parity (PPP), from the World Bank’s World Development Indicators database (World Bank\n\n\n2021a).\n\n\nThird, we use\n\nMention: \"World Development Indicators database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"GDP per capita in purchasing power adjusted terms is merged in as a country control.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000611:5:0", "occurrence_id": "e603a3bda69982599cad4239", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: data on occupational skills profiles is likely to proxy the other three Southeast Asian countries in the\n\n\nstudy more accurately than occupational skills profiles from the U.S. Second, the analysis identifies\n\n\nskills requirements from unstructured data – job postings – rather than from a predefined list of skills\n\n\nor tasks (as done in a firm survey, for example), allowing the data to “speak for itself” when creating\n\n\nskills categories.\n\n\nThe analysis has three conclusions. First, virtually the entire employed population in the four-country\n\n\nsample are in low- and medium-digital occupations. Around two-thirds of the employed population\n\n\nis in low-digital occupations, which means that their tasks require limited use of even the most basic\n\n\ndigital technologies. Less than 3 percent work in highly digital occupations, which require frequent\n\n\nuse of digital technologies.\n\n\nSecond, highly digital occupations require similar levels and kinds of cognitive and socioemotional\n\n\nskills as do low- and medium- digital occupations. Three exercises confirm this conclusion. Using data\n\n\nfrom Malaysia and regression analysis, we observe that low-, medium-, and highly digital occupations\n\n\nhave different shares of digital skills but roughly the same shares of cognitive and socioemotional\n\n\nskills. The few exceptions are that medium- and high-digital occupations more frequently require\n\n\nthinking skills (one of the three cognitive skills subsets that we explore) and low-digital occupations\n\n\nmore frequently require relationship skills (one of the three socioemotional skills subsets in our\n\n\nsample). Furthermore, pairwise correlations and an exploratory factor analysis confirm\n\n\ncomplementarities between levels of digital skills and subsets of cognitive and socioemotional skills.\n\n\nSimilar analysis using occupational data from Cambodia, Thailand, and Vietnam show similar results.\n\n\nThird, all occupations are digital to some degree. In Malaysia, the share of skills that are digital in\n\n\nvery low-, low-, and medium-digital occupations are 5, 9, and 15 percent. At least 1 percent of skills\n\n\nin every occupation in Malaysia is digital, ranging from 1\n\nMention: \"unstructured data – job postings\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Skills requirements are derived from advertised postings rather than a predefined task list.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000611:5:0", "occurrence_id": "a1ca70dc6bb6f04a84bd5628", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: data on occupational skills profiles is likely to proxy the other three Southeast Asian countries in the\n\n\nstudy more accurately than occupational skills profiles from the U.S. Second, the analysis identifies\n\n\nskills requirements from unstructured data – job postings – rather than from a predefined list of skills\n\n\nor tasks (as done in a firm survey, for example), allowing the data to “speak for itself” when creating\n\n\nskills categories.\n\n\nThe analysis has three conclusions. First, virtually the entire employed population in the four-country\n\n\nsample are in low- and medium-digital occupations. Around two-thirds of the employed population\n\n\nis in low-digital occupations, which means that their tasks require limited use of even the most basic\n\n\ndigital technologies. Less than 3 percent work in highly digital occupations, which require frequent\n\n\nuse of digital technologies.\n\n\nSecond, highly digital occupations require similar levels and kinds of cognitive and socioemotional\n\n\nskills as do low- and medium- digital occupations. Three exercises confirm this conclusion. Using data\n\n\nfrom Malaysia and regression analysis, we observe that low-, medium-, and highly digital occupations\n\n\nhave different shares of digital skills but roughly the same shares of cognitive and socioemotional\n\n\nskills. The few exceptions are that medium- and high-digital occupations more frequently require\n\n\nthinking skills (one of the three cognitive skills subsets that we explore) and low-digital occupations\n\n\nmore frequently require relationship skills (one of the three socioemotional skills subsets in our\n\n\nsample). Furthermore, pairwise correlations and an exploratory factor analysis confirm\n\n\ncomplementarities between levels of digital skills and subsets of cognitive and socioemotional skills.\n\n\nSimilar analysis using occupational data from Cambodia, Thailand, and Vietnam show similar results.\n\n\nThird, all occupations are digital to some degree. In Malaysia, the share of skills that are digital in\n\n\nvery low-, low-, and medium-digital occupations are 5, 9, and 15 percent. At least 1 percent of skills\n\n\nin every occupation in Malaysia is digital, ranging from 1\n\nMention: \"data\n\n\nfrom Malaysia\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Regression analysis on this country's postings tests whether digital occupations differ in skill composition.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000611:5:0", "occurrence_id": "9609eb3122d21fba96cbcbba", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: data on occupational skills profiles is likely to proxy the other three Southeast Asian countries in the\n\n\nstudy more accurately than occupational skills profiles from the U.S. Second, the analysis identifies\n\n\nskills requirements from unstructured data – job postings – rather than from a predefined list of skills\n\n\nor tasks (as done in a firm survey, for example), allowing the data to “speak for itself” when creating\n\n\nskills categories.\n\n\nThe analysis has three conclusions. First, virtually the entire employed population in the four-country\n\n\nsample are in low- and medium-digital occupations. Around two-thirds of the employed population\n\n\nis in low-digital occupations, which means that their tasks require limited use of even the most basic\n\n\ndigital technologies. Less than 3 percent work in highly digital occupations, which require frequent\n\n\nuse of digital technologies.\n\n\nSecond, highly digital occupations require similar levels and kinds of cognitive and socioemotional\n\n\nskills as do low- and medium- digital occupations. Three exercises confirm this conclusion. Using data\n\n\nfrom Malaysia and regression analysis, we observe that low-, medium-, and highly digital occupations\n\n\nhave different shares of digital skills but roughly the same shares of cognitive and socioemotional\n\n\nskills. The few exceptions are that medium- and high-digital occupations more frequently require\n\n\nthinking skills (one of the three cognitive skills subsets that we explore) and low-digital occupations\n\n\nmore frequently require relationship skills (one of the three socioemotional skills subsets in our\n\n\nsample). Furthermore, pairwise correlations and an exploratory factor analysis confirm\n\n\ncomplementarities between levels of digital skills and subsets of cognitive and socioemotional skills.\n\n\nSimilar analysis using occupational data from Cambodia, Thailand, and Vietnam show similar results.\n\n\nThird, all occupations are digital to some degree. In Malaysia, the share of skills that are digital in\n\n\nvery low-, low-, and medium-digital occupations are 5, 9, and 15 percent. At least 1 percent of skills\n\n\nin every occupation in Malaysia is digital, ranging from 1\n\nMention: \"occupational data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"The same analysis run on three further countries reproduces the skill complementarity results.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000612:9:1", "occurrence_id": "5a1a77abe1a2bdf6bb7b4fb0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ://github.com/takaakimasaki/DisplacementDataGap)</u>\n24 The oldest dataset available in WB/UNHCR MDL was collected in 1995 at the time of this study.\n25 See more recent developments in UNHCR MDL: <u>[https://www.unhcr.org/blogs/responsible-and-timely-sharing-](https://www.unhcr.org/blogs/responsible-and-timely-sharing-data-on-unhcrs-microdata-library-in-2022/)</u>\n<u>[data-on-unhcrs-microdata-library-in-2022/.](https://www.unhcr.org/blogs/responsible-and-timely-sharing-data-on-unhcrs-microdata-library-in-2022/)</u>\n\nMention: \"WB/UNHCR MDL\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"A footnote notes the oldest dataset available in this joint microdata library dates to 1995.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000617:6:0", "occurrence_id": "7114680b4afcaf04ee1dd540", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 1: Log of Number of Connected Firms.\n\n###### **2.2 PTA Data**\n\n\nThe Deep Trade Agreement Dataset of the World Bank is the most comprehensive\ndatabase of PTAs. It includes 279 treaties from 1958 to 2015. Even more interestingly, this database also includes a full text analysis of each treaty and a detailed\ncodification on the inclusion of 52 different provisions.\n\nGiven our empirical specification, which will include country pair fixed effects, we\ncan only identify the effects of those treaties that came into force during the period\n2007-2015. Nevertheless, it is the case that this includes almost half of the treaties\n(112 new treaties) that generate around 4,000 new dyadic relations (see table 1) that\ninvolve a rich variety of countries (see table 2).\n\n\n5\n\nMention: \"Deep Trade Agreement Dataset\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its codification of 279 treaties and full-text analysis supplies the preferential trade agreement variables.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000618:26:1", "occurrence_id": "da350318d408acec803a6409", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ,366 60,654 195,809\nR-squared 0.102 0.133 0.099 0.163 0.102 0.147\nEvent FE Yes Yes Yes Yes Yes Yes\nYear FE Yes No Yes No Yes Yes\nCountry*Year FE No Yes No Yes No No\nNever treated Regions No Yes No Yes No No\nDep Var Mean 0.611 0.611 0.597 0.596 0.612 0.618\nEvents 150 150 37 37 108 108\nYears 12 12 12 12 12 12\nRegions 132 697 34 232 104 529\nCountries 41 41 18 18 35 35\n\n\nNotes:    - significant at 10%; ** significant at 5%; *** significant at 1%. The dependent variable is\nthe attitudes towards immigrants question, equal to 1 if the respondent answers yes to the question\n“is the city or area where you live a good place to live for immigrants from other countries?”, 0 for\ndisagreement, and missing for blanks or refusals. Regional controls for the inverse hyperbolic sine\ntransformation of region population and refugee population at the year of the event, and individual\ncontrols for age, age squared, sex, educational level and city size were used. The sample includes\nindividual respondents to the Gallup World Poll in region-years (columns 1, 3, and 5) and countryyears (columns 2 and 6) who were surveyed within 4 years before or after an event. An event is\ndefined as a region with an increase of at least 5,000 refugees (row 1), 10,000 refugees (row 2) or\n50,000 refugees (row 3) in a calendar year. Results use the Gallup sampling weights and standard\nerrors are clustered at the event level.\n\n\n24\n\nMention: \"Gallup World Poll\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Individual respondents near refugee influx events form the sample for the attitude regressions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000621:16:0", "occurrence_id": "6f461e5ae510dae961494466", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: industry, and region—collected data for Egypt in 2004, 2007, 2008, 2013, 2016, and 2020. Each\n\n\nsurvey year is representative of the business environment that year. Of these, this analysis will\n\n\nfocus on a panel of 198 firms in 2013, 2016, and 2020. See Table 2 for mean values of our main\n\n\nvariables. <sup>9</sup> [^9: Appendix A reports descriptive statistics for firms that survived between 2016 and 2020 with those that did not.\nOverall, the firms in both groups are broadly similar except in terms of size. Larger firms are more likely to survive.]\n\n\nWe supplement the ES data with information on the universe of firms captured by the 2017\n\n\nCensus of Establishments in Egypt and with data from the United Nations International Trade\n\n\nStatistics Database (UN COMTRADE), specifically information on world imports from Egypt\n\n\nover the period of study.\n\n\n**_4.2_** **_Stylized Facts of Egypt_**\n\n\n**_a._** **_Exporter Characteristics_**\n\n\nWe begin the data analysis at the aggregate level, by comparing total Egyptian exports\n\n\nfrom the ES with world imports from Egypt from UN COMTRADE. As Figure 5 shows, the ES\n\n\ndata represent a reasonable proxy for Egypt’s exports to the world. In 2020, wood manufacturing\n\n\n(ISIC=20) and vehicle manufacturing (ISIC = 34) stand out as the only two sectors over\n\nrepresented in the ES relative to the UN COMTRADE data.\n\n\nFigure 5 aggregates exports from all ES firms. Looking at the data more closely, we see\n\n\nthat few Egyptian firms export, either directly or indirectly (Table 3). <sup>10</sup> [^10: Firms exporting directly are those that sell to other countries without intermediaries. Firms exporting indirectly\nare those which sell to a wholesaler or other who then resells to customers in other countries.] In 2020, only 27 of the 198\n\n\nfirms in the panel exported.\n\n\n**_b.\n\nMention: \"ES data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Surviving and non-surviving firms are compared on characteristics from these survey rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000621:16:0", "occurrence_id": "a9bc9751db918531f5047096", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: industry, and region—collected data for Egypt in 2004, 2007, 2008, 2013, 2016, and 2020. Each\n\n\nsurvey year is representative of the business environment that year. Of these, this analysis will\n\n\nfocus on a panel of 198 firms in 2013, 2016, and 2020. See Table 2 for mean values of our main\n\n\nvariables. <sup>9</sup> [^9: Appendix A reports descriptive statistics for firms that survived between 2016 and 2020 with those that did not.\nOverall, the firms in both groups are broadly similar except in terms of size. Larger firms are more likely to survive.]\n\n\nWe supplement the ES data with information on the universe of firms captured by the 2017\n\n\nCensus of Establishments in Egypt and with data from the United Nations International Trade\n\n\nStatistics Database (UN COMTRADE), specifically information on world imports from Egypt\n\n\nover the period of study.\n\n\n**_4.2_** **_Stylized Facts of Egypt_**\n\n\n**_a._** **_Exporter Characteristics_**\n\n\nWe begin the data analysis at the aggregate level, by comparing total Egyptian exports\n\n\nfrom the ES with world imports from Egypt from UN COMTRADE. As Figure 5 shows, the ES\n\n\ndata represent a reasonable proxy for Egypt’s exports to the world. In 2020, wood manufacturing\n\n\n(ISIC=20) and vehicle manufacturing (ISIC = 34) stand out as the only two sectors over\n\nrepresented in the ES relative to the UN COMTRADE data.\n\n\nFigure 5 aggregates exports from all ES firms. Looking at the data more closely, we see\n\n\nthat few Egyptian firms export, either directly or indirectly (Table 3). <sup>10</sup> [^10: Firms exporting directly are those that sell to other countries without intermediaries. Firms exporting indirectly\nare those which sell to a wholesaler or other who then resells to customers in other countries.] In 2020, only 27 of the 198\n\n\nfirms in the panel exported.\n\n\n**_b.\n\nMention: \"2017\n\n\nCensus of Establishments in Egypt\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The universe of establishments is used to check how well the survey covers the firm population.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000621:16:0", "occurrence_id": "6180a420b76d060180d8903d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: industry, and region—collected data for Egypt in 2004, 2007, 2008, 2013, 2016, and 2020. Each\n\n\nsurvey year is representative of the business environment that year. Of these, this analysis will\n\n\nfocus on a panel of 198 firms in 2013, 2016, and 2020. See Table 2 for mean values of our main\n\n\nvariables. <sup>9</sup> [^9: Appendix A reports descriptive statistics for firms that survived between 2016 and 2020 with those that did not.\nOverall, the firms in both groups are broadly similar except in terms of size. Larger firms are more likely to survive.]\n\n\nWe supplement the ES data with information on the universe of firms captured by the 2017\n\n\nCensus of Establishments in Egypt and with data from the United Nations International Trade\n\n\nStatistics Database (UN COMTRADE), specifically information on world imports from Egypt\n\n\nover the period of study.\n\n\n**_4.2_** **_Stylized Facts of Egypt_**\n\n\n**_a._** **_Exporter Characteristics_**\n\n\nWe begin the data analysis at the aggregate level, by comparing total Egyptian exports\n\n\nfrom the ES with world imports from Egypt from UN COMTRADE. As Figure 5 shows, the ES\n\n\ndata represent a reasonable proxy for Egypt’s exports to the world. In 2020, wood manufacturing\n\n\n(ISIC=20) and vehicle manufacturing (ISIC = 34) stand out as the only two sectors over\n\nrepresented in the ES relative to the UN COMTRADE data.\n\n\nFigure 5 aggregates exports from all ES firms. Looking at the data more closely, we see\n\n\nthat few Egyptian firms export, either directly or indirectly (Table 3). <sup>10</sup> [^10: Firms exporting directly are those that sell to other countries without intermediaries. Firms exporting indirectly\nare those which sell to a wholesaler or other who then resells to customers in other countries.] In 2020, only 27 of the 198\n\n\nfirms in the panel exported.\n\n\n**_b.\n\nMention: \"United Nations International Trade\n\n\nStatistics Database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"World imports from Egypt over the study period are drawn from this trade database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000621:16:0", "occurrence_id": "87037cf2ce2b9ea91aecb562", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: industry, and region—collected data for Egypt in 2004, 2007, 2008, 2013, 2016, and 2020. Each\n\n\nsurvey year is representative of the business environment that year. Of these, this analysis will\n\n\nfocus on a panel of 198 firms in 2013, 2016, and 2020. See Table 2 for mean values of our main\n\n\nvariables. <sup>9</sup> [^9: Appendix A reports descriptive statistics for firms that survived between 2016 and 2020 with those that did not.\nOverall, the firms in both groups are broadly similar except in terms of size. Larger firms are more likely to survive.]\n\n\nWe supplement the ES data with information on the universe of firms captured by the 2017\n\n\nCensus of Establishments in Egypt and with data from the United Nations International Trade\n\n\nStatistics Database (UN COMTRADE), specifically information on world imports from Egypt\n\n\nover the period of study.\n\n\n**_4.2_** **_Stylized Facts of Egypt_**\n\n\n**_a._** **_Exporter Characteristics_**\n\n\nWe begin the data analysis at the aggregate level, by comparing total Egyptian exports\n\n\nfrom the ES with world imports from Egypt from UN COMTRADE. As Figure 5 shows, the ES\n\n\ndata represent a reasonable proxy for Egypt’s exports to the world. In 2020, wood manufacturing\n\n\n(ISIC=20) and vehicle manufacturing (ISIC = 34) stand out as the only two sectors over\n\nrepresented in the ES relative to the UN COMTRADE data.\n\n\nFigure 5 aggregates exports from all ES firms. Looking at the data more closely, we see\n\n\nthat few Egyptian firms export, either directly or indirectly (Table 3). <sup>10</sup> [^10: Firms exporting directly are those that sell to other countries without intermediaries. Firms exporting indirectly\nare those which sell to a wholesaler or other who then resells to customers in other countries.] In 2020, only 27 of the 198\n\n\nfirms in the panel exported.\n\n\n**_b.\n\nMention: \"UN COMTRADE data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Sectoral representation in the survey is compared with this trade series to judge its representativeness.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000625:11:0", "occurrence_id": "2c24731d2e1c02fc6c5a31ba", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ###### **3.2 Data**\n\nOur analysis considers all 25 administrative regions (oblasts) in Ukraine, including Kyiv\n\n\ncity as a special region. For the purposes of this paper, the pre-conflict period (2008-2012)\n\n\nis defined as the 5-year period before the Maidan protests in 2013, and the post-conflict\n\n\nperiod is between 2014 and 2019 (the last year of our data). All regional statistics for the\n\n\npre-conflict period, including population and average per capita income numbers, are based\n\n\non the official information gathered by the State Statistics Services of Ukraine (UkrStat).\n\n\nOur data also includes gross bilateral migration flows between oblasts for the same period,\n\n\nwhich were reported in UkrStat’s demographic yearbooks. Information on these flows reflects\n\n\nthe official registrations of residence recorded by the State Migration Service of Ukraine.\n\n\nDescriptive statistics show that Donetsk and Luhansk regions were among the most\n\n\npopulous oblasts in Ukraine before the onset of the conflict in 2014 (Figure 1b). With\n\n\n4.5 million and 2.3 million residents, they were ranked 1st and 7th among all regions,\n\n\nrespectively. They also had relatively high real gross regional product per capita (2nd and\n\n\n9th, respectively). Compared to other regions, however, migration flows into Donbas were\n\n\nrelatively small. Incoming migrants constituted only 0.26 percent of local population in\n\n\nDonetsk, and 0.28 percent in Luhansk, putting them among the least attractive regions\n\n\n(24th and 25th among all) for migrant arrivals (1c). Among the most important sources\n\n\nof migrants were neighboring regions. Donetsk, for example received 22.8 percent of all its\n\n\ninflows from Kharkiv, 20.7 percent from Luhansk, and 12.3 percent from Dnipropetrovsk.\n\n\nMigration outflows from Donbas were also small. While the nation-wide interregional\n\n\nmigration outflow rate stood at 0.56 percent\n\nMention: \"demographic yearbooks\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Gross bilateral migration flows between regions are reported in these official yearbooks.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000628:32:1", "occurrence_id": "9663b017fba0797e4496e84c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: </sup> <sup>k.</sup> <sup>_Baseline_</sup>\n\nmodel is run in the full sample, at 3 digits of sectoral disaggregation and including origin-destination, plant, government-region and\n\n\ngovernment-sector fixed effects. _Baseline_ _4d_ is the baseline model at 4 digits of sectoral disaggregation. _Gov-reg-sec_ adds government\n\nlevel _×_ destination _×_ sector fixed effects. _No_ _capital_ _buyers_ excludes government agencies in capital regions. _No_ _capitals_ _(B_ _or_ _S)_ excludes\n\n\ngovernment agencies as well as firms located in capital regions.\n\n\nand hence compete for it. Second, an important part of our analysis is done at the intensive\n\nmargin. That is, the intensive margin home bias is identified non-local plants already selling to\n\n\nboth government levels within a given region. Therefore, at least one agency in each government\n\n\nlevel already knows the plant and is buying procurement goods or services from it. And third,\n\n\nwe provide some further evidence to show that subnational governments tend to systematically\n\nbuy from local firms even when non-local firms are competing for a specific project. Although the\n\n\nTED website provides detailed information about the characteristics of the contracts, it does not\n\n\nprovide the identity of the bidders for the project that did not win. To overcome this limitation, we\n\n\nidentified around 50 subnational government agencies that started providing detailed information\n\nabout their projects in 2013. Putting all these agencies together, we were able to uncover the\n\n\nidentity and location of the firms competing for the same projects as well as their final rankings\n\n\nfor around 1,000 contracts over the 2013-2016 period.\n\n\nWe find that local firms, i.e., firms located within the same region, win around 40% of the\n\nauctions. Importantly, we also see that at least\n\nMention: \"TED website\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"It gives contract details but not losing bidders, so agency records had to be collected separately.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000629:17:2", "occurrence_id": "cbd7141c18dd024f2056a454", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  sign from the short-run effect.\n\n\nAnother example relates to the revenue effects of tariff reforms. Devarajan, Go, and Li (1999) show\nhow the fiscal impact of tariff liberalization also depends on the substitution elasticity between foreign and\ndomestic goods. Unless there is an upward shift in output productivity, a reduction in tariffs will invariably\ninvolve losses in revenue for much of the plausible range of the trade elasticities unless compensated by\nincreases in domestic taxes. There is no Laffer Curve for import tariffs.\n\n\n19 We also tried two altermative income per capita measurements from the World Bank WDI – the 2015 gross national income per capita using\nthe Atlas and purchasing power parity methods. The patterns of the scatter plots remain largely the same. Because some countries have missing\nobservations using these methods, we did not choose them.\n20 The R-squared of the regression line is about 0.50 in both cases. The regression coefficient in each case is positive and significant at\nprob=0.05.\n21 This is confirmed in the empirical estimation of substitution elasticities in Devarajan, Go and Li (1994).\n\n\n16\n\nMention: \"World Bank WDI\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Alternative income per capita measures are tested and leave the patterns unchanged, but have missing countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000630:13:0", "occurrence_id": "ed738111312ca3ac340b76c1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **_4.1.2._** **_Data_**\nData come from four sources, the World Bank Private Participation in Infrastructure Project\nDatabase (PPI), <sup>12</sup> [^12: Version 2019 H1, available at https://ppi.worldbank.org/en/ppidata.] the Polity IV Project, <sup>13</sup> [^13: Polity IV Project, Political Regime Characteristics and Transitions, 1800-2018, available at\n<u>[http://www.systemicpeace.org/inscrdata.html.](http://www.systemicpeace.org/inscrdata.html)</u>\n<u>[14 Available at http://datatopics.worldbank.org/world-development-indicators/.](http://datatopics.worldbank.org/world-development-indicators/)</u>] World Bank’s World Development Indicators (WDI), <sup>14</sup>\nand Laeven and Valencia (2018).\n\nThe PPI database includes data on project characteristics as they were agreed at the time of the\nsigning of the PPP contract or at the time of financial closure. These characteristics include the\ntype of project, sector, contract period, government level granting the contract, identities of the\nsponsors, types of government support, amount of investment commitments and financing\ninformation. The PPI also provides the current status of the project as active, concluded, distressed\nor canceled.\n\nA considerable number of projects in the PPI database lack the data for all characteristics. Missing\ndata on the variables essential for the analysis, namely contract period and the level of government\nwhich granted the contract, were added for all projects using the individual project descriptions\nprovided in the database if available.\n\nThe Polity IV data is used to identify the institutional characteristics of a country. To identify these\ncharacteristics, yearly executive recruitment, constraints on the executive and political competition\nconcept variables are used. The values\n\nMention: \"World Bank Private Participation in Infrastructure Project\nDatabase\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"One of four sources, providing project-level infrastructure contract information.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000630:13:0", "occurrence_id": "ca6f1192fd74f608b25aaf22", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **_4.1.2._** **_Data_**\nData come from four sources, the World Bank Private Participation in Infrastructure Project\nDatabase (PPI), <sup>12</sup> [^12: Version 2019 H1, available at https://ppi.worldbank.org/en/ppidata.] the Polity IV Project, <sup>13</sup> [^13: Polity IV Project, Political Regime Characteristics and Transitions, 1800-2018, available at\n<u>[http://www.systemicpeace.org/inscrdata.html.](http://www.systemicpeace.org/inscrdata.html)</u>\n<u>[14 Available at http://datatopics.worldbank.org/world-development-indicators/.](http://datatopics.worldbank.org/world-development-indicators/)</u>] World Bank’s World Development Indicators (WDI), <sup>14</sup>\nand Laeven and Valencia (2018).\n\nThe PPI database includes data on project characteristics as they were agreed at the time of the\nsigning of the PPP contract or at the time of financial closure. These characteristics include the\ntype of project, sector, contract period, government level granting the contract, identities of the\nsponsors, types of government support, amount of investment commitments and financing\ninformation. The PPI also provides the current status of the project as active, concluded, distressed\nor canceled.\n\nA considerable number of projects in the PPI database lack the data for all characteristics. Missing\ndata on the variables essential for the analysis, namely contract period and the level of government\nwhich granted the contract, were added for all projects using the individual project descriptions\nprovided in the database if available.\n\nThe Polity IV data is used to identify the institutional characteristics of a country. To identify these\ncharacteristics, yearly executive recruitment, constraints on the executive and political competition\nconcept variables are used. The values\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Macroeconomic country characteristics are added from this indicator series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000630:13:0", "occurrence_id": "8d4a329fe55c461f5ffdd543", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **_4.1.2._** **_Data_**\nData come from four sources, the World Bank Private Participation in Infrastructure Project\nDatabase (PPI), <sup>12</sup> [^12: Version 2019 H1, available at https://ppi.worldbank.org/en/ppidata.] the Polity IV Project, <sup>13</sup> [^13: Polity IV Project, Political Regime Characteristics and Transitions, 1800-2018, available at\n<u>[http://www.systemicpeace.org/inscrdata.html.](http://www.systemicpeace.org/inscrdata.html)</u>\n<u>[14 Available at http://datatopics.worldbank.org/world-development-indicators/.](http://datatopics.worldbank.org/world-development-indicators/)</u>] World Bank’s World Development Indicators (WDI), <sup>14</sup>\nand Laeven and Valencia (2018).\n\nThe PPI database includes data on project characteristics as they were agreed at the time of the\nsigning of the PPP contract or at the time of financial closure. These characteristics include the\ntype of project, sector, contract period, government level granting the contract, identities of the\nsponsors, types of government support, amount of investment commitments and financing\ninformation. The PPI also provides the current status of the project as active, concluded, distressed\nor canceled.\n\nA considerable number of projects in the PPI database lack the data for all characteristics. Missing\ndata on the variables essential for the analysis, namely contract period and the level of government\nwhich granted the contract, were added for all projects using the individual project descriptions\nprovided in the database if available.\n\nThe Polity IV data is used to identify the institutional characteristics of a country. To identify these\ncharacteristics, yearly executive recruitment, constraints on the executive and political competition\nconcept variables are used. The values\n\nMention: \"PPI database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"Missing contract period and granting government level were coded from individual project descriptions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000630:13:0", "occurrence_id": "f22562f9f36afd15db6b9054", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **_4.1.2._** **_Data_**\nData come from four sources, the World Bank Private Participation in Infrastructure Project\nDatabase (PPI), <sup>12</sup> [^12: Version 2019 H1, available at https://ppi.worldbank.org/en/ppidata.] the Polity IV Project, <sup>13</sup> [^13: Polity IV Project, Political Regime Characteristics and Transitions, 1800-2018, available at\n<u>[http://www.systemicpeace.org/inscrdata.html.](http://www.systemicpeace.org/inscrdata.html)</u>\n<u>[14 Available at http://datatopics.worldbank.org/world-development-indicators/.](http://datatopics.worldbank.org/world-development-indicators/)</u>] World Bank’s World Development Indicators (WDI), <sup>14</sup>\nand Laeven and Valencia (2018).\n\nThe PPI database includes data on project characteristics as they were agreed at the time of the\nsigning of the PPP contract or at the time of financial closure. These characteristics include the\ntype of project, sector, contract period, government level granting the contract, identities of the\nsponsors, types of government support, amount of investment commitments and financing\ninformation. The PPI also provides the current status of the project as active, concluded, distressed\nor canceled.\n\nA considerable number of projects in the PPI database lack the data for all characteristics. Missing\ndata on the variables essential for the analysis, namely contract period and the level of government\nwhich granted the contract, were added for all projects using the individual project descriptions\nprovided in the database if available.\n\nThe Polity IV data is used to identify the institutional characteristics of a country. To identify these\ncharacteristics, yearly executive recruitment, constraints on the executive and political competition\nconcept variables are used. The values\n\nMention: \"Polity IV data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Executive recruitment, constraints and competition variables identify a country's institutional characteristics.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000635:25:0", "occurrence_id": "9320d54109a934c3c2f294b0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: #### Figure 1. Global output growth and relative per capita incomes\n\nA. Actual output growth B. Potential output growth\n\n\n\nPercent 2000-2021 average\n\n8\n\n\n4\n\n\n0\n\n\n-4\n\n\nC. Per capita income in EMDEs relative to\nadvanced economies\n\nPercent\n20\n\n\n15\n\n\n10\n\n\n5\n\n\n0\n\n\n\nPercent\n\n10\n\n8\n\n6\n\n4\n\n2\n\n0\n\n\n\nPotential growth\n\nActual growth\n\n\n\nD. Per capita income relative to advanced\neconomies, 2022\n\nPercent\n\n25\n\n\n20\n\n\n15\n\n\n10\n\n\n5\n\n\n\n0\n\n\n\nEAP ECA LAC MNA SAR SSA\n\n\n\nSources: Penn World Tables, World Bank.\nNote: AEs = advanced economies; EAP = East Asia and Pacific, ECA = Europe and Central Asia,\nEMDE = emerging market and\ndeveloping economy, LAC = Latin America and the Caribbean, MNA = Middle East and North Africa,\nSAR = South Asia, SSA = SubSaharan Africa.\nA. Sample of 181 countries.\nB.-D Based on production function approach. GDP-weighted averages for a sample of 29 advanced\neconomies and 53 EMDEs.\nC. Per capita income differential between EMDEs and advanced economies are defined as the GDPweighted average of percent of\nGDP per capita of the EMDEs as a ratio of GDP-weighted average of advanced economies.\n\nMention: \"Penn World Tables\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Actual and potential output growth and relative per capita incomes are taken from these tables.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000638:6:1", "occurrence_id": "f80667976f10be5502b9a2a0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: izing factor for its own business. As such, it is likely to be accounted for when firms\n\n\nmake their decisions and form their own expectations about the future. <sup>5</sup>\n\n\nIn this regard, the WBES dataset represents the only available survey providing forward-looking expec\n\ntations on firms’ future sales for MENA regions. This is a critical piece of information that made our study\n\n\npossible and allows to shed light on the mechanism at stake. We employ a continuous measure for firms\n\n\nexpected sales growth in the next year (E( _Sales_ _growth_ )), as well as an ordinal measure capturing increasing,\n\n\ndecreasing, or stable expected earnings (respectively, E( _Sales_ _growth_ ): _Positive_, _Negative_, or _Stable_ ).\n\n\nWe match all this information with a wide set of financial characteristics of the company. First of all,\n\n\nwe exploit data on the availability of outstanding loans or credit lines ( _Loan_ _availability_ ) at the time of the\n\n\nsurvey, which is a synthetic measure for firms’ access to bank finance. Because we are interested in the\n\n\n3The original number of waves for countries in the MENA region ranges between two and five, from 2007 to 2020. However,\nonly in more recent years (i.e., the global dataset) firms are attached to a consistent panel identifier across waves that allows\nfor longitudinal analyses. For most countries, two waves with panel identifiers are available: 2013–2019 for Jordan, Lebanon,\nMorocco, West Bank and Gaza, and 2013–2020 for Tunisia. The only exception is Egypt, for which three waves are available:\n2013, 2016, and 2020. We account for differences across countries and timing with the inclusion of country-specific time fixed\neffects that purge the model from economy-wide factors that vary over time.\n\n4We collapsed the two answers because there is no clear ranking between\n\nMention: \"WBES dataset\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"It is described as the only survey providing firms' forward-looking sales expectations for the region.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000639:39:3", "occurrence_id": "f846eb2517105db0e95ca558", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: *2** Cameroon only has four datapoints, in 2000-2002 and 2008.\n\n**3** We use census data from the LMSM survey for Ethiopia because the UNIDO data from 2014-2015 are estimates,\nand the census data extend past 2015 allowing us to include 2016-17. The LMSM series match well to the UNIDO\nseries pre-2010 and the LMSM is the source of the UNIDO data.\n\n**4** Lao’s export data is only available from 2010-2019.\n**5** Lesotho only has employment data from 2001-2009, but not output or value-added data--we therefore only\nreport employment and export (2000-2017) growth results.\n**6** Nigeria does not have UNIDO INDSTAT2 data post-2000, so we estimated formal employment from a report\nreleased by Nigeria’s National Bureau of Statistics that has employment statistics for 2010-2012 (NBS, 2014); we\nextrapolated values up through 2017 using a linear trend. Because of the limited data availability in UNIDO, we\nonly report employment and export (2001-2019) growth results.\n\n\n38\n\nMention: \"UNIDO data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its recent years are estimates rather than census observations, so survey census data are preferred.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000639:39:3", "occurrence_id": "35bcb20739f0378008793408", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: *2** Cameroon only has four datapoints, in 2000-2002 and 2008.\n\n**3** We use census data from the LMSM survey for Ethiopia because the UNIDO data from 2014-2015 are estimates,\nand the census data extend past 2015 allowing us to include 2016-17. The LMSM series match well to the UNIDO\nseries pre-2010 and the LMSM is the source of the UNIDO data.\n\n**4** Lao’s export data is only available from 2010-2019.\n**5** Lesotho only has employment data from 2001-2009, but not output or value-added data--we therefore only\nreport employment and export (2000-2017) growth results.\n**6** Nigeria does not have UNIDO INDSTAT2 data post-2000, so we estimated formal employment from a report\nreleased by Nigeria’s National Bureau of Statistics that has employment statistics for 2010-2012 (NBS, 2014); we\nextrapolated values up through 2017 using a linear trend. Because of the limited data availability in UNIDO, we\nonly report employment and export (2001-2019) growth results.\n\n\n38\n\nMention: \"UNIDO INDSTAT2 data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Absence of data after 2000 for one country forces reliance on a national statistical report.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000639:39:3", "occurrence_id": "d885e4149ba56edfbfada6d1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: *2** Cameroon only has four datapoints, in 2000-2002 and 2008.\n\n**3** We use census data from the LMSM survey for Ethiopia because the UNIDO data from 2014-2015 are estimates,\nand the census data extend past 2015 allowing us to include 2016-17. The LMSM series match well to the UNIDO\nseries pre-2010 and the LMSM is the source of the UNIDO data.\n\n**4** Lao’s export data is only available from 2010-2019.\n**5** Lesotho only has employment data from 2001-2009, but not output or value-added data--we therefore only\nreport employment and export (2000-2017) growth results.\n**6** Nigeria does not have UNIDO INDSTAT2 data post-2000, so we estimated formal employment from a report\nreleased by Nigeria’s National Bureau of Statistics that has employment statistics for 2010-2012 (NBS, 2014); we\nextrapolated values up through 2017 using a linear trend. Because of the limited data availability in UNIDO, we\nonly report employment and export (2001-2019) growth results.\n\n\n38\n\nMention: \"employment statistics for 2010-2012\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Formal employment is estimated from this report and extrapolated to later years by linear trend.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000627:3:0", "occurrence_id": "24f7db3a5e98a40313f330f4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **1 Introduction**\n\n\nThis study estimates the monetary benefits and costs of public schooling in Ghana. We conduct\n\n\nthis analysis for the quantity of schooling, or years of schooling completed, and for the gain in\n\n\nthe quality of schooling (in terms of measured achievement) produced by specific education\n\n\ninterventions. Our motivation is twofold: to understand the extent to which public spending on\n\n\nschooling provides good value for money; and to provide benefit-cost ratios that can be\n\n\ncompared to similar estimates for other public spending in the style of the Copenhagen\n\n\nConsensus. Regarding the latter, our work is complementary to the recent Ghana Priorities\n\n\nproject conducted by the National Development Planning Commission in partnership with the\n\nCopenhagen Consensus Center. <sup>1</sup> [^1: See https://www.copenhagenconsensus.com/ghana-priorities (accessed January 24, 2022).]\n\n\nOur estimates of monetary costs of schooling are reasonably comprehensive, covering\n\n\nboth public expenditure data drawn from government administrative sources and data on the\n\n\ncomplementary private expenditures needed to access public schooling drawn from the 2016/17\n\n\nGhana Living Standards Survey, the latest available round of the national household sample\n\n\nsurvey administered by Ghana’s statistical agency (GSS 2019). Our estimates of the benefits of\n\n\npublic schooling are more limited, however, something that should be kept in mind when\n\n\ninterpreting our results. In particular, we only measure the benefits to students of estimated\n\n\nfuture labor earnings increases generated by additional years of schooling. This ignores other\n\n\nbenefits of public schooling. For example, Adu Boahen and Yamauchi (2018) find that\n\n\nelimination of fees for primary schooling in 1995 led to reduced teenage fertility and higher age\n\n\nat first marriage in Ghana. Duflo, Dupas, and Kremer (2021) find that tuition-free secondary\n\n\nschooling in Ghana reduced fertility as well. Other authors have posited that public schooling\n\nhelps to form better citizens, though this is more difficult to measure. <sup>2</sup>\n\n\nThere are many studies on the impact of public schooling in Ghana\n\nMention: \"Ghana Living Standards Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Private expenditures households incur to access public schooling are taken from this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000640:3:1", "occurrence_id": "ac641a84358182adea0857fe", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  environment for doing business.\nFinally, difficulties in accessing critical resources often make it even more challenging for SMEs\n[to operate in FCS countries (Peschka 2010). Chief among key inputs is external finance,](http://web.worldbank.org/archive/website01306/web/pdf/wdr_background_paper_peschka_0.pdf)\nespecially bank finance, which is needed to fund SMEs’ working capital and investment needs.\n\n\nThis paper seeks to identify the main constraints to SME financial inclusion in FCS countries,\nhighlighting the key priority areas to reduce the financial inclusion gap of SMEs. Though access\nto finance is widely understood to represent a major obstacle to the development of SMEs in\nFCS countries, to our knowledge the topic remains relatively unexplored in the literature. Yet\nunderstanding the key drivers of SME financial inclusion in FCS countries can help policy makers\ndesign and implement more targeted interventions. <sup>1</sup> [^1: Increasing support for private sector development is, among others, a critical pillar of the World Bank Group’s\n[strategy for FCS countries (WBG 2020).](http://documents.worldbank.org/curated/en/844591582815510521/pdf/World-Bank-Group-Strategy-for-Fragility-Conflict-and-Violence-2020-2025.pdf)] To that end, this paper looks at the impact\nof macroeconomic and financial sector variables, the legal and institutional constraints, and the\nrole of the business environment. In particular, using the responses from formal SMEs in the\nWorld Bank Enterprise Surveys (WBES), this paper attempts an answer to the following question:\nto what extent do economic fundamentals and institutional characteristics affect SME financial\ninclusion in FCS countries _relative_ to non-FCS countries?\n\n\nEconomic fundamentals matter for SME financial inclusion. Higher incomes and better physical\ninfrastructure increase savings and the pool of funds in the economy and improve access to\n[finance (Dabla-Norris et al 2015), while macroeconomic and financial stability can positively](https://www.imf.org/external/pubs/ft/wp/2015/wp15206.pdf\n\nMention: \"World Bank Enterprise Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Responses from formal small and medium firms are used to compare financial inclusion in fragile and other countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000642:5:0", "occurrence_id": "42338393fb0605292557347a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: (a) Trade (% of GDP) (b) GVC participation rate\n\n\n**Source:** (a) World Development Indicators. (b) Borin et al. (2021) through WITS.\n\n\nFigure 1: Globalization proxies\n\n\ncountries after 2008, with China being below the world average during the whole period.\n\n\nBut the literature has shown that the GVC participation rate depends on various factors\n\n\n(geography, market size, endowments, and institutions) beyond trade integration efforts\n\n\n(Fernandes et al., 2022). <sup>2</sup>\n\n\nWhile globalization is a complex phenomenon, a clear definition is necessary to un\n\nderstand it. The term \"globalization\" began to be used more commonly in the 1980s,\n\n\nreflecting technological advances that made it easier to complete international transac\n\ntions. Now globalization is better understood by referring to itas an extension beyond\n\n\nnational borders of the same market forces operating at all economic activity levels (IMF,\n\n\n2008). This definition is well-defined and anchors the discussions around globalization.\n\n\nMore importantly, it has the potential to offer deep and intuitive insights thanks to its\n\n\nempirical applications. This paper contributes to the literature by: (i) Sticking to this\n\n\ndefinition to estimate long-run globalization dynamics (1965-2021) to understand when\n\n\nglobalization \"took off\" and how the world economy stands compared to that moment.\n\n\n(ii) Disentangling the heterogeneous dynamics across countries and the role played by\n\n\ntrade agreements. (iii) Helping to fill the gap between economic policy, which continues\n\n\nto use measures like the trade-to-GDP ratio, and the recent contributions to the litera\n\nture showing that this type of analysis would benefit from adhering to theory and use\n\n\ndomestic (in addition to international) trade flows (Yotov, 2022b).\n\n\nDuring the early 2000s, there was a common perception that the world was becoming\n\n\n2It is important to mention that\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Trade openness as a share of GDP is taken from this indicator series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000644:9:0", "occurrence_id": "ff9014f13bafdf6d03c97cfa", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: information for each social protection program listed in the GDSPRC that was announced before\n\nJune 1, 2021 from official documents (including IMF Article IV revisions and related documents\n\nfrom other international organizations), government websites, and news sources. We quantified\n\nthe expenditures of social protection programs in 154 countries. This set of countries forms the\n\ncore sample for the empirical analysis in this paper (see <u>Appendix Table A.1</u> for the list of\n\ncountries).\n\n\nWe also used several auxiliary data sources. Information on countries’ pre-pandemic social\n\nprotection landscape comes from the ASPIRE database (World Bank 2021a) and OECD’s Social\n\nSpending database (OECD 2021b). Data on employment, inactivity, and growth outcomes comes\n\nfrom <u>ILO (2021), the</u> <u>IMF (2021), and the World Bank</u> <u>(World Bank 2021b, 2022b). Poverty</u>\n\nprojections are from Lakner et al. (2020). Information on the size of the informal sector is from\n\nthe Informal Economy Database produced by the World Bank (Elgin et al. 2021).\n\n\nAn essential feature of our analysis is the distinction between social programs focused on income\n\nprotection and programs focused on job protection. <u>Table 1</u> shows a correspondence between\n\nsocial protection areas, categories of social protection, as defined in GDSPRC, and the policy focus\n\n(on either income protection or job protection). All categories included in the social assistance\n\npolicy areas focus on income protection; the categories of social insurance and labor market policy\n\nareas are split between those focusing on income protection and those focusing on job protection.\n\n\n**4.** **The nature of the social protection response and the impact of the pandemic**\n\n\nCountries across the world implemented substantial economic measures in response to the\n\neconomic shock of the pandemic. The size of the\n\nMention: \"ASPIRE database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The pre-pandemic social protection landscape is characterised using this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000644:9:0", "occurrence_id": "fb15794120e72fbf6be0ed00", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: information for each social protection program listed in the GDSPRC that was announced before\n\nJune 1, 2021 from official documents (including IMF Article IV revisions and related documents\n\nfrom other international organizations), government websites, and news sources. We quantified\n\nthe expenditures of social protection programs in 154 countries. This set of countries forms the\n\ncore sample for the empirical analysis in this paper (see <u>Appendix Table A.1</u> for the list of\n\ncountries).\n\n\nWe also used several auxiliary data sources. Information on countries’ pre-pandemic social\n\nprotection landscape comes from the ASPIRE database (World Bank 2021a) and OECD’s Social\n\nSpending database (OECD 2021b). Data on employment, inactivity, and growth outcomes comes\n\nfrom <u>ILO (2021), the</u> <u>IMF (2021), and the World Bank</u> <u>(World Bank 2021b, 2022b). Poverty</u>\n\nprojections are from Lakner et al. (2020). Information on the size of the informal sector is from\n\nthe Informal Economy Database produced by the World Bank (Elgin et al. 2021).\n\n\nAn essential feature of our analysis is the distinction between social programs focused on income\n\nprotection and programs focused on job protection. <u>Table 1</u> shows a correspondence between\n\nsocial protection areas, categories of social protection, as defined in GDSPRC, and the policy focus\n\n(on either income protection or job protection). All categories included in the social assistance\n\npolicy areas focus on income protection; the categories of social insurance and labor market policy\n\nareas are split between those focusing on income protection and those focusing on job protection.\n\n\n**4.** **The nature of the social protection response and the impact of the pandemic**\n\n\nCountries across the world implemented substantial economic measures in response to the\n\neconomic shock of the pandemic. The size of the\n\nMention: \"Social\n\nSpending database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Public social spending levels are taken from this OECD compilation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000644:9:0", "occurrence_id": "b4425a3819672ccfa302182c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: information for each social protection program listed in the GDSPRC that was announced before\n\nJune 1, 2021 from official documents (including IMF Article IV revisions and related documents\n\nfrom other international organizations), government websites, and news sources. We quantified\n\nthe expenditures of social protection programs in 154 countries. This set of countries forms the\n\ncore sample for the empirical analysis in this paper (see <u>Appendix Table A.1</u> for the list of\n\ncountries).\n\n\nWe also used several auxiliary data sources. Information on countries’ pre-pandemic social\n\nprotection landscape comes from the ASPIRE database (World Bank 2021a) and OECD’s Social\n\nSpending database (OECD 2021b). Data on employment, inactivity, and growth outcomes comes\n\nfrom <u>ILO (2021), the</u> <u>IMF (2021), and the World Bank</u> <u>(World Bank 2021b, 2022b). Poverty</u>\n\nprojections are from Lakner et al. (2020). Information on the size of the informal sector is from\n\nthe Informal Economy Database produced by the World Bank (Elgin et al. 2021).\n\n\nAn essential feature of our analysis is the distinction between social programs focused on income\n\nprotection and programs focused on job protection. <u>Table 1</u> shows a correspondence between\n\nsocial protection areas, categories of social protection, as defined in GDSPRC, and the policy focus\n\n(on either income protection or job protection). All categories included in the social assistance\n\npolicy areas focus on income protection; the categories of social insurance and labor market policy\n\nareas are split between those focusing on income protection and those focusing on job protection.\n\n\n**4.** **The nature of the social protection response and the impact of the pandemic**\n\n\nCountries across the world implemented substantial economic measures in response to the\n\neconomic shock of the pandemic. The size of the\n\nMention: \"Informal Economy Database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The size of the informal sector is measured using this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000646:30:0", "occurrence_id": "7d8008ec1fa4f2ceb4c2ec71", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 13:** **Better or worse than expected?** **(Monthly frequency)**\n\n\nThis figure shows the expected outcomes (blue dashed lines in the box plots) and the observed deforestation\nlevels (colored bars), which are red if deforestation is worse and green if better than expected. Note that\nthe results are not directly comparable with those from Figure 6, which uses annual PRODES data. While\nPRODES assess land cover changes by using satellite images of previous years, DETER is a daily alert\n\n\n\n2,000\n\n\n1,500\n\n\n1,000\n\n\n500\n\n\n\n\n|em that is updated every five days. Deforested area in km²|Col2|Col3|Col4|Col5|Col6|Col7|Col8|Col9|Col10|Col11|Col12|Col13|Col14|Col15|Col16|Col17|Col18|Col19|Col20|\n|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|\n|Expected deforestation<br>95% confidence interval<br>~~50% confidence interval~~|Expected deforestation<br>95% confidence interval<br>~~50% confidence interval~~|Expected deforestation<br>95% confidence interval<br>~~50% confidence interval~~|Expected deforestation<br>95% confidence interval<br>~~50% confidence interval~~|Expected deforestation<br>95% confidence interval<br>~~50% confidence interval~~|Expected\n\nMention: \"PRODES data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its annual satellite-based deforestation measurement differs in frequency from the daily alert system used here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000648:5:0", "occurrence_id": "006d15984ff9b3d0f45e07a5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 2019), an essential predictor of stunting later in childhood (Christian et al., 2013). However, to the best of our\nknowledge no evidence yet exists that links the impacts of deforestation to stunting rates.\n\n\nIn this paper, we contribute to this emerging literature by taking a closer look at the water-borne channels and\nthe causal effects of forest loss on human health. Such an endeavor is challenging due to the multiple factors\ndetermining health outcomes and the degree of complementarity that must be accounted for by including\nsocioeconomic, demographic, infrastructure, environmental, behavioral, and institutional factors (Pattanayak &\nPfaff, 2009). We exploit the wealth of data available in the Demographic and Household Surveys (DHS), which\nincludes not only contemporaneous health and socioeconomic outcomes, but also full birth histories for all live\nbirths that occur in the years before the survey. We restrict our sample to 0.7 million children from rural areas <sup>1</sup> [^1: We restrict our sample to non-migrant children, i.e., children whose households have either always lived in their current\nlocation, or that have migrated to their current location before the child was conceived.]\nin 46 countries from 2000-2020 for which we have spatial coordinates. This allows us to link the health outcomes\nwith high-resolution deforestation data and study the effects of forest loss on the incidence of DD (by linking\nlags of forest loss to the survey year) as well as stunting (by linking lags of forest loss to the birth year).\n\n\nIdentifying health impacts of forest loss is challenging given a trade-off between health costs and economic\nbenefits that may be associated with agricultural expansion or urbanization following forest loss. To address the\npotential bias from non-random forest loss and confounding trends, we examine a possible water pathway from\ndeforestation to health outcomes by exploiting quasi-random variation in deforestation upstream to assess the\nimpacts on waterborne disease outcomes for rural households downstream. This allows us to isolate a waterspecific channel through which forest loss impacts health. We also\n\nMention: \"Demographic and Household Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Health, socioeconomic outcomes and full birth histories are used, restricted to rural children.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000652:12:1", "occurrence_id": "781f169a94c684ce15cafa65", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .\n\n\n### **4 Data and Sample**\n\nThe sample includes 780 farmers from 103 villages located in the districts of Manafa,\n\nMbale, and Tororo in Eastern Uganda. This sample had been selected for a field experi\nment carried out in previous years by a research team that included the author (Burchardi\n\net al., 2022). The three contiguous districts were selected for their high suitability to maize\ncultivation according to the FAO GAEZ data. <sup>10</sup> The 103 villages <sup>11</sup> were selected by a\n\nrandomization algorithm from the pool of villages in the 2010 national census with a\n\npopulation density lower than 100 households per square kilometer (to ensure they were\n\nin rural areas), stratifying by population density. In each village, we enlisted all resid\ning households through a researcher-run census survey. We selected six to nine research\n\nparticipants (maximum one per enlisted household, always the head of the household) ac\ncording to the following criteria: _i)_ engaging in commercial farming; _ii)_ having cultivated\n\nor planning to cultivate maize; _iii)_ reporting land holdings from two to six acres; _iv)_ hav\ning a mobile money account. The participants in the Burchardi et al. (2022) experiment\n\nwere enrolled at the end of 2016 and had participated in up to four survey interviews and\n\none willingness-to-pay elicitation by the time the field work for the experiment in this\n\npaper started; all research activities connected to Burchardi et al. (2022) were completed\n\nby 2017.\n\n\nThe data I use come from four surveys: an in-person baseline survey, an in-person\n\nwillingness-to-pay elicitation survey, a phone follow-up survey, an in-person follow-up\n\nsurvey. All surveys have been written by the research team and approved by the U\n\nMention: \"FAO GAEZ data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Agro-ecological suitability for maize justified the choice of the three study districts.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000652:12:1", "occurrence_id": "f584cda998e49e4b5651b81d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .\n\n\n### **4 Data and Sample**\n\nThe sample includes 780 farmers from 103 villages located in the districts of Manafa,\n\nMbale, and Tororo in Eastern Uganda. This sample had been selected for a field experi\nment carried out in previous years by a research team that included the author (Burchardi\n\net al., 2022). The three contiguous districts were selected for their high suitability to maize\ncultivation according to the FAO GAEZ data. <sup>10</sup> The 103 villages <sup>11</sup> were selected by a\n\nrandomization algorithm from the pool of villages in the 2010 national census with a\n\npopulation density lower than 100 households per square kilometer (to ensure they were\n\nin rural areas), stratifying by population density. In each village, we enlisted all resid\ning households through a researcher-run census survey. We selected six to nine research\n\nparticipants (maximum one per enlisted household, always the head of the household) ac\ncording to the following criteria: _i)_ engaging in commercial farming; _ii)_ having cultivated\n\nor planning to cultivate maize; _iii)_ reporting land holdings from two to six acres; _iv)_ hav\ning a mobile money account. The participants in the Burchardi et al. (2022) experiment\n\nwere enrolled at the end of 2016 and had participated in up to four survey interviews and\n\none willingness-to-pay elicitation by the time the field work for the experiment in this\n\npaper started; all research activities connected to Burchardi et al. (2022) were completed\n\nby 2017.\n\n\nThe data I use come from four surveys: an in-person baseline survey, an in-person\n\nwillingness-to-pay elicitation survey, a phone follow-up survey, an in-person follow-up\n\nsurvey. All surveys have been written by the research team and approved by the U\n\nMention: \"2010 national census\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Villages were drawn from this enumeration and stratified by population density.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000655:22:1", "occurrence_id": "eed5d741019d5b6e42376b36", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  of a more qualitative\n\n\nnature, such as the benefits gained from operating informally, linkages with the formal sector,\n\n\nperceived costs or barriers to formalization, and the use of government services. Existing\n\n\nenterprise surveys can be used to inform the exact framing of these questions. For example, the\n\n\nWorld Bank Enterprise Survey asks firms to identify their challenges to doing business based on a\n\n\n21\n\nMention: \"World Bank Enterprise Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Its business environment constraints module is offered as a model for framing survey questions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000657:19:0", "occurrence_id": "31e216e927085127852fe181", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 16: Household Food and Non-food Consumption, 2014/15\n\n\n\n\n\n\n\n\n\n_Source:_ Authors’ calculations using NBHS 2009 and NHBPS 2014/15.\n\n\nBoth food and non-food expenditures remain lower in rural Sudan compared to urban Sudan (Figure\n17). These differences are significantly higher for non-food expenditure, whereas only small differences\nexist for food expenditure. However, as there are differences in prices between the rural and urban\nareas, food consumption is likely higher in rural areas, something which is to be expected, due to the\nincreased physical demands of agrarian work. Differences in non-food consumption, where the urban\npopulation on average consumes almost twice the monetary amount as the rural sector, are unlikely to\nbe due to different prices. As we note in the next subsection, such non-food expenditure may account\nfor differences in asset ownership between urban and rural Sudan. Some assets, such as those that aid\nin communication and transport, can improve the productive capacity of farmers. Therefore, improving\nmarket infrastructure, to increase the opportunity for both food and non-food consumption in rural\nSudan, is vital for improving rural livelihoods and alleviating poverty.\n\n\nFigure 17: Per Capita Food and Non-food Consumption, 2014/15\n\n\n(a) Per Capita Non-food Consumption\n\n\n\n\n\n\n\n18\n\nMention: \"NBHS 2009\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Household food and non-food consumption is compared with the later round from this baseline.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000657:19:0", "occurrence_id": "a3ca9e7d020baf29ca54161f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 16: Household Food and Non-food Consumption, 2014/15\n\n\n\n\n\n\n\n\n\n_Source:_ Authors’ calculations using NBHS 2009 and NHBPS 2014/15.\n\n\nBoth food and non-food expenditures remain lower in rural Sudan compared to urban Sudan (Figure\n17). These differences are significantly higher for non-food expenditure, whereas only small differences\nexist for food expenditure. However, as there are differences in prices between the rural and urban\nareas, food consumption is likely higher in rural areas, something which is to be expected, due to the\nincreased physical demands of agrarian work. Differences in non-food consumption, where the urban\npopulation on average consumes almost twice the monetary amount as the rural sector, are unlikely to\nbe due to different prices. As we note in the next subsection, such non-food expenditure may account\nfor differences in asset ownership between urban and rural Sudan. Some assets, such as those that aid\nin communication and transport, can improve the productive capacity of farmers. Therefore, improving\nmarket infrastructure, to increase the opportunity for both food and non-food consumption in rural\nSudan, is vital for improving rural livelihoods and alleviating poverty.\n\n\nFigure 17: Per Capita Food and Non-food Consumption, 2014/15\n\n\n(a) Per Capita Non-food Consumption\n\n\n\n\n\n\n\n18\n\nMention: \"NHBPS 2014/15\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The 2014/15 consumption figures plotted are calculated from this household panel baseline.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000660:53:0", "occurrence_id": "9b7c70ca4341f1c583433e3d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Bank.\n\nNote: Income classification as per 2021 World Bank classification. The yearly median is computed over a balanced\nsample of 24 African countries and 27 non-African developing countries, for which data were available for all years.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n**Figure 4** Access to and use of financial services in international comparison, 2018-2021. Source: Global Financial\nDevelopment Database (September 2022 Version), World Bank\n\nNote: Data are as per end of year 2020 except for adults with an account at a formal financial institution. Median for\nlow- and lower-middle-income countries in SSA and non-SSA countries.\n\n[Bank accounts per 1,000 adults (former issue of the handbook) adjusted for exposition.]\n\n\n52\n\nMention: \"Global Financial\nDevelopment Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Access to and use of financial services across countries are taken from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000649:40:2", "occurrence_id": "40c1168efc0a6f22f557a6cf", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  small flows (never above 1000 USD). Columns\n\n3–4 present the same estimates excluding large flows (top 1% of ODPs). Outcomes are\n\ndefined at the origin-destination-product-year level. FE stands for fixed effects. Tariffs are\n\ncontrolled for. Source: WITS and ALADI databases.\n\n\n39\n\nMention: \"WITS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Origin-destination-product trade outcomes and tariffs are compiled from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000649:40:2", "occurrence_id": "8e14e2b8509089ba3be538c8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  small flows (never above 1000 USD). Columns\n\n3–4 present the same estimates excluding large flows (top 1% of ODPs). Outcomes are\n\ndefined at the origin-destination-product-year level. FE stands for fixed effects. Tariffs are\n\ncontrolled for. Source: WITS and ALADI databases.\n\n\n39\n\nMention: \"ALADI databases\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The regional integration body's records complement the trade database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000661:7:0", "occurrence_id": "5ae89970d3b354c8565ab5aa", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Looking at primary energy supply, combustible fuels (including coal, oil and gas) remain the most\npredominant energy source for MENA, accounting for over 87% of the region’s electricity in 2018.\nMeanwhile, the share of renewable energy remains relatively small and fell by 2.5 percentage points over\nthe observed timeframe, owing to a marginal rise in the use of other energy sources. On the other hand,\nECA presents a different profile. While combustible fuels are widespread in the region, they also utilize\nnuclear and other renewable energy sources for their supply. In 2018, the share of renewable energy\nincreased to 16.4% due to a decline in nuclear energy and combustible fuels. While this is positive, further\nefforts are required to transform the energy supply and cut emissions significantly. Given that renewables\nrepresent only a small portion of the total energy supply in both ECA and MENA, there is a large scope for\nintervention to switch to cleaner energy sources, especially in MENA. This will become increasingly\nimportant as there is growing pressure from governments and their citizens to respond to the climate\ncrisis.\n\n\nThe ECA region has experienced a steep and steady decline in energy intensity since 1990, having\ndecreased by almost 27% and narrowing the gap between energy intensity values compared to MENA.\nMeanwhile, similar to the trajectory of carbon emissions in absolute, per capita, and per US dollar of GDP\nterms, energy intensity in MENA has remained stable oscillating at around 100 and experiencing a 9.4%\ndecrease between 2005 and 2018. The discrepancy in the magnitude of the change between the two\nregions indicates that there is still much room for improvement for the MENA region, which could benefit\nsignificantly from further investments in energy efficiency.\n\n\n**_Figure 3. Breakdown of primary energy supply by fuel type (%)_**\n\n\nNote: Data represent unweighted averages across countries covered in Enterprise Surveys. Combustible fuels include\ncoal and peat, crude, natural gas liquids (NGL) and feedstocks\n\nMention: \"Enterprise Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The energy breakdown is an unweighted average across the countries these surveys cover.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000667:39:0", "occurrence_id": "4f46335342dc65e69416a721", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ##### **Tables**\n\n\n\n**<u>Table 1: Overview of Informal Sector Enterprise Survey data</u>**\n\n\n\n<u>Universe</u> <u>Starting</u> <u>Total blocks</u> <u>Informal business</u> <u>Interviews</u>\n\n<u>Country</u> <u>Fieldwork</u> <u>City</u> <u>of blocks</u> <u>blocks</u> <u>enumerated</u> <u>units enumerated</u> <u>completed</u>\n\nIndia Sept 2021 – Apr 2022 <u>Hyderabad</u> 29,308 1419 2,045 6,414 2,191\nJaipur 17,569 1276 1,620 4,312 1,526\nKochi 3,622 416 424 1,213 779\nLudhiana 9,296 575 658 2,091 712\nMumbai 16,177 1700 3,400 8,400 2,358\nSehore 2,100 305 314 1032 536\nSurat 11,052 1314 1,854 6,334 1,427\nTezpur 1,058 250 336 611 357\nVaranasi 4,113 311 352 1376 786\n<u>TOTAL</u> <u>94,295</u> <u>7,566</u> <u>11,003</u> <u>31,783</u> <u>10,672</u>\n<u>Iraq</u> <u>Sept 2021 – Apr 2022</u> <u>Baghdad</u> 30,524 1272 1,352 2,282 680\nBasrah 12,611 520 554 1,609 599\nNajaf 3\n\nMention: \"Informal Sector Enterprise Survey data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Universes, blocks enumerated and interviews completed are documented for this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000668:7:1", "occurrence_id": "e3ab1353df4f26783307eddc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: , The SDS drops this option from the\ninstrument.\n**Module II.** Module II adapts the skills component of the O*NET Worker Requirement module. For\nO*NET, skills are the procedures or ways of working given acquired knowledge. Skills can be acquired\non the job – in the same occupation or not – or through formal or informal education. The SDS skills\nmodule asks detailed questions about 9 types of skills: reading (5 items), writing (3 items), math (4\nitems), social (6 items), physical (3 items), problem-solving (6 items), technical (6 items),\nmanagement (4 items), and digital (13 items). The questionnaire also pilots questions related to\nenvironmental skills (4 items) and the ability to work from home (1 item). The module requests that\nrespondents rate each of the skills according to the frequency with which they are performed ranging\nfrom 1 for “Never” to 7 for “Hourly or more often.”\nThe SDS skills module was redesigned after carefully reviewing O*NET and other skills measurement\nsurveys, particularly PIAAC and STEP, and taking into consideration experience from the Vietnam\nand Indonesia pilots. The module also reflects issues raised in the literature about the vagueness and\ncomplexity of O*NET wording and responses (Handel 2016). **Appendix 1** provides a comparison of\nthe O* - <u>NET and SDS skills instrument. Several important changes were made.</u>\n\n<u>Additional skills. The SDS instrument includes 56 skills while O*NET includes 35. Changes</u>\n\n\n\n\n- <u>NET and SDS skills instrument. Several important changes were made.</u>\n<u>Additional skills. The SDS instrument includes 56 skills while O*NET includes 35. Changes</u>\n~~were made afte~~ r reviewing the skills concept and the PIAAC\n\nMention: \"PIAAC\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"This skills assessment was reviewed alongside others when redesigning the skills module.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000668:7:1", "occurrence_id": "4e25c07e65078c5aee5642b5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: , The SDS drops this option from the\ninstrument.\n**Module II.** Module II adapts the skills component of the O*NET Worker Requirement module. For\nO*NET, skills are the procedures or ways of working given acquired knowledge. Skills can be acquired\non the job – in the same occupation or not – or through formal or informal education. The SDS skills\nmodule asks detailed questions about 9 types of skills: reading (5 items), writing (3 items), math (4\nitems), social (6 items), physical (3 items), problem-solving (6 items), technical (6 items),\nmanagement (4 items), and digital (13 items). The questionnaire also pilots questions related to\nenvironmental skills (4 items) and the ability to work from home (1 item). The module requests that\nrespondents rate each of the skills according to the frequency with which they are performed ranging\nfrom 1 for “Never” to 7 for “Hourly or more often.”\nThe SDS skills module was redesigned after carefully reviewing O*NET and other skills measurement\nsurveys, particularly PIAAC and STEP, and taking into consideration experience from the Vietnam\nand Indonesia pilots. The module also reflects issues raised in the literature about the vagueness and\ncomplexity of O*NET wording and responses (Handel 2016). **Appendix 1** provides a comparison of\nthe O* - <u>NET and SDS skills instrument. Several important changes were made.</u>\n\n<u>Additional skills. The SDS instrument includes 56 skills while O*NET includes 35. Changes</u>\n\n\n\n\n- <u>NET and SDS skills instrument. Several important changes were made.</u>\n<u>Additional skills. The SDS instrument includes 56 skills while O*NET includes 35. Changes</u>\n~~were made afte~~ r reviewing the skills concept and the PIAAC\n\nMention: \"STEP\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Its skills measurement approach was among those reviewed for the new module.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000669:5:0", "occurrence_id": "98181c01f64456713b60556a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: where _αik,t_, _βijk,t_, and _γjk,t_ are parameters capturing sales/supply-potential factors of _k_ in _i_, the\ntrade freedom of _k_ between _i_ and _j_, and the demand/consumption potential of _k_ in _j_, all in logs\nand at time/month _t_ . When only trade but no domestic sales data are available, the unilateral\nfrictions on _k_ enter _γjk,t_ only.\n\nWe proceed in two steps to identify the parameters _ηk_ and _ξk_ of interest: (i) estimate _γjk,t_ as\nimporter-product-month fixed effects from monthly bilateral product-level export data _xijk,t_ on 180\nsensitive products; (ii) estimate the impact of unilateral trade frictions on product _k_ from\n\n\n\n_γ_ - _jk,t_ = _gjk,t_ + ˜ _γjk_ + _λ_ <sup>_γ_</sup> _t_\n\n\n\n\n<sup>_γ_</sup> _t_ <sup>+</sup> <sup>_uγ_</sup>\n\n\n\n_ik,t_ <sup>_γ_</sup> <sup>_,_</sup> (4)\n\n\n\n\n<sup>_γ_</sup> _t_ <sup>and</sup> <sup>_uγ_</sup>\n\n\n\nwhere _gjk,t_ is defined as in (2), and _γ_ ˜ _jk_, _λ_ <sup>_γ_</sup> _t_\n\n\n\nwhere _gjk,\n\nMention: \"monthly bilateral product-level export data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Importer-product-month fixed effects are estimated on this data for sensitive products to recover trade frictions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000670:18:1", "occurrence_id": "d9b3ca8aadfce7a8c87f97df", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Philippines exhibits a similar pattern. Income reductions</mark>\nare around -7% for quintiles 2 to 4, while it is -4.4% for quintile 1.\n\n\n<mark>In Türkiye, income changes were very small for all quintiles of the per capita household income</mark>\n<mark>distribution. The smallest changes were between -0.4% and -0.5% in quintiles 2 to 4, while the</mark>\nlargest change was a reduction of 1.1% in quintile 1.\n\n\n<mark>Overall, the shock caused income declines across the distribution, but with no pattern across</mark>\n<mark>countries. This suggests that differences in the structure of the labor market captured by the</mark>\n<mark>household survey as well as the sectoral growth patterns taken from National Accounts lead to</mark>\ndiverse patterns of projected distributional impacts. This would require further study.\n\n\n<mark>We also study how income changes correlate with the educational level and occupation of the</mark>\n<mark>household head. Panel A of Table 1 shows that in four out of five countries —Türkiye, Philippines,</mark>\n<mark>South Africa, and Brazil—income losses were larger the lower the educational level of the</mark>\n<mark>household head. The magnitude of the losses, however, differ between countries. For instance,</mark>\n<mark>household where the head has up to complete primary education faced an average income loss of</mark>\n<mark>8.3% in South Africa and of 11.6% in Brazil. For households where the head has tertiary education,</mark>\n<mark>the loss was of 6.3% in South Africa and 8.4% in Brazil. Sri Lanka is the only exception to the</mark>\n<\n\nMention: \"National Accounts\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Sectoral growth patterns taken from these accounts drive the projected distributional impacts.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000671:10:1", "occurrence_id": "f5f1a4bf492efe79d911ec35", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 2 Formal versus informal firms and performance variables: Descriptive statistics_**\n\n\n\nAn initial question is whether most informal firms are associated with lower performance outcomes than\nformal firms. The main performance variables are labor productivity, sales, employment, and exporting.\nLabor productivity is measured as value added (total sales minus raw materials & intermediate inputs plus\nwater & electricity used in production) divided by the total number of full-time working people including\nowners. The value of sales is asked in three complementary ways in the survey: total sales, turnover, and\n“revenues money received by the business.” Employment is the number of full-time employees plus the\nnumber of owners. Exporting is a non-continuous indicator variable taking the value of 0 or 1 depending\non whether the firm reports having customers located in other countries or selling goods or services\nabroad; it is typically reported as a share of the total number of firms in a particular sub-sample.\n\n\nThe graphs in Figure 1 present conditional kernel densities (residuals of OLS regressions on urban and\nindustry dummies) for formal and informal firms of the three main performance variables: log labor\nproductivity, log sales, and log employment. The unconditional kernel densities of productivity and sales\nvariables look similar. The densities of all three performance variables for formal firms are to the right of\nthose for informal firms, although there are large areas of overlap. The second humps in the log\nproductivity and log sales densities reflect the existence of a higher concentration of firms at relatively\n\n\nbelow FCFA 50 million are eligible to pay a unique global contribution or CGU, and thereby are exempt from separate VAT or\nsales taxes in Senegal. If formal firms that are required to pay local and municipal taxes do not pay them, this makes them nontax-compliant rather than informal. And while the keeping of formal accounts is often included in national definitions of\nformality, the relevant RIA survey question does not allow a direct correspondence with Senegal’s\n\nMention: \"RIA survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its question on formal accounts cannot be mapped directly onto the national definition of informality.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000675:33:0", "occurrence_id": "876d2435c2a1397d570ecf7b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 4.2: Regional Contributions to the Global Prosperity Gap**\n\n\n**Notes:** This figure reports the decomposition of the global estimate of 𝑊(𝑦, 𝑧) into contributions by regions using\nthe prosperity threshold of $25 per person per day in 2017 $PPP. See also Table 4.1.\n\n\nAlthough they measure different concepts, empirically changes over time in the prosperity\ngap and growth in the bottom 40 percent are similar at the country level. This can be seen in Figure\n4.3, which compares average annual growth in the bottom 40 percent (on the vertical axis) against\ngrowth in the prosperity gap (on the horizontal axis and multiplied by -1 so that positive growth\ncorresponds to an improvement in the prosperity gap). The left panel of Figure 4.3 considers the\nfull period 1990-2019 using the lined-up data discussed in Section 3.1, while the right panel\nfocuses on a smaller set of 53 “spells” of various lengths calculated between actual survey years\naround 2014 and around 2019. <sup>31</sup> Each panel includes the 45-degree line, with points above (below)\nthe line corresponding to episodes where average incomes in the bottom 40 percent improved\nfaster (slower) than the prosperity gap. Overall, the correlation between the two measures is high,\nwith most countries close to the 45-degree line.\n\n\n31 The comparable spells can be found in the Global Dataset for Shared Prosperity here:\n<mark>https://www.worldbank.org/en/topic/poverty/brief/global-database -of-shared-prosperity.</mark> Fifty-three countries have\ngrowth spells between -5% and 5%.\n\n\n31\n\nMention: \"Global Dataset for Shared Prosperity\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Comparable growth spells are available in this dataset for the countries studied.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000678:9:2", "occurrence_id": "99619d89c8c70f2a265eb712", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  a control and proxy\nfor the facility's resources.\n\n\n3.2. Key Variables and Summary Statistics\n\n\nWe first merge the SDI data set and the Master List of facilities which results in a subset of 2,578 facilities.\nThis means that 15% of the 2018 SDI facilities could not be found in Kenya's Master List of facilities. This final\nsample was complemented with information on road, electricity, and cellphone internet access, with\nadditional data described in the previous subsection. <sup>11</sup> [^11: The restricted sample works with the facilities that merges with the reviewed Master List. Moreover, an initial inclusion of “Distance to the nearest\nFacility” made it necessary to work with facilities in the 10,131 list.]\n\n\n7 D.J. Weiss, A. Nelson, H.S. Gibson, W. Temperley, S. Peedell, A. Lieber, M. Hancher, E. Poyart, S. Belchior, N. Fullman, B. Mappin, U. Dalrymple, J.\nRozier, T.C.D. Lucas, R.E. Howes, L.S. Tusting, S.Y. Kang, E. Cameron, D. Bisanzio, K.E. Battle, S. Bhatt, and P.W. Gething. A global map of travel time\nto cities to assess inequalities in accessibility in 2015. (2018). Nature. doi:10.1038/nature25181.\n8 For quality road we followed the definition of the Rural Access Index: “ _Good roads are paved roads that look dark and are not eroded. Fair roads_\n_are either paved roads in fair condition or good gravel roads that are motorable_ ” (Iimi, et al., 2016).\n9 10% of the Kenya Population and Housing Census (KPHC), randomly selected.\n10 The Kenya Integrated Household Budget Survey (KIHBS) conducted between 2015 and 2016.\n11 The restricted sample works with the facilities that merges\n\nMention: \"SDI data set\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Facility locations are merged with the national registry, leaving a sample of 2,578 matched facilities.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000678:9:2", "occurrence_id": "0612a2cb2b24f9aa2168aaf2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  a control and proxy\nfor the facility's resources.\n\n\n3.2. Key Variables and Summary Statistics\n\n\nWe first merge the SDI data set and the Master List of facilities which results in a subset of 2,578 facilities.\nThis means that 15% of the 2018 SDI facilities could not be found in Kenya's Master List of facilities. This final\nsample was complemented with information on road, electricity, and cellphone internet access, with\nadditional data described in the previous subsection. <sup>11</sup> [^11: The restricted sample works with the facilities that merges with the reviewed Master List. Moreover, an initial inclusion of “Distance to the nearest\nFacility” made it necessary to work with facilities in the 10,131 list.]\n\n\n7 D.J. Weiss, A. Nelson, H.S. Gibson, W. Temperley, S. Peedell, A. Lieber, M. Hancher, E. Poyart, S. Belchior, N. Fullman, B. Mappin, U. Dalrymple, J.\nRozier, T.C.D. Lucas, R.E. Howes, L.S. Tusting, S.Y. Kang, E. Cameron, D. Bisanzio, K.E. Battle, S. Bhatt, and P.W. Gething. A global map of travel time\nto cities to assess inequalities in accessibility in 2015. (2018). Nature. doi:10.1038/nature25181.\n8 For quality road we followed the definition of the Rural Access Index: “ _Good roads are paved roads that look dark and are not eroded. Fair roads_\n_are either paved roads in fair condition or good gravel roads that are motorable_ ” (Iimi, et al., 2016).\n9 10% of the Kenya Population and Housing Census (KPHC), randomly selected.\n10 The Kenya Integrated Household Budget Survey (KIHBS) conducted between 2015 and 2016.\n11 The restricted sample works with the facilities that merges\n\nMention: \"Master List of facilities\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Fifteen percent of mapped facilities could not be matched to this official registry.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000679:19:0", "occurrence_id": "d19ee2b8aebc5c1e37fe876a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: (1) Source: 2018/19 BIHS. On feed, similar data were not available for other livestock or poultry rearing.\n(2) For sources of feed purchase, respondents were able to select multiple options.\n\n\nTable 2 reinforces that among owners/managers of animals in the 2018/19 BIHS, a substantial share of\n\nproduction across activities, and disproportionately for dairy cows and poultry, is for the household’s own\n\nconsumption as well as for sale. About 85 percent of men and women owners/managers raise animals for\n\nboth consumption as well as sale for meat; this share is about 93 percent for poultry. <sup>18</sup> [^18: Additional data from the BIHS shows that among dairy producers, about half who sell milk from their production sell to\nfriends/neighbors, and the remaining half at the market.] When considering\n\nanimal products such as milk and eggs, 82 percent of poultry owners/managers are engaged entirely in\n\nown-consumption activities, and about 50 percent of dairy owners/managers. A larger share of men and\n\nwomen who raise sheep/goats or bulls sell these animals in the market, but a substantial share continue to\n\nraise these animals for their own consumption along with sale (about 45 percent for sheep/goats, and 22\n\npercent for bulls).  The BIHS also reveals interesting gender differences in ownership and management of\n\nanimals overall. Men, for example, are much more likely to own dairy cows, while women are significantly\n\nmore likely to be involved in managing the care of these animals, including feed and vaccinations. Poultry,\n\non the other hand, is much more likely to be owned/managed among women (59 percent of women are\n\nowners/managers, compared to 8 percent of men).  Men are also more likely than women to own other\n\nlivestock, such as bulls and sheep/goats, although the magnitude of the gender difference is narrower.\n\n\nAmong those selling their products, the 2018/19 BIHS also shows that\n\nMention: \"2018/19 BIHS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Livestock feed use and purchase behaviour are tabulated from this household livestock survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000666:29:2", "occurrence_id": "1ba026e3d3c350d67fd0bcd3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: , at least 27 indigenous languages spoken. Our sample had at least 10\n<u>of these indigenuous languages. Rural localities with forest coverage had at least 15. Nevertheless, in terms of proportion of people</u>\n<u>speaking indiginous languages, the sampled localities had a higher average proportion of people speaking an indigenous language</u>\n<u>(0.62) than the average locality in these two states (0.45) . Rural localities with forest coverage had 0.59, which is statistically</u>\n<u>equivalent to our sampled localities.</u>\n<u>[3 CONAPO (2010) Indice de marginalización por localidad 2010. Taken from](http://www.conapo.gob.mx/es/CONAPO/Indice_de_Marginacion_por_Localidad_2010)</u>\n<u>[http://www.conapo.gob.mx/es/CONAPO/Indice_de_Marginacion_por_Localidad_2010](http://www.conapo.gob.mx/es/CONAPO/Indice_de_Marginacion_por_Localidad_2010)</u>\n~~[The marginalization index combines a series of socio-economic indicators: percentage](http://www.conapo.gob.mx/es/CONAPO/Indice_de_Marginacion_por_Localidad_2010)~~ of population 15 years or older illiterate or\n~~[without completing primary education, precentage of inhabited houses without a toilet](http://www.conapo.gob.mx/es/CONAPO/Indice_de_Marginacion_por_\n\nMention: \"Indice de marginalización por localidad 2010\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Locality-level marginalisation scores are used to compare the sampled localities with the state average.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000680:6:0", "occurrence_id": "0dd3f273a2ea8ee0f5f7299a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: phones. <sup>3</sup> [^3: https://www.google.com/covid19/mobility/] The data presents changes in visits to these different places relative to some baseline\nvalue for a given geographic location, and it is available at the subnational level in some countries.\nThese data are only obtained from mobile phone users with location services turned on but\nnonetheless can reflect useful signals in the data. We extracted Google mobility and search trends\nat the subnational level. The raw GPS data on mobility were first grouped by categories of places\nwith similar characteristics, namely: grocery and pharmacy, parks, transit stations, retail and\nrecreation, residential areas, and workplaces. <sup>4</sup> [^4: This preprocessing takes place at Google.] The resulting data shows how visits to categorized\nplaces change relative to the baseline period, which is the region-specific median value picked\nfrom the 5-week period between January 3 and February 6, 2020, just before the onset of the global\npandemic. The Google mobility information is available only from February 2020, which naturally\nlimits its usage—when tracking changes in economic activity—to the COVID-19 period.\n\nGoogle search data—that is also available at the subnational level in some countries—provide\ninformation regarding the search intensity of a given combination of keywords for the target\nlocation and time period. Each data point can be interpreted as a share of a given request in the\ntotal searches in a given location and time period. However, as Abay et al. (2020) point out, the\nraw Google search series are not comparable across terms and geographic areas, which in our case\nare subnational regions. The primary reason is the following: when requesting Google search data\nfor a given search term, it is possible to obtain the information only for a limited number of\ngeographic locations per request. Thus, to obtain the information for the desired number of\ngeographic locations, one has to file multiple requests in succession\n\nMention: \"Google mobility and search trends\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Subnational mobility and search series are extracted as activity signals.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000680:6:0", "occurrence_id": "b3915a21c762214b35b261af", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: phones. <sup>3</sup> [^3: https://www.google.com/covid19/mobility/] The data presents changes in visits to these different places relative to some baseline\nvalue for a given geographic location, and it is available at the subnational level in some countries.\nThese data are only obtained from mobile phone users with location services turned on but\nnonetheless can reflect useful signals in the data. We extracted Google mobility and search trends\nat the subnational level. The raw GPS data on mobility were first grouped by categories of places\nwith similar characteristics, namely: grocery and pharmacy, parks, transit stations, retail and\nrecreation, residential areas, and workplaces. <sup>4</sup> [^4: This preprocessing takes place at Google.] The resulting data shows how visits to categorized\nplaces change relative to the baseline period, which is the region-specific median value picked\nfrom the 5-week period between January 3 and February 6, 2020, just before the onset of the global\npandemic. The Google mobility information is available only from February 2020, which naturally\nlimits its usage—when tracking changes in economic activity—to the COVID-19 period.\n\nGoogle search data—that is also available at the subnational level in some countries—provide\ninformation regarding the search intensity of a given combination of keywords for the target\nlocation and time period. Each data point can be interpreted as a share of a given request in the\ntotal searches in a given location and time period. However, as Abay et al. (2020) point out, the\nraw Google search series are not comparable across terms and geographic areas, which in our case\nare subnational regions. The primary reason is the following: when requesting Google search data\nfor a given search term, it is possible to obtain the information only for a limited number of\ngeographic locations per request. Thus, to obtain the information for the desired number of\ngeographic locations, one has to file multiple requests in succession\n\nMention: \"Google mobility information\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Availability only from February 2020 bounds the period over which it can track economic activity.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000680:6:0", "occurrence_id": "5b39e9b767e0be8f4ec3701c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: phones. <sup>3</sup> [^3: https://www.google.com/covid19/mobility/] The data presents changes in visits to these different places relative to some baseline\nvalue for a given geographic location, and it is available at the subnational level in some countries.\nThese data are only obtained from mobile phone users with location services turned on but\nnonetheless can reflect useful signals in the data. We extracted Google mobility and search trends\nat the subnational level. The raw GPS data on mobility were first grouped by categories of places\nwith similar characteristics, namely: grocery and pharmacy, parks, transit stations, retail and\nrecreation, residential areas, and workplaces. <sup>4</sup> [^4: This preprocessing takes place at Google.] The resulting data shows how visits to categorized\nplaces change relative to the baseline period, which is the region-specific median value picked\nfrom the 5-week period between January 3 and February 6, 2020, just before the onset of the global\npandemic. The Google mobility information is available only from February 2020, which naturally\nlimits its usage—when tracking changes in economic activity—to the COVID-19 period.\n\nGoogle search data—that is also available at the subnational level in some countries—provide\ninformation regarding the search intensity of a given combination of keywords for the target\nlocation and time period. Each data point can be interpreted as a share of a given request in the\ntotal searches in a given location and time period. However, as Abay et al. (2020) point out, the\nraw Google search series are not comparable across terms and geographic areas, which in our case\nare subnational regions. The primary reason is the following: when requesting Google search data\nfor a given search term, it is possible to obtain the information only for a limited number of\ngeographic locations per request. Thus, to obtain the information for the desired number of\ngeographic locations, one has to file multiple requests in succession\n\nMention: \"Google search data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Keyword search intensity for a location and period is interpreted as a share of all searches.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000682:15:0", "occurrence_id": "7b855a4fba93a6b5a1e3357c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ###### **5.1 Construction of Distorted Allocations**\n\nThis section discusses the calibration of the sectoral productivity dynamics and the\n\ndistortions. Following the literature (for instance, Duarte and Restuccia 2010), the\n\npaths of relative sectoral labor productivity are measured directly from the data as\n\nreal value-added per worker in each sector\n\n\n\n_<u>Ai</u>_\n_Aj_\n\n\n\n=\n\n\n\n_<u>Real−Value−Addedi</u>_\n\n_<u>Employmenti</u>_\n_<u>Real−Value−Addedj</u>_\n\n_Employmentj_\n\n\n\nWe normalize the levels of productivity at the beginning of each country’s sample\nto be equal to 1, letting the taste parameter Ω _i_ in each country be the adjustment\nvariable to replicate the observed sectoral employment shares at the onset of the\n\nstructural transformation.\n\n\nThe sectoral allocation of employment is also driven by the dynamics of aggre\ngate real consumption. We read this directly from the Penn World Tables and nor\nmalize the initial level of consumption to be equal to 1 at the onset of the transfor\nmation period, as discussed in section 4.\n\n\nThe time series of manufacturing and service wedges are identified to match the\n\nevolution of sectoral employment shares. Given the estimated paths of relative pro\nductivity and aggregate real consumption, we identify the sequences of wedges by\n\nrequiring that the equilibrium employment ratios across sectors match the employ\nment ratios in the data for each country. This is achieved by solving for the pair of\n\nwedges in equations 8 and 9 imposing the observed values of the employment ra\ntios in the left-hand sides of the equations. In addition to this baseline experiment,\n\nwe shall also consider a distorted economy where we introduce a single wedge to\n\nmatch the industrial employment share without requiring us to replicate the service\n\nand agricultural employment ratios.\n\n\nFinally, the replication of the observed sectoral employment and\n\nMention: \"Penn World Tables\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Aggregate real consumption is read directly from these tables to drive the calibration.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000684:34:1", "occurrence_id": "dee8e4ee205a6a8b3cd896ea", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .0340*** 0.0365** 0.00415\n10.pot ~~e~~ xp#c.indepvar 0.258** 0.0485*** 0.0252* 0.00530**\nindepvar -0.00147 0.00867 0.00990 0.00243\ncohort -0.00888*** -0.00927*** -0.00697*** -0.00701***\n\n\n# obs 15,378 15,378 15,378 15,378\nR-squared 0.722 0.725 0.723 0.723\nIndependent Variable youth ~~e~~ mployment log ~~o~~ ilprice log ~~w~~ heatprice Gdp ~~g~~ rowth\n\n\nNotes: The dummy variable of being employed in the agricutural sector or not is regressed in each column on country\nlevel economic condition indicators at the year of graduation: youth employment rate (15-24 years old), oil price, wheat\nprice and gdp growth; interacted with potential years of experience 0 to 10. The potential experience is captured by\nthe difference between the individuals’ age at the year of the survey and their age at the year of graduation.\n(ii) Governorate and potential experience fixed effects are included in each regression. Controls for eudcational attainment are also included.\n(iii) Sample of males who graduated between 1990-2019. Data are constructed from LFS 2006-2019, and the economic\nconditions indicators are extracted from World Bank data.\n(iv) Cluster-robust standard errors in parentheses (clustering on cohort and governorate of residence).\n(v) *, ** and ***refer to significance at the 10, 5 and 1% levels respectively.\n\n\n32\n\nMention: \"LFS 2006-2019\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Male graduates' outcomes are measured on data constructed from these labour force survey rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000684:34:1", "occurrence_id": "973e4fb5d28f0835ad91a7ce", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .0340*** 0.0365** 0.00415\n10.pot ~~e~~ xp#c.indepvar 0.258** 0.0485*** 0.0252* 0.00530**\nindepvar -0.00147 0.00867 0.00990 0.00243\ncohort -0.00888*** -0.00927*** -0.00697*** -0.00701***\n\n\n# obs 15,378 15,378 15,378 15,378\nR-squared 0.722 0.725 0.723 0.723\nIndependent Variable youth ~~e~~ mployment log ~~o~~ ilprice log ~~w~~ heatprice Gdp ~~g~~ rowth\n\n\nNotes: The dummy variable of being employed in the agricutural sector or not is regressed in each column on country\nlevel economic condition indicators at the year of graduation: youth employment rate (15-24 years old), oil price, wheat\nprice and gdp growth; interacted with potential years of experience 0 to 10. The potential experience is captured by\nthe difference between the individuals’ age at the year of the survey and their age at the year of graduation.\n(ii) Governorate and potential experience fixed effects are included in each regression. Controls for eudcational attainment are also included.\n(iii) Sample of males who graduated between 1990-2019. Data are constructed from LFS 2006-2019, and the economic\nconditions indicators are extracted from World Bank data.\n(iv) Cluster-robust standard errors in parentheses (clustering on cohort and governorate of residence).\n(v) *, ** and ***refer to significance at the 10, 5 and 1% levels respectively.\n\n\n32\n\nMention: \"World Bank data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Regional economic condition indicators are extracted from this source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000685:31:2", "occurrence_id": "0b3634057b8a8a9dab3fe9c9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: >1</u>\n_λ_ 11\n\n\n\nThis is the familiar expression for gains from trade that is obtained in a wide class of models\n\n\n(Arkolakis et al., 2012). Once we introduce multiple regions with internal trade costs and mul\n\ntiple sectors ( _γ_ _<_ 1), the expression changes.\n\n###### **B. Data: Sources and cleaning**\n\n\n**Annual Survey of Industries (ASI)**\n\n\nUnder the Indian Factory Act of 1948, all establishments without power and employing more\n\n\nthan 20 workers or those with power and employing more than 10 workers are required to be\n\n\nregistered with the Chief Inspector of Factories in each state. This register is used as the sam\n\npling frame for the ASI. The ASI extends to the entire country, except the states of Arunachal\n\n\nPradesh, Mizoram and Sikkim and the Union Territory (UT) of Lakshadweep. The ASI provides\n\n\nstatistical information to assess changes in the growth, composition, and structure of the orga\n\nnized manufacturing sector, comprising activities related to manufacturing processes, repair\n\n\nservices, gas and water supply, and cold storage. Following data cleaning in prior research, cer\n\ntain states such as the seven North-Eastern states and the erstwhile state of Jammu and Kash\n\nmir are also excluded from the analysis. At the same time, districts whose total plant count was\n\n\nless than 50 were dropped.\n\n\n**Creating consistent districts**\n\n\nOne challenge with using district identifiers is that the set of districts evolved quite significantly\n\n\nover the period of study (1989-2009). Accordingly, we create consistent district boundaries for\n\n\nthe entire period. Our source of GIS data for district-boundaries is DataMeet (datameet.org).\n\n\nThe number of districts in India increased from 466 in the 1991 Census to 640 in the 2011\n\n\nCensus. Given that the total area of the country did not change, the new districts were carved\n\n\nout of existing districts\n\nMention: \"Annual Survey of Industries\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Registered manufacturing establishments provide the plant-level data under the factory legislation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000685:31:2", "occurrence_id": "4f84740522933273236c359a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: >1</u>\n_λ_ 11\n\n\n\nThis is the familiar expression for gains from trade that is obtained in a wide class of models\n\n\n(Arkolakis et al., 2012). Once we introduce multiple regions with internal trade costs and mul\n\ntiple sectors ( _γ_ _<_ 1), the expression changes.\n\n###### **B. Data: Sources and cleaning**\n\n\n**Annual Survey of Industries (ASI)**\n\n\nUnder the Indian Factory Act of 1948, all establishments without power and employing more\n\n\nthan 20 workers or those with power and employing more than 10 workers are required to be\n\n\nregistered with the Chief Inspector of Factories in each state. This register is used as the sam\n\npling frame for the ASI. The ASI extends to the entire country, except the states of Arunachal\n\n\nPradesh, Mizoram and Sikkim and the Union Territory (UT) of Lakshadweep. The ASI provides\n\n\nstatistical information to assess changes in the growth, composition, and structure of the orga\n\nnized manufacturing sector, comprising activities related to manufacturing processes, repair\n\n\nservices, gas and water supply, and cold storage. Following data cleaning in prior research, cer\n\ntain states such as the seven North-Eastern states and the erstwhile state of Jammu and Kash\n\nmir are also excluded from the analysis. At the same time, districts whose total plant count was\n\n\nless than 50 were dropped.\n\n\n**Creating consistent districts**\n\n\nOne challenge with using district identifiers is that the set of districts evolved quite significantly\n\n\nover the period of study (1989-2009). Accordingly, we create consistent district boundaries for\n\n\nthe entire period. Our source of GIS data for district-boundaries is DataMeet (datameet.org).\n\n\nThe number of districts in India increased from 466 in the 1991 Census to 640 in the 2011\n\n\nCensus. Given that the total area of the country did not change, the new districts were carved\n\n\nout of existing districts\n\nMention: \"DataMeet\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"District boundary geometries are taken from this open collection to build consistent boundaries over time.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000688:8:0", "occurrence_id": "b1048f907d2883fd2be3e883", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: offered a 100% subsidy to lifeline (low-income) customers who consume 0-50 kWh of electricity a month, while non-lifeline customers (consuming above 50 kWh a month) would\nenjoy a 50% subsidy (Berkouwer et al., 2022). The subsidies were announced on March 29\nand took effect from April 1. Eligibility status was determined based on March consumption. The policy was initially announced to last for three months, however, the subsidy for\n\n[only] lifeline customers was eventually extended for 12 months.\n\n#### **3 Data**\n\n\nThis paper uses administrative data on electricity billing records from Ghana and Rwanda.\nThe Ghana data comes from the Electricity Company of Ghana (ECG), which is the largest\ndistributor in the country with operations in the southern and middle belts. It accounts\nfor nearly 70% of all electricity customers in the country. We use data on billing records\nof the universe of electricity customers of the ECG from January 2018 to December 2020.\nThe data identifies customer types based on the tariff applicable: residential (households),\nnon-residential, and heavy industries. For each customer and year-month, it records the\namount (kWh) of electricity consumed, the monetary value in Ghana Cedis (GHS), meter\ntype (postpaid vs prepaid), and location (district) of the customer. In all, the data contains\n42 million customer-year-month observations.\nThe Rwanda data comes from the Energy Utility Corporation Limited (EUCL), the\nmain distributor, via the Rwanda Utilities Regulatory Authority (RURA). The dataset contains the billing records of the universe of electricity customers in Rwanda from January\n2018 to December 2020. The data identifies customer type based on the tariff applicable:\nresidential, non-residential (commercial, hotels, health centers, and public works ( water\nstorage and pump stations, broadcasters)), and small-and-medium industries.\n\nMention: \"administrative data on electricity billing records\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Billing records for the universe of customers in two countries measure the effect of the subsidy.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000688:8:0", "occurrence_id": "0467962ab47377ad7acf8f9e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: offered a 100% subsidy to lifeline (low-income) customers who consume 0-50 kWh of electricity a month, while non-lifeline customers (consuming above 50 kWh a month) would\nenjoy a 50% subsidy (Berkouwer et al., 2022). The subsidies were announced on March 29\nand took effect from April 1. Eligibility status was determined based on March consumption. The policy was initially announced to last for three months, however, the subsidy for\n\n[only] lifeline customers was eventually extended for 12 months.\n\n#### **3 Data**\n\n\nThis paper uses administrative data on electricity billing records from Ghana and Rwanda.\nThe Ghana data comes from the Electricity Company of Ghana (ECG), which is the largest\ndistributor in the country with operations in the southern and middle belts. It accounts\nfor nearly 70% of all electricity customers in the country. We use data on billing records\nof the universe of electricity customers of the ECG from January 2018 to December 2020.\nThe data identifies customer types based on the tariff applicable: residential (households),\nnon-residential, and heavy industries. For each customer and year-month, it records the\namount (kWh) of electricity consumed, the monetary value in Ghana Cedis (GHS), meter\ntype (postpaid vs prepaid), and location (district) of the customer. In all, the data contains\n42 million customer-year-month observations.\nThe Rwanda data comes from the Energy Utility Corporation Limited (EUCL), the\nmain distributor, via the Rwanda Utilities Regulatory Authority (RURA). The dataset contains the billing records of the universe of electricity customers in Rwanda from January\n2018 to December 2020. The data identifies customer type based on the tariff applicable:\nresidential, non-residential (commercial, hotels, health centers, and public works ( water\nstorage and pump stations, broadcasters)), and small-and-medium industries.\n\nMention: \"Electricity Company of Ghana\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"This distributor supplies the Ghanaian half of the panel, covering nearly 70 percent of customers.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000690:33:1", "occurrence_id": "7d9d08a49d7d17e360e378a0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 285*** 0.349*** 0.350*** 0.592***\n(0.086) (0.049) (0.032) (0.082)\nPrecipitation -0.156** 0.474*** 0.267** 0.699***\n(0.072) (0.100) (0.110) (0.170)\nSubnational FE Yes No Yes No\nYear FE Yes No Yes No\nMean dependent var. 35.605 35.605 25.383 25.383\nObservations 4,129 1,019 4,129 1,019\nEquality test (Panel vs.\np = 0.383 p = 0.029\n<u>long differences)</u>\n\n<u>Equality test (country</u>\np = 0.000 p = 0.000 p = 0.000 p = 0.000\nvs. subnational)\n\n\n\nNumber of countries 128 90 128 90\n<u>Number of regions</u> <u>1,484</u> <u>1,019</u> <u>1,484</u> <u>1,019</u>\n_<u>Notes:</u>_ <u>Robust standard errors in parentheses. Standard errors are clustered at the</u>\ncountry level in Panel A and subnational level in Panel B. Regressions are\nweighted by region population. Inequality data are taken from the GSAP.\nInequality and weather variables in the long-differences model are measured by\nthe difference between averages of the earliest 3-year period and averages of the\nlatest 3-year period. The long differences estimation is based on cross-sectional\ndata with a smaller sample size compared with panel data. The data for the countrylevel analysis and the subnational-level analysis comes from our newly\nconstructed\n\nMention: \"GSAP\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Subnational inequality measures are taken from this atlas for the weather regressions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000691:8:0", "occurrence_id": "0c0bbf71c476d74a12a11f29", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Somali regions. <sup>10</sup> Several terrorist attacks occurred during fieldwork of the SHFS. <sup>11</sup> The\n\nanalysis exploits detailed household data with dates of interviews and GPS positions to\n\nevaluate the immediate impact of attacks against civilians in Somalia. The types of\n\nincidents considered correspond to attacks from Al-Shabaab against civilians. <sup>12</sup> We\n\nconcentrate on measuring the effect within a week due to data availability given that i)\n\nthe questionnaires used a recall period of 7 days for food items, which is the main\n\ncomponent of the consumption aggregate; and ii) only one household was interviewed 8\n\ndays or more after an incident.\n\n\n3.1 Data sources\n\n\nThe main sources of data correspond to detailed household information on socio\ndemographic characteristics, perceptions and poverty conditions from Waves 1 and 2 of\n\nthe Somali High Frequency Survey, as well as location and time of attacks from The\n\nArmed Conflict Location & Event Data Project (ACLED). In this way, household data from\n\nthe SHFS is combined with geo-tagged information on attacks to identify households\n\nexposed to terrorist incidents.\n\n\nWave 1 includes 4,117 households interviewed between February and March 2016,\n\nwhich are representative of 9 of the 18 Somali pre‐war regions, as the remaining areas\n\nwere inaccessible for security reasons at the time of fieldwork. Wave 2 expanded the\n\ncoverage to 17 pre‐war regions, interviewing 6,092 households between December 2017\n\nand January 2018. <sup>13</sup> ACLED data records locations and intensity of armed conflict coded\n\nby researchers who collect information from secondary sources, NGO reports, local and\n\ninternational news reports and research publications (Raleigh and Dowd 2015). The\n\ndatabase is unique due to its geographical level of precision when reporting the latitude\n\nand longitude of the attacks, indicating the location of an incident (Figure 2). All the\n\nviolent incidents reported in ACLED that occurred a\n\nMention: \"Somali High Frequency Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Two waves provide household characteristics, perceptions and poverty for the exposure analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000691:8:0", "occurrence_id": "a61c68be1562530a704538cd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Somali regions. <sup>10</sup> Several terrorist attacks occurred during fieldwork of the SHFS. <sup>11</sup> The\n\nanalysis exploits detailed household data with dates of interviews and GPS positions to\n\nevaluate the immediate impact of attacks against civilians in Somalia. The types of\n\nincidents considered correspond to attacks from Al-Shabaab against civilians. <sup>12</sup> We\n\nconcentrate on measuring the effect within a week due to data availability given that i)\n\nthe questionnaires used a recall period of 7 days for food items, which is the main\n\ncomponent of the consumption aggregate; and ii) only one household was interviewed 8\n\ndays or more after an incident.\n\n\n3.1 Data sources\n\n\nThe main sources of data correspond to detailed household information on socio\ndemographic characteristics, perceptions and poverty conditions from Waves 1 and 2 of\n\nthe Somali High Frequency Survey, as well as location and time of attacks from The\n\nArmed Conflict Location & Event Data Project (ACLED). In this way, household data from\n\nthe SHFS is combined with geo-tagged information on attacks to identify households\n\nexposed to terrorist incidents.\n\n\nWave 1 includes 4,117 households interviewed between February and March 2016,\n\nwhich are representative of 9 of the 18 Somali pre‐war regions, as the remaining areas\n\nwere inaccessible for security reasons at the time of fieldwork. Wave 2 expanded the\n\ncoverage to 17 pre‐war regions, interviewing 6,092 households between December 2017\n\nand January 2018. <sup>13</sup> ACLED data records locations and intensity of armed conflict coded\n\nby researchers who collect information from secondary sources, NGO reports, local and\n\ninternational news reports and research publications (Raleigh and Dowd 2015). The\n\ndatabase is unique due to its geographical level of precision when reporting the latitude\n\nand longitude of the attacks, indicating the location of an incident (Figure 2). All the\n\nviolent incidents reported in ACLED that occurred a\n\nMention: \"Armed Conflict Location & Event Data Project\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Attack locations and times are geo-tagged to households to identify exposure to incidents.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000691:8:0", "occurrence_id": "12b7bd60c14565a2fd660325", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Somali regions. <sup>10</sup> Several terrorist attacks occurred during fieldwork of the SHFS. <sup>11</sup> The\n\nanalysis exploits detailed household data with dates of interviews and GPS positions to\n\nevaluate the immediate impact of attacks against civilians in Somalia. The types of\n\nincidents considered correspond to attacks from Al-Shabaab against civilians. <sup>12</sup> We\n\nconcentrate on measuring the effect within a week due to data availability given that i)\n\nthe questionnaires used a recall period of 7 days for food items, which is the main\n\ncomponent of the consumption aggregate; and ii) only one household was interviewed 8\n\ndays or more after an incident.\n\n\n3.1 Data sources\n\n\nThe main sources of data correspond to detailed household information on socio\ndemographic characteristics, perceptions and poverty conditions from Waves 1 and 2 of\n\nthe Somali High Frequency Survey, as well as location and time of attacks from The\n\nArmed Conflict Location & Event Data Project (ACLED). In this way, household data from\n\nthe SHFS is combined with geo-tagged information on attacks to identify households\n\nexposed to terrorist incidents.\n\n\nWave 1 includes 4,117 households interviewed between February and March 2016,\n\nwhich are representative of 9 of the 18 Somali pre‐war regions, as the remaining areas\n\nwere inaccessible for security reasons at the time of fieldwork. Wave 2 expanded the\n\ncoverage to 17 pre‐war regions, interviewing 6,092 households between December 2017\n\nand January 2018. <sup>13</sup> ACLED data records locations and intensity of armed conflict coded\n\nby researchers who collect information from secondary sources, NGO reports, local and\n\ninternational news reports and research publications (Raleigh and Dowd 2015). The\n\ndatabase is unique due to its geographical level of precision when reporting the latitude\n\nand longitude of the attacks, indicating the location of an incident (Figure 2). All the\n\nviolent incidents reported in ACLED that occurred a\n\nMention: \"ACLED data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its coding from secondary sources and reports is described to justify its use.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000693:15:0", "occurrence_id": "b848180be17c5950a5d664be", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: discounts; (iii) whether the states offered females stamp duty discounts; and (iv) the feasibility and cost\nof data collection. Each aspect is discussed in detail below.\n\n###### 4.1.1 City size\n\n\nIndia is both very large geographically and very diverse culturally, socially, and economically. More than\n4,000 ULBs have discretion to define their own policies on recurrent property taxes. Further, India does\nnot have a centralized repository of all legislation relating to property taxes that might provide a\ncomprehensive review identifying cities that offer property tax incentives to females. A further\ncomplication is India’s linguistic diversity: Many of its 4,000 ULBs use languages other than Hindi or English\nto write their legislation, thus requiring translations for study. In view of these issues, the sample of cities\nwas limited to India’s 20 largest, based on total population as measured by the UN World Urbanization\nProspects 2018, and the ease of obtaining relevant legal information. The selected cities are listed (by\nurban area population) in Appendix 4.\n\n###### 4.1.2 Recurrent property tax discounts to females\n\n\nThe review of legislation, media reports, and telephone conversations with government officials found\nthat only 4 of the 20 cities selected provide females with recurrent discounts on property taxes: Pune,\nPimpri-Chinchwad, Delhi, and Chennai.\n\n###### 4.1.3 Stamp duty discounts to females\n\n\nUnlike cities, most states publicize their Stamp Acts electronically on dedicated websites, allowing legal\nreview of the state-level Stamp Acts of all 28 states and Delhi. This review, and telephone conversations\nwith government officials, identified nine states and Delhi that provided stamp duty and registration\ncharge discounts to females: Assam, Bihar, Gujarat, Haryana, Maharashtra, Odisha, Punjab, Uttarakhand,\nand Uttar Pradesh.\n\n###### 4.1.4 Feasibility and cost of data collection\n\n\nDue\n\nMention: \"UN World Urbanization\nProspects 2018\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Urban area population from this projection determined which twenty cities entered the sample.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000694:5:0", "occurrence_id": "2df4a08b7bd24093e35016f4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Before the pandemic hit, the labor market in Egypt faced persistent challenges. While Egypt’s GDP\ngrowth reached 5.6 percent in 2019, labor force participation continued to decline. This partly\nreflected challenges in absorbing the sustained growth in the country’s working-age population.\nEgypt's unemployment rate was high, particularly for women, youth, and the most highly educated,\nsignaling a difficulty in transitioning from school to work and in integrating women into the labor\nmarket. Furthermore, informality was high and increasing, reflecting the need to improve the quality\nof jobs. Jobs with no social security rose to encompass 67 percent of all jobs in 2018, up from 57\npercent in 2010. These trends reflected difficulties from the demand side of the labor market, where\nthe most productive sectors were not the most important in creating jobs. Pre-COVID labor market\ntrends pointed to the importance of accelerating private sector job creation. While about 78 percent\nof jobs in Egypt were in the private sector, this share remained well below the global average of 90\npercent, and about 3 in 4 private sector workers in Egypt were engaged in informal, low-quality\nactivities characterized by low productivity and low pay (Cortes et al. 2022).\n\nThe COVID-19 pandemic exerted its greatest impact on Egypt during the final quarter of FY20, with the\ndeclaration of a national public health emergency in mid-March 2020. That quarter overlapped with\nthe main lockdown implemented by the GoE. Mobility trends show how activity fell during the last\nquarter of FY20 and subsequently in workplaces, transit stations, and areas related to retail and trade.\nMobility and economic activity started to recover during FY21 (Figure 2).\n\n\nFigure 2: Mobility trends in Egypt, relative to February 2020 (pre-COVID-19)\n\n\n\n\n\n\n\nSource: Authors’ calculations based on Google COVID-19 Community Mobility Trends (as of March 2022).\n\n\nFollowing its mid-March 2020 emergency declaration, the\n\nMention: \"Google COVID-19 Community Mobility Trends\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Visits to place categories relative to February 2020 are calculated to show the recovery in activity.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000686:5:0", "occurrence_id": "a41e4191af78cb0197dae294", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: We find that firms in the lower 25 <sup>th</sup> percentile of sales and the 75 <sup>th</sup> percentile of employment were more\naffected in the aftermath of Russia’s invasion of Ukraine than other firms. We try to explain the\nheterogeneity by examining two candidate structural factors: firms’ export links to Russia and digital gains\nprior to the invasion. Firms with export links to Russia are estimated to have suffered significantly more\nin terms of dropping sales and employment. We do not find any significant mitigating effect of digitization\nafter the invasion shock that the existing literature would suggest. However, the results suggest that\ndigitization helped firms increase employment on average during the studied period. While export links\nto Russia appear to increase sales on average during the studied period, this effect does not survive all\nthe robustness tests we apply.\n\n\nThe remainder of the paper is organized as follows. Section 2 describes the employed data and their\nsummary statistics. Section 3 explains the estimation approach and methodology. Section 4 discusses the\nresults. Section 5 concludes and highlights some policy implications.\n\n### 2. Data\n\nWe use data from the Business Pulse Survey (BPS) developed by the World Bank to measure the impact\nof the COVID-19 pandemic on the private sector. The Business Pulse Survey (BPS) is a novel dataset that\ntracks the potential impact of the pandemic on the private sector along critical dimensions of business\nperformance, such as sales revenue, liquidity, and solvency, labor adjustments, adoption of technology,\nbusiness expectations and uncertainty, and access to government support programs. In most countries,\nthe survey interviews were conducted over the phone. The data include micro, small, medium, and large\nbusinesses across all main sectors (i.e., agriculture, manufacturing, retail, and other services, including\nconstruction). The sampling frame excluded micro-firms and/or businesses in agriculture in some\ncountries. In other countries, when micro\n\nMention: \"Business Pulse Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"This repeating survey tracks pandemic effects on the private sector across several dimensions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000695:36:0", "occurrence_id": "df11ae95da563052769060a1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure A1: Growing Importance of Minerals in Ghana’s Export Basket, 2006–2018**\n\n\n_Source_ : Authors’ estimates based on information from the Atlas of Economic Complexity by the Growth Lab at Harvard\n[University (https://atlas.cid.harvard.edu/).](https://atlas.cid.harvard.edu/)\n\n\n35\n\nMention: \"Atlas of Economic Complexity\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Export basket composition by product group is estimated from this trade atlas.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000699:7:0", "occurrence_id": "3ab057d4b8db0dc867bd3651", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The dramatic picture emerging from Figure 1 include declines up to 80% and more importantly and differently from\n\n\nother infrastructure sectors that have been bouncing back to pre-pandemic levels by the end of 2020, public transport\n\n\nstill suffers from a 50% decline at the end of 2020.\n\n\nWhereas the decrease in ridership of public transport per se does not necessarily imply a change in intermodal\n\n\ncompetition towards private cars, the implications in terms of congestion remain unclear. That is why we also\n\n\ninvestigate trends in congestion, using high frequency data from Tom Tom. Figure 2 reports the percentage of cities\n\n\nwith low congestion for the sample of cities. The picture portrayed is different from Figure 1. Whereas indeed\n\n\nduring the peak of the pandemic 80 percent of the cities were characterized by low congestion, the percentage drops\n\n\nalmost to an insignificant percentage as lockdowns were relaxed and some of the peaks for the second wave of the\n\n\npandemic were much lower and only of a much shorter duration.\n\n\n**Figure 2: Percentage of cities with low congestion for the sample of cities (1st Jan to 31th Dec 2020)**\n\n\n_Source:_ Elaboration from Tom Tom\n\n\nThe stringency of the national lockdown is measured using both cross sectoral and sectoral indicators. The first\n\n\nindex represents the overall stringency of the lockdown, and it consists of the average value of the government\n\n\nresponse stringency index during the period when a full lockdown was in place, from the Oxford Government\n\n\nResponse Tracker. The evolution of this index over time is depicted in Figure 3, ranging from 0 (less stringent) to\n\n\n100 (most stringent) and is based on the policy decisions taken by governments across several areas: workplace\n\n\nrestrictions, mobility restrictions, school closures, and restrictions on gatherings and public events (Hale _et al._,\n\n\n2020).\n\n\n6\n\nMention: \"Oxford Government\n\n\nResponse Tracker\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Average lockdown stringency during full closure periods is computed from this tracker.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000700:47:1", "occurrence_id": "31e3bdd269eb36319e81cd4e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: /u> <u>480</u> <u>480</u> <u>480</u>\n\n\n\n\n<u><mark>> 15 & <= 50</mark></u> <u><mark>480</mark></u> <u><mark>480</mark></u> <u><mark>480</mark></u>\n\n<u>> 50 & <= 200</u> <u>610</u> <u>720</u> <u>720</u>\n\n<u><mark>> 200</mark></u> <u><mark>740</mark></u> <u><mark>740</mark></u> <u><mark>740</mark></u>\n<u>Source: The Council for the Development of Cambodia (CDC).</u>\n\n\nThe 2019/20 CSES captures information on household spending on electricity. Respondents also provide\ninformation on whether electricity is provided from the public network. Subsidies are estimated only for\nthose households drawing power from the public grid. For each area we assign a standard, non-subsidized\nprice per kWh of consumed electricity. From expenditures on electricity, we calculate consumption in\nkWh, depending on geography (special provinces/other urban/other rural) and actual consumption. We\ncalculate the subsidy as the difference in standard price less subsidized price times consumption in kWh.\n\n\n_Modeling social transfers_\n\n\nIn recent years, Cambodia has increased social spending. However, social transfers remain very limited.\nIn 2019, the government implemented a broad program for supporting pregnant women and infants in\nthe form of a Conditional Cash Transfer (MOE and MOH, 2019). This new Conditional Cash Transfer (CCT)\nprogram is included in\n\nMention: \"CSES\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Household electricity spending and grid connection are used to estimate subsidies for public network users.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000703:8:0", "occurrence_id": "b4a99dd5f484d4313ee5e1b0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: “Other” and then hand-matched bank names from the survey with bank names from Fitch Connect. Hand\n\nmatching was necessary because the survey included spelling mistakes and in several cases the name of the\n\n\nbank was in local language in the survey and in English in Fitch Connect. After dropping observations with\n\n\nbanks for which Fitch Connect does not report any financial information, we were left with a dataset of\n\n\napproximately 13,000 firms and more than 500 banks. While we use the matched dataset to analyze the\n\n\neffect of state-ownership on firms’ access to finance, we retain all the unmatched firm records to test for\n\n\npotential statistically significant differences in firm characteristics between matched and unmatched firms.\n\n\nNote that WBES also includes data from 5 advanced European economies, but we do not use these data in\n\n\nour empirical analysis which focuses on emerging and developing economies. After dropping these five\n\n\ncountries we are left with 11,840 records with a bank-firm match.\n\n###### Summary Statistics\n\n\nWe report summary statistics of firm characteristics in Table 2, 3, and 4. Table 2 includes the full sample of\n\n\nfirms, while Table 3 and Table 4 report the sub-samples of firms matched to private banks, and to state\n\nowned banks, respectively. There is a good degree of similarity among these 3 groups. The average firm\n\n\nsize, measured by the number of workers employed by the firm, is 75 for the full sample, 81 for the\n\n\nsubsample of firms matched to private banks, and 108 for the subsample of firms matched to state-owned\n\n\nbanks. The average age is 18 years for the full sample, and 20 for the 2 subsamples of matched enterprises.\n\n\nFinancing conditions faced by firms also appear to be similar across the three samples. The reported value\n\n\nof collateral needed for a loan, measured as percent of the loan value, is 167 for the full sample, 188 percent\n\n\nfor firms that have a relationship with private banks, and 183 percent for\n\nMention: \"Fitch Connect\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"Bank names in local language and with spelling errors were hand-matched to this financial database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000703:8:0", "occurrence_id": "c231fd26f0d2e0a23122e2f1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: “Other” and then hand-matched bank names from the survey with bank names from Fitch Connect. Hand\n\nmatching was necessary because the survey included spelling mistakes and in several cases the name of the\n\n\nbank was in local language in the survey and in English in Fitch Connect. After dropping observations with\n\n\nbanks for which Fitch Connect does not report any financial information, we were left with a dataset of\n\n\napproximately 13,000 firms and more than 500 banks. While we use the matched dataset to analyze the\n\n\neffect of state-ownership on firms’ access to finance, we retain all the unmatched firm records to test for\n\n\npotential statistically significant differences in firm characteristics between matched and unmatched firms.\n\n\nNote that WBES also includes data from 5 advanced European economies, but we do not use these data in\n\n\nour empirical analysis which focuses on emerging and developing economies. After dropping these five\n\n\ncountries we are left with 11,840 records with a bank-firm match.\n\n###### Summary Statistics\n\n\nWe report summary statistics of firm characteristics in Table 2, 3, and 4. Table 2 includes the full sample of\n\n\nfirms, while Table 3 and Table 4 report the sub-samples of firms matched to private banks, and to state\n\nowned banks, respectively. There is a good degree of similarity among these 3 groups. The average firm\n\n\nsize, measured by the number of workers employed by the firm, is 75 for the full sample, 81 for the\n\n\nsubsample of firms matched to private banks, and 108 for the subsample of firms matched to state-owned\n\n\nbanks. The average age is 18 years for the full sample, and 20 for the 2 subsamples of matched enterprises.\n\n\nFinancing conditions faced by firms also appear to be similar across the three samples. The reported value\n\n\nof collateral needed for a loan, measured as percent of the loan value, is 167 for the full sample, 188 percent\n\n\nfor firms that have a relationship with private banks, and 183 percent for\n\nMention: \"WBES\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Advanced European economies in the survey are excluded to keep the focus on emerging economies.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000705:2:0", "occurrence_id": "030ba8b990f17ebaa44b6728", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: #### **Digital Technology Uses among Microenterprises:** **Why Is Productive Use So Low across Sub-Saharan Africa?** Izak Atiyas and Mark A. Dutz 1\n\nKeywords: internet, smartphone, digital technologies, productivity, jobs, inclusion.\n\n\nJEL codes: D22, J24, L25, L86, O14, O33, O55.\n\n\n1 Affiliations: Atiyas (Economic Research Forum and TUSIAD-Sabanci University Competitiveness Forum), Dutz\n(Consultant, IFC). Corresponding author: mdutz@ifc.org. This research has received funding from the joint\nAfrica Chief Economist-Digital Development Research Program on Digital Transformation for Africa (project ID:\nP170151) and from IFC. The authors are grateful to Alison Gillwald and Onkokame Mothobi for making\navailable the RIA (Research ICT Africa) ICT Access business data for 2017-18 and for their support in the use of\nthese data.\n\nMention: \"ICT Access business data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Business ICT access data made available by the research network underpin the analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000708:3:0", "occurrence_id": "e493a791531dc7b8981421a6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 2\n\n### 1. Background\n\n\nLow levels of engagement among women in MNA’s labor force vary from country to country yet\nfemale unemployment is consistently high across all nations. Female labor force participation\n(FLFP) is as high as 58% in Qatar, 57.5% in Kuwait, 55% in Djibouti, and as low as 14% in Jordan,\n12% in the Syrian Arab Republic and 6% in the Republic of Yemen. <sup>1</sup> [^1: All statistics are from the ILOSTAT database, the latest year for which data is available was 2018. Retrieved from\n<u>[https://data.worldbank.org/](https://data.worldbank.org/)</u> in October, 2019.] In the Kurdistan Region of\nIraq (KRI), according to the labor force and demographic surveys conducted in 2012, 2013,\n2014, 2015 and 2017, FLFP was 12.2%, 12.7%, 13.1%, 14.8% and 13.8% respectively. <sup>2</sup> [^2: Labor force surveys were conducted in 2012, 2013, 2014, and 2015, and a demographic survey was conducted by\nin 2017, the most recent source for the FLFP statistic. All surveys were conducted by the Kurdistan Regional\nStatistics Office.\n[3 ILOSTAT database, retrieved from https://data.worldbank.org/](https://data.worldbank.org/) in October, 2019.] Female\nunemployment in nearly all countries in MNA is at least twice as much as that of their male\npeers. For example, the unemployment rate for women in Iraq is 3 times higher than men, at\n31% compared to 10% in 2017. <sup>3</sup>\n\nSocial norms and cultural beliefs dictate behavior in ways that affect important development\noutcomes for women, including their participation\n\nMention: \"ILOSTAT database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Female labour force participation rates by country are taken from this labour statistics database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000709:7:1", "occurrence_id": "7c734a4ccd979632710c17d7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  of national or subnational governments, but also considers\ncases in which the government has special voting powers (i.e. nominating board of directors). <sup>10</sup> In 2020,\nthe IMF Fiscal Monitor considered state-owned firms as those with state participation as low as 20 percent\nand also relied on other criteria to define SOEs, such as national legal forms (IMF, 2020).\n\n\nTo resolve this lack of harmonization, we propose an economic definition to define a firm as an SOE based\non the government’s control as well as their role in the market. That is, an entity is considered a stateowned enterprise for the purpose of the Global BOS database if it satisfies the following conditions:\n\n\nI. It is controlled by government units or by other public corporations, proxied by a level of\ndirect or indirect (i.e., through subsidiaries) participation of above 10%; <sup>11</sup>\n\n\nII. It is recognized by law as a legal entity separate from its owners; <sup>12</sup>\n\n\nIII. It can generate profit or other financial gain for its owners; <sup>13</sup>\n\n\n8 This classification is described in further detail in Dall’Olio et. al. (2022).\n9 For instance, in Indonesia the state-owned enterprises are denoted as two different legal forms controlled by the governments: Badan Usaha\nMilik Negara (national) and locally owned (Badan Usaha Milik Daerah). For instance, SOEs are defined in Azerbaijani as public interest entities\n(PIEs), and in Mozambique as public enterprises and shareholding companies (World Bank, 2016).\n10 Even within different OECD workstreams, there is no harmonized definition. For example, an SOE survey conducted in 2015 denoted SOEs as\ncorporate entities recognized by the national law in which only\n\nMention: \"Global BOS database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"inform\", \"summary\": \"The database's definition of a state-owned enterprise is set out through control and market-role conditions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000709:7:1", "occurrence_id": "7726f6fa515a4ff5a4700a4f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  of national or subnational governments, but also considers\ncases in which the government has special voting powers (i.e. nominating board of directors). <sup>10</sup> In 2020,\nthe IMF Fiscal Monitor considered state-owned firms as those with state participation as low as 20 percent\nand also relied on other criteria to define SOEs, such as national legal forms (IMF, 2020).\n\n\nTo resolve this lack of harmonization, we propose an economic definition to define a firm as an SOE based\non the government’s control as well as their role in the market. That is, an entity is considered a stateowned enterprise for the purpose of the Global BOS database if it satisfies the following conditions:\n\n\nI. It is controlled by government units or by other public corporations, proxied by a level of\ndirect or indirect (i.e., through subsidiaries) participation of above 10%; <sup>11</sup>\n\n\nII. It is recognized by law as a legal entity separate from its owners; <sup>12</sup>\n\n\nIII. It can generate profit or other financial gain for its owners; <sup>13</sup>\n\n\n8 This classification is described in further detail in Dall’Olio et. al. (2022).\n9 For instance, in Indonesia the state-owned enterprises are denoted as two different legal forms controlled by the governments: Badan Usaha\nMilik Negara (national) and locally owned (Badan Usaha Milik Daerah). For instance, SOEs are defined in Azerbaijani as public interest entities\n(PIEs), and in Mozambique as public enterprises and shareholding companies (World Bank, 2016).\n10 Even within different OECD workstreams, there is no harmonized definition. For example, an SOE survey conducted in 2015 denoted SOEs as\ncorporate entities recognized by the national law in which only\n\nMention: \"SOE survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"An earlier survey's legal definition is quoted to show definitions are not harmonised.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000715:13:1", "occurrence_id": "553ae74a0491e360eecba28f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  and GWP data**\n\n\n**Notes:** Panel A shows the raw observed subjective well-being data circa 2000 from Desmet et al. (2018); Panel B shows\n\nthe scale for adjustment obtained with additional data on subjective well-being reflecting its subnational heterogeneity\n\nwithin countries; and Panel C shows the resulting adjusted well-being which is used in the quantification of migration\n\nfrictions.\n\n\n12 Further details on the data collection and assembling are provided in the appendix.\n\n\n12\n\nMention: \"GWP data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Observed subjective well-being around 2000 and its subnational adjustment are drawn from this poll.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000716:10:1", "occurrence_id": "48423e16228b08f2dfada1ce", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: -19 pandemic is still ongoing, and the surveys add a few\n\nnew interview questions over time, thus using the latest survey could help to examine the impact\n\nof COVID-19 more accurately. We also use the first-round survey to check the robustness and\n\ncompare the results with the results of the second-round survey to explore the dynamic of firm\n\nresponses during COVID-19.\n\n\nWe further collect country-level data from various sources. GDP per capita is from the World\n\nDevelopment Indicators database (WDI). The stringency of government policy to contain COVID\n19 is from Oxford COVID-19 Government Response Tracker (OxCGRT) database, which\n\nprovides information on the government policy responses to the COVID-19 pandemic across more\n\n\n9\n\nMention: \"World\n\nDevelopment Indicators database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Country-level GDP per capita is merged in as a control.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000716:10:1", "occurrence_id": "1e1ac42a3fc011311baf1874", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: -19 pandemic is still ongoing, and the surveys add a few\n\nnew interview questions over time, thus using the latest survey could help to examine the impact\n\nof COVID-19 more accurately. We also use the first-round survey to check the robustness and\n\ncompare the results with the results of the second-round survey to explore the dynamic of firm\n\nresponses during COVID-19.\n\n\nWe further collect country-level data from various sources. GDP per capita is from the World\n\nDevelopment Indicators database (WDI). The stringency of government policy to contain COVID\n19 is from Oxford COVID-19 Government Response Tracker (OxCGRT) database, which\n\nprovides information on the government policy responses to the COVID-19 pandemic across more\n\n\n9\n\nMention: \"Oxford COVID-19 Government Response Tracker\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Policy stringency across many countries is taken from this tracker.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000717:4:0", "occurrence_id": "21ea5f564672a3f3e7e09309", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: From a policy perspective, the comprehensive environmental and tax structure of the\n\nIRENCGE-DF model makes it well suited for analyzing a wide range of policy scenarios. In\n\nparticular, the “Fit for 55” package is analyzed in the model by limiting the level of emissions and\n\nletting the model calculate the endogenous price consistent with this cap. Thanks to the extensive\n\nfocus on tax modeling, the IRENCGE-DF model is able to properly fit scenarios with new structures\n\nof tax rates based on the energy content and environmental performance of the fuels and electricity\n\nand also to simulate broader taxable bases by including more products in the scope and by removing\n\nsome of the current exemptions and reductions. Several simulation scenarios will be discussed in the\n\nfollowing sections.\n\n\nThe remainder of the paper is organized as follows. In Section 2, we present the data and\n\nmethodology, reviewing the main aspects of the IRENCGE-DF model. In the Section 3, we present\n\nthe results of our policy simulations. The final section concludes with some policy implications.\n\n##### **2 Data and Methodology** **2.1 SAM Construction**\n\nThe model is calibrated on the 2017 SAM. We update the benchmark data to 2020 by using\n\nmacroeconomic variations resulting from the latest public economic and financial documents (i.e.\n\nItalian annual budget law, etc.).\n\n\nThe information contained in the SAM is obtained by combining national account data, such as the\n\nnational account matrix for 2014, supply-use tables for 2017, and Eurostat data. In addition, we\n\nintegrate missing information with data from tax returns available at the Department of Finance.\n\n\nNational accounts data provide detailed information on the final and intermediate consumption at\n\nactivity and commodity levels, though they do not contain detailed information on taxation. Hence,\n\nwe use the tax return data to distribute taxes and subsidies per commodity. In Table 1, we provide a\n\ndescription of the sets and structure used in the analysis. For\n\nMention: \"Eurostat data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"European statistics are combined with national accounts to construct the social accounting matrix.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000721:6:1", "occurrence_id": "b943f0a57dbfc72d0674a4a1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: /sup> This appears to stem from a change in\ngovernment in The Gambia reducing the chance of asylum in Europe, coupled with the effects of\nborder closures and a reduction in the desire to migrate due to the COVID-19 pandemic in the\nsecond half of our follow-up period (Bah et al, 2021). The consequence was that, despite focusing\ngeographically on the highest migration regions and then studying the demographic group most\nlikely to migrate, only 1.1 percent of our control group attempted backway migration in the 18month follow-up period. This limited the scope for our interventions to reduce irregular migration,\nsince ex post there was very little to reduce.\n\n\n4 Frontex data may underestimate the total amount of irregular migration by not collecting data on migrants who are\nnot intercepted by border patrol, those who change their country of origin, or who do not file for asylum status upon\narrival. However, even if the levels are an understatement, we believe they capture the trend in migration well. Our\nsurvey data questions on migration over time closely track the Frontex trends.\n\n5\n\nMention: \"Frontex data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Interception-based counts omit migrants not stopped or who do not claim asylum, so totals are understated.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000723:28:0", "occurrence_id": "5816a0f521591edcfee49101", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Table A2. Dataset for country characteristics**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n|Characteristics|Measures|Description|Source|\n|---|---|---|---|\n|Commodity<br>import|Commodity importer|An economy is defined as commodity exporter when, either (i) total commodities<br>exports accounted for 30 percent or more of total goods exports or (ii) exports of<br>any single commodity accounted for 20 percent or more of total goods exports. All<br>other countries are considered commodity importers.|World Bank (2018)|\n|Commodity<br>import|Net energy import|Net fuel imports, defined as fuel imports minus fuel exports as percent of GDP|World Development Indicators (WDI)|\n|Financial<br>openness|International asset<br>and liability|De facto financial openness, defined as the sum of international assets and<br>liabilities as a percent of GDP|Lane and Milesi-Ferretti 2007; IMF Balance<br>of Payments and International Investment<br>Position Statistics|\n|Financial<br>openness|Chinn-Ito Index|De jure financial openness as in Chinn-Ito Index (2006)|Chinn and Ito (2006)|\n|Financial<br>openness|Capital control|Capital Control Measures data set of restrictions on capital account inflows and<br>outflows for 10 categories of assets|Fernandez et al. (2016)|\n|Trade<br>openness<br>|Trade-to-GDP|Sum of exports and imports of goods and services as percent of GDP|WDI; IMF WEO|\n|Trade<br>openness<br>|Trade concentration|Product concentration and diversification indexes of exports and imports|UNCTAD|\n|Trade<br>\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Net energy imports and trade-to-GDP in the country characteristics table are taken from this series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000723:28:0", "occurrence_id": "127569ad651b2800d5d89006", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Table A2. Dataset for country characteristics**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n|Characteristics|Measures|Description|Source|\n|---|---|---|---|\n|Commodity<br>import|Commodity importer|An economy is defined as commodity exporter when, either (i) total commodities<br>exports accounted for 30 percent or more of total goods exports or (ii) exports of<br>any single commodity accounted for 20 percent or more of total goods exports. All<br>other countries are considered commodity importers.|World Bank (2018)|\n|Commodity<br>import|Net energy import|Net fuel imports, defined as fuel imports minus fuel exports as percent of GDP|World Development Indicators (WDI)|\n|Financial<br>openness|International asset<br>and liability|De facto financial openness, defined as the sum of international assets and<br>liabilities as a percent of GDP|Lane and Milesi-Ferretti 2007; IMF Balance<br>of Payments and International Investment<br>Position Statistics|\n|Financial<br>openness|Chinn-Ito Index|De jure financial openness as in Chinn-Ito Index (2006)|Chinn and Ito (2006)|\n|Financial<br>openness|Capital control|Capital Control Measures data set of restrictions on capital account inflows and<br>outflows for 10 categories of assets|Fernandez et al. (2016)|\n|Trade<br>openness<br>|Trade-to-GDP|Sum of exports and imports of goods and services as percent of GDP|WDI; IMF WEO|\n|Trade<br>openness<br>|Trade concentration|Product concentration and diversification indexes of exports and imports|UNCTAD|\n|Trade<br>\n\nMention: \"Chinn-Ito Index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"De jure financial openness is measured by this index.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000723:28:0", "occurrence_id": "96d721d04c9f863bd3ae14dd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Table A2. Dataset for country characteristics**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n|Characteristics|Measures|Description|Source|\n|---|---|---|---|\n|Commodity<br>import|Commodity importer|An economy is defined as commodity exporter when, either (i) total commodities<br>exports accounted for 30 percent or more of total goods exports or (ii) exports of<br>any single commodity accounted for 20 percent or more of total goods exports. All<br>other countries are considered commodity importers.|World Bank (2018)|\n|Commodity<br>import|Net energy import|Net fuel imports, defined as fuel imports minus fuel exports as percent of GDP|World Development Indicators (WDI)|\n|Financial<br>openness|International asset<br>and liability|De facto financial openness, defined as the sum of international assets and<br>liabilities as a percent of GDP|Lane and Milesi-Ferretti 2007; IMF Balance<br>of Payments and International Investment<br>Position Statistics|\n|Financial<br>openness|Chinn-Ito Index|De jure financial openness as in Chinn-Ito Index (2006)|Chinn and Ito (2006)|\n|Financial<br>openness|Capital control|Capital Control Measures data set of restrictions on capital account inflows and<br>outflows for 10 categories of assets|Fernandez et al. (2016)|\n|Trade<br>openness<br>|Trade-to-GDP|Sum of exports and imports of goods and services as percent of GDP|WDI; IMF WEO|\n|Trade<br>openness<br>|Trade concentration|Product concentration and diversification indexes of exports and imports|UNCTAD|\n|Trade<br>\n\nMention: \"Capital Control Measures data set\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Restrictions on capital account inflows and outflows across ten asset categories are coded in this dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000725:12:0", "occurrence_id": "c4d3c8cbd6994b9d6e33a953", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **3.2** **Empirical** **Facts**\n\n\nIn this section, I discuss three empirical findings that show a negative relationship between informal\n\nity rates and the accessibility of formal jobs in Mexico City: 1) informal workers are more sensitive\n\n\nto commuting costs and spend less time commuting; 2) informal workers are located in areas in\n\n\nwhich they have poor access to formal employment; and 3) informality rates decrease with transit\n\n\nimprovements that connect informal workers to formal employment.\n\n\n**3.2.1** **Cross-sectional** **Variation**\n\n\n**_Finding_** **_1:_** _Informal_ _workers_ _spend_ _less_ _time_ _commuting_ _and_ _work_ _closer_ _to_ _their_ _home_ _relative_ _to_\n\n\n_formal_ _workers._\n\n\nTo reach the first finding, I use the 2015 Intercensal Survey. With this data, I observe the residence\n\n\nand workplace and average commuting time of each worker at the municipality level. Exploiting\n\n\ncross-sectional variation, I compare the average commuting time and the workplace decisions of\n\n\ninformal vs. formal workers. I adopt the standard definition of informality based on the contractual\n\n\nrelationship of the worker, I also restrict the sample to individuals who worked the week before\n\n\nthe census interview. I run the following linear probability model to test whether informal workers\n\n\nspend less time commuting:\n\n\n_yi_ = _β_ 0 + _β_ 1Informal _i_ + _γXi_ + _γl_ ( _i_ ) + _γn_ ( _i_ ) + _γm_ ( _i_ ) + _ϵi,_ (3.1)\n\n\nwhere _yi_ is a dummy variable that takes the value of 1 if individual _i\n\nMention: \"2015 Intercensal Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Municipality-level residence, workplace and commuting time are compared across informal and formal workers.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000727:47:14", "occurrence_id": "3acf8d49c197877a22ddaf73", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 146\nsd 1.034 1.064 0.958 1.011 . 1.065 1.039 . . . 0.991 . 0.959\nobs 3325 3325 3324 3324 . 3323 3324 . . . 3324 . 3324\n<u><mark>Cronbach's alpha</mark></u> <u><mark>0.695</mark></u> <u><mark>0.577</mark></u> <u><mark>0.542</mark></u> <u><mark>0.435</mark></u> <u><mark>.</mark></u> <u><mark>.</mark></u> <u><mark>0.400</mark></u> <u><mark>.</mark></u> <u><mark>.</mark></u> <u><mark>.</mark></u> <u><mark>0.421</mark></u> <u><mark>.</mark></u> <u><mark>0.505</mark></u>\n\n\n\nNote: nbr items stands for the number of items aggregated to construct the corresponding skills measure. sd stands for standard deviation and obs for the number of observations. Empty cells indicate that the project\ndoes not have data for the specified skill. PSDM stands for Problem Solving and Decision Making. The Facebook data is from the \"Future of Business (FoB)\" survey. This study includes only 15 Sub Saharan Africa (SSA)\ncountries namely: Angola, Benin, Botswana, Cameroon, Côte d'Ivoire, Ethiopia\n\nMention: \"Future of Business (FoB)\" survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Skill measures for fifteen sub-Saharan countries come from this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000729:8:0", "occurrence_id": "ad721ce31a61a7747a3065f9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: coupled with reforms that strengthen the private sector and improve the business environment. Furthermore,\n\n\nif the prevalence of training is to adapt and update skills of educated workers towards work in the private\n\n\nsector, then the policy implication is not only to increase university education, but also to ensure it serves\n\n\nthe private sector as well. Finally, firms with better management practices tend to provide more training in\n\n\nECA but not in MENA. This warrants further investigation and may be because managerial practices are\n\n\nquite poor in the region. These findings are important for the MENA context given the low provision of\n\n\ntraining in the formal private sector.\n\n\nIn summary, our study makes several contributions to the literature. First it explores the role of management\n\n\npractices and education of workers on formal training both at the intensive and extensive margin. Second\n\n\nit utilizes a panel data set to account for several firm-level characteristics to validate the findings. And\n\n\nfinally, it updates several studies in the literature by employing recent firm-level data that includes\n\n\neconomies across Europe, Central Asia, and the MENA region. The rest of the paper is structured as\n\n\nfollows. Section II describes the data. Section III provides the empirical specification and identification\n\n\nstrategy, while section IV provides the results, section V provides robustness checks, and section VI\n\n\nconcludes.\n\n\n**II.** **Data**\n\nThe main source of firm-level data is the World Bank’s Enterprise Surveys (ES). This includes two samples\n\n\nof medium and large enterprises. <sup>3</sup> The first sample includes cross-sectional firm-level surveys for 35\n\n\neconomies across ECA and MENA, all surveyed circa 2013 and 2019/2020, right before the Covid-19\n\n\n3 Medium and large enterprises are defined as having 20 or more employees. The choice of medium and large\nenterprises is due inclusion of management practices measures that were only collected for medium and large\nfirms. There are some small firms that are in the sample for two reasons\n\nMention: \"Enterprise Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Two samples of medium and large enterprises across two regions form the main firm-level data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000730:26:0", "occurrence_id": "c94936e16d9b71858f4b64f2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: How can European banks exercise this form of market power while keeping a constant level\n\nof new loan originations like their untreated peers? We hypothesize that subsidiaries of European\n\nbanks might exploit the regulatory leniency characterizing certain emerging markets to offload\n\ntheir domestic compliance costs to foreign consumers. If this is the case, we should observe that\n\nthe increased interest income documented in Table 6 is concentrated in banking systems featuring\n\nweaker regulatory environments. We find this to be the case. Using data from the World Bank’s\n\nBank Regulation and Supervision Survey, we identify countries featuring stricter overall\n\nrestrictions on banking activities and providing stricter regulatory decisions for NPLs, we\n\ndocument that this regulatory cost-shifting mechanism is significantly stronger in countries with a\n\nweaker regulatory environment (Table 7). This result is consistent with multinational banks\n\nexploiting the regulatory leniency of foreign countries to off-load – at least part of – the regulatory\n\ncosts imposed by domestic regulations.\n\n\n**6** **Conclusion**\n\n\nThe European calendar provisioning represents not only a major regulatory and supervisory\n\nchange, but also a turning point in banking regulation. Its underpinning idea is disruptive to the\n\nlogic of the regulatory framework established by the Basel Committee on Banking Supervision.\n\nRather than relying on banks’ internal risk-based models, the European calendar provisioning\n\nimposes an old-fashioned regulation that hinges on minimum loss coverage requirements based on\n\nsimple rules depending only on the time _elapsed_ since the default. Furthermore, the process of\n\nadoption and enforcement of this regulation is unique: originally introduced in the form of\n\nsupervisory guidance in 2017 by the ECB, it became stricter in 2018, and it turned into a new Pillar\n\n1 regulation in 2019.\n\n\nThe regulatory process followed for the introduction of this new regulation enables us to\n\nassess for the first time how binding supervisory guidances issued under the comply or explain\n\nframework are as compared to the introduction of a similar regulatory intervention. Our empirical\n\nstrategy relies on a Difference-in-Differences (DID\n\nMention: \"Bank Regulation and Supervision Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Countries are ranked on restrictions on banking activity and non-performing loan rules to test regulatory cost-shifting.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000731:3:1", "occurrence_id": "756dac36f7c6869148429033", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  actual data from a census\nof households in a set of rural Mexican communities, Demombynes and Hoogeveen\n(2007) show that correlations between the estimated and true welfare at the local level\nare highest for mean expenditure and poverty measures and lower for inequality measures. Douidich et al. (2016) obtain accurate estimates for quarterly poverty rates using\na classic survey-to-survey imputation method that combines HBSs and Labor Force\nSurveys (LFSs). Using a log-linear regression to impute a completely missing variable\n(total expenditure) for certain years, they obtain reliable estimates for Head Count Ratio\n(HCR) with an exogenous poverty line. Krafft et al. (2019) impute consumption from\nHBSs onto LFSs and obtain information on poverty and inequality. Resulting measures\nof consumption, poverty, and inequality are similar across survey pairs, particularly for\nthe data in Jordan and the Arab Republic of Egypt. These estimates are obtained based\non the hypothesis that the model is time invariant, that is, the coefficients used to impute\nfrom one survey to another are stable over time (e.g., that _βt_ = _β_ ), and that the error term\nis homoscedastic and normally distributed. However, the concern is that incorrectly assuming normal errors or ignoring heteroscedasticity has the potential to introduce bias\nwhen estimating poverty or inequality. Specifically, in the prediction of inequality, the\nbias can be significant. Dang et al. (2019) provide an alternative method of simulating\nthe final consumption imputation, which also seems extremely accurate when estimat\n\n2\n\nMention: \"Labor Force\nSurveys\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Combined with budget surveys, these enable survey-to-survey imputation of quarterly poverty rates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000733:17:0", "occurrence_id": "a4e85d7e84f529dfea1742eb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <mark>Third, the GSPS indicators can enable scholars to understand change over time in</mark>\n\n\n<mark>recruitment practices in governments, by assessing how recruitment practices interact with</mark>\n\n\n<mark>years of service of public servants.</mark> <sup>2</sup> <mark>To illustrate, Figure 5 plots the share of public servants</mark>\n\n\n<mark>indicating that they found out about their public sector job through a public advertisement in</mark>\n\n\n<mark>several countries (Chile, Croatia, Romania, Estonia), by years of service of public servants. It</mark>\n\n\n<mark>shows that in a range of countries, the share of public servants finding out about their first</mark>\n\n\n<mark>public sector job through an advertisement (e.g., newspaper, government website) has</mark>\n\n\n<mark>increased sharply. For instance, in Chile, only 18% of public servants recruited more than 20</mark>\n\n\n<mark>years ago learned about their first public sector job through an advertisement, relative to 44%</mark>\n\n\n<mark>of public servants recruited in the last decade. With jobs being publicly advertised rather than</mark>\n\n\n<mark>only disseminated through word of mouth, transparency in public sector recruitment has</mark>\n\n\n<mark>arguably increased in these countries, at least in terms of vacancy dissemination. Our data can</mark>\n\n\n<mark>thus also help assess changes in civil service management and reforms where applicable.</mark>\n\n\n**<mark>Figure 5. Recruitment through Job Advertisements in Select Countries, by Years of</mark>**\n\n\n**<mark>Service of Public Servants</mark>**\n\n\n<mark>(% of public servants indicating they found out about their first public sector job vacancy</mark\n\nMention: \"GSPS indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"inform\", \"summary\": \"These indicators are proposed as a way to study how recruitment practices change with years of service.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000734:7:0", "occurrence_id": "5c7e7170b8ee4af613b3e079", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: households in CPHS over 23 months from January 2019 to November 2020. These mean\n\n\nincomes were then used to divide the income distribution into ten “objective” income deciles,\n\n\nand each household was categorized in one of these deciles. The mean and standard deviation\n\n\nof monthly income corresponding to each decile, and the number of households interviewed\n\n\nby decile, is given in supplementary Table S7 and Figure S9.\n\n\nIn the June-August 2021 wave of the CPHS, an additional module on inequality was\n\n\nadministered to the panel. In this round of the CPHS, interviews were successfully completed\n\n\nwith 116,061 households, which comprise the sample for our analysis. <sup>5</sup> [^5: Since interviews were not completed with a significant fraction of the CPHS sample, this sample on which our\nanalysis is based is not nationally representative. More generally, the representativeness of the CPHS sample\nhas been debated by scholars and analysts—a limitation that we acknowledge.] Half of these\n\n\nhouseholds were randomly assigned to the treatment group, and the other half to a control\n\n\ngroup. This randomization was stratified by state and whether the household was in a rural\n\n\narea. We compare the treatment and control groups on a set of observable characteristics and\n\n\nfind no discernible difference between the groups (see supplementary table S4).\n\n\nThe module on inequality began by eliciting respondents’ beliefs about their position\n\n\non the income distribution, _“Suppose we divide the households of India into 10 equal parts,_\n\n\n_where the poorest households are in the first part and the richest households are in the last_\n\n\n_part. See this bar_ [respondents were shown a long rectangle composed of 10 equally sized\n\n\nblocks] _. The first box contains the poorest households. The second box contains the next_\n\n\n_poorest households. The last box contains the richest households. Which box do you think your_\n\n\n_household would fall into?”._ Following this question, respondents in the treatment group were\n\n\ninformed if their estimates coincided with those of the research team. Respondents were told,\n\nMention: \"CPHS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Mean household incomes over 23 months define objective income deciles for classification.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000737:4:0", "occurrence_id": "ec1ed9d357ba1377043b2b10", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 1. Introduction\n\nThe Internal displacement monitoring center estimated that as of the end of 2020, about 55 million people\n\nglobally were Internally displaced persons (IDPs), and 87 percent of them were fleeing from conflict and\n\nviolence. <sup>2</sup> [^2: IDMC 2020 displacement data. See link] During conflicts, IDPs, unlike refugees, take refuge within their countries but away from their\n\nhomes. In theory, they are expected to be protected and assisted by their government, irrespective of the\n\ncause of the conflicts or the IDPs' affiliation, whether with the state or the insurrection. In reality, IDPs are\n\namong the most vulnerable populations in any country (World Bank, 2017). Besides the psychological\n\ntrauma of crises leading to their displacement and the loss of their livelihoods and social capital, IDPs\n\nexperience disproportional social and economic exclusions, struggling to access basic services and\n\nassistance. The vulnerability of IDPs is exacerbated by the fact that they may be undocumented and live\n\nclose to conflict areas and, in some cases, with hosts with whom they do not share similar political, social,\n\nor cultural alignments. Therefore, there is the possibility that some IDPs may voluntarily opt-out from\n\ngovernment social protection due to fear of persecution or lack of documentation.\n\n\nThe need to protect the right of IDPs and provide the assistance they need has been championed by\n\nhumanitarian agencies, including the United Nations High Commissioner for Refugees (UNHCR), World\n\nFood Programme (WFP), and other international Non-Governmental Organizations (NGOs) who assist\n\nIDPs through cash transfer programs. For many IDPs, the cash transfers delivered by these agencies are\n\nthe only source or make up the most significant share of their household income. However, unlike the\n\ngovernment social safety net (SSN), the humanitarian agencies’ cash transfers are usually transient, and\n\nmost agencies are faced with fiscal constraints and shorter service terms. In a protracted crisis, IDPs may\n\nquickly fall back into extreme\n\nMention: \"IDMC 2020 displacement data\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"Global internal displacement totals and the conflict share are quoted from this monitoring source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000740:17:0", "occurrence_id": "999629f1cb0388ae16d08f01", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: redrew the boundaries for those two places, and the segmentation was done again following the process\npreviously described.\n\n\nThe last challenge with using this segmentation method was that buildings demarcated on OpenStreetMaps do not\ntypically correlate directly with population figures due to factors such as the lack of information on building heights\nand data being collected through public reporting. Subsequently, a ground-truthing component was designed in\norder to validate the mauza segmentation. The ground validation exercises yielded several key findings, notably:\n\n\n  - The segmentation of EAs was relatively less accurate in densely populated urban regions as opposed to\nrural mauzas with scattered settlements.\n\n  - The number of households found within these segments varied, with some segments having more\nhouseholds than expected and some having fewer than expected.\n\n\nField protocols were designed in order to address these concerns. The protocols involved combining GPS\ncoordinates and also training enumerators in the use of Google Maps to allow for systematic field-based\nadjustment.\n\n\n_Drawing the host sample_\n\n\nOnce the strata were defined, the first stage of selection was done at the mauza level by strata. The selection\nprocess used probability proportional to the population size of the mauza. As Table 4 shows, the sample for\nhost communities was divided equally between the two strata, so 33 mauzas were selected in each stratum.\nWithin each selected mauza, three segments were designated with equal probability of selection. Mauzas\nwere allowed to be selected more than once. <sup>34</sup> [^34: Since the selection was done using probability proportional to the population size of the mauza with replacement, bigger mauzas\nhad higher probability of being selected more than once.] If this occurred, then the number of selected segments within\nthe mauza increased accordingly. For example, if a mauza was selected twice in the sample, six instead of\nthree segments would be selected from the mauza, each segment containing 100-150 households.\n\n\nThe second-stage selection was made using the enumeration areas created for the selected mauzas. All the\nsegments within a mauza had the same probability of being selected. The field team listed\n\nMention: \"OpenStreetMaps\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Crowdsourced building footprints do not align with population because heights are missing and reporting is voluntary.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000755:3:1", "occurrence_id": "ce59adb54fb4539d2642f8e3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  time, theoretical models have\n\n\nemphasized a different type of heterogeneity, namely that the reduction of trade costs affects\n\n\ndifferent types of firms differently. Larger and more productive firms can more easily take\n\n\nadvantage of the change in trade costs than smaller and less productive firms (Melitz, 2003;\n\n\nMelitz and Redding, 2015). As deep trade agreements can reduce trade costs among members\n\n\nby eliminating tariffs and by reducing other frictions due to regulatory differences and/or policy\n\n\nuncertainty, a natural extension of this literature is that DTAs should affect firms differently\n\n\nbased on their characteristics, promoting the export performance of larger firms and leading to\n\n\na selection effect.\n\n\nThis paper empirically investigates the effect of RTAs’ depth on the export performance of\n\n\nheterogeneous firms. The analysis combines firm-level export data for 31 developing countries\n\n\nfrom the World Bank Exporter Dynamics Database (Fernandes et al., 2016) and information\n\n\non the detailed content of more than 300 RTAs from the World Bank Deep Trade Agreements\n\n\ndatabase (Hofmann et al., 2017). The richness of the data allows to precisely take into account\n\n\nthe heterogeneous consequences of deep RTAs on the exports of different firms. We use a\n\n\ndecomposition à la Berman et al. (2012) to study the extent to which the effects of deep RTAs\n\n\n1\n\nMention: \"World Bank Exporter Dynamics Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Firm-level export data for 31 developing countries are drawn from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000755:3:1", "occurrence_id": "a1a4663aabe7cd44f7557ef3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  time, theoretical models have\n\n\nemphasized a different type of heterogeneity, namely that the reduction of trade costs affects\n\n\ndifferent types of firms differently. Larger and more productive firms can more easily take\n\n\nadvantage of the change in trade costs than smaller and less productive firms (Melitz, 2003;\n\n\nMelitz and Redding, 2015). As deep trade agreements can reduce trade costs among members\n\n\nby eliminating tariffs and by reducing other frictions due to regulatory differences and/or policy\n\n\nuncertainty, a natural extension of this literature is that DTAs should affect firms differently\n\n\nbased on their characteristics, promoting the export performance of larger firms and leading to\n\n\na selection effect.\n\n\nThis paper empirically investigates the effect of RTAs’ depth on the export performance of\n\n\nheterogeneous firms. The analysis combines firm-level export data for 31 developing countries\n\n\nfrom the World Bank Exporter Dynamics Database (Fernandes et al., 2016) and information\n\n\non the detailed content of more than 300 RTAs from the World Bank Deep Trade Agreements\n\n\ndatabase (Hofmann et al., 2017). The richness of the data allows to precisely take into account\n\n\nthe heterogeneous consequences of deep RTAs on the exports of different firms. We use a\n\n\ndecomposition à la Berman et al. (2012) to study the extent to which the effects of deep RTAs\n\n\n1\n\nMention: \"World Bank Deep Trade Agreements\n\n\ndatabase\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Detailed content of more than 300 regional trade agreements is coded in this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000757:13:1", "occurrence_id": "d8a81dc677cd1a89b919148e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: . <sup>21</sup> [^21: This assumption is less realistic for Emissions Intensive Trade Exposed (EITE) industries competing in world markets\nwith limited ability to pass cost increases forward to final consumers.] Non-fuel sector price increases are obtained\nas the sum-product of: i) each sector’s fossil fuel intensity; and ii) the price increase of each fossil fuel\ninduced by the climate mitigation policy. Sectoral fossil fuel intensities are obtained from direct\nrequirements matrices of input-output tables. These tables are sourced from the GTAP-10 database, <sup>22</sup> [^22: See: https://www.gtap.agecon.purdue.edu/databases/v10/index.aspx. Given that the GTAP-10 input-output tables are\nrelatively outdated, they are scaled such that their implicit revenue flows from a hypothetical climate mitigation policy\nmatch those of CPAT (which are primarily based on IEA (2021) data).]\nwhich includes 2014 data for 65 sectors <sup>23</sup> [^23: These cover the following five fossil fuels: coal (‘coa’), electricity (‘ely’), oil (‘oil’), natural gas (‘gas’, ‘gdt’) and petroleum\nproducts (‘p_c’).] that are, in turn, mapped to the CPAT non-fuel consumption\ngood/service categories mentioned above to recompute equation (A). Summing the estimates across\nall non-fuel goods and services yields a measure of the loss in consumption from price increases of\nnon-fossil fuel products (for example, food, clothing, housing, etc.) following the introduction of a\ncarbon tax (that is, the ‘indirect’ incidence effect). Adding up the direct and indirect effects yields an\nestimate of the total incidence effect. We further adjust the total incidence effect downwards to\n\n\n[15 See: http://data.bbs.gov.bd/index.php/catalog/182](\n\nMention: \"GTAP-10 database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Input-output tables across 65 sectors supply the fossil fuel intensity coefficients.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000757:13:1", "occurrence_id": "43d28c2a305d9091096d4057", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: . <sup>21</sup> [^21: This assumption is less realistic for Emissions Intensive Trade Exposed (EITE) industries competing in world markets\nwith limited ability to pass cost increases forward to final consumers.] Non-fuel sector price increases are obtained\nas the sum-product of: i) each sector’s fossil fuel intensity; and ii) the price increase of each fossil fuel\ninduced by the climate mitigation policy. Sectoral fossil fuel intensities are obtained from direct\nrequirements matrices of input-output tables. These tables are sourced from the GTAP-10 database, <sup>22</sup> [^22: See: https://www.gtap.agecon.purdue.edu/databases/v10/index.aspx. Given that the GTAP-10 input-output tables are\nrelatively outdated, they are scaled such that their implicit revenue flows from a hypothetical climate mitigation policy\nmatch those of CPAT (which are primarily based on IEA (2021) data).]\nwhich includes 2014 data for 65 sectors <sup>23</sup> [^23: These cover the following five fossil fuels: coal (‘coa’), electricity (‘ely’), oil (‘oil’), natural gas (‘gas’, ‘gdt’) and petroleum\nproducts (‘p_c’).] that are, in turn, mapped to the CPAT non-fuel consumption\ngood/service categories mentioned above to recompute equation (A). Summing the estimates across\nall non-fuel goods and services yields a measure of the loss in consumption from price increases of\nnon-fossil fuel products (for example, food, clothing, housing, etc.) following the introduction of a\ncarbon tax (that is, the ‘indirect’ incidence effect). Adding up the direct and indirect effects yields an\nestimate of the total incidence effect. We further adjust the total incidence effect downwards to\n\n\n[15 See: http://data.bbs.gov.bd/index.php/catalog/182](\n\nMention: \"GTAP-10 input-output tables\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Because the tables are outdated they are rescaled so implied mitigation revenues match another source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000757:13:1", "occurrence_id": "10ca1ac82f8199576fea6e58", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: . <sup>21</sup> [^21: This assumption is less realistic for Emissions Intensive Trade Exposed (EITE) industries competing in world markets\nwith limited ability to pass cost increases forward to final consumers.] Non-fuel sector price increases are obtained\nas the sum-product of: i) each sector’s fossil fuel intensity; and ii) the price increase of each fossil fuel\ninduced by the climate mitigation policy. Sectoral fossil fuel intensities are obtained from direct\nrequirements matrices of input-output tables. These tables are sourced from the GTAP-10 database, <sup>22</sup> [^22: See: https://www.gtap.agecon.purdue.edu/databases/v10/index.aspx. Given that the GTAP-10 input-output tables are\nrelatively outdated, they are scaled such that their implicit revenue flows from a hypothetical climate mitigation policy\nmatch those of CPAT (which are primarily based on IEA (2021) data).]\nwhich includes 2014 data for 65 sectors <sup>23</sup> [^23: These cover the following five fossil fuels: coal (‘coa’), electricity (‘ely’), oil (‘oil’), natural gas (‘gas’, ‘gdt’) and petroleum\nproducts (‘p_c’).] that are, in turn, mapped to the CPAT non-fuel consumption\ngood/service categories mentioned above to recompute equation (A). Summing the estimates across\nall non-fuel goods and services yields a measure of the loss in consumption from price increases of\nnon-fossil fuel products (for example, food, clothing, housing, etc.) following the introduction of a\ncarbon tax (that is, the ‘indirect’ incidence effect). Adding up the direct and indirect effects yields an\nestimate of the total incidence effect. We further adjust the total incidence effect downwards to\n\n\n[15 See: http://data.bbs.gov.bd/index.php/catalog/182](\n\nMention: \"IEA (2021) data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"These energy data underlie the revenue benchmarks used to rescale the input-output tables.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000761:45:0", "occurrence_id": "2d30a3fb943eca71a21a86e9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 64. <sup>15</sup> [^15: Drawing on ILO data and definitions (ILO 2022b), the extended unemployment rate (in ILO documents referred\nto as “the combined rate of unemployment and potential labor force (LU3)” is defined as 100(unemployed +\npotential labor force) / (labor force + potential labor force), where the denominator according to ILO terminology\nis referred to as the extended labor force. The potential labor force is defined as “persons not in employment who\nexpress an interest in this form of work but for whom existing conditions limit their active job search and/or their\navailability”.] Among these data inputs, population 15-64 is from UN (2019) while unemployment,\nemployment and potential labor force is from ILO (2022). For the mining natural resource\nfactor, stock extraction is assumed virtually stagnant.\n\n\nThe post calculations that generate poverty results by simulation are based on the 2021\nheadcount poverty rate estimated by the World Bank, the assumption of a log-normal\nconsumption distribution, and the 2014 Gini coefficient. Household consumption per capita is\nused as the welfare measure. We considered the 2021 figure as valid also for 2022.\n\n\n**<u>Table A.4. Value-added, trade, and consumption elasticities</u>**\n\n**<u>VA</u>** **<u>CET</u>** **<u>Armington</u>** **<u>LES</u>**\n**Qat** 0.25 2.00 2.00 0.83\n**Other agriculture** 0.25 2.00 2.00 0.58\n**Fishing** 0.20 2.00 2.00 0.58\n**Oil and gas** 0.20 2.00 2.00 0.90\n**Other mining** 0.20 2.00 2.00\n\nMention: \"ILO data and definitions\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"The extended unemployment rate is defined following this agency's combined unemployment and potential labour force measure.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000764:19:0", "occurrence_id": "d4fe4f7a63bbe07bb6b6fb9a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **A.C** **First** **Stage** **Estimates**\n\n\nIn this paper, we leverage a new dataset on compulsory schooling laws in Europe from the World\n\n\nBank, which is one of the largest datatabases on CSLs to date. Figure A1 shows the number of\n\n\ncompulsory schooling law reforms by country.\n\n\nFigure A1: **Number** **of** **compulsory** **schooling** **Law** **(CSL)** **reforms** **by** **country** . The map shows all\nCSL changes that can be mapped to the ESS data. Note that a British reform commonly used in literature\nis excluded from our analysis, because this reform is region-specific and the ESS data does not have enough\ngeographic granularity to accurately assign regional laws to respondent’s individual level climate outcomes.\n\n\nFigure A2 shows all first stages with positive effects, in addition to the first stages with both\n\n\npositive and highly statistically significant effects included in the main analysis in the paper. Ex\n\nact first stage estimates are included in the table below. We provide these estimates to give a\n\n\ncomprehensive picture of where CSLs bind and have large effects versus where effects are smaller.\n\n\nFigure A2: **Positive** **CSL** **changes** **by** **country** . This figure shows the number of compulsory\nschooling law changes with positive (but not necessarily significant) first stage estimates by country.\n\n\n18\n\nMention: \"dataset on compulsory schooling laws in Europe\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Reforms in compulsory schooling laws across countries are catalogued in this dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000764:19:0", "occurrence_id": "534809f3a3839fde420d05ad", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **A.C** **First** **Stage** **Estimates**\n\n\nIn this paper, we leverage a new dataset on compulsory schooling laws in Europe from the World\n\n\nBank, which is one of the largest datatabases on CSLs to date. Figure A1 shows the number of\n\n\ncompulsory schooling law reforms by country.\n\n\nFigure A1: **Number** **of** **compulsory** **schooling** **Law** **(CSL)** **reforms** **by** **country** . The map shows all\nCSL changes that can be mapped to the ESS data. Note that a British reform commonly used in literature\nis excluded from our analysis, because this reform is region-specific and the ESS data does not have enough\ngeographic granularity to accurately assign regional laws to respondent’s individual level climate outcomes.\n\n\nFigure A2 shows all first stages with positive effects, in addition to the first stages with both\n\n\npositive and highly statistically significant effects included in the main analysis in the paper. Ex\n\nact first stage estimates are included in the table below. We provide these estimates to give a\n\n\ncomprehensive picture of where CSLs bind and have large effects versus where effects are smaller.\n\n\nFigure A2: **Positive** **CSL** **changes** **by** **country** . This figure shows the number of compulsory\nschooling law changes with positive (but not necessarily significant) first stage estimates by country.\n\n\n18\n\nMention: \"ESS data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Its lack of geographic granularity means a region-specific British reform must be excluded.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000767:43:0", "occurrence_id": "b8ff8874b3eb1f8b95be18d8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **A.1** **Example of data extraction and analysis**\n\n\nIn this section we exemplify how researchers can use our dataset for generating descriptive statistics about public procurement and budget execution across Brazilian municipalities.\nThe full documentation on using datasets hosted on the Data Basis platform can be found at\nhttps://basedosdados.github.io/mais (in Portuguese), including video tutorials and details on\nhow to create projects at Google BigQuery.\n\nHere we provide a simple example using the ‘basedosdados‘ R package. This package\nallows users to connect to their BigQuery projects (and billing ID) and directly make SQL\nqueries on BigQuery datasets, which are then loaded in R.\n\nThe SQL query presented below does the following: it connects to the _licitacao_ (tender)\ntable including tender-level information for all municipalities and years covered in our data\n(over 2.3 million unique tenders); it creates an indicator for when the modality is defined as\ncategories 8 (waiver) or 10 (non-requirement); then it computes the average share of tenders\nthat are non-competitive and the total number of tenders at the municipality-year level. The\nresulting table, loaded in the ‘data‘ object, includes 19,033 municipality-year observations.\nWhen executed, this query took 3 seconds to process the 2.3 million observations and provide\nus these summarized statistics.\n\nWe then can simply plot a histogram of the share of non-competitive tenders across municipality - year (in this case, we filter the dataset for observations with at least 50 tenders).\nWe present the results in Figure A.7. We can see that the average we describe in Table 2 of\n30%-40% masks substantial heterogeneity: while a substantial amount of municipalities in a\ngiven year do not use non-competitive modalities, in some municipalities the share of these\nis above three-quarters.\n\nMention: \"basedosdados\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"This package connects users to cloud datasets so they can query them directly from the analysis environment.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000772:28:1", "occurrence_id": "6beef0e6a78f2f8695bedbfc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  activities that produce a homogeneous electricity product:\ncoal, gas, nuclear, hydropower, wind, solar and other. The single electricity sector in the original I-O\ntable is divided based on the GTAP electricity database with ad-hoc adjustments for the Turkish\nelectricity balance tables. Nuclear power is introduced with a small share in electricity supply based\non the average cost structure of nuclear power activities in the GTAP power database.\n\n#### 2 MFMOD additional feature\n\n\nThe World Bank’s macrostructural model, MFMod is used in quantifying the economy-wide impacts\nof the energy efficient infrastructure investment. The customized standalone model for Türkiye\nextends the investment channels along the several dimensions. (1) Capital stock is differentiated by\npublic and private contributions and (2) capital stock is differentiated between infrastructure and\nnon-infrastructure capital, <sup>17</sup> and (3) the time horizon is extended to 2040.\n\n\nIn the model, investment decisions are based on the difference between returns and the cost of\ncapital. Specifically, the dynamic investment equation is a function of past investment decisions (to\nreflect the “sticky” nature of asset decisions) and the marginal Tobin’s Q ratio, which reflects the\nreturn vs. the cost of capital. In the long-run, the investment to capital ratio is a function of long-run\neconomic growth plus the rate of capital depreciation. The private sector infrastructure equation is\nwritten as:\n\n\n17 In (Hallegatte, Jooste, & McIsaac, 2022), they use a modified version of the model to explore the vulnerability to\nfloods and earthquakes. <u><mark>[http://hdl.handle.net/10986/37060](http://hdl.handle.net/10986/37060)</mark></u>\n\n\n27\n\nMention: \"Turkish\nelectricity balance tables\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"The single electricity sector in the input-output table is split using these national balance tables.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000772:28:1", "occurrence_id": "deffca3d066e6908eb9ff7c1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  activities that produce a homogeneous electricity product:\ncoal, gas, nuclear, hydropower, wind, solar and other. The single electricity sector in the original I-O\ntable is divided based on the GTAP electricity database with ad-hoc adjustments for the Turkish\nelectricity balance tables. Nuclear power is introduced with a small share in electricity supply based\non the average cost structure of nuclear power activities in the GTAP power database.\n\n#### 2 MFMOD additional feature\n\n\nThe World Bank’s macrostructural model, MFMod is used in quantifying the economy-wide impacts\nof the energy efficient infrastructure investment. The customized standalone model for Türkiye\nextends the investment channels along the several dimensions. (1) Capital stock is differentiated by\npublic and private contributions and (2) capital stock is differentiated between infrastructure and\nnon-infrastructure capital, <sup>17</sup> and (3) the time horizon is extended to 2040.\n\n\nIn the model, investment decisions are based on the difference between returns and the cost of\ncapital. Specifically, the dynamic investment equation is a function of past investment decisions (to\nreflect the “sticky” nature of asset decisions) and the marginal Tobin’s Q ratio, which reflects the\nreturn vs. the cost of capital. In the long-run, the investment to capital ratio is a function of long-run\neconomic growth plus the rate of capital depreciation. The private sector infrastructure equation is\nwritten as:\n\n\n17 In (Hallegatte, Jooste, & McIsaac, 2022), they use a modified version of the model to explore the vulnerability to\nfloods and earthquakes. <u><mark>[http://hdl.handle.net/10986/37060](http://hdl.handle.net/10986/37060)</mark></u>\n\n\n27\n\nMention: \"GTAP power database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Average cost structures for nuclear generation are taken from this power database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000775:36:0", "occurrence_id": "2598fe15eff7fb16ae9f401e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ### Appendix A – Detailed Data Description\n\n###### Street Networks\n\nData for the street networks for pedestrians for both Rio de Janeiro and Kinshasa were extracted from\nOpenStreetMap (OSM). A global collaborative community of mappers who all contribute and maintain\npublicly available data related to roads, points of interest and administrative boundaries (OpenStreetMap,\n2022). Researchers estimate that approximately 80% of the street grid is accounted for on OSM and it is\nwidely accepted as a reliable source of road network data (Barrington-Leigh & Millard-Ball, 2017). A base\nwalking speed of 3.6km/h was used across the road network. Digital Elevation Models (DEM) for both\ncities were also incorporated into the street networks to account for the impact that slopes might have\non walking speed with these being adjusted through Conveyal’s R5 engine using Tobler’s hiking function. <sup>17</sup> [^17: Details of which are described on the Conveyal website and github repository. Available at\n<u>[https://docs.conveyal.com/learn-more/generalized-cost and](https://docs.conveyal.com/learn-more/generalized-cost)</u>\nhttps://github.com/conveyal/r5/blob/43068e168138bc5636d895861209c1c0049e0f1a/src/main/java/com/convey\nal/r5/rastercost/ToblerCalculator.java#L32]\n\n###### Transport Networks\n\nTransit networks in both cities are described in the General Transit Feed Specification (GTFS) format. GTFS\nis a standardized framework for transit service schedules and geographic information. It was created with\nthe intention of standardizing the way public transit data is published and used (particularly for software\napplications) globally and utilizes a series of relational .txt files to efficiently store this data (Google, 2021).\nIt includes data related to the stops, schedule (frequency-based or specific schedules), and routing related\nto different services within a transit system and can often include other supplementary\n\nMention: \"OpenStreetMap\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Pedestrian street networks for both cities are extracted from this volunteer mapping project.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000780:44:2", "occurrence_id": "88c9201bd17ed770cab3f76c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: > We then use the model to predict welfare within the HFPS 𝑦⁡̂ = 𝛽̂2𝑋2𝐻𝐹𝑃𝑆for urban\n\nhouseholds. <sup>40</sup> [^40: For the purposes of the prediction, all HFPS respondents are assumed to own a mobile phone.] This predicted welfare level for urban households is divided into quintiles to distinguish\nbetween poorer and wealthier urban households. Table B2 below reports that the R <sup>2</sup> values for the full\ncountry sample vary from 0.36 in Ethiopia and Gabon to 0.56 in Sierra Leone. Table B3 shows the kernel\ndensity plots of the measured welfare data in the GMD, the predicted welfare in the GMD, and the predicted\nwelfare in the HFPS. In general, the distribution of predicted welfare looks quite similar in the GMD and\nthe HFPS, but it is distinct from the measured consumption data in the GMD. This highlights that the\npoverty measure used is a longer-term measure of welfare that does not necessarily reflect transient shocks.\n\n\n\n_Table B2: Adjusted R-squared from imputation models_\n\n\n<u><mark>Country</mark></u> <u><mark>Adj. R-squared</mark></u>\n\n<mark>Burkina Faso</mark> <mark>0.54</mark>\n\n<mark>Ethiopia</mark> <mark>0.36</mark>\n\n<mark>Gabon</mark> <mark>0.36</mark>\n\n<mark>Ghana</mark> <mark>0.43</mark>\n\n<mark>Gambia, The</mark> <mark>0.43</mark>\n\n<mark>Mali</mark> <mark>0.43</mark>\n\n<mark>Mozambique</mark> <mark>0.43</mark>\n\n<mark>Malawi</mark\n\nMention: \"HFPS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Welfare is predicted for urban respondents and divided into quintiles.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000780:44:2", "occurrence_id": "337eae747b53f8f01669490a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: > We then use the model to predict welfare within the HFPS 𝑦⁡̂ = 𝛽̂2𝑋2𝐻𝐹𝑃𝑆for urban\n\nhouseholds. <sup>40</sup> [^40: For the purposes of the prediction, all HFPS respondents are assumed to own a mobile phone.] This predicted welfare level for urban households is divided into quintiles to distinguish\nbetween poorer and wealthier urban households. Table B2 below reports that the R <sup>2</sup> values for the full\ncountry sample vary from 0.36 in Ethiopia and Gabon to 0.56 in Sierra Leone. Table B3 shows the kernel\ndensity plots of the measured welfare data in the GMD, the predicted welfare in the GMD, and the predicted\nwelfare in the HFPS. In general, the distribution of predicted welfare looks quite similar in the GMD and\nthe HFPS, but it is distinct from the measured consumption data in the GMD. This highlights that the\npoverty measure used is a longer-term measure of welfare that does not necessarily reflect transient shocks.\n\n\n\n_Table B2: Adjusted R-squared from imputation models_\n\n\n<u><mark>Country</mark></u> <u><mark>Adj. R-squared</mark></u>\n\n<mark>Burkina Faso</mark> <mark>0.54</mark>\n\n<mark>Ethiopia</mark> <mark>0.36</mark>\n\n<mark>Gabon</mark> <mark>0.36</mark>\n\n<mark>Ghana</mark> <mark>0.43</mark>\n\n<mark>Gambia, The</mark> <mark>0.43</mark>\n\n<mark>Mali</mark> <mark>0.43</mark>\n\n<mark>Mozambique</mark> <mark>0.43</mark>\n\n<mark>Malawi</mark\n\nMention: \"measured welfare data in the GMD\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Observed consumption distributions are compared with predicted welfare to assess the imputation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000785:42:0", "occurrence_id": "ed75eb92d065f96e72de3480", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Finally, we use the MPD data to obtain the percentage of electoral votes for each of the\n\n\npolitical parties identified above to obtain the right vote share in country _j_ at time _t_ . We are\n\n\nleft with a variable that we call ’Right vote’ as it broadly reflects the share of right vote in\n\n\ncountry _i_ at time _t_, where right vote is defined by our three main topic of interests described\n\n\nabove.\n\n\n**Table** **B1:** Principal component analysis on party preferences\n\n\n**Component** **Eigenvalue** **Difference** **Proportion** **Cumulative**\n\n\n1st Component 1.220 0.294 0.406 0.406\n2nd Component 0.925 0.070 0.308 0.715\n3rd Component 0.855 0.285 1.000\n\n\n**Table** **B2:** Principal components (eigenvectors)\n\n\n**Variable** **1st** **Component** **2nd** **Component** **3rd** **Component**\n\n\nInternationalism 0.523 0.802 0.289\nNational way of life -0.628 0.133 0.767\nMulticulturalism 0.576 -0.582 0.573\n\n##### **C De Chaisemartin & D’Haultfeuille estimator: Controlling for** **heterogeneous treatment timing and dynamic effects.**\n\n\nIn a difference-in-difference setting, the treatment may not always be implemented at the same\n\n\ntime for every unit. The so-called staggered design leads therefore to a bad comparison problem\n\n\nwhere the classical DiD estimator will compare the outcome evolution of the groups that switch\n\n\n(becomes treated from t-1 to t) to groups that can be already treated. In addition and\n\nMention: \"MPD data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Party-level vote shares are aggregated from this party dataset into a right-wing vote measure.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000786:28:0", "occurrence_id": "89c7f52a849f0a3b00bb3d25", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure** **14:** Effect of Domestic and External Lockdowns on Exports in Indonesia\n\n\na. Domestic Lockdown b. External Lockdowns\n\n\nNote: Blue spike means a coefficient is statistically significant\n\nSource: Authors’ compilation using BPS data\n\n\nsignificant on quantities of air exports in its first 12 months, but significantly reduced them\n\n\nby 129% and 65% in the thirteenth and fifteenth months, respectively, and later improved\n\n\nthem by 185% in the sixteenth month. This may signal large variation in prices as the effects\n\n\nare initially seen in values but not quantities. We test this more formally in the next section.\n\n\nExternal lockdowns reduced air export values immediately they were imposed at a\n\n\nmonthly average of 9% in the first three months. The value of air exports later improved\n\n\nfrom the ninth month with a monthly average of 31%. Air export quantities declined by 7%\n\n\nand 8% in the first and second months of the external lockdown, respectively. They later\n\n\ngrew by a monthly average of 25% from the tenth month.\n\n\nIn general, domestic and external lockdowns have had an inverse effect on air exports,\n\n\nwith the impact of domestic lockdowns having a much larger magnitude. The domestic\n\n\nlockdown boosted air exports at its initial stage, while restrictions by partner countries\n\n\ndampened it. However, air exports improved at the latter stages of external lockdowns but\n\n\ndeclined from the domestic containment measures.\n\n\nWe do not find evidence that domestic lockdowns affected exports by sea cargo, which\n\n\ntransported approximately 83% of Indonesia’s exports in the May 2018-August 2021 period\n\n\n(Figure 16a). Figure 16b shows that external lockdowns had a brief disruption on sea export\n\n\nvalues. They shrunk by a monthly average of 4.6% between the fourth and sixth months\n\n\n26\n\nMention: \"BPS data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Export responses to domestic and external lockdowns are estimated on this business pulse survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000787:19:0", "occurrence_id": "13df22d3479714efdfec9673", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: domestic work is extremely low, either migrants or not (Figure 18). Furthermore, most women who\nmigrated work as salaried workers compared to those who did not migrate (71.1 against 66.0 percent). A\nlower proportion of them work as self-employed (18.6 against 22.7 percent). The same holds for men,\nalthough the differences across migration status are more significant (Figure 19).\n\n\nFigure 18. Distribution of women and men (15+) who did and did not migrate by labor force status\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nSource: Own estimates based on Population and Housing Census 2020.\n\n\nFigure 19. Distribution of employed women and men (15+) who did and did not migrate by type of\nemployment\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nSource: Own estimates based on Population and Housing Census 2020.\n\n\n**Unconditional earning gains from internal migration are higher for men than for women.** On average,\nwomen who migrated earn 17.1 percent more than those who did not migrate before controlling for other\nfactors. The unconditional gains from migration are higher for men. Men who migrated make 27.0 percent\nmore than their counterparts (Figure 20). When comparing only internal migrants who moved for\neconomic reasons versus those who stayed, the unconditional earnings gains from migration are higher\n(39.2 percent for men and 33.5 for women), and the migration earnings gains gap between men and\nwomen narrows from almost 10 percentage points to 5.7 percentage points.\n\n\n16\n\nMention: \"Population and Housing Census 2020\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Labour force status and employment type are compared between migrants and non-migrants using this census.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000789:17:0", "occurrence_id": "391e0ac7b3447f87068849dd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Note: Fitted values and 95% confidence interval shown\nSource: Social Sustainability Global Database 2022.\n\n\nSocial sustainability scatterplots also provide insights for redressing inequality. Certain aspects\nof social sustainability are associated with lower income inequality in countries, as shown in\nFigure 5. Higher levels of process legitimacy and, to a lesser extent, stronger social cohesion are\nassociated with greater income equality. Put another way, rule of law, control of corruption, a\nfair judicial system and interpersonal and institutional trust are related with increased equality.\nBy contrast, resilience and inclusion show a weaker, almost neutral, relationship with inequality.\nWhile we cannot claim causality, our results suggest that a highly targeted focus on improving\naccess to services and markets and social protection (which are all captured by inclusion and\nresilience components) is unlikely to boost equality in a marked way. This might reflect the wide\ninequalities that such policies try to address in the first place, and/or the need for long-term\ninvestment in equal opportunities in education or health before results are rendered visible. All in\nall, while more cohesive and legitimate societies are more equal, inclusive and resilient societies\ncan still coexist with high levels of income inequality.\n\n\n\n\n\n\n\n16\n\nMention: \"Social Sustainability Global Database 2022\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Process legitimacy and other social sustainability measures are related to income inequality across countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000794:16:0", "occurrence_id": "05ab3181139213759226d6ef", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 14\n\n\nFinally, for deindustrialization, we use as a proxy the decline in the GDP share of MFG\n\n\nover time. Indeed, if MFG productivity decreases relative to the world, MFG employment\n\n\nshould decrease. The GDP share combines information on productivity and employment.\n\n### **3. Resources, (De-)Industrialization, and Urbanization**\n\n\n**Data.** We study 116 relatively large countries, with a developing status as of 1960, and\n\n\nfor which data on urbanization, resource exports, and the GDP shares of manufacturing\n\n\n(MFG) and services (SERV) are available every 5 years between 1960 and 2020.\n\n\nWe obtain urban shares (%) from the United Nations (2018) and the share of fuel and\n\n\nmining exports in total exports for the period 1960-2010 from GJV16 who rely on data\n\n\nfrom World Bank (2021) and the _Mineral_ _Industry_ _Surveys_ of USGS (2020). We extend\n\n\ntheir data up to 2020, modifying them when needed. The USGS data are important\n\n\ndue to the fact that the World Bank data are incomplete or incorrect in many cases. <sup>17</sup>\n\n\nMeasurement error, if classical, would make us under-estimate the contribution of these\n\n\nresources to urbanization. The export shares of all agricultural products come from FAO\n\n\n(2020). Using these shares and knowing from World Bank (2021) the export-to-GDP ratio\n\n\nof each country in each year, we calculate NRX/GDP (%) as in Sachs and Warner (1995,\n\n\n1997). Next, we obtain the time series of the GDP share of MFG+SERV by relying on\n\n\nWorld Bank (2021), Central Intelligence Agency (2021) and United Nations (1960-1980,\n\n\n2020c) and the GDP share of FIRE c. 2020 from United Nations (2020a). <sup>18</sup>\n\n\n**\n\nMention: \"data\n\n\nfrom World Bank\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Fuel and mining export shares are partly built on this source, extended to 2020 by the authors.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000794:16:0", "occurrence_id": "6ec088b77af45945f35e9be7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 14\n\n\nFinally, for deindustrialization, we use as a proxy the decline in the GDP share of MFG\n\n\nover time. Indeed, if MFG productivity decreases relative to the world, MFG employment\n\n\nshould decrease. The GDP share combines information on productivity and employment.\n\n### **3. Resources, (De-)Industrialization, and Urbanization**\n\n\n**Data.** We study 116 relatively large countries, with a developing status as of 1960, and\n\n\nfor which data on urbanization, resource exports, and the GDP shares of manufacturing\n\n\n(MFG) and services (SERV) are available every 5 years between 1960 and 2020.\n\n\nWe obtain urban shares (%) from the United Nations (2018) and the share of fuel and\n\n\nmining exports in total exports for the period 1960-2010 from GJV16 who rely on data\n\n\nfrom World Bank (2021) and the _Mineral_ _Industry_ _Surveys_ of USGS (2020). We extend\n\n\ntheir data up to 2020, modifying them when needed. The USGS data are important\n\n\ndue to the fact that the World Bank data are incomplete or incorrect in many cases. <sup>17</sup>\n\n\nMeasurement error, if classical, would make us under-estimate the contribution of these\n\n\nresources to urbanization. The export shares of all agricultural products come from FAO\n\n\n(2020). Using these shares and knowing from World Bank (2021) the export-to-GDP ratio\n\n\nof each country in each year, we calculate NRX/GDP (%) as in Sachs and Warner (1995,\n\n\n1997). Next, we obtain the time series of the GDP share of MFG+SERV by relying on\n\n\nWorld Bank (2021), Central Intelligence Agency (2021) and United Nations (1960-1980,\n\n\n2020c) and the GDP share of FIRE c. 2020 from United Nations (2020a). <sup>18</sup>\n\n\n**\n\nMention: \"USGS data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Mineral industry surveys are used where the World Bank series is incomplete or incorrect.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000794:16:0", "occurrence_id": "08115e54953bb1ed442bbb91", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 14\n\n\nFinally, for deindustrialization, we use as a proxy the decline in the GDP share of MFG\n\n\nover time. Indeed, if MFG productivity decreases relative to the world, MFG employment\n\n\nshould decrease. The GDP share combines information on productivity and employment.\n\n### **3. Resources, (De-)Industrialization, and Urbanization**\n\n\n**Data.** We study 116 relatively large countries, with a developing status as of 1960, and\n\n\nfor which data on urbanization, resource exports, and the GDP shares of manufacturing\n\n\n(MFG) and services (SERV) are available every 5 years between 1960 and 2020.\n\n\nWe obtain urban shares (%) from the United Nations (2018) and the share of fuel and\n\n\nmining exports in total exports for the period 1960-2010 from GJV16 who rely on data\n\n\nfrom World Bank (2021) and the _Mineral_ _Industry_ _Surveys_ of USGS (2020). We extend\n\n\ntheir data up to 2020, modifying them when needed. The USGS data are important\n\n\ndue to the fact that the World Bank data are incomplete or incorrect in many cases. <sup>17</sup>\n\n\nMeasurement error, if classical, would make us under-estimate the contribution of these\n\n\nresources to urbanization. The export shares of all agricultural products come from FAO\n\n\n(2020). Using these shares and knowing from World Bank (2021) the export-to-GDP ratio\n\n\nof each country in each year, we calculate NRX/GDP (%) as in Sachs and Warner (1995,\n\n\n1997). Next, we obtain the time series of the GDP share of MFG+SERV by relying on\n\n\nWorld Bank (2021), Central Intelligence Agency (2021) and United Nations (1960-1980,\n\n\n2020c) and the GDP share of FIRE c. 2020 from United Nations (2020a). <sup>18</sup>\n\n\n**\n\nMention: \"World Bank data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its gaps and errors in many cases are the reason another source is preferred.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000798:30:3", "occurrence_id": "ba338fa245bf4c53e0e4c319", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _ (4), 1459–1483.\n\n\n**Field,** **Erica,** **Seema** **Jayachandran,** **Rohini** **Pande,** **and** **Natalia** **Rigol**, “Friendship at\n\nWork: Can Peer Effects Catalyze Female Entrepreneurship?,” _American Economic Journal:_\n_Economic_ _Policy_, May 2016, _8_ (2), 125–153.\n\n\n**Gallup**, “Americans Satisfied With Number of Friends, Closeness of Friendships,” March\n\n2004.\n\n\n**Godlonton, Susan and Rebecca Thornton**, “Peer effects in learning HIV results,” _Journal_\n\n_of_ _Development_ _Economics_, January 2012, _97_ (1), 118–129.\n\n\n**GoI**, “National Family Health Survey, 2015-16,” Technical Report, Ministry of Health and\n\nFamily Welfare, Mumbai 2017.\n\n\n28\n\nMention: \"National Family Health Survey\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"Listed in the references as a government technical report.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000799:12:0", "occurrence_id": "970c7cd7d66bc7cea2041f92", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Aggregate non-tradable production is estimated through a minimal opportunity-cost framework\nin which the annual output of an individual is proxied by the income the worker would have earned if she\nhad been allocated to the traditional sector of the labor market. The cumulative value of the non-tradable\nproduction is given by the yearly number of hours devoted to the domestic chores and housework activities\nby gender. The approach applies the average traditional-sector wage, as not doing so would transfer the\ngender wage inequities from the market to the valuated output (Chadeau, 1985). The model does not\npredict or impute the wages respective to earning-related characteristics such as the employment status or\nage-group of the different shares of the population partaking in non-tradable activities as this is deemed\nproblematic for an aggregate standpoint (Hill, 2009). Instead, the alternative that is adopted ascribes the\ntraditional-sector wage as the single and minimum remuneration with which the non-tradable production\nis valued by the economy.\n\n### **4 Parameterization**\n\n**4.1** **Data Sources**\nA wide range of data are used to calibrate the model. In following the CKW 2017 approach, the\ndemographic module is comprised of total fertility rates (TFR) and projected quinquennial population by\nfive-year age group that are gathered from the World Population Prospects of the United Nations (2010).\nThe module of two-sector production is divided between modern sector and the traditional sector. The\ndata on sectoral share of labor is taken by the Liberian Labor Force Survey (LFS) of 2010 and the sectoral\nshare of GDP are taken from the 2007-2013 estimated average of Abid (2016). The amount of arable land\nand the value rendered by the natural resources are likewise inputted into the production module and\ntaken from the World Development Indicators (WDI) database. The capital stock estimates and the Gross\n\nMention: \"World Population Prospects\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Fertility rates and projected population by age group calibrate the model's demographic module.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000799:12:0", "occurrence_id": "81fc7dd0833105ef75c14c16", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Aggregate non-tradable production is estimated through a minimal opportunity-cost framework\nin which the annual output of an individual is proxied by the income the worker would have earned if she\nhad been allocated to the traditional sector of the labor market. The cumulative value of the non-tradable\nproduction is given by the yearly number of hours devoted to the domestic chores and housework activities\nby gender. The approach applies the average traditional-sector wage, as not doing so would transfer the\ngender wage inequities from the market to the valuated output (Chadeau, 1985). The model does not\npredict or impute the wages respective to earning-related characteristics such as the employment status or\nage-group of the different shares of the population partaking in non-tradable activities as this is deemed\nproblematic for an aggregate standpoint (Hill, 2009). Instead, the alternative that is adopted ascribes the\ntraditional-sector wage as the single and minimum remuneration with which the non-tradable production\nis valued by the economy.\n\n### **4 Parameterization**\n\n**4.1** **Data Sources**\nA wide range of data are used to calibrate the model. In following the CKW 2017 approach, the\ndemographic module is comprised of total fertility rates (TFR) and projected quinquennial population by\nfive-year age group that are gathered from the World Population Prospects of the United Nations (2010).\nThe module of two-sector production is divided between modern sector and the traditional sector. The\ndata on sectoral share of labor is taken by the Liberian Labor Force Survey (LFS) of 2010 and the sectoral\nshare of GDP are taken from the 2007-2013 estimated average of Abid (2016). The amount of arable land\nand the value rendered by the natural resources are likewise inputted into the production module and\ntaken from the World Development Indicators (WDI) database. The capital stock estimates and the Gross\n\nMention: \"Liberian Labor Force Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The sectoral share of labour is taken from this survey to set the production module.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000799:12:0", "occurrence_id": "0909707d66f597e7fd697929", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Aggregate non-tradable production is estimated through a minimal opportunity-cost framework\nin which the annual output of an individual is proxied by the income the worker would have earned if she\nhad been allocated to the traditional sector of the labor market. The cumulative value of the non-tradable\nproduction is given by the yearly number of hours devoted to the domestic chores and housework activities\nby gender. The approach applies the average traditional-sector wage, as not doing so would transfer the\ngender wage inequities from the market to the valuated output (Chadeau, 1985). The model does not\npredict or impute the wages respective to earning-related characteristics such as the employment status or\nage-group of the different shares of the population partaking in non-tradable activities as this is deemed\nproblematic for an aggregate standpoint (Hill, 2009). Instead, the alternative that is adopted ascribes the\ntraditional-sector wage as the single and minimum remuneration with which the non-tradable production\nis valued by the economy.\n\n### **4 Parameterization**\n\n**4.1** **Data Sources**\nA wide range of data are used to calibrate the model. In following the CKW 2017 approach, the\ndemographic module is comprised of total fertility rates (TFR) and projected quinquennial population by\nfive-year age group that are gathered from the World Population Prospects of the United Nations (2010).\nThe module of two-sector production is divided between modern sector and the traditional sector. The\ndata on sectoral share of labor is taken by the Liberian Labor Force Survey (LFS) of 2010 and the sectoral\nshare of GDP are taken from the 2007-2013 estimated average of Abid (2016). The amount of arable land\nand the value rendered by the natural resources are likewise inputted into the production module and\ntaken from the World Development Indicators (WDI) database. The capital stock estimates and the Gross\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Arable land area and natural resource values are taken from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000800:38:1", "occurrence_id": "039f6f768ec88febb938cee8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: www.reuters.com/article/us-climate-change-un/global-temperatures-on-track-for-3-5degree-rise-by-2100-u-n-idUKKCN1NY186?edition-redirect=uk\n\nSanchez, L. and D. Stern. 2016. Drivers of industrial and non-industrial greenhouse gas\nemissions. Ecological Economics, 124:17–24.\n\nSchiavina, M., A. Moreno-Monroy, L. Maffenini and P. Veneri, Paolo. 2019. GHS-FUA\nR2019A - GHS functional urban areas, derived from GHS-UCDB R2019A (2015). European\nCommission, Joint Research Centre (JRC) [Dataset] doi:10.2905/347F0337-F2DA-4592-87B3E25975EC2C95 PID: http://data.europa.eu/89h/347f0337-f2da-4592-87b3-e25975ec2c95.\n\nSteffen, W., J. Rockström and K. Richardson. 2018. Trajectories of the Earth System in the\nAnthropocene. PNAS 115(33): 8252-8259.\n\nTurner, M. and M. Gonzalez-Navarro. 2018. Subways and urban growth: Evidence from earth.\nJournal of Urban Economics 108: 85-106.\n\nUN (United Nations). 2021. GDP and its breakdown at constant 2015 prices in US Dollars - all\ncountries for all years - sorted alphabetically. UN Department of Economic and Social Affairs,\n[Statistics Division, National Accounts. https://unstats.un.org/unsd/snaama/downloads](https://unstats.un.org/unsd/snaama\n\nMention: \"GHS-FUA\nR2019A\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"Cited as a dataset of functional urban areas in the reference list.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000800:38:1", "occurrence_id": "8894b161086973ff2482a847", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: www.reuters.com/article/us-climate-change-un/global-temperatures-on-track-for-3-5degree-rise-by-2100-u-n-idUKKCN1NY186?edition-redirect=uk\n\nSanchez, L. and D. Stern. 2016. Drivers of industrial and non-industrial greenhouse gas\nemissions. Ecological Economics, 124:17–24.\n\nSchiavina, M., A. Moreno-Monroy, L. Maffenini and P. Veneri, Paolo. 2019. GHS-FUA\nR2019A - GHS functional urban areas, derived from GHS-UCDB R2019A (2015). European\nCommission, Joint Research Centre (JRC) [Dataset] doi:10.2905/347F0337-F2DA-4592-87B3E25975EC2C95 PID: http://data.europa.eu/89h/347f0337-f2da-4592-87b3-e25975ec2c95.\n\nSteffen, W., J. Rockström and K. Richardson. 2018. Trajectories of the Earth System in the\nAnthropocene. PNAS 115(33): 8252-8259.\n\nTurner, M. and M. Gonzalez-Navarro. 2018. Subways and urban growth: Evidence from earth.\nJournal of Urban Economics 108: 85-106.\n\nUN (United Nations). 2021. GDP and its breakdown at constant 2015 prices in US Dollars - all\ncountries for all years - sorted alphabetically. UN Department of Economic and Social Affairs,\n[Statistics Division, National Accounts. https://unstats.un.org/unsd/snaama/downloads](https://unstats.un.org/unsd/snaama\n\nMention: \"GHS-UCDB R2019A\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"Named as the underlying urban centre database from which the functional areas are derived.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000803:47:1", "occurrence_id": "13e74444491fe2fbf01deb27", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 022 3729 Malawi 0.745 0.027 2178\n\nDominican\n0.838 0.093 540 Mali 0.699 0.034 1332\nRepublic\n\nGuatemala 0.722 0.019 2690 Mozambique 0.759 0.055 279\nHaiti 0.629 0.021 2144 Namibia 0.823 0.115 279\n\nPeru 0.738 0.028 2645 Niger 0.705 0.056 308\n\nNigeria 0.693 0.011 5396\n\n\nMiddle East & North Africa Rwanda 0.715 0.043 1362\n\nEgypt, Arab Rep. 0.656 0.025 2934 Senegal 0.792 0.069 354\n\nJordan 0.543 0.023 5047 Sierra Leone 0.597 0.039 1139\n\nSouth Africa 0.592 0.072 578\n\n\nSouth Asia Tanzania 0.711 0.052 1213\n\nAfghanistan 0.561 0.017 4857 Togo 0.837 0.073 410\n\nBangladesh 0.718 0.025 1712 Uganda 0.704 0.037 1710\n\nIndia 0.686 0.004 72192 Zambia 0.741 0.027 2011\n\nNepal 0.640 0.032 1317 Zimbabwe 0.721 0.054 518\n\nPakistan 0.762 0.013 5583\n\n\n_Notes_ : This table presents the estimated share of the intergenerational component using the Bingley and Cappellari\n(2019) method for each of the 53 developing countries in the Demographic and Health Surveys (DHS) using the\n1990s birth cohort sample.\n\n\n45\n\nMention: \"Demographic and Health Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The intergenerational component is estimated for 53 developing countries using the 1990s cohort in these surveys.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000804:5:1", "occurrence_id": "4ac033d7a20e03c950b6d484", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  separately in each\neconomy. However, given the independence of the data sources, not all the variables are\navailable for all the Western Balkan economies. For example, Serbia and North Macedonia\ndo not collect information about the economic sector from the workers. To overcome this\nproblem, we combine the I2D2 with the WTS database by calculating the aggregated salaries\nand employment levels for every region and multiplying them by the proportion of wages\nacross the three economic industries calculated at the national level. Additionally, some\nof the regions in the Western Balkans lack information on the salaries for specific sectors.\nTherefore, to circumvent this second problem, we input them using a regression model based\non the industry, education, and age variables.\n\n\nFinally, we use data from the GRIP to calculate the travel times across the Western\nBalkans region and the EU countries. This dataset gathers and harmonizes information on\nall the road infrastructure for over 222 countries worldwide.\n\n\n1Table 12 lists all the sub-national regions of the Western Balkans included in the analysis\n\n\n3\n\nMention: \"I2D2\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"Because some economies omit sector information, this database is combined with another to fill the gap.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000804:5:1", "occurrence_id": "4bdaee2ab3eb5e9a9ab1699a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  separately in each\neconomy. However, given the independence of the data sources, not all the variables are\navailable for all the Western Balkan economies. For example, Serbia and North Macedonia\ndo not collect information about the economic sector from the workers. To overcome this\nproblem, we combine the I2D2 with the WTS database by calculating the aggregated salaries\nand employment levels for every region and multiplying them by the proportion of wages\nacross the three economic industries calculated at the national level. Additionally, some\nof the regions in the Western Balkans lack information on the salaries for specific sectors.\nTherefore, to circumvent this second problem, we input them using a regression model based\non the industry, education, and age variables.\n\n\nFinally, we use data from the GRIP to calculate the travel times across the Western\nBalkans region and the EU countries. This dataset gathers and harmonizes information on\nall the road infrastructure for over 222 countries worldwide.\n\n\n1Table 12 lists all the sub-national regions of the Western Balkans included in the analysis\n\n\n3\n\nMention: \"WTS database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its national wage shares across three industries are used to allocate regional salaries and employment.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000807:22:1", "occurrence_id": "5616ef52509e7070d9dbf880", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: . <sup>44</sup> From\n\n\n38For the computation of fiscal multipliers below, we use quarterly data from IFS (IMF).\n39It is worth mentioning that the commodity price indices from Gruss and Kebhaj’s (2019) are countryspecific. Specifically, for each country, the change in the international price of up to 45 individual commodities is weighted using commodity-level data for the period 1962-2018.\n\n40If we use the Arellano-Bond estimator for dynamic linear panel methods to deal with lagged-dependent\nvariables, the same results hold for regressions (3) and (6). Also, using as controls world trade (as percentage\nof GDP) instead of terms of trade and public debt as percentage of GDP does not change the results.\n\n41Notice that regressions (4)-(6) control for the EMDE response by introducing an interactive term.\n42Since all variables are log-differences, the estimated coeffi cients are elasticities.\n43Comparing our quantitative estimates to those of other studies is not easy since empirical specifications\nvary considerably. With that caveat in mind, our estimates are very much comparable to those of, for\nexample, Richaud et al. (2019). Based on a sample of 87 EMDE commodity exporters, Richaud et al. report\na range of 0.34 to 1.25 percent for the cumulative impulse response of GDP to a one standard deviation shock\nin commodity prices. In our case, the sample is 11 EMDE with a coeffi cient range in the panel regressions\nin Table 3 of 0.063-0.085. Since the standard deviation of changes in commodity prices for EMDE is 14.4\npercent, the range for the impact on GDP is 0.92 to 1.22 percent. Notice that the comparison is valid\nbecause changes in commodity prices tend to be very persistent (see\n\nMention: \"IFS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Quarterly series from this fund statistic are used to compute fiscal multipliers.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000811:12:3", "occurrence_id": "32949cf93aea3697ef82f1ab", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <br>– <br>Sri<br>Lanka;<br> <br>Oct<br>- <br>Dec<br>2021<br>|Household<br>labor force<br>survey;<br> <br>Jan - March<br>2021<br>Apr - June<br>2021|\n\n\n\n<mark>Notes: Wave-1 data were collected between December 10</mark> <sup>th</sup> <mark>and 29</mark> <sup>th</sup> <mark>, 2020, wave-2 between April</mark>\n<mark>28</mark> <sup>th</sup> <mark>and May 9</mark> <sup>th</sup> <mark>, 2021, and wave-3 between August 5</mark> <sup>th</sup> <mark>and 23</mark> <sup>rd</sup> <mark>, 2021. Each wave had over 1000</mark>\n<mark>“completed” surveys per country. A “completed” survey is one where respondents answered</mark>\n<mark>until the eighteenth question, which included all questions related to employment status and</mark>\n<mark>characteristics. For the analysis, we use only completed surveys, and ignored partial-response</mark>\n<mark>surveys. The sample size for these surveys in each wave is presented in panel A in table 3.1.</mark>\n\n<mark>Second, we use the Global Monitoring Database, a probability-based survey, to compare</mark>\n<mark>demographic variables and individual attributes with our survey. The GMD is the repository of</mark>\n<mark>\n\nMention: \"Global Monitoring Database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"This probability-based survey is used to compare demographic variables and individual attributes with the study's sample.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000815:20:0", "occurrence_id": "9f62390edb136040a1bd1bf0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: through purchase but store the foods without consumption during the reference period. In these\ncases, expenditure data fails to reflect the actual amount of consumption.\n\nThe EAT Lancet planetary health diet’s composition as used for this paper is one specific selection\nthat has been made based on the quantified contribution from each food group as proposed by the\nEAT Lancet publication, with the contribution from different foods within each food group as\nobserved in the Bangladesh HIES. Another contribution from different food groups, so as to select\na vegetarian or vegan diet or to adapt to specific dietary patterns in different countries is also\npossible, but as no quantification for those options was provided, we used the quantified option.\n\n##### **5 Conclusion**\n\n\nMany developing countries (48 lower middle-income countries) have a severe double burden <sup>14</sup>\n(Popkin et al. 2020) where both undernutrition and poor diets are equally important to consider.\nFor ensuring access, links need to be drawn between social protection programs and food systems\nto ensure that sustainable approaches are taken to reduce undernutrition and poverty, which at the\nsame time do not harm these populations (i.e., by resulting in overweight/obesity and not resolving\nmicronutrient deficiencies) and the impact on the environment is appropriately balanced with the\nright food production policies and approaches.\n\n\n14 Defined as a prevalence of wasting of >15%, stunting of >30%, and thinness in women of >20%, and an adult or\nchild overweight prevalence >20%.\n\n\n19\n\nMention: \"Bangladesh HIES\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Observed contributions of foods within each group shape the diet selection used to cost a healthy diet.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000816:33:2", "occurrence_id": "b74f540267ec0518ca063d1a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .003) (0.003) (0.067)\nMax Night Lights (log) 2.363 0.177 0.106 0.093*\n(0.053) (0.003) (0.002) (0.053)\n\n\nNotes: This table reports the mean values for various village-level characteristics for various sub-samples used in\nthe study. The variables under panels demographics and public goods come from the 2011 Population Census.\nThe socioeconomic characterictics from the 2013 Socio-Economic and Caste Census. Monthly composites of\nnighttime lights are constructed from data obtained from Visible Infrared Imaging Suite. The NREGA variables\nare our calculations based on the data available from Ministry Of Rural Development, Government Of India. The\nspecific construction of each is explained in greater detail in the data section. Column 2 reports the mean for all\nvillages in the state. In Column 3, we report means for a restricted sample of villages in special talukas. This\nmeans villages in blocks where at least one GP has an elected leader in charge. Throughout our study, we use\nvillages and GPs in special talukas for our analysis. Column 4 reports the mean for all villages in Non-Special or\nnormal talukas. In Column 4, we report the coefficient on the difference in means on the Special-Taluka dummy.\nThe specification includes district fixed effects. Standard errors are robust and clustered at the taluka level.\nSignificant at *10%, **5%, and ***1%.\n\n\n32\n\nMention: \"2011 Population Census\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Village demographic and public goods characteristics are drawn from this census.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000816:33:2", "occurrence_id": "7b1dfa95a28590ffe2b41c8d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .003) (0.003) (0.067)\nMax Night Lights (log) 2.363 0.177 0.106 0.093*\n(0.053) (0.003) (0.002) (0.053)\n\n\nNotes: This table reports the mean values for various village-level characteristics for various sub-samples used in\nthe study. The variables under panels demographics and public goods come from the 2011 Population Census.\nThe socioeconomic characterictics from the 2013 Socio-Economic and Caste Census. Monthly composites of\nnighttime lights are constructed from data obtained from Visible Infrared Imaging Suite. The NREGA variables\nare our calculations based on the data available from Ministry Of Rural Development, Government Of India. The\nspecific construction of each is explained in greater detail in the data section. Column 2 reports the mean for all\nvillages in the state. In Column 3, we report means for a restricted sample of villages in special talukas. This\nmeans villages in blocks where at least one GP has an elected leader in charge. Throughout our study, we use\nvillages and GPs in special talukas for our analysis. Column 4 reports the mean for all villages in Non-Special or\nnormal talukas. In Column 4, we report the coefficient on the difference in means on the Special-Taluka dummy.\nThe specification includes district fixed effects. Standard errors are robust and clustered at the taluka level.\nSignificant at *10%, **5%, and ***1%.\n\n\n32\n\nMention: \"2013 Socio-Economic and Caste Census\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Socioeconomic village characteristics come from this caste and economic census.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000816:33:2", "occurrence_id": "b2412e87e336e4e23f0b0a68", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .003) (0.003) (0.067)\nMax Night Lights (log) 2.363 0.177 0.106 0.093*\n(0.053) (0.003) (0.002) (0.053)\n\n\nNotes: This table reports the mean values for various village-level characteristics for various sub-samples used in\nthe study. The variables under panels demographics and public goods come from the 2011 Population Census.\nThe socioeconomic characterictics from the 2013 Socio-Economic and Caste Census. Monthly composites of\nnighttime lights are constructed from data obtained from Visible Infrared Imaging Suite. The NREGA variables\nare our calculations based on the data available from Ministry Of Rural Development, Government Of India. The\nspecific construction of each is explained in greater detail in the data section. Column 2 reports the mean for all\nvillages in the state. In Column 3, we report means for a restricted sample of villages in special talukas. This\nmeans villages in blocks where at least one GP has an elected leader in charge. Throughout our study, we use\nvillages and GPs in special talukas for our analysis. Column 4 reports the mean for all villages in Non-Special or\nnormal talukas. In Column 4, we report the coefficient on the difference in means on the Special-Taluka dummy.\nThe specification includes district fixed effects. Standard errors are robust and clustered at the taluka level.\nSignificant at *10%, **5%, and ***1%.\n\n\n32\n\nMention: \"data available from Ministry Of Rural Development\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Employment guarantee programme variables are calculated from these administrative records.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000818:47:1", "occurrence_id": "0d87565f1e81da335210840d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: particularly in terms of its air, sea,\nland, and internet connections); <sup>37</sup> and its preparedness for rapid change and innovation (as\nmeasured by R&D expenditure, the number of new patents or the digital skills of the population,\nand the governments’ inability to provide a long-term vision). Since these supply gaps do not\nnecessarily constitute a binding constraint on growth and they are as much a cause as a\nconsequence of growth, the expected growth returns from filling the gaps are uncertain.\nNonetheless, they clearly point toward potential bottlenecks that must be assessed—and, if needed,\naddressed—as a matter of priority. <sup>38</sup>\n\n\n36 The figure was constructed using the World Economic Forum’s Global Competitiveness Index (excluding those\nassociated with macro policy and the financial sector), which classifies indices of 130 indicators into seven categories:\n(i) Rule of law (security, violence, quality of police and the judiciary); (ii) Market support (the quality of laws and\nregulations in support of market activities); (iii) Infrastructure (transportation and communications); (iv) Social\ncapital; (v) Innovation (the country’s and the state’s preparedness for change); (vi) Health and human rights (the\nattention to basic human needs and rights); and (vii) Trade openness (the country’s openness to both foreign goods\nand foreign factors).\n\n37 For a recent assessment of LA’s shortfalls in infrastructure, which highlights the costly implications for growth and\nwelfare of gaps in transport infrastructure, see Perry, Guzmán and Benavides (2020).\n\n38 To better tailor policies to country circumstances, avoid unhelpful “laundry lists” of ill connected actions, and help\nidentify areas more amenable to reform and with larger expected impacts, policy makers can use a variety of tools\n\n\n46\n\nMention: \"Global Competitiveness Index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Selected indicators from 130 categories excluding macro policy and finance are used to build the connectivity figure.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000819:40:0", "occurrence_id": "962dd43593895ff4f97787aa", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 4: Total Number of Flight Departures over the Sample Period**\n\n\n**Notes:** The figure reports the total number of international flight departures observed in each year between\nthe African country pairs in our data sample. The value reported on top of each bar represents the number\nof bilateral country pairs contributing departures data to the reported totals, highlighting the unbalanced\nnature of our estimation sample. The reported number of international departures is measured in thousands.\nThe original data comes from the OAG flights database.\n\n\n39\n\nMention: \"OAG flights database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"International flight departures between country pairs originate in this commercial flight schedule data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000821:9:0", "occurrence_id": "5b3d4ad891cccaa508de3e7b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: administration techniques can reduce sensitivity bias by changing the social context in which sensitive\nquestions are asked so that the identity of the respondent is obscured from interviewers, bystanders, and\nother potential social referents.\n\n\nWhile most of the core topics shared between the PPCS and the Facebook survey could have been\nconsidered sensitive, we expect that in this context the questions on conflict exposure and mental health\nmay have been more sensitive than the questions on food insecurity and COVID-19 exposure.\nRespondents may have perceived the questions on personal exposure to conflict to be sensitive because\nthey brought up memories of traumatic experiences. In the mental health modules, respondents may\nhave perceived questions as sensitive because of the negative attitudes and stigma around mental illness\nthat persist in many parts of the region. <sup>7</sup> [^7: See for example Alyafei et al (2021) and Elyamani et al (2021) for references to negative attitudes and stigma in\nregional context, or World Health Organization: Regional Office for the Eastern Mediterranean (n.d.) for reference\nto stigma in the Palestinian context.]\n\n\nQuestions about food insecurity may be perceived as sensitive because of their association with poverty,\nbut we expect this sensitivity to be reduced in the specific context of the PPCS because most respondents\nhad participated in the last wave of the Socio-Economic and Food Security Survey and might therefore be\naccustomed to answering questions about food insecurity. COVID-19 restrictions are not as severe as they\nwere in earlier stages of the pandemic, and surveys about people's experience of COVID have proliferated\nsince early 2020. We might therefore expect the difference in sensitivity bias between the PPCS and the\nFacebook survey to be more marked in the indicators related to conflict exposure and mental health than\nin the indicators related to food insecurity and COVID-19 exposure.\n\n\n_Hypotheses_\nH2a: People tend to underreport conflict exposure, aggressive tendencies, indicators of risk for mental\nhealth problems, and symptoms consistent with depression because these are perceived as dangerous or\nundesirable traits and experiences.\n\n\nH2b: People are more likely to\n\nMention: \"Facebook survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Overlapping topics allow sensitivity of conflict exposure and mental health questions to be assessed.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000821:9:0", "occurrence_id": "f4a915f273cb0a9da18fb5aa", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: administration techniques can reduce sensitivity bias by changing the social context in which sensitive\nquestions are asked so that the identity of the respondent is obscured from interviewers, bystanders, and\nother potential social referents.\n\n\nWhile most of the core topics shared between the PPCS and the Facebook survey could have been\nconsidered sensitive, we expect that in this context the questions on conflict exposure and mental health\nmay have been more sensitive than the questions on food insecurity and COVID-19 exposure.\nRespondents may have perceived the questions on personal exposure to conflict to be sensitive because\nthey brought up memories of traumatic experiences. In the mental health modules, respondents may\nhave perceived questions as sensitive because of the negative attitudes and stigma around mental illness\nthat persist in many parts of the region. <sup>7</sup> [^7: See for example Alyafei et al (2021) and Elyamani et al (2021) for references to negative attitudes and stigma in\nregional context, or World Health Organization: Regional Office for the Eastern Mediterranean (n.d.) for reference\nto stigma in the Palestinian context.]\n\n\nQuestions about food insecurity may be perceived as sensitive because of their association with poverty,\nbut we expect this sensitivity to be reduced in the specific context of the PPCS because most respondents\nhad participated in the last wave of the Socio-Economic and Food Security Survey and might therefore be\naccustomed to answering questions about food insecurity. COVID-19 restrictions are not as severe as they\nwere in earlier stages of the pandemic, and surveys about people's experience of COVID have proliferated\nsince early 2020. We might therefore expect the difference in sensitivity bias between the PPCS and the\nFacebook survey to be more marked in the indicators related to conflict exposure and mental health than\nin the indicators related to food insecurity and COVID-19 exposure.\n\n\n_Hypotheses_\nH2a: People tend to underreport conflict exposure, aggressive tendencies, indicators of risk for mental\nhealth problems, and symptoms consistent with depression because these are perceived as dangerous or\nundesirable traits and experiences.\n\n\nH2b: People are more likely to\n\nMention: \"Socio-Economic and Food Security Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Prior participation in this survey's latest wave is expected to reduce sensitivity to food insecurity questions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000823:9:0", "occurrence_id": "62a402c0a2492290802eccc3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: escalated, there was a 21 percentage point increase in the share of the population that was\n\n\nexperiencing acute food insecurity or worse relative to 2014 (IPC 2015). Furthermore, the\n\n\nEmergency Food Security and Nutrition Assessment estimated that 65 percent of Yemenis\n\n\nhad poor or borderline food consumption in 2016, a 57 percentage point increase relative\n\n\nto 2014 (FAO 2017; World Bank 2017). Since 2015, the severity of the food emergency\n\n\nhas varied along with variation in the shocks to income and prices described above. And\n\n\nfood insecurity worsened each year after 2015, teetering on the edge of famine at the end\n\n\nof 2018 (e.g., IPC 2017; IPC 2018; OCHA 2019a).\n\n\nHowever, given the importance of violence to displacement, it is important to note\n\n\nthat those who live in close proximity to violence do not have worse well-being on aver\n\nage. There is little correlation between the location of violence and access to food and\n\n\nbasic services, and there is little change in well-being in response to a number of sudden\n\n\nincreases in violence and violent offensives (e.g., Almoayad et al. 2020; Tandon and Vish\n\nwanath 2020; Favari et al. 2022; etc.). One of the reasons for the lack of a relationship\n\n\nis that particular instances of violence have impacts far beyond the location where they\n\n\ntake place (e.g, Tandon and Vishwanath 20202), and many of the other shocks described\n\n\nabove have much stronger and more robust impacts on the humanitarian situation (e.g.,\n\n\nFavari et al. 2021; etc.). Combined, these results suggest that the violence that might\n\n\nbe causing households to become displaced does not necessarily result in a worsening\n\n\nhumanitarian situation experienced in those locations from which displaced households\n\n\nmight be fleeing.\n\n\n**Section** **3a.** **Data**\n\n\nGiven the tumultuous environments in which large-scale displacements occur\n\nMention: \"Emergency Food Security and Nutrition Assessment\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its estimate that 65 percent faced poor or borderline food consumption anchors the deterioration narrative.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000827:19:0", "occurrence_id": "57c3651eb70e18f35adb4dcb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: frequently cited as a barrier. For example, 24 percent and 19 percent of the unbanked in Iraq and\n\n\nMorocco, respectively, cited religion as a barrier. There are also gender differences in citing\n\n\nreligion as a barrier to account ownership. For example, in Morocco, 14 percent of unbanked\n\n\nmen compared to 23 percent of unbanked women cite religion as a barrier. Sharia-compliant\n\n\nsavings and credit products can help address this barrier (Karlan et al. 2021). However, prices on\n\n\nSharia-compliant products are typically more expensive (Hussain et al. 2016), stressing the\n\n\nimportance that users understand the often more complicated terms and fees.\n\n\n**_5.2_** **_Lack of access to technology is a leading barrier to using a mobile money account_**\n\n\nAccount ownership has increased in certain countries due to the adoption of digital technologies,\n\n\nincluding mobile phones. Since 2014, Sub-Saharan Africa has been the global leader in mobile\n\n\nmoney account ownership. As of 2021, 33 percent of adults reported having a mobile money\n\n\naccount in the region, compared with only 10 percent of adults globally. In this way, mobile\n\n\nmoney has played a foundational role in driving increased account ownership in Sub-Saharan\n\n\nAfrica.\n\n\nGiven such high mobile money adoption in Sub-Saharan Africa, the Global Findex 2021 survey\n\n\nincluded a dedicated module that asked unbanked adults in that region (i.e., adults with neither a\n\n\nmobile money account nor an account at a traditional financial institution, like a bank)\n\n\nspecifically why they did not have a mobile money account. As with the general set of questions\n\n\nabout financial institution accounts, respondents could give more than one reason for not having\n\n\na mobile money account, though most gave fewer than two.\n\n\n18\n\nMention: \"Findex 2021 survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"A dedicated module asks unbanked adults in the region why they lack a mobile money account.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000828:20:0", "occurrence_id": "98cd2b665269225072fa5283", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: advantages for businesses when three conditions are exist: (1) customer willingness-to-pay for\n\nlabeled service, (2) label credibility, and (3) label exclusivity. Our findings resonate with the first and\n\n\n\nadvantages for businesses when three conditions are exist: (1) customer willingness-to-pay for\n\nlabeled service, (2) label credibility, and (3) label exclusivity. Our findings resonate with the first and\n\nthird conditions.\n\n\n\nthird conditions.\n\n\n\n\n\n_Notes:_\n\n_1._ _Source: Authors’ calculations based on data obtained from Booking.com_\n_2._ _The line represents the linear prediction of the model, and the shaded area represents the 95% confidence intervals._\n\n\n_Notes:_\n\n_1._ _Source: Authors’ calculations based on data obtained from Booking.com_\n_2._ _The line represents the linear prediction of the model, and the shaded area represents the 95% confidence intervals._\n\n\n19\n\nMention: \"data obtained from Booking.com_\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Platform listing data are used to calculate the model's predicted relationship with confidence bands.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000829:11:2", "occurrence_id": "b7402fb9a5f466f4b860ffc6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: , 𝑢𝑟𝑏𝑎𝑛ℎ𝑐𝑡,\n𝑝𝑜𝑤𝑒𝑟𝑔𝑟𝑖𝑑ℎ𝑐𝑡, 𝑖𝑛𝑡𝑒𝑟𝑛𝑒𝑡ℎ𝑐𝑡, 𝑝𝑢𝑏𝑙𝑖𝑐𝑖ℎ𝑐𝑡, 𝑛𝑐ℎ𝑖𝑙𝑑ℎ𝑐𝑡, 𝑓𝑒𝑒𝑙𝑎𝑛𝑥𝑖𝑜𝑢𝑠ℎ𝑐𝑡. Moreover, including county fixed effects\nis redundant in this case, as very few children moved between waves.\n23 Given that the analysis is limited to two periods, this is equivalent to specifying the model in first differences.\n24 Figure A1, in the appendix, shows the adoption rates of remote and alternative forms of education by different\ngroups of the population.\n25 Data on internet access and computer ownership became available only in January 2021 (wave four), and\ntherefore is not included in the baseline regressions throughout this section. In the 2019 Kenya Population and\nHousing Census, only 22.6 percent of individuals were reported using the internet and 10.4 using computers\n(KNBS 2019).\n26 Between September and November 2020 (wave three), almost 60 percent of children who learned remotely did\nso via the internet.\n27 Including other interaction terms, such as between female and age or female and number of hours spent on house\nchores, did not yield significant results (not shown).\n\n10\n\nMention: \"2019 Kenya Population and\nHousing Census\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Internet and computer use rates reported here frame the remote learning coverage figures.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000834:12:0", "occurrence_id": "94902319e9ee0ce4bf62ab50", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Our key independent variable of interest is the political connectivity of each firm. We follow Faccio\n(2006), and others, and code a firm as politically connected when it has at least one board member or\nCEO who is a politician, a close relative of one, or a publicly known friend. For that purpose, we leverage online business directories and Lebanon’s commercial registry to look up the name of each firm’s\nboard members in addition to collecting data on their size, age, and paid-in capital.\n\nOur approach to identify political connections takes into account that political connections are a complex phenomenon in a country like Lebanon (Leenders, 2012; Diwan and Haidar, 2020). We go beyond\nprevious studies, which establish objective criteria for the identification of connections, such as by name\nmatching of a company’s shareholder or CEO names with those of political actors. As such approaches\nhave tended to underestimate results, <sup>7</sup> [^7: The widely-cited work of Faccio (2006), for example, uses a data set of firms worldwide and finds no politically connected\nfirms in Zimbabwe and Venezuela—two countries with an arguably weak record for the control of corruption. Even for the\nUnited States, where the author’s data set includes more than 7,000 firms, her approach only identifies 14 connected firms\n(p. 374), a number that other works have found to be much higher (Goldman, Rocholl and So, 2009).] we instead review each firm in our data set manually via an\napproach outlined in detail in Mahmalat, Atallah and Maktabi (2021). For each firm, we go through a\nmulti-layered search process that relies on media searches on the names of each board member of a firm\nwith a corresponding name of a politician or their political party. This approach allows us to carefully\nassess a number of common issues in the identification of political connections, such as whether individuals with matching names are related, connections are\n\nMention: \"Lebanon’s commercial registry\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Board membership and firm size, age and paid-in capital are looked up in this registry to identify political connections.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000836:33:1", "occurrence_id": "962da46515974759026e5cac", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 462) (0.444) (0.431) (0.403)\n\nThriving 0.262 0.108 0.180*** 0.278*** 0.414***\n\n(0.440) (0.310) (0.385) (0.448) (0.493)\n\nStruggling 0.606 0.689 0.681*** 0.583*** 0.508***\n\n(0.489) (0.463) (0.466) (0.493) (0.500)\n\nSuffering 0.132 0.204 0.139*** 0.140*** 0.0781***\n\n(0.339) (0.403) (0.346) (0.347) (0.268)\n\nNumber of observations 388,602 69,101 106,410 100,728 112,363\n\n_<u>Notes:</u>_ <u>Means calculated using sample-weighted individual-level data from the 2014-17 Gallup World Poll. Standard errors are in parentheses.</u> <u>Asterisks indicate</u>\n\nwhether the difference in means is statistically significantly different from low-income countries.\n\n\n32\n\nMention: \"Gallup World Poll\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Sample-weighted individual means of thriving and struggling are computed on this poll.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000837:51:2", "occurrence_id": "c30ff34cb41089cae2ae19cb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  83.8 30.9 31.4 30.7\n\n\nSpouse born in the U.S 55.2 13.0 13.5 12.9 63.8 63.1 64.1\n\n\n<u>Married</u> <u>after</u> <u>migrating:</u>\n\n\nSpouse born in the same country 63.4 81.2 78.8 84.7 51.8 53.8 49.0\n\n\nSpouse born in the U.S 28.7 12.5 14.2 10.1 39.3 37.3 42.1\n\n\nObservations 37,381 13,215 5,671 7,544 24,166 10,059 14,107\n\n\n**Notes** Sample: individuals aged 18 to 60 at the time of the survey, born in China, and who migrated\n\n\nto the U.S. when they were between 18 and 40.\n\n\nData source: ACS 2017-2021 (Ruggles et al., 2023).\n\n50\n\nMention: \"ACS 2017-2021\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Characteristics of Chinese migrants aged 18-60 are tabulated from this American community survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000838:29:0", "occurrence_id": "46a557f0512035d7490333e9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: cases where disruptions would be expected to have severe economic implications due to relation\n\nspecific investments. But the experience with Russia revealed that at least Europe was ill\n\nprepared to deal with disruptions due to major geopolitical shocks. The concerns about Europe\n\n\nand Russia have led---by extrapolation---to more general concerns about the resilience of global\n\n\nsupply chains to geopolitical risks, and to demands to fundamentally reorganize international\n\n\nrelationships, so as to decouple from any country that is perceived as “unfriendly” by the US.\n\n\nChina tops the list.\n\n\nA natural question in this context is whether, beyond specific instances of government\n\nled sanctions, market forces have led countries to already orient themselves towards trading with\n\n\n“friends.” In Figure 6 we evaluate whether the US is in fact friendshoring, by relating the import\n\n\nshare of a country in 2022 to the percent of Americans believing the country is a friend or ally in\n\n\nthe most recent poll by YouGov (2017). For many trading partners, a majority of Americans\n\n\nbelieve the country is a friend, and there is a positive and statistically significant association\n\n\nbetween this variable and a country’s share of imports. China, the largest trading partner, is in\n\n\nfact an outlier, with just 35 percent of Americans believing it is a friend.\n\n\n_Figure 6: America is already friend shoring_\n\n\n_Sources:_ YouGov (2017), U.S Census Bureau (2022).\n\n\n_Notes:_ Countries are classified as unfriendly if less than 50 percent of Americans believe country is a friend or ally.\nFigure includes only countries with greater than one tenth of one percent of U.S. imports in 2022. The slope of the\nred dashed line is 4.04 (s.e. = 1.87).\n\n\n28\n\nMention: \"poll by YouGov\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Shares of Americans who regard a country as a friend are related to that country's import share.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000845:19:0", "occurrence_id": "35af25853d93413d11565f52", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Site_ _Selection_ . <sup>3</sup> This journal, in a series of articles called the _The_ _Million_ _Dollar_ _Plant_,\n\n\nreports the final location of a new plant creation, along with the identity of the potential\n\n\nalternate or the _runner-up_ location(s). We refer to each such set of winning and losing\n\n\ncounties as a _deal_ . Our final number of deals varies depending on the outcomes we are\n\n\nlooking at (e.g. labor market or retail inflation) which determine our sample sizes. <sup>4</sup>\n\n###### **3.2 Retail Price Data: IRI Data**\n\n\nOur primary retail price data set covers various grocery stores and drugstores from 2001\n\n\nto 2012 and is provided by IRI Worldwide. The data set includes store-week-UPC sales\n\n\nand quantity information for products in 31 categories, representing roughly 15 percent of\n\n\nhousehold spending in the Consumer Expenditure Survey. In total, the data cover around\n\n\n7,200 stores in about 47 IRI markets, corresponding to 968 US counties. IRI markets are\n\n\ncomposed of U.S. counties, which can represent a single metropolitan area (e.g. Los Angeles)\n\n\nor various metropolitan statistical areas aggregated into one region (e.g. West Texas/New\n\n\nMexico covers all FIPS counties of New Mexico but only some of Western Texas). There are\n\n\na large number of papers in the literature that study regional inflation using the definition of\n\n\nmarket constructed by the IRI (Pandya and Venkatesan, 2016; Bronnenberg et al., 2008). <sup>5</sup>\n\n\nWhile the raw data are sampled weekly, we construct quarterly prices both at category and\n\n\nbarcode level, similar to Stroebel and Vavra (2019), since this reduces high-frequency\n\nMention: \"Retail Price Data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Store-week product sales and quantities across 31 categories provide the price outcomes.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000845:19:0", "occurrence_id": "2db008623a66b45e8b0b5b74", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Site_ _Selection_ . <sup>3</sup> This journal, in a series of articles called the _The_ _Million_ _Dollar_ _Plant_,\n\n\nreports the final location of a new plant creation, along with the identity of the potential\n\n\nalternate or the _runner-up_ location(s). We refer to each such set of winning and losing\n\n\ncounties as a _deal_ . Our final number of deals varies depending on the outcomes we are\n\n\nlooking at (e.g. labor market or retail inflation) which determine our sample sizes. <sup>4</sup>\n\n###### **3.2 Retail Price Data: IRI Data**\n\n\nOur primary retail price data set covers various grocery stores and drugstores from 2001\n\n\nto 2012 and is provided by IRI Worldwide. The data set includes store-week-UPC sales\n\n\nand quantity information for products in 31 categories, representing roughly 15 percent of\n\n\nhousehold spending in the Consumer Expenditure Survey. In total, the data cover around\n\n\n7,200 stores in about 47 IRI markets, corresponding to 968 US counties. IRI markets are\n\n\ncomposed of U.S. counties, which can represent a single metropolitan area (e.g. Los Angeles)\n\n\nor various metropolitan statistical areas aggregated into one region (e.g. West Texas/New\n\n\nMexico covers all FIPS counties of New Mexico but only some of Western Texas). There are\n\n\na large number of papers in the literature that study regional inflation using the definition of\n\n\nmarket constructed by the IRI (Pandya and Venkatesan, 2016; Bronnenberg et al., 2008). <sup>5</sup>\n\n\nWhile the raw data are sampled weekly, we construct quarterly prices both at category and\n\n\nbarcode level, similar to Stroebel and Vavra (2019), since this reduces high-frequency\n\nMention: \"IRI Data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The named dataset covering about 7,200 stores in 968 counties defines the geographic unit of analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000845:19:0", "occurrence_id": "6983b2094981532411ebc383", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Site_ _Selection_ . <sup>3</sup> This journal, in a series of articles called the _The_ _Million_ _Dollar_ _Plant_,\n\n\nreports the final location of a new plant creation, along with the identity of the potential\n\n\nalternate or the _runner-up_ location(s). We refer to each such set of winning and losing\n\n\ncounties as a _deal_ . Our final number of deals varies depending on the outcomes we are\n\n\nlooking at (e.g. labor market or retail inflation) which determine our sample sizes. <sup>4</sup>\n\n###### **3.2 Retail Price Data: IRI Data**\n\n\nOur primary retail price data set covers various grocery stores and drugstores from 2001\n\n\nto 2012 and is provided by IRI Worldwide. The data set includes store-week-UPC sales\n\n\nand quantity information for products in 31 categories, representing roughly 15 percent of\n\n\nhousehold spending in the Consumer Expenditure Survey. In total, the data cover around\n\n\n7,200 stores in about 47 IRI markets, corresponding to 968 US counties. IRI markets are\n\n\ncomposed of U.S. counties, which can represent a single metropolitan area (e.g. Los Angeles)\n\n\nor various metropolitan statistical areas aggregated into one region (e.g. West Texas/New\n\n\nMexico covers all FIPS counties of New Mexico but only some of Western Texas). There are\n\n\na large number of papers in the literature that study regional inflation using the definition of\n\n\nmarket constructed by the IRI (Pandya and Venkatesan, 2016; Bronnenberg et al., 2008). <sup>5</sup>\n\n\nWhile the raw data are sampled weekly, we construct quarterly prices both at category and\n\n\nbarcode level, similar to Stroebel and Vavra (2019), since this reduces high-frequency\n\nMention: \"Consumer Expenditure Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Household spending covered by the scanner panel is compared against this expenditure survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000846:23:2", "occurrence_id": "d54b9762c9e31c4645ac5d83", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: u> <u>125</u>\n\n\n\n<u>Source: Authors’ compilation from the GMD, ASPIRE, and Findex databases.</u>\nNotes: Income refers to having less than 1.5*$2.15 per household member, education refers to no adult having primary\nschooling, social protection refers to not receiving any social protection, and financial inclusion refers to not having a bank or\nmobile money account. Water refers to access to limited standard water, and access to electricity is as written. The coverage\nyear for education, water and electricity is based on the filters of +- 5 years from 2019, and for social protection is based on the\nfilter of +- 5years from 2017.\n\n\n\n24 For China, we used the Household survey CHIP 2013 for dimensions of education and social protection.\n\n\n\n22\n\nMention: \"GMD\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Household deprivations in income, education and social protection are compiled from this microdata repository.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000847:10:0", "occurrence_id": "c75ede797b0a244bf96d7ebb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: household’s recovery. While encouraging, these results show the joint effect\n\n\nof the basic program and top-ups, and not the isolated the effect of the latter.\n\n#### **3 The Pantawid Pamilyang Pilipino Program**\n\n\nThe 4P conditional cash transfer program was first introduced in 2008 by the\n\n\nDepartment of Social Welfare and Development (DSWD). After a very lim\n\nited pilot in 2007, the program started its geographic rollout in March 2008.\n\n\nOver the course of that year, 4P started operating in 160 municipalities in\n\n\n33 provinces representing all 17 regions. Geographic eligibility was based on\n\n\nthe poorest municipalities in the 20 poorest provinces, the poorest provinces\n\n\nin regions not covered by this criterion, and some pockets of poverty in ur\n\nban areas. In the first half of 2009, the program was expanded to cover all\n\n\nmunicipalities with a poverty index of 61.23 percent or more. This added\n\n\nanother 28 provinces and 140 municipalities to the area under coverage. In\n\n\n2010 the threshold for the municipal poverty index was dropped to 57.13 per\n\ncent and in 2011 to 36.99 percent. In 2011, the program also experienced its\n\n\nbiggest expansion by adding over 1.2 million households as beneficiaries as it\n\n\nextended to cover almost all provinces. After a further expansion in 2012 it\n\n\nfinally covered all municipalities in 2013. To determine household eligibility,\n\n\nsince 2010 it employs the proxy means test based _Listahanan_ registry. In ad\n\ndition to being considered poor according to _Listhanan_, eligible households\n\n\n9\n\nMention: \"_Listahanan_ registry\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Household eligibility for transfers is determined by the proxy means test recorded in this registry.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000853:6:0", "occurrence_id": "cfed0b8b4556776394f6659f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: With the data, we compute firm-level export and import values, volumes, prices, number of\n\ntransactions, export destinations, import origin countries, and number of products. The average\n\nprice calculation at the firm-level accounts for the fact that different firms produce different prod\nucts. So, for every disaggregated product (8-digit HS code), we compute the average price per\n\nunit across all firms and all destinations in that month. For every firm, we compute a ratio most\n\ndisaggregated produce-wise: the numerator is the price that the firm charges for that HS-8 product\n\nand the denominator is the average price of that product. Then, we compute a weighted firm-month\n\naverage price per unit of these products in the firm’s export basket.\n\n\nWe define GVC firms as firms that imported and exported at least once in the Covid-19 pre\nperiod (February 2019-November 2019). Non-GVC firms are either pure importers or pure ex\nporters. There are 8,290 GVC firms (23.11%), 22,065 pure importers (61.52%), and 5,511 pure\n\nexporters (15.37%) in our data.\n\n\nThe data shows that 80.4 percent of export value for Indonesia was generated by GVC firms\n\nbefore Covid (i.e., before March 2020) and 82.1 percent after Covid. Overall, about 81.5 percent of\n\nIndonesia’s exports were generated by GVC firms in the sample period. These figures are slightly\n\nhigher than what was reported in other studies where about two-thirds of export values were gen\nerated by importing-exporters (Cal`ı et al., 2022).\n\n\nThe data on non-tariff measures (NTMs) is sourced from the World Bank NTM database for\nIndonesia. <sup>1</sup> This dataset improves on existing data from UN\n\nMention: \"data on non-tariff measures\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Regulatory measures affecting trade are taken from this country-specific database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000853:6:0", "occurrence_id": "ab892977b2d214de09ff9975", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: With the data, we compute firm-level export and import values, volumes, prices, number of\n\ntransactions, export destinations, import origin countries, and number of products. The average\n\nprice calculation at the firm-level accounts for the fact that different firms produce different prod\nucts. So, for every disaggregated product (8-digit HS code), we compute the average price per\n\nunit across all firms and all destinations in that month. For every firm, we compute a ratio most\n\ndisaggregated produce-wise: the numerator is the price that the firm charges for that HS-8 product\n\nand the denominator is the average price of that product. Then, we compute a weighted firm-month\n\naverage price per unit of these products in the firm’s export basket.\n\n\nWe define GVC firms as firms that imported and exported at least once in the Covid-19 pre\nperiod (February 2019-November 2019). Non-GVC firms are either pure importers or pure ex\nporters. There are 8,290 GVC firms (23.11%), 22,065 pure importers (61.52%), and 5,511 pure\n\nexporters (15.37%) in our data.\n\n\nThe data shows that 80.4 percent of export value for Indonesia was generated by GVC firms\n\nbefore Covid (i.e., before March 2020) and 82.1 percent after Covid. Overall, about 81.5 percent of\n\nIndonesia’s exports were generated by GVC firms in the sample period. These figures are slightly\n\nhigher than what was reported in other studies where about two-thirds of export values were gen\nerated by importing-exporters (Cal`ı et al., 2022).\n\n\nThe data on non-tariff measures (NTMs) is sourced from the World Bank NTM database for\nIndonesia. <sup>1</sup> This dataset improves on existing data from UN\n\nMention: \"World Bank NTM database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"It is described as improving on the previously available international source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000854:10:0", "occurrence_id": "1a0ef27d1dadf17f5f9fe744", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: cashiers in a grocery store with 60 employees. When collecting the five data points under the _Parenthood_\n\nindicator, the research applies additional assumptions, such as that mothers and fathers have worked\n\nlong enough to accrue maternity, paternity, and parental benefits. <sup>5</sup> In each data collection cycle, specific\n\ntime frames are set, and only legislation entered into force in each time frame is measured. Only codified\n\nlaws (including codified customary laws) are considered in data collection. WBL data were collected from\n\nprimary legal sources and confirmed by local experts through the administration of standardized\n\nquestionnaires. Historical data on laws and regulations before 2009 and going back to 1970 were collected\n\nthrough desk research of the primary legal sources and consulted with secondary sources where the\n\nprimary ones could not be located. <sup>6</sup>\n\n\nWe follow the WBL methodology in our analysis, and the total number of leave days available to a mother\n\nis calculated as the combination of maternity leave, parental leave days reserved for the mother and any\n\nadditional days that are not specifically reserved for the use of either parent. <sup>7</sup> Fathers’ leave is the sum of\n\npaternity leave days plus any days of parental leave specifically allocated to the father. In our analysis,\n\nonly paid leave is counted, and all leave days are calculated in calendar days. Continuing with the example\n\nof Germany, according to data for 2021, women in Germany are entitled to 98 days of maternity leave, <sup>8</sup>\n\nwhile there is no provision in place granting paternity leave specifically (fathers’ leave is only covered\n\nunder parental leave policy). Thus, according to our method of counting leave days, total leave days for\n\nmothers is 458—of 98 maternity leave days, 60 parental leave days reserved for mothers, and 300\n\nunallocated parental leave days. Total leave days for fathers is 60 days, comprised\n\nMention: \"WBL data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Indicators are coded from primary legal sources and confirmed by local experts using standard questionnaires.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000860:4:1", "occurrence_id": "542c13a9060d8ebc13078b5c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  an equivalent increase in the\nstringency of NPIs, as captured by the Oxford Covid-19 Government Response Tracker (OxCGRT)\nstringency index (Hale et al., 2022).\n\n\nIn lower-income countries, elasticities are estimated to be less negative for cities with higher\npopulation densities, implying less adverse impacts of the Covid-19 crisis on these cities as compared\nto less densely populated ones. Higher population densities are also associated with lower total loss\nof lights for cities in lower-income countries. These results suggest that, for lower-income countries,\nthe tendency of dense agglomerations to persist outweighs the negative impacts of density during an\nepidemic outbreak – i.e., the city resilience effect tends to outweigh the health-preserving behavior\neffect. By contrast, for cities in high-income countries, population density is uncorrelated with the\nloss of lights. This likely reflects differences in the characteristics of lower density locations between\nrich and poor countries. For instance, compared to similar cities in poorer countries, low-density\ncities in richer countries have higher capacities to provide public goods and services that build\nresilience to shocks (World Bank, forthcoming).\n\n\nMethodologically, our paper builds on a well-established and growing literature in economics that\nuses nighttime lights to proxy economic activity (Henderson et al., 2012; Bleakley and Lin, 2012;\nStoreygard, 2016; Jedwab et al., 2021). However, with only a few exceptions, most of these studies\nhave relied on data from Operational Linescan System (OLS) satellite sensors from the US\ngovernment’s Defense Meteorological Satellite Program (DMSP). While these data have the\nadvantage of having a relatively long annual time series spanning 1992–2013, they do not cover more\nrecent years. They also suffer from several drawbacks that undermine their ability to proxy economic\nactivity at finer geographic resolutions. These drawbacks include the inability to detect nighttime\nlight variations\n\nMention: \"Oxford Covid-19 Government Response Tracker\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The stringency index measuring non-pharmaceutical interventions is taken from this tracker.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000860:4:1", "occurrence_id": "8ad0775d9952816f031e4ab9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  an equivalent increase in the\nstringency of NPIs, as captured by the Oxford Covid-19 Government Response Tracker (OxCGRT)\nstringency index (Hale et al., 2022).\n\n\nIn lower-income countries, elasticities are estimated to be less negative for cities with higher\npopulation densities, implying less adverse impacts of the Covid-19 crisis on these cities as compared\nto less densely populated ones. Higher population densities are also associated with lower total loss\nof lights for cities in lower-income countries. These results suggest that, for lower-income countries,\nthe tendency of dense agglomerations to persist outweighs the negative impacts of density during an\nepidemic outbreak – i.e., the city resilience effect tends to outweigh the health-preserving behavior\neffect. By contrast, for cities in high-income countries, population density is uncorrelated with the\nloss of lights. This likely reflects differences in the characteristics of lower density locations between\nrich and poor countries. For instance, compared to similar cities in poorer countries, low-density\ncities in richer countries have higher capacities to provide public goods and services that build\nresilience to shocks (World Bank, forthcoming).\n\n\nMethodologically, our paper builds on a well-established and growing literature in economics that\nuses nighttime lights to proxy economic activity (Henderson et al., 2012; Bleakley and Lin, 2012;\nStoreygard, 2016; Jedwab et al., 2021). However, with only a few exceptions, most of these studies\nhave relied on data from Operational Linescan System (OLS) satellite sensors from the US\ngovernment’s Defense Meteorological Satellite Program (DMSP). While these data have the\nadvantage of having a relatively long annual time series spanning 1992–2013, they do not cover more\nrecent years. They also suffer from several drawbacks that undermine their ability to proxy economic\nactivity at finer geographic resolutions. These drawbacks include the inability to detect nighttime\nlight variations\n\nMention: \"Operational Linescan System\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its long annual series is an advantage but it does not cover recent years.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000864:60:0", "occurrence_id": "e855e40836639c9f0bdece50", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: of 2019, and 7 percent by the end of 2020. Firms assigned to treatment are more likely to\n\nsurvive by 3 percentage points in 2019, and 5 percentage points in 2020. Firms may not\n\nappear in the PILA because they have closed, or because of reporting issues. For example,\n\na couple of the firms appear to be registered only under the owner’s name as a natural\n\nperson, rather than as companies. We therefore consider a second definition of survival,\n\nwhich involves cross-checking data from formal employment reported in the PILA with\n\ndata on the firm from the export database, from the RUES (Mercantile Registry), and\n\nfrom IPA calling and visiting firms to verify their status. We believe this is the most\n\naccurate measure of firm survival as a result. Column 3 looks at impacts on this measure\n\nof survival for the full 200 firms, and again finds treated firms are 5.1 percentage points\n\nmore likely to survive, with this effect significant at the 10 percent level.\n\n\nThe firms that closed were considerably smaller to begin with on average than those\n\nthat survived. For the control group, mean (median) January 2018 employment was 87\n\n(48) for the survivors, compared to 51 (21) for firms that closed; for the treatment group\n\nit was 73 (43) for the survivors, compared to 23 (11) for firms that closed. Moreover,\n\nthe firms that died were much less likely to be engaged in exporting: 0 out of the control\n\nfirms that died had exported in the past three years prior to applying for the program\n\ncompared to 64 percent of the control firms that survived, and only 1 of the treated firms\n\nthat died had exported in the three years prior to applying for the program. Since we\n\ncode exports as zero for firms that are closed, closure effectively results in replacing 0\n\nexports for an open firm with 0 exports for a closed firm.\n\n\nPanel B of Table E2 provides treatment estimates on monthly firm employment\n\nMention: \"PILA\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Firms may be absent because they closed or because of registration and reporting problems.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000864:60:0", "occurrence_id": "9d21f3eea1028a385b5d659f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: of 2019, and 7 percent by the end of 2020. Firms assigned to treatment are more likely to\n\nsurvive by 3 percentage points in 2019, and 5 percentage points in 2020. Firms may not\n\nappear in the PILA because they have closed, or because of reporting issues. For example,\n\na couple of the firms appear to be registered only under the owner’s name as a natural\n\nperson, rather than as companies. We therefore consider a second definition of survival,\n\nwhich involves cross-checking data from formal employment reported in the PILA with\n\ndata on the firm from the export database, from the RUES (Mercantile Registry), and\n\nfrom IPA calling and visiting firms to verify their status. We believe this is the most\n\naccurate measure of firm survival as a result. Column 3 looks at impacts on this measure\n\nof survival for the full 200 firms, and again finds treated firms are 5.1 percentage points\n\nmore likely to survive, with this effect significant at the 10 percent level.\n\n\nThe firms that closed were considerably smaller to begin with on average than those\n\nthat survived. For the control group, mean (median) January 2018 employment was 87\n\n(48) for the survivors, compared to 51 (21) for firms that closed; for the treatment group\n\nit was 73 (43) for the survivors, compared to 23 (11) for firms that closed. Moreover,\n\nthe firms that died were much less likely to be engaged in exporting: 0 out of the control\n\nfirms that died had exported in the past three years prior to applying for the program\n\ncompared to 64 percent of the control firms that survived, and only 1 of the treated firms\n\nthat died had exported in the three years prior to applying for the program. Since we\n\ncode exports as zero for firms that are closed, closure effectively results in replacing 0\n\nexports for an open firm with 0 exports for a closed firm.\n\n\nPanel B of Table E2 provides treatment estimates on monthly firm employment\n\nMention: \"export database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Customs records are cross-checked against reported employment to confirm whether a firm still operates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000864:60:0", "occurrence_id": "defffd0c1ee15c9102b1b676", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: of 2019, and 7 percent by the end of 2020. Firms assigned to treatment are more likely to\n\nsurvive by 3 percentage points in 2019, and 5 percentage points in 2020. Firms may not\n\nappear in the PILA because they have closed, or because of reporting issues. For example,\n\na couple of the firms appear to be registered only under the owner’s name as a natural\n\nperson, rather than as companies. We therefore consider a second definition of survival,\n\nwhich involves cross-checking data from formal employment reported in the PILA with\n\ndata on the firm from the export database, from the RUES (Mercantile Registry), and\n\nfrom IPA calling and visiting firms to verify their status. We believe this is the most\n\naccurate measure of firm survival as a result. Column 3 looks at impacts on this measure\n\nof survival for the full 200 firms, and again finds treated firms are 5.1 percentage points\n\nmore likely to survive, with this effect significant at the 10 percent level.\n\n\nThe firms that closed were considerably smaller to begin with on average than those\n\nthat survived. For the control group, mean (median) January 2018 employment was 87\n\n(48) for the survivors, compared to 51 (21) for firms that closed; for the treatment group\n\nit was 73 (43) for the survivors, compared to 23 (11) for firms that closed. Moreover,\n\nthe firms that died were much less likely to be engaged in exporting: 0 out of the control\n\nfirms that died had exported in the past three years prior to applying for the program\n\ncompared to 64 percent of the control firms that survived, and only 1 of the treated firms\n\nthat died had exported in the three years prior to applying for the program. Since we\n\ncode exports as zero for firms that are closed, closure effectively results in replacing 0\n\nexports for an open firm with 0 exports for a closed firm.\n\n\nPanel B of Table E2 provides treatment estimates on monthly firm employment\n\nMention: \"RUES\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"The mercantile registry supplies a further check on firm status.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000869:19:2", "occurrence_id": "78d83f5d4e0a5505f4af049e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  lower-tax jurisdictions. A similar logic\napplies to top income earners. Consequently, corporate tax rates and personal income tax rates for the\ntop percentile declined in 65 countries over the period 1980-2007, whereas the median income tax\nrate rose, both of which have been attributed to rising globalization (Egger et al. 2019). This is\n\n\n6 “Trade taxes” is defined by the World Development Indicators as follows: “Taxes on international trade include import\nduties, export duties, profits of export or import monopolies, exchange profits, and exchange taxes.” As such, it does not\ninclude value added taxes levied on imported goods.\n\n18\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Its definition of trade taxes is quoted to clarify what the variable does and does not include.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000870:21:0", "occurrence_id": "227e2ce843839f35f022df60", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 4 Average MSE of Poverty Gap at 6.85 USD 2017 PPP, for LAC Countries**\n\n\n\n\n\nSource: Own estimations based on SEDLAC.\n\n\n**Inequality**\n\n\nThis section presents results for the three model variations when estimating the most standard inequality\nmeasure – the Gini coefficient. <sup>22</sup> [^22: This indicator measures how dispersed the income distribution is and takes values from 0 to 1, where 0 represents perfect\nequality and 1 stands for perfect inequality.] Results indicate that, like the case of the international upper poverty\nline, the prediction is similar for the three model variations in the short run, and the fit is generally\nreasonable (Figure 5). Still, errors increase with the distance between the baseline and the estimated\nyear. In addition, results show that the assessed variations tend to constantly overestimate inequality for\nsome countries (i.e., Bolivia, Costa Rica, El Salvador, Honduras, Mexico, Panama, Peru, and Paraguay)\nand underestimate it in others (like the Dominican Republic). Thus, the bias is not always in the same\ndirection or magnitude. For instance, the _new model with rescaling_ has the lowest gap in Bolivia in 2017\nbut the largest from 2018 through 2020.\n\n\nFigure 6 presents the MSEs for LAC countries by year and its trend to compare all the model variations\nover time. In this case, errors seem to increase in a linear pattern instead of the exponential-shaped trend\nobserved in the poverty and vulnerability indicators. Overall, results suggest that the _old model_ is the best\nvariation of the micro-simulation model to estimate inequality since its MSE is constantly lower than the\nother tested variations, except for atypical years such as 2020.\n\n\nIn summary, results suggest that no variation of the model should be considered a best-all-countries/allmeasures fit since country-specificities might arise. Depending on the purpose of the analysis, either the\n_new model without rescaling_ (poverty and vulnerability) or the _old model_ (inequality) might perform\nbetter.\n\nMention: \"SEDLAC\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Poverty gap estimates and their mean squared errors are computed on this regional household survey database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000880:4:0", "occurrence_id": "7bee7fac48ceff289a079583", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ration in a PDS shop other than their home shop. Only 8% believed they could claim ration\n\n\noutside their home state.\n\n\nWe embedded an information experiment into a panel survey, the Consumer Pyramid\n\n\nHousehold Survey (CPHS), carried out three times a year by the Center for Monitoring\n\n\nIndian Economy (CMIE). During the survey wave from September–December 2021, enumer\n\nators read a script explaining the ONORC scheme to households in clusters—corresponding\n\n\napproximately to villages in rural areas and towns in urban areas—randomly assigned to the\n\n\ntreatment group. The script included information about barriers that migrants might face\n\nwhen attempting to claim ration at a non-designated shop. <sup>2</sup> This information was based\n\n\non pre-experimental research we conducted into the program’s implementation frictions, in\n\ncluding sending migrant mystery shoppers to test the ONORC program and phone surveys\n\n\nwith PDS shop owners. We also provided treated households with a toll-free phone number\n\n\nthey could call to reach an information hotline which we staffed for five months. Our staff\n\n\nwere trained to answer questions about the ONORC program or to find ration shops across\n\n\n29 states using a database we created of ration shop addresses and phone numbers.\n\n\nThe information we provided immediately increased beliefs about ration portability\n\n\nwithin district, across district, and across state lines, indicating that many households were\n\n\nunaware of the ONORC program. Four months later, treated households had significantly\n\n\n_lower_ beliefs about ration portability compared to control-group households. We show that\n\n\nconcurrent awareness campaigns by state governments—which informed households about\n\n\nthe ONORC scheme but did not provide the same information about barriers to access\n\ning it—explain the higher beliefs about ration portability in the control group four months\n\nafter our experiment. <sup>3</sup> To measure the intensity of government awareness campaigns, we\n\n\nexploit additional, out-of-sample data on beliefs about ration portability which\n\nMention: \"Consumer Pyramid\n\n\nHousehold Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"An information experiment on ration portability was embedded in this repeating household panel.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000880:4:0", "occurrence_id": "4e07f5a5bfeecf5eca03300e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ration in a PDS shop other than their home shop. Only 8% believed they could claim ration\n\n\noutside their home state.\n\n\nWe embedded an information experiment into a panel survey, the Consumer Pyramid\n\n\nHousehold Survey (CPHS), carried out three times a year by the Center for Monitoring\n\n\nIndian Economy (CMIE). During the survey wave from September–December 2021, enumer\n\nators read a script explaining the ONORC scheme to households in clusters—corresponding\n\n\napproximately to villages in rural areas and towns in urban areas—randomly assigned to the\n\n\ntreatment group. The script included information about barriers that migrants might face\n\nwhen attempting to claim ration at a non-designated shop. <sup>2</sup> This information was based\n\n\non pre-experimental research we conducted into the program’s implementation frictions, in\n\ncluding sending migrant mystery shoppers to test the ONORC program and phone surveys\n\n\nwith PDS shop owners. We also provided treated households with a toll-free phone number\n\n\nthey could call to reach an information hotline which we staffed for five months. Our staff\n\n\nwere trained to answer questions about the ONORC program or to find ration shops across\n\n\n29 states using a database we created of ration shop addresses and phone numbers.\n\n\nThe information we provided immediately increased beliefs about ration portability\n\n\nwithin district, across district, and across state lines, indicating that many households were\n\n\nunaware of the ONORC program. Four months later, treated households had significantly\n\n\n_lower_ beliefs about ration portability compared to control-group households. We show that\n\n\nconcurrent awareness campaigns by state governments—which informed households about\n\n\nthe ONORC scheme but did not provide the same information about barriers to access\n\ning it—explain the higher beliefs about ration portability in the control group four months\n\nafter our experiment. <sup>3</sup> To measure the intensity of government awareness campaigns, we\n\n\nexploit additional, out-of-sample data on beliefs about ration portability which\n\nMention: \"CMIE\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The institute administers the panel three times a year.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000880:4:0", "occurrence_id": "39f9a748258aa0d32e012670", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ration in a PDS shop other than their home shop. Only 8% believed they could claim ration\n\n\noutside their home state.\n\n\nWe embedded an information experiment into a panel survey, the Consumer Pyramid\n\n\nHousehold Survey (CPHS), carried out three times a year by the Center for Monitoring\n\n\nIndian Economy (CMIE). During the survey wave from September–December 2021, enumer\n\nators read a script explaining the ONORC scheme to households in clusters—corresponding\n\n\napproximately to villages in rural areas and towns in urban areas—randomly assigned to the\n\n\ntreatment group. The script included information about barriers that migrants might face\n\nwhen attempting to claim ration at a non-designated shop. <sup>2</sup> This information was based\n\n\non pre-experimental research we conducted into the program’s implementation frictions, in\n\ncluding sending migrant mystery shoppers to test the ONORC program and phone surveys\n\n\nwith PDS shop owners. We also provided treated households with a toll-free phone number\n\n\nthey could call to reach an information hotline which we staffed for five months. Our staff\n\n\nwere trained to answer questions about the ONORC program or to find ration shops across\n\n\n29 states using a database we created of ration shop addresses and phone numbers.\n\n\nThe information we provided immediately increased beliefs about ration portability\n\n\nwithin district, across district, and across state lines, indicating that many households were\n\n\nunaware of the ONORC program. Four months later, treated households had significantly\n\n\n_lower_ beliefs about ration portability compared to control-group households. We show that\n\n\nconcurrent awareness campaigns by state governments—which informed households about\n\n\nthe ONORC scheme but did not provide the same information about barriers to access\n\ning it—explain the higher beliefs about ration portability in the control group four months\n\nafter our experiment. <sup>3</sup> To measure the intensity of government awareness campaigns, we\n\n\nexploit additional, out-of-sample data on beliefs about ration portability which\n\nMention: \"database we created of ration shop addresses and phone numbers\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"The authors compiled shop contact details so hotline staff could answer portability queries across 29 states.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000881:9:0", "occurrence_id": "04500d0530609fb6e5db20b1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: within a household. Consequently, this lack of data can pose a serious challenge for researchers\n\n\nand stakeholders in assessing and monitoring poverty and health for individuals across the world.\n\n\nMost household consumption and dietary data can be classified as incomplete in one of three ways:\n\n\n(1) incomplete household survey data as either completely missing individual consumption data;\n\n\n(2) partially missing such data, as in a case in which the data is only available for certain\n\n\nadministrative areas; or (3) missing such data over time (Dang et al., 2019). The NNS does not\n\n\nexhibit incompleteness in this way and provides an ideal opportunity to investigate questions of\n\n\nindividual food consumption and expenditure within a household.\n\n\nIn our analysis, we use data from 2013 (hereafter NNS 2013), which includes the most recent and\n\n\npublicly available household and individual food consumption data. The NNS 2013 includes 10\n\n\nmodules: (1) Anthropometry, (2) Biochemical, (3) Clinical, (4) Dietary – Household, (5) Dietary\n\n\n- Individual, (6) Food Security, (7) Infant and Young Children Feeding, (8) Maternal Health and\n\n\nNutrition, (9) Socio-Economic – Household, and (10) Socio-Economic – Individual. Note that the\n\n\nDietary – Household module applied standard food weighting procedures and collected data on all\n\n\nmeals (breakfast, lunch, dinner, and snacks) prepared and cooked in the house for all the household\n\n\nmembers, whereas the Dietary – Individual module conducted face-to-face interviews using a 24\n\nhour Food Recall Form to collect food consumption at the individual level. Although more rounds\n\n\nof the NNS (such as the NNS 2015 and NNS 2018) were conducted after 2013, these more recent\n\n\nNNS are not ideal for this study. First, the NNS 2015 did not implement food consumption modules\n\n\nand second, the NNS 2018 was not publicly available at the time at which this work was\n\n\nundertaken\n\nMention: \"NNS 2013\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its ten modules provide the only recent public individual-level food consumption data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000881:9:0", "occurrence_id": "32eef8519879128de9ee53f9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: within a household. Consequently, this lack of data can pose a serious challenge for researchers\n\n\nand stakeholders in assessing and monitoring poverty and health for individuals across the world.\n\n\nMost household consumption and dietary data can be classified as incomplete in one of three ways:\n\n\n(1) incomplete household survey data as either completely missing individual consumption data;\n\n\n(2) partially missing such data, as in a case in which the data is only available for certain\n\n\nadministrative areas; or (3) missing such data over time (Dang et al., 2019). The NNS does not\n\n\nexhibit incompleteness in this way and provides an ideal opportunity to investigate questions of\n\n\nindividual food consumption and expenditure within a household.\n\n\nIn our analysis, we use data from 2013 (hereafter NNS 2013), which includes the most recent and\n\n\npublicly available household and individual food consumption data. The NNS 2013 includes 10\n\n\nmodules: (1) Anthropometry, (2) Biochemical, (3) Clinical, (4) Dietary – Household, (5) Dietary\n\n\n- Individual, (6) Food Security, (7) Infant and Young Children Feeding, (8) Maternal Health and\n\n\nNutrition, (9) Socio-Economic – Household, and (10) Socio-Economic – Individual. Note that the\n\n\nDietary – Household module applied standard food weighting procedures and collected data on all\n\n\nmeals (breakfast, lunch, dinner, and snacks) prepared and cooked in the house for all the household\n\n\nmembers, whereas the Dietary – Individual module conducted face-to-face interviews using a 24\n\nhour Food Recall Form to collect food consumption at the individual level. Although more rounds\n\n\nof the NNS (such as the NNS 2015 and NNS 2018) were conducted after 2013, these more recent\n\n\nNNS are not ideal for this study. First, the NNS 2015 did not implement food consumption modules\n\n\nand second, the NNS 2018 was not publicly available at the time at which this work was\n\n\nundertaken\n\nMention: \"NNS 2015\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"It is excluded because it did not field the food consumption modules.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000881:9:0", "occurrence_id": "e96362988aee3ac06adadc60", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: within a household. Consequently, this lack of data can pose a serious challenge for researchers\n\n\nand stakeholders in assessing and monitoring poverty and health for individuals across the world.\n\n\nMost household consumption and dietary data can be classified as incomplete in one of three ways:\n\n\n(1) incomplete household survey data as either completely missing individual consumption data;\n\n\n(2) partially missing such data, as in a case in which the data is only available for certain\n\n\nadministrative areas; or (3) missing such data over time (Dang et al., 2019). The NNS does not\n\n\nexhibit incompleteness in this way and provides an ideal opportunity to investigate questions of\n\n\nindividual food consumption and expenditure within a household.\n\n\nIn our analysis, we use data from 2013 (hereafter NNS 2013), which includes the most recent and\n\n\npublicly available household and individual food consumption data. The NNS 2013 includes 10\n\n\nmodules: (1) Anthropometry, (2) Biochemical, (3) Clinical, (4) Dietary – Household, (5) Dietary\n\n\n- Individual, (6) Food Security, (7) Infant and Young Children Feeding, (8) Maternal Health and\n\n\nNutrition, (9) Socio-Economic – Household, and (10) Socio-Economic – Individual. Note that the\n\n\nDietary – Household module applied standard food weighting procedures and collected data on all\n\n\nmeals (breakfast, lunch, dinner, and snacks) prepared and cooked in the house for all the household\n\n\nmembers, whereas the Dietary – Individual module conducted face-to-face interviews using a 24\n\nhour Food Recall Form to collect food consumption at the individual level. Although more rounds\n\n\nof the NNS (such as the NNS 2015 and NNS 2018) were conducted after 2013, these more recent\n\n\nNNS are not ideal for this study. First, the NNS 2015 did not implement food consumption modules\n\n\nand second, the NNS 2018 was not publicly available at the time at which this work was\n\n\nundertaken\n\nMention: \"NNS 2018\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Access, not content, is the obstacle: the round was not public when the work was done.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000885:13:0", "occurrence_id": "ea243d358ba22af13d4e00bd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Chod et al. (2019) offers a theory of trade credit and supporting empirical evidence that starts\n\n\nfrom the proposition that, in the presence of multiple suppliers, the benefit of providing trade\n\n\ncredit is not fully internalized by a trade creditor. After obtaining trade credit, a retailer can use\n\n\nfreed-up liquidity to buy more goods from other suppliers as well as the one who provided the\n\n\ncredit. This leads to a free rider problem because a given supplier bears the full cost of providing\n\n\ntrade credit, but the benefits of larger spending by the retailer are shared among all suppliers.\n\n\nThus, Chod et al. (2019) hypothesize that a supplier that is responsible for a larger share of the\n\n\nretailer’s purchases internalizes a larger part of the benefit and is therefore willing to offer a\n\n\nlarger proportion of its goods on credit. This is similar to the equity stake idea from Petersen and\n\n\nRajan (1997) except that it emphasizes the importance of the supplier’s position among firms\n\n\nselling to the _same customer_ rather than its competitive position (market power) among all firms\n\n\nin its industry.\n\n\nUsing annual data from Compustat, complemented with the Compustat Industry Segment\n\n\ndata set, which identifies firms’ principal customers and is used to identify customer-supplier\n\n\nrelationships (following the method in Cohen and Frazzini, 2008), for a sample of U.S. retailers\n\n\nfrom 1996 to 1999, Chod et al. (2019) find a significant positive relationship between (a) the\n\n\namount of trade credit extended by a supplier and its average share of retailers’ purchases and (b)\n\n\nthe amount of trade credit that a retailer receives and the concentration in its suppliers, as\n\n\nreflected in the Herfindahl index of supplier shares of the retailer’s purchases. These findings\n\n\nprovide support for a modified version of the Petersen/Rajan equity stake hypothesis, which\n\n\nhighlights the importance of a buyer’s retail platform for a supplier in relation to that of other\n\n\ncompeting\n\nMention: \"annual data from Compustat\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"A cited study uses these annual firm financials to test its trade credit theory on US retailers.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000885:13:0", "occurrence_id": "380df3332ed4ccf9183f1844", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Chod et al. (2019) offers a theory of trade credit and supporting empirical evidence that starts\n\n\nfrom the proposition that, in the presence of multiple suppliers, the benefit of providing trade\n\n\ncredit is not fully internalized by a trade creditor. After obtaining trade credit, a retailer can use\n\n\nfreed-up liquidity to buy more goods from other suppliers as well as the one who provided the\n\n\ncredit. This leads to a free rider problem because a given supplier bears the full cost of providing\n\n\ntrade credit, but the benefits of larger spending by the retailer are shared among all suppliers.\n\n\nThus, Chod et al. (2019) hypothesize that a supplier that is responsible for a larger share of the\n\n\nretailer’s purchases internalizes a larger part of the benefit and is therefore willing to offer a\n\n\nlarger proportion of its goods on credit. This is similar to the equity stake idea from Petersen and\n\n\nRajan (1997) except that it emphasizes the importance of the supplier’s position among firms\n\n\nselling to the _same customer_ rather than its competitive position (market power) among all firms\n\n\nin its industry.\n\n\nUsing annual data from Compustat, complemented with the Compustat Industry Segment\n\n\ndata set, which identifies firms’ principal customers and is used to identify customer-supplier\n\n\nrelationships (following the method in Cohen and Frazzini, 2008), for a sample of U.S. retailers\n\n\nfrom 1996 to 1999, Chod et al. (2019) find a significant positive relationship between (a) the\n\n\namount of trade credit extended by a supplier and its average share of retailers’ purchases and (b)\n\n\nthe amount of trade credit that a retailer receives and the concentration in its suppliers, as\n\n\nreflected in the Herfindahl index of supplier shares of the retailer’s purchases. These findings\n\n\nprovide support for a modified version of the Petersen/Rajan equity stake hypothesis, which\n\n\nhighlights the importance of a buyer’s retail platform for a supplier in relation to that of other\n\n\ncompeting\n\nMention: \"Compustat Industry Segment\n\n\ndata set\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"Its principal customer identification lets supplier-customer links be constructed.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000888:9:0", "occurrence_id": "417d47094cbec39241448b74", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The ambitious infrastructure development program that the government embarked on around\n\n\n2014–18 significantly contributed to construction sector growth. However, most construction\n\n\nprojects had to be stopped due to the rising budget deficits and increased debt service\n\n\nobligations just before the onset of the pandemic.\n\n\n**Figure** **1:** **Real** **GDP** **growth** **by** **sector** **in** **Zambia** **from** **2011-20**\n\n\n_Source:_ _authors’_ _compilation_ _based_ _on_ _World_ _Bank_ _(2022)._\n\n\nTo get a sense of the firm performance just before the COVID-19 pandemic in Zambia, we\n\n\nuse employment and sales data from the World Bank ES conducted from September 2019 to\n\n\nMarch 2020. While the survey excluded some sectors, such as mining and agriculture, most\n\n\nof Zambia’s largest economic sectors (wholesale and retail, manufacturing, and construction)\n\n\nwere covered, as were rapidly growing sectors such as tourism and ICT. According to the\n\n\nWorld Bank (2020), Zambia’s annual private sector employment grew by an average of 3\n\n\npercent, which was lower than the Sub-Saharan African average employment growth rate of\n\n\n7 percent. Real sales in Zambia declined by an average of 2 percent per annum, while the\n\n\n7\n\nMention: \"World Bank ES\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Employment and sales just before the pandemic are taken from this survey round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000891:15:0", "occurrence_id": "25854aab8cd820e6a44ebba5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: because privacy protections tend to help build trust in digital transactions. The former variable’s\n\n\neffect is ambiguous, at least in the short run. On the one hand, such laws require that digital\n\n\nsignatures have equal recognition under the law as properly signed paper contracts. On the other\n\n\nhand, such regulations require authentication procedures to ensure that parties involved in a\n\n\ntransaction are real, which can deter the use of digital payments in contexts where authentication\n\n\ncomes at the expense of privacy rights. <sup>14</sup> [^14: For additional details on the construction of these indexes, see World Bank (2021, Chapter 6).]\n\n\nFinally, _Capabilitiesi2019_ is included to control for human capabilities that are positively\n\n\nlinked to DFS usage as reflected in educational attainment. In the regressions below, we rely on\n\n\nmean years of schooling among adults aged 25 or older from the UNDP’s Human Development\n\n\nIndex to proxy for this factor, because of its wide country coverage. Again, results are qualitatively\n\n\nsimilar when the schooling variable is replaced by the full HDI or the World Bank Human Capital\n\n\nIndex.\n\n\nEquation 2 adds to equation 1 the share of the of the population that made or received a\n\n\ndigital payment in 2017 ( _DigPayi,2017_ ), which is from the penultimate version of the Global Findex\n\n\ndataset. It is included to control for the level of digital payment usage in a country prior to the\n\n\npandemic. <sup>15</sup> [^15: Note that the Findex data were not collected in 2018 or 2019, so 2017 is the last date prior to the pandemic for which\nindicators of DFS usage are available for a wide sample of countries via Findex.] Recall that our estimation approach for uncovering causal relationships is rooted in\n\n\nthe assumption that the COVID-19 shock was unanticipated, and that its effects on digital\n\n\nparticipation depended on pre-existing conditions. For this reason, the explanatory variables are\n\n\nmeasured in 2019 to the extent possible. By controlling for the level of digital payment usage just\n\n\nprior\n\nMention: \"Human Development\n\n\nIndex\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Adult mean years of schooling from this index proxies human capabilities.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000891:15:0", "occurrence_id": "ddff3c6b70b3e6a13274eedc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: because privacy protections tend to help build trust in digital transactions. The former variable’s\n\n\neffect is ambiguous, at least in the short run. On the one hand, such laws require that digital\n\n\nsignatures have equal recognition under the law as properly signed paper contracts. On the other\n\n\nhand, such regulations require authentication procedures to ensure that parties involved in a\n\n\ntransaction are real, which can deter the use of digital payments in contexts where authentication\n\n\ncomes at the expense of privacy rights. <sup>14</sup> [^14: For additional details on the construction of these indexes, see World Bank (2021, Chapter 6).]\n\n\nFinally, _Capabilitiesi2019_ is included to control for human capabilities that are positively\n\n\nlinked to DFS usage as reflected in educational attainment. In the regressions below, we rely on\n\n\nmean years of schooling among adults aged 25 or older from the UNDP’s Human Development\n\n\nIndex to proxy for this factor, because of its wide country coverage. Again, results are qualitatively\n\n\nsimilar when the schooling variable is replaced by the full HDI or the World Bank Human Capital\n\n\nIndex.\n\n\nEquation 2 adds to equation 1 the share of the of the population that made or received a\n\n\ndigital payment in 2017 ( _DigPayi,2017_ ), which is from the penultimate version of the Global Findex\n\n\ndataset. It is included to control for the level of digital payment usage in a country prior to the\n\n\npandemic. <sup>15</sup> [^15: Note that the Findex data were not collected in 2018 or 2019, so 2017 is the last date prior to the pandemic for which\nindicators of DFS usage are available for a wide sample of countries via Findex.] Recall that our estimation approach for uncovering causal relationships is rooted in\n\n\nthe assumption that the COVID-19 shock was unanticipated, and that its effects on digital\n\n\nparticipation depended on pre-existing conditions. For this reason, the explanatory variables are\n\n\nmeasured in 2019 to the extent possible. By controlling for the level of digital payment usage just\n\n\nprior\n\nMention: \"World Bank Human Capital\n\n\nIndex\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Substituting this index for schooling leaves results qualitatively unchanged.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000891:15:0", "occurrence_id": "624316ebb777b45308e9b121", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: because privacy protections tend to help build trust in digital transactions. The former variable’s\n\n\neffect is ambiguous, at least in the short run. On the one hand, such laws require that digital\n\n\nsignatures have equal recognition under the law as properly signed paper contracts. On the other\n\n\nhand, such regulations require authentication procedures to ensure that parties involved in a\n\n\ntransaction are real, which can deter the use of digital payments in contexts where authentication\n\n\ncomes at the expense of privacy rights. <sup>14</sup> [^14: For additional details on the construction of these indexes, see World Bank (2021, Chapter 6).]\n\n\nFinally, _Capabilitiesi2019_ is included to control for human capabilities that are positively\n\n\nlinked to DFS usage as reflected in educational attainment. In the regressions below, we rely on\n\n\nmean years of schooling among adults aged 25 or older from the UNDP’s Human Development\n\n\nIndex to proxy for this factor, because of its wide country coverage. Again, results are qualitatively\n\n\nsimilar when the schooling variable is replaced by the full HDI or the World Bank Human Capital\n\n\nIndex.\n\n\nEquation 2 adds to equation 1 the share of the of the population that made or received a\n\n\ndigital payment in 2017 ( _DigPayi,2017_ ), which is from the penultimate version of the Global Findex\n\n\ndataset. It is included to control for the level of digital payment usage in a country prior to the\n\n\npandemic. <sup>15</sup> [^15: Note that the Findex data were not collected in 2018 or 2019, so 2017 is the last date prior to the pandemic for which\nindicators of DFS usage are available for a wide sample of countries via Findex.] Recall that our estimation approach for uncovering causal relationships is rooted in\n\n\nthe assumption that the COVID-19 shock was unanticipated, and that its effects on digital\n\n\nparticipation depended on pre-existing conditions. For this reason, the explanatory variables are\n\n\nmeasured in 2019 to the extent possible. By controlling for the level of digital payment usage just\n\n\nprior\n\nMention: \"Findex\n\n\ndataset\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The share of the population making or receiving a digital payment is taken from this earlier wave.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000891:15:0", "occurrence_id": "e2061c9d2b6e1afa784e098c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: because privacy protections tend to help build trust in digital transactions. The former variable’s\n\n\neffect is ambiguous, at least in the short run. On the one hand, such laws require that digital\n\n\nsignatures have equal recognition under the law as properly signed paper contracts. On the other\n\n\nhand, such regulations require authentication procedures to ensure that parties involved in a\n\n\ntransaction are real, which can deter the use of digital payments in contexts where authentication\n\n\ncomes at the expense of privacy rights. <sup>14</sup> [^14: For additional details on the construction of these indexes, see World Bank (2021, Chapter 6).]\n\n\nFinally, _Capabilitiesi2019_ is included to control for human capabilities that are positively\n\n\nlinked to DFS usage as reflected in educational attainment. In the regressions below, we rely on\n\n\nmean years of schooling among adults aged 25 or older from the UNDP’s Human Development\n\n\nIndex to proxy for this factor, because of its wide country coverage. Again, results are qualitatively\n\n\nsimilar when the schooling variable is replaced by the full HDI or the World Bank Human Capital\n\n\nIndex.\n\n\nEquation 2 adds to equation 1 the share of the of the population that made or received a\n\n\ndigital payment in 2017 ( _DigPayi,2017_ ), which is from the penultimate version of the Global Findex\n\n\ndataset. It is included to control for the level of digital payment usage in a country prior to the\n\n\npandemic. <sup>15</sup> [^15: Note that the Findex data were not collected in 2018 or 2019, so 2017 is the last date prior to the pandemic for which\nindicators of DFS usage are available for a wide sample of countries via Findex.] Recall that our estimation approach for uncovering causal relationships is rooted in\n\n\nthe assumption that the COVID-19 shock was unanticipated, and that its effects on digital\n\n\nparticipation depended on pre-existing conditions. For this reason, the explanatory variables are\n\n\nmeasured in 2019 to the extent possible. By controlling for the level of digital payment usage just\n\n\nprior\n\nMention: \"Findex\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"No waves were fielded in 2018 or 2019, so 2017 is the last comparable pre-pandemic year.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000872:9:0", "occurrence_id": "84efce9e19c4ad82b0b01755", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 1 presents the distribution of violent conflicts, specifically battles, across all regions\n\nof Ethiopia before (August 2019-October 2020) and after the outbreak of the war (November 2020\nMay 2021). These figures are based on conflict event records in the Armed Conflict Location and\n\nEvent Data (ACLED). Panel (a) shows that battles were sparsely and evenly distributed across\n\nregions before November 2020, except in Tigray where there were no major battles recorded\n\nduring that time. Panel (b) confirms that during the November 2020 - May 2021 period, there was\n\na dramatic spike in battle events, the vast majority of which were confined to Tigray, with little\n\nchange in battle incidents elsewhere. This figure further highlights that these battles covered all\n\nparts of Tigray. This is not surprising given the number of actors in the war and battle fronts. In\n\nsouthern Tigray the armed conflict was between Tigray regional forces and the federal army and\n\nallied forces. In the north it was between Tigray regional forces and the Eritrean army. In western\n\nTigray, the war involved the federal army (along with their allied Amhara regional forces and\n\nEritrean army) and Tigray regional forces. Thus, most of Tigray was effectively an active war\n\nzone.\n\n\n**Figure 1: Spatial distribution of violent conflicts before and after the outbreak of the war**\n\n\n8\n\nMention: \"Armed Conflict Location and\n\nEvent Data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Battle event records are mapped across regions before and after the war began.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000895:13:1", "occurrence_id": "7e508a04b46ea71fee340a25", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  to the sectors that are used in the trade data, i.e. ITPDE and WIOD.\n\n\nWe use the International Trade and Production Database for Estimation (ITPDE, Release 2)\n\nput together by Borchert et al. (2020) for our baseline results for three reasons: first, to estimate\nEquation 4 theory-consistent and to be able to include trends for globalization, we need data\non international and domestic trade flows (Yotov 2021). In contrast to other standard sources\nfor trade data such as UN Comtrade or CEPII’s Baci (Gaulier and Zignago 2010), the ITPDE\ncovers both, inter- and intranational trade. Second, deep trade agreements typically cover\ntrade in both goods and services (Mattoo, Rocha, et al. 2020) while tariffs are only applicable\nfor trade in goods. It could be possible that trade agreements do not reduce NTBs for goods\ntrade but do so for services. To get a full picture of the trade-creating effect of NTBs, we\nwould therefore like to include also services. Lastly, the ITPDE covers the whole world while\n\nthe comparable WIOD data only report trade for 44 countries (Timmer et al. 2015). The ITPDE\n\n\n5 [WITS is available at https://wits.worldbank.org/.](https://wits.worldbank.org/)\n\n\n11\n\nMention: \"ITPDE\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its worldwide coverage of international and domestic trade flows makes it the baseline dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000895:13:1", "occurrence_id": "29e21c000c94ff67dbd876de", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  to the sectors that are used in the trade data, i.e. ITPDE and WIOD.\n\n\nWe use the International Trade and Production Database for Estimation (ITPDE, Release 2)\n\nput together by Borchert et al. (2020) for our baseline results for three reasons: first, to estimate\nEquation 4 theory-consistent and to be able to include trends for globalization, we need data\non international and domestic trade flows (Yotov 2021). In contrast to other standard sources\nfor trade data such as UN Comtrade or CEPII’s Baci (Gaulier and Zignago 2010), the ITPDE\ncovers both, inter- and intranational trade. Second, deep trade agreements typically cover\ntrade in both goods and services (Mattoo, Rocha, et al. 2020) while tariffs are only applicable\nfor trade in goods. It could be possible that trade agreements do not reduce NTBs for goods\ntrade but do so for services. To get a full picture of the trade-creating effect of NTBs, we\nwould therefore like to include also services. Lastly, the ITPDE covers the whole world while\n\nthe comparable WIOD data only report trade for 44 countries (Timmer et al. 2015). The ITPDE\n\n\n5 [WITS is available at https://wits.worldbank.org/.](https://wits.worldbank.org/)\n\n\n11\n\nMention: \"WIOD data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"The comparable input-output source covers only 44 countries, making it less suitable.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000895:13:1", "occurrence_id": "bd33122c8408a560158c7c16", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  to the sectors that are used in the trade data, i.e. ITPDE and WIOD.\n\n\nWe use the International Trade and Production Database for Estimation (ITPDE, Release 2)\n\nput together by Borchert et al. (2020) for our baseline results for three reasons: first, to estimate\nEquation 4 theory-consistent and to be able to include trends for globalization, we need data\non international and domestic trade flows (Yotov 2021). In contrast to other standard sources\nfor trade data such as UN Comtrade or CEPII’s Baci (Gaulier and Zignago 2010), the ITPDE\ncovers both, inter- and intranational trade. Second, deep trade agreements typically cover\ntrade in both goods and services (Mattoo, Rocha, et al. 2020) while tariffs are only applicable\nfor trade in goods. It could be possible that trade agreements do not reduce NTBs for goods\ntrade but do so for services. To get a full picture of the trade-creating effect of NTBs, we\nwould therefore like to include also services. Lastly, the ITPDE covers the whole world while\n\nthe comparable WIOD data only report trade for 44 countries (Timmer et al. 2015). The ITPDE\n\n\n5 [WITS is available at https://wits.worldbank.org/.](https://wits.worldbank.org/)\n\n\n11\n\nMention: \"ITPDE\n\n\n5\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"The mention anchors the footnote giving the public access location of the database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000897:21:0", "occurrence_id": "b9eaf61e607a04076e84ad92", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ### **7 Policy Implications**\n\nOur paper studies the effect of weather shocks in 2018 on consumption in Afghanistan,\n\na country affected by conflict that is also witnessing extreme weather events driven by\n\nclimate change. We combine a geo-coded cross-sectional household survey (IE-LFS 2019\n20) with satellite-based precipitation, temperature, and vegetation measures. The 2018\n\ndrought measured by positive temperature anomalies and negative vegetation anomalies\n\ndecreased consumption and increased poverty rates.\n\n\nWorking in a data scarce environment, we are able to deliver important insights on the\n\nrelationship between weather shocks and welfare in a conflict affected country, show\ncasing techniques that can be used in similar context where only cross-sectional data\n\nis available. We find evidence that households decreased non-food consumption to save\n\nthemselves from food insecurity. While both food and non-food consumption expendi\ntures decreased with adverse vegetation anomalies, households were able to maintain their\n\nfood consumption by reducing non-food monthly per capita consumption when faced with\n\nbad temperature anomalies.\n\n\nIn terms of resilience, households owning land could mitigate the adverse effects of\n\ndroughts, whereas livestock ownership could not protect households against welfare losses\n\ndue to severe droughts. This confirms that asset ownership may offer a means of insuring\n\nagainst weather shocks by allowing households to either sell or borrow against the as\nset; however, this coping mechanism is often imperfect (Dercon, 2002; Fafchamps et al.,\n\n1998). Given the ongoing conflict at the time in Afghanistan, we also find suggestive\n\nevidence that consumption was disproportionately lower in Taliban districts in 2018 that\n\nexperienced bad vegetation anomalies.\n\n\nIn terms of vulnerability, while 47 percent of households were classified as poor in the\n\nsurvey, a simulation exercise based on historical weather data suggests that 62.5 percent of\n\nindividuals have a greater than 1 in 4 chance of falling into poverty due to weather shocks.\n\nThis is likely a lower bound, given that the methodology does not account for displaced\n\nhouseholds or adaptive strategies. It\n\nMention: \"IE-LFS 2019\n20\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Geo-coded household consumption is linked to local weather anomalies to estimate shock effects.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000897:21:0", "occurrence_id": "54a33630f990cb17b02ce6e2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ### **7 Policy Implications**\n\nOur paper studies the effect of weather shocks in 2018 on consumption in Afghanistan,\n\na country affected by conflict that is also witnessing extreme weather events driven by\n\nclimate change. We combine a geo-coded cross-sectional household survey (IE-LFS 2019\n20) with satellite-based precipitation, temperature, and vegetation measures. The 2018\n\ndrought measured by positive temperature anomalies and negative vegetation anomalies\n\ndecreased consumption and increased poverty rates.\n\n\nWorking in a data scarce environment, we are able to deliver important insights on the\n\nrelationship between weather shocks and welfare in a conflict affected country, show\ncasing techniques that can be used in similar context where only cross-sectional data\n\nis available. We find evidence that households decreased non-food consumption to save\n\nthemselves from food insecurity. While both food and non-food consumption expendi\ntures decreased with adverse vegetation anomalies, households were able to maintain their\n\nfood consumption by reducing non-food monthly per capita consumption when faced with\n\nbad temperature anomalies.\n\n\nIn terms of resilience, households owning land could mitigate the adverse effects of\n\ndroughts, whereas livestock ownership could not protect households against welfare losses\n\ndue to severe droughts. This confirms that asset ownership may offer a means of insuring\n\nagainst weather shocks by allowing households to either sell or borrow against the as\nset; however, this coping mechanism is often imperfect (Dercon, 2002; Fafchamps et al.,\n\n1998). Given the ongoing conflict at the time in Afghanistan, we also find suggestive\n\nevidence that consumption was disproportionately lower in Taliban districts in 2018 that\n\nexperienced bad vegetation anomalies.\n\n\nIn terms of vulnerability, while 47 percent of households were classified as poor in the\n\nsurvey, a simulation exercise based on historical weather data suggests that 62.5 percent of\n\nindividuals have a greater than 1 in 4 chance of falling into poverty due to weather shocks.\n\nThis is likely a lower bound, given that the methodology does not account for displaced\n\nhouseholds or adaptive strategies. It\n\nMention: \"satellite-based precipitation\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Precipitation, temperature and vegetation measures define the drought exposure.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000903:25:2", "occurrence_id": "aef14ab7194bef905d802d0b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: >0.105</u>\n\n\nNotes: This table presents OLS regression results for the impact of the financial education program on\nrepayment delays and credit losses. The outcome variables come from administrative data from the\nCredit Registry System (SCR), housed at the Brazilian Central Bank (BCB) and are defined as follows: an\nindicator variable equal to 1 if the student's longest repayment delay is between one day and one year\n(columns 1 and 3); an indicator variable equal to 1 if the student's longest repayment delay is greater\nthan a year, when credit contracts get written off as a loss (columns 2 and 4). Columns 1 and 2 consider\nall credit contracts in the system since June 2016, including previously realized losses which stay in the\nsystem. For columns 3 and 4, we re-compute the outcome variables after dropping credit contracts that\nwere already a loss in June 2016. Panel A presents regressions of the outcome on a treatment dummy,\ncontrolling for state by month fixed effects. Panel B controls additionally for school pair dummies, and\npanel C controls further for student gender. Panel D breaks the treatment effects into two time periods:\n2016 to 2018, when students may still be in university, and 2019 and 2020, when most students have\nentered the labor market, using all controls. 2020 data includes only January and February (pre COVID-19\npandemic in Brazil). Robust standard errors, clustered at the school level, are in parentheses. *** p<0.01,\n** p<0.05, * p<0.1.\n\n\n24\n\nMention: \"Credit Registry System\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Repayment delays and credit losses of programme participants are read from this administrative registry.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000905:12:0", "occurrence_id": "a01a5fcff3ff58664b5bdef9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Unfortunately, the input-output data for the Central American economies are not usually available.\n\nBarring Costa Rica, for which the input-output tables are available in the OECD database (OECD,\n\n2017), I couldn’t find such data for the other countries. To overcome this challenge, I use the average\n\nintermediate shares from eight regional countries (including Costa Rica). In the data appendix, I show\n\nthat these shares of regional economies lie within close range. Therefore, their average provides a\n\nreasonable benchmark for the Central American countries.\n\nTable 1 reports the country-wise coverage of cleaned data obtained from various sources. Column (6)\n\nshows the effective period of analysis for each country, which corresponds to the overlapping years\n\nin columns (2)–(5).\n\n\n\n**<u>Table 1: Country-wise time coverage</u>**\n\n\n\n**<u>Trade</u>**\n**shares** **Sectoral trade**\n\n\n\n**<u>Effective</u>**\n\n\n\n**Series** **Employment**\n\n\n\n**Employment** **added** **shares** **Sectoral trade** **period**\n\n**<u>shares</u>** **<u>shares</u>** **<u>(aggregate)</u>** **<u>shares</u>** **<u>of coverage</u>**\n\n\n\n**<u>Value</u>**\n**added**\n\n\n\n**<u>Atlas of</u>**\n\n**Source** **ILO** **UNSD** **UNSD** **economic**\n\n**complexity**\n<u>(1)</u> <u>\n\nMention: \"OECD database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its input-output table for one economy supplies the regional intermediate shares used elsewhere.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000906:16:1", "occurrence_id": "175584282bf23675d861635e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  *** p<0.01. Standard errors clustered at the ‘grappe’ level in parentheses. Dependent\nvariable is the log of total monetary consumption per capita. Individual-level sampling weights applied. All\nspecifications include controls for the gender, age, and education level of the household head, household size, the\ndependency ratio, household walls material, household roof material, household floor material, access to\nelectricity, access to improved drinking water, and access to improved sanitation as well as an urban-rural dummy\nand region fixed effects. Source: EHCVM and World Bank estimates.\n\n###### 4.2. Disaggregated differences in consumption\n\nWe now disaggregate the results by different parts of the consumption basket to characterize more fully\nthe deprivation implied by the overall changes in consumption and poverty described in Section 4.1. We\nbegin by breaking down the results by food and non-food consumption; that is, we re-estimate Equation 1\nwith the log of food consumption per capita and the log non-food consumption per capita as the main\nregressors. Pooling the data from Burkina Faso, Niger, and Senegal, food consumption was 11.4 percent\nlower in the lean season wave while non-food consumption was 7.4 percent lower (Table 4). The\ndifference between these two effects is even larger at 11.3 percent for food consumption and 6.3 percent\nfor non-food consumption when full controls are included. <sup>23</sup> [^23: We use Stata’s `suest` command to formally test the difference between the effects of seasonality on food and\nnon-food consumption. When pooling Burkina Faso, Niger, and Senegal, the p-value on the formal test for the\ndifference between the food and non-food coefficients is 0.1190 without controls and 0.0003 with controls.] This in itself suggests that seasonality may\nhave more effect on food consumption. Nevertheless, these results may partly be an artifact of the way\nin which the EHVCM data were collected; the recall period for the consumption of\n\nMention: \"EHCVM\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Housing assets and access to water, sanitation and electricity are taken from this household survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000906:16:1", "occurrence_id": "64395c105a5359ffa78adbd2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  *** p<0.01. Standard errors clustered at the ‘grappe’ level in parentheses. Dependent\nvariable is the log of total monetary consumption per capita. Individual-level sampling weights applied. All\nspecifications include controls for the gender, age, and education level of the household head, household size, the\ndependency ratio, household walls material, household roof material, household floor material, access to\nelectricity, access to improved drinking water, and access to improved sanitation as well as an urban-rural dummy\nand region fixed effects. Source: EHCVM and World Bank estimates.\n\n###### 4.2. Disaggregated differences in consumption\n\nWe now disaggregate the results by different parts of the consumption basket to characterize more fully\nthe deprivation implied by the overall changes in consumption and poverty described in Section 4.1. We\nbegin by breaking down the results by food and non-food consumption; that is, we re-estimate Equation 1\nwith the log of food consumption per capita and the log non-food consumption per capita as the main\nregressors. Pooling the data from Burkina Faso, Niger, and Senegal, food consumption was 11.4 percent\nlower in the lean season wave while non-food consumption was 7.4 percent lower (Table 4). The\ndifference between these two effects is even larger at 11.3 percent for food consumption and 6.3 percent\nfor non-food consumption when full controls are included. <sup>23</sup> [^23: We use Stata’s `suest` command to formally test the difference between the effects of seasonality on food and\nnon-food consumption. When pooling Burkina Faso, Niger, and Senegal, the p-value on the formal test for the\ndifference between the food and non-food coefficients is 0.1190 without controls and 0.0003 with controls.] This in itself suggests that seasonality may\nhave more effect on food consumption. Nevertheless, these results may partly be an artifact of the way\nin which the EHVCM data were collected; the recall period for the consumption of\n\nMention: \"EHVCM data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Seasonality results may reflect the recall periods used in collection rather than true behaviour.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000907:0:0", "occurrence_id": "b046f3c53bfa29ef889bac3f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Policy Research Working Paper\n10165\nInferring COVID-19 Vaccine Attitudes \nfrom Twitter Data\nAn Application to the Arabic Speaking World\nRoy van der Weide\nDevelopment Economics \nDevelopment Research Group\nSeptember 2022 \nPublic Disclosure Authorized\nPublic Disclosure Authorized\nPublic Disclosure Authorized\nPublic Disclosure Authorized\n\nMention: \"Twitter Data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Public posts are the input for inferring vaccine attitudes across the Arabic-speaking world.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000913:21:0", "occurrence_id": "20752ac4bed48ef144576a67", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ### 6 Education Financing and Impact on Learning\n###### 6.1 Officials in the EAP region overestimate public funding for education.\n\nOverall, respondents from EAP countries believe that education financing levels are higher than they\nactually are, see Figure 16. This overestimation of education financing in EAP contrasts with other regions,\nwhere officials tend to underestimate education spending relative to actual spending levels. This finding\nremains consistent even when the education budget figure is taken into account instead of actual\nspending (World Bank, 2023). When asked about their preferred percentage of the total government\nbudget to be spent on education, EAP respondents consistently provided estimates that exceeded their\nown perceptions of the current spending share allocated to education.\n\n\n_Figure 15: Share of public funding for education of total public spending_\n\n\n\n\n\n\n\n\n\n\n\nSources: CGD–World Bank survey, 2022. Survey respondents were asked two financing questions: (i) Now thinking about the\noverall government budget, roughly what percentage do you think is spent on education? and (ii) What percentage of the total\n[government budget do you think should be spent on education? Data on actual budget share from UNESCO for 2021 (http://sdg4-](http://sdg4-data.uis.unesco.org/)\n<u>[data.uis.unesco.org/).](http://sdg4-data.uis.unesco.org/)</u>\n\n\nThe magnitude of the misperceptions is larger in EAP than in other countries, and in a different direction.\nOn average, the gap between the estimated share of the budget and the actual share of the budget is 4\npercentage points in surveyed countries in EAP, while the gap is about 1 percentage point in the other\ndirection in non-EAP countries. As a share of education funding, this gap is very large in real terms, as the\n\nMention: \"CGD–World Bank survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Officials' estimates of the education share of public spending are collected in this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000917:37:5", "occurrence_id": "d5cf0478676bd5db98a280cd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 0.03)<br>(0.03)|\n|Constant<br>0.979***<br>1.160***<br> <br>(0.08)<br>(0.08)|-0.294<br>-0.553***<br>(0.21)<br>(0.21)|\n|Observations<br>54,639<br>54,615<br>R-squared<br>0.24<br>0.23|55,039<br>55,014<br>0.05<br>0.04|\n\n\n\n_Source_ : Authors’ calculations based on the 2018 survey “Getting to Know 8th Grade Students.”\n_Note_ : *The omitted category in aspired and expected education is general high school. Conditional correlations are Ordinary Least Squares\nregressions. Yellow and blue highlights show significant correlations at least at the 95 percent level for the variables of interest and covariates,\nrespectively. Grades are standardized at the school level, with a mean of zero and a standard deviation of one. Measures of socioemotional\nskills are standardized for the entire sample. Standard errors are clustered at the school level.\n\n\n_*p < 0.1 **p < 0.05 ***p < 0.01_\n\n\n36\n\nMention: \"Getting to Know 8th Grade Students\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Correlations between aspired and expected education are estimated on this student survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000921:33:0", "occurrence_id": "ab7572f63344151056f6ee92", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Data and methodology\n\n\nThis exercise builds on a similar one conducted in 2020 to examine child poverty using the GMD (Silwal,\nEngilbertsdottir, Cuesta, Newhouse and Stewart, 2020). The analysis contains records from 10.4 million\nindividuals from 147 countries, taken from the 2022 spring version of the GMD with 2019 as a base year.\nThe GMD is a collection of globally harmonized household survey data compiled by the Data for Goals\ngroup of the World Bank’s Poverty and Equity Global Practice. Further details on the background of the\nGMD are given in Castañeda et al. (2016). A unique feature of the GMD is that the welfare aggregates\nare the same as those used to compute the poverty estimates published by PovcalNet and the World\nDevelopment Indicators. These aggregates are based on household per capita income or consumption,\ndepending on the specific concept that is used to measure national poverty in any given country.\n\n\nFor the effects of this exercise, the rescaling of welfare was performed using the Poverty and Inequality\nPlatform (PIP) which contains data on poverty headcounts for each year (2013-2018). For 2020-2022,\nthe data was provided by the Poverty and Shared Prosperity team. A rescaling of the weights was done\nusing the UNDESA population estimates from 2013-2021. In other words, the number of children is\nestimated by applying the proportion of people that are children in the relevant survey to the total\nUNDESA national population projections for each country. This could lead to discrepancies with UNDESA\nestimates of the number of children in a country. For the 2022 estimates, UNDESA population prospects\nwere used.\n\n\nPoverty estimates for 2020, 2021 and 2022 are “nowcasted”—that is, gross domestic product (GDP)\ngrowth rates have been used to forecast household incomes, assuming all households experience equal\nconsumption growth in percentage terms.\n\n\nTrends from 2013 to 2018 are all based on\n\nMention: \"GMD\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Records for 10.4 million individuals across 147 countries provide the household welfare distribution.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000921:33:0", "occurrence_id": "692d9317349d162548e18396", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Data and methodology\n\n\nThis exercise builds on a similar one conducted in 2020 to examine child poverty using the GMD (Silwal,\nEngilbertsdottir, Cuesta, Newhouse and Stewart, 2020). The analysis contains records from 10.4 million\nindividuals from 147 countries, taken from the 2022 spring version of the GMD with 2019 as a base year.\nThe GMD is a collection of globally harmonized household survey data compiled by the Data for Goals\ngroup of the World Bank’s Poverty and Equity Global Practice. Further details on the background of the\nGMD are given in Castañeda et al. (2016). A unique feature of the GMD is that the welfare aggregates\nare the same as those used to compute the poverty estimates published by PovcalNet and the World\nDevelopment Indicators. These aggregates are based on household per capita income or consumption,\ndepending on the specific concept that is used to measure national poverty in any given country.\n\n\nFor the effects of this exercise, the rescaling of welfare was performed using the Poverty and Inequality\nPlatform (PIP) which contains data on poverty headcounts for each year (2013-2018). For 2020-2022,\nthe data was provided by the Poverty and Shared Prosperity team. A rescaling of the weights was done\nusing the UNDESA population estimates from 2013-2021. In other words, the number of children is\nestimated by applying the proportion of people that are children in the relevant survey to the total\nUNDESA national population projections for each country. This could lead to discrepancies with UNDESA\nestimates of the number of children in a country. For the 2022 estimates, UNDESA population prospects\nwere used.\n\n\nPoverty estimates for 2020, 2021 and 2022 are “nowcasted”—that is, gross domestic product (GDP)\ngrowth rates have been used to forecast household incomes, assuming all households experience equal\nconsumption growth in percentage terms.\n\n\nTrends from 2013 to 2018 are all based on\n\nMention: \"World\nDevelopment Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"The microdata's welfare aggregates match those behind these published poverty estimates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000921:33:0", "occurrence_id": "27558969994a3db95f2bf32c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Data and methodology\n\n\nThis exercise builds on a similar one conducted in 2020 to examine child poverty using the GMD (Silwal,\nEngilbertsdottir, Cuesta, Newhouse and Stewart, 2020). The analysis contains records from 10.4 million\nindividuals from 147 countries, taken from the 2022 spring version of the GMD with 2019 as a base year.\nThe GMD is a collection of globally harmonized household survey data compiled by the Data for Goals\ngroup of the World Bank’s Poverty and Equity Global Practice. Further details on the background of the\nGMD are given in Castañeda et al. (2016). A unique feature of the GMD is that the welfare aggregates\nare the same as those used to compute the poverty estimates published by PovcalNet and the World\nDevelopment Indicators. These aggregates are based on household per capita income or consumption,\ndepending on the specific concept that is used to measure national poverty in any given country.\n\n\nFor the effects of this exercise, the rescaling of welfare was performed using the Poverty and Inequality\nPlatform (PIP) which contains data on poverty headcounts for each year (2013-2018). For 2020-2022,\nthe data was provided by the Poverty and Shared Prosperity team. A rescaling of the weights was done\nusing the UNDESA population estimates from 2013-2021. In other words, the number of children is\nestimated by applying the proportion of people that are children in the relevant survey to the total\nUNDESA national population projections for each country. This could lead to discrepancies with UNDESA\nestimates of the number of children in a country. For the 2022 estimates, UNDESA population prospects\nwere used.\n\n\nPoverty estimates for 2020, 2021 and 2022 are “nowcasted”—that is, gross domestic product (GDP)\ngrowth rates have been used to forecast household incomes, assuming all households experience equal\nconsumption growth in percentage terms.\n\n\nTrends from 2013 to 2018 are all based on\n\nMention: \"Poverty and Inequality\nPlatform\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"National poverty headcounts for these years are used to rescale the survey welfare aggregates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000921:33:0", "occurrence_id": "58918c30515e00e5ce911c8d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Data and methodology\n\n\nThis exercise builds on a similar one conducted in 2020 to examine child poverty using the GMD (Silwal,\nEngilbertsdottir, Cuesta, Newhouse and Stewart, 2020). The analysis contains records from 10.4 million\nindividuals from 147 countries, taken from the 2022 spring version of the GMD with 2019 as a base year.\nThe GMD is a collection of globally harmonized household survey data compiled by the Data for Goals\ngroup of the World Bank’s Poverty and Equity Global Practice. Further details on the background of the\nGMD are given in Castañeda et al. (2016). A unique feature of the GMD is that the welfare aggregates\nare the same as those used to compute the poverty estimates published by PovcalNet and the World\nDevelopment Indicators. These aggregates are based on household per capita income or consumption,\ndepending on the specific concept that is used to measure national poverty in any given country.\n\n\nFor the effects of this exercise, the rescaling of welfare was performed using the Poverty and Inequality\nPlatform (PIP) which contains data on poverty headcounts for each year (2013-2018). For 2020-2022,\nthe data was provided by the Poverty and Shared Prosperity team. A rescaling of the weights was done\nusing the UNDESA population estimates from 2013-2021. In other words, the number of children is\nestimated by applying the proportion of people that are children in the relevant survey to the total\nUNDESA national population projections for each country. This could lead to discrepancies with UNDESA\nestimates of the number of children in a country. For the 2022 estimates, UNDESA population prospects\nwere used.\n\n\nPoverty estimates for 2020, 2021 and 2022 are “nowcasted”—that is, gross domestic product (GDP)\ngrowth rates have been used to forecast household incomes, assuming all households experience equal\nconsumption growth in percentage terms.\n\n\nTrends from 2013 to 2018 are all based on\n\nMention: \"UNDESA population estimates\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Population figures are used to rescale survey weights.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000921:33:0", "occurrence_id": "788ee0c90900db813b5876e2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Data and methodology\n\n\nThis exercise builds on a similar one conducted in 2020 to examine child poverty using the GMD (Silwal,\nEngilbertsdottir, Cuesta, Newhouse and Stewart, 2020). The analysis contains records from 10.4 million\nindividuals from 147 countries, taken from the 2022 spring version of the GMD with 2019 as a base year.\nThe GMD is a collection of globally harmonized household survey data compiled by the Data for Goals\ngroup of the World Bank’s Poverty and Equity Global Practice. Further details on the background of the\nGMD are given in Castañeda et al. (2016). A unique feature of the GMD is that the welfare aggregates\nare the same as those used to compute the poverty estimates published by PovcalNet and the World\nDevelopment Indicators. These aggregates are based on household per capita income or consumption,\ndepending on the specific concept that is used to measure national poverty in any given country.\n\n\nFor the effects of this exercise, the rescaling of welfare was performed using the Poverty and Inequality\nPlatform (PIP) which contains data on poverty headcounts for each year (2013-2018). For 2020-2022,\nthe data was provided by the Poverty and Shared Prosperity team. A rescaling of the weights was done\nusing the UNDESA population estimates from 2013-2021. In other words, the number of children is\nestimated by applying the proportion of people that are children in the relevant survey to the total\nUNDESA national population projections for each country. This could lead to discrepancies with UNDESA\nestimates of the number of children in a country. For the 2022 estimates, UNDESA population prospects\nwere used.\n\n\nPoverty estimates for 2020, 2021 and 2022 are “nowcasted”—that is, gross domestic product (GDP)\ngrowth rates have been used to forecast household incomes, assuming all households experience equal\nconsumption growth in percentage terms.\n\n\nTrends from 2013 to 2018 are all based on\n\nMention: \"UNDESA national population projections\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Applying survey child shares to these projections yields the number of children per country.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000921:33:0", "occurrence_id": "d0982ce2dc6c8d319902f398", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Data and methodology\n\n\nThis exercise builds on a similar one conducted in 2020 to examine child poverty using the GMD (Silwal,\nEngilbertsdottir, Cuesta, Newhouse and Stewart, 2020). The analysis contains records from 10.4 million\nindividuals from 147 countries, taken from the 2022 spring version of the GMD with 2019 as a base year.\nThe GMD is a collection of globally harmonized household survey data compiled by the Data for Goals\ngroup of the World Bank’s Poverty and Equity Global Practice. Further details on the background of the\nGMD are given in Castañeda et al. (2016). A unique feature of the GMD is that the welfare aggregates\nare the same as those used to compute the poverty estimates published by PovcalNet and the World\nDevelopment Indicators. These aggregates are based on household per capita income or consumption,\ndepending on the specific concept that is used to measure national poverty in any given country.\n\n\nFor the effects of this exercise, the rescaling of welfare was performed using the Poverty and Inequality\nPlatform (PIP) which contains data on poverty headcounts for each year (2013-2018). For 2020-2022,\nthe data was provided by the Poverty and Shared Prosperity team. A rescaling of the weights was done\nusing the UNDESA population estimates from 2013-2021. In other words, the number of children is\nestimated by applying the proportion of people that are children in the relevant survey to the total\nUNDESA national population projections for each country. This could lead to discrepancies with UNDESA\nestimates of the number of children in a country. For the 2022 estimates, UNDESA population prospects\nwere used.\n\n\nPoverty estimates for 2020, 2021 and 2022 are “nowcasted”—that is, gross domestic product (GDP)\ngrowth rates have been used to forecast household incomes, assuming all households experience equal\nconsumption growth in percentage terms.\n\n\nTrends from 2013 to 2018 are all based on\n\nMention: \"UNDESA population prospects\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The most recent vintage is used for the 2022 population denominators.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000910:25:1", "occurrence_id": "fbf521f7ea39a3f9c560f768", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Distribution of Patients by Wealth Index and Treatment Status\n\n\n\n\n\n\n\n_Notes._ Kolmogorov-Smirnov test p-value = 0.325. Index range is observed range. Wealth\nindex is estimated following the methodology of the Kenya Demographic and Health Survey\nbased on household ownership of selected assets.\n\n\n24\n\nMention: \"Kenya Demographic and Health Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Its asset-based methodology is followed to estimate the wealth index used for patient distribution.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000912:6:1", "occurrence_id": "01291d171a9cb2cbca2c931e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  advantage (RCA), export similarity with China, and intra\n\nindustry trade with China have been computed using trade data from UN Comtrade. Data on Free Trade\n\n\nAgreements (FTAs) are from Mario Larch’s Regional Trade greements Database based on Egger and Larch\n\n\n(2008). A summary of the main variables is presented in Appendix Table 2.\n\n\nPrima facie evidence on the reshuffling of the top trade partners from 2017 to 2022 is presented in Figure\n\n\n3. China’s share fell 5.3 percentage points (ppts). For strategic products, this decline is even more\n\n\nmassive, from 36.8 percent in 2017 to 23.1 percent in 2022. Focusing on the overall shares, the countries\n\n\nwith the biggest gains in market share were Vietnam (1.9 ppt), Taiwan, China (1 ppt), Canada (0.75 ppt),\n\n\nMexico (0.64 ppt), India (0.57 ppt), and the Republic of Korea (0.53 ppt). These six countries more than\n\n\naccount for China’s 5.3 percentage point decline. Their combined gain does not, however, mean that\n\n\nthese countries are the main or only beneficiaries, as they may be increasing their market share in\n\n\nproducts that China does not export or for reasons unrelated to the tariffs on imports from China. These\n\n\n2 As is common in the literature, we exclude countries with population less than 1 million.\n\n\n5\n\nMention: \"Regional Trade greements Database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Free trade agreement membership is taken from this database to complement the trade flows.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000922:15:0", "occurrence_id": "b314637f360c12d21126e486", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Table 2. Analysis of Variance on the consumption expenditure shares, India\n\n\n\n\n\n\n\n\n\nMeat and fish 0.06 78.71\n\nEggs 0.10 3.67\n\nMilk and milk products 0.96 146.84\n\nOils and fats 0.34 18.07\n\nVegetables 2.45 76.82\n\nFruits 0.06 6.10\n\nPulses 0.85 10.16\n\nSweeteners 0.05 6.99\n\nSpices 0.25 4.44\n\nNonalcoholic beverages 0.11 2.77\n\nPrepared meals, snacks 0.16 7.26\n\nPan, tobacco, and intoxicants 0.11 64.69\n\nClothing and footwear 1.39 52.13\n\nHousing 0.06 27.90\n\nFuel 10.51 181.58\n\nElectricity 0.14 31.93\n\nHousehold goods, furniture 0.96 25.00\n\nHousehold service 0.08 10.07\n\nHealth 1.56 34.72\n\nTransport 0.15 22.38\n\nCommunication 0.14 14.51\n\nRecreation and amusement 2.10 78.18\n\nEducation 0.79 110.72\n\nPersonal care and effects 0.40 10.25\n\n\nSource : CPHS 2015-2019 and authors’ calculations.\n\n\n14\n\nMention: \"CPHS 2015-2019\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Consumption expenditure shares by category are tabulated from this household panel.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000923:23:0", "occurrence_id": "fa5284471880c27033a7e639", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Table 2: Average treatment effects of German acquisition\non firm-level employment\n\n\nTWFE Doubly robust\n(1) (2)\n\n\nAverage treatment effect (German acquisition) 0.075** 0.112**\n(0.035) (0.038)\n\n\nEffect by treatment year\n2005 0.057** 0.059**\n(0.026) (0.028)\n2006 0.043* 0.045\n(0.026) (0.030)\n2007 0.040 0.068**\n(0.033) (0.029)\n2008 0.087** 0.117**\n(0.041) (0.047)\n2009 0.059 0.083*\n(0.039) (0.045)\n2010 0.096* 0.140**\n(0.052) (0.056)\n2011 0.117** 0.161***\n(0.052) (0.062)\n2012 0.070 0.177**\n(0.077) (0.079)\n\n\nNotes: Authors’ calculations based on Orbis data. Results include only firms\nestablished before 2004. The doubly robust estimator of Sant’Anna and Zhao\n(2020) in column (2) has been implemented using pre-treatment levels of employment, age, region and sectoral dummies as covariates. *10% level, **5%\nlevel, and ***1% level.\n\n\nTable 3: Effects of EU accession on firm performance across sectors\n\n\nDep. variable: Log TFPR Log revenue Log employment Log exports\n(1) (2) (3) (4)\n\n\nPost _×_ Germany’s export share (log) 0.048*** 0.082** 0.029 0.162**\n(0.017) (0.032) (0.023) (0.064)\nFirm FE Yes Yes Yes Yes\n\nMention: \"Orbis data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Firm-level employment before and after German acquisition is estimated on this company database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000925:7:2", "occurrence_id": "062f694a53dd85be16ad727a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Bank’s Global Subnational\nPoverty Atlas offers several poverty estimates, which are all derived from the latest available Living\nStandards Measurement Survey (LSMS) for the respective country (World Bank, 2021). Areas where no\npoverty estimates are available tend to be high-income countries and small island states. For the purpose\nof this study, the standard World Bank definitions of poverty are used to determine the number of poor\npeople in a given subnational administrative unit. Specifically, poverty is defined by the daily expenditure\nthresholds of $1.90, $3.20, and $5.50.\n\n**Administrative boundaries:** The definition of national administrative boundaries follows the standard\nWorld Bank global administrative map. However, national boundaries are further disaggregated into\nsubnational units for all countries where World Bank household surveys are available with subnational\nrepresentativeness. These subnational units are typically provinces or states (i.e. admin1) but can also\ninclude custom groupings of subnational regions determined by the sampling strategy of household\nsurveys. Overall, this study covers 211 countries, which are disaggregated into 2,183 subnational units.\n\n\n6\n\nMention: \"World Bank household surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Where these surveys are sub nationally representative, national boundaries are disaggregated into matching units.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000927:8:0", "occurrence_id": "125740a09eb2d717072c720d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: estimated the number of poor people by calculating the number of individuals below the 2019 Guinean\n\n\npoverty line of 5,006,362 Guinean francs or $1,385 international dollars (purchasing power parity (2019);\n\n\n$) per year. <sup>[12]</sup> We considered total consumption expenditure as the wealth of individuals, and total\n\n\nconsumption expenditure minus OOP health expenditure as the impact of health expenditure on the\n\n\nwealth of individuals. We conducted the analyses for the entire population and by quintile. We calculated\n\n\nquintiles before and after health expenditures. For already poor individuals (individuals in quintiles 1 and\n\n\n2), the change in poverty was equal to their total consumption expenditures. <sup>[9]</sup>\n\n\nFor CHE and IHE, we considered all health expenditures as <mark>OOP</mark> health <mark>payments b</mark> ecause of the small\n\n\nproportion (3%) of individuals with some form of prepayment mechanism.\n\n\n**Identification of CHE and IHE drivers and estimation of the potential FRP of subsidies**\n\n\nWe defined the drivers of CHE and IHE as the expenditure categories resulting in the highest prevalences\n\n\nof CHE and IHE cases (previously calculated).\n\n\nWe estimated the FRP associated with THE and expenditure categories as the equivalent of the prevalence\n\n\nof CHE and IHE cases they induce. We estimated FRP for the entire population and by quintile. For IHE,\n\n\nquintiles were calculated before and after health expenditures.\n\n\nStatistical analyses were performed using STATA software (StataCorp LLC, College Station, TX, version\n\n\n16.1). We report the number of individuals at the national level rather than the number of respondents\n\n\nto the LSMS. Expenditure data were converted to 2019 international dollars using the rate 1 international\n\n\ndollar = 3,614 Guinean francs. <sup>[12]</sup>\n\n\n**\n\nMention: \"LSMS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Expenditure data are converted to international dollars and used to count people below the national poverty line.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000929:12:1", "occurrence_id": "b63fc8250b2aee37439c79ff", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  school, and young children and older people are more likely to use health care\nservices relative to older children and younger adults. Poorer households in Vietnam tend to be younger,\nwith more children. As households get richer, they have fewer children, with half or more of the richest\ntwo deciles having none at all (and those that do are more likely to attend non-subsidised private schools).\nAt the same time, there are almost no elderly in the poorer half of the distribution (and relatively few in\nany household, although this will change as the country ages) (see figure in Appendix 2).\n\n\nThe monetary value of public education is allocated as the cost per student per level of education (preprimary, primary, lower and upper secondary, TVET and higher education) calculated from administrative\nbudget and enrollment data. This is a _government cost_ approach, in which the use-value to in-kind benefits\nis given by the average cost of service provision per beneficiary. This represents what the household would\nhave to pay to use the service at the government’s cost. Data from past years on costs of each level of\neducation or each kind of treatment are used, then uprated while maintaining their proportion to the\ntotal government expenditures in 2018. <sup>22</sup>\n\n\nWe deduct costs paid by households (tuition fees and contributions to schools/class) from the benefits.\nThese are reported in the VHLSS but it is not possible to disaggregate them to allocate them to specific\n\n\n21 In addition, there is an electricity benefit transfer for poor households. We do not allocate this specifically and\ntreat it as part of the cash transfers.\n22 Scaling of in-kind benefits was standard in earlier CEQs but there has been a movement away from this practice.\nIn Vietnam, any scaling would be small as the gap between national accounts private consumption and VHLSS\nconsumption is small.\n\n\n11\n\nMention: \"administrative\nbudget and enrollment data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Cost per student by education level is calculated from these records to value in-kind education benefits.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000929:12:1", "occurrence_id": "60621a044f6ab6d97a743d13", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  school, and young children and older people are more likely to use health care\nservices relative to older children and younger adults. Poorer households in Vietnam tend to be younger,\nwith more children. As households get richer, they have fewer children, with half or more of the richest\ntwo deciles having none at all (and those that do are more likely to attend non-subsidised private schools).\nAt the same time, there are almost no elderly in the poorer half of the distribution (and relatively few in\nany household, although this will change as the country ages) (see figure in Appendix 2).\n\n\nThe monetary value of public education is allocated as the cost per student per level of education (preprimary, primary, lower and upper secondary, TVET and higher education) calculated from administrative\nbudget and enrollment data. This is a _government cost_ approach, in which the use-value to in-kind benefits\nis given by the average cost of service provision per beneficiary. This represents what the household would\nhave to pay to use the service at the government’s cost. Data from past years on costs of each level of\neducation or each kind of treatment are used, then uprated while maintaining their proportion to the\ntotal government expenditures in 2018. <sup>22</sup>\n\n\nWe deduct costs paid by households (tuition fees and contributions to schools/class) from the benefits.\nThese are reported in the VHLSS but it is not possible to disaggregate them to allocate them to specific\n\n\n21 In addition, there is an electricity benefit transfer for poor households. We do not allocate this specifically and\ntreat it as part of the cash transfers.\n22 Scaling of in-kind benefits was standard in earlier CEQs but there has been a movement away from this practice.\nIn Vietnam, any scaling would be small as the gap between national accounts private consumption and VHLSS\nconsumption is small.\n\n\n11\n\nMention: \"VHLSS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Household-paid tuition fees and school contributions are reported here and deducted from benefits.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000934:21:0", "occurrence_id": "b601ee962a5f634560a3cbd5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 5: Shapley decomposition employment adjustments\n\n\n\n(a) Extensive margin\n\n\n\n\n\n(b) Intensive margin\n\n\n\n\n\n\n\n\n\n\n|59.8|Col2|Col3|\n|---|---|---|\n|59.8|59.8|59.8|\n||||\n|22|22||\n|1.6<br>5.6|11||\n||||\n\n\n|72.4|Col2|\n|---|---|\n|||\n|||\n||20|\n|1<br>2.3|4.3|\n|||\n\n\n\n\n\n\n|Col1|Informality External attributes<br>Severity of the lockdown (mobility) Severity of demand shock<br>Constraints|\n|---|---|\n|||\n|||\n\n\n|Col1|Informality External attributes<br>Severity of the lockdown (mobility) Severity of demand shock<br>Constraints|\n|---|---|\n|||\n|||\n\n\n\n**Notes:** Firm attributes include sector, age, gender of the CEO, country, and panel firms.\nSource: Authors’ calculations using data from the World Bank Business Pulse Survey (BPS).\n\n\nFigure 6: Employment adjustments in crisis and recovery phases\n\n\n\n(a) Employment adjustments by formal and informal firms\n\n\n\n\n\n(b) Employment adjustments by types of informal firms\n\n\n\n\n\n\n\n**Notes:** Firm attributes include sector, age, gender of the CEO, country, and panel firms.\nSource: Authors’ calculations using data from the World Bank Business Pulse Survey (BPS).\n\n\n20\n\nMention: \"World Bank Business Pulse Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Shapley decompositions of employment adjustments are computed on this repeating firm survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000935:14:0", "occurrence_id": "11bbdad9d834d507bec0e1eb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ###### **3.2 Sample**\n\nThe sample for this paper consists of 150,000 business owners across 50 countries where there\n\n\nwas sufficient sample size per country to conduct gender-disaggregated analysis. Facebook\n\n\nimplements a modified simple random sample for survey invitations within each country. <sup>21</sup>\n\n\nSampling weights are determined for survey non-response patterns using characteristics most\n\n\nclosely associated with non-response to Facebook surveys. <sup>22</sup> We present the descriptive\n\n\nstatistics of our sample by gender in Table G1 in Appendix G and describe them here.\n\n\nThe sample consists of male and female business owners and managers that are similar in\n\n\nage, education and household composition. The average age of male and female entrepreneurs\n\n\nis 37 years old; 57% of women entrepreneurs in the sample have at least a university or\n\n\ncollege degree relative to 54% of male entrepreneurs; and 97% of the sample have at least\n\n\nsome secondary education. In terms of marital status: 78% of men and 75% of women\n\n\nsurveyed report that they have a long-term partner or spouse. In the survey we ask about\n\n\nthe types of dependents in their household: 41% of men and 37% of women in the sample\n\n\nreport that they live with children under 5 years old, 59% of men and 66% of women in the\n\n\nsample report that they live with children aged 6-18 years old, and 30% of men and 25%\n\n\nof women in the sample report that they live with elderly dependents aged over 65 years\n\n\nold. The average income for the households in the sample in the past 30 days is USD2,733\n\n\nfor men; and USD2,729 for women, however this is mostly driven by households in North\n\n\nAmerica and Europe. <sup>23</sup> Within their households, 83% of men and 56% of women report that\n\n\nthey consider themselves\n\nMention: \"Facebook surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"Sampling weights are constructed using the characteristics most associated with non-response in these surveys.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000937:9:2", "occurrence_id": "f55a54fa8ec7a577e8b5dc41", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  A household size greater than four may be more appropriate, especially in lower-income countries, where fertility rates tend\nto be higher. However, Anker (2011) mentions a few considerations. First, data on household size are often not available in labor\nforce surveys, which usually exclude individuals outside the ages of 15–64. Second, using region- or country-specific estimates of\ncompleted family size can be complicated. For instance, observed average household size is affected by the distribution of\nhousehold types and sizes. Single-person households would significantly reduce average household size even though this is\nunrelated to the typical household size of workers with children.\n11 Consumer price index and PPP values are taken from WDI (World Development Indicators),World Bank, Washington, DC,\nhttps://datatopics.worldbank.org/world-development-indicators/.\n12 HDI (Human Development Index) (dashboard), United Nations Development Programme, New York,\nhttps://hdr.undp.org/data-center/human-development-index#/indicies/HDI.\n\n\n8\n\nMention: \"WDI\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Consumer price indices and purchasing power parity values are drawn from this indicator series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000940:5:0", "occurrence_id": "8bdd086ea9e783d9973c4275", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: than the global average in each case. With universal coverage of primary and secondary education,\nwomen achieve relatively high levels of education in all three counties (figure 1), making such notable\ngender pay gaps particularly striking.\n\n**Figure 1: Summary measures of educational attainment in Central Asia**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n_Source: World Development Indicators and National Statistical Offices._\n\n\nTaking Uzbekistan as an example that illustrates challenges across the region, the gender gap in labor\nforce participation is strongly linked to both educational attainment and the number of children living in\nthe household (figure 2). Only 6 percent of women with general secondary education or below, and 12\npercent of women with upper secondary education, are active in the labor market compared to 18 percent\nand 45 percent of men, respectively. However, 52 percent of women with bachelor’s degrees or higher\nare active in the labor market, which is more comparable to the 60 percent labor participation rate for\nmen with bachelor’s degrees. Furthermore, women (ages 16+) in the bottom two wealth quintiles were\nmore than 50 percent likely to be inactive or unemployed, whereas 39 percent of men in the lowest wealth\nquintile and 32 percent of men in the second lowest quintile were either inactive or unemployed. Having\nyoung children also affect female and male labor force participation differently. Having young children\nhas distinct effects on female and male labor force participation. In the case of men, having more young\nchildren is associated with a higher likelihood of working. For women, it is the reverse, women are much\nless likely to work outside the home if more children are present.\n\n\n4\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Educational attainment measures for the region are compiled with national statistical offices from this series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000942:6:0", "occurrence_id": "17fb612a80716b9231599c30", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: properly control for this potential endogeneity issue, a dynamic panel GMM estimation technique is\nimplemented (Arellano and Bond, 1991; Arellano and Bover, 1995). The system GMM technique controls\nfor endogeneity by using lags of the endogenous variables as instruments in the econometric equation. It\nalso controls for eventual autocorrelation issues that could appear after introducing lags of the dependent\nvariable as regressors. 𝐸𝑆𝑝𝑒𝑛𝑑𝑖𝑡−1, 𝐹𝑖𝑠𝑐𝑜𝑛𝑖𝑡−1 𝑎𝑛𝑑 𝐺𝐷𝑃𝑔𝑤ℎ𝑖𝑡 are treated as endogenous variables in\nthe system GMM regression. <sup>3</sup> [^3: It is worth mentioning that debt service obligations exist because of countries’ borrowing activities in the past.\nPast education spending financed by debt could play a role in the relationship between the current debt service\nand education spending. Although equation 4 is dynamic and takes into account education expenditure at time t-1,\nit may not fully capture the long-term relationship between past education spending on the one hand and the\ncurrent debt service and education spending on the other hand.]\n\n**3. Data**\n\nTable 1-Panel A shows the data used to estimate equation 1, and Panel B describes the data used to\nestimate equation 4. Data are collected from the World Economic Outlook and the World Development\nIndicators databases. The data cover the period 1990-2019. To address the data availability issues, all\nvariables are averaged by a three-year interval.\n\nPanel A data show that about 9.5% of the observations experienced a relative decline of total government\nexpenditure over the last decades and the average debt service-to-reserve ratio varies between 0.5% and\n397%. The average debt-to-GDP ratio is over the IMF's recommended benchmark of 40% in about 60% of\nthe countries, showing that several countries may be affected by debt sustainability issues. The data are\navailable for a total of 50 low- and middle-income countries from 1991 to 2019 to estimate equation 1.\n\nEducation spending per school-age individual (in primary, secondary, and tertiary) is the dependent\nvariable in equation 4 (Table 1-Panel B).\n\nMention: \"World Economic Outlook\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Debt service and government spending series for the dynamic panel come from this fund database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000942:6:0", "occurrence_id": "cad580393d193ce128cef7b3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: properly control for this potential endogeneity issue, a dynamic panel GMM estimation technique is\nimplemented (Arellano and Bond, 1991; Arellano and Bover, 1995). The system GMM technique controls\nfor endogeneity by using lags of the endogenous variables as instruments in the econometric equation. It\nalso controls for eventual autocorrelation issues that could appear after introducing lags of the dependent\nvariable as regressors. 𝐸𝑆𝑝𝑒𝑛𝑑𝑖𝑡−1, 𝐹𝑖𝑠𝑐𝑜𝑛𝑖𝑡−1 𝑎𝑛𝑑 𝐺𝐷𝑃𝑔𝑤ℎ𝑖𝑡 are treated as endogenous variables in\nthe system GMM regression. <sup>3</sup> [^3: It is worth mentioning that debt service obligations exist because of countries’ borrowing activities in the past.\nPast education spending financed by debt could play a role in the relationship between the current debt service\nand education spending. Although equation 4 is dynamic and takes into account education expenditure at time t-1,\nit may not fully capture the long-term relationship between past education spending on the one hand and the\ncurrent debt service and education spending on the other hand.]\n\n**3. Data**\n\nTable 1-Panel A shows the data used to estimate equation 1, and Panel B describes the data used to\nestimate equation 4. Data are collected from the World Economic Outlook and the World Development\nIndicators databases. The data cover the period 1990-2019. To address the data availability issues, all\nvariables are averaged by a three-year interval.\n\nPanel A data show that about 9.5% of the observations experienced a relative decline of total government\nexpenditure over the last decades and the average debt service-to-reserve ratio varies between 0.5% and\n397%. The average debt-to-GDP ratio is over the IMF's recommended benchmark of 40% in about 60% of\nthe countries, showing that several countries may be affected by debt sustainability issues. The data are\navailable for a total of 50 low- and middle-income countries from 1991 to 2019 to estimate equation 1.\n\nEducation spending per school-age individual (in primary, secondary, and tertiary) is the dependent\nvariable in equation 4 (Table 1-Panel B).\n\nMention: \"World Development\nIndicators databases\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Education spending and the remaining macro variables are taken from this World Bank database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000944:48:0", "occurrence_id": "37f1c943014b19e376f5789f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Appendix 9: Additional figures**\n\n**Figure 9.1: Trade-employment (left) and trade-income elasticities (right), Western Europe vs. rest of sample**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nSource: Own computations. Data: OECD TiE and TiVA data, covering 48 countries and 45 sectors for the period 1995-2018. Note: Tradeemployment (trade-income, resp.) elasticities at the country-sector level have been estimated using an instrumental variables regression\napproach with employment (labor income, resp.) (in natural logs) as dependent variable and the instrumented trade variables (in levels and\nnatural logs) as independent variable. The panel data regressions control for country-sector, country-time, and sector-time fixed effects. Only\nresults significant at the 10% level and above the critical value for the Kleibergen-Paap test statistic are reported. For more information, see\nAppendix 1. WEuropean = Western European countries which include the EU, Norway, Switzerland and United Kingdom.\n\n\n**Figure 9.2: Export-employment (upper panel) and export-income elasticities (lower panel), by country**\n\n|Col1|0.8<br>0.6<br>0.4<br>0.2<br>0.0<br>-0.2<br>-0.4<br>-0.6|Col3|\n|---|---|---|\n||<br>-0.2<br>0.0<br>0.2<br>0.4<br>0.6<br>0.8<br>1.0<br>DEU<br>BEL<br>SWE<br>CAN<br>AUS<br>IRL<br>SVK<br>NL\n\nMention: \"OECD TiE and TiVA data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Trade-employment and trade-income elasticities are estimated across 48 countries and 45 sectors.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000947:15:1", "occurrence_id": "174e5744ae23c53a988b54ba", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  accessible to pedestrians. A pedestrian walking speed of\n\n\n3.6km/h is used, and the threshold of 20 minutes is used as the maximum walking time _per leg_\n\n\nof the trip when public transport is being used. Second, we use the public transit network that\n\n\nhas been collected. The collected frequency of departure, and speeds by which the public\n\n\ntransport vehicles travel on the route are used for the analysis (see 2.1. for details and maps\n\n\nabout this data). Third, we use the distribution of employment in each city. We base it on Barzin\n\n\net al. (2022), which estimates the relative density of employment across the metropolitan area\n\n\nusing a machine learning method and remote sensing data.\n\n\n21 A given destination will be considered to be reachable from a residential location if it can be reached in under\n60 minutes during at least 50% of the departures within the 8:00 to 9:00 am window.\n\n\n14\n\nMention: \"public transit network\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Collected departure frequencies and vehicle speeds determine reachable destinations within the walking threshold.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000952:23:0", "occurrence_id": "d1fc9d3aeec08b2407996eaa", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: teraction term between international trade and the GDP per capita as a measure of the\n\n\nexporter’s level of development. Again, the estimate for civil conflict is unchanged.\n\n#### **5 Conclusion**\n\n\nThis paper investigates the impact of civil conflicts on international trade. It uses\n\n\ninternational trade and production data from 1989-2006 provided by the BACI data set\n\n\nto carry out this empirical analysis. Additionally, it uses data on civil conflicts from the\n\n\nMPEV data set for the same period. For this purpose, a structural gravity model is esti\n\nmated with exporter-time, importer-time, and country-pair fixed effects. The presence of\n\n\ndirectional time-varying fixed effects is crucial to obtain consistent and reliable estimates.\n\n\nHowever, in this case, country-year specific variables such as civil conflict can no longer\n\n\nbe estimated as they are absorbed by the fixed effects. To fix this issue, a new methodol\n\nogy outlined by Heid et al. (2017) and Beverelli et al. (2018) is used where internal trade\n\n\ndata is used to identify the impact of civil conflicts on international trade.\n\n\nThis paper finds that civil conflicts in a country have a highly detrimental effect\n\n\non trade flows. On average, a civil conflict in a given year decreases international trade by\n\n\nover 40% in this year. Moreover, this effect is highly persistent over time. Distinguishing\n\n\nbetween political and ethnic civil conflicts, this paper shows that ethnic conflicts destroy\n\n\ntrade more than political ones. Thus, one of the main channels through which civil con\n\nflicts impact international trade is by causing social disorder and deepening ethnic and\n\n\nreligious differences within a society. This can lead to a significant deterioration of the\n\n\neconomic environment affecting trade relations with the rest of the world. Additionally,\n\n\nthis paper examines the spillover effects of civil conflicts to neighboring countries. It\n\n\nshows that an additional contiguous country involved in a civil conflict reduces interna\n\ntional trade flows by up to\n\nMention: \"BACI data set\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"International trade and production flows form the dependent variable of the gravity model.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000952:23:0", "occurrence_id": "90c2b5bed376beaafc568427", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: teraction term between international trade and the GDP per capita as a measure of the\n\n\nexporter’s level of development. Again, the estimate for civil conflict is unchanged.\n\n#### **5 Conclusion**\n\n\nThis paper investigates the impact of civil conflicts on international trade. It uses\n\n\ninternational trade and production data from 1989-2006 provided by the BACI data set\n\n\nto carry out this empirical analysis. Additionally, it uses data on civil conflicts from the\n\n\nMPEV data set for the same period. For this purpose, a structural gravity model is esti\n\nmated with exporter-time, importer-time, and country-pair fixed effects. The presence of\n\n\ndirectional time-varying fixed effects is crucial to obtain consistent and reliable estimates.\n\n\nHowever, in this case, country-year specific variables such as civil conflict can no longer\n\n\nbe estimated as they are absorbed by the fixed effects. To fix this issue, a new methodol\n\nogy outlined by Heid et al. (2017) and Beverelli et al. (2018) is used where internal trade\n\n\ndata is used to identify the impact of civil conflicts on international trade.\n\n\nThis paper finds that civil conflicts in a country have a highly detrimental effect\n\n\non trade flows. On average, a civil conflict in a given year decreases international trade by\n\n\nover 40% in this year. Moreover, this effect is highly persistent over time. Distinguishing\n\n\nbetween political and ethnic civil conflicts, this paper shows that ethnic conflicts destroy\n\n\ntrade more than political ones. Thus, one of the main channels through which civil con\n\nflicts impact international trade is by causing social disorder and deepening ethnic and\n\n\nreligious differences within a society. This can lead to a significant deterioration of the\n\n\neconomic environment affecting trade relations with the rest of the world. Additionally,\n\n\nthis paper examines the spillover effects of civil conflicts to neighboring countries. It\n\n\nshows that an additional contiguous country involved in a civil conflict reduces interna\n\ntional trade flows by up to\n\nMention: \"MPEV data set\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Civil conflict incidence over the same period provides the main regressor.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000956:24:0", "occurrence_id": "4c1b6294e5cb945ef3f32ee2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 11. Poverty rate using a poverty line of**\n**US$ 6.85 PPP; female-headed household of women**\n**who are sole providers with at least one young child**\n\n\n\n**Figure 12. Poverty rate using a poverty line of US$ 6.85 PPP;**\n**contributions to family income by gender of the head of**\n**household**\n\n\n\n_Source: Own estimates with data from the EHPM 2019. Male-headed households who are sole providers with_ _young children are_\n_omitted, as they only represent_ _0.1 percent of households in the country._\n\n\n**A critical finding of this study is that fiscal policy has a greater impact on increasing poverty among**\n**households with a sole female breadwinner than among households with a sole male provider,**\n**which widens gender gaps.** Among households with a sole female breadwinner, the poverty rate\nincreases by 4.3 percentage points (from 22.3 percent to 26.6 percent), whereas households with the\nman as the sole provider experience a smaller increase of 2.3 percentage points (from 15.5 percent to\n17.8 percent) (Figure 13). These changes are statistically significant, and the poverty gap between the\ntwo household types increases from around 6.8 percentage points to 8.8 percentage points.\n\n\n**These results suggest that women in households with a female sole breadwinner are disadvantaged**\n**both in the labor market and in terms of fiscal policy** . In terms of fiscal policy, the inequalities arising\nfrom the labor market are not being compensated; as shown in Figure 8, the income from the labor\nmarket of this group of women is lower before fiscal policy. Furthermore, receiving transfers and\nmaking tax payments leads to an increase in poverty. It is also important to consider the role of\nremittances.\n\nMention: \"EHPM 2019\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Poverty rates by household headship and gender of income contributors are estimated on this household survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000961:0:0", "occurrence_id": "afb26b3819fac655fca9a905", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Policy Research Working Paper\n10184\nImpacts of COVID-19 on the Income  \nand Mental Well-Being of Cismen, Ciswomen, \nTransgender, and Non-Binary Individuals\nEvidence from the 2020 COVID-19 Disparities Survey\nTanima Ahmed\nToni Joe Lebbos\nSean Howell\nErik Lamontagne\nJoshua Seth Wimpey\nSocial Sustainability and Inclusion Global Practice\nSeptember 2022 \nPublic Disclosure Authorized\nPublic Disclosure Authorized\nPublic Disclosure Authorized\nPublic Disclosure Authorized\n\nMention: \"2020 COVID-19 Disparities Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Income and mental well-being of gender-diverse respondents are measured in this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000963:9:2", "occurrence_id": "517f19f41fbb6414a091ae18", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  boxes within each psu_\nTotal hours/day 7.97 3.32 9.41 27848 2.34 2.31 1.69 14117\n**(B3) Household Data Obs level:** _household-by-monthly-round_\nLength of outage\n\nover the past 10 days (days) 2.31 2.00 1.86 3726 - - - over the past 10 days (hours/day) 6.96 4.00 7.32 3724 - - - yesterday (hours/day) 10.04 12.00 6.31 2382 2.09 2.00 1.25 462\n<u>day before yesterday (hours/day)</u> <u>10.12</u> <u>12.00</u> <u>6.48</u> <u>2480</u> <u>2.29</u> <u>2.00</u> <u>1.31</u> <u>455</u>\nAll observations are in or after 2016, Nov 1.\n\n\nThese comparisons are drawn out in alternative formulation in table 3, which reports t-tests comparing\nthe average difference between the two data collection methods from each source for the same\nlocation and period. Although the data are structured differently (SSB data provide continuous\ncoverage), SSB data can be harmonized to cover the given reference period (for instance, “yesterday”\nor “the day before yesterday” calculated using the interview date) and compared. Survey-based\nmethods report about 10 percent fewer outages “yesterday” on average, and a similar rate with respect\nto the “day before yesterday.” When asked about the preceding ten-day period, respondents reported\nless, reporting outages at about half the rate that boxes recorded them.\n\n\n8\n\nMention: \"SSB data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Its continuous coverage is harmonised to the reference period so it can be compared with survey-based methods.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000964:28:2", "occurrence_id": "b38ad698b23c5fd1cf901f61", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: R6)|Jul-Nov 2021<br>(R6)|5,765|Is [child] currently enrolled in school or planning to<br>enrol once schools reopen?|\n|Malawi|Apr-May 2021<br>(R10)|Previous<br>survey (IHPS<br>2019)|919|Are you currently attending school (the 2020-21<br>year)?|\n|Nigeria|Oct 2020 (R6)|Previous<br>survey (GHSP<br>2018)|1,762|Is [child] currently attending school either in-<br>person or remotely?<br>Has [child] been attending classes on-site since<br>schools reopened after the closure due to the<br>coronvirus crisis?|\n|Nigeria|Mar 2021 (R11)|Mar 2021 (R11)|1,680|Is [child] currently attending school either in-<br>person or remotely?|\n|Uganda|Oct-Nov 2020<br>(R4)|Previous<br>survey (UNPS<br>2018/19)|2,136|Is [child] currently going to school?|\n|Uganda|Mar-Apr 2021<br>(R6)|Mar-Apr 2021<br>(R6)|2,100|Is [child] currently going to school?|\n|Uganda|Sep-Nov 2021<br>(R7), asking<br>about May 2021|Sep-Nov 2021<br>(R7), asking<br>about May 2021|1,950|Was [child] attending school before schools were<br>closed due to the second lockdown due to<br>coronavirus in June 2021?|\n\n\n27\n\nMention: \"IHPS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its respondents form the earlier wave against which school attendance is re-asked.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000964:28:2", "occurrence_id": "414ef78e82589ed7fb8652db", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: R6)|Jul-Nov 2021<br>(R6)|5,765|Is [child] currently enrolled in school or planning to<br>enrol once schools reopen?|\n|Malawi|Apr-May 2021<br>(R10)|Previous<br>survey (IHPS<br>2019)|919|Are you currently attending school (the 2020-21<br>year)?|\n|Nigeria|Oct 2020 (R6)|Previous<br>survey (GHSP<br>2018)|1,762|Is [child] currently attending school either in-<br>person or remotely?<br>Has [child] been attending classes on-site since<br>schools reopened after the closure due to the<br>coronvirus crisis?|\n|Nigeria|Mar 2021 (R11)|Mar 2021 (R11)|1,680|Is [child] currently attending school either in-<br>person or remotely?|\n|Uganda|Oct-Nov 2020<br>(R4)|Previous<br>survey (UNPS<br>2018/19)|2,136|Is [child] currently going to school?|\n|Uganda|Mar-Apr 2021<br>(R6)|Mar-Apr 2021<br>(R6)|2,100|Is [child] currently going to school?|\n|Uganda|Sep-Nov 2021<br>(R7), asking<br>about May 2021|Sep-Nov 2021<br>(R7), asking<br>about May 2021|1,950|Was [child] attending school before schools were<br>closed due to the second lockdown due to<br>coronavirus in June 2021?|\n\n\n27\n\nMention: \"GHSP\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its earlier households are the frame for the follow-up questions on remote and on-site schooling.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000966:8:2", "occurrence_id": "1b258b1e72ad642bc5fa3561", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: )\nreported receiving support from teachers in terms of interest, encouragement and adaptation of\nschoolwork to individual needs. The results from the REDS support the hypothesis that the national\n\n\n1 UNICEF Rapid Assessment – telephone survey: June – July 2020, SISQE-UNICEF online poll (through\nSISQE web portal) – June – July 2020, U-Report Poll (SMS, Facebook, Telegram, App for smartphones):\nMay 2020, Listening to the Citizens of Uzbekistan – May 2020.\n2 There are slight variations in results by surveys and by student grade.\n\n\n7\n\nMention: \"REDS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Teacher support reported by students comes from this rapid education survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000966:8:2", "occurrence_id": "961d009c1ea21c65ddd2ad7a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: )\nreported receiving support from teachers in terms of interest, encouragement and adaptation of\nschoolwork to individual needs. The results from the REDS support the hypothesis that the national\n\n\n1 UNICEF Rapid Assessment – telephone survey: June – July 2020, SISQE-UNICEF online poll (through\nSISQE web portal) – June – July 2020, U-Report Poll (SMS, Facebook, Telegram, App for smartphones):\nMay 2020, Listening to the Citizens of Uzbekistan – May 2020.\n2 There are slight variations in results by surveys and by student grade.\n\n\n7\n\nMention: \"UNICEF Rapid Assessment – telephone survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Listed among the parallel instruments whose results vary slightly from the main survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000966:8:2", "occurrence_id": "767f4c8eae4336a99d65da1f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: )\nreported receiving support from teachers in terms of interest, encouragement and adaptation of\nschoolwork to individual needs. The results from the REDS support the hypothesis that the national\n\n\n1 UNICEF Rapid Assessment – telephone survey: June – July 2020, SISQE-UNICEF online poll (through\nSISQE web portal) – June – July 2020, U-Report Poll (SMS, Facebook, Telegram, App for smartphones):\nMay 2020, Listening to the Citizens of Uzbekistan – May 2020.\n2 There are slight variations in results by surveys and by student grade.\n\n\n7\n\nMention: \"SISQE-UNICEF online poll\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"An online poll run through the national statistics portal used alongside the main survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000966:8:2", "occurrence_id": "22aeb5618ba2bfb295652f59", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: )\nreported receiving support from teachers in terms of interest, encouragement and adaptation of\nschoolwork to individual needs. The results from the REDS support the hypothesis that the national\n\n\n1 UNICEF Rapid Assessment – telephone survey: June – July 2020, SISQE-UNICEF online poll (through\nSISQE web portal) – June – July 2020, U-Report Poll (SMS, Facebook, Telegram, App for smartphones):\nMay 2020, Listening to the Citizens of Uzbekistan – May 2020.\n2 There are slight variations in results by surveys and by student grade.\n\n\n7\n\nMention: \"U-Report Poll\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"A multi-channel poll listed among the sources with slightly differing results.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000966:8:2", "occurrence_id": "523ba370d65ec3a254cafb85", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: )\nreported receiving support from teachers in terms of interest, encouragement and adaptation of\nschoolwork to individual needs. The results from the REDS support the hypothesis that the national\n\n\n1 UNICEF Rapid Assessment – telephone survey: June – July 2020, SISQE-UNICEF online poll (through\nSISQE web portal) – June – July 2020, U-Report Poll (SMS, Facebook, Telegram, App for smartphones):\nMay 2020, Listening to the Citizens of Uzbekistan – May 2020.\n2 There are slight variations in results by surveys and by student grade.\n\n\n7\n\nMention: \"Listening to the Citizens of Uzbekistan\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"Another contemporaneous poll named in the same comparison.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000969:11:0", "occurrence_id": "42e9e29c5abda29f2b18e8c0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: and ninth-graders, and students in the last grade of high school. The exam assesses students’\n\n\nproficiency in Portuguese and math.\n\n\nSince 1998, INEP has annually applied the National High School Proficiency Test (ENEM). <sup>11</sup>\n\n\nENEM is not mandatory and the students are the ones in charge of registering for the exam.\n\n\nENEM participants are the students enrolled in high school, mostly in the last grade, and the\n\n\nones that have already finished this level of education. Since 2009, the exam started being\n\n\nused by several Brazilian universities as one of their criteria to select their students. As a\n\n\nconsequence, the number of participants significantly increased over time, from 115,000 in 1998\n\n\nto 8.5 million in 2015. The exam aims to assess students’ learning levels in reading, math,\n\n\nhuman sciences, and natural sciences.\n\n\nAll aforementioned administrative surveys are publicly available. INEP discloses identified data\n\n\nat the school level, allowing us to merge the pilot schools with Education Census, _Prova_ _Brasil_,\n\n\nand ENEM. Table B.10 shows that, overall, treated and control schools do not have significant\n\n\ndifferences in average students’ proficiency in Portuguese and Math, grade promotion, dropout,\n\n\nretention rates, and age-grade distortion. Unidentified microdata at the student level is also\n\n\npublicly available and identified data only upon researcher request to access INEP’s facilities. <sup>12</sup>\n\n\nAs presented in section 2, principals’ choice of which classes would receive the intervention\n\n\nor would serve as a control is not necessarily random. In this sense, one may wonder whether\n\n\nschool principals from treatment and control schools face different incentives to select the classes\n\n\nthey did. In treated schools, principals could either select low-performer classes hoping to help\n\n\ntheir students with the intervention, or high-performer classes if they believe these students\n\n\nwould benefit most from the program. In\n\nMention: \"Education Census\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Identified school-level records are merged with the pilot schools to compare treated and control schools.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000970:23:0", "occurrence_id": "5d68bc6e291ed771ca4609e3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Table 2: Village-Level Public Health Infrastructure\n\n\nHealth facility Connection Public program Auxiliary Female\nindex index index Nurse Midwife doctor\n(1) (2) (3) (4) (5)\n\n\nFraction of female leaders 1.730* 2.138*** 0.973** 0.980*** 0.837***\n(0.919) (0.528) (0.455) (0.246) (0.263)\nObservations 7,696 7,700 7,722 7,723 7,722\nOutcome mean 0.000 0.000 0.000 0.646 0.219\n\n\nState fixed effects x x x x x\nDistrict characteristics x x x x x\n\n\n_Notes_ : Data comes from the 2012-13 District-level Household and Facility Survey (DLHS) of India. The columns\npresent 2SLS estimates specified in the following equations:\n\n\n\n\n\n\n_j_\n\n\n\n\n\n\n_j_\n\n\n\n_Yvds_ = _β_ 0 + _β_ 1 _Fds_ + _β_ 2 _TCds_ +\n\n\n\n_φ_ 1 _jIjdsG_ ( _Mjds_ ) +\n\n\n\n_φ_ 2 _jIjds_ + _Zds_ <sup>_′_</sup> <sup>_θ_</sup> <sup>2 +</sup> <sup>_δs_</sup> <sup>+</sup> <sup>_ϵvds_</sup> (4)\n\n\n\n_φ_ 2 _jIjds_ + _Z_ <sup>_′_</sup>\n\n\n\n\n\n\n_j_\n\n\n\n\n\n\n_j_\n\n\n\n_α_ 2 _jIjds_ + _Zds_ <sup>_′_<\n\nMention: \"District-level Household and Facility Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Village-level public health infrastructure indices are constructed from this district survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000972:5:0", "occurrence_id": "d590d641501282c04c2e2ba8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: from socialism to a market economy caused a change in the returns to schooling in transition\ncountries. It extends the analysis beyond other papers in this literature by explicitly looking at the\nimpact of EU accession. Since both events (transition and EU accession) represent critical changes\nin terms of political organization, markets, and trade, among others, this research explores the\ntransmission channel by which each shock could have affected returns to schooling. Given the\nincreased trade and economic integration in Europe post-transition, the premium to skills has\nincreased. We have shown a link between transition and the economic benefits of schooling. We\ntest if the returns to schooling increased significantly as a result of transition and EU integration,\nwith the corollary that the driver of the increase in the returns is openness to trade, eased by\ntransition and EU membership.\n\n\n**Data**\n\n\nThe data for this analysis stems from two main sources. First, the <u>[Comparable Returns to Education](https://datacatalog.worldbank.org/search/dataset/0059847/Comparable-Returns-to-Education)</u>\ndatabase was used as the main source of private returns to education data. This database was\nsupplemented with Psacharopoulos and Patrinos (2018) to make up for the gaps. Meanwhile, the\ndata source of the average years of schooling per country was taken from the Barro-Lee\nEducational Attainment Database (Barro and Lee 2013). This dataset contains the educational\nattainment of the population, including average schooling years, for 146 countries around the\nworld in 5-year intervals. These two key databases were selected as they ensure comparability\nacross countries.\n\n\nWe analyze the returns to education in 28 transition countries (see Annex 1). Here we define postsocialist/transition countries as all European and Central Asian countries that were part of the\nSoviet Union and socialist pact until the fall of the Soviet Union by the end\n\nMention: \"Barro-Lee\nEducational Attainment Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Average years of schooling for 146 countries in five-year intervals are taken from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000974:28:4", "occurrence_id": "f3d9256cbf70b1225c11f23e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: *𝑝** **𝑝** **𝑝** **𝑝** **𝑠** **𝑠**\ntime where the GCI scores are methodically consistent and comparable across years.\n\n\nThe final source, UNCTAD’s Export Concentration Ratio (also known as the Herfindahl-Hirschmann Index),\nmeasures a country’s export market concentration as follows: <sup>𝑥</sup>\n\n𝑋\n\n\n\n𝐻𝐻𝑐𝑐 =\n\n\n\n~~<u>𝑛</u>~~\n~~<u>𝑖</u>~~ 𝑖 ~~<u>=1</u>~~\n\n~~<u>�∑</u>~~ ~~<u>(</u>~~ <sup>~~<u>𝑥𝑖</u>~~</sup> <sup>����</sup>\n\n~~<u>𝑋</u>~~ 𝑋\n\n\n\n~~<u>𝑛</u>~~\n~~<u>𝑖</u>~~ 𝑖 ~~<u>=1</u>~~\n\n~~<u>(</u>~~ <sup>~~<u>𝑥</u>~~</sup> ~~<u>𝑋</u>~~ <sup>~~<u>𝑖</u>~~</sup> <sup>𝑥</sup> 𝑋 <sup>~~<u>)2</u>~~</sup> <u>−</u> ~~<u>�1/𝑛</u>~~\n\n~~1 −�1/𝑛~~\n\n\n\n𝑖\n\n~~<u>\n\nMention: \"UNCTAD’s Export Concentration Ratio\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Export market concentration is computed from this ratio as a Herfindahl index.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000977:11:0", "occurrence_id": "13d123d792d0fb74c026adc5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 3.3. Application to Bhutan\n\n\n**_Modeling of Market Income_**\n\n\nThe most challenging part of the fiscal incidence analysis for Bhutan is the lack of information on labor\nand nonlabor income. Previous studies have dealt with this in several different ways. A popular approach\nis to use the survey-to-survey imputation to match the consumption survey with an income survey—this\nwas done in Argentina (Lopez Del Valle et al. 2021), Ecuador (Llerena Pinto et al. 2015) and Paraguay\n(Canavire et al. 2021), for example. The feasibility and success of this approach depends on a number of\nfactors, including the availability of income and consumption surveys that were conducted not too far apart\nin time and the existence of a sufficient number of common variables. While Labour Force Surveys are\nconducted annually in Bhutan, the questions are less suited to capture the earnings from the agriculture\nsector which is the most important source of livelihoods for the rural poor.\n\n\nAnother approach is to impute market income by assuming disposable income equals consumption\naggregates, which is a reasonable assumption in a lower-middle-income country as it may be rather\nuncommon to have large savings, when information on direct transfers and taxes is available, . However,\nusing this approach directly for Bhutan is challenging since there is no information in the BLSS on direct\ntaxes and social security. Moreover, it is difficult to attribute aggregate household consumption into\nindividual-level consumption unless some simplifying assumptions are made, such as dividing household\nconsumption or income by the number of workers to get an estimate of average individual income (see Lara\nIbarra et al. 2019 for the Arab Republic of Egypt and Hill et al. 2017 for Ethiopia). Since this method would\nnot reflect income differences between household members, it could lead to an imprecise estimate of\nincome tax liabilities.\n\n\nWe develop a more sophisticated approach for Bhutan by adopting a consumption-based approach to\nmodeling market income. Specifically,\n\nMention: \"Labour Force Surveys\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"They are fielded annually but their questions poorly capture agricultural earnings, the main rural livelihood.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000977:11:0", "occurrence_id": "53f63b0befef08183ebf9de2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 3.3. Application to Bhutan\n\n\n**_Modeling of Market Income_**\n\n\nThe most challenging part of the fiscal incidence analysis for Bhutan is the lack of information on labor\nand nonlabor income. Previous studies have dealt with this in several different ways. A popular approach\nis to use the survey-to-survey imputation to match the consumption survey with an income survey—this\nwas done in Argentina (Lopez Del Valle et al. 2021), Ecuador (Llerena Pinto et al. 2015) and Paraguay\n(Canavire et al. 2021), for example. The feasibility and success of this approach depends on a number of\nfactors, including the availability of income and consumption surveys that were conducted not too far apart\nin time and the existence of a sufficient number of common variables. While Labour Force Surveys are\nconducted annually in Bhutan, the questions are less suited to capture the earnings from the agriculture\nsector which is the most important source of livelihoods for the rural poor.\n\n\nAnother approach is to impute market income by assuming disposable income equals consumption\naggregates, which is a reasonable assumption in a lower-middle-income country as it may be rather\nuncommon to have large savings, when information on direct transfers and taxes is available, . However,\nusing this approach directly for Bhutan is challenging since there is no information in the BLSS on direct\ntaxes and social security. Moreover, it is difficult to attribute aggregate household consumption into\nindividual-level consumption unless some simplifying assumptions are made, such as dividing household\nconsumption or income by the number of workers to get an estimate of average individual income (see Lara\nIbarra et al. 2019 for the Arab Republic of Egypt and Hill et al. 2017 for Ethiopia). Since this method would\nnot reflect income differences between household members, it could lead to an imprecise estimate of\nincome tax liabilities.\n\n\nWe develop a more sophisticated approach for Bhutan by adopting a consumption-based approach to\nmodeling market income. Specifically,\n\nMention: \"BLSS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"It carries no information on direct taxes or social security, complicating the income approach.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000980:4:0", "occurrence_id": "00e6bd3b66a7ce45c55c3f16", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: creating more middle-class jobs and harnessing the economic potential of women to build back\n\n\nbetter going forward.\n\n\nThis paper measures the gendered impacts of the COVID-19 pandemic on employment in\n\n\nIndonesia. While the existing literature is replete with studies that focus on the pandemic’s effects\n\n\non female employment in developed country contexts (Couch, Fairlie and Xu, 2020; Alon et al.,\n\n\n2021; Andrew et al., 2020; Adams-Prassl et al., 2020; Albanesi and Kim, 2021; Kikuchi et al.,\n\n\n2020), few studies focus on developing countries, particularly in the Southeast Asia region.\n\n\nMoreover, this paper also studies the pandemic’s effect on the structure of Indonesia’s labor market\n\n\nby analyzing its heterogeneous impacts on employment across sectors, occupations and population\n\n\nsubgroups. In measuring the pandemic’s impact, we employ a seasonal difference-in-differences\n\n\napproach. This approach is akin to that used in identifying “excess deaths” during the pandemic\n\n\n(Karlinsky and Kobak, 2021), in that the pre-pandemic between-season variations are used as a\n\n\ncounterfactual for the between-season variations observed during the pandemic. It then becomes\n\n\npossible to attribute the difference between the actual and counterfactual variations to the impact\n\n\nof the pandemic.\n\n\nOur empirical strategy relies on two identifying assumptions. First, we leverage the\n\n\nexogenous timing of the COVID-19 pandemic since it was largely unanticipated. Second, we\n\n\nassume that, in the absence of the pandemic, the between-season employment outcome variations\n\n\nin 2020 would largely follow the trends observed in the preceding years. For our analysis, we use\n\n\nIndonesia’s nationally representative National Labor Force Survey (Sakernas <sup>2</sup> ), administered\n\n\nbiannually by Statistics Indonesia in February and August. We pool all observations from the\n\n\n2 Sakernas stands for _Survey Angkatan Kerja Nasional_ .\n\n\n3\n\nMention: \"National Labor Force Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Nationally representative employment outcomes are pooled from two rounds a year of this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000980:4:0", "occurrence_id": "66ba3e83e28b8aed36a6cf6c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: creating more middle-class jobs and harnessing the economic potential of women to build back\n\n\nbetter going forward.\n\n\nThis paper measures the gendered impacts of the COVID-19 pandemic on employment in\n\n\nIndonesia. While the existing literature is replete with studies that focus on the pandemic’s effects\n\n\non female employment in developed country contexts (Couch, Fairlie and Xu, 2020; Alon et al.,\n\n\n2021; Andrew et al., 2020; Adams-Prassl et al., 2020; Albanesi and Kim, 2021; Kikuchi et al.,\n\n\n2020), few studies focus on developing countries, particularly in the Southeast Asia region.\n\n\nMoreover, this paper also studies the pandemic’s effect on the structure of Indonesia’s labor market\n\n\nby analyzing its heterogeneous impacts on employment across sectors, occupations and population\n\n\nsubgroups. In measuring the pandemic’s impact, we employ a seasonal difference-in-differences\n\n\napproach. This approach is akin to that used in identifying “excess deaths” during the pandemic\n\n\n(Karlinsky and Kobak, 2021), in that the pre-pandemic between-season variations are used as a\n\n\ncounterfactual for the between-season variations observed during the pandemic. It then becomes\n\n\npossible to attribute the difference between the actual and counterfactual variations to the impact\n\n\nof the pandemic.\n\n\nOur empirical strategy relies on two identifying assumptions. First, we leverage the\n\n\nexogenous timing of the COVID-19 pandemic since it was largely unanticipated. Second, we\n\n\nassume that, in the absence of the pandemic, the between-season employment outcome variations\n\n\nin 2020 would largely follow the trends observed in the preceding years. For our analysis, we use\n\n\nIndonesia’s nationally representative National Labor Force Survey (Sakernas <sup>2</sup> ), administered\n\n\nbiannually by Statistics Indonesia in February and August. We pool all observations from the\n\n\n2 Sakernas stands for _Survey Angkatan Kerja Nasional_ .\n\n\n3\n\nMention: \"Sakernas\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"The local name and its expansion identify the survey and its biannual February and August fielding.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000987:62:0", "occurrence_id": "015b9b1431ccd1da2797dc51", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <u>Medium-term ECT-R2 - Hours worked last week</u>\nStandardized Standardized differences<u>differences- Raw</u> <u>Weighted</u>\n\n\n\nVariance\n<u>Ratio- Raw</u>\n\n\n\nStandardized differences\n\n\nVariance Ratio\n<u>Weighted</u>\n\n\n\n<u>Weighted</u>\n\n\n\nFemale 0.365174 -0.01629 0.824682 1.003837\n\n\nAge 0.111013 0.090337 0.870673 1.052575\nHousehold dependency\n\nratio 0.194286 -0.1805 3.234074 1.336616\n\n\nHousehold head 0.001935 -0.03197 1.022384 1.01169\nCompleted Secondary\n\nSchool -0.16927 0.055904 0.826611 1.056511\nMain income source:\n\nBusiness 0.254888 -0.17184 0.825316 1.073115\nMain income source:\n\nWage/Salary -0.23946 0.068772 0.663537 1.096681\n\n\nAsset score -0.31364 0.007725 0.67199 0.911279\n\n\nBombali -0.13839 0.267097 0.740313 1.52662\n\n\nKenema -0.1917 0.026184 0.613047 1.058022\n\n\nPort Loko 0.128681 -0.03446 1.52749 0.883245\n\n\nWestern Area Urban 0.254694 -0.16893 0.959249 0.977487\n\n\nOveridentification test for covariate balance\n\n\nChi-square 8.025189\n\n\n<u>P-value</u> <u>0.841955</u>\n_<mark>Source</mark>_ <mark>: Authors’ calculations from CIMS (2020)</mark>\n\n\n\n61\n\nMention: \"CIMS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Covariate balance and hours worked are computed from this integrated household survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000988:21:0", "occurrence_id": "58e0d6d43a41f78464793985", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 1. Timeline of Implementation and Data Collection**\n\n\n**Figure 1. Percent of Treatment Students that used ALEKS at least one Minute**\n\n\n\n\n\n\n\n\n\nSource: Authors using take-up data from ALEKS software.\n\n\n20\n\nMention: \"take-up data from ALEKS software\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Software logs show the share of treatment students who used the programme at all.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000989:10:2", "occurrence_id": "4222813f902b15d717602354", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: =en&segment=indicator&id=EMP_DWAP_SEX_AGE_RT_A)</mark></u> and <u><mark>[World Bank (2020)](https://datacatalog.worldbank.org/dataset/global-jobs-indicators-database)</mark></u>, together with the expected long-run return to education. We\nalso compute aggregate results by bringing all expected earnings losses to their present value, assuming\na work life of 45 years and a 3% discount rate. To make these results more realistic, we also adjust the\naggregate loss by the expected adult survival rate (following the World Bank HCI) and the fact that not\nall workers will always be in gainful employment.\n\nSee Annex C for the percentage of population covered for the key outcome variables.\n###### Outcome 1. Learning Poverty\n\nIn the previous release of Learning Poverty results, baseline Learning Poverty was 52.7% for low- and\nmiddle-income <sup>19</sup> [^19: The estimation for low- and middle-income countries refers to the estimation for World Bank Part 2 countries, which also include\nthe following high-income countries for Learning Poverty calculations: Chile, Croatia, Panama, Poland, and Trinidad and Tobago.\nOther high-income Part 2 countries include Antigua and Barbuda, Palau, the Seychelles, St. Kitts and Nevis, and Uruguay.] countries in 2015. In the current release, baseline global Learning Poverty for these\n\n\n18 [To learn more about the Learning Poverty indicator update see Azevedo, Montoya et al. (2021).](http://hdl.handle.net/10986/36082) For Learning Poverty, expected\nlearning gains are adjusted according to the standard deviation for the assessment for each country. See annex E1 for all assessments\nand standard deviations used.\n19 The estimation for low- and middle-income countries refers to the estimation for World Bank Part 2 countries, which also include\nthe following high-income countries for Learning Poverty calculations: Chile,\n\nMention: \"global-jobs-indicators-database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Employment rates by sex and age are combined with expected returns to education to compute earnings losses.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000990:31:0", "occurrence_id": "c4697ddfe46a67bf41838daf", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Table A5:** Effect of female assignment on chat-based and admin-based purchase outcomes\n\n\n<u>Any chat-based</u> <u>Any admin-based</u> <u>Combined purchases</u>\n\n\n(1) (2) (3) (4) (5)\nAll (48h) (24h) (48h) (24h)\n\nFemale -.018 <sup>_∗∗∗_</sup> -.02 <sup>_∗∗_</sup> -.018 <sup>_∗∗_</sup> -.039 <sup>_∗∗∗_</sup> -.037 <sup>_∗∗∗_</sup>\n\n(.0067) (.0097) (.0087) (.012) (.011)\n\n\nControl Mean (wt) .033 .043 .037 .076 .070\nN 2653 2653 2653 2653 2653\n\n\nThis table shows the effect of female name assignment on purchase outcomes for administrative, non-administrative, and combined purchases. Non-administrative purchase records\nare recovered from chats, as they are purchases made by agents for customers. Administrative purchases are purchases made by customers observed through administrative records.\nPurchases are measured within 24 or 48 hours of the start of the chat. Female indicator determined in customer’s first chat of the day. Controls include agent-month, customer location, customer purchase history, and customer chat history fixed effects. The control group\nmean is reweighted by fixed effects cells to match the implied agent-month fixed effects-only\nOLS weights. Standard errors two-way clustered at the agent-day and customer-day level.\n_∗_ _p <_ 0.10, _∗∗_ _p <_\n\nMention: \"Non-administrative purchase records\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Purchases agents made for customers are recovered from chat transcripts as a separate outcome.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000990:31:0", "occurrence_id": "a78fa6fc22249b4b2beaef43", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Table A5:** Effect of female assignment on chat-based and admin-based purchase outcomes\n\n\n<u>Any chat-based</u> <u>Any admin-based</u> <u>Combined purchases</u>\n\n\n(1) (2) (3) (4) (5)\nAll (48h) (24h) (48h) (24h)\n\nFemale -.018 <sup>_∗∗∗_</sup> -.02 <sup>_∗∗_</sup> -.018 <sup>_∗∗_</sup> -.039 <sup>_∗∗∗_</sup> -.037 <sup>_∗∗∗_</sup>\n\n(.0067) (.0097) (.0087) (.012) (.011)\n\n\nControl Mean (wt) .033 .043 .037 .076 .070\nN 2653 2653 2653 2653 2653\n\n\nThis table shows the effect of female name assignment on purchase outcomes for administrative, non-administrative, and combined purchases. Non-administrative purchase records\nare recovered from chats, as they are purchases made by agents for customers. Administrative purchases are purchases made by customers observed through administrative records.\nPurchases are measured within 24 or 48 hours of the start of the chat. Female indicator determined in customer’s first chat of the day. Controls include agent-month, customer location, customer purchase history, and customer chat history fixed effects. The control group\nmean is reweighted by fixed effects cells to match the implied agent-month fixed effects-only\nOLS weights. Standard errors two-way clustered at the agent-day and customer-day level.\n_∗_ _p <_ 0.10, _∗∗_ _p <_\n\nMention: \"administrative records\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Customer purchases observed directly in platform records form the administrative outcome.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000994:49:0", "occurrence_id": "0e6c4cd4af7286822d74614f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure** **A2:** Cross-sectional average earnings and transfers\n\n\nfactory earnings, both in the 3 months prior to the survey (in FCFA).\nEarnings from factory administrative data. Transfers from 2 rounds of\nworker phone surveys. Amount transferred in the 15 days prior to the\nsurvey. Observations residualized from survey fixed effects. Transfers\ntop-coded at the 99th percentile. Line of best linear fit reported. 349\nobservations, from workers not offered a Private savings account.\n\n\n47\n\nMention: \"factory administrative data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Earnings in the three months before the survey are taken from employer records.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000997:13:0", "occurrence_id": "b225014c7e74bfd1b686437b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The regions of Bongolava and Sofia have the highest percentages of children weighing less than 2.5\nkg, with 19% and 17%, respectively. On the other hand, the regions of Ihorombe and Anosy have the\nlowest percentages of low-birth-weight children, at around 8%. In other words, stunting seems\nprevalent in the least poor regions of Madagascar and less prevalent in the poorest regions. These\nfindings emphasize the importance of regional context in addressing child undernutrition, highlighting\nthe need for tailored approaches that consider socioeconomic factors, health care access, and\nnutritional practices to improve child health outcomes across Madagascar.\n\nBut these heterogeneities also exist within regions at the commune level. We therefore applied the\nSAE approach in the DHS data to produce stunting estimates at the commune level and produce the\nmap presented under figure 3. The maps highlight higher stunting rates among communes located in\nhigh plateau of the regions in the center, also known for its better agro-ecological advantage for\nagricultural production. We also note the higher stunting rates among communes in the regions in the\neast and southeast of the country.\n\nLinguistic, historical, and DNA research suggests that the central highlands of Madagascar were settled\nby people with an Austronesian background 1,200 years ago (most likely South Borneo); the coastal\nareas were settled later by people mainly from East Africa (Hurles et al 2005). Can this partly explain\nthe stunting patterns? Our limited data does not allow us to verify this hypothesis.\n\nThe variables selected to predict stunting as well as the resulting coefficients from the linear mixed\nmodel is provided in table A2 of the annex. Table 4 presents a set of model diagnostics relating to\nthe fit of the model and the consistency of the normality assumptions with the sample data. Overall,\nthe model explains 7.5 percent of the variation in heigh for age Z score when not considering the\nconditional random effect, and 11.3 percent when the\n\nMention: \"DHS data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"A small-area estimation approach is applied to produce stunting estimates at commune level.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000998:7:0", "occurrence_id": "5c5e58c5b3bf21afa1e3db7d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: children in Nigeria (Azomahou and Yitbarek 2016; Razzu and Wambile 2021). Northern\n\nNigeria has some of the poorest educational attainment rates for older adults. Only 40\n\npercent of 30-34-year-olds have ever attended school in the North East and North West\n\nzones, compared to 90 percent in the South East and South West regions of the country.\n\n\nAnother factor that may contribute to low school enrollment and deficient learning\n\noutcomes is that the region is also characterized by a strong adherence to traditional\n\nnorms. The formal legal institution of Shari’a law applies in most northern states, cover\ning social, civil, and criminal matters (Suberu 2009; Human Right Watch 2014). This has\n\nreinforced social norms that encourage early marriage and childbearing for Muslim ado\nlescents and in the process create additional barriers to education. This is also reflected in\n\nattitudes toward education for women and labor force participation of mothers. Accord\ning to the World Values Survey (2017 -2021), in Nigeria, 42% of respondents believes that\n\nuniversity is more important for a boy than for a girl and 41% believes that pre-school\n\nchildren suffer if the mother works. These adverse norms contrast with those observed in\n\nother countries where education plays a more prominent role, such as Kenya, where the\n\nproportions of the population holding these views are 18% and 23%, respectively.\n\n\nFurther, the emergence of the militant terrorist group Boko Haram, which translates\n\nas ”Western education is forbidden”, creates an additional barrier for school enrollment\n\nand attendance. As a result, the predominant forms of schooling available to young chil\ndren in the north include: 1) Qur’anic schools, which exclusively focus lessons on Islamic\n\ntheology and religious texts, or 2) Islamiyya schools, which also cover Islamic subjects\nbut follow a more typical school structure with regularized time schedules. <sup>2</sup>\n\n###\n\nMention: \"World Values Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Shares holding restrictive views on women's education and maternal work are reported from this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000999:7:0", "occurrence_id": "cd65dde063777ff99698847b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: way democracy works in El Salvador?” Again the main results do not depend on dichotomization\n\ndecisions.\n\n\nTable 1 shows some basic summary statistics for the aforementioned dependent variables as\n\nwell as for some set of individual-level characteristics. Panel A focuses on LAPOP data while\nPanel B on Latinobar´ometro data. My samples of analysis come from pooling different waves\n\navailable for the time period 2004-2018. Note that the total number of observations for each\n\nvariable vary because some of the questions were not asked every wave and, in some cases, when\n\nasked some respondents choose to not answer. Roughly half of the respondents are men (48%)\n\nand the average age is 39 in both samples. In the LAPOP sample, 38% are individuals living in\n\nrural areas while 18% of the individuals in the Latinobar´ometro sample lives in small urban areas.\n\nAverage years of schooling is above 8 years while, depending on the sample, between 7% and 9%\n\nof the individuals are unemployed (i.e,individuals who are not working and look for a job).\n\n\nThe LAPOP sample reveals substantial heterogeneity in the level of trust depending on which\n\ninstitutions one looks at. According to the respondents, the Catholic Church is the most trusted\n\ninstitution (60% of Salvadoran strongly trust in the Church) followed by mass media and municipal\n\ngovernment (both with 56%). On the other hand, the national legislature (with 39%) and the\n\npresident (with 45%) are the less trusted over the time period of analysis. Finally, only 38%\n\nof Salvadorans think that the judicial system would punish the guilty if they were a victim of a\n\nrobbery or assault.\n\n\nThe Latinobar´ometro sample shows that 52% of respondents thinks that income distribution in\n\nEl Salvador is “unfair” or “very unfair”. Few Salvadorans are optimistic about current economic\n\nconditions (only 16% answered that the economic conditions of the\n\nMention: \"LAPOP data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Summary statistics for the democracy attitudes variables are reported for this survey's pooled waves.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:000999:7:0", "occurrence_id": "e1005902fe3d1c61b1199515", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: way democracy works in El Salvador?” Again the main results do not depend on dichotomization\n\ndecisions.\n\n\nTable 1 shows some basic summary statistics for the aforementioned dependent variables as\n\nwell as for some set of individual-level characteristics. Panel A focuses on LAPOP data while\nPanel B on Latinobar´ometro data. My samples of analysis come from pooling different waves\n\navailable for the time period 2004-2018. Note that the total number of observations for each\n\nvariable vary because some of the questions were not asked every wave and, in some cases, when\n\nasked some respondents choose to not answer. Roughly half of the respondents are men (48%)\n\nand the average age is 39 in both samples. In the LAPOP sample, 38% are individuals living in\n\nrural areas while 18% of the individuals in the Latinobar´ometro sample lives in small urban areas.\n\nAverage years of schooling is above 8 years while, depending on the sample, between 7% and 9%\n\nof the individuals are unemployed (i.e,individuals who are not working and look for a job).\n\n\nThe LAPOP sample reveals substantial heterogeneity in the level of trust depending on which\n\ninstitutions one looks at. According to the respondents, the Catholic Church is the most trusted\n\ninstitution (60% of Salvadoran strongly trust in the Church) followed by mass media and municipal\n\ngovernment (both with 56%). On the other hand, the national legislature (with 39%) and the\n\npresident (with 45%) are the less trusted over the time period of analysis. Finally, only 38%\n\nof Salvadorans think that the judicial system would punish the guilty if they were a victim of a\n\nrobbery or assault.\n\n\nThe Latinobar´ometro sample shows that 52% of respondents thinks that income distribution in\n\nEl Salvador is “unfair” or “very unfair”. Few Salvadorans are optimistic about current economic\n\nconditions (only 16% answered that the economic conditions of the\n\nMention: \"Latinobar´ometro data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its waves supply the second set of dependent variables and individual characteristics.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001000:46:0", "occurrence_id": "2ade3f9d6ca71c82730cc53e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: FIGURE D.1: Rent in surveyed households by exposure to floods, Accra\n\n\n_Source_ : Authors’ calculation based on Disaster-Poverty Household Survey and Barzin et al. (2022) data.\n\n\n45\n\nMention: \"Disaster-Poverty Household Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Rent in flooded and non-flooded households is computed from this disaster-affected household survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001002:26:5", "occurrence_id": "1a6bb6d392532b401fc4eb66", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 6-9<br>10-19<br>20 or more<br>1-5<br>6-9<br>10-19<br>20 or more<br>1-5<br>6-9<br>10-49<br>50-100<br>101 or more|Encuesta Permanente de Hogares 2019<br>Encuesta de Hogares 2019<br>Pesquisa Nacional por Amostra de Domic´ılios Cont´ınua 2019<br>Encuesta de Caracterizaci´on Socioecon´omica Nacional 2017<br>Gran Encuesta Integrada de Hogares 2019<br>Encuesta Nacional de Hogares 2019<br>Encuesta Nacional Continua de la Fuerza de Trabajo 2019<br>Encuesta Nacional de Ingresos y Gastos de los Hogares 2018<br>Encuesta Permanente de Hogares Continua 2019<br>Encuesta Nacional de Hogares 2019<br>Encuesta Continua de Hogares 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<\n\nMention: \"Encuesta Permanente de Hogares 2019\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Named as the national household survey supplying the harmonised employment-size distribution.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001002:26:5", "occurrence_id": "670d46203127bbb5ce728e7f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 6-9<br>10-19<br>20 or more<br>1-5<br>6-9<br>10-19<br>20 or more<br>1-5<br>6-9<br>10-49<br>50-100<br>101 or more|Encuesta Permanente de Hogares 2019<br>Encuesta de Hogares 2019<br>Pesquisa Nacional por Amostra de Domic´ılios Cont´ınua 2019<br>Encuesta de Caracterizaci´on Socioecon´omica Nacional 2017<br>Gran Encuesta Integrada de Hogares 2019<br>Encuesta Nacional de Hogares 2019<br>Encuesta Nacional Continua de la Fuerza de Trabajo 2019<br>Encuesta Nacional de Ingresos y Gastos de los Hogares 2018<br>Encuesta Permanente de Hogares Continua 2019<br>Encuesta Nacional de Hogares 2019<br>Encuesta Continua de Hogares 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<\n\nMention: \"Encuesta de Hogares 2019\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Named as the national household survey supplying the harmonised employment-size distribution.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001002:26:5", "occurrence_id": "e0e357abf965c560288c8931", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 6-9<br>10-19<br>20 or more<br>1-5<br>6-9<br>10-19<br>20 or more<br>1-5<br>6-9<br>10-49<br>50-100<br>101 or more|Encuesta Permanente de Hogares 2019<br>Encuesta de Hogares 2019<br>Pesquisa Nacional por Amostra de Domic´ılios Cont´ınua 2019<br>Encuesta de Caracterizaci´on Socioecon´omica Nacional 2017<br>Gran Encuesta Integrada de Hogares 2019<br>Encuesta Nacional de Hogares 2019<br>Encuesta Nacional Continua de la Fuerza de Trabajo 2019<br>Encuesta Nacional de Ingresos y Gastos de los Hogares 2018<br>Encuesta Permanente de Hogares Continua 2019<br>Encuesta Nacional de Hogares 2019<br>Encuesta Continua de Hogares 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<\n\nMention: \"Pesquisa Nacional por Amostra de Domic\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The continuous national household survey is one of the country sources in the compilation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001002:26:5", "occurrence_id": "60cf947c33d44701794223dc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 6-9<br>10-19<br>20 or more<br>1-5<br>6-9<br>10-19<br>20 or more<br>1-5<br>6-9<br>10-49<br>50-100<br>101 or more|Encuesta Permanente de Hogares 2019<br>Encuesta de Hogares 2019<br>Pesquisa Nacional por Amostra de Domic´ılios Cont´ınua 2019<br>Encuesta de Caracterizaci´on Socioecon´omica Nacional 2017<br>Gran Encuesta Integrada de Hogares 2019<br>Encuesta Nacional de Hogares 2019<br>Encuesta Nacional Continua de la Fuerza de Trabajo 2019<br>Encuesta Nacional de Ingresos y Gastos de los Hogares 2018<br>Encuesta Permanente de Hogares Continua 2019<br>Encuesta Nacional de Hogares 2019<br>Encuesta Continua de Hogares 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<\n\nMention: \"Encuesta de Caracterizaci\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The socio-economic characterisation survey supplies one country's observations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001002:26:5", "occurrence_id": "3ba3bc8eaf17f7abcac59ff1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 6-9<br>10-19<br>20 or more<br>1-5<br>6-9<br>10-19<br>20 or more<br>1-5<br>6-9<br>10-49<br>50-100<br>101 or more|Encuesta Permanente de Hogares 2019<br>Encuesta de Hogares 2019<br>Pesquisa Nacional por Amostra de Domic´ılios Cont´ınua 2019<br>Encuesta de Caracterizaci´on Socioecon´omica Nacional 2017<br>Gran Encuesta Integrada de Hogares 2019<br>Encuesta Nacional de Hogares 2019<br>Encuesta Nacional Continua de la Fuerza de Trabajo 2019<br>Encuesta Nacional de Ingresos y Gastos de los Hogares 2018<br>Encuesta Permanente de Hogares Continua 2019<br>Encuesta Nacional de Hogares 2019<br>Encuesta Continua de Hogares 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<\n\nMention: \"Gran Encuesta Integrada de Hogares 2019\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The integrated household survey supplies one country's observations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001002:26:5", "occurrence_id": "317acfd360b0e6d48d221809", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 6-9<br>10-19<br>20 or more<br>1-5<br>6-9<br>10-19<br>20 or more<br>1-5<br>6-9<br>10-49<br>50-100<br>101 or more|Encuesta Permanente de Hogares 2019<br>Encuesta de Hogares 2019<br>Pesquisa Nacional por Amostra de Domic´ılios Cont´ınua 2019<br>Encuesta de Caracterizaci´on Socioecon´omica Nacional 2017<br>Gran Encuesta Integrada de Hogares 2019<br>Encuesta Nacional de Hogares 2019<br>Encuesta Nacional Continua de la Fuerza de Trabajo 2019<br>Encuesta Nacional de Ingresos y Gastos de los Hogares 2018<br>Encuesta Permanente de Hogares Continua 2019<br>Encuesta Nacional de Hogares 2019<br>Encuesta Continua de Hogares 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<\n\nMention: \"Encuesta Nacional de Hogares 2019\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The national household survey supplies one country's observations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001002:26:5", "occurrence_id": "f6fc089a8207cad82ed6bd73", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 6-9<br>10-19<br>20 or more<br>1-5<br>6-9<br>10-19<br>20 or more<br>1-5<br>6-9<br>10-49<br>50-100<br>101 or more|Encuesta Permanente de Hogares 2019<br>Encuesta de Hogares 2019<br>Pesquisa Nacional por Amostra de Domic´ılios Cont´ınua 2019<br>Encuesta de Caracterizaci´on Socioecon´omica Nacional 2017<br>Gran Encuesta Integrada de Hogares 2019<br>Encuesta Nacional de Hogares 2019<br>Encuesta Nacional Continua de la Fuerza de Trabajo 2019<br>Encuesta Nacional de Ingresos y Gastos de los Hogares 2018<br>Encuesta Permanente de Hogares Continua 2019<br>Encuesta Nacional de Hogares 2019<br>Encuesta Continua de Hogares 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<\n\nMention: \"Encuesta Nacional Continua de la Fuerza de Trabajo 2019\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The continuous labour force survey supplies one country's observations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001002:26:5", "occurrence_id": "b752aebefb87c56fe8683cd8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 6-9<br>10-19<br>20 or more<br>1-5<br>6-9<br>10-19<br>20 or more<br>1-5<br>6-9<br>10-49<br>50-100<br>101 or more|Encuesta Permanente de Hogares 2019<br>Encuesta de Hogares 2019<br>Pesquisa Nacional por Amostra de Domic´ılios Cont´ınua 2019<br>Encuesta de Caracterizaci´on Socioecon´omica Nacional 2017<br>Gran Encuesta Integrada de Hogares 2019<br>Encuesta Nacional de Hogares 2019<br>Encuesta Nacional Continua de la Fuerza de Trabajo 2019<br>Encuesta Nacional de Ingresos y Gastos de los Hogares 2018<br>Encuesta Permanente de Hogares Continua 2019<br>Encuesta Nacional de Hogares 2019<br>Encuesta Continua de Hogares 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<\n\nMention: \"Encuesta Nacional de Ingresos y Gastos de los Hogares 2018\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The national income and expenditure survey supplies one country's observations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001002:26:5", "occurrence_id": "7e7fa1f876a7c811b4256dbd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 6-9<br>10-19<br>20 or more<br>1-5<br>6-9<br>10-19<br>20 or more<br>1-5<br>6-9<br>10-49<br>50-100<br>101 or more|Encuesta Permanente de Hogares 2019<br>Encuesta de Hogares 2019<br>Pesquisa Nacional por Amostra de Domic´ılios Cont´ınua 2019<br>Encuesta de Caracterizaci´on Socioecon´omica Nacional 2017<br>Gran Encuesta Integrada de Hogares 2019<br>Encuesta Nacional de Hogares 2019<br>Encuesta Nacional Continua de la Fuerza de Trabajo 2019<br>Encuesta Nacional de Ingresos y Gastos de los Hogares 2018<br>Encuesta Permanente de Hogares Continua 2019<br>Encuesta Nacional de Hogares 2019<br>Encuesta Continua de Hogares 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<\n\nMention: \"Encuesta Permanente de Hogares Continua 2019\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The continuous permanent household survey supplies one country's observations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001002:26:5", "occurrence_id": "5885755716cb366d578e1b76", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 6-9<br>10-19<br>20 or more<br>1-5<br>6-9<br>10-19<br>20 or more<br>1-5<br>6-9<br>10-49<br>50-100<br>101 or more|Encuesta Permanente de Hogares 2019<br>Encuesta de Hogares 2019<br>Pesquisa Nacional por Amostra de Domic´ılios Cont´ınua 2019<br>Encuesta de Caracterizaci´on Socioecon´omica Nacional 2017<br>Gran Encuesta Integrada de Hogares 2019<br>Encuesta Nacional de Hogares 2019<br>Encuesta Nacional Continua de la Fuerza de Trabajo 2019<br>Encuesta Nacional de Ingresos y Gastos de los Hogares 2018<br>Encuesta Permanente de Hogares Continua 2019<br>Encuesta Nacional de Hogares 2019<br>Encuesta Continua de Hogares 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<br>EU SILC 2019<\n\nMention: \"Encuesta Continua de Hogares 2019\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The continuous household survey supplies one country's observations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001003:26:1", "occurrence_id": "bc734b335f2c1a94ea5ede6b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: > <u>36.2</u>\n\n\n\nSource: Author calculations based on HIES 2018-19 data\n\n\n**<mark>Comparability of expenditure and income data.</mark>** <mark>In most developing countries, consumption expenditure</mark>\n<mark>is used as a proxy for welfare. Theoretically, consumption is likely to remain more stable over time than</mark>\n<mark>income, which can be affected by seasonality and other shocks. Practically, it is extremely difficult to</mark>\n<mark>collect accurate income data in developing economies with high informality, a large prevalence of self-</mark>\n<mark>employment or own-account workers, reliance on foreign remittances, in addition to smallholder</mark>\n<mark>agricultural production for home use. Income measures are especially problematic due to difficulties in</mark>\n<mark>estimating self-employment profits, falsified responses and underestimation (Deaton and Zaidi 2002;</mark>\n<mark>Deaton 2019). For these reasons, even though the HIES collects data on income, it is more reliable to</mark>\n<mark>continue using consumption as a proxy for income in the REM approach.</mark>\n\nFor comparison, household income aggregates were estimated using the data in HIES 2018-19. A\nhousehold’s total income is given by the sum of all the following sources:\n\n\n1) Work or job-related annual income from main, secondary and other occupations;\n2) In-kind payments for employment as well as transfers such as pensions;\n3) Main and secondary income from agriculture (including in-kind transfers) for those in\nagriculture;\n4) Unearned income sources including local and international remittances, Zakat/Ushr,\ninheritances, income from insurance, the Benazir Income Support Programme;\n5) Income earned (rents or sales)\n\nMention: \"HIES 2018-19\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Household income aggregates are built from its multiple income sources to compare with consumption.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001004:14:1", "occurrence_id": "da61439729485f5f64deaad8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: . Yet, formal remittances\n\n\nto several developing countries—including Mexico—increased substantially during the first months of the\n\n\npandemic.\n\n\nThis article produces evidence to explain this paradox. It contends that a sudden acceleration in the\n\n\nformalization of remittances partly explains the increase in recorded remittances observed in several de\n\nveloping countries during the first period of the COVID-19 pandemic. This stemmed from a prohibitive\n\n\nincrease in the costs of sending informal remittances due to mobility restrictions imposed to control con\n\ntagion. We use subnational data from a developing country, Mexico, to show that the increase in formal\n\n\nremittances was driven by municipalities closer to a US border crossing that had been more likely to re\n\nceive informal remittances before March 2020. Consistent with this hypothesis, these areas also experienced\n\n\na disproportionate increase in transaction bank accounts during this period. These accounts were proba\n\nbly opened in order to receive formal remittances via electronic transfer that would have come through\n\n\ninformal channels prior to the pandemic.\n\n\n22On January 31, 2020, the White House 2019 Novel Coronavirus Task Force announced the implementation of new travel policies\neffective on February 2, 2020. For a detailed timeline of the different events since December 2019, see the CDC museum timeline in\n[the following website.](https://www.cdc.gov/museum/timeline/covid19.html)\n\n\n12\n\nMention: \"CDC museum timeline\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"A detailed chronology of events since December 2019 is referenced for readers.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001013:11:0", "occurrence_id": "f8cbb1532a84ed86fdd47093", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: We contribute to this literature by tracking monthly changes in household expenditure\n\n\nand income following the Kerala floods. The results in section 6 show a persistent decline in\n\n\nhousehold expenditure, particularly non-essential expenditure, despite a quick recovery and\n\n\nsubsequent increase household income. The flood also affected the composition of household\n\n\ndebt and savings. In section 7, we show that households were more likely to borrow for\n\n\nhousing and medical purposes, and less likely to borrow for consumption.\n\n\n**4.** **Data,** **Samples,** **and** **Summary** **Statistics**\n\n\nOur analysis employs numerous district-level and household-level variables from various\n\n\nsources. This section describes the data, discusses the sample selection, and reports summary\n\n\nstatistics.\n\n\n**4.1** **Data**\n\n\n_Nighttime_ _Lights_\n\n\nNighttime lights are a widely used proxy for aggregate economic activity. <sup>14</sup> Felbermayr et\n\n\nal. (2022) use nighttime lights to investigate the impact of weather anomalies on economic\n\n\nactivity. In India, they have been used to analyze the spatial impact of demonetization\n\n\n(Chodorow-Reich et al. 2020), regional convergence (Chanda and Kabiraj 2020), and the\n\n\nheterogeneous impact of the COVID-19 pandemic (Beyer, Franco-Bedoya, and Galdo 2021,\n\n\nBeyer, Jain, and Sinha 2021). To translate changes in nighttime lights into changes in\n\n\neconomic activity, we rely on the quarterly elasticity estimated by Beyer, Hu, and Yao (2022).\n\n\nAlthough nighttime lights data are very noisy over time, they are useful for difference-in\n\ndifference analyses since most of the noise is common across the treated and untreated units.\n\n\nWe extract district level nighttime light data from the Visible and Infrared Imaging\n\n\nSuite Day Night Band (VIIRS-DNB) Cloud Free Monthly Composites (\n\nMention: \"Visible and Infrared Imaging\n\n\nSuite Day Night Band\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"District night-time light values are extracted to support the difference-in-differences design.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001013:11:0", "occurrence_id": "f1ec71dc7b0ecf5b5c378296", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: We contribute to this literature by tracking monthly changes in household expenditure\n\n\nand income following the Kerala floods. The results in section 6 show a persistent decline in\n\n\nhousehold expenditure, particularly non-essential expenditure, despite a quick recovery and\n\n\nsubsequent increase household income. The flood also affected the composition of household\n\n\ndebt and savings. In section 7, we show that households were more likely to borrow for\n\n\nhousing and medical purposes, and less likely to borrow for consumption.\n\n\n**4.** **Data,** **Samples,** **and** **Summary** **Statistics**\n\n\nOur analysis employs numerous district-level and household-level variables from various\n\n\nsources. This section describes the data, discusses the sample selection, and reports summary\n\n\nstatistics.\n\n\n**4.1** **Data**\n\n\n_Nighttime_ _Lights_\n\n\nNighttime lights are a widely used proxy for aggregate economic activity. <sup>14</sup> Felbermayr et\n\n\nal. (2022) use nighttime lights to investigate the impact of weather anomalies on economic\n\n\nactivity. In India, they have been used to analyze the spatial impact of demonetization\n\n\n(Chodorow-Reich et al. 2020), regional convergence (Chanda and Kabiraj 2020), and the\n\n\nheterogeneous impact of the COVID-19 pandemic (Beyer, Franco-Bedoya, and Galdo 2021,\n\n\nBeyer, Jain, and Sinha 2021). To translate changes in nighttime lights into changes in\n\n\neconomic activity, we rely on the quarterly elasticity estimated by Beyer, Hu, and Yao (2022).\n\n\nAlthough nighttime lights data are very noisy over time, they are useful for difference-in\n\ndifference analyses since most of the noise is common across the treated and untreated units.\n\n\nWe extract district level nighttime light data from the Visible and Infrared Imaging\n\n\nSuite Day Night Band (VIIRS-DNB) Cloud Free Monthly Composites (\n\nMention: \"Cloud Free Monthly Composites\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The monthly cloud-free product is the specific extract used.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001017:5:1", "occurrence_id": "1e0bf2a58497d053dc3fc7ca", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  some of its funders, through SMS messages and emails sent to a sample\nof firms in a Mexican government database, and through flyers in Mexico City. Guatemala\nwas added as a second country after five rounds of recruitment had taken place, in order\nto test the feasibility of further geographic expansion. Overall, 65% of the Mexican sample\nand 86% of the Guatemalan sample were recruited through Facebook.\n\n\nFacebook usage is high in Mexico, with an estimated 90 million users in 2022, which is 84\npercent of the population aged 10 and over. <sup>2</sup> This illustrates the potential for online recruitment to reach large numbers of microenterprises. To participate in the program, individuals\nhad to click on the advertisement and fill out a short pre-registration form indicating interest, and then they were invited to attend an online information session to find out more\nabout the course. They then registered by filling out a form that serves as our baseline data.\nCREA’s paid advertising campaigns were seen by 3.3 million unique viewers. However, as is\ntypical with online advertising, the conversion rate is low: 52,719 (1.6%) of viewers clicked\non the link, 10,700 pre-registered, and 2,208 registered for the program across all sources\n(1,478 from the Facebook ads). The estimated recruitment cost from advertising was $3.38\nper person in our sample.\n\n\nTable 1 provides summary statistics of the experimental sample, and compares them to a\n\n\n2User data from https://www.statista.com/statistics/282326/number-of-facebook-users-in-mexico/; population of 127 million and 16 percent aged under 10 and assumed to not be Facebook users for this calculation.\n\n\n4\n\nMention: \"Mexican government database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Firms listed in this register were contacted by text and email as part of recruitment.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001020:11:2", "occurrence_id": "06f65b2f38866f88697f0b4f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: br>0.00364|1.59<br>1.65<br>1.66<br>1.61<br>1.63<br>1.65|\n\n\n**Source** : Authors’ calculations based on HIECS 2015 and 2017/18 and Consumer Price Index (general number)\n2015 to 2018, CAPMAS.\n\nThe evolution of poverty was heterogeneous across different regions. Putting aside the Border regions,\nwhere the smaller sample size provides a less reliable picture, Table 6 shows that the largest increase in\npoverty took place in the Metropolitan area, where the poverty rate doubled. This is indeed the region\nin which total expenditures per capita underwent the largest decrease, from about 12,000 Egyptian\npounds in 2015 to about 9,600 Egyptian pounds in 2017/18. As a result, poverty increased from 11\npercent in 2015 to 25 percent in 2017/18. Poverty also increased in all other regions (although to a\nsmaller extent), except in Upper Rural Egypt. The latter region is the only one where real total\nexpenditure per capita did not decrease over the period. In fact, the poverty rate decreased by 3.3\npercentage points in Upper Rural Egypt. However, the decomposition reveals that this reduction is\nalmost entirely due to the reduction in the real poverty lines, and thus does not signal an increase in real\nconsumption in that region. Upper Rural Egypt remained by far the region with the highest poverty rate\nin 2017/18.\n\n\n10\n\nMention: \"HIECS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Poverty evolution across regions is computed from these two survey rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001020:11:2", "occurrence_id": "dfdd9a1903ffc0d92a8528f4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: br>0.00364|1.59<br>1.65<br>1.66<br>1.61<br>1.63<br>1.65|\n\n\n**Source** : Authors’ calculations based on HIECS 2015 and 2017/18 and Consumer Price Index (general number)\n2015 to 2018, CAPMAS.\n\nThe evolution of poverty was heterogeneous across different regions. Putting aside the Border regions,\nwhere the smaller sample size provides a less reliable picture, Table 6 shows that the largest increase in\npoverty took place in the Metropolitan area, where the poverty rate doubled. This is indeed the region\nin which total expenditures per capita underwent the largest decrease, from about 12,000 Egyptian\npounds in 2015 to about 9,600 Egyptian pounds in 2017/18. As a result, poverty increased from 11\npercent in 2015 to 25 percent in 2017/18. Poverty also increased in all other regions (although to a\nsmaller extent), except in Upper Rural Egypt. The latter region is the only one where real total\nexpenditure per capita did not decrease over the period. In fact, the poverty rate decreased by 3.3\npercentage points in Upper Rural Egypt. However, the decomposition reveals that this reduction is\nalmost entirely due to the reduction in the real poverty lines, and thus does not signal an increase in real\nconsumption in that region. Upper Rural Egypt remained by far the region with the highest poverty rate\nin 2017/18.\n\n\n10\n\nMention: \"Consumer Price Index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"The general index deflates nominal values across the survey rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001031:5:1", "occurrence_id": "a3cb0f22785d5bfa28112f30", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  vulnerable to various shocks and lack the capacity to mitigate their impact when\nthey occur. For these households, access to social protection can minimize long-term effects of using\nextreme coping strategies. However, data shows that majority of Sudanese households (particularly the\npoorest) rely on costly coping strategies such as using savings and/or selling assets (World Bank 2014, 24\nand 113). Furthermore, a recent report on the state of social protection in Sudan documents the uneven\ncoverage of existing programs and provides recommendations to expand coverage, improve targeting,\nand evaluate their effectiveness (Kjellgren et al. 2014). Thus, reforming SSN programs to increase access\nand coverage as well as restoration of macroeconomic stability is needed to strengthen the resilience and\nreduce poverty among these households.\n\n\nBox 1: Sudan’s Safety Nets and Social Protection Programs\n\n\n**At the time of the NHBPS 2014/15, existing government and parastatal transfers covered only a small fraction**\n**of the poor** . Food aid recipients were heavily concentrated in North Darfur, benefiting from well-targeted\ntransfers from humanitarian actors. Government transfers and transfers expended by Zakat, a parastatal religious\norganization, on the other hand, covered only a small fraction of the poor and lacked accuracy in targeting.\nHowever, the receipt of transfers, especially the amounts household received, is likely underestimated.\n\n\n**While food aid programs were reasonably targeted to the poor, government transfers, NGO/charity transfers,**\n**and Zakat transfers had comparatively low targeting accuracy** **_._** The bottom 40 percent accounted for around 60\npercent of the recipients of food aid (World Bank 2018). This compares to a share in consumption expenditure\namong this group of around 20 percent. Hence, food aid in Sudan is progressive and pro-poor. While pro-poor\n<u>and other types of transfers lack targeting accuracy, the bottom 40 account only accounted for\n\nMention: \"NHBPS 2014/15\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"It shows government and parastatal transfers reached only a small fraction of the poor.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001032:20:0", "occurrence_id": "76dcbedb52f6889dd738a983", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: - 19 \n\nvariables: About 14,000 and 8,000 observations are used for the inventory and agglomeration\n\nanalyses, respectively.\n\n\nOn the transport connectivity side, as discussed above, three variables are constructed: (i)\n\nconnectivity to the nearest city ( _CITY_ ), (ii) connectivity to the nearest boarder point ( _BP_ ), and\n\n(iii) market access index ( _MAI_ ). According to the World Population Review database, there are\n\n193 populated areas where more than 5,000 people live, which are defined as “cities” in this\n\npaper. <sup>12</sup> For border points, 18 locations are used ( **Table 1** ), at which the CPMM program\n\nregularly collect data on border crossing time and costs. Other border points exist but are\n\nexcluded from the analysis because no relevant data are available.\n\n\nUnlike the traditional dichotomous classification of beneficiaries and non-beneficiaries\n\ndepending on their proximity to improved roads, which is often used in earlier impact evaluation\n\nstudies on road development (e.g., Dercon et al., 2009; Khandker et al., 2009; IMF, 2020), the\n\ncurrent paper takes advantage of continuous variation of transport costs and travel times in the\n\nCPMM data. In addition, our connectivity variables are considered to be closer to actual costs\n\nand times, reflecting the reality on the ground. For instance, the duration of travel includes not\n\nonly drive time and border crossing time but also time spent for weight inspection, police check,\n\nloading and unloading. Transportation costs reflect not only shipping fees but also official and\n\nunofficial payments incurred during the trip, such as border clearance fees, customs and\n\n\n12 Available at https://worldpopulationreview.com/countries/cities/kazakhstan\n\nMention: \"World Population Review database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its counts of populated areas above 5,000 people define which places are treated as cities.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001036:21:0", "occurrence_id": "e1ecb84b66cd6e55f0d721ee", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Survey (ACS) one-year samples for 2010 and 2019 available from IPUMS USA (Ruggles et al. 2022). Data on educational outcomes of origin\ncountries are from the updated Barro-Lee Educational Attainment dataset (Barro and Lee, 2013), and GDP data are from the Penn World\nTables (PWT) 10.0 (Feenstra, Inklaar, and Timmer 2015).\n\n\n_Note_ : The figure traces the change in skill composition among migrants (panel a) and emigration probabilities to the United States by\neducational level (panel b) as low- and middle-income origin countries develop. The horizontal axis shows the increase, in percent, in per\ncapita GDP in constant 2017 US dollars in purchasing power parity terms (2017 PPP$) of origin countries relative to the origin country’s baseline\nyear. The vertical axis shows changes in educational composition (panel a) and emigration probabilities (panel b) relative to the origin country’s\nbaseline year. The baseline year is defined for each origin country as the year with the lowest per capita GDP (2017 PPP$). Low-and middleincome countries were among the bottom two-thirds circa 1960 in terms of per capita GDP (2017 PPP$). CI = confidence interval.\n\n#### 4. Conclusion\n\n\nThis paper shows that the relationship between emigration and development is complex. The emigration\nresponse to development varies along a few dimensions. First, the income level of origin countries matters.\nLow-income countries have a more complex response than middle-income countries, which exhibit an\n“inverted-U” pattern whereby migration first rises, then plateaus, and then falls with development. Second,\nmigration response differs by the type of destination. Migration from low-income origins to low-return\ndestinations—neighboring countries or other low-income destination countries—falls in the initial stages of\ndevelopment. Emigration to high-\n\nMention: \"IPUMS USA\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The one-year ACS samples for 2010 and 2019 are obtained through this harmonising project.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001036:21:0", "occurrence_id": "cca8a79d56a334ef1225a30e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Survey (ACS) one-year samples for 2010 and 2019 available from IPUMS USA (Ruggles et al. 2022). Data on educational outcomes of origin\ncountries are from the updated Barro-Lee Educational Attainment dataset (Barro and Lee, 2013), and GDP data are from the Penn World\nTables (PWT) 10.0 (Feenstra, Inklaar, and Timmer 2015).\n\n\n_Note_ : The figure traces the change in skill composition among migrants (panel a) and emigration probabilities to the United States by\neducational level (panel b) as low- and middle-income origin countries develop. The horizontal axis shows the increase, in percent, in per\ncapita GDP in constant 2017 US dollars in purchasing power parity terms (2017 PPP$) of origin countries relative to the origin country’s baseline\nyear. The vertical axis shows changes in educational composition (panel a) and emigration probabilities (panel b) relative to the origin country’s\nbaseline year. The baseline year is defined for each origin country as the year with the lowest per capita GDP (2017 PPP$). Low-and middleincome countries were among the bottom two-thirds circa 1960 in terms of per capita GDP (2017 PPP$). CI = confidence interval.\n\n#### 4. Conclusion\n\n\nThis paper shows that the relationship between emigration and development is complex. The emigration\nresponse to development varies along a few dimensions. First, the income level of origin countries matters.\nLow-income countries have a more complex response than middle-income countries, which exhibit an\n“inverted-U” pattern whereby migration first rises, then plateaus, and then falls with development. Second,\nmigration response differs by the type of destination. Migration from low-income origins to low-return\ndestinations—neighboring countries or other low-income destination countries—falls in the initial stages of\ndevelopment. Emigration to high-\n\nMention: \"Barro-Lee Educational Attainment dataset\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Educational outcomes in origin countries are taken from this updated attainment dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001036:21:0", "occurrence_id": "e734f579f240769e8ed8eec4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Survey (ACS) one-year samples for 2010 and 2019 available from IPUMS USA (Ruggles et al. 2022). Data on educational outcomes of origin\ncountries are from the updated Barro-Lee Educational Attainment dataset (Barro and Lee, 2013), and GDP data are from the Penn World\nTables (PWT) 10.0 (Feenstra, Inklaar, and Timmer 2015).\n\n\n_Note_ : The figure traces the change in skill composition among migrants (panel a) and emigration probabilities to the United States by\neducational level (panel b) as low- and middle-income origin countries develop. The horizontal axis shows the increase, in percent, in per\ncapita GDP in constant 2017 US dollars in purchasing power parity terms (2017 PPP$) of origin countries relative to the origin country’s baseline\nyear. The vertical axis shows changes in educational composition (panel a) and emigration probabilities (panel b) relative to the origin country’s\nbaseline year. The baseline year is defined for each origin country as the year with the lowest per capita GDP (2017 PPP$). Low-and middleincome countries were among the bottom two-thirds circa 1960 in terms of per capita GDP (2017 PPP$). CI = confidence interval.\n\n#### 4. Conclusion\n\n\nThis paper shows that the relationship between emigration and development is complex. The emigration\nresponse to development varies along a few dimensions. First, the income level of origin countries matters.\nLow-income countries have a more complex response than middle-income countries, which exhibit an\n“inverted-U” pattern whereby migration first rises, then plateaus, and then falls with development. Second,\nmigration response differs by the type of destination. Migration from low-income origins to low-return\ndestinations—neighboring countries or other low-income destination countries—falls in the initial stages of\ndevelopment. Emigration to high-\n\nMention: \"Penn World\nTables\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"GDP levels for origin countries are taken from this national accounting database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001037:5:1", "occurrence_id": "f1adf32765947da643c1b952", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |26718|0.09|0.29|0|1|\n|**Foreign bank**|26053|0.12|0.32|0|1|\n|**Foreign state bank**|26718|0|0.01|0|1|\n|**Bank health index**|7078|0.56|0.42|-2.33|1.11|\n|**Bank account**|28263|0.93|0.26|0|1|\n|**Ln(age)**|28113|2.68|0.76|0|5.32|\n|**20-99 employees**|28444|0.16|0.37|0|1|\n|**>99 employees**|28444|0.03|0.18|0|1|\n|**SOE**|28056|0|0.05|0|1|\n|**Foreign company?**|28037|0.02|0.15|0|1|\n|**Legal origin, France**|27426|0.41|0.49|0|1|\n|**Legal origin, Germany**|27426|0.59|0.49|0|1|\n|**Legal origin, Scandinavia**|27426|0|0|0|1|\n|**Legal origin, UK**|27426|0|0.03|0|1|\n\n\n\nSource: authors' computations based on the World Bank Enterprise Survey Data\n\n\nTo assess whether banking with a state-owned bank affects the probability of innovating, our baseline\nframework is:\n\n\n𝐼𝑛𝑛𝑜𝑣𝑎𝑡𝑒𝑖𝑗𝑡 = 𝛼𝐿𝑖𝑛𝑒 𝑜𝑓 𝑐𝑟𝑒𝑑𝑖𝑡𝑖𝑗𝑡 + 𝛽𝐿𝑖𝑛𝑒 𝑜𝑓 𝑐𝑟𝑒𝑑𝑖𝑡𝑖𝑗𝑡 × 𝑆𝑡𝑎𝑡𝑒𝑜𝑤𝑛𝑒𝑑 𝑏𝑎𝑛𝑘𝑘𝑗𝑡\n\nMention: \"World Bank Enterprise Survey Data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Firm characteristics and bank ownership categories are tabulated from this enterprise survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001041:12:1", "occurrence_id": "73482585071a79d1522bd414", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: it,_ (1)\n\n\n\n_Yit_ =\n\n\n\n\n\n\n_k̸_ = _−_ 2\n\n\n\n_βk ·_ I( _k_ = _t_ ) _·_ Takeover _i_ + _ψi_ + _νt_ +\n\n\n\nwhere _i_ indexes workers, _t_ indexes event time, and _Yit_ is an outcome at the level of an individual\n\n\nworker. Takeover _i_ is an indicator for being an incumbent worker employed at a firm involved in\n\n\na takeover four years prior to the event, the _βk_ are event-time indicators, the _ψi_ are worker fixed\n\n\neffects, the _νt_ are event-time fixed effects, the _γl_ are calendar-year fixed effects. We cluster standard\n\n\nerrors at the event level. Because of the staggered timing of these events, we use the weighting\n\n\napproach proposed by Sun and Abraham (2021). <sup>11</sup>\n\n\n10We use actual skill levels as reported in the Education Database as well as imputed skill values using an inverse\nMincer equation (see Appendix B). Low-skilled refers to workers who have up to and including a low (junior) secondary\neducation. Medium-skilled refers to workers with upper-secondary or post-secondary education excluding tertiary\neducation. Workers designated as high-skilled have tertiary education.\n\n11As demonstrated in Goodman-Bacon (2021) two-way fixed effects models (TWFE) can produce biased estimates in\n\n\n10\n\nMention: \"Education Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Actual skill levels of workers are read from this administrative education record.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001050:12:0", "occurrence_id": "812219f38b7d120025bd55c8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **4.** **CH4 Concentration Anomalies over Irrigated Rice Production Areas**\n\nThe current version of our database provides monthly mean CH4 concentration anomalies in\n266,884 terrestrial grid cells at 25 km resolution for 2020, 2021, and the first half of 2022. The\nmeans are calculated from daily observations that are pre-filtered to remove global background\nconcentrations, trends, and latitudinal clustering. We overlay the IFPRI shapefile on the global\ngrid to extract cells with irrigated rice production. For this exercise, CH4 observations in each cell\nare limited to the months in which fields are actually flooded. We identify those months using\nRiceAtlas, a spatial database of global rice calendars developed by Laborte et al. (2017). RiceAtlas\ndistinguishes up to three rice cultivation seasons and identifies the first flooding and harvest days\nfor each season by day-of-year.\n\nFor each pixel in RiceAtlas, we use this information to identify months that lie fully within periods\nlasting from first field flooding to ten days before harvest (to account for pre-harvest drainage\n(IRRI, 2022)). Then we intersect the RiceAtlas shapefile with our SP5/IFPRI spatial grid and assign\neach grid cell the cultivation calendar with the greatest representation in that cell. <sup>9</sup> [^9: Most grid cells have only one cultivation calendar; this assignment rule applies to grid cells that straddle two or more\nRiceAtlas calendars.] For grid cells\nwith IFPRI irrigated rice areas but no RiceAtlas information, we assign the cultivation calendar in\nthe nearest RiceAtlas cell. We compute mean grid cell CH4 anomalies for all within-calendar\nmonthly observations during the period 2020-2022.\n\n\n**5. Sources of Variation in Concentration Anomalies**\n\n\n**5.1 The Effect of Field Scale**\n\nWe posit a monotone increasing relationship between CH4 anomalies and the scale of IFPRI\nflooded fields in a grid cell and perform a linear regression using percentile values. <sup>10</sup> To test\n\nMention: \"IFPRI shapefile\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"It is overlaid on the global grid to select cells with irrigated rice production.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001050:12:0", "occurrence_id": "dd1feec213bcedc633b756b6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **4.** **CH4 Concentration Anomalies over Irrigated Rice Production Areas**\n\nThe current version of our database provides monthly mean CH4 concentration anomalies in\n266,884 terrestrial grid cells at 25 km resolution for 2020, 2021, and the first half of 2022. The\nmeans are calculated from daily observations that are pre-filtered to remove global background\nconcentrations, trends, and latitudinal clustering. We overlay the IFPRI shapefile on the global\ngrid to extract cells with irrigated rice production. For this exercise, CH4 observations in each cell\nare limited to the months in which fields are actually flooded. We identify those months using\nRiceAtlas, a spatial database of global rice calendars developed by Laborte et al. (2017). RiceAtlas\ndistinguishes up to three rice cultivation seasons and identifies the first flooding and harvest days\nfor each season by day-of-year.\n\nFor each pixel in RiceAtlas, we use this information to identify months that lie fully within periods\nlasting from first field flooding to ten days before harvest (to account for pre-harvest drainage\n(IRRI, 2022)). Then we intersect the RiceAtlas shapefile with our SP5/IFPRI spatial grid and assign\neach grid cell the cultivation calendar with the greatest representation in that cell. <sup>9</sup> [^9: Most grid cells have only one cultivation calendar; this assignment rule applies to grid cells that straddle two or more\nRiceAtlas calendars.] For grid cells\nwith IFPRI irrigated rice areas but no RiceAtlas information, we assign the cultivation calendar in\nthe nearest RiceAtlas cell. We compute mean grid cell CH4 anomalies for all within-calendar\nmonthly observations during the period 2020-2022.\n\n\n**5. Sources of Variation in Concentration Anomalies**\n\n\n**5.1 The Effect of Field Scale**\n\nWe posit a monotone increasing relationship between CH4 anomalies and the scale of IFPRI\nflooded fields in a grid cell and perform a linear regression using percentile values. <sup>10</sup> To test\n\nMention: \"RiceAtlas\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Its cultivation calendars identify the months fields are flooded.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001050:12:0", "occurrence_id": "ffd2d6ade436d55369169221", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **4.** **CH4 Concentration Anomalies over Irrigated Rice Production Areas**\n\nThe current version of our database provides monthly mean CH4 concentration anomalies in\n266,884 terrestrial grid cells at 25 km resolution for 2020, 2021, and the first half of 2022. The\nmeans are calculated from daily observations that are pre-filtered to remove global background\nconcentrations, trends, and latitudinal clustering. We overlay the IFPRI shapefile on the global\ngrid to extract cells with irrigated rice production. For this exercise, CH4 observations in each cell\nare limited to the months in which fields are actually flooded. We identify those months using\nRiceAtlas, a spatial database of global rice calendars developed by Laborte et al. (2017). RiceAtlas\ndistinguishes up to three rice cultivation seasons and identifies the first flooding and harvest days\nfor each season by day-of-year.\n\nFor each pixel in RiceAtlas, we use this information to identify months that lie fully within periods\nlasting from first field flooding to ten days before harvest (to account for pre-harvest drainage\n(IRRI, 2022)). Then we intersect the RiceAtlas shapefile with our SP5/IFPRI spatial grid and assign\neach grid cell the cultivation calendar with the greatest representation in that cell. <sup>9</sup> [^9: Most grid cells have only one cultivation calendar; this assignment rule applies to grid cells that straddle two or more\nRiceAtlas calendars.] For grid cells\nwith IFPRI irrigated rice areas but no RiceAtlas information, we assign the cultivation calendar in\nthe nearest RiceAtlas cell. We compute mean grid cell CH4 anomalies for all within-calendar\nmonthly observations during the period 2020-2022.\n\n\n**5. Sources of Variation in Concentration Anomalies**\n\n\n**5.1 The Effect of Field Scale**\n\nWe posit a monotone increasing relationship between CH4 anomalies and the scale of IFPRI\nflooded fields in a grid cell and perform a linear regression using percentile values. <sup>10</sup> To test\n\nMention: \"spatial database of global rice calendars\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"The description states it distinguishes up to three seasons and gives flooding and harvest days.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001050:12:0", "occurrence_id": "cc3d7e8abd4b083d6b9ecbb3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **4.** **CH4 Concentration Anomalies over Irrigated Rice Production Areas**\n\nThe current version of our database provides monthly mean CH4 concentration anomalies in\n266,884 terrestrial grid cells at 25 km resolution for 2020, 2021, and the first half of 2022. The\nmeans are calculated from daily observations that are pre-filtered to remove global background\nconcentrations, trends, and latitudinal clustering. We overlay the IFPRI shapefile on the global\ngrid to extract cells with irrigated rice production. For this exercise, CH4 observations in each cell\nare limited to the months in which fields are actually flooded. We identify those months using\nRiceAtlas, a spatial database of global rice calendars developed by Laborte et al. (2017). RiceAtlas\ndistinguishes up to three rice cultivation seasons and identifies the first flooding and harvest days\nfor each season by day-of-year.\n\nFor each pixel in RiceAtlas, we use this information to identify months that lie fully within periods\nlasting from first field flooding to ten days before harvest (to account for pre-harvest drainage\n(IRRI, 2022)). Then we intersect the RiceAtlas shapefile with our SP5/IFPRI spatial grid and assign\neach grid cell the cultivation calendar with the greatest representation in that cell. <sup>9</sup> [^9: Most grid cells have only one cultivation calendar; this assignment rule applies to grid cells that straddle two or more\nRiceAtlas calendars.] For grid cells\nwith IFPRI irrigated rice areas but no RiceAtlas information, we assign the cultivation calendar in\nthe nearest RiceAtlas cell. We compute mean grid cell CH4 anomalies for all within-calendar\nmonthly observations during the period 2020-2022.\n\n\n**5. Sources of Variation in Concentration Anomalies**\n\n\n**5.1 The Effect of Field Scale**\n\nWe posit a monotone increasing relationship between CH4 anomalies and the scale of IFPRI\nflooded fields in a grid cell and perform a linear regression using percentile values. <sup>10</sup> To test\n\nMention: \"SP5/IFPRI spatial grid\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The rice calendar shapefile is intersected with this grid to assign a calendar per cell.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001057:9:1", "occurrence_id": "6fa66418dca51da0f123c679", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  from Europe and Central Asia region, one country\n\n\nfrom East Asia Pacific region, one country from Latin America and the Caribbean region, and\n\n\n1 country from the Middle East and North Africa region. In terms of income groups: 10 coun\n\ntries are in the low-income group, 6 countries are in the lower middle-income group, and two\n\n\ncountries are in the upper middle-income group.\n\n\nThe typical LSMS household survey contains questions on expenditures on private transport\n\n\nand transport services during recall periods. The questions on expenditures ask whether the\n\n\nhouseholds purchased any vehicles or paid for transport-related goods or services during the\n\n\nfollowing periods of time: i) 7 days, ii) 1 month, iii) 3 months, iv) 6 months, and v) 1 year, and\n\n\nhow much money was spent. Usually, short recall periods are used for transport expenditure on\n\n\nnondurables, such as public transport fares, fuels and lubricants, and maintenance and repairs,\n\n\nwhile the longer periods are used for transport expenditure on durables, such as purchases of\n\n\nvehicles, spare parts, and accessories.\n\n\nOther household characteristics contained in LSMS surveys are used in the reduced-form\n\n\nregression part. Information on the total amount of expenditure, the size of the household and\n\n\n3These surveys mainly fall into two categories; 10 of the 18 country surveys are from the LSMS collection\nincluded in the World Bank Microdata Library. The other 8 country surveys are from other multi-topic household\nsurveys labelled as “LSMS-tagged” surveys.\n\n\n8\n\nMention: \"LSMS household survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its questions on transport expenditure and recall periods supply the transport cost measures.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001057:9:1", "occurrence_id": "21d99ae94f2615d26e9da455", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  from Europe and Central Asia region, one country\n\n\nfrom East Asia Pacific region, one country from Latin America and the Caribbean region, and\n\n\n1 country from the Middle East and North Africa region. In terms of income groups: 10 coun\n\ntries are in the low-income group, 6 countries are in the lower middle-income group, and two\n\n\ncountries are in the upper middle-income group.\n\n\nThe typical LSMS household survey contains questions on expenditures on private transport\n\n\nand transport services during recall periods. The questions on expenditures ask whether the\n\n\nhouseholds purchased any vehicles or paid for transport-related goods or services during the\n\n\nfollowing periods of time: i) 7 days, ii) 1 month, iii) 3 months, iv) 6 months, and v) 1 year, and\n\n\nhow much money was spent. Usually, short recall periods are used for transport expenditure on\n\n\nnondurables, such as public transport fares, fuels and lubricants, and maintenance and repairs,\n\n\nwhile the longer periods are used for transport expenditure on durables, such as purchases of\n\n\nvehicles, spare parts, and accessories.\n\n\nOther household characteristics contained in LSMS surveys are used in the reduced-form\n\n\nregression part. Information on the total amount of expenditure, the size of the household and\n\n\n3These surveys mainly fall into two categories; 10 of the 18 country surveys are from the LSMS collection\nincluded in the World Bank Microdata Library. The other 8 country surveys are from other multi-topic household\nsurveys labelled as “LSMS-tagged” surveys.\n\n\n8\n\nMention: \"LSMS surveys\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Household total expenditure and size used in the reduced-form regressions are taken from these surveys.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001053:1:0", "occurrence_id": "cde921076b3744af4644a518", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Policy Research Working Paper 10083\n\n### **Abstract**\n\nHardly any evidence exists on the effects of mental illness\non refugee labor outcomes. This paper offers the first study\non this topic in the context of Australia, one of the host\ncountries with the largest number of refugees per capita in\nthe world. Analyzing the _Building a New Life in Australia_\nlongitudinal survey, the paper exploits the variations in\ntraumatic experiences of refugees interacted with post-resettlement time periods to causally identify the impacts of\nrefugee mental health. The findings show that worse mental\n\n\n\nhealth, as measured by a one-standard-deviation increase\nin the Kessler mental health score, reduces the probability of employment by 14.1 percent and labor income by\n26.8 percent. There is also evidence of adverse impacts of\nrefugees’ mental illness on their children’s mental health\nand educational performance. These effects appear to be\nmore pronounced for newly arriving refugees and those\nwithout social networks, but they may be ameliorated with\ngovernment support.\n\n\n\nThis paper is a product of the Development Data Group, Development Economics and the Fragility, Conflict and Violence\n\nGlobal Theme. It is part of a larger effort by the World Bank to provide open access to its research and make a contribution\nto development policy discussions around the world. Policy Research Working Papers are also posted on the Web at http://\nwww.worldbank.org/prwp. The authors may be contacted at hdang@worldbank.org.\n\n\n_The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development_\n_issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the_\n_names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those_\n_of the authors. They do not necessarily represent the views of\n\nMention: \"_Building a New Life in Australia_\nlongitudinal survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Refugees' traumatic experiences interacted with time since resettlement identify mental health effects on labour outcomes.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001060:20:0", "occurrence_id": "2079383a9126b0c85480144d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Notes: Adjusted for inflation using the World Bank World Development Indicators Consumer Price Index with 2010\nas the base year. Data on tax revenue is from the Lagos State Government Digest of Statistics 2006, 2010, 2013. The\nperiod between the red vertical lines indicates when the federal government withheld transfers from Lagos local\ngovernments.\n\n\nIn Lagos, the specific actions taken by the tax administration focused on deploying technology by\nintroducing electronic payments and tax withholding, by creating an autonomous tax agency\nseparate from the civil service, and by hiring and incentivizing skilled professionals. There was\npolitical support for enforcing tax laws, like imposing fines and sealing delinquent businesses. In\naddition, there was significant taxpayer education and outreach, and publicity campaigns to foster\ntax morale by linking the visible public infrastructure investments to the expansion in tax\npayments. Overall, this case suggests that success in increasing tax collection requires a\ncombination of factors, including a well-motivated government and an ability to deploy diverse\ntools, particularly those relating to technology and personnel.\n\nAnother key factor that may influence a government’s decision of how much to enforce taxes is\nthe availability of technology tools that reduce the political and economic costs of taxation by\nmaking tax collection more automated and less salient. For example, using employers to withhold\npersonal income taxes removes the need to require citizens to remit payments, and consumption\ntaxes like the value-added tax may be less salient to citizens. Such tools or instruments may lead\nto less political resistance—but they only work well if the government has access to appropriate\ntechnology tools for implementing them. As an example from the United States context, Cabral\nand Hoxby (2012) show that property taxes are less salient to property owners who use the\ntechnology-aided practice of bundling their taxes with their monthly mortgage payment through a\ntax escrow account, compared to other homeowners with mortgages who make one or two large\ndirect tax payments a year to the government. Areas with higher rates of escrow use also have\nhigher property tax rates, suggesting that the\n\nMention: \"World Bank World Development Indicators Consumer Price Index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Series are deflated to constant 2010 values using this price index.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001060:20:0", "occurrence_id": "54801eeded8e70df68d6a3b8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Notes: Adjusted for inflation using the World Bank World Development Indicators Consumer Price Index with 2010\nas the base year. Data on tax revenue is from the Lagos State Government Digest of Statistics 2006, 2010, 2013. The\nperiod between the red vertical lines indicates when the federal government withheld transfers from Lagos local\ngovernments.\n\n\nIn Lagos, the specific actions taken by the tax administration focused on deploying technology by\nintroducing electronic payments and tax withholding, by creating an autonomous tax agency\nseparate from the civil service, and by hiring and incentivizing skilled professionals. There was\npolitical support for enforcing tax laws, like imposing fines and sealing delinquent businesses. In\naddition, there was significant taxpayer education and outreach, and publicity campaigns to foster\ntax morale by linking the visible public infrastructure investments to the expansion in tax\npayments. Overall, this case suggests that success in increasing tax collection requires a\ncombination of factors, including a well-motivated government and an ability to deploy diverse\ntools, particularly those relating to technology and personnel.\n\nAnother key factor that may influence a government’s decision of how much to enforce taxes is\nthe availability of technology tools that reduce the political and economic costs of taxation by\nmaking tax collection more automated and less salient. For example, using employers to withhold\npersonal income taxes removes the need to require citizens to remit payments, and consumption\ntaxes like the value-added tax may be less salient to citizens. Such tools or instruments may lead\nto less political resistance—but they only work well if the government has access to appropriate\ntechnology tools for implementing them. As an example from the United States context, Cabral\nand Hoxby (2012) show that property taxes are less salient to property owners who use the\ntechnology-aided practice of bundling their taxes with their monthly mortgage payment through a\ntax escrow account, compared to other homeowners with mortgages who make one or two large\ndirect tax payments a year to the government. Areas with higher rates of escrow use also have\nhigher property tax rates, suggesting that the\n\nMention: \"Lagos State Government Digest of Statistics\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Tax revenue figures for these years are compiled in this official digest.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001024:8:0", "occurrence_id": "7cc15f62b9147161424d6297", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: sources. Section 3 presents the empirical approach used to estimate the effects of childcare\n\n\nlaws on FLFP. Section 4 presents the main results and is followed by robustness checks in\n\n\nSection 5. Section 6 concludes the paper.\n\n##### **2 Data**\n\n\nOur analysis is based on data from 80 countries spanning a 20-year period (2000-2020).\n\n\nWe combine information on childcare regulations, labor statistics, and socio-demographic\n\n\nindicators from the following sources to construct a country-year panel for our analysis.\n\n##### **2.1 Data sources**\n\n\n_Childcare regulations._ The World Bank’s WBL database collects information on the legal en\n\nvironment pertaining to women’s economic participation and opportunity in 95 countries\n\n\n(Hyland et al., 2020). In particular, the dataset covers laws and regulations governing the\n\n\nsupply and parental demand for childcare services. The dataset includes information on\n\n\nthe dates of enactment and commencement of a law and other indicators related to ac\n\ncessibility, affordability, and quality of childcare services applicable to children below the\n\n\nformal pre-primary school starting age, as this is where the largest legal policy gap exists.\n\n\nThe WBL database maps current legal and regulatory measures adopted by each country\n\n\nto ensure or strengthen the following three pillars of childcare (see list of indicators within\n\n\neach pillar in Table A1):\n\n\n - Availability: The childcare law expands access to childcare by supporting different\n\n\ntypes of childcare provision and its convenience.\n\n\n6\n\nMention: \"WBL database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Laws and regulations on childcare supply and demand across 95 countries form the legal variables of the country-year panel.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001061:14:1", "occurrence_id": "8ced3b8ae0e7d749fbd69ea8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  7% 4%\nPuglia 4% 5% 6% 7% 9% 6% 6% 4%\nBasilicata 1% 1% 1% 1% 1% 1% 1% 2%\nCalabria 2% 3% 3% 2% 4% 3% 3% 3%\nSicilia 5% 6% 7% 9% 10% 7% 7% 2%\n<u>Sardegna</u> <u>2%</u> <u>2%</u> <u>2%</u> <u>3%</u> <u>3%</u> <u>3%</u> <u>2%</u> <u>1%</u>\n\n\nTable 1: The Table shows for each Italian region (rows) data of Households’ income shares\n(DF estimation). These shares represent the income source for households across the regions.\nThe Italian regions are listed from North regions to South and islands (Sicilia and Sardegna)\n\n\n  - SSC employer from ISTAT data 2016. The total social security contributions by\nemployers are equal to €176.5 billion;\n\n\n  - Taxes on commodities, composed of other net taxes on products from national SAM,\nare redistributed across regions using the demand for the respective commodity as\na share. We took data from the national SAM 2016, which report a total national\namount of taxes on commodities equal to €9 billion;\n\n\n  - VAT is regionalized using data from the DF microsimulation model on VAT (Cirillo\net al., 2021). Only the VAT relating to households is regionalized. The total amount\n(€102 billion) was re-proportioned using the totals from the ISTAT national accounts\n(i.e., those reported in the 2016 national SAM);\n\n\n  - Excises from national SAM (€59,2 billion) reproportioned on the\n\nMention: \"data of Households’ income shares\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Regional income shares estimated from these household data are used to distribute aggregates across regions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001061:14:1", "occurrence_id": "ed80581bf8e20dae0bb6db91", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  7% 4%\nPuglia 4% 5% 6% 7% 9% 6% 6% 4%\nBasilicata 1% 1% 1% 1% 1% 1% 1% 2%\nCalabria 2% 3% 3% 2% 4% 3% 3% 3%\nSicilia 5% 6% 7% 9% 10% 7% 7% 2%\n<u>Sardegna</u> <u>2%</u> <u>2%</u> <u>2%</u> <u>3%</u> <u>3%</u> <u>3%</u> <u>2%</u> <u>1%</u>\n\n\nTable 1: The Table shows for each Italian region (rows) data of Households’ income shares\n(DF estimation). These shares represent the income source for households across the regions.\nThe Italian regions are listed from North regions to South and islands (Sicilia and Sardegna)\n\n\n  - SSC employer from ISTAT data 2016. The total social security contributions by\nemployers are equal to €176.5 billion;\n\n\n  - Taxes on commodities, composed of other net taxes on products from national SAM,\nare redistributed across regions using the demand for the respective commodity as\na share. We took data from the national SAM 2016, which report a total national\namount of taxes on commodities equal to €9 billion;\n\n\n  - VAT is regionalized using data from the DF microsimulation model on VAT (Cirillo\net al., 2021). Only the VAT relating to households is regionalized. The total amount\n(€102 billion) was re-proportioned using the totals from the ISTAT national accounts\n(i.e., those reported in the 2016 national SAM);\n\n\n  - Excises from national SAM (€59,2 billion) reproportioned on the\n\nMention: \"ISTAT data 2016\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Employer social security contributions of €176.5 billion are taken from this national statistics source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001062:5:1", "occurrence_id": "2429ca12c5a28e69ced44410", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: , firms could receive up to 3 rounds of support in 8 possible combinations: from receiving\nsupport in each of the three rounds to none. Most firms which received any government support did so\nonly in Round 1: 20 percent of firms or about one-third of firms receiving support. This was followed by\nfirms receiving support in all three rounds (9 percent), and Rounds 1 and 2 and Rounds 1 and 3 (7\npercent each). The combinations of receiving support in Round 2 only, Round 3 only, or Rounds 2 and 3\nwas reported by fewer than 5 percent of firms each. About 45 percent of firms received no support at\nall. The ordering of combinations is proportional when considering only Rounds 1 and 2 (Panel B).\n\n**2.2 What are the firm characteristics and changes in performance during COVID-19?**\n\nTable 3 Panel A presents the summary statistics of firm background characteristics from the pre-COVID19 ES. To measure productivity, we use labor productivity, defined as (log of) sales over total number of\npermanent full-time employees. <sup>7</sup> [^7: In calculating labor productivity, the outliers are eliminated by first log-transforming total annual sales and the\nnumber of permanent full-time employees, and then trimming at plus and minus three standard deviations from\nthe mean, as described in World Bank (2021a).] Labor productivity is preferred to total factor productivity due to\nlimitations in measuring non-labor inputs in the ES (World Bank 2021a). The average firm in our sample\nhad a log of labor productivity of 10.59. At the same time, the average firm had 28 employees (2.62 in\nlog units) and was 20 years old (2.88 in log units). Only 20 percent of the firms had a female top\nmanager. The sample includes few firms with at least 10 percent state or foreign ownership (less than 1\nand 7 percent, respectively). Less than half the firms reported that they had a\n\nMention: \"pre-COVID19 ES\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Firm background characteristics and labour productivity are summarised from this pre-pandemic round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001063:4:0", "occurrence_id": "dbb43893ab52f01c90b03579", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The Colombian government created in 2018 the _Permiso_ _Especial_ _de_ _Permanencia_ (PEP) to\nfacilitate their integration. Prior to the PEP, the government rolled out a voluntary registration process, the _Registro Administrativo de Migrantes Venezolanos_ (RAMV), with the sole\npurpose of identifying the location and needs of irregular migrants. In July 2018, the government unexpectedly announced that all migrants registered in RAMV, a little more than\n440,000, were eligible for applying to the PEP program through a free and online tool. Of\nthose, 64% registered into PEP. PEP granted legal migratory status for two years, access\nto education and health, eligibility to social programs, job permits, and an identification\ncard to access private services. For a detailed description of the program see Ib´a˜nez et al.\n(2022).\n\n\nThe design and rollout of PEP lend themselves to estimating the causal effects on the\nresilience of forced migrants to negative shocks. The unexpected and broad nature of\nthe program (available to all RAMV migrants and with no eligibility criteria), allows for\nruling out anticipatory effects and strategic behaviors that have affected the evaluations\nof other similar programs.\n\n\n**III** **Data**\n\n\nWe use two waves of the Venezuelan Refugees Panel Survey (VenRePS), administered in\nthe second semesters of 2020 and 2021. Figure A.1 in the appendix depicts the timeline\nfor the PEP roll-out, the VenRePS waves, and the lockdowns put in place by the Colombian government. The baseline sample covered 2,232 forced migrant households who\narrived in Colombia between January 1, 2017 and December 2018, including 1,135 who\nregistered in RAMV and were thus eligible for PEP (from here on RAMV migrants), and\n1,097 irregular forced migrants (non-RAMV migrants from here on).\n\n\nThe sample of RAM\n\nMention: \"Venezuelan Refugees Panel Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Two waves covering 2,232 baseline households provide the outcomes of the permit programme.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001063:4:0", "occurrence_id": "ccb8e502ea55a54c6149c868", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The Colombian government created in 2018 the _Permiso_ _Especial_ _de_ _Permanencia_ (PEP) to\nfacilitate their integration. Prior to the PEP, the government rolled out a voluntary registration process, the _Registro Administrativo de Migrantes Venezolanos_ (RAMV), with the sole\npurpose of identifying the location and needs of irregular migrants. In July 2018, the government unexpectedly announced that all migrants registered in RAMV, a little more than\n440,000, were eligible for applying to the PEP program through a free and online tool. Of\nthose, 64% registered into PEP. PEP granted legal migratory status for two years, access\nto education and health, eligibility to social programs, job permits, and an identification\ncard to access private services. For a detailed description of the program see Ib´a˜nez et al.\n(2022).\n\n\nThe design and rollout of PEP lend themselves to estimating the causal effects on the\nresilience of forced migrants to negative shocks. The unexpected and broad nature of\nthe program (available to all RAMV migrants and with no eligibility criteria), allows for\nruling out anticipatory effects and strategic behaviors that have affected the evaluations\nof other similar programs.\n\n\n**III** **Data**\n\n\nWe use two waves of the Venezuelan Refugees Panel Survey (VenRePS), administered in\nthe second semesters of 2020 and 2021. Figure A.1 in the appendix depicts the timeline\nfor the PEP roll-out, the VenRePS waves, and the lockdowns put in place by the Colombian government. The baseline sample covered 2,232 forced migrant households who\narrived in Colombia between January 1, 2017 and December 2018, including 1,135 who\nregistered in RAMV and were thus eligible for PEP (from here on RAMV migrants), and\n1,097 irregular forced migrants (non-RAMV migrants from here on).\n\n\nThe sample of RAM\n\nMention: \"VenRePS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Its wave timing is aligned with the permit roll-out and lockdowns in the study timeline.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001064:35:0", "occurrence_id": "19cc984a27eda8ef1d99a713", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: detailed analysis herein dispels the worry that this is a high-risk business from inception. The loan\n\namounts are manageable, and it seems AHFC know their customers well given the high-touch\n\nmicrofinance strategies that they appear to be using. However, the exercise to tag lenders and loans\n\nin the available data is not perfect, and the analysis here contains smaller HFC-N loans as well that\n\nmay not be underwritten by AHFC lenders as the detailed credit bureau data do not allow a\n\nseparation of lender types.\n\n\nThe policy implications are to allow these AHFC to continue to lend to those neglected by\n\nlegacy lenders provided that these new lenders have good prudential and corporate governance\n\nstandards. Their novel approach to housing finance should not attract excessive regulation,\n\nespecially within the arc of ‘government as a facilitator’ of housing. The detailed annex shows the\n\nspecial institutional context in which AHFC emerged in India. This context is central to the\n\nnarrative. The unique dual housing finance setup in India has allowed for local currency capital\n\nmarkets to grow and these AHFC have likely found a niche between borrowing long term in local\n\ncurrency and lending—often with the support of government programs—to those deemed\n\nunworthy borrowers by other institutions. Their growth and low delinquency rates are impressive.\n\nIn terms of the size of the housing finance market in India, the injection of capital in public sector\n\nbanks in 2017 increased the number of loans from these lenders, especially in non-Tier 1 markets.\n\nAlthough AHFC have grown rapidly from a low base—especially in the number of loans—they\n\nonly represent between 1.4 to 2.4 percent of the mortgage market.\n\n\nThe implications for other countries trying to replicate this model would be to assess a\n\ncandidate market’s local currency capital market and whether alternative credit screening\n\ntechniques used by AHFC, such as the diary cash log method, can be borrowed from this successful\n\nIndia example. The annex allows for a more detailed peer country institutional comparison. The\n\nexperience\n\nMention: \"credit bureau data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"It cannot separate lender types, so tagging lenders and loans is imperfect.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001066:6:0", "occurrence_id": "34da31540e93b7d7c1157d26", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: descriptive statistics and trends of the data, using some country specific examples. Section 4 outlines the\nmethodology and framework we use for our empirical specifications. Section 5 presents the results of our analysis,\nand section 6 concludes.\n\n\n**2. Data**\n\n\nWe use data from a global survey on _COVID-19 Beliefs, Behaviors and Norms_ jointly conducted by researchers at\nthe Massachusetts Institute of Technology, Facebook, Johns Hopkins University (JHU), the World Health\nOrganization (WHO), and the Global Outbreak Alert and Response Network (GOARN). The survey was\nadministered via Facebook to a sample of its users in 67 countries from August 2020 to March 2021. We restrict\nour main analysis to 18 countries for which data is available over multiple waves. <sup>5</sup> While survey respondents are\nmostly urban, aged between 20-50 years and with tertiary education, the data contain weights to represent each\ncountry's adult or Internet-using population, rather than just Facebook users, and to reduce bias due to\nnonresponse.\n\n\nThe survey contains information on willingness to be vaccinated, perception of community norms around\npreventive behaviors, access to information about the virus and trust in media and the government. It also\nincludes data on individual characteristics and personal experience of the pandemic, with regard to health and\nemployment. Some summary statistics of the data are presented in Table 1. <sup>6</sup> Globally, 69 percent of the sample\nreport that they will accept the vaccine with some regional variation ranging from slightly above 50 percent in\nNorth America to relatively higher acceptance in South Asia at around 75 percent (Figure 1). <sup>7</sup>\n\n\n**Figure 1** : Vaccine acceptance around the world\n\n\nSource: Authors calculations using the Facebook COVID-19 Beliefs, Behaviors & Norms Survey (2021). Data are weighted using sample\nsurvey weights.\n\n\n5\n\nMention: \"Facebook COVID-19 Beliefs, Behaviors & Norms Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Weighted responses give vaccine acceptance by region.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001067:31:1", "occurrence_id": "c11a1d8eb932abeac2808581", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: the more conservative scenario in our\n\n\nsetting). Figure 6 (top panel) shows the point estimates of the Gini index calculated on the\n\n\nRuLIS ‘raw data sets’, that is on data as available from the data provider. Countries are\n\n\nranked low (Kyrgyzstan) to high (Bolivia) accordingly. The bottom panel of Figure 6\n\n\nshows the estimated Gini indices, for the same countries, after excluding outliers (as\n\n\nidentified by `outdetect` ). Even a cursory inspection of the two graphs bring to light a\n\n\nnumber of rerankings: countries move up or down the ladder as a consequence of how\n\n\noutliers are treated.\n\n\n30\n\nMention: \"RuLIS ‘raw data sets’\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Gini indices are calculated on the microdata as delivered by the provider before any outlier treatment.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001068:10:0", "occurrence_id": "326a0972a2628ff8adbc447d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **<mark>3.2.</mark>** **<mark>Establishing Causality in Educational Research</mark>**\n\n\n<mark>Establishing causal relationships between education and later life outcomes such as incomes,</mark>\n<mark>employability, and even voting behavior has been of interest for many decades. While there is</mark>\n<mark>overwhelming evidence for the positive impact that schooling can have, researchers have been</mark>\n<mark>cautious in drawing strong inferences about the causal effect of schooling especially in the absence</mark>\nof experimental evidence.\n\n\n<mark>However, the emergence of large-scale microeconomic datasets such as OECD’s Programme for</mark>\n<mark>International Student Assessment (PISA), Trends in International Mathematics and Science Study</mark>\n<mark>(TIMSS), and Progress in International Reading Literacy Study (PIRLS) has provided researchers</mark>\n<mark>with more tools to study these relationships. It is now possible to deploy econometric methods</mark>\n<mark>such as instrumental variables, regression-discontinuity, propensity score matching, difference-in-</mark>\n<mark>difference, and different sorts of fixed-effects specifications to establish causality. Cordero and</mark>\n<mark>Cristóbal (2017) provide a comprehensive overview of literature that uses such quasi-experimental</mark>\n<mark>techniques. The authors also provide examples of studies that have used such techniques to</mark>\n<mark>establish the impact of various school education policies. Particularly, the difference-in-difference</mark>\n<mark>approach and instrument variables strategy have been most frequently used for comparison</mark>\n<mark>between public and private schools, or to study the effects of class size, tracking, instructional\n\nMention: \"International Student Assessment\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"Cited as one large-scale dataset that made credible causal designs feasible in education research.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001068:10:0", "occurrence_id": "20e24878737245d0b694f61a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **<mark>3.2.</mark>** **<mark>Establishing Causality in Educational Research</mark>**\n\n\n<mark>Establishing causal relationships between education and later life outcomes such as incomes,</mark>\n<mark>employability, and even voting behavior has been of interest for many decades. While there is</mark>\n<mark>overwhelming evidence for the positive impact that schooling can have, researchers have been</mark>\n<mark>cautious in drawing strong inferences about the causal effect of schooling especially in the absence</mark>\nof experimental evidence.\n\n\n<mark>However, the emergence of large-scale microeconomic datasets such as OECD’s Programme for</mark>\n<mark>International Student Assessment (PISA), Trends in International Mathematics and Science Study</mark>\n<mark>(TIMSS), and Progress in International Reading Literacy Study (PIRLS) has provided researchers</mark>\n<mark>with more tools to study these relationships. It is now possible to deploy econometric methods</mark>\n<mark>such as instrumental variables, regression-discontinuity, propensity score matching, difference-in-</mark>\n<mark>difference, and different sorts of fixed-effects specifications to establish causality. Cordero and</mark>\n<mark>Cristóbal (2017) provide a comprehensive overview of literature that uses such quasi-experimental</mark>\n<mark>techniques. The authors also provide examples of studies that have used such techniques to</mark>\n<mark>establish the impact of various school education policies. Particularly, the difference-in-difference</mark>\n<mark>approach and instrument variables strategy have been most frequently used for comparison</mark>\n<mark>between public and private schools, or to study the effects of class size, tracking, instructional\n\nMention: \"Trends in International Mathematics and Science Study\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"Listed among the comparative assessments that expanded methodological options.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001068:10:0", "occurrence_id": "7379d2ece615865d4861febd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **<mark>3.2.</mark>** **<mark>Establishing Causality in Educational Research</mark>**\n\n\n<mark>Establishing causal relationships between education and later life outcomes such as incomes,</mark>\n<mark>employability, and even voting behavior has been of interest for many decades. While there is</mark>\n<mark>overwhelming evidence for the positive impact that schooling can have, researchers have been</mark>\n<mark>cautious in drawing strong inferences about the causal effect of schooling especially in the absence</mark>\nof experimental evidence.\n\n\n<mark>However, the emergence of large-scale microeconomic datasets such as OECD’s Programme for</mark>\n<mark>International Student Assessment (PISA), Trends in International Mathematics and Science Study</mark>\n<mark>(TIMSS), and Progress in International Reading Literacy Study (PIRLS) has provided researchers</mark>\n<mark>with more tools to study these relationships. It is now possible to deploy econometric methods</mark>\n<mark>such as instrumental variables, regression-discontinuity, propensity score matching, difference-in-</mark>\n<mark>difference, and different sorts of fixed-effects specifications to establish causality. Cordero and</mark>\n<mark>Cristóbal (2017) provide a comprehensive overview of literature that uses such quasi-experimental</mark>\n<mark>techniques. The authors also provide examples of studies that have used such techniques to</mark>\n<mark>establish the impact of various school education policies. Particularly, the difference-in-difference</mark>\n<mark>approach and instrument variables strategy have been most frequently used for comparison</mark>\n<mark>between public and private schools, or to study the effects of class size, tracking, instructional\n\nMention: \"Progress in International Reading Literacy Study\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"The third comparative study named as a source of tools for causal analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001076:4:0", "occurrence_id": "e7171cf6e8e60646a77b5554", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: size of the carbon benefit.\n\n\n**1.1** **Our** **approach**\n\n\nTo estimate the impact of adoption of no-till agriculture on local air pollution levels we leverage\n\n\ntwo recent advances in data collection and modeling. One obstacle to estimating the benefits of\n\n\nno-till agricultural has been poor data on the uptake of no-till over the entire U.S. Corn Belt\n\n\ngrowing region. Historically, analysis of the uptake of no-till practices has depended on five\n\n\nyear surveys conducted by the Conservation Technology information Center and the National\n\nResources Inventory database (e.g. ref. <sup>19,20</sup> ). These sources do not provide data that is spatially\n\n\nor temporally granular. Instead here we rely on advances in remote sensing and machine learning\n\nto measure uptake of no-till at approximately the field level and at an annual frequency. <sup>21</sup> This\n\n\nenables us to conduct an examination of how year-to-year changes in no-till usage in a given\n\n\ncounty impact the air quality of that county.\n\n\nA second obstacle has been the lack of data on PM2 _._ 5 levels with sufficient spatial and\n\n\ntemporal resolution. Data based on ground monitors generally offers daily, or even hourly,\n\n\ninformation on pollution levels but over a very limited geographic region. These monitors are\n\n\nrarely located in the rural regions that are most likely to be impacted by the changes in tillage\n\n\npractices that we examine here. The problem of limited geographic scope can be solved by\n\n\nusing remotely sensed data on pollution levels that takes advantage of satellite data to provide\n\n\nestimates of pollutant levels across a large geographic area. However, the increase in geographic\n\n\nscope comes at the cost of low temporal frequency. Because we expect changes to local air\n\n\npollutants due to the uptake of no-till to be concentrated at the times when no-till practices are\n\n\nmost likely to occur, the lack of temporal\n\nMention: \"National\n\nResources Inventory database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its five-year surveys are too coarse spatially and temporally for measuring practice uptake.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001077:6:0", "occurrence_id": "5237f302c40c9890c43b4a55", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: each country. The estimated coefficient indicates how student results depart from the time trend\ndepending on the school closure duration.\n\nTo investigate heterogeneity in the impact of school closures on achievement by gender and by\nachievement distribution, we introduced interaction with gender and ran the quantile\nregressions. To deal with many dummy variables and interactions, we use recent\nimplementations of fast quantile regression algorithms (Chernozhukov et al. 2022).\n\nPIRLS data are collected through a complex stratified survey with schools sampled as primary\nsampling units and classrooms with students sampled at the second stage. We used the\njackknife replicate weights provided by the PIRLS database to estimate the sampling error. We\nuse plausible values for reading achievement to adjust for the measurement error.\n\nThe paper also provides estimates for individual countries. We first compare achievement\nchange between 2016 and 2021. Then we use a similar model as for the populations of students\nin EU or ECA+, but with individual interactions between country dummies and indicator for\n2021 data. The general model is given by equation below:\n\n**𝑖** **𝑡** **𝑖** **𝑖**\n\n\n\n𝑛𝑛\n\n**𝑖**\n\n\n\n𝑛𝑛\n\n𝑌𝑌𝑖 **𝑖** 𝑖𝑖 = �𝛼𝛼𝑘𝑘\n\n𝑘𝑘=1\n\n\n\n𝑛𝑛\n\n+ �𝛽𝛽𝑘𝑘 ∗𝑡 **𝑡** 𝑡\n\n𝑘𝑘=1\n\n\n\n**𝑡**\n\n𝑘𝑘=1\n\n\n\n𝑛𝑛\n\n**𝑡**\n\n\n\n𝑛𝑛\n\n+ �𝜏𝜏𝑘𝑘 ∗𝐷𝐷2021\n\n𝑘𝑘=1\n\n\n\n𝑘𝑘=1\n\n\n\n𝑛𝑛\n\n\n\n+ 𝛾𝛾𝑋𝑋𝑖 **𝑖** 𝑖𝑖 + 𝜀𝜀𝑖 **𝑖** 𝑖𝑖\n\n\n\n**𝑖**\n\n𝑘𝑘=1\n\n\n\n**𝑖** **𝑡** **𝑖** **𝑖**\n\n𝑘𝑘=1 𝑘𝑘=1 𝑘𝑘=1\n\nfor each country. We estimate this model with and without controlling for age and grade. We\nWhich is similar to equation [\n\nMention: \"PIRLS data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Reading achievement across countries is analysed on this stratified two-stage survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001077:6:0", "occurrence_id": "3212892970c3e308e290c2e2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: each country. The estimated coefficient indicates how student results depart from the time trend\ndepending on the school closure duration.\n\nTo investigate heterogeneity in the impact of school closures on achievement by gender and by\nachievement distribution, we introduced interaction with gender and ran the quantile\nregressions. To deal with many dummy variables and interactions, we use recent\nimplementations of fast quantile regression algorithms (Chernozhukov et al. 2022).\n\nPIRLS data are collected through a complex stratified survey with schools sampled as primary\nsampling units and classrooms with students sampled at the second stage. We used the\njackknife replicate weights provided by the PIRLS database to estimate the sampling error. We\nuse plausible values for reading achievement to adjust for the measurement error.\n\nThe paper also provides estimates for individual countries. We first compare achievement\nchange between 2016 and 2021. Then we use a similar model as for the populations of students\nin EU or ECA+, but with individual interactions between country dummies and indicator for\n2021 data. The general model is given by equation below:\n\n**𝑖** **𝑡** **𝑖** **𝑖**\n\n\n\n𝑛𝑛\n\n**𝑖**\n\n\n\n𝑛𝑛\n\n𝑌𝑌𝑖 **𝑖** 𝑖𝑖 = �𝛼𝛼𝑘𝑘\n\n𝑘𝑘=1\n\n\n\n𝑛𝑛\n\n+ �𝛽𝛽𝑘𝑘 ∗𝑡 **𝑡** 𝑡\n\n𝑘𝑘=1\n\n\n\n**𝑡**\n\n𝑘𝑘=1\n\n\n\n𝑛𝑛\n\n**𝑡**\n\n\n\n𝑛𝑛\n\n+ �𝜏𝜏𝑘𝑘 ∗𝐷𝐷2021\n\n𝑘𝑘=1\n\n\n\n𝑘𝑘=1\n\n\n\n𝑛𝑛\n\n\n\n+ 𝛾𝛾𝑋𝑋𝑖 **𝑖** 𝑖𝑖 + 𝜀𝜀𝑖 **𝑖** 𝑖𝑖\n\n\n\n**𝑖**\n\n𝑘𝑘=1\n\n\n\n**𝑖** **𝑡** **𝑖** **𝑖**\n\n𝑘𝑘=1 𝑘𝑘=1 𝑘𝑘=1\n\nfor each country. We estimate this model with and without controlling for age and grade. We\nWhich is similar to equation [\n\nMention: \"PIRLS database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Jackknife replicate weights from this database are used to estimate sampling error.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001078:15:0", "occurrence_id": "ffa6f19d8219e789ecc8b4fd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: #### **5 Calibration**\n\nFirst, I calibrate the model in order to match the U.S. plant size distribution by assuming that\n\n\nthe U.S. is free of distortions and bribery. Then, I also calibrate this model to match Türkiye’s\n\n\nplant-size distribution as well as the bribery payments of different size groups as observed\n\n\nin the Enterprise Surveys.\n\n\nFor the U.S., I use a version of the model where population and technology parameter\n\n\ngrow at a constant rate. I borrow the following parameters from Guner _et al._ (2008): popu\n\nlation and technology parameter growth rates are set to 0.011 and 0.0255, _ν =_ 0.41, _β =_ 0.94,\n\n\n_δ =_ 0.04 and _γ =_ 0.80. Then I assume a composite lognormal-Pareto distribution for the dis\n\ntribution of managerial ability, _F_ ( _z_ ). The composite lognormal-Pareto distribution is char\n\nacterized by three parameters: a standard deviation of the lognormal distribution, _σ_, a shape\n\nparameter of the Pareto distribution, _s_, and a threshold value for distributions, _zT_ . <sup>10</sup> These\n\n\nparameter values are found to match the mean size for plants, the fraction of plants with 1–\n\n\n49 employees, and the proportion of workers in big plants in the US. Tables 4 and 5 display\n\n\nthe calibrated parameter values and the model performance.\n\n\nFor the calibration of the Turkish economy, I borrow some of the parameters from the\n\n\nliterature which are calculated using National Income and Product Accounts. I choose the\n\n\nshare of capital, _αγ =_ 0.34, depreciation, _δ =_ 0.055 and discount rate, _β =_ 0.93 consistent\n\n\nwith Torul & Öz\n\nMention: \"Enterprise Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Plant-size distribution and bribery payments by firm size are matched in the calibration.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001078:15:0", "occurrence_id": "f6ce9d67b1a96bca1769896b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: #### **5 Calibration**\n\nFirst, I calibrate the model in order to match the U.S. plant size distribution by assuming that\n\n\nthe U.S. is free of distortions and bribery. Then, I also calibrate this model to match Türkiye’s\n\n\nplant-size distribution as well as the bribery payments of different size groups as observed\n\n\nin the Enterprise Surveys.\n\n\nFor the U.S., I use a version of the model where population and technology parameter\n\n\ngrow at a constant rate. I borrow the following parameters from Guner _et al._ (2008): popu\n\nlation and technology parameter growth rates are set to 0.011 and 0.0255, _ν =_ 0.41, _β =_ 0.94,\n\n\n_δ =_ 0.04 and _γ =_ 0.80. Then I assume a composite lognormal-Pareto distribution for the dis\n\ntribution of managerial ability, _F_ ( _z_ ). The composite lognormal-Pareto distribution is char\n\nacterized by three parameters: a standard deviation of the lognormal distribution, _σ_, a shape\n\nparameter of the Pareto distribution, _s_, and a threshold value for distributions, _zT_ . <sup>10</sup> These\n\n\nparameter values are found to match the mean size for plants, the fraction of plants with 1–\n\n\n49 employees, and the proportion of workers in big plants in the US. Tables 4 and 5 display\n\n\nthe calibrated parameter values and the model performance.\n\n\nFor the calibration of the Turkish economy, I borrow some of the parameters from the\n\n\nliterature which are calculated using National Income and Product Accounts. I choose the\n\n\nshare of capital, _αγ =_ 0.34, depreciation, _δ =_ 0.055 and discount rate, _β =_ 0.93 consistent\n\n\nwith Torul & Öz\n\nMention: \"National Income and Product Accounts\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Parameters such as the capital share and depreciation are taken from literature estimates based on these accounts.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001081:11:0", "occurrence_id": "97c1e1770330d5de63877702", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: indicators affect changes in winter crop area; in fact, reductions in winter crop area in VCs unaffected by\n\nconflict are higher (though not significantly so) than in those affected by conflict. Determining whether the\n\ndecline in winter crop area is due to conflict-related shortages of labor, fuel, or other inputs needed to carry\n\nout the necessary crop management activities or climatic factors will require econometric analysis.\n\n\nAs the literature suggests that for winter wheat, satellite-based estimates of NDVI can reliably approximate\n\nyield, <sup>16</sup> [^16: Dhillon _et al._ (2020) report that the best fit is obtained most to the Feekes scale (Large 1954), i.e., between 97 and 112 GDD>0] we use the crop masks obtained from the above land cover analysis for 2019-22, respectively, to\n\ncompute NDVI on all fields planted to winter wheat for each of the 10-day periods ending on May11, 21,\n\n31, and June 11 and aggregate these values at village level. Calculations were performed on the Google\n\nEarth Engine cloud computing platform. <sup>17</sup> [^17: Due to a change in Sentinel 2 data processing that was introduced with version 4.0.0. end of January 2022, GEE’s S2_SR_HARMONIZED\ncollection was used for 2022 NDVI values.] Figure 3 illustrates NDVI densities for different periods in each\n\nyear. For all years except 2021, peak NDVI values are attained in the first 10 days of June. Tabulating VC\nlevel figures in panel C of table 2 confirms that NDVI in 2022 was uniformly lower than in earlier years.\n\n\n**3.3 Weather conditions**\n\n\n<mark>To control for time-varying effects that are unrelated to conflict, we use information on precipitation as</mark>\n\n<mark>well as minimum and maximum temperature derived from the United States National Centers for</mark>\n\n<mark>Environmental Prediction Global Forecast System (GFS) that produces weather forecasts every 6 hours.<\n\nMention: \"Sentinel 2 data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Vegetation indices are computed on the cloud platform to track winter crop area.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001082:20:0", "occurrence_id": "c58859e2633e645a798dc0de", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Nonetheless, despite all this good recent work on inequality at the Bank and the Fund, the growth in\nresearch and public debate on inequality elsewhere has been so large that the Bretton Woods institutions\nhave become relatively smaller intellectual players in the field. In and of itself, this is clearly no bad thing.\nInequality truly is one of the fundamental challenges facing today’s human societies, and the greater\nvisibility and popularity of the topic is hugely welcome. Research institutes such as the World Inequality\nLab at the Paris School of Economics, which hosts the World Inequality Database and publishes an annual\n_World Inequality Report_, are producing important, pathbreaking work extending our knowledge of\ninequality around the world. Other examples include the team at UNU-WIDER, who host the World\nIncome Inequality Database (WIID) and have a had a long tradition of work on both income and wealth\ninequality, as well as the Stone Center on Socio-Economic Inequality at the City University of New York.\nOf course, most academic research on inequality is produced at universities around the world, whether\nthey host one of these institutes or not.\n\n\nFrom the perspective of the two institutions, however, two features of this particular surge in external\nresearch pose a challenge. First, the bulk of the inequality research currently conducted at the world’s\nleading research universities and institutes focuses on rich countries. This is not exceptional; the same is\ntrue of most other topics. Authors, journal editors and referees residing in the United States and Europe\ntend to be very interested in what is happening in their own countries. Second, a considerable share of\nthe new work on inequality appears to focus on rich people. This too is perhaps unsurprising, given the\nrapid rise in top incomes – and top income shares – in most countries for which reliable data is available.\nIt is also part of the key innovation that spurred much of this research which, as noted earlier, was the\nuse of alternative data sources to supplement the inadequate\n\nMention: \"World Inequality Database\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"Named as flagship research outside the Bank and Fund that has expanded knowledge of inequality.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001082:20:0", "occurrence_id": "d7b47ce384cb9d7fd9e91bab", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Nonetheless, despite all this good recent work on inequality at the Bank and the Fund, the growth in\nresearch and public debate on inequality elsewhere has been so large that the Bretton Woods institutions\nhave become relatively smaller intellectual players in the field. In and of itself, this is clearly no bad thing.\nInequality truly is one of the fundamental challenges facing today’s human societies, and the greater\nvisibility and popularity of the topic is hugely welcome. Research institutes such as the World Inequality\nLab at the Paris School of Economics, which hosts the World Inequality Database and publishes an annual\n_World Inequality Report_, are producing important, pathbreaking work extending our knowledge of\ninequality around the world. Other examples include the team at UNU-WIDER, who host the World\nIncome Inequality Database (WIID) and have a had a long tradition of work on both income and wealth\ninequality, as well as the Stone Center on Socio-Economic Inequality at the City University of New York.\nOf course, most academic research on inequality is produced at universities around the world, whether\nthey host one of these institutes or not.\n\n\nFrom the perspective of the two institutions, however, two features of this particular surge in external\nresearch pose a challenge. First, the bulk of the inequality research currently conducted at the world’s\nleading research universities and institutes focuses on rich countries. This is not exceptional; the same is\ntrue of most other topics. Authors, journal editors and referees residing in the United States and Europe\ntend to be very interested in what is happening in their own countries. Second, a considerable share of\nthe new work on inequality appears to focus on rich people. This too is perhaps unsurprising, given the\nrapid rise in top incomes – and top income shares – in most countries for which reliable data is available.\nIt is also part of the key innovation that spurred much of this research which, as noted earlier, was the\nuse of alternative data sources to supplement the inadequate\n\nMention: \"World\nIncome Inequality Database\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"Listed with its host institute as a long-standing source of income and wealth inequality work.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001083:17:1", "occurrence_id": "733db22aeb97f509a7323e69", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: >model.|Endogenous LBD|\n|**WITCH**|transport, industry and<br>residential/commercial|Passengers and freight vehicles (both distinguished<br>according to traditional combustion, hybrid, plug in<br>hybrid and electric). The composition of vehicles is<br>determined by a Leontief function of a range of costs<br>(including battery, O&M, fuel and carbon costs)|Partly endogenous|\n|**C3IAM**|transport, industry and<br>residential/commercial|Intercity passenger transport, urban passenger<br>transport and freight transport. Urban passenger<br>transport is divided into public transport (bus and rail<br>transit), taxi, and private vehicles (cars and electric<br>bikes). Intercity passenger transport is divided into<br>private vehicles and four main business intercity<br>passenger transport (road transport, railway<br>transport, aviation and waterway transport). Freight<br>transport is divided into five main types: road<br>transport, railway transport, aviation transport,<br>waterway and pipeline transport.|Exogenous|\n|**BET EMF33**|Transport, Power, Heat|Passenger and freight|Exogenous|\n|**AIM-CGE**|Industry, residential,<br>commercial, transport|Road, Railway, Waterways, Air transport|Exogenous|\n|**GEM-E3-ICCS**|Firms, households|As regards the transport modes, the sectorial split<br>follows and further extends the classification used by<br>the GTAP database. This extension splits the GTAP<br>data in passenger and freight transport by mode and<br>further disaggregates the modes into rail and road as|Exogenous\n\nMention: \"GTAP database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Its sectoral classification is the base that the model's transport split extends.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001084:8:1", "occurrence_id": "495c8bc630ce79474ff6b118", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  calendar year.”_ The sample covers 107 armed conflicts between 1963\n\nand 2019 in 61 developing countries.\n\n\n**_c._** **_Data on debt restructurings_**\n\n\nData on debt restructurings are based on Asonuma and Trebesch (2016), which provide information on\n\nthe occurrence and duration of 204 restructurings globally between 1978 and 2019, 197 of which have\n\nbeen completed. The cases are further differentiated into two categories: preemptive restructurings and\n\npost-default restructurings. In preemptive restructurings, governments renegotiate with lenders while\n\nthey are still current on their loan payments. In post-default restructurings, governments unilaterally\n\ndefault and then renegotiate their debt. Among the 197 finished restructurings, 81 were preemptive, and\n\n116 were post-default. Many countries have multiple restructurings. All countries in the data set are\n\n\n4 According to EM-DAT, this is the value of all damages and economic losses directly or indirectly related to the\ndisaster.\n\n\n7\n\nMention: \"Data on debt restructurings\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Occurrence and duration of 204 restructurings, split into preemptive and post-default cases, are coded here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001084:8:1", "occurrence_id": "8b06067946fd7ace40bbe556", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  calendar year.”_ The sample covers 107 armed conflicts between 1963\n\nand 2019 in 61 developing countries.\n\n\n**_c._** **_Data on debt restructurings_**\n\n\nData on debt restructurings are based on Asonuma and Trebesch (2016), which provide information on\n\nthe occurrence and duration of 204 restructurings globally between 1978 and 2019, 197 of which have\n\nbeen completed. The cases are further differentiated into two categories: preemptive restructurings and\n\npost-default restructurings. In preemptive restructurings, governments renegotiate with lenders while\n\nthey are still current on their loan payments. In post-default restructurings, governments unilaterally\n\ndefault and then renegotiate their debt. Among the 197 finished restructurings, 81 were preemptive, and\n\n116 were post-default. Many countries have multiple restructurings. All countries in the data set are\n\n\n4 According to EM-DAT, this is the value of all damages and economic losses directly or indirectly related to the\ndisaster.\n\n\n7\n\nMention: \"EM-DAT\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Its definition covers damages and economic losses directly or indirectly related to a disaster.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001085:11:3", "occurrence_id": "3008bdc69ca0b8cd3b1d3573", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |<br>1.9% <br>|<br>Tea <br>|<br>0.4% <br>|\n|<br>Other bakery products, e.g. quiches, pizzas, pies (excluding<br>meat pies, fish pies and sweet corn) <br>|<br>1.7% <br>|<br>Other food products, e.g. homogenized baby food <br>|<br>0.3% <br>|\n|<br>* Catering services provided by restaurants, cafés, etc. in<br>places providing recreational, cultural, sporting or<br>entertainment services <br>|1.6% <br>|Salt <br>|0.0% <br>|\n|<br>Sugar, unrefined or refined, powdered, crystallized or in<br>lumps <br>|1.6% <br>|Mineral or spring waters; all drinking water sold<br>in containers <br>|0.0% <br>|\n|<br>Soft drinks<br>|1.6%|<br>||\n\n\n_Source: Pacific Poverty Assessment, Republic of the Marshall Islands chapter_\n\n\n\n10\n\nMention: \"Pacific Poverty Assessment\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Food item weights in the consumption basket are taken from this country chapter.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001087:6:0", "occurrence_id": "86199edfc621d2fa6c757627", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: also expect our results to contribute to the conversation on the costs of the pandemic not only\n\n\nacross space, but educational levels. Our findings indicate that the youngest children were the most\n\n\naffected. Given the importance of early childhood and primary education for human capital\n\n\ndevelopment, this evidence provides input for policy makers to design interventions that could\n\n\nhelp mitigate the medium- and long-term consequences of the pandemic ( <mark>Fuchs-Schündeln</mark> _<mark>et al</mark>_ <mark>.,</mark>\n\n\n<mark>2022)</mark> . Therefore, we hope that this research will be a timely guide for educational policy in\n\n\nGuatemala and other developing countries with similar education systems. While we are unable to\n\n\nstudy the consequences of the pandemic on learning, documenting that some students were more\n\n\naffected than others is a necessary step to consider when defining recovery policies.\n\n\nThe remainder of this paper is organized as follows. The next section describes trends in\n\n\neducational outcomes in Guatemala before and after the COVID-19 pandemic and introduces the\n\n\nwarning index for infection rates developed by the government and used by the Ministry of\n\n\nEducation to determine school restrictions on in-person instruction. Section 3 describes the\n\n\nadministrative panel data and our selected educational outcomes. Section 4 discusses our\n\n\nidentification strategy to estimate the effects of differential exposures to COVID-19. Section 5\n\n\npresents our findings and tests their robustness. We conclude in Section 6.\n\n#### **2. COVID-19 and educational outcome trends in Guatemala**\n\n\n_2.1. Changes in educational outcomes before and after the pandemic_\n\n\nEstimates using Guatemalan administrative records indicate that between-year school dropout\n\n\nincreased between 2018 and 2021. <sup>7</sup> While 7.8% of school-age children dropped out in 2019, this\n\n\n7 We define between-year school dropout when a student is observed in one school year but does\n\nMention: \"Guatemalan administrative records\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Between-year school dropout rates before and after the pandemic are computed from these school records.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001094:31:7", "occurrence_id": "cba6471f75f9e7b121a2682d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 045</u> <u>0.040</u>\n<u>Source: World Bank staff analysis using Romania MoF firm-level data, 2011-2019. Entry rate in industry j in year t is the ratio of the number of entrant firms to the average number</u>\nof firms in t and t-1. Exit rate is the ratio of the number of exiting firms to the average number of firms in t and t-1. Gross job creation is the ratio of the sum of employment created\nby firms to the average number of employments in t and t-1. Gross job destruction is the ratio of the sum of employment reductions of firms to the average number of employments\nin t and t-1. Net job creation rate is the ratio of the difference between employment created and reductions by firms to the average number of employments in t and t-1. Job reallocation\n(churn) rate is the sum of the absolute value of changes in employment divided by the average employment between years t and t-1. Employment share of young firms is the fraction\nof employees in total employment who are employed by firms aged less than 5 years. Employment growth dispersion is the standard deviation of employment growth of firms in an\nindustry, where employment growth is the change in employment divided by the average employment between years t and t-1. HHI is the Herfindahl-Hirschman Index of sector.\nGrowth of total assets is the change in total assets divided by the average of total assets between t and t-1. Concentration calculated as the sum of squared market (revenue) share of\neach firm operating in a sector. SOE variables are dummy variables taking the value one if national or subnational governments have ownership stake of 10-100% in the firm. The\nreference group is private-owned enterprises (POEs), defined as firms where national or subnational governments own\n\nMention: \"Romania MoF firm-level data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Industry entry and exit rates are constructed as ratios of entrants and exits to the average firm stock.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001095:52:0", "occurrence_id": "0b7bf7689dac13ec1a41ff5d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: (0.128)\n65-69 1.097\n(0.147)\n70-74 0.963\n(0.202)\n75-79 0.765\n(0.402)\n80-84 1.313\n(0.679)\n85+ 1.387\n(0.637)\nfemale 0.059\n(0.032)\nurban -0.025\n(0.020)\nEducation\n(Aggregate\nlevel)\n\n2 - Basic 0.402\n(0.032)\n3 Intermediate 0.642\n(0.038)\n4 Advanced 1.038\n(0.025)\n5 - Level not\nstated 0.561\n(0.158)\n4 regions l\nNorth -0.124\n(0.029)\nCenter -0.125\n(0.028)\nSouth 0.016\n(0.029)\nIntercept -2.254\n(0.106)\nNumber of\n<u>observations</u> <u>20518</u>\n<u>Source: author estimates using LFS 2021.</u>\n\n\n51\n\nMention: \"LFS 2021\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Over 20,000 observations are used to estimate returns by age, education, gender and region.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001096:6:0", "occurrence_id": "bcc0e3e188139515cd1fb050", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ranges from a small number of dwellings in rural areas to conurbations comprising multiple cities. These\ncluster boundaries function as spatial masks for which road networks were extracted from OSM and\nmodeled as directed graphs/digraphs (a set of vertices and a collection of directed edges that each connects\nan ordered pair of vertices). An evaluation of the OSM road networks in different functional hierarchies\nreveals high heterogeneity in the completeness of mapping. We focus on drivable road networks with higher\nfunctional hierarchy due to incomplete mapping of roads with lower hierarchy–namely residential roads\nand local pathways. Next, we integrate both pluvial and fluvial flood maps under 10 return periods from\nFathom’s latest global flood modeling effort and overlay them on road networks to evaluate local and\nnetwork-level flood hazard exposure <sup>35</sup> . We disrupt the road network based on location-specific flood\ninundation conditions and run travel simulations from sampled origins and destinations recording travel\ntrajectory, travel time, and travel distance. Finally, we compare the simulation results between dry and wet\n(flooded) conditions and across different flood intensities (i.e., return periods) to evaluate the resilience of\nroad transportation networks to flooding. The following sections detail our approach for network boundary\ndelineation, road network model creation, flood mapping, and travel simulation.\n\n\nFig. 1: Overall workflow diagram. This study consists of three main components: 1) delineate network boundaries\nleveraging global settlement dataset and generate convex hulls based on these boundaries (later used as spatial masks\nfor extracting the road network from OSM dataset), 2) process global pluvial and fluvial flood mapping products from\nFathom global flood models, and 3) simulate travel activities in dry as well as disrupted networks under 10 flood\nscenarios.\n###### Network boundary delineation\n\nTo tackle the aforementioned challenge of defining network boundaries that capture the functional network\nstructure, we use settlement cluster boundaries as a proxy. Since the goal of this study is\n\nMention: \"OSM\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Road networks within settlement masks are extracted and modelled as directed graphs.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001096:6:0", "occurrence_id": "c3a764fce51451249dba4b4e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ranges from a small number of dwellings in rural areas to conurbations comprising multiple cities. These\ncluster boundaries function as spatial masks for which road networks were extracted from OSM and\nmodeled as directed graphs/digraphs (a set of vertices and a collection of directed edges that each connects\nan ordered pair of vertices). An evaluation of the OSM road networks in different functional hierarchies\nreveals high heterogeneity in the completeness of mapping. We focus on drivable road networks with higher\nfunctional hierarchy due to incomplete mapping of roads with lower hierarchy–namely residential roads\nand local pathways. Next, we integrate both pluvial and fluvial flood maps under 10 return periods from\nFathom’s latest global flood modeling effort and overlay them on road networks to evaluate local and\nnetwork-level flood hazard exposure <sup>35</sup> . We disrupt the road network based on location-specific flood\ninundation conditions and run travel simulations from sampled origins and destinations recording travel\ntrajectory, travel time, and travel distance. Finally, we compare the simulation results between dry and wet\n(flooded) conditions and across different flood intensities (i.e., return periods) to evaluate the resilience of\nroad transportation networks to flooding. The following sections detail our approach for network boundary\ndelineation, road network model creation, flood mapping, and travel simulation.\n\n\nFig. 1: Overall workflow diagram. This study consists of three main components: 1) delineate network boundaries\nleveraging global settlement dataset and generate convex hulls based on these boundaries (later used as spatial masks\nfor extracting the road network from OSM dataset), 2) process global pluvial and fluvial flood mapping products from\nFathom global flood models, and 3) simulate travel activities in dry as well as disrupted networks under 10 flood\nscenarios.\n###### Network boundary delineation\n\nTo tackle the aforementioned challenge of defining network boundaries that capture the functional network\nstructure, we use settlement cluster boundaries as a proxy. Since the goal of this study is\n\nMention: \"OSM dataset\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Convex hulls of settlement boundaries define the extent from which the network is pulled.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001102:25:0", "occurrence_id": "979b494ba94bcaa052a2418f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **APPENDIX 1**\n\n**Incidence of indirect taxes from a life-cycle perspective: VAT corrections**\n\nThe main results presented in the paper are based on household income as the main measure\nof welfare (as used for official welfare statistics in Serbia), thus showing the distributional\nimpact of taxes with respect to current household income. This is considered <mark>relevant for policy</mark>\n<mark>making purposes in the short to medium term.</mark> An alternative question, explored further in this\nAppendix, is to understand the incidence of taxes over the life cycle with respect to household\nconsumption as a proxy for permanent income. Assessing the incidence of indirect taxes from\na life-cycle perspective is possible through adjustments by the income/consumption ratio,\nwhich is the VAT correction as per the Commitment to Equity guidebook.\n\n<mark>For the permanent income concept of welfare, s</mark> tandard <mark>life-cycle theory argues that</mark>\n<mark>consumption is a good proxy. With respect to consumption, taxes at a fixed rate on</mark>\n<mark>consumption like VAT should in theory be a neutral tax over the life cycle.</mark> The other\nmotivation for adjustments comes from negative consumable income observations. In cases\nwhen the reported consumption is much higher than reported income, either due to\nunderreporting of income, dissaving, or borrowing, some of the households with consumption\nmuch higher than reported income end up with negative consumable income. In the case of\nSILC 2017 data for Serbia, the number of <mark>those with negative consumable income is very small,</mark>\n<mark>only 0.2 percent.</mark>\n\n<mark>Adjustments take the form of scaling VAT values by the income/consumption ratio. After</mark>\n<mark>computing the amount of indirect taxes paid as in the main\n\nMention: \"SILC 2017 data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Households with negative consumable income are identified and VAT values rescaled by the income-to-consumption ratio.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001103:18:1", "occurrence_id": "a6a0f11b65785bdcb98ecbef", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  on agricultural productivity, inputs, and production</mark>\n\n<mark>practices including those that are relevant for climate change adaptation and mitigation.</mark> <sup>25</sup> [^25: This is done through the <mark>LSMS Integrated Survey on Agriculture (LSMS-ISA) project in collaboration with national statistical</mark>\n<mark>offices of client countries. F</mark> or more information about the LSMS-ISA project, see:\n<u>[https://www.worldbank.org/en/programs/lsms/initiatives/lsms-ISA.](https://www.worldbank.org/en/programs/lsms/initiatives/lsms-ISA)</u>\n<u>26 This is part of the Living Standards Measurement Study – Plus (LSMS+) program, established in 2016. For more information</u>\n[about the LSMS+ program see https://www.worldbank.org/en/programs/lsms/initiatives/lsms-plus.](https://www.worldbank.org/en/programs/lsms/initiatives/lsms-plus)] <mark>This is critical</mark>\n\n<mark>information in the context of climate change, given that most of the global poor are engaged in agriculture</mark>\n\n<mark>and that this sector is highly vulnerable to climatic variability – especially in low-income countries where</mark>\n\n<mark>smallholder farmers rely on rainfall as the primary source of irrigation. In these settings, agriculture is</mark>\n\n<mark>possibly the leading channel to explain climate migration and, as such, detailed agricultural data collected</mark>\n\n<mark>at the household level are key information for studying this nexus. The community survey complementing</mark>\n\n<mark>the household level data collection in the LSMS survey system is also an important source of information</mark>\n\n<mark>for\n\nMention: \"LSMS survey system\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"inform\", \"summary\": \"Its community survey complements household data as a key source for the agriculture-migration nexus.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001104:7:0", "occurrence_id": "edb7559d5671f73047002aea", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: participation in different markets through an economic lens, abstracting from the political and social\npreferences of individual policy makers. This also allows to provide guidance on the different alternatives\nand tailored solutions for SOE reform beyond ownership transformation and which can be strongly related\nto the type of sectors where these companies operate. <sup>11</sup> [^11: Sanchez-Navarro, Goodwin, & Kikeri (2021) present a full description of the different alternatives for reform building on this taxonomy is\npresented in the CPSD SOE Knowledge note indicating the different menu of options of reform. For instance, it proposes management\narrangements and concessions can be more suitable for natural monopoly sectors, whereas divestiture may be more suitable for large SOEs in\ncompetitive sectors that could be served by (or in competition with) the private sector.]\n\n\nThe proposed taxonomy is also critical to allow for cross-country comparison of the SOE footprint. Political\neconomy and social objectives vary from country to country, and, as such, it is impossible to objectively\njustify the presence of the state in the market on the basis of those objectives. However, since economic\nmarket failures are uniquely identified based on the characteristics of the markets in which goods and\nservices are produced and the technologies used, the proposed taxonomy allows analysts to compare\nSOEs’ presence in the economy across countries according to a homogeneous classification system. This\ntaxonomy is already feeding into the Businesses of the State Database of the World Bank Group which is\ndescribed in a companion paper to this work (Dall’Olio et al., 2022).\n\n\nThe remainder of this paper is organized as follows. Section II provides a review of the economic literature\non SOEs and market failures. Section III provides the conceptual framework according to which economic\nsectors are classified. Section IV analyzes the economic sectors, disaggregated to the level of the NACE\nfour-digit classification, and classifies each of them into the three proposed categories. Section V presents\na data driven exercise to validate the taxonomy classification. Section VI concludes.\n\n#####\n\nMention: \"Businesses of the State Database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"The proposed state-ownership taxonomy is already being fed into this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001106:4:0", "occurrence_id": "08a84250d79737c1e88348bd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: They assume private returns to a year of schooling of 8%, schooling time lost to be 0.33 years,\n\n\nimplying 2.67 percent lower future earnings. After considering the assumed mitigating effects\n\n\nof distance learning, the average annual loss per student is (2011 PPP) $458. Finally, Hanushek\n\n\nand Woessmann (2020: 8–9) use the OECD’s Survey of Adult Skills predominantly conducted in\n\n\nhigh-income countries and estimate that a loss of 0.25 school years predicts a 1.9 percent\n\n\ndecrease of income over one’s career, while 1 lost year predicts a 7.7 percent decrease.\n\n\nThese simulations are certainly useful. However, they also have inherent weaknesses\n\n\nwhen assessing long-term impacts because the underlying parameters are not directly\n\n\nestimated from a school-disruption event. Instead, the parameters are the best guesses of what\n\n\nthe overall disruption might look like, by necessity simplifying and omitting the complex system\n\n\nof response feedback loops initiated by the disruption. Among the key omissions are the\n\n\nmitigating bottom-up activities of parents and students in response to school closures, such as\n\n\nprivate tutoring and mentoring. Moreover, the simulations are by design assessing the global\n\n\neffects of the schooling disruption, setting the issue of more disaggregate effects aside.\n\n\nThis paper estimates the present value of lost income for a future Kuwaiti civil servant\n\n\nwhose studies are disrupted by the COVID-19 pandemic. We do so by building on Bilo et al.\n\n\n(2021), who use a unique data set and the natural experiment of the Gulf War induced school\n\n\ndisruption, to estimate the private costs of the pandemic. We argue that the Gulf War induced\n\n\na school disruption that is similar to the schooling disruption caused by the COVID-19 pandemic\n\n\nand therefore, the methodology is well suited to simulating the long-term impacts of schooling\n\n\ndisruptions in the context of the pandemic. Equally importantly, the Gulf War disruption\n\n\nhappened almost 30 years ago, allowing us\n\nMention: \"OECD’s Survey of Adult Skills\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"A cited study uses it, mostly in high-income countries, to estimate career earnings losses from lost schooling.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001110:28:0", "occurrence_id": "6345ed2eaf0b68d21b015b4c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 5: Aggregate tax revenue-based progressivity indicators**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nSource: authors calculations based on UNU-WIDER World Revenue Dataset\n\n\n27\n\nMention: \"UNU-WIDER World Revenue Dataset\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Aggregate tax revenue aggregates underlying the progressivity indicators come from this dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001113:34:1", "occurrence_id": "d6a2bc36b7eb9c4c98c9cc23", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: /u> <u>(0.130)</u> <u>(0.166)</u>\n**Admin. 1 fixed effects** Yes Yes Yes Yes Yes Yes Yes Yes Yes\n**Urban/rural residence** Yes Yes Yes Yes Yes Yes Yes Yes Yes\n**Baseline hazard (age in year)** Yes Yes Yes Yes Yes Yes Yes Yes Yes\n**Calendar year fixed effects** Yes Yes Yes Yes Yes Yes Yes Yes Yes\n**<u>N obs.</u>** <u>2920224</u> <u>8615</u> <u>79127</u> <u>60028</u> <u>80370</u> <u>177099</u> <u>384767</u> <u>81718</u> <u>106033</u>\n<u>Source: Authors’ calculations based on DHS surveys and UCDP conflict data.</u>\nNotes: *p<0.05; **p<0.01; ***p<0.001. Clustered standard errors in parentheses.\n\n\n33\n\nMention: \"DHS surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Nearly three million individual observations provide the outcomes in the regressions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001113:34:1", "occurrence_id": "5bd9755a2548edca4be9ad9e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: /u> <u>(0.130)</u> <u>(0.166)</u>\n**Admin. 1 fixed effects** Yes Yes Yes Yes Yes Yes Yes Yes Yes\n**Urban/rural residence** Yes Yes Yes Yes Yes Yes Yes Yes Yes\n**Baseline hazard (age in year)** Yes Yes Yes Yes Yes Yes Yes Yes Yes\n**Calendar year fixed effects** Yes Yes Yes Yes Yes Yes Yes Yes Yes\n**<u>N obs.</u>** <u>2920224</u> <u>8615</u> <u>79127</u> <u>60028</u> <u>80370</u> <u>177099</u> <u>384767</u> <u>81718</u> <u>106033</u>\n<u>Source: Authors’ calculations based on DHS surveys and UCDP conflict data.</u>\nNotes: *p<0.05; **p<0.01; ***p<0.001. Clustered standard errors in parentheses.\n\n\n33\n\nMention: \"UCDP conflict data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Conflict event locations are merged to the survey respondents.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001114:6:1", "occurrence_id": "5426c14b4e485ed65905693d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  utility estimation provides researchers and practitioners\nwith a unified approach to diagnosing and mitigating bias while correctly accounting for\nchoice error. Our Stata and SurveyCTO templates facilitate the measurement of WTP\nusing MPL.\n\nThe next section describes WTP elicitation via MPL. It discusses common patterns\nin MPL data, presents a framework to think about error in such data, and summarizes\nhow the existing literature has addressed MSB, NSB and interval identification. Section\n3 proposes to combine order randomization with a random utility model for MPL choices\nand uses data from an experiment in Cape Town, South Africa to show how this approach\naffects measured WTP. We demonstrate in particular that order biases can significantly\naffect elicited WTP. We also describe how researchers might apply our framework to\ninvestigate other sources of bias that arise through MPL design decisions. In section 4, we\ndiscuss features of MPLs for WTP elicitation that we identify as useful for improving data\nquality. These features are implemented in an accompanying template for implementing\nan MPL in SurveyCTO and estimating WTP from the data, described in detail in a\ntechnical appendix (Appendix S2).\n\n\n4\n\nMention: \"data from an experiment in Cape Town, South Africa\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Elicited willingness to pay is used to show how order biases affect measured values.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001115:8:0", "occurrence_id": "59c6230863662a20f006251a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: business activities. As a result, the share of sectors that were exposed to foreign equity limit (FEL)\n\n\nfell from 34.1% in 2016 to 7.6% in 2021 (figure 1a). Thus the reform left several sectors fully\n\n\nopen to FDI, such as: (i) relatively small activities fully or partially reserved for Micro, Small,\n\n\nand Medium Enterprises (MSMEs) and (ii) sectors considered to be of public interest that require\n\n\nminority or no foreign ownership. With the remaining restrictions, the share of business activities\n\n\nsubject to at least one investment restriction decreased from 38.5% to 20.4% of all economic sectors\n\n\n(figure 1b).\n\n\nFigure 1: Evolution of Investment Restriction in Indonesia\n\n\na. Share of sector with FEL restriction b. Share of sector with any restriction\nSource: Author calculation based on series of Presidential Decree on Investment\n\n###### **2.2 Data**\n\n\nTwo sets of data are used in this analysis: realized FDI and DDI values, sourced from Indonesia’s\n\n\nMinistry of Investment (henceforth BKPM); and planned FDI sourced from fDi markets data which\n\n\nalso proxies investor sentiments/expectations. The BKPM data includes all types of FDI while the\n\n\nfDi markets data included in this analysis does not include mergers and acquisitions. BKPM data\n\n\nis at a quarterly level up to September 2022 while fDi market data is at a monthly level up to March\n\n\n2022.\n\n\n_Klasifikasi Baku Lapangan Usaha Indonesia_ or KBLI is the official business classification in\n\n\nIndonesia, published by the Indonesia Centre of Statistics with reference to the International Stan\n\ndard Classification of All Economic Activities (ISIC). The regulation was at a highly disaggregated\n\n\nlevel at business activities of 5-digit KBLI (see list in the Annex Table A1). However, both data\n\n\n6\n\nMention: \"fDi markets data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Monthly planned investment announcements proxy investor expectations, excluding mergers and acquisitions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001115:8:0", "occurrence_id": "9dd04ab3ce61daea2d29f4b1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: business activities. As a result, the share of sectors that were exposed to foreign equity limit (FEL)\n\n\nfell from 34.1% in 2016 to 7.6% in 2021 (figure 1a). Thus the reform left several sectors fully\n\n\nopen to FDI, such as: (i) relatively small activities fully or partially reserved for Micro, Small,\n\n\nand Medium Enterprises (MSMEs) and (ii) sectors considered to be of public interest that require\n\n\nminority or no foreign ownership. With the remaining restrictions, the share of business activities\n\n\nsubject to at least one investment restriction decreased from 38.5% to 20.4% of all economic sectors\n\n\n(figure 1b).\n\n\nFigure 1: Evolution of Investment Restriction in Indonesia\n\n\na. Share of sector with FEL restriction b. Share of sector with any restriction\nSource: Author calculation based on series of Presidential Decree on Investment\n\n###### **2.2 Data**\n\n\nTwo sets of data are used in this analysis: realized FDI and DDI values, sourced from Indonesia’s\n\n\nMinistry of Investment (henceforth BKPM); and planned FDI sourced from fDi markets data which\n\n\nalso proxies investor sentiments/expectations. The BKPM data includes all types of FDI while the\n\n\nfDi markets data included in this analysis does not include mergers and acquisitions. BKPM data\n\n\nis at a quarterly level up to September 2022 while fDi market data is at a monthly level up to March\n\n\n2022.\n\n\n_Klasifikasi Baku Lapangan Usaha Indonesia_ or KBLI is the official business classification in\n\n\nIndonesia, published by the Indonesia Centre of Statistics with reference to the International Stan\n\ndard Classification of All Economic Activities (ISIC). The regulation was at a highly disaggregated\n\n\nlevel at business activities of 5-digit KBLI (see list in the Annex Table A1). However, both data\n\n\n6\n\nMention: \"BKPM data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Quarterly realised investment values, covering all investment types, are reported by this ministry.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001116:16:0", "occurrence_id": "5f9a22c7ad0154ff50861de2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 3 shows, for rural areas, the main driver for the almost 7 percentage point reduction in poverty was the\n\nincrease in consumption of food and non-food items, while higher asset ownership and improvement in\n\ndwelling characteristics played a minor role. In urban areas, consumption of food and non-food items\n\ncontributed to a lesser extent to reduce poverty. However, the higher household consumption could not\n\ncompensate for the increase in average household size, nor the deterioration of dwelling conditions.\n\nFurther validating the urban results, we see that no change in an individual component of the model\n\nwould alter the net poverty increase.\n\n\n**Figure 3: Decomposition of changes in welfare using Urban and Rural models**\n\n\n\n\n\n**Note** : Results use a Shapley decomposition on the different components of the prediction of the welfare vector. The different\ncategories include the variables of the urban and rural model independently as shown in Table 3. Model error is the average\nerror across the multiple imputations.\n**Source:** IE-LFS 2019-20 and AWMS 2023 (R3).\n\n\nDiscussion of economic trends in Afghanistan and their effect on poverty levels\n\nBefore socioeconomic data stopped being collected, close to half of the population in Afghanistan was\nliving below the national poverty line. As shown in Figure 4, while the national poverty rate in 2019-20\nwas lower than the 55 percent estimated in 2016-17, <sup>14</sup> [^14: World Bank (2018) _<u>[Poverty in Afghanistan : Results based on ALCS 2016-17.](https://documents1.worldbank.org/curated/en/451111535402851523/pdf/AUS0000426-REVISED-ALCS-Poverty-Chapter-upload-v2.pdf)</u>_ Washington, D.C. : World Bank Group.] poverty trends were showing diverging patterns\nin urban and rural areas. In urban areas, the observed increase in poverty is the result of the compounding\neffect of the progressive decline in aid as well as the onset of the COVID-19 crisis, while in rural areas,\npoverty was on a declining path spurred\n\nMention: \"IE-LFS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Urban and rural welfare models are estimated on this survey to predict consumption.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001116:16:0", "occurrence_id": "f5ef7cae5d9819b2d9bef8d0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 3 shows, for rural areas, the main driver for the almost 7 percentage point reduction in poverty was the\n\nincrease in consumption of food and non-food items, while higher asset ownership and improvement in\n\ndwelling characteristics played a minor role. In urban areas, consumption of food and non-food items\n\ncontributed to a lesser extent to reduce poverty. However, the higher household consumption could not\n\ncompensate for the increase in average household size, nor the deterioration of dwelling conditions.\n\nFurther validating the urban results, we see that no change in an individual component of the model\n\nwould alter the net poverty increase.\n\n\n**Figure 3: Decomposition of changes in welfare using Urban and Rural models**\n\n\n\n\n\n**Note** : Results use a Shapley decomposition on the different components of the prediction of the welfare vector. The different\ncategories include the variables of the urban and rural model independently as shown in Table 3. Model error is the average\nerror across the multiple imputations.\n**Source:** IE-LFS 2019-20 and AWMS 2023 (R3).\n\n\nDiscussion of economic trends in Afghanistan and their effect on poverty levels\n\nBefore socioeconomic data stopped being collected, close to half of the population in Afghanistan was\nliving below the national poverty line. As shown in Figure 4, while the national poverty rate in 2019-20\nwas lower than the 55 percent estimated in 2016-17, <sup>14</sup> [^14: World Bank (2018) _<u>[Poverty in Afghanistan : Results based on ALCS 2016-17.](https://documents1.worldbank.org/curated/en/451111535402851523/pdf/AUS0000426-REVISED-ALCS-Poverty-Chapter-upload-v2.pdf)</u>_ Washington, D.C. : World Bank Group.] poverty trends were showing diverging patterns\nin urban and rural areas. In urban areas, the observed increase in poverty is the result of the compounding\neffect of the progressive decline in aid as well as the onset of the COVID-19 crisis, while in rural areas,\npoverty was on a declining path spurred\n\nMention: \"AWMS 2023\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The third round supplies the more recent observations to which welfare is predicted.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001120:11:0", "occurrence_id": "5f4937a15286b781f506ba7f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The analysis using the customs data, which helps validate the results from BPS data, estimates\n\na simpler version of Equation (1) described in Appendix B. The results of this estimation are\n\nrepresented graphically, as before showing marginal effects along with 95% confidence intervals.\nThere are two important differences across the BPS and the customs data analyses. _First_, monthly\ncustoms outcomes are aggregated and the analysis is conducted with data at the country-firmquarterly level. _Second_, since the customs data comprise only trading firms, only trade-related\n\noutcomes from BPS data are compared using the customs data. Further, when drawing comparisons\n\n\nwith the customs data, we restrict the sample of BPS data to include only exporters or importers.\n\n\nTo capture the global engagement of firms, we construct several measures using the BPS data. Two\n\n\nstandard measures are indicator variables, one for whether the firm exports and one for whether the\n\n\nfirm imports before the pandemic in 2019. These two measures capture firms’ direct engagement in\n\nglobal markets. Another measure is an indicator variable for firms that both export _and_ import, that\n\nare designated as GVC firms. The final and more novel measure captures firms’ indirect engagement\n\n\nin global markets through an indicator variable for whether the firm sold to domestic exporters or\n\nmultinationals in 2019 but did not directly export or import. We refer to these firms as suppliers\n\nof GVC firms. Since these four measures may not be mutually exclusive, we also construct four\n\n\nmutually exclusive categories of global engagement to contrast with firms that had no engagement\n\n\nat all: only indirect engagement; only direct engagement (export or import); both indirect and direct\n\nengagement (with direct engagement being measured with either export or import), and finally\n\n\nfirms that are indirectly engaged and are also a GVC firm in 2019. Except for the indicator variable\n\nfor firms that export, other measures of global engagement are only available for the countries\n\n\nsurveyed in round 3 including the trade module. Hence, the analysis\n\nMention: \"BPS data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Customs records are used to check the results previously obtained from this business survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001105:18:0", "occurrence_id": "f5719ea47384b5ec7f1830d4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: with a large margin of error. Our preschool interaction is only significant at 15% (t-statistic is 1.61)\n\n\nin the larger sample (column 3). One additional preschool per 1,000 children appears to increase\n\n\nthe number of female workers by 1 percent. The coefficient remains positive but becomes\n\n\ninstead close to zero in the restricted TFP sample (column 4).\n\n\nSecond, we corroborate the finding that plants are more able to tap into female workers’\n\n\nskills by using the data from SAKERNAS. We use repeated cross-section, spanning the years from\n\n\n2007 to 2014, and estimate individual-level regression:\n\n\n𝑦𝑖𝑑𝑡 = 𝜃𝐹𝑒𝑚𝑎𝑙𝑒𝑖 + 𝜑𝐹𝑒𝑚𝑎𝑙𝑒𝑖𝑃𝑟𝑒𝑠𝑐ℎ𝑜𝑜𝑙𝑑𝑡+𝚪𝐗𝐢𝐝𝐭 + 𝛾𝑑𝑡 + 𝜖𝑖𝑑𝑡 (5)\n\n\nwhere 𝑦𝑖𝑑𝑡 is a labor market outcome for individual _i_ in district _d_ at time _t._ We analyze three\n\n\nbinary outcomes: employment in manufacturing, employment in services and wage\n\n\nemployment. 𝐹𝑒𝑚𝑎𝑙𝑒𝑖 is a dummy variable equal to 1 if the respondent is a woman, and\n\n\n𝑃𝑟𝑒𝑠𝑐ℎ𝑜𝑜𝑙𝑑𝑡 is defined based on PODES data: number of preschools per child in district _d_ at time\n\n\n_t._ 𝐗𝐢𝐝𝐭 is a vector of time-variant variables, which include education, experience, and experience\n\n\nsquared. 𝛾𝑑𝑡 is a district-time fixed effect.\n\n\nWhile we cannot establish a causal impact, we can explore associations between\n\n\npreschool availability and women’s employment. Table 10 shows that being a woman is\n\n\nnegatively associated with employment in manufacturing, services and wage employment.\n\n\nHowever, the availability of preschools in the district appears to reduce this gender penalty. The\n\n\ncoefficient on interaction between 𝐹𝑒𝑚𝑎𝑙𝑒𝑖 and our measure of preschool density is positive and\n\n\nstrongly significant. An increase in one standard deviation (0.003 additional preschool per child)\n\n\nincreases the likelihood of employment in manufacturing by 0.27 percentage points more for\n\n\nwomen.\n\n\nWe also limit the sample to women, and check whether preschool availability affects\n\n\nwomen\n\nMention: \"data from SAKERNAS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Repeated cross-sections are used to estimate individual-level labour market outcomes by gender and preschool availability.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001105:18:0", "occurrence_id": "aa7198d365a19d3611001254", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: with a large margin of error. Our preschool interaction is only significant at 15% (t-statistic is 1.61)\n\n\nin the larger sample (column 3). One additional preschool per 1,000 children appears to increase\n\n\nthe number of female workers by 1 percent. The coefficient remains positive but becomes\n\n\ninstead close to zero in the restricted TFP sample (column 4).\n\n\nSecond, we corroborate the finding that plants are more able to tap into female workers’\n\n\nskills by using the data from SAKERNAS. We use repeated cross-section, spanning the years from\n\n\n2007 to 2014, and estimate individual-level regression:\n\n\n𝑦𝑖𝑑𝑡 = 𝜃𝐹𝑒𝑚𝑎𝑙𝑒𝑖 + 𝜑𝐹𝑒𝑚𝑎𝑙𝑒𝑖𝑃𝑟𝑒𝑠𝑐ℎ𝑜𝑜𝑙𝑑𝑡+𝚪𝐗𝐢𝐝𝐭 + 𝛾𝑑𝑡 + 𝜖𝑖𝑑𝑡 (5)\n\n\nwhere 𝑦𝑖𝑑𝑡 is a labor market outcome for individual _i_ in district _d_ at time _t._ We analyze three\n\n\nbinary outcomes: employment in manufacturing, employment in services and wage\n\n\nemployment. 𝐹𝑒𝑚𝑎𝑙𝑒𝑖 is a dummy variable equal to 1 if the respondent is a woman, and\n\n\n𝑃𝑟𝑒𝑠𝑐ℎ𝑜𝑜𝑙𝑑𝑡 is defined based on PODES data: number of preschools per child in district _d_ at time\n\n\n_t._ 𝐗𝐢𝐝𝐭 is a vector of time-variant variables, which include education, experience, and experience\n\n\nsquared. 𝛾𝑑𝑡 is a district-time fixed effect.\n\n\nWhile we cannot establish a causal impact, we can explore associations between\n\n\npreschool availability and women’s employment. Table 10 shows that being a woman is\n\n\nnegatively associated with employment in manufacturing, services and wage employment.\n\n\nHowever, the availability of preschools in the district appears to reduce this gender penalty. The\n\n\ncoefficient on interaction between 𝐹𝑒𝑚𝑎𝑙𝑒𝑖 and our measure of preschool density is positive and\n\n\nstrongly significant. An increase in one standard deviation (0.003 additional preschool per child)\n\n\nincreases the likelihood of employment in manufacturing by 0.27 percentage points more for\n\n\nwomen.\n\n\nWe also limit the sample to women, and check whether preschool availability affects\n\n\nwomen\n\nMention: \"PODES data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"District preschool counts per child are constructed from these village potential records.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001125:8:0", "occurrence_id": "c5d0ef0e6080348bb347f2fd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: follow-up sample). On the other hand, several demographic indicators are quite different between the\ntwo samples: while women, pregnant women and women who had given birth in the previous 6 months\nhad been oversampled during the first survey, the follow-up sample was selected more randomly. In terms\nof socio-economic data, the two samples are relatively similar (Table 1). This also includes access to private\nand public means of transportation, with approximately a fifth of the respondents’ households owning a\nmotorcycle, about three-quarters having no private transport vehicles, and one in ten respondents not\nhaving access to any public transport. However, even among the respondents whose households own at\nleast one transport vehicle, not all (especially women) have the vehicle available for their use, the share\nbeing much higher in the 2021 sample than the 2022 one.\n\n**_Table 1:_** _Basic statistics and socio-economic characteristics of the survey samples collected in 2021 and 2022_\n\n\n\n_First survey_\n\n\n\n_Follow-up_\n\n\n\n_<u>(2021)</u>_ _<u>(2022)</u>_\n\n\n\n_<u>(2021)</u>_\n\n\n\nSource: Household surveys collected by the team in Sud and Nippes in 2021 and 2022\n\n\nBased on the responses provided in the two survey rounds, the study analyzed the relative importance of\ntransport related barriers among all the various constraints that inhibit household access to health care,\nschooling, and livelihood opportunities. Based on the results of the survey implemented in 2022, statistical\nanalysis was also conducted to assess whether there are statistically significant differences in health,\neducation and market access behaviors between the households living located in villages within 2 km of\na road damaged by the earthquake versus those living in the unaffected villages.\n\nComplementing the surveys, interviews were conducted with local representatives of a number of\ndevelopment organizations in Haiti, including international and local (UNICEF, Action Aid,\n\nMention: \"Household surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Two rounds fielded in these departments provide the sample characteristics and constraint responses.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001126:27:1", "occurrence_id": "70b011f069f235e0557ea9a9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  0.99 0.99\nKleibergen-Paap rk LM statistic 11.90*** 12.55*** 11.47***\n<u>Cragg-Donald Wald F statistic</u> <u>12.37</u> <u>14.48</u> <u>11.63</u>\nNotes: Standard errors clustered at the country level in parentheses, *** p<0.01, ** p<0.05, *\np<0.1. Here common controls and constant terms are omitted for brevity. All regressions include\nboth country fixed effects and year fixed effects. The dependent variable at the second stage is\nDGE-based output informality (in percent of GDP). Domestic credit to private sector is in percent\nof GDP (taken from WDI). See sections 3 and 4 for detailed data descriptions and empirical\nmethods. GDP per capita is in constant 2010 U.S. dollars. Data are between 2000 and 2018.\n\n\n26\n\nMention: \"WDI\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Domestic credit to the private sector as a share of GDP is taken from this indicator series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001131:4:1", "occurrence_id": "79b5edb466e436ca61dec611", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  the last 50 years. Our idea was to elicit their views\n\n\nwithout specifically referring to certain crises. The second question asked them to order from first to third the worst\n\n\ncrises in Colombia during the last 10 years, with the following options: illegal drug trafficking, internal armed conflict,\n\n\nand COVID-19. <sup>4</sup> Responses to both questions confirmed that our treatment increased the salience of the pandemic.\n\n\nIndividuals in the treatment group raised the COVID-19 crisis in the open-ended question more often than did the\n\n\ncontrol group. They also ranked COVID-19 as a worse crisis than illegal drug trafficking and internal armed conflict.\n\n\nWe document four main findings. First, our results strongly support the validity of the _resentment channel_ whereby\n\n\ntreated natives resent migrants during economic crises. In particular, treated individuals consistently reported more\n\n\nnegative attitudes towards migrants than did the control group. Natives primed with the COVID treatment were less\n\n\nlikely to think an immigrant was poor due to circumstances beyond their control (0.07 standard deviations (sd) lower\n\n\nthan the control group). These same natives were more likely to hold negative opinions about both the economic\n\n\nimpacts of migrants (0.07 sd lower than the control group) and about their tax contributions (0.19 sd lower than the\n\n\ncontrol group). The effects are small for the first two outcomes but larger for the latter case. These responses may\n\n\n2In practice, migration from Venezuela to Colombia is likely higher still as many migrants may never formally register due to fear of deportation.\n3There is also evidence that attitudes towards migrants in Colombia have deteriorated: according to Gallup’s Migrant Acceptance Index, it ranks\nthird among countries where acceptance dropped the most between 2016 and 2019.\n\n4Of course, there have been other crises in Colombia during this period; this list did not intend to be comprehensive but only to offer relevant\noptions besides COVID-19).\n\nMention: \"Migrant Acceptance Index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its ranking documents the fall in acceptance of migrants over this period.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001133:4:0", "occurrence_id": "6f0a94046a745de5f15a59f0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Campaign improves childhood cognitive skills. The direct role of sanitation on educational outcomes suggests that\n\n\nlack of access to sanitation is a threat to human capital formation, which is an important driver of long-term growth\n\n\nand development. Despite its benefits, governments in developing countries have not provided sufficient investment,\n\n\nadministrative support, and continuous oversight for the substantial provision of basic sanitation services.\n\n\nTable 1 provides a first look at the progress made in access to electricity and water compared to sanitation\n\n\nservices in low-income countries. Over the 2010 to 2019 period, access to electricity and water services increased\n\n\nby 14.07 and 7.97 percentage points, respectively. Meanwhile, access to sanitation services increased by only 0.68\n\n\npercentage points in low-income countries. This marginal increase in access to sanitation may be partly driven by the\n\n\nfact that there is less public awareness of, and perhaps, lower priority given to, the sanitation problem— i.e., sanitation\n\n\nis a “low visibility good” in developing countries— compared to electricity or water services.\n\n\n**Table 1:** The change in access to basic public goods in low-income countries, 2010-2019\n\n\n<u>Public good/service</u> <u>2010</u> <u>2019</u> <u>Change (% Points)</u>\n\n\nElectricity (% of population) 26.85 40.92 14.07\nWater services (% of population) 49.85 57.80 7.95\nSanitation services (% of population) 44.56 45.24 0.68\n\n\nSource: Authors’ calculation of World Bank’s World Development Indicators. Access to electricity is the percentage of population with access to electricity. Water services (% of population) is the percentage of people\nusing at least basic water services. This indicator encompasses both people using basic water services as well as\nthose using safely managed water services. Basic drinking water services is defined as\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Access to electricity, water and sanitation services over the decade are compiled from these indicators.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001136:11:0", "occurrence_id": "2a08029e506c13ef1394c9ae", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **4. The Broken Relationship between Imports and Labor Markets: A Bartik Analysis**\n\n\n  - **Overview of the Bartik Approach**\n\nCanonical trade models predict that increased trade increases factor income and alters the wage\nbill. However, recent empirical evidence suggests that labor markets are not as flexible as most\nmodels assume them to be, implying that adjustment to changes in trade volumes can be very\nslow (Autor, Dorn & Hanson, 2013; Dix-Carneiro & Kovak, 2014). Furthermore, since the effects\nof trade are typically heterogeneous across sectors, individuals and regions are differentially\nexposed to sector-specific shocks, giving rise to distributional concerns.\n\nIn the previous section we analyzed how trade agreements impact trade flows from Egypt to the\nrest of the world. Liberalization policies do seem to have a positive effect on exports but\nnegligible on imports. Yet, labor market outcomes have worsened. To estimate the relationship\nbetween labor markets and trade, we follow a common methodological framework to specific\nlocal conditions and employ a geography-based Bartik (1991) approach to quantify exogenous\nexport demand shocks from the US and EU countries for Egyptian imports.\n\nBartik models exploit the spatial variation in sectoral-composition assuming that changes in\nsector-specific foreign demand is different across regions. In other words, regions more exposed\nto trade for a particular sector and with major employment for the same sector will be more\naffected from a trade shock particular to the sector. Differences in trade exposure of regions to\nthis shock serves as an identification tool. For the trade exposure to be regionally identifiable,\nsome barriers or rigidities in the labor market should exist. Sluggish or segmented regional labor\nmarkets are a fundamental principal of the approach (Raymond et al, 2020).\n\nFigures 8 and 9 provide evidence of the concentrated trade exposure and segmented labor\nmarkets. Using the Enterprise Census (EC) from 2017 and the Labor Force Survey (LFS) from the\nsame year,\n\nMention: \"Enterprise Census\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Establishment locations and employment are used to map trade exposure across local labour markets.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001136:11:0", "occurrence_id": "3be0e14733bd870cf3676bf5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **4. The Broken Relationship between Imports and Labor Markets: A Bartik Analysis**\n\n\n  - **Overview of the Bartik Approach**\n\nCanonical trade models predict that increased trade increases factor income and alters the wage\nbill. However, recent empirical evidence suggests that labor markets are not as flexible as most\nmodels assume them to be, implying that adjustment to changes in trade volumes can be very\nslow (Autor, Dorn & Hanson, 2013; Dix-Carneiro & Kovak, 2014). Furthermore, since the effects\nof trade are typically heterogeneous across sectors, individuals and regions are differentially\nexposed to sector-specific shocks, giving rise to distributional concerns.\n\nIn the previous section we analyzed how trade agreements impact trade flows from Egypt to the\nrest of the world. Liberalization policies do seem to have a positive effect on exports but\nnegligible on imports. Yet, labor market outcomes have worsened. To estimate the relationship\nbetween labor markets and trade, we follow a common methodological framework to specific\nlocal conditions and employ a geography-based Bartik (1991) approach to quantify exogenous\nexport demand shocks from the US and EU countries for Egyptian imports.\n\nBartik models exploit the spatial variation in sectoral-composition assuming that changes in\nsector-specific foreign demand is different across regions. In other words, regions more exposed\nto trade for a particular sector and with major employment for the same sector will be more\naffected from a trade shock particular to the sector. Differences in trade exposure of regions to\nthis shock serves as an identification tool. For the trade exposure to be regionally identifiable,\nsome barriers or rigidities in the labor market should exist. Sluggish or segmented regional labor\nmarkets are a fundamental principal of the approach (Raymond et al, 2020).\n\nFigures 8 and 9 provide evidence of the concentrated trade exposure and segmented labor\nmarkets. Using the Enterprise Census (EC) from 2017 and the Labor Force Survey (LFS) from the\nsame year,\n\nMention: \"Labor Force Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Worker locations in the same year measure labour market segmentation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001148:24:0", "occurrence_id": "e7f02816f1503f4e8f5c3ca3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: raising interest rates and results in lower investment by crowding out private investment and\n\nhence lowering infrastructure development (Aschauer, 2000; Aizenman et al., 2007; Winter,\n\n2017).\n\n\nAmong other variables affecting physical infrastructure, urbanization seems to have a\n\npositive and significant effect on infrastructure growth. This implies that rising urbanization\n\nincreases infrastructure facilities, possibly due to increased demand for the latter (Démurger,\n\n2001 and Li, 2017; Maparu et al. 2021). As per the 2011 census, around 31% of the population\n\nlives in urban areas, and is expected to rise to more than 50% by 2050 (UN, 2014). The\n\ndemographic shift will require huge infrastructure expenditures to increase and upgrade the\n\nexisting stock of infrastructure facilities such as transport, telecom, water, and energy. On the\n\nother hand. Population density has a negative impact indicating the dominance of scale impact\n\nover the demand impact (Randolph et al.,1999). Financial development proxies - measured as\n\nbank credit to GSDP (CRE) are found to impact infrastructure development positively. The\n\ndevelopment of the financial sector facilitates long-term project financing with a reasonable\n\ncost which is vital for infrastructure development (Agrawal, 2000). Similarly, the social\n\ninfrastructure index (SINFRA) also positively affects physical infrastructure development\n\nsupporting the complementarity hypothesis among different types of infrastructure facilities\n\n(Cerra et al., 2017).\n\n\nWe also find the coefficient of the dummy variable positive and statistically significant.\n\nThis indicates that if the center and state governments belong to the same political party or in\n\na coalition government, it would have a positive impact on the state’s infrastructure\n\ndevelopment. State governments with the same political party or coalition in the central\n\ngovernment tend to get higher support in terms of financial resources and relevant policies\n\n(e.g., the central government’s support on national highways that pass through the state,\n\nbuilding electric power plants and transmission lines, central\n\nMention: \"2011 census\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The urban share of population reported here anchors the urbanisation projection.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001150:6:0", "occurrence_id": "db806eca418634260004f790", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: To better understand how our experimental sample compared to the average property in\n\n\nDar es Salaam, we matched our experimental data to a set of data comprising every parcel in\n\n\nTRA’s database for the city. Out of approximately 830,000 parcels in the city, nearly 30 percent\n\n\nare owned by a taxpayer in our experimental sample. Parcels included in our experiment were\n\n\nonly slightly (1.4 percentage points) less likely to have been issued a bill. Of those that were\n\n\nissued bills, only 35 percent of properties in the experimental sample had been valued by the\n\n\nTRA, as opposed to 74 percent in the rest of the city (those that were not valued were charged\n\n\na TSh 10,000 flat tax). Conditional on being billed and rated, the median value for properties\n\n\nincluded in the experiment was higher, approximately 31.2 million TSh (13,500 USD) versus 20\n\n\nmillion TSh (8,600 USD) for the rest of the city. Properties that were billed and rated faced,\n\n\non average, an annual tax rate of approximately 0.16% of the property’s rated value.\n\n\n**3.2** **Treatments**\n\n\nIn collaboration with the Tanzania Revenue Authority, we randomized each property owner\n\ninto one of five groups. <sup>3</sup> The treatments are summarized in Table 1. The control was not to\n\n\nreceive any message from the TRA. Group T1 received a simple message reminding them to pay\n\n\ntheir property tax, indicating the due date (June 30th) and providing information the taxpayers\n\n\ncould use to contact the TRA in case they had any questions. All other treatments included\n\n\nthis reminder message.\n\n\nTreatment 2 added a ‘reciprocity’ message, where taxpayers were reminded that taxes fund\n\n\nsocial services and infrastructure and finished with the TRA’s slogan “Together we build our\n\n\nNation.” Treatment 3 included the simple reminder as well as a ’social pressure’\n\nMention: \"TRA’s database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Every parcel in the city is compared with the experimental sample to assess representativeness.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001155:11:0", "occurrence_id": "b40a4d669a6aa90ad83df8f2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: As Latinobarómetro does not collect information on the general mood of the\nrespondent at the time of the interview, we control for the optimistic or pessimistic\ntrait of the respondent (proxied by her current satisfaction with life; see Annex E). <sup>9</sup> [^9: <mark>As demonstrated there, including the life satisfaction variable has no effect on the results, and</mark>\n<mark>the variable does not appear to be significant in any of the three models compared.</mark>]\n\nFor the majority of variables, missing values account for less than 5 percent\nof the sample. We test for patterns among those missing values and confirm\nthat missing observations across variables are, in effect, random and are not\nassociated with any of the sample’s demographic characteristics. The one\nexception is trust in the armed forces, for which around 15 percent of\nobservations are missing; as a result, we avoid adding that variable in our\nregressions.\n\n###### 3.1. Trust in Government and Citizen Participation in the Latin America and Caribbean Region\n\n\nTrust in the government is low across the LAC region (Table 1). Less than a\nthird (27.3 percent) of the population has confidence in the national\ngovernment. The few exceptions include El Salvador (72.4 percent), the\nDominican Republic (52.3 percent) and Uruguay (57.4 percent). <sup>10</sup> [^10: In Uruguay, high levels of trust in the government have prevailed consistently for almost\ntwo decades. El Salvador, instead, has seen abrupt surges and declines in trust over the same\nperiod, likely responding to the rise and fall of populist governments.] However,\nfor most countries in the region trust levels are below 40 percent, falling to\nas low as 10 percent in some cases (Latinobarómetro Corporation, 2021).\nAmong the worst performers, by the time the 2020 Latinobarómetro data\n\nMention: \"2020 Latinobarómetro data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Trust levels and respondents' mood are measured in this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001122:16:0", "occurrence_id": "7983e97b49bb485934f9d2e5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: of about 8 points. These results also suggest that in many contexts adding deep learning estimates to\nexisting predictions based on gradient boosting offers limited additional improvement in accuracy.\n\n###### b. Household or sub-area models are usually preferable to area-level models when sub-area data are available.\n\n\nThe pros and cons of different models for small area estimation in different contexts has been a subject\nof contention for many years and there is not yet a consensus between different statisticians and\npractitioners. For the purposes of this discussion, we assume that recent census data are unavailable,\nnecessitating the use of geospatial data. In most cases, geospatial data are available in the form of zonal\nstatistics at the “sub-area level”, where the sub-area is a geographic area such as a grid or village. A\nzonal statistic, for example, could be the average night-time luminosity in the village or grid. Linking\nsurvey and geospatial data at the level of the enumeration area or village is quite common in practice.\nThe Demographic and Health Surveys publicly release jittered geocoordinates for each EA in most cases,\nfacilitating this type of linking. Meanwhile, the target area is typically a more aggregate administrative\nunit, such as a district or subdistrict. Finally, we define the regional level as a level above the target area\nfor which the survey is considered to be representative.\n\n\nMany analysts use Bayesian modeling for small area estimation. We focus here, however, on empirical\nbest predictor models (Jiang and Lahiri, 2006, Molina and Rao, 2010), rather than purely Bayesian\nmodels, for three reasons. First, when they have been compared, EBP models give similar results to\nhierarchical Bayesian models (Guadarrama et al, 2016). Second, national statistics offices may be more\ncomfortable with empirical Bayesian than Bayesian methods, partly due to discomfort with assuming a\nprior distribution. Finally, multiple well-documented and user-friendly software packages employ E\n\nMention: \"geospatial data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Zonal statistics such as average night-time luminosity substitute for unavailable recent census data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001122:16:0", "occurrence_id": "90cf29d8c70e6f5edcacb69e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: of about 8 points. These results also suggest that in many contexts adding deep learning estimates to\nexisting predictions based on gradient boosting offers limited additional improvement in accuracy.\n\n###### b. Household or sub-area models are usually preferable to area-level models when sub-area data are available.\n\n\nThe pros and cons of different models for small area estimation in different contexts has been a subject\nof contention for many years and there is not yet a consensus between different statisticians and\npractitioners. For the purposes of this discussion, we assume that recent census data are unavailable,\nnecessitating the use of geospatial data. In most cases, geospatial data are available in the form of zonal\nstatistics at the “sub-area level”, where the sub-area is a geographic area such as a grid or village. A\nzonal statistic, for example, could be the average night-time luminosity in the village or grid. Linking\nsurvey and geospatial data at the level of the enumeration area or village is quite common in practice.\nThe Demographic and Health Surveys publicly release jittered geocoordinates for each EA in most cases,\nfacilitating this type of linking. Meanwhile, the target area is typically a more aggregate administrative\nunit, such as a district or subdistrict. Finally, we define the regional level as a level above the target area\nfor which the survey is considered to be representative.\n\n\nMany analysts use Bayesian modeling for small area estimation. We focus here, however, on empirical\nbest predictor models (Jiang and Lahiri, 2006, Molina and Rao, 2010), rather than purely Bayesian\nmodels, for three reasons. First, when they have been compared, EBP models give similar results to\nhierarchical Bayesian models (Guadarrama et al, 2016). Second, national statistics offices may be more\ncomfortable with empirical Bayesian than Bayesian methods, partly due to discomfort with assuming a\nprior distribution. Finally, multiple well-documented and user-friendly software packages employ E\n\nMention: \"Demographic and Health Surveys\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Jittered coordinates released for most enumeration areas allow survey data to be linked to spatial layers.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001132:1:0", "occurrence_id": "a229b88ed9f7eff2509c3cb3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Policy Research Working Paper 1394\n\n##### **Abstract**\n\nThis paper quantifies the impact of effective corporate tax\n\nrates on aggregate total factor productivity. Using Chilean manufacturing data, the paper documents a large\ndispersion in the effective tax rate faced by firms and a\nmass of firms facing a 0 percent tax rate. These empirical\npatterns are incorporated into a standard monopolistic\ncompetition model with corporate tax rates. The paper’s\nquantitative findings show that the TFP gains between the\n\n\n\neconomy implied by the Chilean tax code of 1998–2007\nand a hypothetical economy without effective corporate\ntax rate inefficiencies are between 4 and 11 percent. The\npaper considers counterfactual policies in which all firms\nface the same tax rate and finds a monotonically decreasing\nrelationship between the level of the tax rate and total factor\nproductivity.\n\n\n\nThis paper is a product of the Finance, Competitiveness and Innovation Global Practice. It is part of a larger effort by the\nWorld Bank to provide open access to its research and make a contribution to development policy discussions around the\n\nworld. Policy Research Working Papers are also posted on the Web at http://www.worldbank.org/prwp. The authors may\nbe contacted at fpatinopena@worldbank.org and mdiner@wharton.upenn.edu.\n\n\n_The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development_\n_issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the_\n_names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those_\n_of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and_\n_its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent._\n\n\nProduced by the Research Support\n\nMention: \"Chilean manufacturing data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Firm-level effective tax rates and productivity are documented from these plant records.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001158:1:0", "occurrence_id": "dc26d060d080d854231b4779", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Policy Research Working Paper 10621\n\n#### **Abstract**\n\nThis paper provides a detailed descriptive analysis of pat\nterns of urbanization across Sub-Saharan Africa for the\nyear circa 2015. Despite the rapidity and importance of\nSub-Saharan Africa’s urbanization, little is known about\nthe anatomy of patterns of urbanization across the region\ndue to a lack of detailed and accurate official data on urban\nsettlements and populations. To address this gap, the paper\napplies a modified version of the “dartboard” algorithm\nto high-resolution gridded population data for the region,\nwhich is derived from digitized maps of the footprints\nof all buildings in the region from very high-resolution\nsatellite imagery. This allows for a consistent definition of\nurban areas across all countries in the region, overcoming\n\n\n\nthe measurement problems that arise from relying on official definitions of urban areas, which vary markedly across\ncountries. Using this definition, the paper presents evidence\non key empirical regularities that are related to disparities\nacross the urban hierarchies, such as the extent of urban\nprimacy and Zipf’s law, as well as on the internal structures of cities, such as population density gradients and\nthe number of centers that cities possess. The paper also\nanalyzes how these characteristics are related to key country\ncharacteristics. Finally, the paper compares the results with\nthose that arise from the use of an alternative definition of\nurban areas—the degree of urbanization.\n\n\n\nThis paper is a product of the Urban, Disaster Risk Management, Resilience and Land Global Practice and the Poverty\n\nand Equity Global Practice.. It is part of a larger effort by the World Bank to provide open access to its research and make\na contribution to development policy discussions around the world. Policy Research Working Papers are also posted on\nthe Web at http://www.worldbank.org/prwp. The authors may be contacted at mroberts1@worldbank.org.\n\n\n_The Policy Research Working Paper Series disseminates the findings\n\nMention: \"high-resolution gridded population data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"A modified dartboard algorithm applied to these grids yields a consistent definition of urban areas.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001166:31:0", "occurrence_id": "809eda632366ec6cee982ba6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: were ultimately reviewed by the European Union. A 27% increase in IF take up implies\n\n\nthat roughly 25 jurisdictions joined the Inclusive Framework directly as a result of the\n\n\npeer review process.\n\n\nBack-of-the-envelope calculations suggest that this has not led to a large shift in the\n\n\ncomposition of the Inclusive Framework. For example, assuming no heterogeneity in the\n\n\nEU selection treatment effect, the average GDP per capita (in nominal 2018 dollars) for an\n\n\nIF member is only $370 lower than it would have been without the EU’s involvement. The\n\n\naverage Tax Justice Network Financial Secrecy Index value of IF members is currently\n\n\n291, versus 288 in the counterfactual. The EU review and listing process has made the\n\n\nIF more representative of poorer countries and those who potential contribute to global\n\n\nprofit shifting, but only slightly.\n\n\nThe EU made two choices in designing its review process that have arguably had a\n\n\nlarge impact on the IF’s composition. The first is in the setting of its thresholds for\n\n\nselection into the review process. The median GDP per capita, in 2018 nominal dollars,\n\n\nof jurisdictions within 10 percentile points of the cutoff is approximately $4,200. The\n\n\nsecond is excluding least developed countries (LDCs) from the review process. Taken\n\n\ntogether, a significant number of developing countries that might have been induced to\n\n\ntake part in the Inclusive Framework are left out of the current negotiations. One of the\n\n\nfactors that led the EU to exclude LDCs and set thresholds where it did was the fact that\n\n\nmany developing countries do not currently have the capacity to implement the BEPS\n\n\nminimum standards.\n\n\nHowever, while it is not clear that the EU review process would have led the same\n\n\npressure to join as it did for many tax havens, one potential benefit ceded by the design\n\n\nof the review process was greater involvement in the IF by developing countries, although\n\n\nthere is an ongoing debate as whether their participation in the IF has translated into\n\n\ngreater taxing right for developing countries (Christians and Van\n\nMention: \"Tax Justice Network Financial Secrecy Index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Average index values of framework members are compared with the counterfactual.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001169:32:0", "occurrence_id": "de30e5015aa87e05b778c8bf", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |&Business<br>Operation<br>Income|2.2 Net Profit of Own<br>Business|2% effective rate|74|PIT of Self-<br>Employed Business<br>Income =78.5|\n|---|---|---|---|---|\n|<br>3. Household<br>Property Income|3.1 Financial Investment<br>Profit|20%|88.6|<br>PIT of Interest,<br>Dividend, and<br>Bonus=114.2|\n|<br>3. Household<br>Property Income|3.2 Rent Income: House,<br>Land, Other Properties|Assume Full<br>Evasion|NA|<br>PIT of Leasing<br>Properties=8.6|\n\n\n\nNote: Admin. data is collected from China Tax Yearbook 2019.\n\n\n**Contributions to social security to the employee system**\n\n\nThe survey has information on whether individuals contribute to the employee’s system and the amount\nthey contribute. Respondents who do not declare as contributing to any of the systems, we assume that\nthose who work in government units, administrative institutions, and stated-own enterprises, as well as\nother employees who reported having a formal job contract, are formal employees contributing to the\nemployee system and we impute the amount contributed. Moreover, there is a significant share of\ncontributors to the employee system who did not report the amount and therefore needs to be imputed; we\nimpute the amounts for the formal employees among those. <sup>14</sup> The contributions are imputed using reported\nsalary and contribution schemes of the capital city corresponding to the individual’s province of residence.\nWe further impute the contributions from the employer side, while self-employed are assumed to pay all\ncontributions out-\n\nMention: \"China Tax Yearbook 2019\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Administrative tax totals used to calibrate the imputation of personal income tax are taken from this yearbook.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001171:44:0", "occurrence_id": "8cfbe9fb2aee5178567a1081", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _For labor or for divorce?_\n\n\ninstitutions of each country, UNDP and ESCWA open data portals. Unilateral divorce reforms were introduced in Egypt in 2000, in Jordan in 2001, and in\nMorocco in 2004. The red dotted lines represent the years that Egypt and Jordan introduced additional reforms to dispositions on child custody in their\nrespective Family Codes.\n\n\n**Figure A. 3. Evolution of the World Bank Index of Laws in Morocco, Egypt, and Jordan**\n\n\nOECD high-income, South Asia, Latin America, East Asia, Sub-Saharan Africa, and Europe and Central Asia regions. The latest year, 2023, is considered. The\nindex scores are based on the average of each economy’s scores for the 8 topics included in the aggregate score: mobility, workplace, pay, marriage,\nparenthood, entrepreneurship, assets, and pension. The index scores range between 0 and 100 and a higher score indicates more gender equal laws.\n\n43\n\nMention: \"World Bank Index of Laws\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Index scores averaged across comparator regions trace the evolution of family law.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001172:41:0", "occurrence_id": "3d5085b474aec661cf65e0ef", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: tionnaire from Falk et al. (2022), along with a question on how much do the subjects in\n\n\nour sample trust the Colombian government. Additionally, we inquired about the ado\n\nlescents’ social networks by asking about their number of friends of each nationality and\n\n\nwhether they have felt discriminated in school. While these outcomes were not the pri\n\nmary focus of the VenReps Kids survey, they collectively provide a general assessment of\n\n\nsocial cohesion, a crucial aspect in understanding the overall well-being of children and\n\n\nadolescents.\n\n\nWe measure the average differences in altruism between Venezuelan and Colombian ado\n\nlescents following the estimation of equation 1. Table 8 presents the results for altruism,\n\n\ntrust, discrimination and social ties in panels A, B and C respectively. As in the previous\n\n\ntables, the first column for each outcome reports the results of the estimates of equation\n\n\n1 without controls, and the second and third columns report the results of the estimation\n\n\nwith all controls and the propensity score matching respectively. As for altruism, we see\n\n\nthat Venezuelan adolescents are on average willing to donate 17% more of their imagi\n\nnary money endowment to a good cause relative to Colombian adolescents. As for trust,\n\n\nthe results are mixed when analyzing the trust items separately. On average, Venezuelan\n\n\nadolescents tend to be more benevolent towards an injustice, are more likely to think that\n\n\npeople have good intentions and have higher trust in the Colombian government. How\n\never, they are less willing to return a favor (although not significantly more) and trust\n\n\nless in Colombian unknown people. Moreover, Venezuelan adolescents have felt more\n\n\ndiscriminated in school and tend to have less friends. The findings suggest that forced\n\n\nmigration may have negative impacts on the accumulation of social capital due to barri\n\ners in social norms that hinder social integration. Adolescent migrants face challenges in\n\n\nbuilding connections with the host community given their weakened social ties and the\n\n\npresence of exclusionary attitudes.\n\n\nAll in all, our analysis suggests that there are meaningful\n\nMention: \"VenReps Kids survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Adolescents' networks, discrimination and altruism measures are collected in this child survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001175:3:1", "occurrence_id": "ce28411da4e03e3df41e6cc6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  activity, owing to Nepal being a highly rural economy. And while the\neconomy is shifting away from agriculture, and jobs created over the last decade have been a mix of low- and\nhigh-productivity in industry and service sectors, men have benefitted disproportionately. Of the 3.8 million\nnew wage jobs added to the economy by 2018, men took up over two-thirds of those jobs (Ruppert Bulmer,\n2020), including jobs in manufacturing, finance, and business services sectors. These outcomes cannot be\nexplained by Nepali women’s education levels – progress in educational attainment of girls is sizable and\nwomen enjoy high returns to education (Ruppert Bulmer et al., 2020). <sup>4</sup> [^4: Over the last three decades, women achieved strong education gains. Average years of schooling rose from 2 years for\nfemales born in 1960 to over 9 years for those born after 1990. A woman with a college degree, on average, earns double\nwhat a woman with incomplete primary earns, controlling for sector of work and other factors (Ruppert Bulmer et al. 2020).]\n\n\nYoung women, particularly married ones, have low attachment to the labor market. Nearly half (47 percent)\nof them are not in employment, education, or training (NEET), vs. 22 percent of young men. <sup>5</sup> [^5: Nepal Labor Force Survey 2017/2018.] Over a third\nof young married women aged 20 and 24 report to not have done any work other than housework in the 12\nmonths before the 2022 Nepal Demographics and Health Survey (DHS), as compared to three percent of\nyoung married men within the same age group. Social norms related to the unequal gender roles in families may\ncompel women to shoulder a disproportionate burden of household work, thereby reducing their availability\nfor wage employment and restricting them to jobs with lower employment potential than men. Women are\nmore likely than men to be in unpaid farm work\n\nMention: \"Nepal Labor Force Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The share of young people not in employment, education or training is reported from this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001175:3:1", "occurrence_id": "b5af68a3cee1b92c72585527", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  activity, owing to Nepal being a highly rural economy. And while the\neconomy is shifting away from agriculture, and jobs created over the last decade have been a mix of low- and\nhigh-productivity in industry and service sectors, men have benefitted disproportionately. Of the 3.8 million\nnew wage jobs added to the economy by 2018, men took up over two-thirds of those jobs (Ruppert Bulmer,\n2020), including jobs in manufacturing, finance, and business services sectors. These outcomes cannot be\nexplained by Nepali women’s education levels – progress in educational attainment of girls is sizable and\nwomen enjoy high returns to education (Ruppert Bulmer et al., 2020). <sup>4</sup> [^4: Over the last three decades, women achieved strong education gains. Average years of schooling rose from 2 years for\nfemales born in 1960 to over 9 years for those born after 1990. A woman with a college degree, on average, earns double\nwhat a woman with incomplete primary earns, controlling for sector of work and other factors (Ruppert Bulmer et al. 2020).]\n\n\nYoung women, particularly married ones, have low attachment to the labor market. Nearly half (47 percent)\nof them are not in employment, education, or training (NEET), vs. 22 percent of young men. <sup>5</sup> [^5: Nepal Labor Force Survey 2017/2018.] Over a third\nof young married women aged 20 and 24 report to not have done any work other than housework in the 12\nmonths before the 2022 Nepal Demographics and Health Survey (DHS), as compared to three percent of\nyoung married men within the same age group. Social norms related to the unequal gender roles in families may\ncompel women to shoulder a disproportionate burden of household work, thereby reducing their availability\nfor wage employment and restricting them to jobs with lower employment potential than men. Women are\nmore likely than men to be in unpaid farm work\n\nMention: \"2022 Nepal Demographics and Health Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Unpaid housework reported by young married women and men is measured in this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001177:17:0", "occurrence_id": "94b0f88f1b5498e457a7cd8d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: social protection for workers by ensuring access to social security for the unemployed through\nvarious funds and solidarity mechanisms.\n\n\nA notable finding from this review of programs is the lack of national measures addressing barriers\nto accessing care and redistributing care work within households, including the promotion of the\ncare economy, which disproportionately falls on women with lower levels of education. Existing\nprograms, managed by the Colombian Institute of Family Welfare (ICBF), primarily focus on\nchildcare and time release, without providing options for elderly care or complementary measures\nthat facilitate the effective integration of caregivers into the care and labor sectors.\n\n\n**8.** **Conclusions**\n\n\nThe gender gap in labor markets presents a significant challenge, particularly impacting vulnerable\nwomen, and Colombia is no exception, with the pandemic exacerbating this issue. This study\nexamines recent trends in Colombian labor outcomes within the context of COVID-19. Our focus\nis primarily on gender gaps in the labor market and the factors influencing them, particularly in\nsectors where women are more vulnerable, such as home care.\n\n\nUtilizing data from the GEIH, decomposition methodologies, and probabilistic models, we find\nthat the economic recovery observed in 2021 did not close the gender gaps, indicating underlying\nstructural factors. Moreover, the pre-pandemic gender gaps in terms of participation and\nunemployment were intensified by the crisis. Interestingly, gender gaps in labor income and hours\nworked decreased during the crisis, but this was primarily driven by a significant decline in men's\nlevels rather than improved conditions for women. Another contributing factor to women's slow\nrecovery in the labor market is the moderate recovery of the informal sector in 2021, in which\nwomen have a significant presence. However, there are indications of a more favorable labor\nmarket recovery for women in 2022.\n\n\nThe analysis reveals that several factors disproportionately affect women and impact their\nprobability of participating in the labor market. Having children in the household, low levels of\nschooling, and household income are among the key determinants\n\nMention: \"GEIH\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Decompositions and probabilistic models of gender gaps in labour outcomes are estimated on this household survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001179:23:0", "occurrence_id": "fe6531785e0da2b5e09df7f7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Spatial distribution_\n\nTo obtain a more granular geographical understanding, we present bi-maps for the region in this\nsection. These maps combine the vulnerability individual incidence map with a map of a specific\nindex related to a hazard. The graphs are just an illustration using data from some, not all, of the\nhazards considered in the model (we present the flood and landslides TH index, SPEI 6, and for\nMexico the cyclones/hurricanes). As previously mentioned, the vulnerability measure has\nrepresentativeness at the state level, and the corresponding granular geographical map of vulnerability\ncorresponds to provinces (Figure B2). However, the specific index can be mapped in a lower level\nthan the province level. The aim is to leverage provincial-level vulnerability data into lower levels to\nidentify districts with elevated indexes of hazards, facilitating the target of policy initiatives. For\ninstance, in the case of a state with a high percentage of individuals vulnerable, the available data does\nnot provide information about the specific location of these persons within the state. In such a scenario,\nit may be prudent to initiate policies in localities that could be more exposed and affected by natural\ndisasters or hazards due to a high hazard index. This will help to reduce the impact of disasters on\nthese areas.\n\nIn Figure 11, we present the vulnerability individual incidence at the state level and the TH flood index\nat the district level for the countries analyzed. We can observe that at the center west of Andean\ncountries as well as in Central America, there are some districts in dark blue or purple color, indicating\na high vulnerability incidence at the state level and high values of the TH flood index at the district\nlevel, suggesting that these localities are the ones with more significant potential damage from floods\nin the region.\n\n\n**Figure 11.** Bi-map considering vulnerability to poverty individual incidence and TH flood index in\nthe region.\n\n\n22\n\nMention: \"SPEI 6\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"The six-month drought index is mapped against individual vulnerability to produce bi-maps.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001179:23:0", "occurrence_id": "7649c0e85b6d1820b10965d5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Spatial distribution_\n\nTo obtain a more granular geographical understanding, we present bi-maps for the region in this\nsection. These maps combine the vulnerability individual incidence map with a map of a specific\nindex related to a hazard. The graphs are just an illustration using data from some, not all, of the\nhazards considered in the model (we present the flood and landslides TH index, SPEI 6, and for\nMexico the cyclones/hurricanes). As previously mentioned, the vulnerability measure has\nrepresentativeness at the state level, and the corresponding granular geographical map of vulnerability\ncorresponds to provinces (Figure B2). However, the specific index can be mapped in a lower level\nthan the province level. The aim is to leverage provincial-level vulnerability data into lower levels to\nidentify districts with elevated indexes of hazards, facilitating the target of policy initiatives. For\ninstance, in the case of a state with a high percentage of individuals vulnerable, the available data does\nnot provide information about the specific location of these persons within the state. In such a scenario,\nit may be prudent to initiate policies in localities that could be more exposed and affected by natural\ndisasters or hazards due to a high hazard index. This will help to reduce the impact of disasters on\nthese areas.\n\nIn Figure 11, we present the vulnerability individual incidence at the state level and the TH flood index\nat the district level for the countries analyzed. We can observe that at the center west of Andean\ncountries as well as in Central America, there are some districts in dark blue or purple color, indicating\na high vulnerability incidence at the state level and high values of the TH flood index at the district\nlevel, suggesting that these localities are the ones with more significant potential damage from floods\nin the region.\n\n\n**Figure 11.** Bi-map considering vulnerability to poverty individual incidence and TH flood index in\nthe region.\n\n\n22\n\nMention: \"provincial-level vulnerability data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Vulnerability measured at province level is down-scaled to districts with elevated hazard indexes.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001180:6:0", "occurrence_id": "ab5ec6469f88f558b26d63bd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The second set of findings focuses on the magnitude and persistence of the parenthood penalty\n\n\nacross various subgroups in Indonesia. These results offer important insights for policymakers\n\n\nseeking to target interventions effectively. First, we found that households with an extended fam\n\nily member living in the household experienced the fastest recovery from the birth penalty. For\n\n\nthese households, the penalty became insignificant by the second year after childbirth. In con\n\ntrast, for households without extended family members, the penalty persisted until year seven.\n\n\nSecond, we observe significant differences in the magnitude and persistence of the birth penalty\n\n\nbased on education level. Highly educated individuals faced larger penalties (approximately 30%\n\n\nin the first year) and for a longer period compared to their less educated counterparts (around\n\n\n20% in the first year). Furthermore, we find that rural women experienced the highest birth\n\n\npenalty in year three, after which the trend began to reverse. In contrast, urban women saw\n\n\nthis reversal begin in year four. These results suggest that expanding access to childcare could\n\n\nsignificantly reduce the persistence of the birth penalty, particularly for highly educated urban\n\n\nwomen, who experienced the largest and most persistent penalties. This subgroup represents\n\n\nan important target for policy interventions, where childcare provision could have the greatest\n\n\nimpact.\n\n\nThe rest of the paper is structured as follows. Section 2 provides an overview of the Indonesia\n\n\nFamily Life Survey (IFLS) data and the institutional context of the Indonesian labor market.\n\n\nSection 3 outlines our empirical strategy, including the DID methodology and the identification\n\n\nassumptions. Section 4 presents our main findings, including a detailed heterogeneity analysis\n\n\nand robustness checks. Finally, Section 5 discusses the policy implications of our findings and\n\n\nconcludes with suggestions for future research.\n\n##### **2 Institutional Context, Data, and Descriptive Statistics**\n\n\n**2.1** **The Indonesia Family Life Survey**\n\n\nThe Indonesian Family Life Survey (IFLS) is a longitudinal multitopic household survey that is\n\n\nrepresentative of 83% of\n\nMention: \"Indonesia\n\n\nFamily Life Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its household panel provides the data underlying the difference-in-differences design on parenthood penalties.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001182:10:0", "occurrence_id": "07ecc28dc8bed3ce6ab6abe8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: by the Earth Observation Group at the Colorado School of Mines. <sup>5</sup> [^5: https://developers.google.com/earth-engine/datasets/catalog/NOAA_VIIRS_DNB_MONTHLY_V1_VCMSLCFG] To measure pollution, we\n\nuse Nitrogen Dioxide (NO2) concentrations measured by Sentinel 5-P satellites. <sup>6</sup> [^6: “Sentinel-5P NRTI NO2 : Near Real-Time Nitrogen Dioxide”, Google Earth Engine.] We expect\n\nthat NO2 levels increase in response to human activity, namely burning fossil fuels and biomass,\n\nwhich may have fallen during the crisis. EVI was taken from the Landsat 8 Collection 1 Tier\n\n1 composite. <sup>7</sup> [^7: https://developers.google.com/earth-engine/datasets/catalog/LANDSAT_LC08_C01_T1_8DAY_EVI]\n\n\nThe Google search index at subnational levels reflects the relative popularity of any given\n\nsearch term. The total search volume for each term is divided by total Google search volume\n\nfor that week and subnational region. The search index for each term is rescaled such that the\n\nweek-subnational region observation with the highest relative search score, for all weeks in\n\nthat country, is assigned a value of 100 with all other weeks expressed relative to this peak.\n\nThe baseline (weeks with a value of 100) itself is a mechanism for rescaling search term data\n\nsuch that values are comparable across regions and over time. <sup>8</sup> [^8: To be more precise, the index for a particular month and term is constructed as following:] We aggregate data into monthly\n\naverages, taking the average of the weekly data points for each subnational district.\n\n\nGoogle restricts the quantity of data that can be downloaded in one batch. Therefore, the\n\ndatasets are extracted in multiple rounds to cover all subnational districts for all countries with\n\nHFPS data. This process requires indexing all downloads against a common administrative\n\ndistrict which serves as a reference point. The reference district is included in all rounds of\n\ndownloads\n\nMention: \"Sentinel-5P NRTI NO2\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Near real-time nitrogen dioxide concentrations measure pollution response to reduced activity.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001182:10:0", "occurrence_id": "793d3d6b19c507a56797d663", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: by the Earth Observation Group at the Colorado School of Mines. <sup>5</sup> [^5: https://developers.google.com/earth-engine/datasets/catalog/NOAA_VIIRS_DNB_MONTHLY_V1_VCMSLCFG] To measure pollution, we\n\nuse Nitrogen Dioxide (NO2) concentrations measured by Sentinel 5-P satellites. <sup>6</sup> [^6: “Sentinel-5P NRTI NO2 : Near Real-Time Nitrogen Dioxide”, Google Earth Engine.] We expect\n\nthat NO2 levels increase in response to human activity, namely burning fossil fuels and biomass,\n\nwhich may have fallen during the crisis. EVI was taken from the Landsat 8 Collection 1 Tier\n\n1 composite. <sup>7</sup> [^7: https://developers.google.com/earth-engine/datasets/catalog/LANDSAT_LC08_C01_T1_8DAY_EVI]\n\n\nThe Google search index at subnational levels reflects the relative popularity of any given\n\nsearch term. The total search volume for each term is divided by total Google search volume\n\nfor that week and subnational region. The search index for each term is rescaled such that the\n\nweek-subnational region observation with the highest relative search score, for all weeks in\n\nthat country, is assigned a value of 100 with all other weeks expressed relative to this peak.\n\nThe baseline (weeks with a value of 100) itself is a mechanism for rescaling search term data\n\nsuch that values are comparable across regions and over time. <sup>8</sup> [^8: To be more precise, the index for a particular month and term is constructed as following:] We aggregate data into monthly\n\naverages, taking the average of the weekly data points for each subnational district.\n\n\nGoogle restricts the quantity of data that can be downloaded in one batch. Therefore, the\n\ndatasets are extracted in multiple rounds to cover all subnational districts for all countries with\n\nHFPS data. This process requires indexing all downloads against a common administrative\n\ndistrict which serves as a reference point. The reference district is included in all rounds of\n\ndownloads\n\nMention: \"Landsat 8 Collection 1 Tier\n\n1 composite\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The enhanced vegetation index is taken from this eight-day composite product.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001182:10:0", "occurrence_id": "4cde5ac34134d5daca1ee59f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: by the Earth Observation Group at the Colorado School of Mines. <sup>5</sup> [^5: https://developers.google.com/earth-engine/datasets/catalog/NOAA_VIIRS_DNB_MONTHLY_V1_VCMSLCFG] To measure pollution, we\n\nuse Nitrogen Dioxide (NO2) concentrations measured by Sentinel 5-P satellites. <sup>6</sup> [^6: “Sentinel-5P NRTI NO2 : Near Real-Time Nitrogen Dioxide”, Google Earth Engine.] We expect\n\nthat NO2 levels increase in response to human activity, namely burning fossil fuels and biomass,\n\nwhich may have fallen during the crisis. EVI was taken from the Landsat 8 Collection 1 Tier\n\n1 composite. <sup>7</sup> [^7: https://developers.google.com/earth-engine/datasets/catalog/LANDSAT_LC08_C01_T1_8DAY_EVI]\n\n\nThe Google search index at subnational levels reflects the relative popularity of any given\n\nsearch term. The total search volume for each term is divided by total Google search volume\n\nfor that week and subnational region. The search index for each term is rescaled such that the\n\nweek-subnational region observation with the highest relative search score, for all weeks in\n\nthat country, is assigned a value of 100 with all other weeks expressed relative to this peak.\n\nThe baseline (weeks with a value of 100) itself is a mechanism for rescaling search term data\n\nsuch that values are comparable across regions and over time. <sup>8</sup> [^8: To be more precise, the index for a particular month and term is constructed as following:] We aggregate data into monthly\n\naverages, taking the average of the weekly data points for each subnational district.\n\n\nGoogle restricts the quantity of data that can be downloaded in one batch. Therefore, the\n\ndatasets are extracted in multiple rounds to cover all subnational districts for all countries with\n\nHFPS data. This process requires indexing all downloads against a common administrative\n\ndistrict which serves as a reference point. The reference district is included in all rounds of\n\ndownloads\n\nMention: \"HFPS data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"The countries covered by this survey determine which subnational districts were downloaded.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001185:24:0", "occurrence_id": "1917605a10ed2946947df45e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ##### **Appendix A Additional Figures**\n\n**Figure A.1:** Percent Change in Speed based on Google Data and Uber Data\n\n\n**Figure A.2:** Speed versus Congestion Hourly, Before and After the Curfew Policy in 2020\n\n\n23\n\nMention: \"Google Data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Travel speed changes are computed from this navigation data alongside ride-hailing records.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001185:24:0", "occurrence_id": "231feec0b6b50812cbf6265e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ##### **Appendix A Additional Figures**\n\n**Figure A.1:** Percent Change in Speed based on Google Data and Uber Data\n\n\n**Figure A.2:** Speed versus Congestion Hourly, Before and After the Curfew Policy in 2020\n\n\n23\n\nMention: \"Uber Data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Trip speeds from this operator are used with the other source to assess the curfew's congestion effects.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001186:32:1", "occurrence_id": "729080578377467fdc3fac88", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  (Total population aged\n15 and more). Note: Employment data from ILOSTAT are used in all cases, except for those Latin American countries were data from ILOSTAT\nwere not available (Colombia, Costa Rica, and Nicaragua) or the time series was longer on SEDLAC (Argentina, Chile, Panama, and Peru). Data\nfor these 7 countries come from SEDLAC.\n\n\n21 See World Bank (2020) <u>[https://pubdocs.worldbank.org/en/406191621616642876/Indotask-TR-English.pdf.](https://pubdocs.worldbank.org/en/406191621616642876/Indotask-TR-English.pdf)</u>\n\n\n31\n\nMention: \"Employment data from ILOSTAT\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Employment relative to population aged 15 and over is taken from this labour statistics database in all but seven countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001186:32:1", "occurrence_id": "9cca5681a504e99a93f7cbea", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  (Total population aged\n15 and more). Note: Employment data from ILOSTAT are used in all cases, except for those Latin American countries were data from ILOSTAT\nwere not available (Colombia, Costa Rica, and Nicaragua) or the time series was longer on SEDLAC (Argentina, Chile, Panama, and Peru). Data\nfor these 7 countries come from SEDLAC.\n\n\n21 See World Bank (2020) <u>[https://pubdocs.worldbank.org/en/406191621616642876/Indotask-TR-English.pdf.](https://pubdocs.worldbank.org/en/406191621616642876/Indotask-TR-English.pdf)</u>\n\n\n31\n\nMention: \"SEDLAC\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It supplies longer series for the countries where the primary source is unavailable or shorter.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001188:25:0", "occurrence_id": "ec0e592efdd679308765a30a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: #### **4 Could an Increase in Green Trade Affect Workers?**\n\nThere are multiple mechanisms for why green goods trade could affect labor markets differ\n\nently. Sector-specific skills needs for the production process involving green goods are not\n\n\nevenly distributed across genders and skill groups. Green industries often require specialized\n\n\nskills and knowledge related to sustainability and environmental technologies. The availabil\n\nity of these skills and the preferences of female workers towards these sectors can influence\n\n\nthe share of female workers within green industries. If women possess the necessary skills\n\n\nand are more inclined towards environmentally focused sectors, the share of female workers\n\n\nin these industries may be relatively higher. For example, a study by International Renew\n\nable Energy Agency shows that renewable energy employs about 32% women, compared to\n\n\n22% in the energy sector overall (IRENA, 2019). <sup>2</sup> Moreover, the skill level required in the\n\n\nproduction of green goods could be higher. The 2022 LinkedIn Global Green Skill Report\n\n\n(LinkedInEconomicGraph, 2022) found that in 2021 there were only 62 women for every 100\n\n\nmen considered green talent - a number that has been stagnant since 2015. <sup>3</sup>\n\n\nThe sorting pattern of workers into industries with a higher share of green goods exports can\n\n\nbe influenced by social norms and stereotypes that shape career choices and the represen\n\ntation of different genders and skill levels. For instance, the perception of green industries\n\n\nas environmentally conscious and socially responsible may attract a specific pool of workers,\n\n\nsuch as skilled college graduates (OCDE, 2014). <sup>4</sup>\n\n\nThe expansion of green goods exports has the potential to increase the demand for labor\n\n\nacross all types of industries, and it can be particularly advantageous for previously under\n\nrepresented groups. As the demand for labor in the green goods sector grows, the likelihood\n\n\nof employment opportunities for these underrepresented groups also increases like\n\nMention: \"2022 LinkedIn Global Green Skill Report\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its tally of 62 women per 100 men in green talent is cited as evidence of gender gaps.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001190:18:0", "occurrence_id": "72a5b182992de6978be64071", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: interaction term between bank credit and a proxy measure of the level of competition in the\n\n\nbanking industry which equals the assets of the five largest banks in the country-year as a share of\n\n\ntotal commercial banking assets ( _5-bank asset concentration_ ). For robustness, we also report\n\n\nresults using a similar index for the three largest banks (3 _-bank asset concentration_ ). Higher values\n\n\nof the concentration indices imply less competition in the banking industry. The data source is\n\n\nGFDD, World Bank.\n\n\nIn all the heterogeneity estimations, we include as controls interaction terms between bank\n\n\ncredit and the following variables (defined above): log of GDP per capita, log of number of\n\n\nworkers at the firm (firm size), log of age of the firm., and labor productivity of the firm (logs).\n\n\nThese controls guard against the possibility that our main interaction term (as stated in Hypotheses\n\n\n1-3) is spuriously picking up the differential effect of bank credit in countries with high vs. low\n\n\nGDP per capita, and firms of different size, age, and labor productivity.\n\n\n**4. Base regression results**\n\n\nBase regression results are provided in table 1. Panel A in table 1 contains the estimated log odds\n\n\nratios while Panel B contains the estimated marginal effects of the bank credit to GDP ratio. <sup>4</sup> [^4: Unless stated otherwise, all marginal effects throughout the paper are evaluated at the mean value of the various\nexplanatory variables.] Table\n\n\n1 shows that regardless of the controls, the relationship between the bank credit to GDP ratio and\n\n\nthe likelihood (log odds ratios and marginal effect) of a having a woman top manager is positive\n\n\nand statistically significant at the 1 percent level. Without any controls (except for country, year,\n\n\nand industry fixed effects), the odds of having a woman top manager increase by 2.0 log points\n\n\ndue a unit increase in the bank credit to GDP ratio (column 1, Panel A). The corresponding increase\n\n\nwith various\n\nMention: \"GFDD\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Bank asset concentration indices measuring competition are constructed from this financial structure database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001195:30:0", "occurrence_id": "d24f9127b981b890a7d2ce4c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Table 7: Chichewa Scores vs School Exposure\n\n\n(1) (2) (3)\nModel 1 Model 2 Model 3\n<u>b/se</u> <u>b/se</u> <u>b/se</u>\nCovid -56.436*** -44.603** -22.837\n(3.96) (18.93) (15.02)\nTime 11.774*** 11.876***\n(0.63) (0.66)\nTime & Covid -1.310\n\n(2.11)\nTime 11.876***\n(Excluding Closure) (0.66)\nTime & Covid -1.310\n<u>(Excluding Closure)</u> <u>(2.11)</u>\nObservations 4436.000 4436.000 4436.000\n<u>R-sqr</u> <u>0.080</u> <u>0.080</u> <u>0.080</u>\n\n\nData Source: MLSS 2016/18/19/20/21.\n\n\nNote: Standard Errors are in parenthesis. See Tables 2 and 5 for details of the sample and models.\n\n\n- p<0.10, ** p<0.05, *** p<0.01\n\n\n29\n\nMention: \"MLSS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Repeated rounds provide the test scores used to estimate closure effects.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001196:11:1", "occurrence_id": "6ef8db7878723f309ac212da", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  be adversely affected by events (IPCC, 2014). Following this definition, we include\npopulation density (people per sq. km of the district or state area), percentage of cropland,\nand urbanization of the district or state area. The data source of type of land is from the\nMODIS 500m Landcover Dataset NASA LP DAAC at the USGS EROS Center, retrieved\nfrom Google Earth Engine Data Catalog based on the 2018 Images LC type 1 with 17 classes\nof land use. We employ class 12 for the cropland area (where at least 60% of the area is\ncultivated), class 13 for the urban area (at least 30% is impervious surface area, including\nbuilding materials, asphalt, and vehicles), and the sum of all the classes for the total area of\nthe administrative units at level 2.\n\nTable 3 summarizes the average and standard deviation of the main variables of hazards and\nexposure at administrative level used in the analysis. The results indicate that Central America\nand Mexico face an elevated potential damage from cyclones and a higher frequency of\noccurrences of this hazard. In South America, floods are more prevalent, but countries like\nPeru and Ecuador evidence on average a higher potential damage from landslides than floods.\n\n\n10\n\nMention: \"MODIS 500m Landcover Dataset\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Classes 12 and 13 define cropland and urban shares used as vulnerability covariates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001196:11:1", "occurrence_id": "f3ee0981d053486ebbe1e455", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  be adversely affected by events (IPCC, 2014). Following this definition, we include\npopulation density (people per sq. km of the district or state area), percentage of cropland,\nand urbanization of the district or state area. The data source of type of land is from the\nMODIS 500m Landcover Dataset NASA LP DAAC at the USGS EROS Center, retrieved\nfrom Google Earth Engine Data Catalog based on the 2018 Images LC type 1 with 17 classes\nof land use. We employ class 12 for the cropland area (where at least 60% of the area is\ncultivated), class 13 for the urban area (at least 30% is impervious surface area, including\nbuilding materials, asphalt, and vehicles), and the sum of all the classes for the total area of\nthe administrative units at level 2.\n\nTable 3 summarizes the average and standard deviation of the main variables of hazards and\nexposure at administrative level used in the analysis. The results indicate that Central America\nand Mexico face an elevated potential damage from cyclones and a higher frequency of\noccurrences of this hazard. In South America, floods are more prevalent, but countries like\nPeru and Ecuador evidence on average a higher potential damage from landslides than floods.\n\n\n10\n\nMention: \"Google Earth Engine Data Catalog\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The imagery product was retrieved through this catalogue rather than directly from the provider.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001202:16:1", "occurrence_id": "b328ddde6003b6c6de9cec7e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Digital Gender Divide Include Upskill, Innovate”.\nOsundina, O.A. (2020) “Sustainable development: Does improvement in education and health\n\nof women improve female labour force participation rate?” Sustainable Development.\n2020; 28:13–24. DOI: 10.1002/sd.1961\nPatricia, C. and Cortés, J.P. (2020). “Children and The Remaining Gender Gaps in The Labor\n\nMarket”. NBER Working Paper 27980\nRamadan, R. (2022) “Drivers of Income Inequality during COVID-19 Pandemic in Jordan”.\n\nEconomic Research Forum Working Paper No. 1546.\nThe Economist Intelligence Unit (2021a) “The Inclusive Internet Index 2021- Methodology\n\nreport”. Sponsored by Facebook\nThe Economist Intelligence Unit (2021b) “The Inclusive Internet Index 2021- Bridging Digital\n\nDivides”. Internet.org by Facebook\nValberg, S. (2020) “ICT, Gender, and the Labor Market: A Cross-Country Analysis” in\n\n“Digitalisation and Development. Issues for India and Beyond”. Edited by Dibyendu\nMaiti, Fulvio Castellacci and Arne Melchior. Springer.\nViollaz, M. and Winkler, H. (2020) “Does the Internet Reduce Gender Gaps? The Case of\n\nJordan”. Social Protection and Jobs Global Practice & Middle East and North Africa\nRegion. Policy Research Working Paper 9183\nWodon, Q. T., de la Brière, B. (2018). “The Cost of Gender Inequality Unrealized Potential:\n\nThe High Cost of Gender Inequality in Earnings”. World Bank Group\nWorld Development Report (2016) “Digital Dividends”. World Bank Group\nWorld Development Indicators (2023): <u>[https://databank.worldbank.org/source/world-](https://databank.worldbank.org/source/world-\n\nMention: \"World Bank Group\nWorld Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"Listed with its data bank link among the sources and references of the study.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001203:6:1", "occurrence_id": "4632e2fe8ec7102ce95b020e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \nnot have a council permit. The WBISES captures informal businesses, while the WBES and\n\nWBMES survey formal firms. <sup>7</sup>\n\n\nThe WBES is a nationally representative survey where a stratified random sampling approach is\n\napplied to formally registered firms that have five or more employees, have at least 1 percent of\n\nprivate ownership, and do not have legal status as cooperatives. The sample size in Zambia is 601\n\nfirms. Formal micro-enterprises, which are defined as employing fewer than five workers,\n\n\n7 See Appendix Table A.1 for details on the registration status by survey types.\n\n5\n\nMention: \"WBISES\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"It captures the informal businesses that the formal-firm surveys omit.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001203:6:1", "occurrence_id": "9b68417321a0fcbc1b6b59be", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \nnot have a council permit. The WBISES captures informal businesses, while the WBES and\n\nWBMES survey formal firms. <sup>7</sup>\n\n\nThe WBES is a nationally representative survey where a stratified random sampling approach is\n\napplied to formally registered firms that have five or more employees, have at least 1 percent of\n\nprivate ownership, and do not have legal status as cooperatives. The sample size in Zambia is 601\n\nfirms. Formal micro-enterprises, which are defined as employing fewer than five workers,\n\n\n7 See Appendix Table A.1 for details on the registration status by survey types.\n\n5\n\nMention: \"WBES\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Nationally representative stratified sampling of formally registered firms with five or more employees.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001203:6:1", "occurrence_id": "ad3affbdd1e6e3170b0bc6c7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \nnot have a council permit. The WBISES captures informal businesses, while the WBES and\n\nWBMES survey formal firms. <sup>7</sup>\n\n\nThe WBES is a nationally representative survey where a stratified random sampling approach is\n\napplied to formally registered firms that have five or more employees, have at least 1 percent of\n\nprivate ownership, and do not have legal status as cooperatives. The sample size in Zambia is 601\n\nfirms. Formal micro-enterprises, which are defined as employing fewer than five workers,\n\n\n7 See Appendix Table A.1 for details on the registration status by survey types.\n\n5\n\nMention: \"WBMES\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It covers a further set of formal firms alongside the main enterprise survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001207:9:1", "occurrence_id": "4186fe37dfe7d5baa61cffe0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  inflation experienced by the top and bottom deciles between November 2018\nand May 2022 is shown in Figure 4. The figure suggests that poor families experience higher\ninflation overall.\n\n**Figure 4: Inflation Inequality across Income Deciles**\n\n18\n\n16\n\n14\n\n12\n\n10\n\n8\n\n6\n\n4\n\n2\n\n0\n\n\nDecile 1 (Poorest) Decile 10 (Richest) YoY_Inflation\n\n\nSource: Authors’ calculations from PBS data.\n\nThe poorest households (decile 1), on average, experienced one percentage point higher inflation\nrates than the richest households (decile 10) over this period. The maximum difference of 3.50\npercentage points between these two deciles was observed in July 2020. The highest levels of\ninflation inequality were observed between April 2020 and January 2021, when inflation was on\nthe decline due to COVID-19. This gap can be explained by the fact that during this period, the\naverage energy inflation was -2.4 percent and 0.5 percent for urban and rural regions,\nrespectively (Figure 3). <sup>4</sup> The food inflation, however, was in double digits during the same period\n(Figure 2). Since food has the highest budget share for the poorest of the households, while\nutilities and transport are used the most by the top decile, the gap between these extreme\n\n\n4 This was due to reduced oil and utility consumption resulting from the fall in transportation and industrial activities\ncaused by COVID lockdowns.\n\n\n8\n\nMention: \"PBS data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Inflation experienced by income deciles is computed from this household survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001211:9:0", "occurrence_id": "1a82089bf0a66ea898240d96", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Data merging. A key aspect of our research methodology is the merging of the Orbis and\n\n\nBERKEARTH datasets using latitude and longitude information. This common element allows\n\n\nus to associate firm-level financial performance data from Orbis with the corresponding weather data\n\n\nfrom BERKEARTH, both mapped to the same geographic region.\n\n\nGiven that the weather data is gridded with one degree of latitude and longitude, we first assign a\n\n\nunique index to each grid. For example, Index 1 is assigned to the upper leftmost grid (longitude −180\n\n\nto −179 and latitude −90 to −89). Subsequently, we assign the same index to the location of each firm.\n\n\nFor instance, a firm located at longitude −179 _._ 3 and latitude −89 _._ 3 would also be assigned Index 1.\n\n\nThese indices are then used to merge the two datasets. As a result, the final firm-level data includes\n\n\nannual observations of the firm, its location, sector, and fixed assets, as well as the mean, variance and\n\n\nmaximum of the temperature for the gridded location of the firm.\n\n\n3.2 Data pre-processing\n\n\nWe carry out several cleaning procedures to enhance the quality and consistency of the data. First,\n\n\nwe exclude firms with non-positive assets from our dataset. This exclusion is based on the accounting\n\n\nprinciple that fixed assets should always be non-negative. While it is technically possible to have zero\n\n\nfixed assets, these cases are excluded to enable the calculation of the growth rate of fixed assets. Second,\n\n\nto avoid potential double counting, our analysis is limited to unconsolidated data. We select U1 and U2\n\nas the consolidation codes in Orbis data. <sup>11</sup> This choice ensures that the data reflects each firm’s inde\n\npendent operations, preventing distortion from inter-company transactions or accounts. This is crucial\n\n\nfor examining the impact of location-specific factors,\n\nMention: \"Orbis\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Firm financial performance is geolocated by latitude and longitude so it can be matched to weather grids.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001211:9:0", "occurrence_id": "b037662f2e1d0dbaac58310e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Data merging. A key aspect of our research methodology is the merging of the Orbis and\n\n\nBERKEARTH datasets using latitude and longitude information. This common element allows\n\n\nus to associate firm-level financial performance data from Orbis with the corresponding weather data\n\n\nfrom BERKEARTH, both mapped to the same geographic region.\n\n\nGiven that the weather data is gridded with one degree of latitude and longitude, we first assign a\n\n\nunique index to each grid. For example, Index 1 is assigned to the upper leftmost grid (longitude −180\n\n\nto −179 and latitude −90 to −89). Subsequently, we assign the same index to the location of each firm.\n\n\nFor instance, a firm located at longitude −179 _._ 3 and latitude −89 _._ 3 would also be assigned Index 1.\n\n\nThese indices are then used to merge the two datasets. As a result, the final firm-level data includes\n\n\nannual observations of the firm, its location, sector, and fixed assets, as well as the mean, variance and\n\n\nmaximum of the temperature for the gridded location of the firm.\n\n\n3.2 Data pre-processing\n\n\nWe carry out several cleaning procedures to enhance the quality and consistency of the data. First,\n\n\nwe exclude firms with non-positive assets from our dataset. This exclusion is based on the accounting\n\n\nprinciple that fixed assets should always be non-negative. While it is technically possible to have zero\n\n\nfixed assets, these cases are excluded to enable the calculation of the growth rate of fixed assets. Second,\n\n\nto avoid potential double counting, our analysis is limited to unconsolidated data. We select U1 and U2\n\nas the consolidation codes in Orbis data. <sup>11</sup> This choice ensures that the data reflects each firm’s inde\n\npendent operations, preventing distortion from inter-company transactions or accounts. This is crucial\n\n\nfor examining the impact of location-specific factors,\n\nMention: \"BERKEARTH datasets\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Gridded weather observations are assigned to firms through this product.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001211:9:0", "occurrence_id": "f68398143af60cfe00f72f12", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Data merging. A key aspect of our research methodology is the merging of the Orbis and\n\n\nBERKEARTH datasets using latitude and longitude information. This common element allows\n\n\nus to associate firm-level financial performance data from Orbis with the corresponding weather data\n\n\nfrom BERKEARTH, both mapped to the same geographic region.\n\n\nGiven that the weather data is gridded with one degree of latitude and longitude, we first assign a\n\n\nunique index to each grid. For example, Index 1 is assigned to the upper leftmost grid (longitude −180\n\n\nto −179 and latitude −90 to −89). Subsequently, we assign the same index to the location of each firm.\n\n\nFor instance, a firm located at longitude −179 _._ 3 and latitude −89 _._ 3 would also be assigned Index 1.\n\n\nThese indices are then used to merge the two datasets. As a result, the final firm-level data includes\n\n\nannual observations of the firm, its location, sector, and fixed assets, as well as the mean, variance and\n\n\nmaximum of the temperature for the gridded location of the firm.\n\n\n3.2 Data pre-processing\n\n\nWe carry out several cleaning procedures to enhance the quality and consistency of the data. First,\n\n\nwe exclude firms with non-positive assets from our dataset. This exclusion is based on the accounting\n\n\nprinciple that fixed assets should always be non-negative. While it is technically possible to have zero\n\n\nfixed assets, these cases are excluded to enable the calculation of the growth rate of fixed assets. Second,\n\n\nto avoid potential double counting, our analysis is limited to unconsolidated data. We select U1 and U2\n\nas the consolidation codes in Orbis data. <sup>11</sup> This choice ensures that the data reflects each firm’s inde\n\npendent operations, preventing distortion from inter-company transactions or accounts. This is crucial\n\n\nfor examining the impact of location-specific factors,\n\nMention: \"BERKEARTH\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Its one-degree grid cells are indexed so each firm can be assigned to a unique cell.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001211:9:0", "occurrence_id": "37e1c1af694a3d9b97fb659c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Data merging. A key aspect of our research methodology is the merging of the Orbis and\n\n\nBERKEARTH datasets using latitude and longitude information. This common element allows\n\n\nus to associate firm-level financial performance data from Orbis with the corresponding weather data\n\n\nfrom BERKEARTH, both mapped to the same geographic region.\n\n\nGiven that the weather data is gridded with one degree of latitude and longitude, we first assign a\n\n\nunique index to each grid. For example, Index 1 is assigned to the upper leftmost grid (longitude −180\n\n\nto −179 and latitude −90 to −89). Subsequently, we assign the same index to the location of each firm.\n\n\nFor instance, a firm located at longitude −179 _._ 3 and latitude −89 _._ 3 would also be assigned Index 1.\n\n\nThese indices are then used to merge the two datasets. As a result, the final firm-level data includes\n\n\nannual observations of the firm, its location, sector, and fixed assets, as well as the mean, variance and\n\n\nmaximum of the temperature for the gridded location of the firm.\n\n\n3.2 Data pre-processing\n\n\nWe carry out several cleaning procedures to enhance the quality and consistency of the data. First,\n\n\nwe exclude firms with non-positive assets from our dataset. This exclusion is based on the accounting\n\n\nprinciple that fixed assets should always be non-negative. While it is technically possible to have zero\n\n\nfixed assets, these cases are excluded to enable the calculation of the growth rate of fixed assets. Second,\n\n\nto avoid potential double counting, our analysis is limited to unconsolidated data. We select U1 and U2\n\nas the consolidation codes in Orbis data. <sup>11</sup> This choice ensures that the data reflects each firm’s inde\n\npendent operations, preventing distortion from inter-company transactions or accounts. This is crucial\n\n\nfor examining the impact of location-specific factors,\n\nMention: \"Orbis data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Unconsolidated records with codes U1 and U2 are kept to avoid double counting.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001216:25:7", "occurrence_id": "dc808079fbe48f97feb6c81e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 𝑖𝑖</sup> + 𝑆 **𝑆** 𝑆 𝑆𝑆_𝐹 **𝐹** 𝐹 **𝐹** _𝑇 **𝑇** 𝑇𝑇_𝐿 **𝐿** 𝐿 **𝐿** 𝑙𝑙 <sup>𝑖𝑖</sup> + 𝑆 **𝑆** 𝑆 𝑆𝑆_𝐶 **𝐶** 𝐶 **𝐶** 𝐶 **𝐶** _𝑡 **𝑡** 𝑥𝑥 <sup>𝑖𝑖</sup>\n\n\n\n**𝑆** **𝐶** **𝐶** **𝐶** **𝐶** **𝑆**\n\n**_8. Technology costs: Capital_** **𝑆**\n\n\n\nThe capital cost ($/MWh) of a technology and its standard deviation are initially taken from the IEA's\nglobal database (assuming discount rate of 7%). A learning rate term is introduced to capture the\nreduction in capital cost over time, in line with the global accumulation of the technology (which is\nexogenous). The accumulation of global technology can be used to represent different global energy\npathways, such as business as usual or net zero. **𝐶** **𝐶** **𝐶** **𝐶** **𝐶** **𝐶** **𝐶** **𝐶** **𝑙** **𝑙** **𝑙** **𝑙** **𝑖** **𝐶** **𝐶** **𝐶** **𝐶** **𝐶** **𝑎**\n\n\n\npathways, such as business as usual or net zero.\n\n𝑖𝑖\n\n𝐶 **𝐶** 𝐶 **𝐶** 𝐶 **𝐶*\n\nMention: \"IEA's\nglobal database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Initial capital cost per megawatt-hour and its dispersion are taken from this technology database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001217:36:1", "occurrence_id": "9ffa2af9deb8eff2d1c425cc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  year, the variable equals _M_ /12, where _M_ represents the DPI 2020\nmonth of the election, and (12 – _M_ )/12 in the preceding year. For all other\nyears, the variable is set to zero. The type of election considered\n(legislative or executive) depends on the political system in place\n(presidential, assembly-elected president, or parliamentary).\n\n\n\nDPI 2020\n\n\n\nDemocratic accountability 0-6 indicator assessing how responsive government is to its people, on\nthe basis that the less responsive it is, the more likely it is that the\ngovernment will fall, peacefully in a democratic society, but possibly\nviolently in a non-democratic one.\n\nLaw and Order 0-6 indicator of the assessment of established law and order in a country.\nLaw and Order are assessed separately, with each sub-component\ncomprising zero to three points. The Law sub-component is an\nassessment of the strength and impartiality of the legal system, while the\nOrder sub-component is an assessment of popular observance of the law.\n\n\n34\n\n\n\nICRG Database\n\n\nICRG Database\n\nMention: \"ICRG Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its Law and Order sub-components score the strength and impartiality of the legal system.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001221:11:2", "occurrence_id": "94929088141e12876780c9d6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  observing whether\n\n\n\nthird of the urban labor force across the six countries studied (Table 2). The small share of\n\nmigrants in urban Uganda is largely definitional; the data only allows for observing whether\n\n13 Ideally each person in the individual datasets could be linked directly with Africapolis data, but individuals\nare not geo-referenced in the censuses (Ghana, Kenya, and Mali) and the labor force survey (Ethiopia), and for\nconfidentiality purposes, geo-references are offset in the household survey data (Tanzania, and Uganda).\n14 In Kenya, too many districts had agglomerations with mixed city sizes to warrant their automatic exclusion.\nTherefore, if 70 percent of the total agglomeration area from Africapolis in the specific district consisted of one\ncity size classification, then that district was assigned the dominant city size classification. Nine districts were\nstill dropped because the composition of agglomeration city sizes did not satisfy this criterion.\n15 The city size distribution deviates most from the Africapolis city size distribution for Uganda (between 12\nand 35 percentage points), while it is almost identical in Ethiopia. In the other countries, the difference across\nthe various size categories mostly ranges between 5 and 15 percentage points. There is a better (albeit still\nimperfect) match with the city size distribution reported in the World Development Indicators (WDI), which\nare based on official definitions of urban areas.\n\nMention: \"Africapolis data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Direct linkage to individuals is impossible because censuses and surveys lack or offset geo-references.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001221:11:2", "occurrence_id": "2659af79a92db40e52f2bf0d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  observing whether\n\n\n\nthird of the urban labor force across the six countries studied (Table 2). The small share of\n\nmigrants in urban Uganda is largely definitional; the data only allows for observing whether\n\n13 Ideally each person in the individual datasets could be linked directly with Africapolis data, but individuals\nare not geo-referenced in the censuses (Ghana, Kenya, and Mali) and the labor force survey (Ethiopia), and for\nconfidentiality purposes, geo-references are offset in the household survey data (Tanzania, and Uganda).\n14 In Kenya, too many districts had agglomerations with mixed city sizes to warrant their automatic exclusion.\nTherefore, if 70 percent of the total agglomeration area from Africapolis in the specific district consisted of one\ncity size classification, then that district was assigned the dominant city size classification. Nine districts were\nstill dropped because the composition of agglomeration city sizes did not satisfy this criterion.\n15 The city size distribution deviates most from the Africapolis city size distribution for Uganda (between 12\nand 35 percentage points), while it is almost identical in Ethiopia. In the other countries, the difference across\nthe various size categories mostly ranges between 5 and 15 percentage points. There is a better (albeit still\nimperfect) match with the city size distribution reported in the World Development Indicators (WDI), which\nare based on official definitions of urban areas.\n\nMention: \"labor force survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its respondents are not geo-referenced, blocking spatial linkage to agglomerations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001221:11:2", "occurrence_id": "6044974e708762967ea3ace7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  observing whether\n\n\n\nthird of the urban labor force across the six countries studied (Table 2). The small share of\n\nmigrants in urban Uganda is largely definitional; the data only allows for observing whether\n\n13 Ideally each person in the individual datasets could be linked directly with Africapolis data, but individuals\nare not geo-referenced in the censuses (Ghana, Kenya, and Mali) and the labor force survey (Ethiopia), and for\nconfidentiality purposes, geo-references are offset in the household survey data (Tanzania, and Uganda).\n14 In Kenya, too many districts had agglomerations with mixed city sizes to warrant their automatic exclusion.\nTherefore, if 70 percent of the total agglomeration area from Africapolis in the specific district consisted of one\ncity size classification, then that district was assigned the dominant city size classification. Nine districts were\nstill dropped because the composition of agglomeration city sizes did not satisfy this criterion.\n15 The city size distribution deviates most from the Africapolis city size distribution for Uganda (between 12\nand 35 percentage points), while it is almost identical in Ethiopia. In the other countries, the difference across\nthe various size categories mostly ranges between 5 and 15 percentage points. There is a better (albeit still\nimperfect) match with the city size distribution reported in the World Development Indicators (WDI), which\nare based on official definitions of urban areas.\n\nMention: \"household survey data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Coordinates are deliberately offset for confidentiality, so precise spatial joining is not possible.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001227:8:0", "occurrence_id": "d2046f7bd1af622064ecfb36", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Notes: Stars indicates levels of significance of t-test on equality of means between the two years: * p<0.1, ** p<0.05, *** p<0.01\nSource: LFS 2019-20, 2020-21\n\n\n\nIn order to better gauge the differential labor market impact of the COVID-19 crisis between comparable\nmen and women, we estimate the following linear OLS model on the pooled sample of individuals in the\nworking age population observed in the 2019-20 and 2020-21 LFS survey waves: **𝑖** **𝐹** **𝐹** **𝐹** **𝐶** **𝐶** **𝐹** **𝐹** **𝐹** **𝐶** **𝐶** **𝐹** **𝐹**\n\n\n\nworking age population observed in the 2019-20 and 2020-21 LFS survey waves: (1)   𝑌𝑌𝑖 **𝑖** 𝑖𝑖 = 𝛼𝛼0 + 𝛽𝛽 𝐹 **𝐹** 𝐹 **𝐹** 𝐹 **𝐹** 𝑖𝑖 + 𝛾𝛾 𝐶 **𝐶** 𝐶 **𝐶** 𝐶𝐶𝑖𝑖 + 𝛿𝛿 (𝐹 **𝐹** 𝐹 **𝐹** 𝐹 **𝐹** 𝑖𝑖 ∗𝐶 **𝐶** 𝐶 **𝐶** 𝐶𝐶𝑖𝑖) + 𝜂𝜂 𝑋𝑋𝑖𝑖 + 𝐹 **𝐹** 𝑝𝑝 + 𝐹 **𝐹** 𝑠𝑠 + 𝜖𝜖𝑖𝑖\n\nwhere 𝑌𝑌𝑖𝑖 is the outcome of interest (labor force participation, employment and unemployment\n\n\n\n(1)   𝑌𝑌𝑖 **𝑖** 𝑖𝑖 = 𝛼𝛼0 + 𝛽𝛽 𝐹 **𝐹** 𝐹 **𝐹** 𝐹 **𝐹** 𝑖𝑖 + 𝛾𝛾\n\nMention: \"LFS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Means are compared between two consecutive years to gauge differential impacts.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001227:8:0", "occurrence_id": "3cb7c51bc1cd6ce5eac0a482", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Notes: Stars indicates levels of significance of t-test on equality of means between the two years: * p<0.1, ** p<0.05, *** p<0.01\nSource: LFS 2019-20, 2020-21\n\n\n\nIn order to better gauge the differential labor market impact of the COVID-19 crisis between comparable\nmen and women, we estimate the following linear OLS model on the pooled sample of individuals in the\nworking age population observed in the 2019-20 and 2020-21 LFS survey waves: **𝑖** **𝐹** **𝐹** **𝐹** **𝐶** **𝐶** **𝐹** **𝐹** **𝐹** **𝐶** **𝐶** **𝐹** **𝐹**\n\n\n\nworking age population observed in the 2019-20 and 2020-21 LFS survey waves: (1)   𝑌𝑌𝑖 **𝑖** 𝑖𝑖 = 𝛼𝛼0 + 𝛽𝛽 𝐹 **𝐹** 𝐹 **𝐹** 𝐹 **𝐹** 𝑖𝑖 + 𝛾𝛾 𝐶 **𝐶** 𝐶 **𝐶** 𝐶𝐶𝑖𝑖 + 𝛿𝛿 (𝐹 **𝐹** 𝐹 **𝐹** 𝐹 **𝐹** 𝑖𝑖 ∗𝐶 **𝐶** 𝐶 **𝐶** 𝐶𝐶𝑖𝑖) + 𝜂𝜂 𝑋𝑋𝑖𝑖 + 𝐹 **𝐹** 𝑝𝑝 + 𝐹 **𝐹** 𝑠𝑠 + 𝜖𝜖𝑖𝑖\n\nwhere 𝑌𝑌𝑖𝑖 is the outcome of interest (labor force participation, employment and unemployment\n\n\n\n(1)   𝑌𝑌𝑖 **𝑖** 𝑖𝑖 = 𝛼𝛼0 + 𝛽𝛽 𝐹 **𝐹** 𝐹 **𝐹** 𝐹 **𝐹** 𝑖𝑖 + 𝛾𝛾\n\nMention: \"LFS survey waves\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The pooled working-age sample from both waves is used for the gendered impact regression.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001231:25:0", "occurrence_id": "92a239422faf4e66129b930d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 24\n\n\n\n**Table** **SI5-2:** Impact of heat on the difference in reported crimes and arrests in Houston\n\n\nContemporaneous arrests 3-day pooled arrests\n\n\nT above 100F 0.045 0.047\n(0.010) (0.012)\nT 95-100F 0.034 0.040\n(0.005) (0.007)\nT 90-95F 0.022 0.023\n(0.005) (0.007)\nT 85-90F 0.016 0.021\n(0.004) (0.006)\nT 80-85F 0.018 0.020\n(0.004) (0.006)\nT 75-80F 0.015 0.015\n(0.004) (0.005)\nN 1,840,860 1,839,600\nOutcome mean, T60-65 0.33 0.03\n**Fixed** **Effects:**\nTract Yes Yes\nMonth Yes Yes\nYear Yes Yes\nDOW Yes Yes\n\n\nNotes: All columns report the results of a linear fixed effects specification. We estimate the impact of\na hot day on the difference between the number of incidents reported to the Houston Police Department\n(Houston PD) and the number of arrests reported to the Texas Department of Public Safety (TDPS). In\nall cases we aggregate the count of incidents (Houston PD) data or arrests (TDPS data) to the tract-day\nlevel and conduct analysis at that level of aggregation. The sample in all cases is a balanced panel of tracts\nthat contain at least one Houston PD crime report at the daily level from 2010 to 2017. In column 2, we\npool arrests across the day of interest and the following two days. Errors are clustered at the tract level\nand are reported in parentheses. All regressions are weighted by the total population in each tract-year.\nAll regressions include the full set\n\nMention: \"TDPS data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Arrest counts are aggregated to the census tract-day level for the heat-crime analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001233:9:1", "occurrence_id": "c5758c976d46f8fe726a1a2f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  𝑑𝑑ℎ�= Pr (𝑘𝑘ℎ−1 < 𝜶𝜶𝑑𝑑𝑿𝑿𝑖 **𝑖** + 𝜸𝜸𝑑𝑑𝑮𝑮𝑖 **𝑖** + 𝜗𝜗𝑐𝑐 + 𝜀𝜀𝑖 **𝑖** ≤𝑘𝑘ℎ)\n\n(3).\n\nAssuming 𝜀𝜀𝑖 **𝑖** is normally distributed, we can use an ordered probit to estimate the parameters of\n\n\n\n(3).\n\n**𝑖**\n\n\n\n**𝑖**\n\n𝑑𝑑 have a large number of “ _do not know_ ” responses. We assume there\n\n𝑖\n\n\n\n(3).\n\n**𝑖**\n\n𝑑𝑑\n\nare two reasons why respondents would choose “\n\nQuestions used to construct 𝑂𝑂𝑖\n\n\n\nare two reasons why respondents would choose “ _do not know_ ”. They may be genuinely unaware\n\nof the number of fatalities in a disaster, or they may simply want to finish the interview sooner and\n\ngive that response to speed up the process. If “ _do not know_ ” responses are non-random, for\n\nexample, if respondents who are more likely to overestimate fatalities are also more likely to be\n\nimpatient about the interview process, the estimates in equation (3) could be biased.\n\n\nTo address this potential source of bias, we model the non-response process by estimating the\n\nprobability of giving a response (other than “ _do not know_ ”) as a function of the respondent\n\ncharacteristics jointly with estimating the ordered probit in equation (3). We use the number of\n\n“ _do not know_ ” responses to other questions in the section of the LiTS that contains the disaster\n\nquestion (module 4) as an instrument for the selection equation. The exclusion restriction here is\n\nbased on the assumption that the desire to finish the interview sooner is not related to factual\n\nknowledge about disaster fatalities but is correlated with the probability of giving “do not know”\n\n\n\nanswers to the disaster questions. The selection equation could be then expressed as follows:\n\n**𝑖** **\n\nMention: \"LiTS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Counts of \\\"do not know\\\" responses in the disaster module serve as the instrument in the selection equation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001234:11:0", "occurrence_id": "1d85531d17e1289458d0e624", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: coverage and improved reporting (as noted by EM-DAT), we focus our analysis on data from 2000\n\nonwards.\n\n\nWe define an intensity shock, measuring the number of people affected by natural disasters using\n\ndata from EM-DAT following Fomby, Ikeda, and Loayza (2013), as:\n\n\n𝐼𝑛𝑡𝑒𝑛𝑠𝑖𝑡𝑦𝑖,𝑡 = <sup>𝑇𝑜𝑡𝑎𝑙 𝑑𝑒𝑎𝑡ℎ𝑠𝑖,𝑡</sup> <sup>+ 0.3 ∗𝑇𝑜𝑡𝑎𝑙 𝑎𝑓𝑓𝑒𝑐𝑡𝑒𝑑𝑖,𝑡</sup>\n\n𝑃𝑜𝑝𝑢𝑙𝑎𝑡𝑖𝑜𝑛𝑖,𝑡\n\n\nwhere _I_ is the country and _t_ is years. <sup>2</sup> [^2: It is also possible to define a damages shock which looks at the economic cost of a natural disaster. We prefer an\nintensity shock, however, for several reasons. First, there are less than half as many damages observations as\nobservations on people affected by natural disasters. Second, damages are likely based on insured damages. Third,\nestimates of economic cost are harder to determine and more subjective.] Unlike in Fomby, Ikeda, and Loayza (2013), and more\n\nrecently Bayoumi, Quayyum, and Das (2021), we use the actual intensity measure as the climate\n\nshock rather than a dummy variable based on some threshold value. Table 1 provides the name,\n\ndescription, and source of the other variables used in this section. Loayza et al. (2012) show that\n\nthe type of disaster may matter for its impact on the economy, however, we study all disasters.\n\n\nWe assume that any unreported event is considered to be a non-occurrence, and hence assigned an\n\nintensity value of zero. This decision is supported by several factors. First, the improved data\n\ncoverage by EM-DAT from 2000 onwards significantly reduces the likelihood of major events\n\ngoing unreported. Also, within our sample we notice an association between missing damage\n\nvalues and lower climate shock intensities, suggesting that unreported events are likely to be\n\nsmaller in scale. Second, we underscore the importance of establishing a counterfactual scenario\n\nwithout any shocks. Solely analyzing climate event datasets might not\n\nMention: \"EM-DAT\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"People affected by natural disasters define the intensity shock, using records from 2000 onwards.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001235:57:2", "occurrence_id": "65bec53e9e7a3f234fbf24d4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  <sup>the benchmark price of output.</sup>\n\n\nData for calibration includes the 2017 China’s input-output table and a firm-level dataset\n\n<sup>_<u>v</u>_</sup> _<u>i</u>_\nfrom the Ministry of Ecology and Environment of 2017, which can be used to derive the “ _V_ <sup>”</sup>\n\ncomponent in Equation (A4). The elasticities of substitution ( σ ) at different levels of the nested\n\nCES structure are obtained from calibrations and various sources. Table B2 provides the\n\n\n56\n\nMention: \"2017 China’s input-output table\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Together with a firm-level environmental dataset it provides the coefficients for calibration.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001236:8:1", "occurrence_id": "be21e9572793ea9c6c06dfa9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: e sur la Population et l’Emploi [ENPE]). The survey is a repeated cross-section of\nhouseholds with its main goal producing statistics regarding the social, educational, and\neconomic characteristics of the population that is either working, unemployed, or out of the\nlabor force.\n\nAmong other information, the ENPE reports individuals’ employment status by industry (if\nany), gender, and province for intermittent years between 2006 and 2016. It does not provide\ndata for wages or any other variables of interest. ENPE provides individual level weights that\naggregate to representative statistics of the Tunisian population. To construct the labor\nmarket indicators used in the empirical part of this paper, we excluded individuals under age\n15 and aggregated the data at the district-year level using the provided weights. The emission\nintensity data at the sectoral level comes from IMF climate dashboard, providing data\nbetween 1995 and 2018. We then harmonize industries to merge the sectoral codes in ENPE\nwith emission intensity data from IMF climate dashboard. We also merge <mark>energy price data</mark>\n<mark>for Tunisia obtained from the CEIC database. This captures the price at which energy is sold</mark>\n<mark>to industrial customers, such as manufacturing plants, factories, or large commercial</mark>\n<mark>enterprises. W</mark> e were then able to construct a provincial panel for labor market and price\nexposure.\n\n#### **4. Results**\n\n\n**<mark>4.1. E</mark>** **ffects of trade-related environmental policies on carbon-intensive industries and**\n**trade flows**\n\n\n7\n\nMention: \"IMF climate dashboard\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Sectoral emission intensity over this span is taken from this dashboard and harmonised to the survey's industry codes.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001236:8:1", "occurrence_id": "ed1b74d1f145fa6bb51a5ee8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: e sur la Population et l’Emploi [ENPE]). The survey is a repeated cross-section of\nhouseholds with its main goal producing statistics regarding the social, educational, and\neconomic characteristics of the population that is either working, unemployed, or out of the\nlabor force.\n\nAmong other information, the ENPE reports individuals’ employment status by industry (if\nany), gender, and province for intermittent years between 2006 and 2016. It does not provide\ndata for wages or any other variables of interest. ENPE provides individual level weights that\naggregate to representative statistics of the Tunisian population. To construct the labor\nmarket indicators used in the empirical part of this paper, we excluded individuals under age\n15 and aggregated the data at the district-year level using the provided weights. The emission\nintensity data at the sectoral level comes from IMF climate dashboard, providing data\nbetween 1995 and 2018. We then harmonize industries to merge the sectoral codes in ENPE\nwith emission intensity data from IMF climate dashboard. We also merge <mark>energy price data</mark>\n<mark>for Tunisia obtained from the CEIC database. This captures the price at which energy is sold</mark>\n<mark>to industrial customers, such as manufacturing plants, factories, or large commercial</mark>\n<mark>enterprises. W</mark> e were then able to construct a provincial panel for labor market and price\nexposure.\n\n#### **4. Results**\n\n\n**<mark>4.1. E</mark>** **ffects of trade-related environmental policies on carbon-intensive industries and**\n**trade flows**\n\n\n7\n\nMention: \"CEIC database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Prices at which energy is sold to industrial customers are obtained from this commercial database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001238:45:0", "occurrence_id": "30f1cefcf817d1ee3e5203f8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: #### **Appendix B: A new panel of road investments**\n\nSignificant investments in the Democratic Republic of Congo (DRC) have been made since\n2000, particularly focused on large infrastructure projects around 2010. Yet, there seems\nto be a lack of quantitative evidence on their benefits, partly due to the lack of available\ngranular data on these infrastructure projects. An important contribution of this paper is\nthe construction of the first panel of road investments in the DRC since 2002. We collect data\nfrom multiple sources to create the first database that gathers the exact location, timing,\nand amount spent on transport projects since 2000. First, we create a database of road\ntransport projects in DRC. Second, we geolocalize as many investment projects as possible.\nThird, we create a panel of data that can be used with the rest of the data.\n\n###### **B.1 Create a database of road transport projects**\n\n\nWe rely on multiple sources of information: universities (William & Mary, Boston University\n& Johns Hopkins), government bodies (Plateforme de Gestion de l’Aide et des Investissements, Cellule Infrastructure), and donors’ websites (World Bank, ADB).\n\n\n1. AidData  - DRC AIMS Geocoded Research Release, Version 1.3.1 <sup>1</sup>\n\n\nAiddata is a development research lab housed at William & Mary’s Global Research\nInstitute. They provide granular data on topics related to development, in particular\non the financing of infrastructure projects and China’s expanding portfolio of overseas investments and soft power activities. One of their projects aimed at listing all\nprojects that have happened in African countries, including their commitment, the\nstarting and end date of the projects, the sector, and the main donor. The DRC\nAIMS dataset was part of this effort and includes 6842 projects in all sectors covering\nthe period 1998-2014. Among these\n\nMention: \"DRC\nAIMS dataset\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its 6,842 projects with commitments, dates, sectors and donors are used to build the road investment panel.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001240:3:1", "occurrence_id": "a45d7ec0f0613dcf20eccfbb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  previous work by greatly expanding representation for plants,\ninvertebrates and other non-vertebrate species. Our algorithm allows rapid updates and new\nmaps as the GBIF data increase.\n\n\nIn this paper, we draw on the previous work to explore some implications for 30x30 in a sample\nof 10 countries in Latin America, Africa and the Asia-Pacific region. We focus on individual\ncountries to highlight the role of conservation stewardship in local settings. Our country analyses\nbegin with current protected areas, which exhibit great cross-country variation in territorial\ncoverage and representation of species taxa. We consider both terrestrial and marine cases,\nfocusing on species that are endemic to each country. Drawing on our GBIF spatial database for\nnearly 600,000 species, we overlay species occurrence maps on the most recent protected-area\nmaps from the World Database of Protected Areas. This enables identification of endemic species\nwhose coverage by existing protected areas is either non-existent or extremely sparse.\n\n\nOur approach departs from many previous exercises by giving equal weight to all vertebrates,\ninvertebrates, plants, and other species whose occurrence regions are mapped in our database.\nWe use a spatially-efficient algorithm to identify a hierarchy of priorities for localities where new\n\n\n1 Pimm, S. L., Jenkins, C. N., Abell, R., Brooks, T. M., Gittleman, J. L., Joppa, L. N., ... & Sexton, J. O. 2014.\nThe biodiversity of species and their rates of extinction, distribution, and protection. Science, 344(6187),\n1246752.\n\n\n2\n\nMention: \"GBIF data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"inform\", \"summary\": \"The mapping algorithm can be refreshed quickly as this occurrence data grows.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001240:3:1", "occurrence_id": "099163d13b8f3d00ec4f5ffc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  previous work by greatly expanding representation for plants,\ninvertebrates and other non-vertebrate species. Our algorithm allows rapid updates and new\nmaps as the GBIF data increase.\n\n\nIn this paper, we draw on the previous work to explore some implications for 30x30 in a sample\nof 10 countries in Latin America, Africa and the Asia-Pacific region. We focus on individual\ncountries to highlight the role of conservation stewardship in local settings. Our country analyses\nbegin with current protected areas, which exhibit great cross-country variation in territorial\ncoverage and representation of species taxa. We consider both terrestrial and marine cases,\nfocusing on species that are endemic to each country. Drawing on our GBIF spatial database for\nnearly 600,000 species, we overlay species occurrence maps on the most recent protected-area\nmaps from the World Database of Protected Areas. This enables identification of endemic species\nwhose coverage by existing protected areas is either non-existent or extremely sparse.\n\n\nOur approach departs from many previous exercises by giving equal weight to all vertebrates,\ninvertebrates, plants, and other species whose occurrence regions are mapped in our database.\nWe use a spatially-efficient algorithm to identify a hierarchy of priorities for localities where new\n\n\n1 Pimm, S. L., Jenkins, C. N., Abell, R., Brooks, T. M., Gittleman, J. L., Joppa, L. N., ... & Sexton, J. O. 2014.\nThe biodiversity of species and their rates of extinction, distribution, and protection. Science, 344(6187),\n1246752.\n\n\n2\n\nMention: \"GBIF spatial database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Occurrences for nearly 600,000 species are turned into range maps and overlaid on protected areas.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001240:3:1", "occurrence_id": "31c61552207da28045abf18a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  previous work by greatly expanding representation for plants,\ninvertebrates and other non-vertebrate species. Our algorithm allows rapid updates and new\nmaps as the GBIF data increase.\n\n\nIn this paper, we draw on the previous work to explore some implications for 30x30 in a sample\nof 10 countries in Latin America, Africa and the Asia-Pacific region. We focus on individual\ncountries to highlight the role of conservation stewardship in local settings. Our country analyses\nbegin with current protected areas, which exhibit great cross-country variation in territorial\ncoverage and representation of species taxa. We consider both terrestrial and marine cases,\nfocusing on species that are endemic to each country. Drawing on our GBIF spatial database for\nnearly 600,000 species, we overlay species occurrence maps on the most recent protected-area\nmaps from the World Database of Protected Areas. This enables identification of endemic species\nwhose coverage by existing protected areas is either non-existent or extremely sparse.\n\n\nOur approach departs from many previous exercises by giving equal weight to all vertebrates,\ninvertebrates, plants, and other species whose occurrence regions are mapped in our database.\nWe use a spatially-efficient algorithm to identify a hierarchy of priorities for localities where new\n\n\n1 Pimm, S. L., Jenkins, C. N., Abell, R., Brooks, T. M., Gittleman, J. L., Joppa, L. N., ... & Sexton, J. O. 2014.\nThe biodiversity of species and their rates of extinction, distribution, and protection. Science, 344(6187),\n1246752.\n\n\n2\n\nMention: \"World Database of Protected Areas\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The most recent protected-area boundaries provide the layer species coverage is measured against.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001243:23:0", "occurrence_id": "32c861d4ee5551df103b43fb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **<u>Table 3. Shifts in education level and profession for middle class households in 2005–08 and 2015–18</u>**\n\n\n\n<u>Shares (%) by</u>\neducation level and\n\ntype of occupation\n\n%\n2005–08 2015–18\n<u>change</u>\n_Education of household head_\n\n\n\nPrimary or less 24 13 –43\nLower-secondary 17 15 –14\nUpper-secondary 35 40 13\nPost-secondary 2 1 –49\n<u>Tertiary</u> <u>22</u> <u>31</u> <u>41</u>\n\n_Occupation of_ _household head_\n\nManagers, professionals, technicians 48 52 8\nSupport, service, sales workers 23 21 –11\nCrafts, trades, elementary occupations 18 17 –11\nPlant/machine operators/assemblers 7 8 15\n<u>Skilled and unskilled agricultural/etc.</u> <u>2</u> <u>2</u> <u>–10</u>\n_<u>Source</u>_ <u>: Authors’ calculations using data from the EU-SILC UDB-L.</u>\n_Note_ : This table the composition in terms of education and occupation of the household head for the European\nhouseholds that belong to the middle class under the baseline definition (lower threshold = income associated to a\n4 percent probability of falling into poverty; upper threshold = income associated to a 1 percent probability of falling\ninto poverty). Values are pooled for all countries in the sample, population weighted.\n\n\n22\n\nMention: \"EU-SILC UDB-L\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Education and occupation of the household head are compared across periods for middle-class households.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001244:15:1", "occurrence_id": "92e150f68d49b560acd0869b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  costs and profits but excluding taxes, as discussed below. It includes producer\nsubsidies to the extent to which they are reflected in the supply cost by producers.\n\n  - **VAT rates,** including the standard VAT rate in each country and the reduced rates granted\n\nto specific fuels for a limited set of end-uses. The effective VAT rate on energy consumed\nin the industrial, power generation, and service sectors is zero because these users reclaim\npaid VAT. The output of these sectors, e.g., the electricity produced by the power sector,\ncarries VAT when consumed by households.\n\n<u>Retail prices</u>\n\nAnnual average retail prices are widely available for coal, natural gas, and electricity at an\neconomywide level, as well as disaggregated by main end-user—industrial, residential, and power\ngeneration. However, for other fuels (e.g., gasoline and diesel) only economywide annual average\nretail prices are available (Parry et al, 2021). The dataset from Parry et al., (2021) does not include\nspatially specific fuel prices. Accordingly, retail prices for some fuels are taken from IMF and\nWorld Bank country desk datasets. For cases where such data was not available, they use a simple\naverage across various third-party sources, including Eurostat, IEA, World Bank Doing Business\nIndicators, Global Petrol Prices Retail Energy Price Data, and Enerdata Global Energy & CO2\n\n\n31 When different carbon tax rates are applied in the sector they cover, this paper uses the highest rate in the\ncomputation of the TCP as WB (2022a) does not provide quantitative estimates of emissions covered by each specific\ntax rate.\n\n12\n\nMention: \"Annual average retail prices\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Economy-wide and end-user prices for coal, gas and electricity enter the effective price calculation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001244:15:1", "occurrence_id": "3cfe9bae57d0da275d6f1ae8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  costs and profits but excluding taxes, as discussed below. It includes producer\nsubsidies to the extent to which they are reflected in the supply cost by producers.\n\n  - **VAT rates,** including the standard VAT rate in each country and the reduced rates granted\n\nto specific fuels for a limited set of end-uses. The effective VAT rate on energy consumed\nin the industrial, power generation, and service sectors is zero because these users reclaim\npaid VAT. The output of these sectors, e.g., the electricity produced by the power sector,\ncarries VAT when consumed by households.\n\n<u>Retail prices</u>\n\nAnnual average retail prices are widely available for coal, natural gas, and electricity at an\neconomywide level, as well as disaggregated by main end-user—industrial, residential, and power\ngeneration. However, for other fuels (e.g., gasoline and diesel) only economywide annual average\nretail prices are available (Parry et al, 2021). The dataset from Parry et al., (2021) does not include\nspatially specific fuel prices. Accordingly, retail prices for some fuels are taken from IMF and\nWorld Bank country desk datasets. For cases where such data was not available, they use a simple\naverage across various third-party sources, including Eurostat, IEA, World Bank Doing Business\nIndicators, Global Petrol Prices Retail Energy Price Data, and Enerdata Global Energy & CO2\n\n\n31 When different carbon tax rates are applied in the sector they cover, this paper uses the highest rate in the\ncomputation of the TCP as WB (2022a) does not provide quantitative estimates of emissions covered by each specific\ntax rate.\n\n12\n\nMention: \"annual average\nretail prices\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"For gasoline and diesel only economy-wide values exist, with no spatial detail.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001244:15:1", "occurrence_id": "eda47c15a7648fe227316b59", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  costs and profits but excluding taxes, as discussed below. It includes producer\nsubsidies to the extent to which they are reflected in the supply cost by producers.\n\n  - **VAT rates,** including the standard VAT rate in each country and the reduced rates granted\n\nto specific fuels for a limited set of end-uses. The effective VAT rate on energy consumed\nin the industrial, power generation, and service sectors is zero because these users reclaim\npaid VAT. The output of these sectors, e.g., the electricity produced by the power sector,\ncarries VAT when consumed by households.\n\n<u>Retail prices</u>\n\nAnnual average retail prices are widely available for coal, natural gas, and electricity at an\neconomywide level, as well as disaggregated by main end-user—industrial, residential, and power\ngeneration. However, for other fuels (e.g., gasoline and diesel) only economywide annual average\nretail prices are available (Parry et al, 2021). The dataset from Parry et al., (2021) does not include\nspatially specific fuel prices. Accordingly, retail prices for some fuels are taken from IMF and\nWorld Bank country desk datasets. For cases where such data was not available, they use a simple\naverage across various third-party sources, including Eurostat, IEA, World Bank Doing Business\nIndicators, Global Petrol Prices Retail Energy Price Data, and Enerdata Global Energy & CO2\n\n\n31 When different carbon tax rates are applied in the sector they cover, this paper uses the highest rate in the\ncomputation of the TCP as WB (2022a) does not provide quantitative estimates of emissions covered by each specific\ntax rate.\n\n12\n\nMention: \"dataset from Parry et al.\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"It lacks spatially specific fuel prices, so other sources are used to fill in.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001244:15:1", "occurrence_id": "68d5cd82fd77ec07346fe1d9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  costs and profits but excluding taxes, as discussed below. It includes producer\nsubsidies to the extent to which they are reflected in the supply cost by producers.\n\n  - **VAT rates,** including the standard VAT rate in each country and the reduced rates granted\n\nto specific fuels for a limited set of end-uses. The effective VAT rate on energy consumed\nin the industrial, power generation, and service sectors is zero because these users reclaim\npaid VAT. The output of these sectors, e.g., the electricity produced by the power sector,\ncarries VAT when consumed by households.\n\n<u>Retail prices</u>\n\nAnnual average retail prices are widely available for coal, natural gas, and electricity at an\neconomywide level, as well as disaggregated by main end-user—industrial, residential, and power\ngeneration. However, for other fuels (e.g., gasoline and diesel) only economywide annual average\nretail prices are available (Parry et al, 2021). The dataset from Parry et al., (2021) does not include\nspatially specific fuel prices. Accordingly, retail prices for some fuels are taken from IMF and\nWorld Bank country desk datasets. For cases where such data was not available, they use a simple\naverage across various third-party sources, including Eurostat, IEA, World Bank Doing Business\nIndicators, Global Petrol Prices Retail Energy Price Data, and Enerdata Global Energy & CO2\n\n\n31 When different carbon tax rates are applied in the sector they cover, this paper uses the highest rate in the\ncomputation of the TCP as WB (2022a) does not provide quantitative estimates of emissions covered by each specific\ntax rate.\n\n12\n\nMention: \"Eurostat\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"One of the third-party sources averaged to obtain missing retail prices.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001244:15:1", "occurrence_id": "b5c4aea7642a1ac0f9acb4c3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  costs and profits but excluding taxes, as discussed below. It includes producer\nsubsidies to the extent to which they are reflected in the supply cost by producers.\n\n  - **VAT rates,** including the standard VAT rate in each country and the reduced rates granted\n\nto specific fuels for a limited set of end-uses. The effective VAT rate on energy consumed\nin the industrial, power generation, and service sectors is zero because these users reclaim\npaid VAT. The output of these sectors, e.g., the electricity produced by the power sector,\ncarries VAT when consumed by households.\n\n<u>Retail prices</u>\n\nAnnual average retail prices are widely available for coal, natural gas, and electricity at an\neconomywide level, as well as disaggregated by main end-user—industrial, residential, and power\ngeneration. However, for other fuels (e.g., gasoline and diesel) only economywide annual average\nretail prices are available (Parry et al, 2021). The dataset from Parry et al., (2021) does not include\nspatially specific fuel prices. Accordingly, retail prices for some fuels are taken from IMF and\nWorld Bank country desk datasets. For cases where such data was not available, they use a simple\naverage across various third-party sources, including Eurostat, IEA, World Bank Doing Business\nIndicators, Global Petrol Prices Retail Energy Price Data, and Enerdata Global Energy & CO2\n\n\n31 When different carbon tax rates are applied in the sector they cover, this paper uses the highest rate in the\ncomputation of the TCP as WB (2022a) does not provide quantitative estimates of emissions covered by each specific\ntax rate.\n\n12\n\nMention: \"IEA\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Another third-party source averaged for missing fuel prices.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001244:15:1", "occurrence_id": "2db34b7923bbd41ee7e64850", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  costs and profits but excluding taxes, as discussed below. It includes producer\nsubsidies to the extent to which they are reflected in the supply cost by producers.\n\n  - **VAT rates,** including the standard VAT rate in each country and the reduced rates granted\n\nto specific fuels for a limited set of end-uses. The effective VAT rate on energy consumed\nin the industrial, power generation, and service sectors is zero because these users reclaim\npaid VAT. The output of these sectors, e.g., the electricity produced by the power sector,\ncarries VAT when consumed by households.\n\n<u>Retail prices</u>\n\nAnnual average retail prices are widely available for coal, natural gas, and electricity at an\neconomywide level, as well as disaggregated by main end-user—industrial, residential, and power\ngeneration. However, for other fuels (e.g., gasoline and diesel) only economywide annual average\nretail prices are available (Parry et al, 2021). The dataset from Parry et al., (2021) does not include\nspatially specific fuel prices. Accordingly, retail prices for some fuels are taken from IMF and\nWorld Bank country desk datasets. For cases where such data was not available, they use a simple\naverage across various third-party sources, including Eurostat, IEA, World Bank Doing Business\nIndicators, Global Petrol Prices Retail Energy Price Data, and Enerdata Global Energy & CO2\n\n\n31 When different carbon tax rates are applied in the sector they cover, this paper uses the highest rate in the\ncomputation of the TCP as WB (2022a) does not provide quantitative estimates of emissions covered by each specific\ntax rate.\n\n12\n\nMention: \"World Bank Doing Business\nIndicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its energy price indicators are among those averaged when data are missing.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001244:15:1", "occurrence_id": "a458f8b3e9deea9bce448342", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  costs and profits but excluding taxes, as discussed below. It includes producer\nsubsidies to the extent to which they are reflected in the supply cost by producers.\n\n  - **VAT rates,** including the standard VAT rate in each country and the reduced rates granted\n\nto specific fuels for a limited set of end-uses. The effective VAT rate on energy consumed\nin the industrial, power generation, and service sectors is zero because these users reclaim\npaid VAT. The output of these sectors, e.g., the electricity produced by the power sector,\ncarries VAT when consumed by households.\n\n<u>Retail prices</u>\n\nAnnual average retail prices are widely available for coal, natural gas, and electricity at an\neconomywide level, as well as disaggregated by main end-user—industrial, residential, and power\ngeneration. However, for other fuels (e.g., gasoline and diesel) only economywide annual average\nretail prices are available (Parry et al, 2021). The dataset from Parry et al., (2021) does not include\nspatially specific fuel prices. Accordingly, retail prices for some fuels are taken from IMF and\nWorld Bank country desk datasets. For cases where such data was not available, they use a simple\naverage across various third-party sources, including Eurostat, IEA, World Bank Doing Business\nIndicators, Global Petrol Prices Retail Energy Price Data, and Enerdata Global Energy & CO2\n\n\n31 When different carbon tax rates are applied in the sector they cover, this paper uses the highest rate in the\ncomputation of the TCP as WB (2022a) does not provide quantitative estimates of emissions covered by each specific\ntax rate.\n\n12\n\nMention: \"Global Petrol Prices Retail Energy Price Data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"A commercial retail energy price series used in the simple average.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001244:15:1", "occurrence_id": "e3c2d4feafc882d1aaa53ef0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  costs and profits but excluding taxes, as discussed below. It includes producer\nsubsidies to the extent to which they are reflected in the supply cost by producers.\n\n  - **VAT rates,** including the standard VAT rate in each country and the reduced rates granted\n\nto specific fuels for a limited set of end-uses. The effective VAT rate on energy consumed\nin the industrial, power generation, and service sectors is zero because these users reclaim\npaid VAT. The output of these sectors, e.g., the electricity produced by the power sector,\ncarries VAT when consumed by households.\n\n<u>Retail prices</u>\n\nAnnual average retail prices are widely available for coal, natural gas, and electricity at an\neconomywide level, as well as disaggregated by main end-user—industrial, residential, and power\ngeneration. However, for other fuels (e.g., gasoline and diesel) only economywide annual average\nretail prices are available (Parry et al, 2021). The dataset from Parry et al., (2021) does not include\nspatially specific fuel prices. Accordingly, retail prices for some fuels are taken from IMF and\nWorld Bank country desk datasets. For cases where such data was not available, they use a simple\naverage across various third-party sources, including Eurostat, IEA, World Bank Doing Business\nIndicators, Global Petrol Prices Retail Energy Price Data, and Enerdata Global Energy & CO2\n\n\n31 When different carbon tax rates are applied in the sector they cover, this paper uses the highest rate in the\ncomputation of the TCP as WB (2022a) does not provide quantitative estimates of emissions covered by each specific\ntax rate.\n\n12\n\nMention: \"Enerdata Global Energy & CO2\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"A final commercial source averaged into missing price observations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001245:5:0", "occurrence_id": "1ce521a3b7b67faef6387e6f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: to inclusion in commercial lending? Finally, what are the implementation conditions needed to promote\n\nadditionality and effective targeting of MSME finance operations?\n\n\nTo answer these questions, we use administrative loan-level data from the KUR program itself, which\n\nincludes information on over 8 million debtors between 2015 and 2020. Using the program database as a\n\nsampling frame, we complement the administrative data with surveys of a representative sample of 1,402\n\nKUR borrowers as well as qualitative interviews with separate samples of 100 KUR borrowers and 100 non\nborrowers. In sum, these data sources provide a rich and novel set of insights into the profiles, perspectives,\n\nand experiences of KUR borrowers since 2015 and shed light on the ways in which the publicly supported\n\nscheme plays out in supporting MSMEs on the ground.\n\n\nOur main and novel finding is that the subsidized credit provided through KUR enhances financial inclusion by\n\nreaching previously excluded micro and small businesses; however, it does not serve as a stepping- stone to\n\nunsubsidized commercial lending. Prior to receiving a subsidized KUR loan for the first time, only 11 percent\n\nof borrowers had ever received a commercial loan from a bank. Although successful at reaching first-time\n\nborrowers, the longer-term impacts on their financial inclusion are uncertain because KUR borrowers do\n\nnot move into unsubsidized commercial lending. Approximately half of KUR borrowers went on to receive\n\nKUR loans multiple times within a 5-year period, and only 3 percent of firms who received KUR financing\n\ngraduated to other forms of commercial lending. Until recent regulatory changes, firms had little incentive\n\nto choose commercial sources of financing over KUR’s below-market interest rates, and banks preferred to\n\nencourage repeat lending to minimize risk.\n\n\nWe also identify key design features of the KUR scheme which pose challenges for inclusivity and consumer\n\nprotection and hold important lessons for the design of public MSME finance support schemes. First,\n\nwe find that although the KUR program targets firms with insufficient collateral, in practice, 96% of KUR\n\nborrowers submit collateral for loans,\n\nMention: \"administrative loan-level data from the KUR program\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Over eight million debtors form both the analytical sample and the sampling frame for the borrower surveys.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001249:6:1", "occurrence_id": "51e0aa99a201ed7463ed0dea", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  from the nationally representative Demographic and Health Survey from multiple countries\n\nlinked to additional data such as satellite imagery (Chi et al. 2022). The Demographic and Health\n\nSurvey (DHS) is a series of nationally representative surveys conducted in multiple countries,\n\nincluding South Asian countries.\n\n\n**Firm size in India.** We use the most recent cross-sectional firm-level data from the Sixth\n\nEconomic Census of India, conducted in 2013, which captures information for over 58 million\n\nnon-agricultural firms across India, including employee counts for each firm (Government of India\n\n\n5\n\nMention: \"Demographic and Health\n\nSurvey\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"A cited study links its nationally representative responses to satellite imagery.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001249:6:1", "occurrence_id": "d8467b261f101c47f073719e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  from the nationally representative Demographic and Health Survey from multiple countries\n\nlinked to additional data such as satellite imagery (Chi et al. 2022). The Demographic and Health\n\nSurvey (DHS) is a series of nationally representative surveys conducted in multiple countries,\n\nincluding South Asian countries.\n\n\n**Firm size in India.** We use the most recent cross-sectional firm-level data from the Sixth\n\nEconomic Census of India, conducted in 2013, which captures information for over 58 million\n\nnon-agricultural firms across India, including employee counts for each firm (Government of India\n\n\n5\n\nMention: \"Sixth\n\nEconomic Census of India\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Employee counts for over 58 million non-agricultural firms measure firm size.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001252:13:1", "occurrence_id": "2380c35c5c03fd9ffe352c31", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  which is\n\n\nour preferred model for interpretation. Table 4 shows that the estimated results for per capita\n\n\nexpenditure vary somewhat between model 1 and model 2, which is perhaps due to unobserved\n\n\ntrends occurring in different years. But more importantly, once we control for both the province\n\n\nand year FE, overall, the results remain similar between model 2 and model 3.\n\n\nIn particular, Table 4 shows that, holding all other factors constant, a 1 percent rise in a\n\n\nprovince’s per capita expenditure reduced its poverty rate by 0.26 percentage points (column 3),\n\n\nits poverty gap by 0.10 percentage points (column 6), and its poverty severity index by 0.05\n\n\npercentage points (column 9). Interestingly, holding fixed the per capita expenditure levels, a 1\n\n\n11 We also estimate a variant of equation (1) where we include the explanatory variables in lagged forms using both\nprovince FE and GMM models. The results are qualitatively similar (Dang et al., 2023). We further show the\nestimation results with additional control variables for public infrastructure such as the share of villages with yearround passable roads, the share of villages with a daily market, and the distance from the village to the nearest town\nin Appendix A, Tables A.3 and A.4. Since these variables are computed from the commune module of the VHLSSs,\nwe do not have complete data for these control variables for the whole period of our study. Yet, these tables show that\nprovinces with more villages that have year-round passable roads could reduce poverty and have faster economic\ngrowth.\n\n12\n\nMention: \"VHLSSs\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Its commune module yields road access, market presence and distance controls, though not for the whole period.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001253:20:1", "occurrence_id": "60473217bb429d7ea41d0710", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  is logged per capita monthly\n\nhousehold consumption; in columns (4)-(6) a dummy equaling 1 if the household is situated\n\nbelow the poverty line. Columns (1) and (4) include dummies indicating whether any house\nhold member is engaged in non-farm activities. Columns (2) and (5) disaggregates non-farm\n\nemployment by employment type, while columns (3) and (6) does likewise for employment\n\nsectors. All specifications restrict the sample to rural households, and use household weights.\n\nStandard errors continue to be clustered by district, allowing for arbitrary correlation in con\nsumption/poverty outcomes across households located in the same district.\n\n\nTable 6 shows robust correlations between non-farm employment, education and house\nhold consumption. Column (1) shows that while having a literate member in the household\n\nincreases per capita monthly consumption by almost 7 percent, the presence of a member\n\n\n9 The consumption data obtained through the NSS survey rounds are used to construct state-specific\npoverty line thresholds corresponding to each round. Households with monthly per capita consumption less\nthan this threshold are classified as below the poverty line or BPL.\n\n\n20\n\nMention: \"NSS survey rounds\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"State-specific poverty line thresholds are constructed for each round from the consumption data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001254:28:0", "occurrence_id": "a80ad86800a5cfffbdc8adb5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: high inflation, and an overvalued exchange rate; (ii) a highly distorted incentive structure derived\nfrom an overly protective, arbitrary, and non-transparent trade regime, a multiple exchange rate\nsystem that penalized exports, domestic price controls and administrative impediments, a complex\ntax system, restrictive foreign investment rules, and widespread corruption; and (iii) financial\nrepression and an unstable and inefficient banking sector. Economic reform efforts supported by\nthe IMF and World Bank were half-hearted and short-lived, as popular unrest (notably in April\n1984) or balance of payments crises (notably in June 1987) forced the government to backtrack.\nAccording to the CPIA index, the state was in fragile territory throughout this period. In 1990, the\neconomic situation deteriorated further owing to a pre-electoral loosening of financial policies, a\nsharp deterioration in the terms of trade, and a drought, leading to a spike of year-on-year inflation\nto 80 percent, a widening balance of payments deficit, and external arrears, including to the IMF\nand World Bank. In August 1990, the government launched the “New Economic Program,” a raft\nof reforms including fiscal consolidation, trade reforms to enhance the outward orientation of the\neconomy, unification of the multiple exchange rates and a devaluation, liberalization of interest\nrates, and steps to strengthen the banking system. Refinancing and rescheduling agreements with\nthe Paris Club (1991) and with private creditors (1994) resulted in a halving of public external\ndebt, and a new foreign investment law (1995) led to a surge in foreign direct investment. Growth\naccelerated and inflation fell sharply and stayed in the single digits. International pressure after the\nflawed 1994 elections led to a new government in 1996, which reduced tensions with Cuba,\nintroduced judicial reforms, and signed the Free Trade Agreement for Central America. Although\neconomic volatility continued, the cumulative impact of economic and political reforms led to the\nDominican Republic’s exit from fragility in 1997.\n\n\n**Rwanda\n\nMention: \"CPIA index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its rating is cited to show the country was in fragile territory throughout the period.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001260:52:0", "occurrence_id": "0f0915477f0e24a592168aaf", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Table A.3. Summary statistics of Green and Carbon-intensive occupations in India**\n\n\nSource: Own elaboration based on microdata from the 2019 Periodic Labour Force Survey (PLFS) for\nIndia.\nNotes: The column shows in parenthesis the sample standard deviations of the means and the standard\nerror of the difference in sample means estimated as the robust standard error of the coefficient of a\nregression of the green/carbon-intensive/rest job variable on the group membership dummy variable. (1)\nGreen occupations are those 4-digit ISCO-08 occupations with any of the 44 green qualifications defined\nby the Skill Council for Green Jobs (see Table 0). To estimate the proportion of green jobs for each 3digit ISCO-88 (as specified in the Indian labor force survey), 4-digit ISCO-08 occupations are mapped to\nthe 4-digit ISCO-88 classification using the ILO's crosswalk (when a single ISCO-88 occupation code is\nmapped to more than one ISCO-08, a green proportion measure is used to reflect the proportion of the\ncorresponding ISCO-08 occupations that have been classified as green) and then estimated at the ISCO88 3 digit level assuming an equal distribution of workers in 4-digits occupations within each 3-digit\noccupation. (2) Carbon-intensive occupations are those defined by Vona et al. (2018) at the 6-digits level\nof the 2010-SOC, mapped to the 4-digit ISCO-08 classification using a crosswalk from the US Bureau of\nLabor Statistics (when a single ISCO occupation code is mapped to more than one 6-digit SOC, a carbonintensive proportion measure is used that reflect the proportion of corresponding 6-digit SOC occupations\nthat have been classified as carbon-intensive), converted to ISCO-88 at 4-digits using ILO's crosswalk\nand then estimated at the ISCO-\n\nMention: \"2019 Periodic Labour Force Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Microdata on green and carbon-intensive occupations are summarised from this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001261:14:1", "occurrence_id": "28daba77e1c2ab6b065af9a8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: sup> <sup>_GDPct_</sup> <sup>) +</sup> <sup>_β_</sup> <sup>3</sup> <sup>_kCDiff k_</sup>\n\n\n\n_ijcst_ <sup>_k_</sup> <sup>_× Empijcst_</sup> <sup>+</sup>\n\n\n\n\n- _β_ 1 _kCDiff_ <sup>_k_</sup>\n\n\n\n_ijcst_ <sup>_k_</sup> <sup>+</sup> <sup>_β_</sup> <sup>2</sup> <sup>_kCDiff k_</sup>\n\n\n\n_yijcst_ =\n\n\n\n\n\n\n_k_\n\n\n\n_β_ 4 _kClim_ <sup>_k_</sup> _ijcst_ <sup>_×_</sup> <sup>log(</sup> <sup>_GDPct_</sup> <sup>)</sup> <sup>_× Empijcst_</sup>\n\n\n\n\n+ _δEmpijcst_ + _CHistijcs_ + _γjct_ + _ϵjsc._ (1)\n\n\n\nHere, _CDiffijcst_ is the difference between the relevant climate variable for firm _i_ in year\n\n\n_t_ (when the Enterprise Survey was conducted) relative to that climate variable in firm _i_ ’s\n\n\nlocation from 1980–2008. The variable log( _GDPct_ )\n\nMention: \"Enterprise Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"inform\", \"summary\": \"Its interview year defines when the firm's climate variables are measured relative to the 1980-2008 baseline.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001273:29:0", "occurrence_id": "71c19b01aee805f6cc56af8c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Logistics Performance**\n\n\nFigure A. 6 EU Import shares in 2023 against logistics performance ranking in 2018, (1= highest performer)\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n|Overall logistics performance<br>6<br>5<br>4<br>3<br>2<br>1<br>0<br>0 20 40 60 80 100 120 140 160|Trade & transport infrastructure quality<br>6<br>5<br>4<br>3<br>2<br>1<br>0<br>0 20 40 60 80 100 120 140 160|Ability to track & trace consignments<br>6<br>5<br>4<br>3<br>2<br>1<br>0<br>0 20 40 60 80 100 120 140 160|\n|---|---|---|\n|0<br>1<br>2<br>3<br>4<br>5<br>6<br>0<br>20<br>40<br>60<br>80<br>100 120 140 160<br>Quality of logistics services|0<br>1<br>2<br>3<br>4<br>5<br>6<br>0<br>20<br>40<br>60<br>80<br>100 120 140 160<br>Ease of arranging competitive shipments<br>|0<br>1<br>2<br>3<br>4<br>5<br>6<br>0<br>20<br>40<br>60<br>80<br>100 120 140 160<br>Efficiency of the clearance process|\n\n\n\n_Note_ : Calculations based on data from UN Comtrade and WD\n\nMention: \"data from UN Comtrade and WD\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Import shares are combined with logistics rankings to produce the scatter.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001276:30:0", "occurrence_id": "3c3e4bca55e12c142e7dc339", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 5. Change in preparedness to cope with a disaster or emergency after COVID-19, household\n(%)\n\n\n\n\n\n\n\n_Source:_ World Bank – UNDP LAC High-Frequency Phone Survey, 2021 - Wave 2.\n_Note:_ The sample is restricted to households exposed to at least one risk.\n\n\n28\n\nMention: \"LAC High-Frequency Phone Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its second wave measures changes in disaster preparedness among risk-exposed households.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001277:24:0", "occurrence_id": "debe7f68226edb7377a9485c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Source: World Bank Survey\nNotes: Figure presents the percentage of countries in each of four income categories which allow a type of\ntravel, with or without a permit or license. On average, a higher share of higher income countries allow\ntravel rights compared to middle and low income countries.\n\n\nFigure 9: Prevalence of Rights Across Income Categories\n\n\na result, carriers from the destination country will have lower rates of empty trips, while\n\nother carriers will have elevated rates of empty trips. Even for the remaining non-strategic\n\ncommodities, the IST allows for freight-sharing agreements that enforce a minimum share\n\nof international shipments be carried by carriers from the destination country.\n\n\nEvidence from the taxicab market suggests that price regulations may also generate empty\n\ntrips. Fréchette et al. (2019) consider the market for taxicabs in New York City, where prices\n\nare fixed by regulation. As a result, the market clears on the dimension of waiting time,\n\nrather than price. When demand is low, drivers spend large amounts of time searching for\n\npassengers rather than lowering their prices, and cabs are empty between 30 and 70 percent\n\nof the time. In addition, because taxi drivers drive in discrete shifts, they cannot adjust\n\nto low demand by stopping early, and continue to drive empty. These regulatory frictions\n\ncan also amplify the effects of other frictions. Buchholz (2022) finds that, in the taxi cab\n\ncontext, price regulations exacerbate search and matching frictions by creating artificially\n\nthin markets. Relaxing pricing regulations can achieve a significant share of the welfare gains\n\nfrom eliminating all search frictions and optimally allocating drivers and passengers. On the\n\nother hand, regulations which limit the size of late fees may decrease the number of empty\n\ntrips. OECD (2017) argues that strong pro-carrier regulations in Mexico, which limit the\n\nability of shippers to punish carriers for delivering products late, causes carriers to spend\n\nmore time searching for return loads. By making search time less costly, this decreases the\n\n\n23\n\nMention: \"World Bank Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Country rules on whether travel types require permits are compiled in this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001279:8:1", "occurrence_id": "2b9318708e04cf037dc0ae31", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .\nAt the same time, it rationalized spending not directly linked to the pandemic, such as public wages, to\nbuffer the negative impact on fiscal accounts. In addition, a reform to improve the long-term sustainability\nof the pension system was approved in 2023. As the post-pandemic recovery went underway, Uruguay\nfound itself as the country with the lowest sovereign spreads of the region, and improved credit ratings\nfrom the main credit agencies.\n\n\n**Figure 3: Uruguay's fiscal outcomes**\na) Fiscal Balance of the non-monetary public sector       b) Debt-to-GDP ratio of the non-monetary public sector\nIn percent of GDP                                           In percent of GDP\n\n\n\n\n\n\n\nSource: World Bank database, based on BCU and MEF\n\n\n3 Marandino and Oddone (2022).\n4 See World Bank Group (2022) for more details.\n\n\n7\n\nMention: \"World Bank database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Fiscal balance and debt ratios are compiled from national central bank and finance ministry figures into this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001282:4:0", "occurrence_id": "7d37883da52ba3d040fbcd09", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: collecting qualitative observations from public online spaces in Jordan. First, I analyze hundreds\n\n\nof business marketing posts in Facebook groups and document significant differences in male and\n\n\nfemale users’ posting behavior. Women are significantly more likely than men to conceal parts of their\n\n\nidentity by using full names and real pictures on the profiles from which they market their businesses.\n\n\nSecond, I document the existence of dozens of regional and national private women-only Facebook\n\n\ntrading groups in Jordan, with thousands of members in each. These gender-segregated groups are\n\n\noften used as ‘safe spaces’ for female entreprenerus to interact and transact without male presence.\n\n\nTheir popularity is consistent with existing evidence on female entrepreneurs generally segregating\n\n\ninto female-dominated industries with less interactions with men (Ashraf, Delfino, and Glaeser 2019).\n\n\nTogether, these observations suggest that cultural constraints on women’s visibility and interactions\n\n\nare often mirrored online, but also that female entrepreneurs use digital platforms’ features to operate\n\n\n_within_ these constraints, potentially at the expense of their businesses’ visibility.\n\n\nBased on these findings, I set up an experiment aiming to increase businesses’ online exposure\n\n\nwithout compromising owners’ visibility. This is possible by setting up public Facebook business\n\n\npages, which allow marketing a business without publicizing its owner’s identity. To understand the\n\n\nimpact of online markets on low-mobility women, including those with limited digital literacy, I also\n\n\ndesigned the intervention to lift informational, logistical, and financial barriers to technology adoption.\n\n\nParticularly, I work with 1,122 female entrepreneurs recruited from a database of microfinance clients,\n\n\nand I randomize their access to logistical support in setting up virtual storefronts for their businesses\n\n\non Facebook. The support offered creating a public Facebook business page, marketing it for up to 6\n\n\nmonths, and covering paid advertisements of up to $20 for active pages. The support package therefore\n\n\nprovided a digital marketing outsourcing service for the treatment group. To help treated participants\n\n\nmaintain their online presence, the\n\nMention: \"database of microfinance clients\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"1,122 female entrepreneurs were recruited from this client list for the randomised trial.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001291:20:1", "occurrence_id": "cab16ff3a9075a7052f4bb02", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  homicide rate and women's labor participation. These coefficients\n\nhighlight the limited impact of changes in the male homicide rate on women's labor participation from\n\n1960 to 1989.\n\n\nDuring the 1990s and the early 2000s, a positive relationship existed between these two variables. As\n\nthe violence related to drug trafficking escalated in Colombia, resulting in the murder of many men,\n\nwomen began to enter the labor market at a higher pace (Iregui-Bohórquez et al., 2021). This increase\n\nin women's labor participation can be attributed to the need to offset income losses and the demand\n\nfor women's skills in urban labor markets (Calderón, Gáfaro, and Ibáñez, 2011); hence, their\n\nparticipation showed a slight improvement compared to men. The employment rate also displayed a\n\npositive trend, reaching approximately 50% by the end of the period, indicating a modest\n\n\n10 Data from Global Burden of Disease Collaborative Network. Global Burden of Disease Study 2019 (GBD\n2019) Results. Seattle, United States: Institute for Health Metrics and Evaluation (IHME), 2021.\n<u>[https://ourworldindata.org/homicides.](https://ourworldindata.org/homicides)</u>\n\n19\n\nMention: \"Global Burden of Disease Study 2019\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Homicide rates used alongside labour participation come from this study's results.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001293:18:0", "occurrence_id": "c67424897f060df45919f53b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Income and consumption index:_ includes the logarithm of monthly household income and\n\n\nconsumption. Because these variables are continuous the index is constructed following\n\n\nthe methodology outlined in Kling, Liebman and Katz (2007).\n\n\n_Labor market index:_ includes a dichotomous variable for employment, and the logarithm\n\n\nof monthly wages and weekly hours worked.\n\n\nDescriptive statistics for each of the outcomes and their components variables are de\n\nscribed in Table 1.\n\n\n**III.B** **Data for the Taliban’s Geographical Presence**\n\n\nTo obtain information on the Taliban’s geographical presence, we scraped online data\n\n\nfrom the Foundation for Defense of Democracies’ Long War Journal on the districts con\ntrolled by Taliban, by U.S. Allies, and contested territories between the two groups. <sup>9</sup> The\n\n\ndata includes a yearly evolution of the territorial control in Afghanistan. The classification\n\n\nis based on open-source information from NATO’s data in Afghanistan, press reports, in\n\nformation provided by government agencies, and the Taliban. To categorize the districts,\n\n\nthey identify who provides government services, ensures security within the district, ad\n\nministers the district openly, and oversees local courts (see Appendix B.B for a detail\n\n\ndescription of the variable’s construction). <sup>10</sup>\n\n\nFigures 3 to 5 report province-yearly changes in the share of districts controlled by the\n\n\nTaliban, the U.S. allies, and the contested districts in dispute between the Taliban and\n\nthe U.S. allies. <sup>11</sup> As seen in the figures, Taliban control was relatively stable from 2017\n\n\nthrough 2019, after which drastic shifts took place across regions, in parallel with the U.S.\n\n\nwithdrawal from Afghanistan.\n\n\n9The Foundation for Defense of Democracies is Washington, DC-based nonpartisan research institute\nfocusing on national security and foreign policy. It aims to strengthen US national security\n\nMention: \"NATO’s data in Afghanistan\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Open-source NATO information helps classify yearly district territorial control.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001295:22:1", "occurrence_id": "06cb21438c8cfe388282ce26", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  (0.087) (0.023) (0.021) (0.009) (0.010) (0.005)\nShare of population with a primary 0.021 0.018** -0.006 0.000 -0.001 -0.000\neducation or less (0.023) (0.008) (0.007) (0.004) (0.003) (0.002)\nGRP per capita (constant 2010 IDR) 0.402* 0.113 0.037 -0.037 0.030 -0.060**\n(0.239) (0.142) (0.102) (0.100) (0.035) (0.026)\nGrowth in GRP per capita -0.009 -0.014 -0.006 -0.001 0.000 0.002\n<u>(0.013)</u> <u>(0.009)</u> <u>(0.011)</u> <u>(0.006)</u> <u>(0.001)</u> <u>(0.002)</u>\nObservations 271 272 272 272 272 272\nR-squared 0.589 0.489 0.837 0.561 0.393 0.837\n<u>Year dummies</u> <u>Yes</u> <u>Yes</u> <u>Yes</u> <u>Yes</u> <u>Yes</u> <u>Yes</u>\nSource: Author’s calculations using Sakernas data\nNotes: Robust standard errors in parentheses. ***p<0.01, **p<0.05, *p<0.1\n\n\n\n21\n\nMention: \"Sakernas data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"District-level regressions with year dummies are estimated from this labour force survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001297:7:1", "occurrence_id": "b2fcf8460e4ae33e9002d3f8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .).\n\n\nAs discussed in the Introduction, we estimate the change in well-being between two\n\n\nnationally representative surveys conducted during the last resource boom. Specifically,\n\n\nwe estimate the change in monetary poverty and access to several essential services using\n\n\nthe 2009 Household Income and Expenditure Survey (HIES) and the 2016-2018 DHS.\n\n\nFigure 1 reports both the GDP per capita in current US $ over this time period and the\n\n\nreal per capita GDP growth rates in each year. <sup>5</sup> The figure illustrates that per capita\n\n\nGDP in constant US $ increased by over 30 percent between 2010 and 2017 and that the\n\n\naverage annual growth in real per capita GDP was 3.3 percentage points. And in a subset\n\n\n5The GDP figures are obtained from the World Development Indicators; and the real per capita\ngrowth rates are obtained from the World Bank’s 2023 Macroeconomic and Poverty Outlook.\n\n\n6\n\nMention: \"2009 Household Income and Expenditure Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Monetary poverty and access to essential services at the start of the boom are measured here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001297:7:1", "occurrence_id": "e8a88fd5b4c614e91d172247", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .).\n\n\nAs discussed in the Introduction, we estimate the change in well-being between two\n\n\nnationally representative surveys conducted during the last resource boom. Specifically,\n\n\nwe estimate the change in monetary poverty and access to several essential services using\n\n\nthe 2009 Household Income and Expenditure Survey (HIES) and the 2016-2018 DHS.\n\n\nFigure 1 reports both the GDP per capita in current US $ over this time period and the\n\n\nreal per capita GDP growth rates in each year. <sup>5</sup> The figure illustrates that per capita\n\n\nGDP in constant US $ increased by over 30 percent between 2010 and 2017 and that the\n\n\naverage annual growth in real per capita GDP was 3.3 percentage points. And in a subset\n\n\n5The GDP figures are obtained from the World Development Indicators; and the real per capita\ngrowth rates are obtained from the World Bank’s 2023 Macroeconomic and Poverty Outlook.\n\n\n6\n\nMention: \"2016-2018 DHS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The later round supplies the end-point values for poverty and service access.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001297:7:1", "occurrence_id": "c1b414d75e4dae478a9cdc87", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .).\n\n\nAs discussed in the Introduction, we estimate the change in well-being between two\n\n\nnationally representative surveys conducted during the last resource boom. Specifically,\n\n\nwe estimate the change in monetary poverty and access to several essential services using\n\n\nthe 2009 Household Income and Expenditure Survey (HIES) and the 2016-2018 DHS.\n\n\nFigure 1 reports both the GDP per capita in current US $ over this time period and the\n\n\nreal per capita GDP growth rates in each year. <sup>5</sup> The figure illustrates that per capita\n\n\nGDP in constant US $ increased by over 30 percent between 2010 and 2017 and that the\n\n\naverage annual growth in real per capita GDP was 3.3 percentage points. And in a subset\n\n\n5The GDP figures are obtained from the World Development Indicators; and the real per capita\ngrowth rates are obtained from the World Bank’s 2023 Macroeconomic and Poverty Outlook.\n\n\n6\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"GDP per capita in current dollars is taken from this series to frame the growth narrative.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001262:28:1", "occurrence_id": "b214dd19761f15c77368cafc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: *\n**or improving take-up rates, might impact social assistance programs' performance and effectiveness in**\n**alleviating poverty and reducing inequality in Bulgaria** . It allows policymakers to assess the potential\noutcomes of these changes and make informed decisions regarding the design and implementation of\nsocial policies.\n\n\n_Policy Scenario 1: GMI increase and constant take-up._\n**In this policy scenario, we undertake a simulation consistent with the reform of the Monthly Social**\n**Allowance (MSA) program.** This implies an increase in the GMI base from its previous value of 75\nBulgarian Leva (BGN) to 30% of the at-risk-of-poverty (AROP) poverty line. This adjustment assumes that\nthe estimated take-up rate from survey data remains constant. This scenario is not limited to the MSA\nalone. It also considers its interaction with the heating allowance, which is instrumental in addressing\nseasonal affordability challenges, particularly those related to heating expenses. With the higher\nthreshold, more households would now meet the criteria to access these benefits, based on data from\nthe 2021 Survey on Income and Living Conditions (SILC). We assumed that the take-up rates would remain\nconsistent with our earlier estimates of 56% for MSA and 99% for the heating allowance.\n\n**Under this government reform, the increase in the MSA's base value results in a tangible expansion of**\n**its coverage.** This coverage now extends up to the second decile, encompassing a more significant\nproportion of the population in need (Figure 9, panel a). Additionally, the financial support provided\nthrough these programs witnessed a notable uptick. According to our estimates, the number of\nhouseholds benefiting from these schemes would rise from 24,000 to 84,000, and the total benefits\ndisbursed would increase substantially, reaching nearly BGN 230 million.\n\n**However, it is important to\n\nMention: \"2021 Survey on Income and Living Conditions\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"A higher eligibility threshold is simulated against this survey to count newly qualifying households.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001298:9:0", "occurrence_id": "decea0bd0cf6940c9f64434a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: working- age adult level of $ 80,000 to $ 40,000 is associated with an approximate decline\n\n\nin the time available to top managers of about 6.2 percentage points.\n\n\nFigure 1: Relationship between the Time Tax and RGDP per Working Age Pop.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nNotes: This figure shows the relationship between the log of 1 minus the time tax and RGDP\nper working age adult. Each dot represents a country at a survey year. The solid line is the\nsimple weighted regression line where dependent variable is the log of one minus the time tax\nand the independent variable is the RGDP per working-age adult. The countries are weighted\naccording to their employment sizes. Sources: WBES and WDI.\n\n###### **2.1 General Characterization**\n\n\nWe now attempt a more general empirical characterization of the relationship between the\n\n\n(net) time available to managers and the level of development. We estimate the following\n\n\nrelationship:\n\n\nlog(1 _−_ _ti,c,t_ ) = _α_ + _β_ log _GDP c,t_ + _F_ ( _i, c, t_ ) + _SFEi_ + _ui,c,t_ (1)\n\n\nwhere 1 _−_ _ti,c,t_ stands for 1 minus the time tax at establishment _i_ in country _c_ at year\n\n\n_t._ _GDP c,t_ stands for the log of RGDP per working-adult in country _c_ at year _t_, _SFEi_\n\n\nis a dummy variable associated to the two digit ISIC 3.1 sector, and _F_ ( _i, c, t_ ) stands\n\n\nfor a set of establishment level characteristics for establishment _i_ in country _c_ at year _t_ .\n\n\nWe control for the experience of the top manager (years), the number of permanent\n\nMention: \"WBES\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Manager time available is regressed on development level using these firm data, weighted by employment size.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001299:11:1", "occurrence_id": "11484ad73fc9b5afb61eae5a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  <sup>15</sup> [^15: This is denoted as the layer of ownership. For instance, a firm linked to the state through another BOS has 2 layers or degree of connection to the\nstate.]\n\n\n_Table 5. Degree of state ownership – WB BOS database_\n\n\n\n\n\n\n\n|Degree of State<br>participation|Definition|\n|---|---|\n|Minority|Firms in which the state holds 10-24.9% of the stakes|\n|Blocking minority|Firms in which the state holds 25-49.9% of the stakes|\n|Majority ownership|Firms in which the state holds 50-99.9% of the stakes|\n|Full ownership|Firms in which the state holds 100% of the stakes|\n|Direct|Firms directly owned by a Public Authority (national or subnational), which holds at<br>least 10% participation.|\n|Indirect|Firms participated by the government through another firm or chain of firms owned<br>by the State where it holds at least 10% participation.|\n\n\nSource: Author’s elaboration based on World Bank (BOS) database and (Dall'Olio, et al., 2022).\n\n#### **5. EMPIRICAL METHODOLOGY**\n\n\nThis paper explores two main econometric exercises. First, it estimates the conditional correlation\nof the state participation and the performance differentials between BOSs and POEs. For this\npurpose, we analyze the role of state ownership on firm performance through the following\nspecification:\n\n\n(1) 𝑌𝑖𝑡= 𝛼+ 𝛽1𝐵𝑂𝑆𝑖 + 𝛽2𝑿𝑖𝑡 + 𝜕𝑐 + 𝜕𝑠 + 𝜕𝑡 + 𝜀𝑖𝑗𝑡\n\n\nWhere, 𝑌𝑖 denotes the performance indicator for the firm 𝑖 using several performance\nmeasurements to have a comprehensive overview of the soundness of BOSs vis-à-vis POEs. First,\nwe assess the operational performance in terms of productivity, followed by the financial\nperformance measured by indicators of profitability, liquidity, and returns on investment. Thus,\n\n\n14 Alternative\n\nMention: \"World Bank (BOS) database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Direct and indirect state ownership through chains of firms with at least ten percent participation is identified here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001300:10:4", "occurrence_id": "9e03d1ec115fc88fca2b547b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: >31.4<br>16.4<br>30.2<br>7.2<br>4.3<br>69.8<br>88.9<br>90.7<br>45.8<br>6.0<br>9.3<br>31.4<br>38.4<br>15.6<br>6.3<br>61.6<br>5.7<br>8.4<br>1.7<br>0.8<br>91.6<br>46.2<br>55.6<br>23.5<br>9.3<br>44.4<br>|\n\n\n_expressed in percent of the poverty line (****) Survey data for Lebanon is available from the 2022_\n_Vulnerability Assessment Panel. There are large variations in refugee poverty estimates for Lebanon_\n_ranging from 1 percent using the LMIC poverty line (PPP$ 3.65) and 16 percent using the UMIC poverty line_\n_(PPP$ 6.85) to 40 percent using the national poverty line. As Lebanon became a LMIC in July 2022, the first_\n_estimate should apply, even though, since Lebanon only recently became a LMIC a case could be made for_\n_using the UMIC poverty line._ <sup>_4_</sup>\n\n\n4 Yet even the poverty estimate derived for the UMIC line is on the low side, certainly in comparison to other\ncountries in the region. Lebanon’s low PPP conversion factor, which leads to low poverty lines, is behind this. For\ninstance, the UMIC poverty line is only 38 percent of the national poverty line, whereas it is about 70 percent of the\n\n\n9\n\nMention: \"Survey data for Lebanon\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Refugee poverty rates expressed relative to different poverty lines are computed from these data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001300:10:4", "occurrence_id": "f4b71a583f678ee5fb73033e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: >31.4<br>16.4<br>30.2<br>7.2<br>4.3<br>69.8<br>88.9<br>90.7<br>45.8<br>6.0<br>9.3<br>31.4<br>38.4<br>15.6<br>6.3<br>61.6<br>5.7<br>8.4<br>1.7<br>0.8<br>91.6<br>46.2<br>55.6<br>23.5<br>9.3<br>44.4<br>|\n\n\n_expressed in percent of the poverty line (****) Survey data for Lebanon is available from the 2022_\n_Vulnerability Assessment Panel. There are large variations in refugee poverty estimates for Lebanon_\n_ranging from 1 percent using the LMIC poverty line (PPP$ 3.65) and 16 percent using the UMIC poverty line_\n_(PPP$ 6.85) to 40 percent using the national poverty line. As Lebanon became a LMIC in July 2022, the first_\n_estimate should apply, even though, since Lebanon only recently became a LMIC a case could be made for_\n_using the UMIC poverty line._ <sup>_4_</sup>\n\n\n4 Yet even the poverty estimate derived for the UMIC line is on the low side, certainly in comparison to other\ncountries in the region. Lebanon’s low PPP conversion factor, which leads to low poverty lines, is behind this. For\ninstance, the UMIC poverty line is only 38 percent of the national poverty line, whereas it is about 70 percent of the\n\n\n9\n\nMention: \"2022_\n_Vulnerability Assessment Panel\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The named panel round is identified as the source behind the wide range of refugee poverty estimates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001301:14:0", "occurrence_id": "6e800ae8a8739a5719784ee1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Panel C. New Normal (2015-2019)\n\n\nSource: based on World Bank national accounts data. Note: dots in red correspond to Bottom Billion countries, dots in\nblue to countries classified as Emerging Markets, and dots in green to OECD members. The line corresponds to the regression\nline equation indicated in the legend, where ‘gdp_pc_gr’ is the average GDP per capita growth of the decade and ‘log_gdp_pc’\nis the GDP per capita in log form in the decade’s last year.\n\n**<mark>What happened during the Golden Decade?</mark>**\n\nDuring the golden decade, from 2004 to 2014, the Bottom Billion and the world experienced an\nunprecedented economic boom, driven by China's growth and its insatiable demand for raw\nmaterials. This led to high prices and a sustained surge in natural resource exports. Growth\neven among the bottom billion surged.\n\nThis period provided the Bottom Billion with an exceptional opportunity to accelerate growth\nthrough commodity exports and increased capital market liquidity. These favorable conditions\noffered lower-income countries the chance to finally embark on economic transformation. This\nGolden Decade marked a departure from the Old Normal.\n\nDuring the Golden Decade, commodity prices, such as those for oil and gas, reached historical\nhighs (Figure 10), and government revenues from natural resources in resource-rich countries\n\n\n13\n\nMention: \"World Bank national accounts data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Growth comparisons across low-income, emerging and OECD countries are drawn from these accounts.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001303:22:0", "occurrence_id": "455bd8a8a7a99c4919bb8a39", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Data availability statement\n\n\nThe data underlying this article are available in the article and in its online supplementary\nmaterial. Data availalbe from the ICRG database were provided under licence and cannot be\nshared.\n\n\nReferences\n\n\nAlesina, Alberto and Roberto Perotti. 1995. “Fiscal Expansions and Adjustments in OECD\n\nCountries.” Economic Policy 10: 205-248.\nAvom, Désiré, Brice Kamguia, and Henri Njangang. 2021. “Understand Growth Episodes in Sub\nSaharan Africa: Do Exogenous Shocks Matters?” The Journal of International Trade &\nEconomic Development 30: 596–624.\nAiyar, Shekhar Shekhar, Romain Duval, Damien Puy, Yiqun Wu, and Longmei Zhang. 2018.\n\n“Growth Slowdowns and the Middle-income Trap.” Japan and the World Economy 48:\n22-37.\nBarro, Robert and Xavier Sala-i-Martin. 1991. “Convergence Across States and Regions.”\n\nBrookings Papers on Economic Activity 22(1): 107    - 182.\nBarro, Robert and Xavier Sala-i-Martin. 1992. “Convergence.” Journal of Political Economy\n\n100(2): 223-251.\nBen-David Dan, and David H. Papell. 1998. “Slowdowns and Meltdowns: Postwar Growth\n\nEvidence from 74 Countries.” Review of Economics and Statistics 80(4): 561–571.\nBerg, Andrew, Jonathan Ostry, and Jeromin Zettelmeyer. 2012. “What Makes Growth\n\nSustained?” Journal of Development Economics 98 (149–166).\nBosworth, Barry, and Susan Collins. 2008. “Accounting for Growth: Comparing China and\n\nIndia.” Journal of Economic Perspectives 22: 45–66.\nCaselli, Franceso, Gerardo Esquivel, and Fernando Lefort. 1996. “Reopening the Convergence\n\nDebate: A New Look at Cross-\n\nMention: \"ICRG database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"inform\", \"summary\": \"Licence terms prevent the underlying data from being shared with readers.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001305:16:2", "occurrence_id": "a7696d749666bcfb2b636cb5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: . We\nuse the commands feols and sunab from the R package fixest to perform the estimation\nprocedure [94]. We present all relative period estimates in the SI figures. The Average\nTreatment effect on the Treated (ATT) estimates shown in the main text tables are\nthe weighted averages over the post-transference relative treatment period coefficients\n(0 _≤_ _ℓ_ _≤_ 10), where the weights for each relative period coefficient are based on the\nproportion of the treated group (areas managed by AP) in the overall population\nduring that specific relative period [49].\n\nOur control group comprises protected areas that AP has determined meet their\nestablished criteria for potential future management, which AP refers to as “anchor\nareas.” Polygons demarcating the boundaries of AP’s anchor areas are publicly available [51]. We identify individual protected areas in our control group in three steps.\nFirst, we used Quantum GIS to manually select all polygons from each of the World\nDatabase of Protected Areas’ (WDPA) shapefiles that overlapped with the boundaries of AP’s anchor areas. Second, we eliminated duplicate entries from the selected\nWDPA boundaries. Finally, we manually validated the remaining WDPA boundaries\nto confirm that protected areas in our control group match the set of polygons displayed on AP’s map of anchor areas. In three cases, we used QGIS and georeferenced\npolygons from either UNESCO or the literature [95, 96] to manually create shapefiles\nfor AP anchor areas that did not have shapefiles in the WDPA.\n\nSeveral protected areas in our control group are partly managed by governments\nand partly managed by an NGO other than AP. We retain these protected areas in\nour control group because they do not employ a “delegated” management model like\nAP, where the NGO has full control over management decisions [29]. Several of AP’s\n\n\n15\n\nMention: \"World\nDatabase of Protected Areas\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"Polygons overlapping the intervention areas are selected, de-duplicated and manually validated to build the control group.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001312:43:1", "occurrence_id": "f72c1a61e8243941fbe02dd6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <br>**consumption**<br>|Proportion of the population who<br>consumed no vegetables or fruits during<br>the previous day or night.<br>|World Gallup Poll<br>(Global Diet Quality<br>Project)<br>|2021-2022<br>|N=55<br>countries<br>|<br> <br>|\n|**Soft drink**<br>**consumption**<br>|<br>Proportion of the population who<br>consumed a sugar-sweetened soft drink<br>during the previous day or night. Sugar-<br>sweetened soft drinks include soda,<br>energy drinks, and sports drinks.<br>|<br>World Gallup Poll<br>(Global Diet Quality<br>Project)|2021-2022|N=55<br>countries||\n\n\n\n42\n\nMention: \"World Gallup Poll\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its diet quality module measures the share consuming no vegetables or fruit the previous day.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001313:21:1", "occurrence_id": "0e57513f783b2067401d0e01", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 0.06%|0.06%|0.06%|\n\n\n\nSource: Estimates based on POF 2017/2018; Global Burden of Disease 2019; and SIH/SUS. Estimates assume a price shock of 20%.\n\n\nFigure 5 depicts the progressive pattern of the long-term estimated medical expenditure changes after\nimplementing a taxation on processed and ultra-processed foods. The lowest consumption decile presents\nthe highest gains, with estimated changes in medical expenditure being almost twice as large as the\nsecond decile, three times as large as the third, and 113 times as large as the top consumption decile.\nWhile our analysis does not allow us to estimate fiscal impacts on SUS, the public health care system, it\nsuggests that a taxation on processed and ultra-processed foods would have a positive impact in terms of\nreduced demands and medical expenditure.\n\n_Figure 5_ - Changes in medical expenditure\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n20\n\nMention: \"Global Burden of Disease 2019\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Disease burden parameters enter the estimated medical expenditure response to a price shock.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001316:4:1", "occurrence_id": "8492ad8a29583aad13601317", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  5 bandwidth of a multi-dimensional poverty score cut\n\n\nof Punjab, Pakistan. All surveys were conducted by female enumerators. <sup>2</sup> Respondents belong\n\nto poor households which were within a +/- 5 bandwidth of a multi-dimensional poverty score cut\noff used to determine eligibility for Pakistan’s national unconditional cash transfers program. All\n\n\n\nto poor households which were within a +/- 5 bandwidth of a multi-dimensional poverty score cut\noff used to determine eligibility for Pakistan’s national unconditional cash transfers program. All\n\nrespondents are under the age of 60 years and the average age is 48 years. The vast majority of\n\n\n\noff used to determine eligibility for Pakistan’s national unconditional cash transfers program. All\n\nrespondents are under the age of 60 years and the average age is 48 years. The vast majority of\n\nthese women (93%) are illiterate. We relied on questions from the standard IPV module of the\n\n\n\nrespondents are under the age of 60 years and the average age is 48 years. The vast majority of\n\nthese women (93%) are illiterate. We relied on questions from the standard IPV module of the\n\nDemographic and Health Surveys (DHS), and asked 7 questions on experience of physical violence,\n\n\n\nthese women (93%) are illiterate. We relied on questions from the standard IPV module of the\n\nDemographic and Health Surveys (DHS), and asked 7 questions on experience of physical violence,\n\n4 on sexual violence and 4 related to injuries due to violence.\n\n\n\n4 on sexual violence and 4 related to injuries due to violence.\n\n\n\n4 on sexual violence and 4 related to injuries due to violence.\n\nThe literature so far has focused on comparing the measurement of IPV using ACASI with\n\nF2F interviews by randomizing respondents to _either_ answer questions using ACASI or F2F surveys\n\n\n\nThe literature so far has focused on comparing the measurement of IPV using ACASI with\n\nF2F interviews by\n\nMention: \"Demographic and Health Surveys\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Its standard intimate partner violence module supplies the question wording adopted in the study.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001317:37:0", "occurrence_id": "682ed7983990de10ecc5c96f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ##### **C Wealth, liquidity and drilling**\n\nUsing the 2010 household survey, we compute the total value of household assets in 2009, including\n\nagricultural land, livestock, agricultural machinery, household durable goods, and savings in the form\n\nof bank deposits, cash and jewelry. Gross wealth should be a good proxy for liquidity and thus allows\n\nus to test for the importance of financial constraints on well-drilling. Table C.1 presents estimates of\n\nthe determinants of drilling attempts using the 2012-16 representative plot panel. Columns 1-3 report\n\nlinear probability models with standard errors clustered at the plot level whereas columns 4-6 report\n\nplot-level random effects logit models. Inferences about the association between pre-sample wealth and\n\ndrilling over the 5-year panel from the two sets of estimates are virtually identical.\n\n\nIn column 1 (and 4), which includes only mandal dummies as additional controls, we see that\n\nwealthier households (in 2009) are significantly more likely to drill during 2012-2016. However, once\n\nwe also control for plot size with area quintile dummies (cols. 2 and 5), this effect largely disappears.\n\nEvidently, wealth and plot area are positively correlated and there is a greater incentive to drill on\n\nlarger plots. Finally, in column 3 (and 6), we control for the initial number of functioning borewells on\n\nthe plot entering the panel period, which further attenuates the wealth effect toward zero. Admittedly,\n\nthese liquidity effects, or lack thereof, are associations, i.e., not necessarily causal. For example, despite\n\ncontrolling for initial borewells, unobserved (to us) suitability for groundwater development (and thus\n\ndrilling) may have augmented pre-sample wealth through past groundwater use. Such correlations,\n\nhowever, would bias liquidity effects away from zero, contrary to what we find in Table C.1.\n\n\nFurther, according to our 2017 survey of borewell owners,\n\nMention: \"2010 household survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Total pre-sample household assets including land, livestock, machinery and savings are valued from this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001317:37:0", "occurrence_id": "8a02c07c80ce2d1064661c83", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ##### **C Wealth, liquidity and drilling**\n\nUsing the 2010 household survey, we compute the total value of household assets in 2009, including\n\nagricultural land, livestock, agricultural machinery, household durable goods, and savings in the form\n\nof bank deposits, cash and jewelry. Gross wealth should be a good proxy for liquidity and thus allows\n\nus to test for the importance of financial constraints on well-drilling. Table C.1 presents estimates of\n\nthe determinants of drilling attempts using the 2012-16 representative plot panel. Columns 1-3 report\n\nlinear probability models with standard errors clustered at the plot level whereas columns 4-6 report\n\nplot-level random effects logit models. Inferences about the association between pre-sample wealth and\n\ndrilling over the 5-year panel from the two sets of estimates are virtually identical.\n\n\nIn column 1 (and 4), which includes only mandal dummies as additional controls, we see that\n\nwealthier households (in 2009) are significantly more likely to drill during 2012-2016. However, once\n\nwe also control for plot size with area quintile dummies (cols. 2 and 5), this effect largely disappears.\n\nEvidently, wealth and plot area are positively correlated and there is a greater incentive to drill on\n\nlarger plots. Finally, in column 3 (and 6), we control for the initial number of functioning borewells on\n\nthe plot entering the panel period, which further attenuates the wealth effect toward zero. Admittedly,\n\nthese liquidity effects, or lack thereof, are associations, i.e., not necessarily causal. For example, despite\n\ncontrolling for initial borewells, unobserved (to us) suitability for groundwater development (and thus\n\ndrilling) may have augmented pre-sample wealth through past groundwater use. Such correlations,\n\nhowever, would bias liquidity effects away from zero, contrary to what we find in Table C.1.\n\n\nFurther, according to our 2017 survey of borewell owners,\n\nMention: \"2012-16 representative plot panel\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Plot-level models test whether pre-sample wealth predicts well-drilling attempts.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001317:37:0", "occurrence_id": "e20a8c5791bc829db0d5db18", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ##### **C Wealth, liquidity and drilling**\n\nUsing the 2010 household survey, we compute the total value of household assets in 2009, including\n\nagricultural land, livestock, agricultural machinery, household durable goods, and savings in the form\n\nof bank deposits, cash and jewelry. Gross wealth should be a good proxy for liquidity and thus allows\n\nus to test for the importance of financial constraints on well-drilling. Table C.1 presents estimates of\n\nthe determinants of drilling attempts using the 2012-16 representative plot panel. Columns 1-3 report\n\nlinear probability models with standard errors clustered at the plot level whereas columns 4-6 report\n\nplot-level random effects logit models. Inferences about the association between pre-sample wealth and\n\ndrilling over the 5-year panel from the two sets of estimates are virtually identical.\n\n\nIn column 1 (and 4), which includes only mandal dummies as additional controls, we see that\n\nwealthier households (in 2009) are significantly more likely to drill during 2012-2016. However, once\n\nwe also control for plot size with area quintile dummies (cols. 2 and 5), this effect largely disappears.\n\nEvidently, wealth and plot area are positively correlated and there is a greater incentive to drill on\n\nlarger plots. Finally, in column 3 (and 6), we control for the initial number of functioning borewells on\n\nthe plot entering the panel period, which further attenuates the wealth effect toward zero. Admittedly,\n\nthese liquidity effects, or lack thereof, are associations, i.e., not necessarily causal. For example, despite\n\ncontrolling for initial borewells, unobserved (to us) suitability for groundwater development (and thus\n\ndrilling) may have augmented pre-sample wealth through past groundwater use. Such correlations,\n\nhowever, would bias liquidity effects away from zero, contrary to what we find in Table C.1.\n\n\nFurther, according to our 2017 survey of borewell owners,\n\nMention: \"2017 survey of borewell owners\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Owner-reported information is used to discuss reverse causality from groundwater development.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001319:37:1", "occurrence_id": "d48f2be5ad44e41a94d44917", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: (𝑠𝑠−4); 𝑖 **𝑖** 4 < 𝑠𝑠≤8\n0.134 ∙4 + 0.101 ∙4 + 0.068 ∙(𝑠𝑠−8); 𝑖 **𝑖** 𝑠𝑠> 8\n\n\n\n**𝑒**\n\n𝑓𝑓(𝑠𝑠) = �\n\n**𝑖**\n\n\n\n**𝑖**\n\nIf the wage-schooling relationship is log-linear, the relationship between h and s should also be log-linear. **𝑖**\nThe PWT10.1 constructs the h index for 150 countries using data on schooling years from Barro and Lee\n(2013) for 95 countries, and from Cohen and Soto (2007) and Cohen and Leker (2014) for an additional 55\ncountries. International data on education-wage profiles suggest that the return to an additional year of\neducation in Sub-Saharan Africa, the region with the lowest level of education, is about 13.4 percent. In\ncomparison, the return to an additional year of education is 10.1 percent for the world and 6.8 percent for the\nOECD countries (Psacharopoulos 1994). This measure of human capital tries to reconcile the properties of a\nlog-linear relationship between education and income at the country level with the concavity of that\nrelationship across countries. The h index from the PWT assumes homogeneous returns across countries.\n\n_Labor share of income_ . Finally, the PWT has estimated the labor share (or the share of labor income in\neconomic activity) for a wide array of countries and years. There is broad availability of information on labor\ncompensation of employees; however, a separate estimation is needed for the labor compensation of selfemployed workers. The cross-country estimates of the labor share yield some stylized facts (Feenstra, Inklaar,\nand Timmer 2015): (a) the global average of the labor share\n\nMention: \"PWT10.1\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its human capital index for 150 countries is built from these schooling series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001321:11:0", "occurrence_id": "67eedb9bc072842c66f3cf2d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: electricity connections per 1000 people. <sup>15</sup> Figure 3 shows the temporal trends in electrification\n\n\nat the municipality level.\n\n\nNext, to understand consumption patterns, we use data on electricity purchases to con\n\nstruct customer-level panel data on electricity consumption (purchases). Specifically, for each\n\n\nquarter-year, we aggregate the data on the frequency of electricity purchases, quantity (kWh)\n\n\nof electricity, quantity (kWh) per transaction, and amount (monetary value in RWF) of elec\n\ntricity purchased. Further, we complement our analysis with spatial data showing the location\n\n\nof the universe of all electricity transformers, as well as the geographic footprints of low and\n\nmedium-voltage grid lines in Rwanda. <sup>16</sup> This data is used to examine whether coffee-growing\n\n\nmunicipalities are disproportionately favored in the rollout of grid infrastructure.\n\n\n_Coffee Data_\n\n\nTo measure the intensity of coffee cultivation at baseline, we rely on data compiled by Guariso\n\n\nand Verpoorten (2018) from the 1999 Rwandan Coffee Census. The data contains detailed in\n\nformation on the coffee-growing municipalities, including the number of coffee trees in several\n\n\nage groups ( _<=_ 3, 3 _−_ 10, 10 _−_ 30, and _≥_ 30 years) per municipality. Using these data, for each mu\nnicipality, we compute the total number of coffee trees per km <sup>2</sup> in 1999 as the sum of all coffee\n\ntrees in each municipality divided by the land area (km <sup>2</sup> ). We use this as our measure of coffee\n\n\nintensity at baseline, reflecting the intensity of coffee cultivation per unit of land area across\n\n\nmunicipalities. Figure 2 shows the spatial variations in coffee intensity across municipalities in\n\n\nRwanda.\n\n\nWe complement our analysis with data on coffee prices from the World Bank Commodity\n\nPrice database. <s\n\nMention: \"1999 Rwandan Coffee Census\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Coffee tree counts by age group per municipality give baseline cultivation intensity.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001321:11:0", "occurrence_id": "50f3dae4451652fb86a00a3b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: electricity connections per 1000 people. <sup>15</sup> Figure 3 shows the temporal trends in electrification\n\n\nat the municipality level.\n\n\nNext, to understand consumption patterns, we use data on electricity purchases to con\n\nstruct customer-level panel data on electricity consumption (purchases). Specifically, for each\n\n\nquarter-year, we aggregate the data on the frequency of electricity purchases, quantity (kWh)\n\n\nof electricity, quantity (kWh) per transaction, and amount (monetary value in RWF) of elec\n\ntricity purchased. Further, we complement our analysis with spatial data showing the location\n\n\nof the universe of all electricity transformers, as well as the geographic footprints of low and\n\nmedium-voltage grid lines in Rwanda. <sup>16</sup> This data is used to examine whether coffee-growing\n\n\nmunicipalities are disproportionately favored in the rollout of grid infrastructure.\n\n\n_Coffee Data_\n\n\nTo measure the intensity of coffee cultivation at baseline, we rely on data compiled by Guariso\n\n\nand Verpoorten (2018) from the 1999 Rwandan Coffee Census. The data contains detailed in\n\nformation on the coffee-growing municipalities, including the number of coffee trees in several\n\n\nage groups ( _<=_ 3, 3 _−_ 10, 10 _−_ 30, and _≥_ 30 years) per municipality. Using these data, for each mu\nnicipality, we compute the total number of coffee trees per km <sup>2</sup> in 1999 as the sum of all coffee\n\ntrees in each municipality divided by the land area (km <sup>2</sup> ). We use this as our measure of coffee\n\n\nintensity at baseline, reflecting the intensity of coffee cultivation per unit of land area across\n\n\nmunicipalities. Figure 2 shows the spatial variations in coffee intensity across municipalities in\n\n\nRwanda.\n\n\nWe complement our analysis with data on coffee prices from the World Bank Commodity\n\nPrice database. <s\n\nMention: \"World Bank Commodity\n\nPrice database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Coffee price movements are added to the municipality-level cultivation data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001322:22:2", "occurrence_id": "ac70c8db754b2beb8b86efc4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Hill area<br>_Percent_<br>47<br>75<br>34<br>Urban area<br>_Percent_<br>20<br>30<br>15|\n\n\n_Source: Nepal NLSS III 2010/11_\n\n\n21\n\nMention: \"Nepal NLSS III\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Hill and urban area percentages in the table are taken from this living standards survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001323:55:6", "occurrence_id": "b01791eef5e1cfc85868d6b4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  upstreamness index _>_ 1 _._ 3. The LHS variable on all models is _Mj,t_ <sup>_k_</sup> <sup>, the total value imported</sup>\n\nin Dominican Republic _i_ from place of origin _j_ of product _k_ in year _t_ . All models are estimated using the PPML\nestimator on Dominican Republic’s import data pooled across observations at the year, HS6 digit product, and placeof-origin level, with year fixed effects included in the estimation model. _σ_ <sup>_k_</sup> estimates from Fontagné et al. (2022) are\ninteracted with geographic frictions, and enter separately in the regression itself. Pharmaceutical products are excluded\nfrom the estimation. IHST denotes the Inverse Hyperbolic Sine Transformation. Model (5) is estimated using fewer\nobservations due to missing trade tariff data.\n\n\n53\n\nMention: \"import data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Import values by origin, year and six-digit product are pooled for the gravity estimation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001323:55:6", "occurrence_id": "e554ecf52df949e3a463ce68", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  upstreamness index _>_ 1 _._ 3. The LHS variable on all models is _Mj,t_ <sup>_k_</sup> <sup>, the total value imported</sup>\n\nin Dominican Republic _i_ from place of origin _j_ of product _k_ in year _t_ . All models are estimated using the PPML\nestimator on Dominican Republic’s import data pooled across observations at the year, HS6 digit product, and placeof-origin level, with year fixed effects included in the estimation model. _σ_ <sup>_k_</sup> estimates from Fontagné et al. (2022) are\ninteracted with geographic frictions, and enter separately in the regression itself. Pharmaceutical products are excluded\nfrom the estimation. IHST denotes the Inverse Hyperbolic Sine Transformation. Model (5) is estimated using fewer\nobservations due to missing trade tariff data.\n\n\n53\n\nMention: \"trade tariff data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Missing entries reduce the observations available for one specification.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001328:50:1", "occurrence_id": "e55f48e19c2df67f7d77b666", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Enterprise Surveys, World Bank.<br>Data available at:www.etnerprisesurveys.org <br>|\n|Super efficiency<br>|<br>Slack-based input-oriented non-radial pure<br>technical efficiency based on the methodology<br>in Tone (2002). The estimation assumes<br>variable returns to scale and a single output and<br>three inputs. The output is total annual sales of<br>the hotel during the last complete fiscal year.<br>Inputs are total labor cost incurred by the hotel<br>during the last complete fiscal year, total cost<br>of electricity incurred by the hotel during the<br>last complete fiscal year, and total number of<br>rooms at the hotel at the end of the last<br>complete fiscal year. Stata “dea_supersbm”<br>program was used to compute the efficiency<br>scores. Higher values of the variable imply<br>higher efficiency.<br>Source: Authors’ own calculations based on<br>data from Enterprise Surveys, World Bank.<br>Data available at:www.etnerprisesurveys.org <br>|\n|BCC efficiency|<br>Input-oriented radial pure technical efficiency<br>obtained using the Data Envelopment Analysis<br>(DEA). The methodology is based on the work|\n\n\n49\n\nMention: \"Enterprise Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Firm inputs underlying the efficiency estimates are drawn from this survey and released publicly.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001331:3:0", "occurrence_id": "475a7af8f19d070ec31f2bf6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ##### **1 Introduction**\n\nGenerative AI (GenAI) holds the potential to transform economies and societies. Taking traditional\n\n\nAI’s predictive power one step further, generative AI is capable of creating new content in all forms\n\n\nof media - text, code, images, audio, video, and more. Generative AI tools like ChatGPT, Co-Pilot,\n\n\nand Midjourney are expected to revolutionize how certain tasks are performed, leading to significant\n\n\nefficiency gains and new opportunities for innovation (Eloundou, Manning, Mishkin, and Rock 2023;\n\n\nHumlum and Vestergaard 2024). Since the debut of ChatGPT in November 2022, various types of\n\n\ngenerative AI tools have proliferated and many have amassed a huge user base within record time.\n\n\nThe widespread use of generative AI offers the possibility to transform economic development, social\n\n\nstructures, and global competitiveness. Despite extensive discussions about the applications and\n\n\npotentials of generative AI on societies (Korinek 2023; Chui, Hazan, Roberts, Singla, and Smaje\n\n\n2023; Brynjolfsson, Li, and Raymond 2023; Jha, Qian, Weber, and Yang 2024; Kim, Muhn, and\n\n\nNikolaev 2024), research on the scale of generative AI usage is rare. This paper seeks to unveil the\n\n\nreal-time global scale of generative AI usage, including its demographic and country distribution,\n\n\nand to explore the barriers to and impacts of its adoption. As these technologies continue to evolve,\n\n\nunderstanding their adoption patterns, both in advanced and developing economies, becomes crucial\n\n\nfor policymakers, businesses, and researchers.\n\n\nTo measure AI adoption globally, several organizations have been publishing AI indexes and\n\n\nreports, tracking trends in AI development, and focusing on assessing countries’ readiness for AI\n\n\nintegration. Since 2017, Stanford University has been publishing the AI Index Report (Maslej,\n\n\nFattorini, Perrault, et al. 2024) annually,\n\nMention: \"AI Index Report\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"An annual index tracking AI development and national readiness since 2017.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001333:10:1", "occurrence_id": "cac145a6b72dff05414cac9d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  shock in 2020Q3 that carries over to Q4. We expect the\n\ncommuters’ outcomes to instead track events in Israel and occupied territories more\n\nclosely, with a major shock in 2020Q2, recovery in Q3, and another dip in Q4.\n\n### **3 Data**\n\n##### **3.1 Data sources and definitions**\n\n\nWe employ data from the Labor Force Surveys (LFS) of the West Bank and Gaza,\n\ncollected by the Palestinian Central Bureau of Statistics and prepared by the Economic\n\nResearch Forum. The surveys are conducted on a quarterly basis, covering periods\n\nfrom 2000 onwards. The LFS is meant to represent all households whose ordinary\n\nresidence is in the West Bank and Gaza, though their place of work need not be. The\n\nLFS is representative at the region level (respectively of the West Bank and Gaza),\n\nas well as at the level of locality types (urban, rural, and refugee camps) ( _Palestine_ _-_\n\n_Labor_ _Force_ _Survey,_ _LFS_, 2021).\n\n\nImportantly for the purpose of our analysis, the data have a panel dimension, en\nabling the study of labor market transitions. The sample rotation scheme is described\n\n\n9\n\nMention: \"Labor Force Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Quarterly surveys from 2000 onwards, prepared by the Economic Research Forum, supply the commuter outcomes.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001335:11:0", "occurrence_id": "b8bb06cb8dd3a1d91795e57e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The data were sourced from the National Health Information System (NHIS). This system\n\n\nprovides monthly data at the health center level on the quantity of each contraceptive method\n\n\ndistributed within their designated catchment area by the Burundian government as well as the\n\n\nnumber of CHW referrals to health center facilities. An important limitation of the NHIS data\n\n\nis the difficulty of distinguishing between true zeros and missing values when a health center\n\n\nreports no data for a contraceptive method in a given month. To partly address this issue, we\n\n\nclassify observations as missing if all contraceptive methods show zero quantities for a given\n\n\nhealth center and month. This classification applies to approximately 5% of our sample.\n\n\nThe primary outcome of our study is the quantity of SP injections reported by health centers\n\n\nin the NHIS data. As secondary outcomes, we employ the quantities of relevant alternative\n\n\ncontraceptive methods to investigate potential substitution effects.\n\n##### **Study Sample**\n\n\nTable 1 summarizes key information on the 138 study health centers, revealing the limited\n\n\ncapacity of health centers and low contraceptive utilization (column 1). Panel A shows that\n\n\nonly 38 health centers (28%) had HPO availability. Among these health centers, each had only\n\n\none HPO, indicating very limited in-home administration capacity before the CHW intervention\n\n\nrollout. On average, there were 16 CHWs serving a population of 15,775 within the health\n\n\ncenters’ catchment areas. Of this population, an estimated 3,739 are women of childbearing\n\n\nage, translating to roughly 230 women per CHW.\n\n\nPanel B presents the baseline number of SP injections administered per month within the\n\n\ncatchment area of health centers as well the overall quantity of all contraceptives dispensed,\n\n\nincluding injectables (SP and DMPA-IM), condoms, birth control pills, emergency contraceptive\n\n\npills, implants, and intra-uterine devices. It is worth noting that this measure should be\n\n\ninterpreted with caution given the varying protection duration across contraceptive methods. A\n\nMention: \"National Health Information System\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Monthly contraceptive quantities distributed by health centre and referrals are recorded in this system.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001336:47:0", "occurrence_id": "29c0aa4edf2d47ec6f73a8ec", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Table A3.2: Total number of concentrations and share of treated concentrations by academic\nprogram\n\n\nNumber of concentrations % ever treated\n\nAgriculture 136 23\n\nComputer Science 37 51\n\nEconomics 73 0\n\nEducation 274 12\n\nEngineering 73 49\n\nGeology and Environmental studies 33 61\n\nIslamic Studies 26 0\n\nJournalism 29 0\n\nLaw and Political Sci 46 0\n\nLiterature 95 5\n\nMedical Studies 23 83\n\nPsychology 13 0\n\nPublic Administration and Policy 16 100\n\nScience Education 16 81\n\nSocial Science 63 11\n\nTechnician (e.g., chemical tech, food tech) 9 44\n\nVeterinary 12 0\n\nTotal 975 21\n\n\nNote: Author created the academic programs according to the NEXA selection sheet classification.\nSource: Kankor matriculation dataset 2013-2018.\n\n\n44\n\nMention: \"Kankor matriculation dataset\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Concentrations of treated students across academic programmes are counted from these matriculation records.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001320:3:1", "occurrence_id": "f9ed268a39bb3d05297eb951", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \npart due to population structure (Demombynes et al. 2022). This suggests that the extent to\nwhich the gender gap in mortality increased likely differs across country income levels. In\npart due to the lack of data, previous detailed analysis of COVID-19 mortality patterns by age\nhave combined both sexes, neglecting this well-established fact that mortality rates of men\nare higher than those of women, and that this might vary by country income level.\nThis study draws on recent data which were not available for earlier studies to examine the\nextent to which the gender gap in mortality shifted in high- and middle-income countries\nduring the COVID-19 pandemic. Unfortunately, underdeveloped civil registration systems\nfor reporting on male and female deaths limits data availability in low-income countries, a\npersistent limitation in the study of global population-level mortality patterns.\n\n\n\n**Introduction**\n\n\n\n**Methods**\n\n\n\n_Data sources_ We use the “Global excess deaths associated with COVID-19 (modelled estimates)” dataset\n\n(May 19, 2023 update), produced by WHO’s Technical Advisory Group (TAG) on COVID-19\nmortality assessment. This dataset contains estimates of expected all-cause deaths, all-cause\ndeaths (actual or predicted if actual was not available), and excess deaths by age-group and\nsex for 194 countries for 2020 and 2021. Expected all-cause deaths were forecasted using\nhistoric country-level monthly mortality data prior to the pandemic and serve as reference\npoint in the absence of COVID-19. Excess deaths were computed as \"the mortality above\nwhat would be expected based on the non-crisis mortality rate in the population of interest”\nby differencing actual/predicted all-cause deaths from expected all-cause deaths (World\n\n\n\n_Data sources_\n\n\n\n2\n\nMention: \"Global excess deaths associated with COVID-19\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Modelled expected and actual all-cause deaths provide the mortality series analysed by sex.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001347:36:0", "occurrence_id": "894fc6fc43c352c80eaa6ae4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 5. Geographical Distribution – Night Light Density\n\nBefore and After the Rohingya Influx\n\n\n\n**Before 2017 (Average 1992 - 2016)**\n**Night Light Density**\n\n\n\n**After 2017 (Average 2017-2021)**\n**Night Light Density**\n\n\n\n0.000\n0.001 - 1.400\n1.401 - 1.750\n1.751 - 2.700\n2.701 - 17.000\n**Refugee Camps**\n\n\n\n0.000\n0.001 - 1.400\n1.401 - 1.750\n1.751 - 2.700\n2.701 - 17.000\n**Refugee Camps**\n\n\n\nRefugee Camps\n\n\n\nRefugee Camps\n\n\n\n_Notes:_ The maps illustrate the Cox’s Bazar district in Bangladesh segmented into one-square-kilometer grids. The Cox’s Bazar area has 2,392\ngrid cells. The left panel displays the distribution of night light density between 1992 and 2016, prior to the influx of Rohingya individuals into\nBangladesh. The right panel portrays the distribution of night light density between 2017 and 2021, following the start of the Rohingya influx. The\nmaps were generated using the harmonized night light data series spanning from 1992 to 2021, as developed by Li et al. (2020).\n\nMention: \"harmonized night light data series\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"A single continuous series lets light density be compared before and after the influx.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001350:24:1", "occurrence_id": "1b76474df9491b6e91b3c5f6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Coca Cola, 650ml” was included in the other half. In some countries with relatively\n\n\nfew products satisfying our selection criteria, we adjusted prices to reflect size- or unit\n\nequivalent prices applying the product’s unit price (e.g., price per ml). Finally, we left\n\n\nout products which were not broadly available across locations and periods, or products\n\n\nwhose prices did not change at all over the sample period. Tables 7 to 12 describe the\n\n\nfinal set of products we consider for each country in our sample.\n\n\n**Identification** **of** **origin** **locations** After normalizing product descriptions and brands,\n\n\nmanufacturers were identified. For local manufacturers, manufacturing plants were identi\n\nfied through internet searches, verifying that establishments did not correspond to corpo\n\nrate offices only. In some cases (mostly for beverages manufacturers), multiple manufac\n\nturing plants were identified. In these cases, for each destination, the product was assumed\n\n\nto have been distributed from the closest plant. Imported products were identified either\n\n\nbecause product listings indicated so (in some cases specifying the product’s country of\n\n\norigin), or because the manufacturer does not have a production facility in the country.\n\n\nFor imported products, origin locations are determined by identifying locations through\n\n\nwhich the product is more likely to have entered the country. For landlocked countries,\n\n\norigins were border crossings near the main ports that serve those countries. For countries\n\n\nwith ports, origins of imported products were the main sea ports. These were verified\n\n\nusing Logcluster.org, World Ports Source, Volza Exportgenius.\n\n\n23\n\nMention: \"World Ports Source\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"It helps verify the main sea ports used as origins for imported products.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001354:8:0", "occurrence_id": "9e8d85926067ecd2c4d2a78a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <u>Figure 8. Yearly costs of basic needs per one refugee to be covered in US$across different scenarios</u>\n\n\nThe yearly costs of basic needs assistance to be covered under the “current” scenario and across refugee household\ncharacteristics are reported in Figure 9. There are some groups of refugees with much lower needs for basic needs\nassistance than the average of the “current” scenario costs of US$193 (bars are highlighted in green). For instance,\nas Figure 9 illustrates, the costs for basic needs for refugees who own land are estimated to be around US$149 or\n23 percent lower than the overall average costs.\n\n\n<u>Figure 9. Yearly costs of basic needs per one refugee to be covered in US$for “current” scenario and across different</u>\n\n\n<u>Source: RHCS 2018, WB staff calculations.</u>\n<u>Note: Costs for refugees with different land ownership and land use exclude refugees residing in Kampala.</u>\n\nThe costs of basic needs assistance for refugees having land per capita above 0.05 hectare per capita are US$113 or\n41 percent lower than the average. Costs for refugees with a working head of household are about US$167 or 14\npercent lower than the average. Costs for basic needs assistance for refugees who stayed in Uganda three or more\nyears are US$123 or 37 percent lower than the average. Lastly, basic needs assistance costs for refugees with\nchildren having Ugandan friends are US$176 or 9 percent lower than the average. These descriptive results\ndemonstrate how better economic and social integration of refugees and access to productive assets such as land\nare associated with lower costs for meeting basic needs of refugees.\n\n\nTo evaluate the correlations above more formally, we ran a simple regression model using as left-hand side variable\nthe logarithm of the ratio between pre-assistance consumption to international poverty line. This left-hand side\nvariable measures the gap to be\n\nMention: \"RHCS 2018\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Refugee land ownership and use inform the scenario costs of basic needs assistance.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001358:10:3", "occurrence_id": "f4478f66f5e8fd73c872e0df", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  <u>0.01</u> <u>0.01</u> <u>0.00</u> <u>0.01</u> <u>0.01</u> <u>0.01</u> <u>0.00</u>\n<u>Western Europe</u> <u>0.19</u> <u>0.16</u> <u>0.19</u> <u>0.15</u> <u>0.18</u> <u>0.17</u> <u>0.18</u> <u>0.20</u> <u>0.20</u> <u>0.20</u> <u>0.19</u>\n\n~~Source: Author’s calculation based on the original Colombian customs data.~~\n\n\n\n<u>Western Europe</u> <u>0.19</u> <u>0.16</u> <u>0.19</u> <u>0.15</u> <u>0.18</u> <u>0.17</u> <u>0.18</u> <u>0.20</u> <u>0.20</u>\n\n~~Source: Author’s calculation based on the original Colombian customs data.~~\n<u>Note: This table displays Colombian exports as a share of total exports and is normalized in 2007.</u>\n\n\n\n<u>Note: This table displays Colombian exports as a share of total exports and is normalized in 2007.</u>\n\nSince NTM data are only available for\n\nMention: \"Colombian customs data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Shares by origin region are calculated from the original customs records.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001359:28:1", "occurrence_id": "ff9bead0f3b7ee8ac5aba793", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  in a shorter time frame. While 55% of\n\n\nmunicipal schools in municipalities that did not extend the winter break covered more than\n\n\n80% of the school curriculum, only 45% did so in municipalities that extended the winter break.\n\n\nData from _Prova_ _Brasil_ questionnaire show that teachers faced many barriers to successfully\n\n\ncovering the entire school curriculum. For instance, students in treated schools had fewer hours\n\n\nof classes per day (4.8 hours _×_ 5 hours). The 12-minute daily difference resulted in children from\n\n\nthe affected schools having two days less of school content between the reopening of the schools\n\n\nand the proficiency test. <sup>30</sup> Also, we do not find any source of information of an extension of\n\n\nschool days later in the year to account for the missing days during the winter break extension.\n\n\n29There are between 44 and 53 business days between the reopening of the schools (August 17, 2009) and the\nproficiency test (which took place between October 19 and October 31, 2009). Therefore, the 24-minute difference\nwould result in 1,056 to 1,272 additional minutes of class, equivalent to 17.6 to 21.2 hours. If the average number\nof class hours in the state-managed network in _𝐺_ = 1 is 5.3, the estimate in terms of school days is between three\nand four.\n\n30There are between 44 and 53 business days between the reopening of the schools (August 17, 2009) and the\nproficiency test (which took place between October 19 and October 31, 2009). Therefore, the 12-minute difference\nwould result in 528 to 636 fewer minutes of class, equivalent to 8.8 to 10.6 hours. If the average number of class\nhours in the affected network is 4.8, the estimate in terms of school days is roughly two.\n\n\n26\n\nMention: \"_Prova_ _Brasil_ questionnaire\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Teacher-reported barriers and class hours per day come from this assessment questionnaire.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001367:4:1", "occurrence_id": "0d44f13773807f846a95ba97", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  employees are analyzed. The\n\nliterature has typically focused on manufacturing firms with 50 or more employees.\n\n\nFor around 9,000 medium and large firms across manufacturing and service sectors in 41 countries, our\n\nfindings are consistent with that of Bloom et al (2015) that family management is negatively correlated\n\nwith the quality of management practices. This is an important finding given the differences of our sample\n\nmentioned earlier – we explore both manufacturing and service firms and firms with 20 or more employees,\n\nwhile Bloom et al (2015) only study manufacturing firms with 50 or more employees. Our sample also\n\nconsists of firms in Eastern Europe and the Middle East and North Africa that are largely omitted in the\n\nBloom (2015) sample. The extensive nature of the WBES allows us to enrich the analysis in several ways.\n\nFirst, the surveys collect information on the share of managers that are family, providing some nuance on\n\nwhether any family management or all family management is detrimental to management practices. We\n\nfind evidence towards the latter. Second, the surveys allow for the exploration of whether certain factors\n\neither exacerbate or alleviate the negative relationship between family managers and the quality of\n\nmanagement practices. These include a range of internal and external operating factors. We particularly\n\n\n3\n\nMention: \"WBES\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Its breadth across regions and manager-family information extends the earlier analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001368:22:0", "occurrence_id": "2b9b9f2986a69f6442bb2390", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: economy, leading to aggregate consequences on output and productivity. Fifth, the inclusion\n\n\nof the home production sector is indeed relevant for assessing the efficiency gains of a rise or\n\n\nfall in the premium paid in the public sector.\n\n#### **4 Calibration strategy**\n\n\nWe calibrate the parameters in our model to match key moments of the allocation of talent in\n\n\nEgypt. Our data were extracted from ILOSTAT databases. <sup>10</sup>\n\n\nWe select _α_ such that _ηeα_ (the share of capital) equals a capital income share of 30%,\n\n\nas is standard in the literature (e.g., Gollin, 2002). We assume that the capital income share\n\n\nin the public firm is the same as in the private sector. Hence, we determine _κ_ such that _ηgκ_\n\n\nequals 0.3. The fractions of women and men in the population, _ϕf_ and _ϕm_, are directly taken\n\nfrom the data: _ϕf_ = _ϕm_ = 0 _._ 5. For the parameters of the Euler equation used to set the location\n\n\nof the infinitely elastic supply of capital ( _β_ and _δ_ ), we take the standard value of 0.95 for the\n\n\ndiscount factor _β_, as reported in the literature, and we follow Cagetti and De Nardi (2006) and\n\nBuera and Shin (2013) to choose _δ_ = 0 _._ 06. <sup>11</sup>\n\n\nIn line with Guner et al. (2008), we assume that log-managerial ability follows a (trun\n\ncated) normal distribution with a mean of _m_ and a standard deviation of _sd_ . Additionally, we\n\n\nimpose that this distribution represents the majority of production units, constituting a total\n\nmass of 1 − _p_ ˆ\n\nMention: \"ILOSTAT databases\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Moments of talent allocation across sectors are taken from these labour statistics to calibrate the model.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001371:6:2", "occurrence_id": "f7c449b14b6f5770ab9d7da2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  levels.\nMost low- and middle-income countries feature procyclical or cyclically neutral expenditures, whereas\nmost high-income countries feature countercyclical expenditures. At the same time, high-income\ncountries also saw their revenues and fiscal balances move more strongly in line with the business cycle,\ncorresponding to the desired outcome under countercyclical fiscal policy. This indicates that richer\ncountries were generally more successful in achieving countercyclical fiscal policy than countries with\n\n\nassistance benefits, and employer social benefits in cash and in kind (World Bank 2023). For the Arab Republic of Egypt, we amend\nthis variable with data from World Bank internal sources as follows (for the years 2003-2023): 4.9%, 5.1%, 5.5%, 11.2%, 7.8%,\n10.3%, 12.2%, 8.5%, 9.0%, 9.0%, 10.6%, 10.7%, 8.1%, 7.4%, 7.6%, 7.1%, 5.1%, 3.7%, 4.0%, 4.4%, 4.4%.\n\n~ 5 ~\n\nMention: \"data from World Bank internal sources\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"A published expenditure variable is amended with these internal annual values.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001343:32:1", "occurrence_id": "2555f9a307eb0f718f694030", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  institutions.\n\n\nWe test these two hypotheses. First, we study whether intermediate inputs that are differ\nentiated (i.e., relationship-specific) display higher complementarity. To test this hypothesis, we\n\nestimate our firm-level elasticities of substitution across suppliers conditional on whether the seller\n\noffers an intermediate input classified as either “Differentiated\" or “Homogeneous\" according to\n\nRauch (1999).\n\n\nSecond, we study whether intermediate inputs that require highly-specified contracts and,\n\ntherefore, are likely to be traded in locations with better institutional quality display higher com\nplementarity (Boehm, 2022; Boehm and Oberfield, 2020). To test this hypothesis, we use the\n\ngeographic variation in institutional quality across Indian districts. We classify districts into ter\nciles ( _Low_, _Medium_, and _High_ ) according to the average time it takes for courts in each Indian\n\ndistrict until a first hearing from the [DevData](https://www.devdatalab.org/judicial-data) Lab database, and we estimate our firm-level elas\n\n28For instance, in Elliott and Golub (2022), firms need specific knowledge about input requirements and capabilities\nfrom their trading partners and therefore need an ongoing relationship to overcome moral hazard problems.\n\n\n30\n\nMention: \"DevData\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Average time to first hearing in each district classifies districts into low, medium and high judicial speed.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001364:18:0", "occurrence_id": "7a1fbce81a53748e653eb324", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: effect of the pandemic is insignificant. The results for reading and science should be interpreted\n\n\nwith caution. The negative trends in student achievement started before the pandemic. Thus,\n\n\nour identification strategy, which relies on stable long-term trends, cannot be applied without\n\n\ndoubts to reading and science. Our robustness checks for reading and science results confirm\n\n\nthat indeed in these two domains the time trends are not linear, and one cannot distinguish\n\n\nbetween the effects of the pandemic and the long-term decline in achievement. What causes\n\n\nthese long-term declines is beyond the scope of this paper. We can only note that both reading\n\n\nand science were minor domains in PISA 2022, meaning the measurement of student\n\n\nachievement was less precise than in mathematics.\n\n\n**Table 4. Impact of Pandemic on Student Reading and Science Achievement**\n\n\nReading (2000 to 2022) Science (2006 to 2022)\n\n(1) (2) (3) (4) (5) (6)\n\n\n\n-6.96***\n\n(0.75)\n\n-0.33***\n\n(0.04)\n\n\n\n-8.43***\n\n(1.10)\n\n\n-0.14***\n\n(0.04)\n\n\n\n-0.78\n\n(0.46)\n\n\n\n6.16***\n\n(0.84)\n\n-0.44***\n\n(0.04)\n\n\n\n8.13***\n\n(1.15)\n\n\n-0.38***\n\n(0.04)\n\n\n\nDeparture from linear\ntrend in 2022\n\n\nUNESCO: number of\nweeks with full closures\n\nPISA: country average\nstudent-provided number\nof weeks of closures\n\n_Additional controls_\n\nIndividual level: gender,\nimmigrant background,\nSES\n\nCountry average: gender,\nimmigrant background,\nSES\n\n\n\n-12.10***\n\n(0.43)\n\n\n\nYes Yes Yes Yes Yes Yes\n\n\nYes Yes Yes Yes Yes Yes\n\n\n\n\n- country fixed effects Yes Yes Yes Yes Yes Yes\n\n- country*time Yes Yes Yes Yes Yes Yes\n\nN 3,\n\nMention: \"PISA 2022\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Reading and science were minor domains, so achievement is measured less precisely than mathematics.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001375:16:1", "occurrence_id": "b99f43b01f3d0431ca0968c7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  build the SOE variable, we first clean the raw Links files and eliminate branches\n\n\n(See Appendix C for details). Then, we merge the Links and Entities files to identify\n\n\nthe shareholders that are public authorities. We do so using the entity category S,\n\n\nwhich corresponds to governmental agencies, public departments, and local authori\n\nties. We exclude sovereign wealth funds and institutional investors from this set, as\n\n\n7Each BvD firm ID has two components: a country code that refers to the country where the\nfirm operates and a number.\n\n8Austria, Belgium, Bosnia and Herzegovina, Bulgaria, Croatia, Czechia, Estonia, Finland, France,\nGermany, Hungary, Italy, Luxembourg, Montenegro, North Macedonia, Norway, Poland, Portugal,\nRomania, Slovak Republic, Slovenia, Spain, Serbia and Ukraine.\n\n9The authors develop a methodology and SQL code to recreate the ownership trees for all the\nfirms in the ORBIS database.\n\n\n15\n\nMention: \"ORBIS database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"Ownership trees are recreated for all firms with bespoke SQL code after branches are removed.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001378:17:0", "occurrence_id": "ac8be1d19b970bd36dfe8b5f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 9. Fertilizer prices**\n\n\nIn addition to the reports of relative fertilizer price changes from season to season,\n\nrespondents were also asked for quantitative fertilizer prices at certain points in time. Figure 9\n\nplots the average price of inorganic fertilizer per kilogram. Additional data on fertilizer prices will\n\nbecome available through the HFPS over time, lengthening the time series. In Ethiopia the average\n\nprice of fertilizer increased by roughly 15 ETB/kg (unadjusted), or approximately 33%, from\n\nJanuary 2023 to February 2024. Prices in Uganda reportedly rose from 6,480 UGX/kg to 9,134\n\nUGX/kg on average, an increase of about 41%, between March and August 2023. In Burkina Faso,\n\nfertilizer prices fluctuated but within a relatively tight range, growing by roughly 2.5% between\n\nNovember 2022 and January 2024. Contrary to the experience in other countries, households in\n\nMalawi reported a decline in fertilizer prices of about 6% (December 2022 to June 2023).\n\nHouseholds reporting an increase in fertilizer prices were asked how they coped with such\n\nincreases. While for faming households in Tanzania reduction in the quantity of fertilizer applied\n\nper unit area is the single most important strategy employed to cope with the high prices of\n\ninorganic fertilizer, farming households in Burkina Faso and Ethiopia adopted a number of\n\n\n16\n\nMention: \"data on fertilizer prices\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Reported prices per kilogram of inorganic fertilizer are averaged into a price series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001378:17:0", "occurrence_id": "d881fe35ed9905ea6657588a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 9. Fertilizer prices**\n\n\nIn addition to the reports of relative fertilizer price changes from season to season,\n\nrespondents were also asked for quantitative fertilizer prices at certain points in time. Figure 9\n\nplots the average price of inorganic fertilizer per kilogram. Additional data on fertilizer prices will\n\nbecome available through the HFPS over time, lengthening the time series. In Ethiopia the average\n\nprice of fertilizer increased by roughly 15 ETB/kg (unadjusted), or approximately 33%, from\n\nJanuary 2023 to February 2024. Prices in Uganda reportedly rose from 6,480 UGX/kg to 9,134\n\nUGX/kg on average, an increase of about 41%, between March and August 2023. In Burkina Faso,\n\nfertilizer prices fluctuated but within a relatively tight range, growing by roughly 2.5% between\n\nNovember 2022 and January 2024. Contrary to the experience in other countries, households in\n\nMalawi reported a decline in fertilizer prices of about 6% (December 2022 to June 2023).\n\nHouseholds reporting an increase in fertilizer prices were asked how they coped with such\n\nincreases. While for faming households in Tanzania reduction in the quantity of fertilizer applied\n\nper unit area is the single most important strategy employed to cope with the high prices of\n\ninorganic fertilizer, farming households in Burkina Faso and Ethiopia adopted a number of\n\n\n16\n\nMention: \"HFPS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Future waves will extend the price series as more rounds are fielded.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001384:1:0", "occurrence_id": "af0d18386c336e3dedafc234", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Policy Research Working Paper 11046\n\n### **Abstract**\n\nThis paper investigates the role of gender bias in artificial\n\nintelligence–driven analyses of citizen participation, using\ndata from the 2023 Latinobarómetro Survey. The paper\nproposes that gender bias—whether societal, data driven, or\nalgorithmic—significantly affects civic engagement. Using\nmachine learning, particularly decision trees, the analysis\nexplores how self-reported societal bias (machismo norms)\ninteracts with personal characteristics and circumstances to\nshape civic participation. The findings show that individuals with reportedly low levels of gender bias, who express\npolitical interest, have high levels of education, and align\n\n\n\nwith left-wing views, are more likely to participate. The\npaper also explores different strategies to mitigate gender\nbias in both the data and the algorithms, demonstrating that\ngender bias remains a persistent factor even after applying\ncorrective measures. Notably, lower machismo thresholds\nare required for participation in more egalitarian societies, with men needing to exhibit especially low machismo\nlevels. Ultimately, the findings emphasize the importance\nof integrated strategies to tackle gender bias and increase\nparticipation, offering a framework for future studies to\nexpand on nonlinear and complex social dynamics.\n\n\n\nThis paper is a product of the Social Development Global Department. It is part of a larger effort by the World Bank to\n\nprovide open access to its research and make a contribution to development policy discussions around the world. Policy\nResearch Working Papers are also posted on the Web at http://www.worldbank.org/prwp. The authors may be contacted\n[at jcuesta@worldbank.org. A verified reproducibility package for this paper is available at http://reproducibility.worldbank.](http://reproducibility.worldbank.org)\n[org, click](http://reproducibility.worldbank.org) **[here](https://reproducibility.worldbank.org\n\nMention: \"2023 Latinobarómetro Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Self-reported civic engagement and societal bias are modelled with decision trees on this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001385:20:0", "occurrence_id": "ce661e5d6a774f1f8301c1e0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: sector. Since the WBES includes only formal firms, we cannot directly test for these two\n\nalternative explanations. Yet, suggestive evidence indicates that our effect is driven by\n\nformal firms becoming informal. First, Table A23 columns 1 and 3 show that conflict\n\nintensity does not change the (self-reported) total number of competitors of the firm.\n\nAt the same time, conflict is associated with higher exit rates among formal firms (see\n\nTable A5). In sum, the overall level of competitive pressure faced by the firm remains\n\nunchanged, conflict makes some formal firms to exit the market, and competition from\n\ninformal firms increases. These three effect are consistent with conflict inducing some of\n\nthe previously formal firms to become informal.\n\n\nThese results align with prior studies showing that informality tends to rise in response\n\nto negative shocks, as firms seek to exploit the cost advantages of non-compliance with\n\nformal regulations (Ulyssea, 2018). This is also the case for an increase in conflict expo\nsure. For instance, Neri-Lain´e (2025) documents that —in the context of Afghanistan—\n\nan increase in insecurity reduces formal firms’ existence, increasing informality. These re\nsults documenting the shock-induced increase in informality are relevant for our analysis\n\nbecause of the evidence showing that more competition of informal firms can have nega\ntive consequences for formal ones. An increase in informal competition reduces the output\n\nof formal firms (Rozo and Winkler, 2021), lowers their employment (Amin, 2023), weak\nens innovation incentives (Avenyo et al., 2021), and dampens demand for credit, thereby\n\ncurbing investment (Brancati et al., 2024). More broadly, the expansion of informality\n\nposes challenges for the functioning of the overall economy. It contributes to resource mis\nallocation, reduces total factor productivity, and is often associated with broader welfare\n\nlosses (Ulyssea, 2018\n\nMention: \"WBES\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Because it covers only registered firms, informalisation cannot be tested directly.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001386:8:0", "occurrence_id": "6139d992660bd549aeb10ac5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 6\n\n\n3. Data\n\n\n3.1. **Outcomes.** To measure our outcome variables, we will combine two adminis\n\ntrative datasets. First, we will compute school-level enrollment, dropout rates and\n\n\npromotion rates using the administrative registry of all the students in Colombia, en\n\nrolled in either public or private schools. This dataset is called _R166-SIMAT_ and its\n\n\nsource is the Colombian Ministry of Education. <sup>7</sup> It is available for the period 2012\n\n2018, which shapes our sample period.\n\n\nImportantly, R166-SIMAT includes the student ID that allows us to distinguish be\n\ntween Colombian and foreign students in order to explore the effects of the Venezuelan\n\n\nmigration shock on both native and migrant students. Specifically, we identified as\n\n\n‘migrant’ those students with IDs different to the standard ID that the government\n\n\nissues to underage natives. These include special residence permits, visas, and border\n\n\nmobility cards. We also classified as migrant students the individuals who enrolled\n\n\nusing a provisional ID, provided by the municipal Secretary of Education to undoc\n\numented children who want to enroll in a public institution. The vast majority of\n\n\nundocumented children are foreigners, most of whom are Venezuelans (see section 2.\n\n\nUsing R166-SIMAT we can construct the following school (or school/grade)-level vari\n\nables:\n\n\n1. **(Log)** **Enrollment:** the (log of the) total number of students enrolled per\n\n\nschool (or school/grade) at the beginning of each academic year.\n\n\n7The name of the dataset originated in the Ministry’s Resolution 166 of 2004, which created the\nNational Enrollment System (SIMAT, from the Spanish acronym) that mandated all education institutions to report to the Ministry individual-level enrollment each year as well as the condition of\neach student at the end of the academic year. Importantly, this registry\n\nMention: \"administrative registry of all the students in Colombia\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Enrolment, dropout and promotion rates for public and private schools are computed from this registry.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001386:8:0", "occurrence_id": "18cce493188089f514d1bc66", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 6\n\n\n3. Data\n\n\n3.1. **Outcomes.** To measure our outcome variables, we will combine two adminis\n\ntrative datasets. First, we will compute school-level enrollment, dropout rates and\n\n\npromotion rates using the administrative registry of all the students in Colombia, en\n\nrolled in either public or private schools. This dataset is called _R166-SIMAT_ and its\n\n\nsource is the Colombian Ministry of Education. <sup>7</sup> It is available for the period 2012\n\n2018, which shapes our sample period.\n\n\nImportantly, R166-SIMAT includes the student ID that allows us to distinguish be\n\ntween Colombian and foreign students in order to explore the effects of the Venezuelan\n\n\nmigration shock on both native and migrant students. Specifically, we identified as\n\n\n‘migrant’ those students with IDs different to the standard ID that the government\n\n\nissues to underage natives. These include special residence permits, visas, and border\n\n\nmobility cards. We also classified as migrant students the individuals who enrolled\n\n\nusing a provisional ID, provided by the municipal Secretary of Education to undoc\n\numented children who want to enroll in a public institution. The vast majority of\n\n\nundocumented children are foreigners, most of whom are Venezuelans (see section 2.\n\n\nUsing R166-SIMAT we can construct the following school (or school/grade)-level vari\n\nables:\n\n\n1. **(Log)** **Enrollment:** the (log of the) total number of students enrolled per\n\n\nschool (or school/grade) at the beginning of each academic year.\n\n\n7The name of the dataset originated in the Ministry’s Resolution 166 of 2004, which created the\nNational Enrollment System (SIMAT, from the Spanish acronym) that mandated all education institutions to report to the Ministry individual-level enrollment each year as well as the condition of\neach student at the end of the academic year. Importantly, this registry\n\nMention: \"_R166-SIMAT_\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"The named dataset and its ministry source define the available sample period.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001386:8:0", "occurrence_id": "d3b7a7edeb4b710e6d17586c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 6\n\n\n3. Data\n\n\n3.1. **Outcomes.** To measure our outcome variables, we will combine two adminis\n\ntrative datasets. First, we will compute school-level enrollment, dropout rates and\n\n\npromotion rates using the administrative registry of all the students in Colombia, en\n\nrolled in either public or private schools. This dataset is called _R166-SIMAT_ and its\n\n\nsource is the Colombian Ministry of Education. <sup>7</sup> It is available for the period 2012\n\n2018, which shapes our sample period.\n\n\nImportantly, R166-SIMAT includes the student ID that allows us to distinguish be\n\ntween Colombian and foreign students in order to explore the effects of the Venezuelan\n\n\nmigration shock on both native and migrant students. Specifically, we identified as\n\n\n‘migrant’ those students with IDs different to the standard ID that the government\n\n\nissues to underage natives. These include special residence permits, visas, and border\n\n\nmobility cards. We also classified as migrant students the individuals who enrolled\n\n\nusing a provisional ID, provided by the municipal Secretary of Education to undoc\n\numented children who want to enroll in a public institution. The vast majority of\n\n\nundocumented children are foreigners, most of whom are Venezuelans (see section 2.\n\n\nUsing R166-SIMAT we can construct the following school (or school/grade)-level vari\n\nables:\n\n\n1. **(Log)** **Enrollment:** the (log of the) total number of students enrolled per\n\n\nschool (or school/grade) at the beginning of each academic year.\n\n\n7The name of the dataset originated in the Ministry’s Resolution 166 of 2004, which created the\nNational Enrollment System (SIMAT, from the Spanish acronym) that mandated all education institutions to report to the Ministry individual-level enrollment each year as well as the condition of\neach student at the end of the academic year. Importantly, this registry\n\nMention: \"R166-SIMAT\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Student identifiers allow migrant and native students to be distinguished.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001387:38:0", "occurrence_id": "dc55f25a0fd05594bcfcf7ea", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure A3: Percentage of Natives by Departament. Buenos Aires Metropolitan Area\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n_Sources:_ REDATAM INDEC Census Argentina 2010 (Instituto Nacional de Estad´ıstica y Censos,\n2010). _Notes:_ Buenos Aires metropolitan area includes the 15 communes of the city of Buenos Aires,\nand the following departments in the province of Buenos Aires: Almirante Brown, Avellaneda,\nBerisso, Berazategui, Canuelas, Ensenada, Escobar, Esteban Echeverr´ıa, Ezeiza, Florencio Varela,\nGeneral Rodr´ıguez, General San Mart´ın, Hurlingham, Ituzaing´o, Jos´e C. Paz, La Matanza, La\nPlata, Lan´us, Lomas de Zamora, Malvinas Argentinas, Marcos Paz, Merlo, Moreno, Mor´on, Pilar,\nPresidente Per´on, Quilmes, San Fernando, San Isidro, San Miguel, San Vicente, Tigre, Tres de\nFebrero y Vicente L´opez.\n\n\n37\n\nMention: \"REDATAM INDEC Census Argentina 2010\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Share of natives by department in the metropolitan area is tabulated from this census.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001389:39:1", "occurrence_id": "80af563a524170ee12d1d506", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: square</u> <u>0.803</u> <u>0.802</u> <u>0.825</u> <u>0.804</u> <u>0.523</u> <u>0.851</u> <u>0.828</u>\n\n_<u>Standard errors in parentheses. * p < 0.05, ** p < 0.01, *** p < 0.001. The models use administrative data from INVALSI and the</u>_ <u>Obs</u> <u>2873</u> <u>1851</u> <u>2880</u> <u>2890</u> <u>13242</u> <u>2073</u> <u>1500</u>\n\n\n\n_<u>Standard errors in parentheses. * p < 0.05, ** p < 0.01, *** p < 0.001. The models use administrative data from INVALSI and the</u>_\n_Ministry of Education for the 2022-23 academic year, including Ukrainian refugees and foreigners arrived in the country after_\n_February 2022 and enrolled in Grades 8 through 13. “Ukr Post Feb 2022 is a variable identifying Ukrainian refugees. The_\n_“recommended for high-track” variable is relevant for grade 8 only._\n\n\n\n38\n\nMention: \"administrative data from INVALSI\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Student test records supply the outcomes in the estimated models.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001391:38:0", "occurrence_id": "e68198dd09fa6e61117682d3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 4: Effect of U.S. sanctions on patent applications\n\n\nSource: PATSTAT Dataset, Spring 2022 version.\nNotes: The red dots are the estimated differences in the number of patent applications between sanctioned\nand unsanctioned firms five years before and after the sanctioned firms were added to the Entity List.\nThe blue bar is the 95% confidence interval. The estimates are based on Callaway and Sant’ Anna (2021)\napproach.\n\n\nFigure 5: Number of Chinese Patenting Firms Under Sanctioned IPC Field\n\n\nSource: PATSTAT Dataset, Spring 2022 version.\nNotes: The bar charts shows the number of firms in the newly added sanctioned IPC fields across different\nyears and different technology fields. We identified 40 sanctioned 4 digit IPC field, and then regroup these\nIPC fields into 12 broader technology fields listed in legend\n\n\n36\n\nMention: \"PATSTAT Dataset\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Patent application counts are compared between sanctioned and unsanctioned firms around listing.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001394:7:1", "occurrence_id": "02e90bfd1513016d52c74bf6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Enterprise\n\n\nSurveys (WBES) conducted two other surveys in Ghana in 2013 along with ISES: Micro-WBES\n\n\nand WBES. Only formal (registered) plants are covered in both surveys. Micro-WBES inter\n\nviews plants with fewer than 5 workers, while WBES interviews plants with at least 5 workers.\n\n\nI refer to plants in both Micro-WBES and WBES as formal plants.\n\n\nTable 2 presents the comparison of informal and formal plants in Ghana. The mean size\n\n\nof informal plants is smaller than formal plants: there are 2.3 workers in informal plants\n\n\nwhereas 3 and 34.7 workers are involved in production in the Micro-WBES and WBES re\nspectively. <sup>7</sup> Moreover, informal managers have less experience in the sector they operate in\n\n\ncompared to formal managers. While 42.8% of informal managers have at least a secondary\n\n\nschool degree, around 65% of workers in formal plants have at least a secondary school de\n\ngree. To sum up, on average, informal plants are smaller and younger compared to formal\n\n\nplants; their managers have fewer experience and less education; the fraction of female man\n\nagers is higher among informal plants.\n\n\n6Similar analysis is done by La Porta & Shleifer (2014) across a few countries using the same dataset, and\nMeghir _et al._ (2015) and Ulyssea (2018) in Brazil.\n\n7As Section 4 discusses in detail, the Ghana Statistical Service’s Integrated Business Establishment Survey\nreports the mean size of informal plants as 2.3 workers and the mean size of formal plants as 33.8. Similar\nmean sizes of the World Bank’s plant level survey data sets with that of the Ghana Statistical Service’s census\nbuild confidence in the usage of ISES and WBES to compare informal and formal plants in Ghana.\n\n\n5\n\nMention: \"WBES\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Plants with at least five workers are interviewed, defining the formal sample.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001394:7:1", "occurrence_id": "e212a90a6d3c4423dc170e3b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Enterprise\n\n\nSurveys (WBES) conducted two other surveys in Ghana in 2013 along with ISES: Micro-WBES\n\n\nand WBES. Only formal (registered) plants are covered in both surveys. Micro-WBES inter\n\nviews plants with fewer than 5 workers, while WBES interviews plants with at least 5 workers.\n\n\nI refer to plants in both Micro-WBES and WBES as formal plants.\n\n\nTable 2 presents the comparison of informal and formal plants in Ghana. The mean size\n\n\nof informal plants is smaller than formal plants: there are 2.3 workers in informal plants\n\n\nwhereas 3 and 34.7 workers are involved in production in the Micro-WBES and WBES re\nspectively. <sup>7</sup> Moreover, informal managers have less experience in the sector they operate in\n\n\ncompared to formal managers. While 42.8% of informal managers have at least a secondary\n\n\nschool degree, around 65% of workers in formal plants have at least a secondary school de\n\ngree. To sum up, on average, informal plants are smaller and younger compared to formal\n\n\nplants; their managers have fewer experience and less education; the fraction of female man\n\nagers is higher among informal plants.\n\n\n6Similar analysis is done by La Porta & Shleifer (2014) across a few countries using the same dataset, and\nMeghir _et al._ (2015) and Ulyssea (2018) in Brazil.\n\n7As Section 4 discusses in detail, the Ghana Statistical Service’s Integrated Business Establishment Survey\nreports the mean size of informal plants as 2.3 workers and the mean size of formal plants as 33.8. Similar\nmean sizes of the World Bank’s plant level survey data sets with that of the Ghana Statistical Service’s census\nbuild confidence in the usage of ISES and WBES to compare informal and formal plants in Ghana.\n\n\n5\n\nMention: \"Integrated Business Establishment Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its reported mean sizes of informal and formal plants are compared with the survey estimates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001394:7:1", "occurrence_id": "6d24d40f25fd882e90d3660f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Enterprise\n\n\nSurveys (WBES) conducted two other surveys in Ghana in 2013 along with ISES: Micro-WBES\n\n\nand WBES. Only formal (registered) plants are covered in both surveys. Micro-WBES inter\n\nviews plants with fewer than 5 workers, while WBES interviews plants with at least 5 workers.\n\n\nI refer to plants in both Micro-WBES and WBES as formal plants.\n\n\nTable 2 presents the comparison of informal and formal plants in Ghana. The mean size\n\n\nof informal plants is smaller than formal plants: there are 2.3 workers in informal plants\n\n\nwhereas 3 and 34.7 workers are involved in production in the Micro-WBES and WBES re\nspectively. <sup>7</sup> Moreover, informal managers have less experience in the sector they operate in\n\n\ncompared to formal managers. While 42.8% of informal managers have at least a secondary\n\n\nschool degree, around 65% of workers in formal plants have at least a secondary school de\n\ngree. To sum up, on average, informal plants are smaller and younger compared to formal\n\n\nplants; their managers have fewer experience and less education; the fraction of female man\n\nagers is higher among informal plants.\n\n\n6Similar analysis is done by La Porta & Shleifer (2014) across a few countries using the same dataset, and\nMeghir _et al._ (2015) and Ulyssea (2018) in Brazil.\n\n7As Section 4 discusses in detail, the Ghana Statistical Service’s Integrated Business Establishment Survey\nreports the mean size of informal plants as 2.3 workers and the mean size of formal plants as 33.8. Similar\nmean sizes of the World Bank’s plant level survey data sets with that of the Ghana Statistical Service’s census\nbuild confidence in the usage of ISES and WBES to compare informal and formal plants in Ghana.\n\n\n5\n\nMention: \"ISES\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Together with the enterprise survey it enables comparison of informal and formal plants.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001401:19:0", "occurrence_id": "baa27fd020b65e487dc6f3dd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: countries will result in differences that are small relative to the confidence intervals. A useful rule of\n\n\nthumb is that the observed difference in estimates of governance between two countries is not\n\n\nstatistically significant if the confidence intervals overlap. For example, comparing Guatemala with\n\n\nZambia on Government Effectiveness (ge), it is clear from the graph that the 90 percent confidence\n\n\nintervals overlap. This means that although Zambia has a higher estimate of governance than\n\n\nGuatemala, the difference between the two countries should not be considered as statistically\n\n\nsignificant. In contrast, Mauritius has a much higher estimate of Government Effectiveness (ge) than\n\n\neither Guatemala or Zambia. For these comparisons, 90 percent confidence intervals do not overlap and\n\n\nthe difference in estimates of governance can be considered statistically significant.\n\n\nLooking more systematically across all six aggregate indicators and all years since 1996, 61 percent\n\n\nof all pairwise comparisons of countries within an indicator and year will result in estimates of\n\n\ngovernance that are significantly different in the sense that the 90 percent confidence intervals do not\n\n\noverlap. There are some differences across indicators and over time. For example, looking across the six\n\n\naggregate indicators, the share of significant pairwise comparisons ranges from 52 percent (for Political\n\n\nStability and Absence of Violence/Terrorism) to 70 percent (for Voice and Accountability), in part\n\n\nreflecting the considerably smaller number of data sources that feed into the former versus the latter (23\n\n\nvs 10, see Annex 3). Looking over time and averaging across all six indicators, the share of significant\n\n\npairwise comparisons increased from 52 percent in 1996 to 62 percent in 2005, again in part reflecting a\n\n\nsteady increase in the number of WGI data sources over this period from 12 to 31 over this period (see\n\n\nTable 1). From 2005 onwards, the share of significant pairwise comparisons has fluctuated between 61\n\n\nand 65 percent. This indicates that while margins of error are both substantial and important to consider\n\n\nwhen comparing countries, it also is possible to use the WGI data to identify many significant pairwise\n\n\ndifferences in estimates\n\nMention: \"WGI data sources\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"inform\", \"summary\": \"The number of underlying sources rose from twelve to thirty-one, affecting the precision of comparisons.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001409:13:0", "occurrence_id": "6916e6c464dbfabda65355ab", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **c) Women**\n\n\n\n\n\n\n\n100%\n\n\n80%\n\n\n60%\n\n\n40%\n\n\n20%\n\n\n\n\n\n0%\n\n2009 2014 2019\nAgriculture and forestry Mining Food, beverage and tobacco\nTextile and apparel Wood Construction\nSales Transport Public administration\nOthers\n\n**Source:** Authors’ elaboration using data the CSES household survey.\n\n**3. Trade Policies and Trade Flows**\n\nCambodia's dedication to trade liberalization is apparent through the implementation of a\nseries of trade-related policies. Besides the measures noted, in 1996, the U.S. granted\nCambodia most favored nation (MFN) status, followed by the Generalized System of\nPreferences (GSP) in 1997, and the European Union's Everything-but-Arms (EBA)\nagreement in 2001. It is also a signatory of the 2006 Trade and Investment Framework\nAgreement (TIFA) with the U.S. These agreements facilitated intra-regional and\ninternational exchange of garment manufacturing and the entry of new investors from\nSoutheast Asia into Cambodia.\n\nOther measures further promoted private sector development, including the 1994 law on\ninvestment, which allowed 100% foreign-owned investment and provided guarantees against\nnationalization and price regulation. To facilitate foreign investment, the country created the\nCouncil for the Development of Cambodia (CDC), which functioned as a one-stop for\ninvestors. In recent years, Cambodia has also introduced significant cross-border trade\nfacilitation improvements, resulting in faster export and import procedures. Compared to\nother Southeast Asian countries, Cambodia is known for having the least trade restrictions in\nservices.\n\nTrade plays a critical role in Cambodia’s economy. Since Cambodia’s economic and political\ntransition of the 1990s, it has fostered an export-led economy. Since 2000, gross domestic\nproduct (GDP) and GDP per capita have increased significantly. In fact, several estimates\nhave projected that Cambodia will transition out of Least Developed Country (LDC) status\nas early as 2027. Table 3.1 summarizes Cambodia’s trade\n\nMention: \"CSES household survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Sectoral shares of female employment across three survey years are tabulated from this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001411:41:1", "occurrence_id": "75c2c8ffd38d40a2be21af3a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  (2021). _WHO global air quality guidelines: particulate matter (PM2.5 and_\n\n_PM10),_ _ozone,_ _nitrogen_ _dioxide,_ _sulfur_ _dioxide_ _and_ _carbon_ _monoxide_ . World\nHealth Organization.\n\n\nWorld Bank (2023a). World development indicators: Fraction of population 65 or\n\nolder. Accessed on May 17, 2023.\n\n\nWorld Bank (2023b). World Development Indicators: GNI per capita, Atlas\nmethod. Accessed on May 17, 2023.\n\n\nZivin, J. G. and Neidell, M. (2012). The impact of pollution on worker productiv\nity. _American Economic Review_, 102(7):3652–3673.\n\n\n40\n\nMention: \"World development indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The share of population aged 65 or older is drawn from this series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001411:41:1", "occurrence_id": "39357e5d70d1e22ee3ab8eda", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  (2021). _WHO global air quality guidelines: particulate matter (PM2.5 and_\n\n_PM10),_ _ozone,_ _nitrogen_ _dioxide,_ _sulfur_ _dioxide_ _and_ _carbon_ _monoxide_ . World\nHealth Organization.\n\n\nWorld Bank (2023a). World development indicators: Fraction of population 65 or\n\nolder. Accessed on May 17, 2023.\n\n\nWorld Bank (2023b). World Development Indicators: GNI per capita, Atlas\nmethod. Accessed on May 17, 2023.\n\n\nZivin, J. G. and Neidell, M. (2012). The impact of pollution on worker productiv\nity. _American Economic Review_, 102(7):3652–3673.\n\n\n40\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"GNI per capita on the Atlas method is taken from this series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001412:42:1", "occurrence_id": "a818c3bbab952f7654828149", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  the sample to competitive\n\nindustries at the 2-digit level. Table 17 reports that all of the results described previously\n\n- considering the entirety of the economy - persist. Additionally, we find a statistically\n\nsignificant negative correlation between exit rates and the concentration of BOS in com\npetitive industries. Our results suggest that industries with a higher presence of BOS are\n\ncharacterized by lower shares of young firms, lower levels of selection and higher market\n\nconcentration, all of which contribute to restricting business dynamism and firm life growth\n\n(Hsieh and Klenow, 2014). However, we do not find a statistically significant relationship\n\nbetween the employment share of BOS and entry rates, which would corroborate the nega\ntive relationship reported for entrepreneurship activity, measured by the employment share\n\nof young firms. In terms of the labor market, a positive correlation between the presence\n\nof BOS and job creation could be reflecting the ability of these firms to grow even in an\n\nenvironment of overall lower business dynamics. As we have shown in section 3, firms with\n\nstate participation have higher employment growth rates relative to private firms.\n\n\n16Since no information about firm output or sales is contained in RAIS, we compute the HHI using\nfirm-level employment.\n\n\n41\n\nMention: \"RAIS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"It contains no output or sales information, so concentration is computed from employment.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001415:3:1", "occurrence_id": "526fa1c08c7057cfa2ea86bf", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  (2019), and Vona et al.\n(2018). The taxonomy has also been adapted for various regions, including the European Union\n(Bowen and Hancké 2019), the Netherlands (Elliott et al. 2021), the United Kingdom (Valero et\nal. 2021), Organisation for Economic Cooperation and Development (OECD) member countries\n(OECD 2023), Viet Nam (Doan et al. 2023), and Argentina (de la Vega, Porto, and Cerimelo\n2024). After classifying occupations as green, studies delve into the specific skills and tasks\nrequired for green jobs, analyze trends in green job creation and distribution, and assess the broader\neconomic impacts, such as productivity, innovation, and growth, associated with the green\ntransition.\n\nHowever, two main issues make O*NET less relevant for worldwide use, particularly for green\njobs. First, O*NET was built in 2009, with the last major revision of the taxonomy completed in\n2011 (Dierdorff et al. 2011) and the associated reference book last updated in 2013 (O*NET 2013).\nThe literature on green jobs has expanded significantly since 2009. Second, O*NET is designed\nfor the U.S. labor market, identifying tasks within occupations based on the U.S. context. The tasks\nand skills required to perform these jobs depend on the production technology, which can differ\nsignificantly between the United States and other economies, such as low- and middle-income\ncountries.\n\nOur study aims to expand the inventory of green job titles by integrating a global perspective and\nincorporating contemporary sources. Our literature review comprised a search for articles\npublished after 2008 using Scopus and Web of Science—two leading bibliographic databases that\nare widely used by the academic community for accessing an extensive global collection of peerreviewed publications across various disciplines (Zhu and Liu 2020). The year 2008 marked a\n\n\n2\n\nMention: \"Scopus\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Articles published after 2008 were searched in this bibliographic database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001415:3:1", "occurrence_id": "1c3ba3e71f466a0f4f57d1e6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  (2019), and Vona et al.\n(2018). The taxonomy has also been adapted for various regions, including the European Union\n(Bowen and Hancké 2019), the Netherlands (Elliott et al. 2021), the United Kingdom (Valero et\nal. 2021), Organisation for Economic Cooperation and Development (OECD) member countries\n(OECD 2023), Viet Nam (Doan et al. 2023), and Argentina (de la Vega, Porto, and Cerimelo\n2024). After classifying occupations as green, studies delve into the specific skills and tasks\nrequired for green jobs, analyze trends in green job creation and distribution, and assess the broader\neconomic impacts, such as productivity, innovation, and growth, associated with the green\ntransition.\n\nHowever, two main issues make O*NET less relevant for worldwide use, particularly for green\njobs. First, O*NET was built in 2009, with the last major revision of the taxonomy completed in\n2011 (Dierdorff et al. 2011) and the associated reference book last updated in 2013 (O*NET 2013).\nThe literature on green jobs has expanded significantly since 2009. Second, O*NET is designed\nfor the U.S. labor market, identifying tasks within occupations based on the U.S. context. The tasks\nand skills required to perform these jobs depend on the production technology, which can differ\nsignificantly between the United States and other economies, such as low- and middle-income\ncountries.\n\nOur study aims to expand the inventory of green job titles by integrating a global perspective and\nincorporating contemporary sources. Our literature review comprised a search for articles\npublished after 2008 using Scopus and Web of Science—two leading bibliographic databases that\nare widely used by the academic community for accessing an extensive global collection of peerreviewed publications across various disciplines (Zhu and Liu 2020). The year 2008 marked a\n\n\n2\n\nMention: \"Web of Science\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The second bibliographic database searched for the review.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001417:5:1", "occurrence_id": "b94795d3c9634c8d79970f78", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  World\nEconomic Outlook of the International Monetary Fund, which contains dozens of variables on\nmacroeconomic indicators, and all data from the UN’s World Population Prospects, which\ncontains dozens of variables on population, health, and demographics. We compliment GDP data\nfrom the sources above with estimates from the Madison Database (Bolt and Van Zanden 2024).\nWe also use country and region classifications by the World Bank and data on political rights, civil\nliberties, and freedom status from the Freedom House.\n\nIn addition, we use remote sensing data available from the Google Earth Engine. We use data on\nnighttime lights, precipitation, temperature, impervious surface, cropland, the normalized\ndifference water/snow/vegetation index, and the enhanced vegetation index. While spatial\ncoverage of remote sensing variables is not a concern as they cover the earth’s entire surface,\ntemporal coverage is at times limited. Nighttime light data, for example, dates back to 1992. To fit\ninto this exercise, the remote sensing data need to be aggregated to the country-year level. We\nfirst aggregate them to annual data by calculating the mean, max, min, and standard deviation of\neach location (e.g., a pixel) over a year. Afterwards, we aggregate them spatially by taking the\nmean, max, min, and standard deviation of the annual data for a country. This gives 16 features\nfor each type of variable. Nighttime lights are also converted to a per capita level by dividing the\nsum of lights by the population size. We weigh each grid equally, and hence the indices reflect,\nfor example, the mean temperature in the territory of a country, not the mean temperature\nexperienced by a person in a country. Given that many of the variables impact welfare through\nagriculture, it is not clear that population weighting (which would give a dominance to urban areas)\nwould make them more related to welfare distributions. Yet we also add population-grid weighted\nestimates of temperature and\n\nMention: \"World Population Prospects\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Dozens of population, health and demographic variables are taken from this UN compilation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001417:5:1", "occurrence_id": "64c57688fcdeab2fb41659a0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  World\nEconomic Outlook of the International Monetary Fund, which contains dozens of variables on\nmacroeconomic indicators, and all data from the UN’s World Population Prospects, which\ncontains dozens of variables on population, health, and demographics. We compliment GDP data\nfrom the sources above with estimates from the Madison Database (Bolt and Van Zanden 2024).\nWe also use country and region classifications by the World Bank and data on political rights, civil\nliberties, and freedom status from the Freedom House.\n\nIn addition, we use remote sensing data available from the Google Earth Engine. We use data on\nnighttime lights, precipitation, temperature, impervious surface, cropland, the normalized\ndifference water/snow/vegetation index, and the enhanced vegetation index. While spatial\ncoverage of remote sensing variables is not a concern as they cover the earth’s entire surface,\ntemporal coverage is at times limited. Nighttime light data, for example, dates back to 1992. To fit\ninto this exercise, the remote sensing data need to be aggregated to the country-year level. We\nfirst aggregate them to annual data by calculating the mean, max, min, and standard deviation of\neach location (e.g., a pixel) over a year. Afterwards, we aggregate them spatially by taking the\nmean, max, min, and standard deviation of the annual data for a country. This gives 16 features\nfor each type of variable. Nighttime lights are also converted to a per capita level by dividing the\nsum of lights by the population size. We weigh each grid equally, and hence the indices reflect,\nfor example, the mean temperature in the territory of a country, not the mean temperature\nexperienced by a person in a country. Given that many of the variables impact welfare through\nagriculture, it is not clear that population weighting (which would give a dominance to urban areas)\nwould make them more related to welfare distributions. Yet we also add population-grid weighted\nestimates of temperature and\n\nMention: \"Madison Database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"GDP estimates complement the macro sources.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001417:5:1", "occurrence_id": "87ff3814dbcf41fe470bd93b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  World\nEconomic Outlook of the International Monetary Fund, which contains dozens of variables on\nmacroeconomic indicators, and all data from the UN’s World Population Prospects, which\ncontains dozens of variables on population, health, and demographics. We compliment GDP data\nfrom the sources above with estimates from the Madison Database (Bolt and Van Zanden 2024).\nWe also use country and region classifications by the World Bank and data on political rights, civil\nliberties, and freedom status from the Freedom House.\n\nIn addition, we use remote sensing data available from the Google Earth Engine. We use data on\nnighttime lights, precipitation, temperature, impervious surface, cropland, the normalized\ndifference water/snow/vegetation index, and the enhanced vegetation index. While spatial\ncoverage of remote sensing variables is not a concern as they cover the earth’s entire surface,\ntemporal coverage is at times limited. Nighttime light data, for example, dates back to 1992. To fit\ninto this exercise, the remote sensing data need to be aggregated to the country-year level. We\nfirst aggregate them to annual data by calculating the mean, max, min, and standard deviation of\neach location (e.g., a pixel) over a year. Afterwards, we aggregate them spatially by taking the\nmean, max, min, and standard deviation of the annual data for a country. This gives 16 features\nfor each type of variable. Nighttime lights are also converted to a per capita level by dividing the\nsum of lights by the population size. We weigh each grid equally, and hence the indices reflect,\nfor example, the mean temperature in the territory of a country, not the mean temperature\nexperienced by a person in a country. Given that many of the variables impact welfare through\nagriculture, it is not clear that population weighting (which would give a dominance to urban areas)\nwould make them more related to welfare distributions. Yet we also add population-grid weighted\nestimates of temperature and\n\nMention: \"Nighttime light data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its record begins only in 1992, so temporal coverage limits its use for long panels.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001418:5:0", "occurrence_id": "269767bd46f4e593e5b58c4f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ##### **2 Data and Methodology**\n\nBy way of background, zones and other cluster-based programs have steadily increased\n\n\nin South Asia. Starting from the first export processing zone (EPZ) in Gujarat in 1965,\n\n\nIndia has developed over 3,350 zones, according to the Ministry of Commerce and Industry.\n\n\nBangladesh, Nepal, and Bhutan have all announced plans to establish and expand Special\n\n\nEconomic Zones (SEZs), industrial districts and parks. The passage of SEZ Acts and\n\n\nPolicies signal more to come (Galal, 2021).\n\n\nHowever, despite the growing prevalence of these programs, a lack of data has ob\n\nstructed their systematic assessment. This paper is able to do so thanks to a unique\n\n\nWorld Bank survey designed to provide detailed information on firms within and outside\n\n\nzones. The survey covered 1,201 firms inside different types of zones and 1,166 firms out\n\nside zones in Bangladesh, Bhutan, India, and Nepal. For data collection, the sample frame\n\n\nfor firms outside zones was based on administrative data from government sources. <sup>1</sup> The\n\n\nsample frame for firms inside zones relied on a database collected from zone operators.\n\n\nThe same process was followed in all four countries. Fieldwork was conducted between\n\n\nSeptember 2020 and January 2021 and successfully surveyed 2,367 of the 2,400 target en\n\nterprises. Maps of the number and types of zones are presented in Figure 1. For the sake of\n\n\nsimplification, the term zone is used in the rest of the paper as a generic umbrella to cover\n\n\nall zone types, but when necessary, the specific zone type will be mentioned explicitly.\n\n\nTable 1 presents sample descriptives for firms inside and outside zones. The survey\n\n\nwas designed to ensure that firms outside the zones were from the same districts as the\n\n\nzones and representative of their district. Within each district, on average 52% of firms\n\n\nwere located inside zones. Most firms in the\n\nMention: \"World Bank survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"1,201 firms inside zones and 1,166 outside were interviewed across four countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001424:3:0", "occurrence_id": "3c80641736d45d1f758362f6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ##### **1. Introduction**\n\nHsieh and Klenow (2014) have documented that plants in the U.S. grow more over their\n\nlife cycle compared to plants in Mexico and India. <sup>1</sup> <sup>_,_</sup> <sup>2</sup> Less developed countries also face\nwidespread informality. <sup>3</sup> This paper argues that informality, arising from incomplete\n\nenforcement, distorts incentives for growth over the life cycle. <sup>4</sup> Since tax compliance\n\nincreases with plant size, informal plants may optimally choose to keep operating at\n\n\nsmaller scales to escape detection. Hence, we study the role of informality in explaining\n\nthe life cycle of plants. We ask how aggregate economic outcomes such as output and\n\nplant-size distribution, as well as the life cycle of plants, would change if informality\n\nwere reduced by improving enforcement or reducing the burden of formality.\n\n\nOne of the key challenges in studying how informality distorts plant life cycles is\n\n\nthe limited availability of plant-level panel data containing information on registration\n\nstatus. This study overcomes this challenge by analyzing Mexico’s census plant-level\n\ndata from Instituto Nacional de Estadística, Geografía e Informática (INEGI). The\n\ndata tracks plants over 20 years, from 1998 to 2018, and includes information about\n\n\nestablishments’ registration with different tax authorities. Therefore, unlike Hsieh and\n\n\nKlenow (2014), this allows for analysis of a balanced panel of registered and unregistered\n\nplants.\n\n\nIn our analysis, an informal establishment is a plant which is not registered with\n\neither the central tax authority or the social security administration. In contrast, we\n\nrefer to plants as formal if they are registered with either the central tax authority or\n\n\nthe social security administration. We first show that young formal plants are bigger in\n\nterms of employment\n\nMention: \"census plant-level\n\ndata\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Twenty years of plant tracks with registration status allow life cycles to be compared by formality.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001426:23:0", "occurrence_id": "61546f1f352eadf1cf3fbd76", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Four, of specific interest to us is how these numbers would differ when splitting the\n\n\nsample of studies according to the level of development of countries. Again here, we find\n\n\nimportant heterogeneity. While effects appear larger in developed countries for digital\n\n\nand energy studies, cross-sectoral and transport studies based on developing countries\n\n\nproduce larger elasticities. Notably, larger estimates are found when looking at specific\n\n\noutcomes in developing countries, such as the impact of digital investments on micro-level\n\n\noutput, at 0.16, or of transport investments on labor market outcomes, at 0.18.\n\n\nFinally, there is an important issue regarding how to interpret these elasticities for pol\n\nicy recommendations. Deciding on which policies or sectors to prioritize, and possibly\n\n\non potential financing strategies, would require that we translate elasticities into specific\n\n\nrates of returns (see Gardner and Henry (2023)). This in turn implies the need for infras\n\ntructure capital stock figures. Using a basic Cobb-Douglas production function approach\n\n\nfor illustration, the marginal rate of return for a specific type of infrastructure can be ap\n\nproximated by the following formula: _mrr_ = γ _In f_ . _In f_ <sup>_<u>Y</u>_</sup> <sup>, where γ</sup> <sup>_In f_</sup> <sup>is the elasticity,</sup> <sup>_Y_</sup> <sup>is GDP,</sup>\n\n\nand _In f_ is the stock of infrastructure.\n\n\nThere are however several difficulties involved in finding suitable proxies for _Inf_ . Re\n\ngarding public capital, different methods combine national account data, public budget\n\n\ndata, and information from the Private Participation in Infrastructure (PPI) database to es\n\ntimate infrastructure investments, which can then be used through inventory methods\n\nMention: \"Private Participation in Infrastructure (PPI) database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its project information is combined with accounts and budget data to estimate infrastructure investment.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001435:9:1", "occurrence_id": "88d7d314e8584b93bab3fa8d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: HC also\n\n\nrepresents the main data source for official poverty estimations in Paraguay, with poverty being\n\n\nmeasured on an annual basis. For our purposes, we use data from 2004 to 2019, from which\n\n\nwe generate a pooled cross-sectional data set of 379,170 individual observations (Table A.1 in\n\n\nAppendix A). Since 2017, the EPHC has been collected on a quarterly basis, while previously,\n\n\nthe EPHC was collected only on the fourth quarter of each year. While data for the years 2020\n\n2022 is available, the observed increases in poverty or decreases in incomes during the pandemic\n\n\nand subsequent economic recession may not be primarily attributed to weather shocks, and\n\n\nincluding such data may introduce contamination in our estimations. Therefore, our primary\n\n\nanalyses in this study focuses on pre-pandemic years. It is also important to note that the\n\n\npoverty estimates used in this study are based on weights derived from national projections,\n\n\n2015 revision. In 2022, the statistical office conducted a Population and Housing Census, and\n\n\nupdated the weights for household surveys, but at the completion of this study, data for 2022\n\n\nand 2023 were only available. Therefore, the historical series used in this study are not strictly\n\n\n7\n\nMention: \"Population and Housing Census\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"It was used to update the weights applied to household surveys, though recent data were not yet available.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001441:21:0", "occurrence_id": "fbf27e3fdd70f7a695be4bae", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Suryahadi, A. (2001). International economic integration and labor markets: The case of Indonesia.\n\n\n**Annex A: Data**\n\n_Sakernas_\n\nThe Sakernas data, also known as the National Labor Force Survey (Survei Angkatan Kerja\nNasional), is a comprehensive dataset for Indonesia. It is a survey conducted by the Indonesian\nCentral Bureau of Statistics (Badan Pusat Statistik or BPS) and provides detailed information on\nvarious aspects of the labor force in the country. The survey is typically conducted bi-annually and\ncollects data from households across Indonesia.\n\nThe Sakernas data covers a wide range of topics related to employment, unemployment, and labor\nforce participation. It includes information on individuals' demographic characteristics,\neducational attainment, occupation, industry, employment status, working hours, wages, and other\nrelevant labor-related variables.\n\nFor this study, we built a repeated cross-sectional dataset between 1993 and 2014 using the August\nrounds of data. We harmonized industry codes to the Klasifikasi Baku Lapangan Usaha Indonesia\n(KBLI) 1990 classification, the Indonesian industrial classification based on ISIC Revision 2.\n\n\n_UN COMTRADE_\n\nThe UN Comtrade dataset, officially known as the United Nations Commodity Trade Statistics\nDatabase, is a comprehensive international trade database maintained by the United Nations\nStatistics Division (UNSD). This dataset provides detailed information on global merchandise\ntrade, encompassing the import and export of goods and services between countries and regions.\nIn this study, we limit the analysis to trade in merchandise only, since data on trade in services is\nvery limited for the pre-2000s period.\n\n\n_UN Statistics Division (ISIC-HS Concordances)_\n\nThe UN Statistics Division (ISIC – HS) concordances refer to a set of tables or mapping systems\nthat establish a link between two international classification systems: the International Standard\nIndustrial Classification of All Economic Activities (ISIC\n\nMention: \"Sakernas data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"August rounds are used, with industry codes harmonised to the Indonesian classification.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001441:21:0", "occurrence_id": "656aa0a196fbe1d1f13ddef1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Suryahadi, A. (2001). International economic integration and labor markets: The case of Indonesia.\n\n\n**Annex A: Data**\n\n_Sakernas_\n\nThe Sakernas data, also known as the National Labor Force Survey (Survei Angkatan Kerja\nNasional), is a comprehensive dataset for Indonesia. It is a survey conducted by the Indonesian\nCentral Bureau of Statistics (Badan Pusat Statistik or BPS) and provides detailed information on\nvarious aspects of the labor force in the country. The survey is typically conducted bi-annually and\ncollects data from households across Indonesia.\n\nThe Sakernas data covers a wide range of topics related to employment, unemployment, and labor\nforce participation. It includes information on individuals' demographic characteristics,\neducational attainment, occupation, industry, employment status, working hours, wages, and other\nrelevant labor-related variables.\n\nFor this study, we built a repeated cross-sectional dataset between 1993 and 2014 using the August\nrounds of data. We harmonized industry codes to the Klasifikasi Baku Lapangan Usaha Indonesia\n(KBLI) 1990 classification, the Indonesian industrial classification based on ISIC Revision 2.\n\n\n_UN COMTRADE_\n\nThe UN Comtrade dataset, officially known as the United Nations Commodity Trade Statistics\nDatabase, is a comprehensive international trade database maintained by the United Nations\nStatistics Division (UNSD). This dataset provides detailed information on global merchandise\ntrade, encompassing the import and export of goods and services between countries and regions.\nIn this study, we limit the analysis to trade in merchandise only, since data on trade in services is\nvery limited for the pre-2000s period.\n\n\n_UN Statistics Division (ISIC-HS Concordances)_\n\nThe UN Statistics Division (ISIC – HS) concordances refer to a set of tables or mapping systems\nthat establish a link between two international classification systems: the International Standard\nIndustrial Classification of All Economic Activities (ISIC\n\nMention: \"National Labor Force Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"The English name and its producing statistical office are given for the dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001441:21:0", "occurrence_id": "604d25ed67b27c2955cb7d29", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Suryahadi, A. (2001). International economic integration and labor markets: The case of Indonesia.\n\n\n**Annex A: Data**\n\n_Sakernas_\n\nThe Sakernas data, also known as the National Labor Force Survey (Survei Angkatan Kerja\nNasional), is a comprehensive dataset for Indonesia. It is a survey conducted by the Indonesian\nCentral Bureau of Statistics (Badan Pusat Statistik or BPS) and provides detailed information on\nvarious aspects of the labor force in the country. The survey is typically conducted bi-annually and\ncollects data from households across Indonesia.\n\nThe Sakernas data covers a wide range of topics related to employment, unemployment, and labor\nforce participation. It includes information on individuals' demographic characteristics,\neducational attainment, occupation, industry, employment status, working hours, wages, and other\nrelevant labor-related variables.\n\nFor this study, we built a repeated cross-sectional dataset between 1993 and 2014 using the August\nrounds of data. We harmonized industry codes to the Klasifikasi Baku Lapangan Usaha Indonesia\n(KBLI) 1990 classification, the Indonesian industrial classification based on ISIC Revision 2.\n\n\n_UN COMTRADE_\n\nThe UN Comtrade dataset, officially known as the United Nations Commodity Trade Statistics\nDatabase, is a comprehensive international trade database maintained by the United Nations\nStatistics Division (UNSD). This dataset provides detailed information on global merchandise\ntrade, encompassing the import and export of goods and services between countries and regions.\nIn this study, we limit the analysis to trade in merchandise only, since data on trade in services is\nvery limited for the pre-2000s period.\n\n\n_UN Statistics Division (ISIC-HS Concordances)_\n\nThe UN Statistics Division (ISIC – HS) concordances refer to a set of tables or mapping systems\nthat establish a link between two international classification systems: the International Standard\nIndustrial Classification of All Economic Activities (ISIC\n\nMention: \"Survei Angkatan Kerja\nNasional\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"The local name is supplied as the expansion of the acronym used throughout.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001441:21:0", "occurrence_id": "82269a3d198401b0f3827d22", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Suryahadi, A. (2001). International economic integration and labor markets: The case of Indonesia.\n\n\n**Annex A: Data**\n\n_Sakernas_\n\nThe Sakernas data, also known as the National Labor Force Survey (Survei Angkatan Kerja\nNasional), is a comprehensive dataset for Indonesia. It is a survey conducted by the Indonesian\nCentral Bureau of Statistics (Badan Pusat Statistik or BPS) and provides detailed information on\nvarious aspects of the labor force in the country. The survey is typically conducted bi-annually and\ncollects data from households across Indonesia.\n\nThe Sakernas data covers a wide range of topics related to employment, unemployment, and labor\nforce participation. It includes information on individuals' demographic characteristics,\neducational attainment, occupation, industry, employment status, working hours, wages, and other\nrelevant labor-related variables.\n\nFor this study, we built a repeated cross-sectional dataset between 1993 and 2014 using the August\nrounds of data. We harmonized industry codes to the Klasifikasi Baku Lapangan Usaha Indonesia\n(KBLI) 1990 classification, the Indonesian industrial classification based on ISIC Revision 2.\n\n\n_UN COMTRADE_\n\nThe UN Comtrade dataset, officially known as the United Nations Commodity Trade Statistics\nDatabase, is a comprehensive international trade database maintained by the United Nations\nStatistics Division (UNSD). This dataset provides detailed information on global merchandise\ntrade, encompassing the import and export of goods and services between countries and regions.\nIn this study, we limit the analysis to trade in merchandise only, since data on trade in services is\nvery limited for the pre-2000s period.\n\n\n_UN Statistics Division (ISIC-HS Concordances)_\n\nThe UN Statistics Division (ISIC – HS) concordances refer to a set of tables or mapping systems\nthat establish a link between two international classification systems: the International Standard\nIndustrial Classification of All Economic Activities (ISIC\n\nMention: \"UN Comtrade dataset\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Detailed commodity trade flows are drawn from this international database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001441:21:0", "occurrence_id": "9b6fb0eefda5be40cfb957b1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Suryahadi, A. (2001). International economic integration and labor markets: The case of Indonesia.\n\n\n**Annex A: Data**\n\n_Sakernas_\n\nThe Sakernas data, also known as the National Labor Force Survey (Survei Angkatan Kerja\nNasional), is a comprehensive dataset for Indonesia. It is a survey conducted by the Indonesian\nCentral Bureau of Statistics (Badan Pusat Statistik or BPS) and provides detailed information on\nvarious aspects of the labor force in the country. The survey is typically conducted bi-annually and\ncollects data from households across Indonesia.\n\nThe Sakernas data covers a wide range of topics related to employment, unemployment, and labor\nforce participation. It includes information on individuals' demographic characteristics,\neducational attainment, occupation, industry, employment status, working hours, wages, and other\nrelevant labor-related variables.\n\nFor this study, we built a repeated cross-sectional dataset between 1993 and 2014 using the August\nrounds of data. We harmonized industry codes to the Klasifikasi Baku Lapangan Usaha Indonesia\n(KBLI) 1990 classification, the Indonesian industrial classification based on ISIC Revision 2.\n\n\n_UN COMTRADE_\n\nThe UN Comtrade dataset, officially known as the United Nations Commodity Trade Statistics\nDatabase, is a comprehensive international trade database maintained by the United Nations\nStatistics Division (UNSD). This dataset provides detailed information on global merchandise\ntrade, encompassing the import and export of goods and services between countries and regions.\nIn this study, we limit the analysis to trade in merchandise only, since data on trade in services is\nvery limited for the pre-2000s period.\n\n\n_UN Statistics Division (ISIC-HS Concordances)_\n\nThe UN Statistics Division (ISIC – HS) concordances refer to a set of tables or mapping systems\nthat establish a link between two international classification systems: the International Standard\nIndustrial Classification of All Economic Activities (ISIC\n\nMention: \"United Nations Commodity Trade Statistics\nDatabase\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"The formal title and its custodian division are given for the trade dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001447:17:1", "occurrence_id": "8644edc0b211d177cf395a50", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  where beliefs that educational and\n\nprofessional advancement are not based on meritocracy remain widespread. Indeed, data from\n\nthe 2016 round of the LiTS reveal that more than two-thirds of adults in the region report that\n\nhaving informal connections is important to secure a good government job, and almost 40\n\n\n16\n\nMention: \"data from\n\nthe 2016 round of the LiTS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Shares reporting that informal connections matter for a government job come from this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001448:17:0", "occurrence_id": "7557f2352b3c84b6c78be6a3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: While unprecedented for containerized shipping, freight rate spikes have periodically occurred in other\nshipping segments like dry bulk prior to the 2008 financial crisis when soaring commodity demand over a\nmulti-year span quintupled the Baltic Dry Index (Figure 12). This historical pattern suggests that shipping\nrates have a “low normal” state, interrupted by bursts of high rates when capacity cannot match demand\nsurges.\n\n##### **Conceptual Framework of Stress and Rate Formation**\n\n\nHistorical freight rate patterns suggest two market regimes: Under “normal” conditions, shipping capacity\naccommodates shipper demand. Complex price mechanisms exist, which have, in some regions, replaced\nsystems such as freight rate conferences, where prices were fixed. The industry structure is concentrated\nwith six major carriers providing the bulk of the capacity in global container shipping. The existing global\noperational alliance system, accepted by regulatory authorities, allows carriers to consolidate and optimize\nvessel utilization with partners while alliance members remain commercially independent.\n\n\nThe nature of the business and investment size limits the number of entrants impacting the competitive\nenvironment in oligopolistic behavior. However, in a normal regime and in the absence of price collusion,\nlines are price takers, and rates correspond to the marginal price considering the imbalance in demand (full\ncontainers) across directions. The conventional wisdom in the industry is that shipping lines cannot recover\ntheir investment but for the period of tensions when rate spikes happen. Classical models of price formation\nsuch as Bertrand’s apply to this regime.\n\n\nIn times of capacity tensions, shippers compete for scarce remaining slots on vessels, becoming rate-takers\nrather than setters. This is especially true for shippers not engaged in longer-term contracts. Willingnessto-pay dynamics shapes the markets: price formation à la Ramsey-Boiteux-Dessus hinging on freight’s ad\nvalorem markup over product value. Lower value goods suffer more acute increases proportional to cargo\nvalue. This first manifests for export loads out of\n\nMention: \"Baltic Dry Index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its quintupling before the financial crisis is the historical comparator for rate spikes.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001455:35:1", "occurrence_id": "82979a97b71b3cfeda6a2ffe", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 031 0.149 10,776\nPrimary education 0.018 0.100 10,776\nSecondary education 0.125 0.142 10,776\nMore than secondary education 0.276 0.447 10,776\nDecision making index 0.330 0.336 11,763\nAttitude toward domestic violence index 0.188 0.192 11,763\nAge gap with partner 3.062 1.507 6,760\nAge at marriage 21.274 3.178 6,228\n\n\nNote: Data on cohorts of women from Bangladesh, India, Maldives, Nepal DHS. Decision\nmaking index based on the sum of decisions in which women are involved individually\nor jointly. Attitude toward domestic violence index based on the sum of statements\nrespondents believe violence is justified.\n\n\n34\n\nMention: \"Data on cohorts of women\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Education, decision-making and attitudes to domestic violence are summarised for women's cohorts.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001459:44:0", "occurrence_id": "4f46f457cbca1b7cdfc1c0d8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Appendix B** **Data and Additional Tables and Figures**\n\n\n**Data** **Sources** The highly disaggregated data on the tariffs and AVEs of import-related border NTMs\n\n\ndata for importer-exporter-products at the Harmonized System (HS) 6-digit level are from Kee and Nicita\n\n\n(2022). These data reflect the cross-sectional trade policy pattern in 2018. There are 44 importing countries\n\n\nand 117 exporting countries in our sample, including both developed and developing countries, as well\n\n\nas 5,124 HS-6-digit products.\n\n\nThe tariffs data are obtained from WITS. A further data issue concerns the reporting of preferential\n\n\ntariffs in WITS. As noted by Teti (2024), WITS may in some cases report the MFN tariff as the effectively\n\n\napplied tariff when preferential tariffs are not reported by the importing country. To mitigate this con\n\ncern, we follow a procedure similar to Teti (2024) and use adjacent-year information to recover missing\n\n\npreferential tariffs whenever possible.\n\n\nDue to the lack of comparable data on NTMs across countries and over time, this paper uses cross\n\nsectional variations to study the relationship between tariffs and NTMs. Despite missing the temporal\n\n\nvariation, the broad coverage of our data across countries and products, coupled with a special focus on\n\n\nborder NTMs, enables us examining the factors influencing the relationship between tariffs and NTMs in\n\ncomprehensively. <sup>42</sup> Furthermore, most NTMs are regulations and laws which seldom change from year\n\n\nto year, thereby reducing the impact of missing time-series variation on our identification strategy.\n\n\nOther data used in this paper are the updated WTO data set on the content of PTAs (Hofmann et al.\n\n\n2017), the WTO NTMs Notification Database, the World Bank World Development Indicators, GVC In\n\ndicators (Fernandes et al. 2022)\n\nMention: \"WTO data set\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Coded content of preferential trade agreements is taken from this updated dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001459:44:0", "occurrence_id": "c5fda8489318557812568212", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Appendix B** **Data and Additional Tables and Figures**\n\n\n**Data** **Sources** The highly disaggregated data on the tariffs and AVEs of import-related border NTMs\n\n\ndata for importer-exporter-products at the Harmonized System (HS) 6-digit level are from Kee and Nicita\n\n\n(2022). These data reflect the cross-sectional trade policy pattern in 2018. There are 44 importing countries\n\n\nand 117 exporting countries in our sample, including both developed and developing countries, as well\n\n\nas 5,124 HS-6-digit products.\n\n\nThe tariffs data are obtained from WITS. A further data issue concerns the reporting of preferential\n\n\ntariffs in WITS. As noted by Teti (2024), WITS may in some cases report the MFN tariff as the effectively\n\n\napplied tariff when preferential tariffs are not reported by the importing country. To mitigate this con\n\ncern, we follow a procedure similar to Teti (2024) and use adjacent-year information to recover missing\n\n\npreferential tariffs whenever possible.\n\n\nDue to the lack of comparable data on NTMs across countries and over time, this paper uses cross\n\nsectional variations to study the relationship between tariffs and NTMs. Despite missing the temporal\n\n\nvariation, the broad coverage of our data across countries and products, coupled with a special focus on\n\n\nborder NTMs, enables us examining the factors influencing the relationship between tariffs and NTMs in\n\ncomprehensively. <sup>42</sup> Furthermore, most NTMs are regulations and laws which seldom change from year\n\n\nto year, thereby reducing the impact of missing time-series variation on our identification strategy.\n\n\nOther data used in this paper are the updated WTO data set on the content of PTAs (Hofmann et al.\n\n\n2017), the WTO NTMs Notification Database, the World Bank World Development Indicators, GVC In\n\ndicators (Fernandes et al. 2022)\n\nMention: \"WTO NTMs Notification Database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Notifications of non-tariff measures are drawn from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001459:44:0", "occurrence_id": "12f2984bedf6a3f648667d08", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Appendix B** **Data and Additional Tables and Figures**\n\n\n**Data** **Sources** The highly disaggregated data on the tariffs and AVEs of import-related border NTMs\n\n\ndata for importer-exporter-products at the Harmonized System (HS) 6-digit level are from Kee and Nicita\n\n\n(2022). These data reflect the cross-sectional trade policy pattern in 2018. There are 44 importing countries\n\n\nand 117 exporting countries in our sample, including both developed and developing countries, as well\n\n\nas 5,124 HS-6-digit products.\n\n\nThe tariffs data are obtained from WITS. A further data issue concerns the reporting of preferential\n\n\ntariffs in WITS. As noted by Teti (2024), WITS may in some cases report the MFN tariff as the effectively\n\n\napplied tariff when preferential tariffs are not reported by the importing country. To mitigate this con\n\ncern, we follow a procedure similar to Teti (2024) and use adjacent-year information to recover missing\n\n\npreferential tariffs whenever possible.\n\n\nDue to the lack of comparable data on NTMs across countries and over time, this paper uses cross\n\nsectional variations to study the relationship between tariffs and NTMs. Despite missing the temporal\n\n\nvariation, the broad coverage of our data across countries and products, coupled with a special focus on\n\n\nborder NTMs, enables us examining the factors influencing the relationship between tariffs and NTMs in\n\ncomprehensively. <sup>42</sup> Furthermore, most NTMs are regulations and laws which seldom change from year\n\n\nto year, thereby reducing the impact of missing time-series variation on our identification strategy.\n\n\nOther data used in this paper are the updated WTO data set on the content of PTAs (Hofmann et al.\n\n\n2017), the WTO NTMs Notification Database, the World Bank World Development Indicators, GVC In\n\ndicators (Fernandes et al. 2022)\n\nMention: \"World Bank World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Standard macro controls are taken from these indicators.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001459:44:0", "occurrence_id": "85e13e5f560d9ff7f6e3a97a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Appendix B** **Data and Additional Tables and Figures**\n\n\n**Data** **Sources** The highly disaggregated data on the tariffs and AVEs of import-related border NTMs\n\n\ndata for importer-exporter-products at the Harmonized System (HS) 6-digit level are from Kee and Nicita\n\n\n(2022). These data reflect the cross-sectional trade policy pattern in 2018. There are 44 importing countries\n\n\nand 117 exporting countries in our sample, including both developed and developing countries, as well\n\n\nas 5,124 HS-6-digit products.\n\n\nThe tariffs data are obtained from WITS. A further data issue concerns the reporting of preferential\n\n\ntariffs in WITS. As noted by Teti (2024), WITS may in some cases report the MFN tariff as the effectively\n\n\napplied tariff when preferential tariffs are not reported by the importing country. To mitigate this con\n\ncern, we follow a procedure similar to Teti (2024) and use adjacent-year information to recover missing\n\n\npreferential tariffs whenever possible.\n\n\nDue to the lack of comparable data on NTMs across countries and over time, this paper uses cross\n\nsectional variations to study the relationship between tariffs and NTMs. Despite missing the temporal\n\n\nvariation, the broad coverage of our data across countries and products, coupled with a special focus on\n\n\nborder NTMs, enables us examining the factors influencing the relationship between tariffs and NTMs in\n\ncomprehensively. <sup>42</sup> Furthermore, most NTMs are regulations and laws which seldom change from year\n\n\nto year, thereby reducing the impact of missing time-series variation on our identification strategy.\n\n\nOther data used in this paper are the updated WTO data set on the content of PTAs (Hofmann et al.\n\n\n2017), the WTO NTMs Notification Database, the World Bank World Development Indicators, GVC In\n\ndicators (Fernandes et al. 2022)\n\nMention: \"GVC In\n\ndicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Global value chain participation measures are added from this indicator set.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001464:3:1", "occurrence_id": "c46fb4f5cae6fa6b5d1d5996", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  unfair impacts on competitiveness, particularly with respect to\n\n\nemissions-intensive and trade-exposed (EITE) industries.\n\n\nAs a consequence, policy makers in several countries with more ambitious carbon pricing\n\n\npolicies in place have linked the emissions price to output-based allocations of emission per\nmits for their EITE industries. <sup>3</sup> Output-based rebating of revenues from emissions pricing\n\n\nacts as an implicit production subsidy (Fischer, 2001; B¨ohringer et al., 1998), and thus may\n\n\n1As of mid-2024, less than a quarter of global emissions are subject to an emissions price. See `[https:](https://carbonpricingdashboard.worldbank.org/)`\n`[//carbonpricingdashboard.worldbank.org/](https://carbonpricingdashboard.worldbank.org/)` .\n\n2As of mid-2024, the World Bank in its carbon pricing dashboard reports a price range of US$0.46/tCO2to\nUS$167/tCO2for those countries that charge emissions prices.\n\n3For example, output-based allocations of emission permits are used in Canada ( `[https:](https://laws-lois.justice.gc.ca/eng/acts/g-11.55/)`\n`[//laws-lois.justice.gc.ca/eng/acts/g-11.55/](https://laws-lois.justice.gc.ca/eng/acts/g-11.55/)`, California ( `[https://ww2.arb.ca.gov/our-work/](https://ww2.arb.ca.gov/our-work/programs/cap-and-\n\nMention: \"carbon pricing dashboard\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It reports the observed range of emissions prices across countries as of mid-2024.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001471:70:5", "occurrence_id": "a3a24b56ddcc8f67a91fe93a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: mark></u> <u><mark>361</mark></u> <u><mark>361</mark></u> <u><mark>361</mark></u> <u><mark>92</mark></u>\n<u>Note: * p<0.10, ** p<0.05, *** p<0.01. Standard errors are clustered at the country level. Data from the World Bank's World</u>\nDevelopment Indicators (WDI), Open Data Watch (ODIN), Global Data Barometer (GDB), and Open Data Barometer (ODB).\nIn cases where data are missing for a particular covariate, the data are imputed forward using the nearest available value.\nEstimates with country fixed effects not available for the Global Data Barometer, because the indicator contains only one time\nperiod.\n\n\n\n69\n\nMention: \"Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Country covariates in the regressions come from this indicator series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001471:70:5", "occurrence_id": "f13e37b965671604470c39d8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: mark></u> <u><mark>361</mark></u> <u><mark>361</mark></u> <u><mark>361</mark></u> <u><mark>92</mark></u>\n<u>Note: * p<0.10, ** p<0.05, *** p<0.01. Standard errors are clustered at the country level. Data from the World Bank's World</u>\nDevelopment Indicators (WDI), Open Data Watch (ODIN), Global Data Barometer (GDB), and Open Data Barometer (ODB).\nIn cases where data are missing for a particular covariate, the data are imputed forward using the nearest available value.\nEstimates with country fixed effects not available for the Global Data Barometer, because the indicator contains only one time\nperiod.\n\n\n\n69\n\nMention: \"Open Data Watch\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its open data readiness indicator is one of the outcomes compared.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001471:70:5", "occurrence_id": "3232ba8f4b21c2b673c5ba97", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: mark></u> <u><mark>361</mark></u> <u><mark>361</mark></u> <u><mark>361</mark></u> <u><mark>92</mark></u>\n<u>Note: * p<0.10, ** p<0.05, *** p<0.01. Standard errors are clustered at the country level. Data from the World Bank's World</u>\nDevelopment Indicators (WDI), Open Data Watch (ODIN), Global Data Barometer (GDB), and Open Data Barometer (ODB).\nIn cases where data are missing for a particular covariate, the data are imputed forward using the nearest available value.\nEstimates with country fixed effects not available for the Global Data Barometer, because the indicator contains only one time\nperiod.\n\n\n\n69\n\nMention: \"Global Data Barometer\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Country fixed effects cannot be estimated for it because of how the indicator is constructed.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001471:70:5", "occurrence_id": "67e892956a377ddbc03daeb8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: mark></u> <u><mark>361</mark></u> <u><mark>361</mark></u> <u><mark>361</mark></u> <u><mark>92</mark></u>\n<u>Note: * p<0.10, ** p<0.05, *** p<0.01. Standard errors are clustered at the country level. Data from the World Bank's World</u>\nDevelopment Indicators (WDI), Open Data Watch (ODIN), Global Data Barometer (GDB), and Open Data Barometer (ODB).\nIn cases where data are missing for a particular covariate, the data are imputed forward using the nearest available value.\nEstimates with country fixed effects not available for the Global Data Barometer, because the indicator contains only one time\nperiod.\n\n\n\n69\n\nMention: \"Open Data Barometer\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its open data scores form another outcome in the comparison.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001475:12:4", "occurrence_id": "4b7563ef0d3155defedb1853", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: >**7.2**<br>0.00<br>**472.6**<br>**148.2**<br>0.00<br>(1532.5)<br>(760.4)<br> <br>**711**<br>**663**<br>|\n|<br> <br> <br> <br> % of enterprises in presumptive tax net<br>**44.1**<br>**29.9**<br>0.01<br>Average presumptive tax (birr, imputed)<br>**641.9**<br>**224.7**<br>0.00<br>(1825.3)<br>(724.7)<br> <br> % in sectors with tax-free threshold<br>68.6<br>74.6<br>0.26<br>**Number of enterprises**<br>**478**<br>**458**<br>|<br> <br> <br>**61.0**<br>**16.8**<br>0.00<br>**1222.0**<br>**262.5**<br>0.00<br>(472.6)<br>(148.2)<br> <br>**63.2**<br>**91.9**<br>0.00<br>**498**<br>**449**<br>|\n\n\nSource: Authors’ calculations using ESS 2018/2019 (period 1) and ESS2021\n\nMention: \"ESS2021\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Enterprise survey results are compared with the earlier 2018/2019 period.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001476:4:0", "occurrence_id": "bf75ba272510e4d88ecbb579", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: agricultural trade policy that encourages private investment, and investments in public goods that\nencourage productivity growth (World Bank, 2019).\n\n\n**Figure 2: Poverty among agricultural households in 2014**\n\n\n\n\n\n\n\n\n\n\n\n_Source:_ 2014 Timor-Leste Survey of Living Standards\n\n\nNotwithstanding the role that agriculture can potentially play in reducing poverty in Timor-Leste,\nthe country has one of the least developed agricultural production systems in the world. All\nsubsectors are dominated by subsistence systems and by farmers who are very risk averse\n(Denning 2013; World Bank 2017). Only 30 percent of the country’s arable land is currently under\ncrop production and grazing, and the sector’s productivity has been very low. Agricultural\nproductivity as measured by the value of crop output per hectare in Timor-Leste is well below that\nof other small island developing states and has fallen below the average for other low-income,\nfood-deficit countries in recent years (Figure 3a). Also, agricultural value added per worker has\nbeen falling since 2012 (Figure 3b). The estimated 64,000 metric tons of maize and 65,000 metric\ntons of rice produced in 2015 are 30 percent below the five-year average and well below the\ncountry’s estimated requirements (245,000 metric tons) of these staples (FAO statistics). The\ncountry’s cereal yield is also among the lowest in the East Asia and Pacific region (UN Women\nand World Bank 2018). The agriculture sector is therefore underperforming in terms of its\ncontribution to food security, sufficient livelihood, and overall economic growth.\n\n\n3\n\nMention: \"2014 Timor-Leste Survey of Living Standards\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Poverty among agricultural households is measured from this living standards survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001476:4:0", "occurrence_id": "b26afc9706dea570f60ae547", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: agricultural trade policy that encourages private investment, and investments in public goods that\nencourage productivity growth (World Bank, 2019).\n\n\n**Figure 2: Poverty among agricultural households in 2014**\n\n\n\n\n\n\n\n\n\n\n\n_Source:_ 2014 Timor-Leste Survey of Living Standards\n\n\nNotwithstanding the role that agriculture can potentially play in reducing poverty in Timor-Leste,\nthe country has one of the least developed agricultural production systems in the world. All\nsubsectors are dominated by subsistence systems and by farmers who are very risk averse\n(Denning 2013; World Bank 2017). Only 30 percent of the country’s arable land is currently under\ncrop production and grazing, and the sector’s productivity has been very low. Agricultural\nproductivity as measured by the value of crop output per hectare in Timor-Leste is well below that\nof other small island developing states and has fallen below the average for other low-income,\nfood-deficit countries in recent years (Figure 3a). Also, agricultural value added per worker has\nbeen falling since 2012 (Figure 3b). The estimated 64,000 metric tons of maize and 65,000 metric\ntons of rice produced in 2015 are 30 percent below the five-year average and well below the\ncountry’s estimated requirements (245,000 metric tons) of these staples (FAO statistics). The\ncountry’s cereal yield is also among the lowest in the East Asia and Pacific region (UN Women\nand World Bank 2018). The agriculture sector is therefore underperforming in terms of its\ncontribution to food security, sufficient livelihood, and overall economic growth.\n\n\n3\n\nMention: \"FAO statistics\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Maize and rice production estimates are compared with the five-year average and requirements.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001478:4:1", "occurrence_id": "421a59103987bbd4eefea934", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  are fixed at birth and outside an individual’s control, IOp\nestimates are often seen as a lower bound estimate of true IOp (Ferreira et al. 2011). As\noften pointed out, when the set of observed circumstances is limited, the degree of\nunderestimation is large and lower bound estimates of IOp end up being of limited use\nfor policy (Kanbur & Wagstaff 2014).\n\n\n3 The negative relationship between IOp and growth has been documented using variation over time and\nacross regions of the U.S. (Marrero et al. 2013, Bradbury et al. 2016) and in Europe (Carranza 2020). Evidence\nincluding developing countries is less conclusive and constrained by data limitations (Ferreira et al. 2018).\n\n4 In another noteworthy attempt, Ferreira et al. (2018) estimate IOp across 75 countries between 1980-2005,\nincluding 22 in Sub-Saharan Africa using data from the Demographic and Health Surveys (DHS). However,\nthere are two large disadvantages to using the DHS to estimate IOp: welfare is measured using a wealth\nindex because household consumption is not available, and parental information such as parents’\neducation is unavailable. Their IOp estimates in SSA range from 1-39 percent, though excluding parental\ninformation likely leads to substantial downward bias. A World Bank report (2022a) implemented an IOp\nanalysis in 5 countries in Southern Africa, though data limitations meant it was also unable to consider\nparental characteristics. This report found that IOp ranged from 15-26 percent, and increased to 46 percent\nin South Africa after race was included as a circumstance. Other studies examine drivers of inequality more\ngenerally – for example, Tetteh-Baah et al. (2024) use the DHS to calculate horizontal inequality in various\noutcomes across Africa based on location, ethnicity, gender, and religion – but they often do not include\nincome or consumption as an outcome or incorporate information on\n\nMention: \"Demographic and Health Surveys\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"A cited study uses it across 75 countries, though welfare is proxied by a wealth index and parental data are missing.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001481:7:1", "occurrence_id": "27ecb9b4d2f488692878366a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <br>10<br>20<br>30<br>40<br>50<br>**Employment rate**<br>High imports<br>Low imports|\n\n\n\nSource: Own elaboration based on Population Census, Integrated Public Use Microdata Series and Customs data. Notes: (1) The\nsample is restricted to the urban and semi-urban municipalities with complete data in the whole period. (2) A municipality has\nrelatively high exports (imports) if its exports (imports) per 2000 worker ratio in 2015 is above the median across municipalities\nand low exports (imports) if its ratio lies below the median.\n\n\nThird, a municipality’s trade integration could strongly matter for migration patterns, both across\ndomestic municipalities and with other countries, specifically the United States. More trade opportunities\nin a municipality may deter migration outflows, while attracting more migrants from other countries or\npulling back Mexican migrants from the United States. In fact, the number of unauthorized migrants from\nMexico to the United States declined from 6.9 million in 2007 to 5.6 million by 2015 (Gonzalez-Barrera\nand Krogstad 2019). Trade in a municipality may also have effects on remittances and the working age\npopulation. We therefore investigate the migration channel and again detect a clear difference between\nmunicipalities that are highly integrated in trade and those that are not. The average number of\nimmigrants from other Mexican states in municipalities with high exports per worker is 142, while that in\nmunicipalities with low exports per worker is less than half this magnitude, namely 63 (Figure 4, upper\nleft panel). Again, the relative difference is larger for imports (upper right panel). The analysis also shows\na difference between the number of immigrants from abroad, which is less pronounced (bottom panel)\nand visually confirmed in the scatterplots across all municipalities (Appendix 3).\n\n\n5\n\nMention: \"Population Census\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Municipality employment rates are computed for urban and semi-urban areas with complete data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001481:7:1", "occurrence_id": "8f5bffe51c4eda7c699116a5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <br>10<br>20<br>30<br>40<br>50<br>**Employment rate**<br>High imports<br>Low imports|\n\n\n\nSource: Own elaboration based on Population Census, Integrated Public Use Microdata Series and Customs data. Notes: (1) The\nsample is restricted to the urban and semi-urban municipalities with complete data in the whole period. (2) A municipality has\nrelatively high exports (imports) if its exports (imports) per 2000 worker ratio in 2015 is above the median across municipalities\nand low exports (imports) if its ratio lies below the median.\n\n\nThird, a municipality’s trade integration could strongly matter for migration patterns, both across\ndomestic municipalities and with other countries, specifically the United States. More trade opportunities\nin a municipality may deter migration outflows, while attracting more migrants from other countries or\npulling back Mexican migrants from the United States. In fact, the number of unauthorized migrants from\nMexico to the United States declined from 6.9 million in 2007 to 5.6 million by 2015 (Gonzalez-Barrera\nand Krogstad 2019). Trade in a municipality may also have effects on remittances and the working age\npopulation. We therefore investigate the migration channel and again detect a clear difference between\nmunicipalities that are highly integrated in trade and those that are not. The average number of\nimmigrants from other Mexican states in municipalities with high exports per worker is 142, while that in\nmunicipalities with low exports per worker is less than half this magnitude, namely 63 (Figure 4, upper\nleft panel). Again, the relative difference is larger for imports (upper right panel). The analysis also shows\na difference between the number of immigrants from abroad, which is less pronounced (bottom panel)\nand visually confirmed in the scatterplots across all municipalities (Appendix 3).\n\n\n5\n\nMention: \"Integrated Public Use Microdata Series\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Harmonised census microdata supplement the national census for the municipality panel.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001481:7:1", "occurrence_id": "dbdca2232da2acc283a805a9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <br>10<br>20<br>30<br>40<br>50<br>**Employment rate**<br>High imports<br>Low imports|\n\n\n\nSource: Own elaboration based on Population Census, Integrated Public Use Microdata Series and Customs data. Notes: (1) The\nsample is restricted to the urban and semi-urban municipalities with complete data in the whole period. (2) A municipality has\nrelatively high exports (imports) if its exports (imports) per 2000 worker ratio in 2015 is above the median across municipalities\nand low exports (imports) if its ratio lies below the median.\n\n\nThird, a municipality’s trade integration could strongly matter for migration patterns, both across\ndomestic municipalities and with other countries, specifically the United States. More trade opportunities\nin a municipality may deter migration outflows, while attracting more migrants from other countries or\npulling back Mexican migrants from the United States. In fact, the number of unauthorized migrants from\nMexico to the United States declined from 6.9 million in 2007 to 5.6 million by 2015 (Gonzalez-Barrera\nand Krogstad 2019). Trade in a municipality may also have effects on remittances and the working age\npopulation. We therefore investigate the migration channel and again detect a clear difference between\nmunicipalities that are highly integrated in trade and those that are not. The average number of\nimmigrants from other Mexican states in municipalities with high exports per worker is 142, while that in\nmunicipalities with low exports per worker is less than half this magnitude, namely 63 (Figure 4, upper\nleft panel). Again, the relative difference is larger for imports (upper right panel). The analysis also shows\na difference between the number of immigrants from abroad, which is less pronounced (bottom panel)\nand visually confirmed in the scatterplots across all municipalities (Appendix 3).\n\n\n5\n\nMention: \"Customs data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Municipality export and import values are used to rank places by relative trade exposure.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001484:1:0", "occurrence_id": "6f24f06682b7748b4cb94f1d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Policy Research Working Paper 10715\n\n##### **Abstract**\n\nEnvironmental degradation is the largest public health\nchallenge of the century and is likely to be exacerbated by\nclimate change. This study undertakes a comprehensive\nexamination of the health implications of environmental\nhazards in Cambodia, simultaneously addressing extreme\ntemperatures, precipitation patterns, and air pollution. It\nleverages data from the Demographic and Health Surveys\nand satellite-derived metrics on temperature, precipitation,\nand fine particulate matter. The analysis identifies a positive association between temperature and the occurrence\nof diarrhea and cough among children and a nonlinear\nrelationship between precipitation and these health outcomes. Furthermore, the study demonstrates that pollution\nsignificantly impacts cough incidence. To anticipate future\n\n\n\ntrends, climate simulations are employed to forecast the\nincidence of child diarrhea in Cambodia under different\nclimate and development scenarios. The projections indicate that diarrhea incidence could increase to 19 percent by\n2040 without significant adaptation measures that would\nlessen the adverse impact of weather. For instance, the acceleration in toilet ownership over the last decade reduced\ndiarrhea incidence by at least 1.2 to 1.4 percentage points.\nNevertheless, the path ahead requires proactive efforts to\nimprove sanitation and hygiene. The forecasts suggest that,\nwithout additional strategies to counter climate change’s\nadverse effects, only universal toilet ownership would\ncontain the climate-driven increase in diarrhea incidence\nexpected by 2040.\n\n\n\nThis paper is a product of the Health, Nutrition and Population Global Practice. It is part of a larger effort by the World\n\nBank to provide open access to its research and make a contribution to development policy discussions around the world.\nPolicy Research Working Papers are also posted on the Web at http://www.worldbank.org/prwp. The authors may be\ncontacted at sbaquie@worldbank.org.\n\n\n_The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development_\n_issues. An objective of the series is to get the findings out quickly, even if\n\nMention: \"Demographic and Health Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Child diarrhoea and cough outcomes are linked to satellite climate and pollution measures.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001486:18:1", "occurrence_id": "b69e4cf1f2335ae9716e63b6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .65 2017 PPP 69.34 70.41\n<u>(0.011)</u> <u>(0.001)</u>\n\n_Note:_ Standard errors in parenthesis\n_Source_ : World Bank staff calculations, based on HNAP summer 2022 Demographic and WASH survey\n\nAs shown in Figure 6, the distribution by governorate and the welfare quintiles constructed on each of the\ntwo welfare metrics are substantially different, with clear implications in terms of poverty profile (Figure\n7).\n\nMention: \"HNAP summer 2022 Demographic and WASH survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Welfare metrics and their distribution by governorate and quintile are computed from this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001488:7:1", "occurrence_id": "80bcfcf66fb78e19327f820c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  to improve\neducational outcomes for children by better preparing them for primary school.\n\nMeanwhile, primary education in Uzbekistan is free and compulsory, starting at the age\nof 6-7, with an enrollment rate of 94%. <sup>5</sup> The overall goal of expanding childcare facilities,\nparticularly through PPP preschools, was to increase access to early childhood education and\nenhance children’s readiness for primary education, especially in regions\nwhere such services were previously limited.\n##### **3. Data**\n\n_Survey data._ For our main estimation analyses, we use Listening to Citizens of Uzbekistan\nsurvey (L2CU) data conducted by the World Bank and Development Strategy Center\n(2022). The study comprises a 4,000-household nationally representative baseline survey\nconducted in June/July 2018, and since then a monthly “panel” survey of a subset of 1,500\nhouseholds from the baseline survey until 2023. This survey is used to measure quar- terly\nfemale labor supply, individual and other household-level covariates since June 2018 in\n<u>each district. Furthermore, Household Budget Survey (HBS) data – yearly pooled cross-</u>\n<u>sectional data from 2015-2022 – is used to do robustness test estimations. The HBS serves</u>\n<u>as the fundamental data source for examining people’s standard of living and for devising</u>\n<u>additional measures to enhance the well-being of the population. Furthermore,</u>\n\n~~3Source:~~ ~~https://cabar.asia/ru/uzbekist~~ an-gosudarstvennyh-detskih-sadov-ne-hvataet-chastnye-dorogi\n\n\n##### **3. Data**\n\n\n\n<u>additional measures to enhance the well-\n\nMention: \"Listening to Citizens of Uzbekistan\nsurvey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"A 4,000-household baseline and a monthly sub-panel measure quarterly female labour supply.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001488:7:1", "occurrence_id": "c0813fd0b4e8d48a3ac30a0f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  to improve\neducational outcomes for children by better preparing them for primary school.\n\nMeanwhile, primary education in Uzbekistan is free and compulsory, starting at the age\nof 6-7, with an enrollment rate of 94%. <sup>5</sup> The overall goal of expanding childcare facilities,\nparticularly through PPP preschools, was to increase access to early childhood education and\nenhance children’s readiness for primary education, especially in regions\nwhere such services were previously limited.\n##### **3. Data**\n\n_Survey data._ For our main estimation analyses, we use Listening to Citizens of Uzbekistan\nsurvey (L2CU) data conducted by the World Bank and Development Strategy Center\n(2022). The study comprises a 4,000-household nationally representative baseline survey\nconducted in June/July 2018, and since then a monthly “panel” survey of a subset of 1,500\nhouseholds from the baseline survey until 2023. This survey is used to measure quar- terly\nfemale labor supply, individual and other household-level covariates since June 2018 in\n<u>each district. Furthermore, Household Budget Survey (HBS) data – yearly pooled cross-</u>\n<u>sectional data from 2015-2022 – is used to do robustness test estimations. The HBS serves</u>\n<u>as the fundamental data source for examining people’s standard of living and for devising</u>\n<u>additional measures to enhance the well-being of the population. Furthermore,</u>\n\n~~3Source:~~ ~~https://cabar.asia/ru/uzbekist~~ an-gosudarstvennyh-detskih-sadov-ne-hvataet-chastnye-dorogi\n\n\n##### **3. Data**\n\n\n\n<u>additional measures to enhance the well-\n\nMention: \"Household Budget Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Pooled yearly cross-sections supply the robustness estimates and living standard measures.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001489:5:0", "occurrence_id": "c9a29ef6ffc03532e03d4952", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: First, we analyze the pattern of US investment over time through an event study approach\n\nthat investigates whether the coefficients on distance, border, NATO, and same country\n\nvariables have changed. We find limited evidence of nearshoring or friend-shoring as the\n\ncoefficients of the dummy variables for border and NATO appear relatively stable over\n\ntime. However, the data suggest an intensification in reshoring activities among US\n\ncompanies. Specifically, the same country dummy, which captures how much companies\n\nheadquartered in the US are investing domestically, became positive and statistically\n\nsignificantly different in 2020 and further increased in magnitude in 2023.\n\n\nSecond, we implement a triple-difference approach comparing US investments to EU\n\ninvestments over time to identify differences in their patterns following the beginning of\n\nthe US-China trade war, the Covid-19 pandemic, and the Russian invasion of Ukraine. This\n\napproach allows us to control for all source country-time, destination country-time, and\n\ncountry-pair unobservable characteristics. The results confirm that US companies\n\nintensified their reshoring activities. In addition, we find that, compared to EU firms, US\n\ncompanies also increased nearshoring following the Covid-19 shock. The estimates\n\nindicate that US investment in neighboring countries and within the US states doubled\n\nduring the post Russia’s invasion of Ukraine period, with slightly larger effects for US\n\ninvestment in the US.\n\n\nFinally, we exploit the richness of the fDi Markets database to assess the impact of the\n\nshocks on direct job creation associated with greenfield investment announcements. This\n\nallows us to determine whether the increase in investment announcements is associated\n\nwith job creation in the US or if these investments are likely leading to jobless growth in\n\nthe manufacturing sector with highly automated factories. The results suggest that recent\n\ninvestments are associated with job creation in both the US and neighboring countries. For\n\nexample, following the invasion of Ukraine, the number of direct jobs in greenfield\n\ninvestment announcements were more than double in the US and its neighbors. These\n\nresults combined with the ones\n\nMention: \"fDi Markets database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Greenfield announcements are used to estimate direct job creation following the shocks.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001492:11:0", "occurrence_id": "011c574a6f1fdec0f0822417", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The VHLSSs provide a wealth of information at both the household and individual\n\n\nlevels. Household-level data includes details on assets, production, income, housing\n\n\nconditions, and participation in government programs. Meanwhile, individual-level data\n\n\nencompass demographics, education, and employment information. Additionally, the\n\n\nVHLSSs offer insights into the basic characteristics and infrastructure of the communes and\n\n\nvillages where the sampled households reside.\n\n\nIn addition to the VHLSSs, we utilize two other data sources to measure local\n\n\ncommunities’ development. The first source is nighttime light data. Nighttime light intensity\n\n\nhas been widely used as a reliable proxy for GDP (e.g., Henderson _et al._, 2011; Hu and Yao,\n\n\n2022). One advantage of using nighttime light data is its extensive temporal and spatial\n\n\ncoverage, making it less susceptible to sampling errors compared to household surveys. We\n\n\nobtain nighttime light data from the Defense Meteorological Satellite Program\n\n\n(DMSP)/Operational Linescan System (OLS) and the Visible Infrared Imaging Radiometer\n\n\nSuite (VIIRS) on the Suomi National Polar-orbiting Partnership satellite. The annual\n\n\nnighttime light intensity is measured at the district level during the period 2001- 2017. The\n\n\nsecond data source is firm data at the district level. We compute the number of firms and\n\n\ntheir density using the annual Vietnam Enterprise Censuses during 2000- 2017, which were\n\n\nconducted by Viet Nam’s GSO. This data source provides indicators of wage and nonfarm\n\n\nemployment opportunities at the district level.\n\n\n**3.2. Descriptive analysis**\nWe next present descriptive statistics on the main outcome variables of households in the\n\n\ntreatment (program) districts participating in the 30A Programs (i.e., districts with a poverty\n\n\nrate greater than 50% in 2006) and the control districts (i.e., districts without the 30A\n\n\nPrograms but with a poverty rate above 40%).\n\n10\n\nMention: \"VHLSSs\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Household assets, production, income and housing and individual characteristics are drawn from these surveys.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001492:11:0", "occurrence_id": "49a137d603d93dc8f6c01cae", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The VHLSSs provide a wealth of information at both the household and individual\n\n\nlevels. Household-level data includes details on assets, production, income, housing\n\n\nconditions, and participation in government programs. Meanwhile, individual-level data\n\n\nencompass demographics, education, and employment information. Additionally, the\n\n\nVHLSSs offer insights into the basic characteristics and infrastructure of the communes and\n\n\nvillages where the sampled households reside.\n\n\nIn addition to the VHLSSs, we utilize two other data sources to measure local\n\n\ncommunities’ development. The first source is nighttime light data. Nighttime light intensity\n\n\nhas been widely used as a reliable proxy for GDP (e.g., Henderson _et al._, 2011; Hu and Yao,\n\n\n2022). One advantage of using nighttime light data is its extensive temporal and spatial\n\n\ncoverage, making it less susceptible to sampling errors compared to household surveys. We\n\n\nobtain nighttime light data from the Defense Meteorological Satellite Program\n\n\n(DMSP)/Operational Linescan System (OLS) and the Visible Infrared Imaging Radiometer\n\n\nSuite (VIIRS) on the Suomi National Polar-orbiting Partnership satellite. The annual\n\n\nnighttime light intensity is measured at the district level during the period 2001- 2017. The\n\n\nsecond data source is firm data at the district level. We compute the number of firms and\n\n\ntheir density using the annual Vietnam Enterprise Censuses during 2000- 2017, which were\n\n\nconducted by Viet Nam’s GSO. This data source provides indicators of wage and nonfarm\n\n\nemployment opportunities at the district level.\n\n\n**3.2. Descriptive analysis**\nWe next present descriptive statistics on the main outcome variables of households in the\n\n\ntreatment (program) districts participating in the 30A Programs (i.e., districts with a poverty\n\n\nrate greater than 50% in 2006) and the control districts (i.e., districts without the 30A\n\n\nPrograms but with a poverty rate above 40%).\n\n10\n\nMention: \"nighttime light data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"District light intensity proxies local development with wide temporal and spatial coverage.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001492:11:0", "occurrence_id": "8befc181095e5faaf581c8b3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The VHLSSs provide a wealth of information at both the household and individual\n\n\nlevels. Household-level data includes details on assets, production, income, housing\n\n\nconditions, and participation in government programs. Meanwhile, individual-level data\n\n\nencompass demographics, education, and employment information. Additionally, the\n\n\nVHLSSs offer insights into the basic characteristics and infrastructure of the communes and\n\n\nvillages where the sampled households reside.\n\n\nIn addition to the VHLSSs, we utilize two other data sources to measure local\n\n\ncommunities’ development. The first source is nighttime light data. Nighttime light intensity\n\n\nhas been widely used as a reliable proxy for GDP (e.g., Henderson _et al._, 2011; Hu and Yao,\n\n\n2022). One advantage of using nighttime light data is its extensive temporal and spatial\n\n\ncoverage, making it less susceptible to sampling errors compared to household surveys. We\n\n\nobtain nighttime light data from the Defense Meteorological Satellite Program\n\n\n(DMSP)/Operational Linescan System (OLS) and the Visible Infrared Imaging Radiometer\n\n\nSuite (VIIRS) on the Suomi National Polar-orbiting Partnership satellite. The annual\n\n\nnighttime light intensity is measured at the district level during the period 2001- 2017. The\n\n\nsecond data source is firm data at the district level. We compute the number of firms and\n\n\ntheir density using the annual Vietnam Enterprise Censuses during 2000- 2017, which were\n\n\nconducted by Viet Nam’s GSO. This data source provides indicators of wage and nonfarm\n\n\nemployment opportunities at the district level.\n\n\n**3.2. Descriptive analysis**\nWe next present descriptive statistics on the main outcome variables of households in the\n\n\ntreatment (program) districts participating in the 30A Programs (i.e., districts with a poverty\n\n\nrate greater than 50% in 2006) and the control districts (i.e., districts without the 30A\n\n\nPrograms but with a poverty rate above 40%).\n\n10\n\nMention: \"firm data at the district level\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Firm counts, density, wages and non-farm employment are computed per district.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001492:11:0", "occurrence_id": "6fc463781b86eaf7369a4a8d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The VHLSSs provide a wealth of information at both the household and individual\n\n\nlevels. Household-level data includes details on assets, production, income, housing\n\n\nconditions, and participation in government programs. Meanwhile, individual-level data\n\n\nencompass demographics, education, and employment information. Additionally, the\n\n\nVHLSSs offer insights into the basic characteristics and infrastructure of the communes and\n\n\nvillages where the sampled households reside.\n\n\nIn addition to the VHLSSs, we utilize two other data sources to measure local\n\n\ncommunities’ development. The first source is nighttime light data. Nighttime light intensity\n\n\nhas been widely used as a reliable proxy for GDP (e.g., Henderson _et al._, 2011; Hu and Yao,\n\n\n2022). One advantage of using nighttime light data is its extensive temporal and spatial\n\n\ncoverage, making it less susceptible to sampling errors compared to household surveys. We\n\n\nobtain nighttime light data from the Defense Meteorological Satellite Program\n\n\n(DMSP)/Operational Linescan System (OLS) and the Visible Infrared Imaging Radiometer\n\n\nSuite (VIIRS) on the Suomi National Polar-orbiting Partnership satellite. The annual\n\n\nnighttime light intensity is measured at the district level during the period 2001- 2017. The\n\n\nsecond data source is firm data at the district level. We compute the number of firms and\n\n\ntheir density using the annual Vietnam Enterprise Censuses during 2000- 2017, which were\n\n\nconducted by Viet Nam’s GSO. This data source provides indicators of wage and nonfarm\n\n\nemployment opportunities at the district level.\n\n\n**3.2. Descriptive analysis**\nWe next present descriptive statistics on the main outcome variables of households in the\n\n\ntreatment (program) districts participating in the 30A Programs (i.e., districts with a poverty\n\n\nrate greater than 50% in 2006) and the control districts (i.e., districts without the 30A\n\n\nPrograms but with a poverty rate above 40%).\n\n10\n\nMention: \"Vietnam Enterprise Censuses\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The annual censuses conducted by the statistical office are the origin of the district firm measures.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001494:9:0", "occurrence_id": "1411f88084ba296f77248355", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: twice the growth rate in high-skill and low-skill services (around 1.5%). In upper middle-income countries,\n\n\nagriculture and manufacturing report the highest labor productivity growth of around 3%, followed by\n\n\n2% in high-skill services, and 1.2% in low-skill services. In lower middle-income and low-income countries,\n\n\nagricultural productivity growth is only 1.5%, followed by high-skill services at 1.2%. Productivity growth\n\n\nin manufacturing is only 0.7%, and a mere 0.4% in low-skill services.\n\n\nFigure 2: Labor productivity growth across sectors\n\n\nSource: GGDC 10-sector Database. Labor productivity is calculated as value-added at 2005 fixed prices\n\n\nin each sector divided by sectoral employment. See Appendix A for more details.\n\n\n**International** **production** **specialization** **and** **premature** **de-industrialization**\n\n\nTechnology-enabled globalization has impacted the dynamics of employment and output shares through\n\n\npatterns of international specialization. Advances in transportation and communication technologies in\n\n\nthe late 20th century enabled production unbundling, outsourcing, and offshoring on a massive scale,\n\n\nleading to unprecedented growth in international trade and investment. Many studies have documented\n\n\nthe effects of trade and multinational corporations’ offshoring on de-industrialization in the US (D. H. Au\n\ntor, Dorn, and Hanson 2013; Acemoglu, D. Autor, et al. 2016; Boehm, Flaaen, and Pandalai-Nayar 2020).\n\n\nSuch forces may be even stronger for smaller developing countries, which typically act as price takers in\n\n\nworld markets. Theory suggests that international trade can create patterns of structural change dis\n\ntinct from closed economy models (Matsuyama 2009; Matsuyama 2019). Globalization amplifies, rather\n\n\nthan reduces, the power of endogenous domestic demand composition differences\n\nMention: \"GGDC 10-sector Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Value added at 2005 prices and sectoral employment are combined to compute labour productivity growth.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001495:14:0", "occurrence_id": "403925540589f9599687dc7c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Fig. 4 Inventory of landslide and non-landslide locations for the CHT and Sylhet divisions.\n\n\nGoogle Earth satellite images were used to ensure the randomly generated non-landslide\n\n\npoints are not located at past landslide locations that are not captured in the landslide inventory.\n\n\nIf a point was suspected to be located over a past landslide location based on satellite images,\n\n\nthe point was removed from the non-landslide locations inventory. The red points in Fig. 4\n\n\nshow the non-landslide locations included in the inventory. The inventory includes 301\n\n\n13\n\nMention: \"Inventory of landslide and non-landslide locations\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"Randomly generated non-landslide points were checked against satellite imagery to remove misclassified sites.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001495:14:0", "occurrence_id": "e3afd28cea24bac132526cbf", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Fig. 4 Inventory of landslide and non-landslide locations for the CHT and Sylhet divisions.\n\n\nGoogle Earth satellite images were used to ensure the randomly generated non-landslide\n\n\npoints are not located at past landslide locations that are not captured in the landslide inventory.\n\n\nIf a point was suspected to be located over a past landslide location based on satellite images,\n\n\nthe point was removed from the non-landslide locations inventory. The red points in Fig. 4\n\n\nshow the non-landslide locations included in the inventory. The inventory includes 301\n\n\n13\n\nMention: \"landslide inventory\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Known past events are incomplete, so satellite imagery is used to catch unrecorded landslides.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001497:22:1", "occurrence_id": "e4063f4cb1df3d022dc59033", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  constitutes a random sample of EAs, households, and\nindividuals.\n\nThe LSMS-ISA samples are meant to be nationally representative of households and of\n\n\n\nindividuals.\n\nThe LSMS-ISA samples are meant to be nationally representative of households and of\nindividuals. Ideally, longitudinal studies preserve representativeness over time, indicating that\nthe sample should represent both the current population at each survey occasion and the\ndynamics over time of the initial population. To maintain both types of representativeness,\nlongitudinal surveys follow up with people interviewed in previous survey rounds and add new\nindividuals to ensure that new members of the population such as migrants and newborns are\nincluded (Glewwe and Jacoby 2000). To this end, panel surveys establish rules to define\ninterview targets in follow-up rounds and create specific protocols to track households and\nindividuals that move away from their original location. Moreover, they adopt refreshed\nsamples (i.e., rotating panel or split panel) and protocols for the inclusion in the survey of\nindividuals not previously involved in the study.\n\nWhen discussing tracking in panel surveys, Witoelar (2011) introduced the concept of\n\n\n\nindividuals not previously involved in the study.\n\nWhen discussing tracking in panel surveys, Witoelar (2011) introduced the concept of\ninterview targets and tracking targets. Interview targets are those that need to be interviewed;\ntracking targets are those individuals or households that should be tracked if they have moved\nand, if found, should be interviewed. Similarly, the author distinguished between following-up\nrules and tracking rules in defining the criteria for individuals or households to be interviewed\ndepending on whether they are found in the same location or are movers. To some extent, the\nLSMS-ISA panel surveys can be seen as a panel of both households and individuals. All\nhouseholds and individuals found and interviewed during the baseline are interview targets\nacross the entire history of the panel sample.\n\nThe LSMS-ISA tracking approaches are more heterogeneous and mainly diverge in terms of: (i)\n\nMention: \"LSMS-ISA panel surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"inform\", \"summary\": \"Its follow-up and tracking rules make it a panel of both households and individuals.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001498:15:3", "occurrence_id": "6a4647b7dec3f77cf8cad6a2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \n1 _/_ (1 _−θ_ ) = 0 _._ 67 (complements). We set the depreciation rate _δ_ = 0 _._ 039 (the average rate in the US\n\n\nfor the past 20 years) and assume a rate of population growth _n_ = 0 _._ 03 (UN World Population\n\nProspects). <sup>21</sup> For simplicity, we set _A_ = 1 and _µ_ = _λ_ = 0 _._ 5 (this way, only the two parameters _ρ_\n\n\nand _θ_ will govern the choice of _k_ given _s_ and _u_ ). Finally, we calibrate the savings rate _v_ to match\n\n\nthe capital-output ratio in the US (around 3), which yields _v_ = 0 _._ 21.\n\n\nTo obtain the shares of skilled workers in each country in the baseline, we multiply the share of\n\n\n21The rate of depreciation is the average for US for the period 1999-2019 in PWT 10.01.\n\n\n14\n\nMention: \"UN World Population\n\nProspects\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The population growth rate used in the calibration is taken from these projections.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001507:9:0", "occurrence_id": "c4f83c047175a88d26b39b26", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: increased from 2.0 to 2.5. Similarly, for the case of women, the ratio in manufacturing increased\nfrom 1.8 to 2.9.\n\n\n**Figure 5: Relative Syrian Prevalence, by sector and gender (2017-2019)**\n\n\n**Source:** Authors’ calculations with data from Jordan LFS and UNCOMTRADE\n\nThe limitations for Syrians entering the job market to a larger extent may stem from a skill\nmismatch, as highlighted in Visintin et al. (2015) and McGowan and Andrews (2015). Although\nworking-age Syrians are less educated than their Jordanian counterparts (see Figure 8), this\nmismatch can manifest in highly educated Syrians performing low-skilled jobs and less educated\nSyrians struggling to compete. To analyze this, we can compare the distribution of Jordanians and\nSyrians across sectors for a given educational level. The distribution of native Jordanians can serve\nas a benchmark, reflecting the structural characteristics of the labor market. Significant\ndifferences in the allocation of Syrian workers, compared to Jordanians with the same educational\nbackground, would suggest a skill migration mismatch. Figures 6 and 7 support this hypothesis:\n66% of Jordanians in manufacturing, less than 8% in FIRE (Finance, Insurance, and Real Estate),\nand 25% in EGW (Electricity, Gas, and Water) have not completed secondary education. In\ncontrast, 81% of Syrians in manufacturing, 36% in FIRE, and 91% in EGW have not completed\nsecondary education, indicating a significant deviation from the structural benchmark.\n\n\n8 | P a g e\n\nMention: \"Jordan LFS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Sectoral prevalence of Syrian workers by gender is computed from this labour force survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001507:9:0", "occurrence_id": "f269c68f1a1318276c675e56", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: increased from 2.0 to 2.5. Similarly, for the case of women, the ratio in manufacturing increased\nfrom 1.8 to 2.9.\n\n\n**Figure 5: Relative Syrian Prevalence, by sector and gender (2017-2019)**\n\n\n**Source:** Authors’ calculations with data from Jordan LFS and UNCOMTRADE\n\nThe limitations for Syrians entering the job market to a larger extent may stem from a skill\nmismatch, as highlighted in Visintin et al. (2015) and McGowan and Andrews (2015). Although\nworking-age Syrians are less educated than their Jordanian counterparts (see Figure 8), this\nmismatch can manifest in highly educated Syrians performing low-skilled jobs and less educated\nSyrians struggling to compete. To analyze this, we can compare the distribution of Jordanians and\nSyrians across sectors for a given educational level. The distribution of native Jordanians can serve\nas a benchmark, reflecting the structural characteristics of the labor market. Significant\ndifferences in the allocation of Syrian workers, compared to Jordanians with the same educational\nbackground, would suggest a skill migration mismatch. Figures 6 and 7 support this hypothesis:\n66% of Jordanians in manufacturing, less than 8% in FIRE (Finance, Insurance, and Real Estate),\nand 25% in EGW (Electricity, Gas, and Water) have not completed secondary education. In\ncontrast, 81% of Syrians in manufacturing, 36% in FIRE, and 91% in EGW have not completed\nsecondary education, indicating a significant deviation from the structural benchmark.\n\n\n8 | P a g e\n\nMention: \"UNCOMTRADE\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Trade flows accompany the labour data in the sectoral comparison.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001508:3:1", "occurrence_id": "65157f684dbd46e83b145978", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  plan in place that relies on a realistic\nunderstanding of the existing investments in pre-primary education and scenarios for additional\ninvestments in the upcoming years.\n\nDespite the gaps in existing data, there are mechanisms through which this information gap can be\nfilled in the short to medium term until such systems are developed and implemented. This paper\nis one attempt to fill this information gap. It presents a simulation exercise using an adapted version\nof the ECE Accelerator Costing and Simulation Model <sup>1</sup> to outline the ECE costs, financing gap,\nand human resources and infrastructure needed to achieve government goals and SDG 4.2 targets.\nThe aim is to facilitate evidence-based planning and decision-making for the ECE subsector in the\nupcoming years. The model also considers alternative scenarios for meeting targets by 2035. The\ndiscussion section considers system capacities and additional support mechanisms required for\nthese endeavors.\n\nThis paper draws on administrative data sources, the population census, and publicly available\ndata on government spending to consider what a core package of ECE services might cost in\nPakistan. It considers goals such as those articulated in SDG 4.2 as well as prospective goals being\nconsidered by the government. <sup>2</sup> This paper presents cost estimates for three packages: (i) a\nbusiness-as-usual package, (ii) a core service delivery package, and (iii) an augmented service\ndelivery package. Various approaches to service delivery are costed including community\nconstruction and vouchers.\n\n\n1 The development of this tool was supported by UNICEF in collaboration with others. The original tool is available\nat https://www.ece-accelerator.org/resources/early-childhood-accelerator-simulation-model.\n2 The analyses presented in this paper were motivated by a government request to understand the costs of expanding\nECE provision across Pakistan.\n\n2\n\nMention: \"administrative data sources\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Service and workforce records underpin the costing of a core early childhood package.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001508:3:1", "occurrence_id": "1887109d52bdf5bf496a0c57", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  plan in place that relies on a realistic\nunderstanding of the existing investments in pre-primary education and scenarios for additional\ninvestments in the upcoming years.\n\nDespite the gaps in existing data, there are mechanisms through which this information gap can be\nfilled in the short to medium term until such systems are developed and implemented. This paper\nis one attempt to fill this information gap. It presents a simulation exercise using an adapted version\nof the ECE Accelerator Costing and Simulation Model <sup>1</sup> to outline the ECE costs, financing gap,\nand human resources and infrastructure needed to achieve government goals and SDG 4.2 targets.\nThe aim is to facilitate evidence-based planning and decision-making for the ECE subsector in the\nupcoming years. The model also considers alternative scenarios for meeting targets by 2035. The\ndiscussion section considers system capacities and additional support mechanisms required for\nthese endeavors.\n\nThis paper draws on administrative data sources, the population census, and publicly available\ndata on government spending to consider what a core package of ECE services might cost in\nPakistan. It considers goals such as those articulated in SDG 4.2 as well as prospective goals being\nconsidered by the government. <sup>2</sup> This paper presents cost estimates for three packages: (i) a\nbusiness-as-usual package, (ii) a core service delivery package, and (iii) an augmented service\ndelivery package. Various approaches to service delivery are costed including community\nconstruction and vouchers.\n\n\n1 The development of this tool was supported by UNICEF in collaboration with others. The original tool is available\nat https://www.ece-accelerator.org/resources/early-childhood-accelerator-simulation-model.\n2 The analyses presented in this paper were motivated by a government request to understand the costs of expanding\nECE provision across Pakistan.\n\n2\n\nMention: \"population census\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Child population counts by location determine the scale of the package.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001508:3:1", "occurrence_id": "be67567fdbf538ca3409944e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  plan in place that relies on a realistic\nunderstanding of the existing investments in pre-primary education and scenarios for additional\ninvestments in the upcoming years.\n\nDespite the gaps in existing data, there are mechanisms through which this information gap can be\nfilled in the short to medium term until such systems are developed and implemented. This paper\nis one attempt to fill this information gap. It presents a simulation exercise using an adapted version\nof the ECE Accelerator Costing and Simulation Model <sup>1</sup> to outline the ECE costs, financing gap,\nand human resources and infrastructure needed to achieve government goals and SDG 4.2 targets.\nThe aim is to facilitate evidence-based planning and decision-making for the ECE subsector in the\nupcoming years. The model also considers alternative scenarios for meeting targets by 2035. The\ndiscussion section considers system capacities and additional support mechanisms required for\nthese endeavors.\n\nThis paper draws on administrative data sources, the population census, and publicly available\ndata on government spending to consider what a core package of ECE services might cost in\nPakistan. It considers goals such as those articulated in SDG 4.2 as well as prospective goals being\nconsidered by the government. <sup>2</sup> This paper presents cost estimates for three packages: (i) a\nbusiness-as-usual package, (ii) a core service delivery package, and (iii) an augmented service\ndelivery package. Various approaches to service delivery are costed including community\nconstruction and vouchers.\n\n\n1 The development of this tool was supported by UNICEF in collaboration with others. The original tool is available\nat https://www.ece-accelerator.org/resources/early-childhood-accelerator-simulation-model.\n2 The analyses presented in this paper were motivated by a government request to understand the costs of expanding\nECE provision across Pakistan.\n\n2\n\nMention: \"publicly available\ndata on government spending\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Existing spending levels provide the baseline against which costs are compared.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001510:14:0", "occurrence_id": "27f3aebfc37e2ca5a42ca6d0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: stimulating activities with any caregiver in the previous three days). A Z-score of 0 would indicate\nthat a child in this sample has a similar developmental status to that of a child in the CREDI\nreference sample who is the same age. As demonstrated by Figure 1, the average child in this\nsample has a developmental status almost 2 standard deviations (1.72) below the mean, indicating\nmuch lower levels of child development in this sample compared to the global population.\nDistributions varied by various characteristics of interest, such as maternal education, living in a\nrural area and refugee status (Figures 2-4).\n\n**Figure 1**\n\n\n\n**Figure 2**\n\n\n\n_Sample Distribution for Children with Mothers with Less than a Primary Education Compared_\n\n\n\n\n\n13\n\nMention: \"CREDI\nreference sample\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Its age-specific distributions define the zero point against which children's z-scores are read.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001513:16:1", "occurrence_id": "36ceefc156e11d049f1e68ac", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  score of 31.2 <sup>8</sup> (out of a possible score of 100).\n\n\nConsequently, there is an urgent need to provide funds specifically for health security. Thus, there\n\n\nis a dire need to make a dedicated financial investment in health security. Policy makers in Côte\n\n\nd'Ivoire could, therefore, allocate pandemic funds in national budgets and explore opportunities to\n\n\ncollaborate with philanthropic and multilateral organizations to increase domestic resource\n\n\nmatching. As a result of their collaboration, for every dollar spent by policy makers in Côte d'Ivoire\n\n\non pandemic preparedness and response, the matching donor contributes U$2 (a 2:1 matching\n\n\nratio). Policy makers could use these funds to invest more in the surveillance system, allowing\n\n\noutbreaks to be contained before they spread. Furthermore, modern, and strong laboratory\n\n\n7 <u>[https://data.who.int/dashboards/covid19/vaccines?m49=384&n=c](https://data.who.int/dashboards/covid19/vaccines?m49=384&n=c)</u>\n\n8 [The data are from the 2021 Global Health Security index (https://www.ghsindex.org/country/cote-divoire/). The data](https://www.ghsindex.org/country/cote-divoire/)\nsuggest that Côte d’Ivoire has low score in preventing (12.4), detecting (29.6), responding (35.3), health (17.1), norms\n(47.7), and risk (45.2).\n\n\n15\n\nMention: \"2021 Global Health Security index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its country score is cited as evidence of low preparedness.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001514:3:1", "occurrence_id": "2b7409bc49e95393eda14ab7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  blueprint exists, for any solution to function well, the country, political and economic context\nmatters. At this juncture, the experimentation continues. However, valuable lessons can be drawn from past\nand current experience of countries under different power market design options.\n\nThis paper undertakes a comprehensive stock-take of the evolution of the power market structure in every\neconomy in the world between 1989 and 2024. We rely on a new unique database, the “Global Power\nMarket Structures Database <sup>1</sup> ” developed by one of the co-authors of this paper. The database examines all\n230 global economies, documenting the changes in the power market structure for every year between 1989\nand 2024. The database’s objective is to present different designs pursued by countries globally in\nstructuring their power markets. To the best of our knowledge this is the first such database and this is the\nfirst paper to analyze power markets in such a comprehensive global manner.\n\nThe paper is structured as follows: Section 2 provides a global overview of each power market structure;\nwe also delve into the different design options adopted by countries globally within each structure. Section\n3 presents the global experience of countries transitioning between different market structures, highlighting\nboth enablers and challenges countries faced during these transformations. Section 4 discusses the\nexperience of countries that experimented with different designs within their respective power market\nstructure. Section 5 contains our conclusions and distills the key policy recommendations for each power\nmarket structure based on 35 years of global experience.\n\n\n1 <u>[Global Power Market Structures Database](https://datacatalog.worldbank.org/search/dataset/0065245/global-power-market-structures-database)</u> (Akcura 2024).\n\n\n2\n\nMention: \"Global Power\nMarket Structures Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Changes in power market structure across 230 economies are documented in this new database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001517:26:0", "occurrence_id": "e4cc2cd0d3b6a5a929d0d37f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **6. Global Trends for High-Priority GGUs**\n\n\nIn this section, we extend our sampling and modeling approach to global scale. Among 51,976\nGlobal Geographic Units, we have identified 1,131 GGUs that account for 50% of global methane\nemissions in the EDGAR database. After limiting our exercise to GGU sampling areas with at least\n20 Sentinel-5P observations per year, we estimate trend models (1) and (2) by fixed effects for\n775 high-priority GGUs.\n\n\n**6.1 Overall Results**\n\n\nTable 10 provides a summary that aggregates the econometric results into four categories: No\nTrend (statistically-insignificant results for both models); Mixed (statistically-insignificant results\nfor one model or oppositely-signed but significant results for two models); Decreasing (significant\nnegative results for both models); and Increasing (significant positive results for both). We find\nNo Trend for 56 GGUs because neither model attains classical significance (p=0.05). Results are\nMixed for 150 GGUs, Decreasing for 106 and Increasing for 453. In summary, a substantial\nmajority of 775 GGUs have experienced unambiguous increases in CH4 emissions during the past\nfour years, whether judged by the overall trend or change in the most recent year. The same\npicture emerges if we weight the regression results by EDGAR emissions.\n\n\n**Table 10: GGU estimation results (fixed effects)**\n\n\nEmissions\nWeighted\n\nResults Count % %\n\nNo Trend 56 7.2 6.5\nMixed 160 20.6 23.3\nDecreasing 106 13.7 13.1\nIncreasing 453 58.5 57.1\n\n\n**6.2 Sectoral Results**\n\n\nTable 11 explores the sectoral distribution of results weighted by EDGAR emissions. Soils\nand Livestock exhibit enormous disparities between Increasing and Decreasing cases (90.2 % vs.\n0.6% for the former; 63.6\n\nMention: \"EDGAR database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Emissions by global geographic unit identify the units accounting for half of global methane.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001525:28:2", "occurrence_id": "c0e38d3e0c1a6126d80e6179", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <u>Table 5.7. Robustness check: Impact of PqTR reduction on repetition rate (PqTR 80)</u>**\n\n**<u>Overall</u>** **<u>Lower Primary</u>**\n<u>Control in 2018</u> <u>0.260***</u> <u>0.272***</u>\n(0.007) (0.008)\nTime (year=2019) -0.009 -0.010\n(0.010) (0.011)\nLagged Treatment 0.012 0.016\n(0.009) (0.010)\nLagged Treatment (DiD) -0.034*** -0.040***\n<u>(0.012)</u> <u>(0.014)</u>\n<u>Control (N)</u> <u>216</u> <u>216</u>\nTreatment (N) 424 424\n<u>r2</u> <u>0.03</u> <u>0.03</u>\n_<u>Data Source: EMIS 2018 and 2019</u>_\n_Note: Robust standard errors in parentheses. * p<0.10, ** p<0.05, *** p<0.01_\n_Compares rates at end 2017/18 and end 2018/19. Treated schools are those where PqTR brought below 90 in_\n_2018._\n\n\n27\n\nMention: \"EMIS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Rates at the end of two school years are compared between treated and control schools.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001526:1:0", "occurrence_id": "99939b3215cc12f880db60eb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Policy Research Working Paper 10911\n\n##### **Abstract**\n\nThe arrival of cloud computing provides firms a new way\n\nto access digital technologies as digital services. Yet, capital\nincentive policies present in every OECD country are still\ntargeted towards investments in information technology\n(IT) capital. If cloud services are partial substitutes for\nIT investments, the presence of capital incentive policies\nmay unintentionally discourage the adoption of cloud and\ntechnologies that rely on the cloud, such as artificial intelligence (AI) and big data analytics. This paper exploits a tax\n\n\n\nincentive in the UK for capital investment as a quasi-natural\nexperiment to examine the impact on firm adoption of\ncloud computing, big data analytics and AI. The empirical results find that the policy increased investment in IT\ncapital as would be expected; but it slowed firm adoption\nof cloud, big data and AI. Matched employer-employee\ndata shows that the policy also led firms to reduce their\ndemand for workers that perform data analytics, but not\nother types of workers.\n\n\n\nThis paper is a product of the Office of the Chief Economist, East Asia and the Pacific Region. It is part of a larger effort by\n\nthe World Bank to provide open access to its research and make a contribution to development policy discussions around\nthe world. Policy Research Working Papers are also posted on the Web at http://www.worldbank.org/prwp. The authors\nmay be contacted at timothy.destefano@georgetown.edu, nick.johnstone@iea.org, richard.kneller@nottingham.ac.uk, and\njtimmis@worldbank.org.\n\n\n_The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development_\n_issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the_\n_names of the authors and should\n\nMention: \"Matched employer-employee\ndata\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Linked firm-worker records show demand fell for data analytics workers but not others.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001528:19:0", "occurrence_id": "a9e68744a112014178dcc1c7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: plays a role. We created institutional tiles based on the average government effectiveness index\n\n\nby country from the World Governance Indicator database. Government effectiveness measures\n\n\nthe quality of public services, the civil service, policy formulation and implementation, and the\n\n\ncredibility of a government’s commitment to improving or maintaining these aspects (Kauf\n\nmann, Kraay and Mastruzzi, 2010). Our general intuition is that developing countries with low\n\n\ngovernment effectiveness are not only less able to engage in BLAs but also less able to enforce\n\n\nthe BLAs they do sign with partner countries. This likely translates into less effective BLAs\n\n\nand, consequently, lower migration flows.\n\n\nTo test this hypothesis, we created two, three, and four tiles corresponding to halves, terciles,\n\n\nand quartiles. Countries in the top tiles are characterized by more effective government. Results\n\n\nreported in table 5 strongly suggest that, although the migration response to BLAs in Africa is\n\n\nnull, African countries belonging to the upper tiers of the government indicators do experience\n\n\na positive migration response to BLAs. Conversely, African countries with poor governance\n\n\nindicators show no significant migration response to BLAs.\n\n##### **4.3 Persistence of the impacts**\n\n\nWe now turn our attention to the timing of the effect of BLAs on migration flows using an\n\n\nevent study design. It is worth noting that one of the key advantages of the event study design\n\n\nis the possibility to check if the estimated effect is long lasting or if it quickly vanishes after a\n\n\ncertain period of time. Figure 1 depicts the benchmark estimate timing of the effect of BLAs on\n\n\nmigration flows. First, there is no pre-trend prior to the treatment. Second, it appears that the\n\n\npositive effect of BLAs on migration flows lasts up to 3 decades after BLAs are signed before\n\n\nvanishing.\n\n\nConsistently, heterogeneity results show that this positive effect is driven by non-African\n\n\norigins (panel 2a of figure 2).\n\nMention: \"average government effectiveness index\n\n\nby country\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Country averages are converted into institutional tiles for the comparison.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001528:19:0", "occurrence_id": "ca1573b3bada62692f0dc6b4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: plays a role. We created institutional tiles based on the average government effectiveness index\n\n\nby country from the World Governance Indicator database. Government effectiveness measures\n\n\nthe quality of public services, the civil service, policy formulation and implementation, and the\n\n\ncredibility of a government’s commitment to improving or maintaining these aspects (Kauf\n\nmann, Kraay and Mastruzzi, 2010). Our general intuition is that developing countries with low\n\n\ngovernment effectiveness are not only less able to engage in BLAs but also less able to enforce\n\n\nthe BLAs they do sign with partner countries. This likely translates into less effective BLAs\n\n\nand, consequently, lower migration flows.\n\n\nTo test this hypothesis, we created two, three, and four tiles corresponding to halves, terciles,\n\n\nand quartiles. Countries in the top tiles are characterized by more effective government. Results\n\n\nreported in table 5 strongly suggest that, although the migration response to BLAs in Africa is\n\n\nnull, African countries belonging to the upper tiers of the government indicators do experience\n\n\na positive migration response to BLAs. Conversely, African countries with poor governance\n\n\nindicators show no significant migration response to BLAs.\n\n##### **4.3 Persistence of the impacts**\n\n\nWe now turn our attention to the timing of the effect of BLAs on migration flows using an\n\n\nevent study design. It is worth noting that one of the key advantages of the event study design\n\n\nis the possibility to check if the estimated effect is long lasting or if it quickly vanishes after a\n\n\ncertain period of time. Figure 1 depicts the benchmark estimate timing of the effect of BLAs on\n\n\nmigration flows. First, there is no pre-trend prior to the treatment. Second, it appears that the\n\n\npositive effect of BLAs on migration flows lasts up to 3 decades after BLAs are signed before\n\n\nvanishing.\n\n\nConsistently, heterogeneity results show that this positive effect is driven by non-African\n\n\norigins (panel 2a of figure 2).\n\nMention: \"World Governance Indicator database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It supplies the effectiveness scores measuring public service quality and policy credibility.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001529:15:0", "occurrence_id": "dff59abba15cf69ca2a9c64f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _14_\n\n\neffects, while those across different categories or bins tend to experience distinctly\ndifferent impacts as shown by the strong negative correlation. In the next section,\nwe estimate formal regression models to quantify the negative correlation between\nWBT and firm productivity.\n\n\nFigure 5. : Correlation between Sales per Worker and WBT Deviation\n\n\nSource: Original figure for this paper, based on data from the World Bank Enterprise Surveys\n\nand the Copernicus Climate Change Service.\nNote: WBT deviations are the difference between WBT in month _m_ and year _t_ and the\nlong-term average within a radius of 30 _km_ around the establishment. PPP = purchasing\n\npower parity; WBT = wet-bulb temperature.\n\nMention: \"World Bank Enterprise Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Sales per worker are correlated with local wet-bulb temperature deviations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001529:15:0", "occurrence_id": "764bbcacc1c211b95500487f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _14_\n\n\neffects, while those across different categories or bins tend to experience distinctly\ndifferent impacts as shown by the strong negative correlation. In the next section,\nwe estimate formal regression models to quantify the negative correlation between\nWBT and firm productivity.\n\n\nFigure 5. : Correlation between Sales per Worker and WBT Deviation\n\n\nSource: Original figure for this paper, based on data from the World Bank Enterprise Surveys\n\nand the Copernicus Climate Change Service.\nNote: WBT deviations are the difference between WBT in month _m_ and year _t_ and the\nlong-term average within a radius of 30 _km_ around the establishment. PPP = purchasing\n\npower parity; WBT = wet-bulb temperature.\n\nMention: \"Copernicus Climate Change Service\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Wet-bulb temperature within 30 km of each establishment is taken from this service.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001519:32:0", "occurrence_id": "c9b3ec4a059057550f037cfd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 3. Output since the COVID-19 pandemic in small states and other EMDEs**\n\n_The pandemic disproportionately impacted small states, with an average economic contraction of_\n_about 11 percent in 2020, much greater than in other EMDEs. Tourism-dependent small states_\n_were hit particularly hard. Recoveries have also been slower in small states: their output in 2023_\n_was only 5 percent higher than in 2019, far below the 15 percent increase in other EMDEs._\n_Moreover, output in small states is set to continue on a lower path than expected before the_\n_pandemic._\n\n**A. GDP growth** **B. GDP growth in small states**\n\n\n\nPerc <mark>en</mark> t\nSmall states EMDEs excl. small states\n\n10\n\n\n5\n\n\n0\n\n\n-5\n\n\n-10\n\n\n-15\n\n\n**C. GDP change since 2019**\n\n\n\nPercent\n\n\n10\n\n5\n\n0\n\n-5\n\n-10\n\n-15\n\n\n\n**D. GDP compared to pre-pandemic trends in**\n**small states**\n\n\n\nSmall states: Tourism-reliant\nSmall states: Commodity exporters\n\n\n\nPercent\n\n35 Small states EMDEs excl. small states\n\n\n\n125\n\n\n\n35\n30\n25\n20\n15\n10\n5\n0\n\n-5\n\n\n\n115\n\n\n105\n\n\n95\n\n\n85\n\n\n\nIndex, 100 = 2019\n\n\n\n\n\n-10\n-15\n\n\n\n_Source_ : World Bank.\n_Note_ : f = forecast; EMDEs = emerging market and developing economies. Guyana is excluded from the\nsmall states sample.\nA.-C. Country groups are GDP weighted at average 2010-19 prices and market exchange rates. Data for\n2024-26 are World Bank forecasts. Based on a sample of up to 34 small states and 119 other EMDEs.\nB. Small states commodity exporters include 13 economies and tourism-reliant small states include 22\neconomies.\nD. Blue line indicates GDP outcomes and current World Bank forecasts. Red line indicates the\n\nMention: \"Data for\n2024-26\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Forward years are not observations but forecasts produced by the institution.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001531:36:1", "occurrence_id": "9af671eedce6bf616e57d3ee", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ] Interest rate 0.00 0.73 0.01 575\n(0.01)\n\n[13] Kindness -0.01 0.77 0.38 575\n(0.04)\n\n[14] Comfortable installments 0.00 1.00 0.04 575\n(0.02)\n\n\nNotes: Data based on phone survey conducted by the lender. From December 2015 to April 2017 the lender called a random 5% sample of clients\nin the study at that time per month (stratified additionally by loan officer and credit type, with one client minimum per credit officer, month and\ncredit type). Questions about reasons for pass use were open-ended with both pre-coding of answers by enumerators and free text detail\nexplanation. The knowledge and pass use questions from Panel A were only asked to clients with a flexible loan. A total of 285 flexible loan clients\nwere surveyed, for a total of 345 survey responses (clients could be selected in more than one month's sample). Mean pass use for the December\n2015 to April 2017 period was 0.187 for all flexible loan clients according to lender administrative data. This is similar to the self-reported pass use\nmean reported in row 2. Out of the 345 survey responses, 320 (93%) had pass use recall that was congruent with the lender administrative data.\nAn additional 3% of the 345 survey responses had discrepancies between self-reported pass use and pass use from administrative data that were\nlikely due to minor lags in the reporting of pass use in the administrative records. In these instances, clients claimed to have used a pass already\nand the administrative records indicated they had not. The following month the adminsitrative records indicated the clients had indeed used a\npass, which is an indication that these discrepancies were due to minor lags in recording pass use. Panel B: Columns 1 and 2 show results for\nregressions with Flexible Contracts (pooled Std-Flex and Flex\n\nMention: \"phone survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"A monthly random five percent sample of clients was called to check self-reported behaviour.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001531:36:1", "occurrence_id": "60ea7ac522325b193f808ebe", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ] Interest rate 0.00 0.73 0.01 575\n(0.01)\n\n[13] Kindness -0.01 0.77 0.38 575\n(0.04)\n\n[14] Comfortable installments 0.00 1.00 0.04 575\n(0.02)\n\n\nNotes: Data based on phone survey conducted by the lender. From December 2015 to April 2017 the lender called a random 5% sample of clients\nin the study at that time per month (stratified additionally by loan officer and credit type, with one client minimum per credit officer, month and\ncredit type). Questions about reasons for pass use were open-ended with both pre-coding of answers by enumerators and free text detail\nexplanation. The knowledge and pass use questions from Panel A were only asked to clients with a flexible loan. A total of 285 flexible loan clients\nwere surveyed, for a total of 345 survey responses (clients could be selected in more than one month's sample). Mean pass use for the December\n2015 to April 2017 period was 0.187 for all flexible loan clients according to lender administrative data. This is similar to the self-reported pass use\nmean reported in row 2. Out of the 345 survey responses, 320 (93%) had pass use recall that was congruent with the lender administrative data.\nAn additional 3% of the 345 survey responses had discrepancies between self-reported pass use and pass use from administrative data that were\nlikely due to minor lags in the reporting of pass use in the administrative records. In these instances, clients claimed to have used a pass already\nand the administrative records indicated they had not. The following month the adminsitrative records indicated the clients had indeed used a\npass, which is an indication that these discrepancies were due to minor lags in recording pass use. Panel B: Columns 1 and 2 show results for\nregressions with Flexible Contracts (pooled Std-Flex and Flex\n\nMention: \"lender administrative data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its recorded pass use of 0.187 is compared with clients' recall in the survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001538:33:0", "occurrence_id": "83fa534d413cbc7ef1b31185", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Sources_ : Arroyo Marioli and Vegh (2023); International Monetary Fund; PRS Group (database).\n_Note_ : Bars show average correlation between the (Hodrick Prescott-filtered) cyclical components of real GDP and\nreal government spending within groups. Vertical lines show 25th and 75th percentiles.\nA. Countries with high (low) political risk are defined as those with political risk above (below) the sample\nmedian. Sample has 49 countries with high political risk and 24 countries with low political risk. *** indicates\nthat the difference between the means for the two country groups is statistically significant at the 10 percent\nlevel or better.\nB. Sample has 48 countries with low bureaucracy quality and 25 with high bureaucracy quality. The difference\nbetween the means for the two country groups is not statistically significant.\nC. Sample has 45 countries with low control of corruption and 28 with high control of corruption. *** indicates\nthat the difference between the means for the two country groups is statistically significant at the 10 percent\nlevel or better.\nD. Sample has 32 countries with high government stability and four with low government stability. The\ndifference between the means for the two country groups is not statistically significant.\nE. Sample has 24 countries with high scores (above sample median) for the law-and-order index and 49\ncountries with low (below sample median) scores. *** indicates that the difference between the means for the\ntwo country groups is statistically significant at the 10 percent level or better.\nF. For commodity exporters, the sample includes 28 countries with fiscal rules and 20 countries without fiscal\nrules. The entire sample includes 58 countries with fiscal rules and 47 countries without fiscal rules. The\ndifference between the means for the two country groups (with and without fiscal rules) is statistically\nsignificant (indicated by ***) for the entire sample but not for the sample of commodity exporters.\n\n\n32\n\nMention: \"PRS Group (database)\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its country data are used alongside the fund's to compute cyclical correlations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001541:7:0", "occurrence_id": "5d2a6dadc73cc7cc0fa64c31", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 2: Mapping different measures of the COVID shock\n\n\n_Notes:_ The figure on the left shows the district-wise variation in the COVID-19 death rate (per 100,000\npopulation). The figure on the right shows the district-wise variation in the “leave-one-out” COVID-19\nexposure of an average respondent. The numbers for Bhiwani and Charkhi Dadri refer to the two districts\njointly.\n\n\nthe data on deaths from the Socioeconomic High-resolution Rural-Urban Geographic Plat\nform for India (SHRUG) and adjusted it for district population using data from the 2011\n\nCensus of India. As Figure 2 shows, our “leave-one-out” COVID-19 measure has a similar ge\nographic variation as, and is significantly correlated with, the district-level COVID-19 death\n\nrate, boosting confidence in our shock measure. The p-value of the correlation between our\n\nself-reported shock and district-level COVID-19 death rate is _<_ 0.001. Later, we validate our\n\nresults by conducting robustness checks using the district-level COVID-19 shock measure\n\ninstead of the individual-level “leave-one-out” exposure variable.\n\n\nWe estimate the effect of exposure to COVID-19 on our outcomes of interest using the\n\nfollowing regression specification:\n\n\n_′_\n\n_Yitd_ = _α_ + _βCOV IDi_ + **X** _i_ <sup>_γ_</sup> <sup>+</sup> <sup>_δd_</sup> <sup>+</sup> <sup>_ϵitd_</sup> (1)\n\n\nwhere _Yitd_ is the outcome variable for woman _i_ studying at ITI _t_ located in\n\nMention: \"Socioeconomic High-resolution Rural-Urban Geographic Plat\nform for India\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"District death counts are taken from this platform to build the exposure measure.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001541:7:0", "occurrence_id": "a04e006325a4f4092a564dbf", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 2: Mapping different measures of the COVID shock\n\n\n_Notes:_ The figure on the left shows the district-wise variation in the COVID-19 death rate (per 100,000\npopulation). The figure on the right shows the district-wise variation in the “leave-one-out” COVID-19\nexposure of an average respondent. The numbers for Bhiwani and Charkhi Dadri refer to the two districts\njointly.\n\n\nthe data on deaths from the Socioeconomic High-resolution Rural-Urban Geographic Plat\nform for India (SHRUG) and adjusted it for district population using data from the 2011\n\nCensus of India. As Figure 2 shows, our “leave-one-out” COVID-19 measure has a similar ge\nographic variation as, and is significantly correlated with, the district-level COVID-19 death\n\nrate, boosting confidence in our shock measure. The p-value of the correlation between our\n\nself-reported shock and district-level COVID-19 death rate is _<_ 0.001. Later, we validate our\n\nresults by conducting robustness checks using the district-level COVID-19 shock measure\n\ninstead of the individual-level “leave-one-out” exposure variable.\n\n\nWe estimate the effect of exposure to COVID-19 on our outcomes of interest using the\n\nfollowing regression specification:\n\n\n_′_\n\n_Yitd_ = _α_ + _βCOV IDi_ + **X** _i_ <sup>_γ_</sup> <sup>+</sup> <sup>_δd_</sup> <sup>+</sup> <sup>_ϵitd_</sup> (1)\n\n\nwhere _Yitd_ is the outcome variable for woman _i_ studying at ITI _t_ located in\n\nMention: \"2011\n\nCensus of India\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"District populations from this enumeration convert death counts into rates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001543:8:0", "occurrence_id": "940093ee0de8e30bfcd3bb05", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **3.** **Growth drivers in the business-as-usual baseline**\n\n\nIn this section we calibrate the LTGM-NR model to simulate Peru’s business-as-usual baseline over the\n2023-50 period. To create a baseline for the Peruvian economy in the long run, we calibrate the future\npaths of the growth drivers such as investment, human capital, the growth of TFP in the non-resource and\nnatural resource sectors, <sup>6</sup> as well as other key parameters. These assumptions are summarized in Table\n1. We calibrate the LTGM-NR as the foundation of the analysis and add the LTGM-TFP extension and the\nLTGM-HC extension to simulate non-resource sector TFP growth and human capital growth, respectively.\nWe use population growth projections (by age cohort) from the UN for demographic trends until 2050.\n\n\n6 The resource sector is composed of the copper mining sector, which is around 8 percent of the total GDP and is\nexpected to grow in the short run by around 2 percent due to the entrance of the Quellaveco mine. In the long\nrun, the depletion in reserves will shrink the sector to 6 percent of GDP and grow below 1 percent on average.\n\n\n7\n\nMention: \"population growth projections\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Projections by age cohort drive demographic trends in the long-term simulation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001544:6:0", "occurrence_id": "4ae83206fd308e062502cb6a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **2.1** **WIPO** **Green** **Patents**\n\n\nWe use information from the WIPO green patents data set for the period 2000-2021. This data\n\n\nset contains detailed information on green technology patents, inventions, and technologies from\n\n\nthe WIPO Patentscope database, covering seven sectors: Energy, Water, Farming and Forestry,\n\n\nPollution Waste, Transportation, Construction, and Products, Materials and Processes (PMP).\n\n\nGreen technologies are defined by WIPO as “climate friendly technologies that protect the\n\n\nenvironment, are less polluting, use resources in more sustainable manner, recycle their wastes\n\n\nand products and handle residual waste in a more sustainable manner”. These technologies\n\n\ninclude know-how procedures, goods and services, and equipment as well as organizational and\n\n\nmanagerial procedures. WIPO systematically gathers patent data from 193 national Intellectual\n\n\nProperty (IP) offices worldwide. The database provides information about the year of filling of\n\n\nthe patent, the country where the IP office and the inventor are located, and technical description\n\n\nand status of the technology (e.g. in development, commercialization, etc.).\n\n\n**2.2** **Earnings** **calls** **transcripts**\n\n\nWe use data from Refinitiv Eikon (formerly Thompson Reuters) on the transcripts from quarterly\n\n\nshareholder earnings calls meetings of publicly listed firms from 2012 to 2021, covering 10,554\n\n\nfirms in 82 countries. Earnings calls are key corporate events on the investor relations agenda in\n\n\nwhich senior management responds directly to questions from financial analysts and other market\n\n\nparticipants about the firm’s financial performance over the past quarter and, more broadly,\n\n\ndiscuss current developments (Hollander, Pronk, and Roelofsen, 2010; Hassan, Hollander, van\n\n\nLent, and Tahoun, 2020). They consist of a management presentation and, importantly, a Q&A\n\n\nsession which requires management to comment on subjects they might not otherwise\n\nMention: \"WIPO green patents data set\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Green technology patents across seven sectors are drawn from this dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001544:6:0", "occurrence_id": "d248f9a61e11d1080fd52c18", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **2.1** **WIPO** **Green** **Patents**\n\n\nWe use information from the WIPO green patents data set for the period 2000-2021. This data\n\n\nset contains detailed information on green technology patents, inventions, and technologies from\n\n\nthe WIPO Patentscope database, covering seven sectors: Energy, Water, Farming and Forestry,\n\n\nPollution Waste, Transportation, Construction, and Products, Materials and Processes (PMP).\n\n\nGreen technologies are defined by WIPO as “climate friendly technologies that protect the\n\n\nenvironment, are less polluting, use resources in more sustainable manner, recycle their wastes\n\n\nand products and handle residual waste in a more sustainable manner”. These technologies\n\n\ninclude know-how procedures, goods and services, and equipment as well as organizational and\n\n\nmanagerial procedures. WIPO systematically gathers patent data from 193 national Intellectual\n\n\nProperty (IP) offices worldwide. The database provides information about the year of filling of\n\n\nthe patent, the country where the IP office and the inventor are located, and technical description\n\n\nand status of the technology (e.g. in development, commercialization, etc.).\n\n\n**2.2** **Earnings** **calls** **transcripts**\n\n\nWe use data from Refinitiv Eikon (formerly Thompson Reuters) on the transcripts from quarterly\n\n\nshareholder earnings calls meetings of publicly listed firms from 2012 to 2021, covering 10,554\n\n\nfirms in 82 countries. Earnings calls are key corporate events on the investor relations agenda in\n\n\nwhich senior management responds directly to questions from financial analysts and other market\n\n\nparticipants about the firm’s financial performance over the past quarter and, more broadly,\n\n\ndiscuss current developments (Hollander, Pronk, and Roelofsen, 2010; Hassan, Hollander, van\n\n\nLent, and Tahoun, 2020). They consist of a management presentation and, importantly, a Q&A\n\n\nsession which requires management to comment on subjects they might not otherwise\n\nMention: \"WIPO Patentscope database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It is the underlying patent register from which the green subset is compiled.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001544:6:0", "occurrence_id": "71194041c045ae08c8a97752", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **2.1** **WIPO** **Green** **Patents**\n\n\nWe use information from the WIPO green patents data set for the period 2000-2021. This data\n\n\nset contains detailed information on green technology patents, inventions, and technologies from\n\n\nthe WIPO Patentscope database, covering seven sectors: Energy, Water, Farming and Forestry,\n\n\nPollution Waste, Transportation, Construction, and Products, Materials and Processes (PMP).\n\n\nGreen technologies are defined by WIPO as “climate friendly technologies that protect the\n\n\nenvironment, are less polluting, use resources in more sustainable manner, recycle their wastes\n\n\nand products and handle residual waste in a more sustainable manner”. These technologies\n\n\ninclude know-how procedures, goods and services, and equipment as well as organizational and\n\n\nmanagerial procedures. WIPO systematically gathers patent data from 193 national Intellectual\n\n\nProperty (IP) offices worldwide. The database provides information about the year of filling of\n\n\nthe patent, the country where the IP office and the inventor are located, and technical description\n\n\nand status of the technology (e.g. in development, commercialization, etc.).\n\n\n**2.2** **Earnings** **calls** **transcripts**\n\n\nWe use data from Refinitiv Eikon (formerly Thompson Reuters) on the transcripts from quarterly\n\n\nshareholder earnings calls meetings of publicly listed firms from 2012 to 2021, covering 10,554\n\n\nfirms in 82 countries. Earnings calls are key corporate events on the investor relations agenda in\n\n\nwhich senior management responds directly to questions from financial analysts and other market\n\n\nparticipants about the firm’s financial performance over the past quarter and, more broadly,\n\n\ndiscuss current developments (Hollander, Pronk, and Roelofsen, 2010; Hassan, Hollander, van\n\n\nLent, and Tahoun, 2020). They consist of a management presentation and, importantly, a Q&A\n\n\nsession which requires management to comment on subjects they might not otherwise\n\nMention: \"data from Refinitiv Eikon\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Transcripts of quarterly earnings calls of listed firms are analysed from this provider.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001545:25:1", "occurrence_id": "f88a9735c51f7aad75038796", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ) (5.47) (10.8) (11.5)\nCRI mean .098** .051**\n(6.70) (11.1)\nGPP dummy * -.258** -.655**\n<u>CRI>0.5</u> <u>(-2.86)</u> <u>(-5.42)</u>\n<u>4-digit sector FE</u> <u>Yes</u> <u>Yes</u> <u>Yes</u> <u>Yes</u> <u>Yes</u> <u>Yes</u>\nYear FE Yes Yes Yes Yes Yes Yes\nSize, Age, SOE Yes Yes Yes Yes Yes Yes\nR2 0.740 0.739 0.740 0.370 0.369 0.370\nAdjustd R2 0.740 0.739 0.739 0.369 0.368 0.368\n<u>Obs total</u> <u>780,653</u> <u>780,653</u> <u>780,653</u> <u>1,439,689</u> <u>1,439,689</u> <u>1,439,689</u>\n\n_Source_ : Firm census panel data from 2010-2018 matched with public procurement transaction data. _Note_ :\n\n\nSecond, we restrict the sample only to firms that obtained (any type of) public procurement\ncontracts to estimate if GPP contracts are more often allocated to more productive firms\nthan non-GPP contracts. Again, we test if that relation depends on the corruption risk\ninvolved in awarded GPP and other contracts.\n\n\nTable 6 shows that firms awarded GPP contracts tend to be more productive than same\n\nMention: \"Firm census panel data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Over 1.4 million firm-year observations are matched to procurement contracts.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001545:25:1", "occurrence_id": "fbc408f4d1912fa0edeae31e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ) (5.47) (10.8) (11.5)\nCRI mean .098** .051**\n(6.70) (11.1)\nGPP dummy * -.258** -.655**\n<u>CRI>0.5</u> <u>(-2.86)</u> <u>(-5.42)</u>\n<u>4-digit sector FE</u> <u>Yes</u> <u>Yes</u> <u>Yes</u> <u>Yes</u> <u>Yes</u> <u>Yes</u>\nYear FE Yes Yes Yes Yes Yes Yes\nSize, Age, SOE Yes Yes Yes Yes Yes Yes\nR2 0.740 0.739 0.740 0.370 0.369 0.370\nAdjustd R2 0.740 0.739 0.739 0.369 0.368 0.368\n<u>Obs total</u> <u>780,653</u> <u>780,653</u> <u>780,653</u> <u>1,439,689</u> <u>1,439,689</u> <u>1,439,689</u>\n\n_Source_ : Firm census panel data from 2010-2018 matched with public procurement transaction data. _Note_ :\n\n\nSecond, we restrict the sample only to firms that obtained (any type of) public procurement\ncontracts to estimate if GPP contracts are more often allocated to more productive firms\nthan non-GPP contracts. Again, we test if that relation depends on the corruption risk\ninvolved in awarded GPP and other contracts.\n\n\nTable 6 shows that firms awarded GPP contracts tend to be more productive than same\n\nMention: \"public procurement transaction data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Contract awards identify which firms received green public procurement.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001547:6:0", "occurrence_id": "765368e71c47557deede6dad", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: of the population of Myanmar (around 10 million people) had migrated, either internally or\nabroad (MAPSA 2024).\n\n\nAs economic and security prospects remain bleak, there may be sustained pressure on the\npopulation to migrate outside Myanmar. This may be particularly concerning in the case of highskilled workers that play a critical role in key professions and occupations in Myanmar, especially\nbecause of the low share of the population with higher education, relative to regional and\nneighboring peers (figure 1). High levels of emigration among the already limited numbers of\nhigh-skilled workers, such as medical professionals, educators, and engineers, can result in\nshortages in critical sectors of activity, which in turn can be detrimental to long-term\ndevelopment.\n\n\nFigure 1. Educational attainment, Bachelor’s or equivalent / at least Master's or equivalent,\npopulation 25+, total (%) – Myanmar and selected regional and neighboring comparator countries\n\n\n\n\n\n\n\n\n\nSource: Authors’ calculations using World Development Indicators, for latest available year since 2010. *Data for population 25+\nwith bachelor’s degree or equivalent not available for Cambodia.\n\n\nAgainst this backdrop, we attempt to answer the following research questions:\n\n\n(i) How much do high-skilled youth value the prospects of migrating abroad for undertaking\nwork that is similar to their current jobs?\n\n\n(ii) How would the emigration of high-skilled youth respond to prospects of political or\neconomic stabilization? What form of stabilization – political, or economic – matters (more) for\nthe emigration considerations of high-skilled youth?\n\n\n(iii) Which individual characteristics determine migration responses to signals of political or\neconomic stabilization?\n\n\n5\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Educational attainment shares for the region are compiled from these indicators.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001548:3:0", "occurrence_id": "6aa2c7b95af0a4877772d064", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: #### **1 Introduction**\n\nClimate change will have significant and heterogeneous impacts on water availability throughout\n\n\nthe world. In higher latitudes, historical data and climate models indicate that rainfall is likely\n\n\nto increase, while in lower latitudes overall precipitation is projected to decline. Additional\n\n\nwarming caused by climate change will also increase overall rainfall variability, inducing longer\n\n\nperiods of drought and increasing water stress as the agricultural, energy, and residential sectors\n\n\nextract more water from ground and surface reserves in order to combat increasing exposure to\n\n\nextreme heat and dryness (IPCC (2021)).\n\n\nAlthough there is a growing literature on the impacts of water variability on aggregate\n\n\neconomic outcomes such as GDP (Damania et al. (2017); Russ (2020)), there has been limited\n\n\nwork documenting the causal impact of water variability on labor market outcomes. This\n\n\npaper aims to quantitatively analyze the impacts of water shocks on individual employment\n\n\noutcomes in Sub-Saharan Africa. Farming is the livelihood of two-thirds of the population in the\n\n\nregion. With a high dependence on rainfed agriculture and limited resources and investment for\n\n\nclimate adaptation, this region is particularly susceptible to climate change. Notwithstanding,\n\n\nSub-Saharan Africa is growing rapidly, with its population projected to double by 2050. This\n\n\ndemographic trend puts enormous pressure on the labor market to create enough jobs to absorb\n\n\nthe growing workforce. All of these factors underscore the importance of understanding and\n\n\nmitigating the risks brought about by climate-related water shocks on jobs in Sub-Saharan\n\n\nAfrica.\n\n\nOur analysis relies on extensive, georeferenced individual socio-demographic and employ\n\nment information collected by the Demographic and Health Surveys (DHS) Program. The final\n\n\ndataset includes over half a million individuals from 16 countries across Sub-Saharan Africa\n\n\nover the period from 2005 to 2018. By using a series of pooled cross-sections of DHS surveys,\n\n\nthis paper analyzes the respondent’s employment status at the time of the survey and exploit a\n\n\nrich econometric specification that controls for a\n\nMention: \"Demographic and Health Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Georeferenced socio-demographic and employment information for over half a million people in 16 countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001553:31:0", "occurrence_id": "4a1f64f713881b6e86f29a7c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ##### **B Data**\n\nTable 1: Sources and definition of the main variables\n\n\nVariable Source Definition\n\n\nFixed_broadband_USD_log ITU’s ICT Price Basket Price of the fixed broadband basket is the\n\n\nleast expensive offer with a minimum monthly\n\n\nusage of 5 Gigabytes (GB) and an advertised\n\n\ndownload speed above 256 Kilobits per sec\n\nond (Kbps). Price is expressed in USD PPP\n\n\nand used in logarithm form.\n\n\nData_only_mobile_bdb_USD_log ITU’s ICT Price Basket Price of data-only mobile broadband basket\n\n\nis the least expensive offer with a minimum\n\n\nmonthly usage of 2 GB based on 3G or newer\n\n\ntechnologies. Price is expressed in USD PPP\n\n\nand used in logarithm form.\n\n\nn_cables /n_cables_log Telegeography Number of Submarine cables laid in a given\n\n\ncountry.\n\n\nBandwidth / Bandwidth_log Telegeography International used Internet bandwidth corre\n\nsponding to the inventory of capacity used\n\n\nby four categories of users: Internet back\n\nbone providers, content providers, research\n\neducational networks, and enterprises and\n\n\nothers. The capacity is measured in Gbps.\n\n\nRegulationScore ITU ICT regulatory tracker is an index pinpoint\n\ning the changes taking place in the ICT reg\n\nulatory environment based on 4 sub-indexes\n\n\n(regulatory authority, regulatory mandate,\n\n\nregulatory regime, and competition frame\n\nwork) and 50 indicators.\n\n\nHHI_fixedBdb / HHI_fixedBdb_log Telegeography Herfindahl-Hirschman Index (HHI) of the top\n\n\n5 retail fixed-broadband operators.\n\n\nHHI_mobileBdb / HHI_mobileBdb_log GSMA Intelligence Herfindahl-Hirschman Index (HHI) in the re\n\ntail mobile broadband markets.\n\n\nGDP_log WDI GDP in 2015 constant USD\n\n\nGDPpC_log WDI GDP per capita in 2015 constant USD\n\n\nPopulation_log WDI Total Population size, proxi\n\nMention: \"ITU ICT regulatory tracker\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Four sub-indexes on regulatory authority and management are combined into a regulation score.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001555:8:0", "occurrence_id": "99a450bbbc2a73cfd0c733ff", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: weighted to adjust for non-response and other sampling errors. See **Table 0** for the response rates\n\n\nby treatment.\n\n\nThe survey was implemented using the World Bank’s _Survey Solutions_ software for\n\n\nComputer Assisted Personal Interviews (CAPI). This allowed for daily upload of completed\n\n\ninterviews by the field teams, as well as frequent data download and review by the NBS and OCGS\n\n\nheadquarters teams to ensure the quality of the data being collected, while providing feedback to\n\n\nthe enumerators for further quality control.\n\n\n**3.2. Study design**\n\n\nThree approaches to collecting household food consumption data – 14-day diary, 7-day recall with\n\n\nshort items list, and 7-day recall with long items list – are compared in this study. To better\n\n\nunderstand the differences in household food consumption data collection, this methodological\n\n\nstudy replicated the HBS 2017/18 (14-day diary) and the NPS (7-day recall with short list)\n\n\nmethodologies, as well as a variant/combination of the NPS (7-day recall with long list). Given\n\n\nthat the HBS is the survey that NBS and OCGS use to produce official poverty statistics, the diary\n\n\nmethodology served as the benchmark and households assigned to this group served as the control\n\n\ngroup, and the two 7-day recall arms are compared to the diary. In addition, all households in the\n\n\nstudy received a Food Away from Home (FAFH) module implemented at the individual level to\n\n\ncapture expenditure on food consumed outside the home, such as from restaurants, joints, etc.\n\n\nHouseholds were randomly assigned to each study arm in all clusters to ensure non-contamination\n\n\n(Beegle et al. 2012; Brzozowski et al 2017). The design details of each method are described\n\n\nbelow.\n\n\n7\n\nMention: \"HBS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Because it produces the official poverty statistics, the diary method is benchmarked against it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001556:19:0", "occurrence_id": "5cfdb12fd7aad4e4aa370b8d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 6: Value added by VAT firms vs GDP\n\n\nThis graph compares the sector-level contribution to national GDP to the value added (sales - purchases) of firms\ncovered in the administrative tax records for 2019.\n\n\nThese results align with findings by B¨ohme and Thiele (2014); Zhou (2022) who document similar\n\n\nlinking patterns between formal and informal firms in Benin, Burkina Faso, Cˆote d’Ivoire, Mali,\n\nS´en´egal, Togo (B¨ohme and Thiele, 2014) and India (Zhou, 2022) respectively. <sup>20</sup> Gadenne et al.\n\n\n(2022) use granular data covering a wide range of sectors to document that non-VAT paying firms\n\n\nthat participate in a simplified tax scheme in West Bengal, India, also sell little to VAT-paying\n\n\nfirms, but purchase between 50-75% of their inputs from VAT-paying businesses. The higher\n\n\nincidence of informality in downstream sectors, as well as informal firms being more likely to\n\n\npurchase from larger firms rather than vice versa, are consistent with the underlying enforcement\n\n\nstructure of VAT systems. The VAT system incentivises downstream firms to request receipts\n\n\nfrom suppliers to claim input VAT deductions against their output VAT obligations. However,\n\n\nend consumers and VAT-exempt entities lack incentives to demand receipts as they cannot claim\n\n\nVAT refunds (Naritomi, 2019). Consequently, we indeed expect downstream sectors to exhibit\n\n\nhigher shares of economic activity outside the VAT system.\n\n\nsmall, and medium firms. Small enterprises, defined as those with up to 50 employees and KShs five million\nannual turnover (KNBS, 2016) may fall below the VAT threshold, but medium-sized firms often exceed it.\n\n20B¨ohme and Thiele (2014) look at multiple sectors, while Zhou (2022) focuses on manufacturing. For a review\non links between the formal and informal economy, see Meagher\n\nMention: \"administrative tax records\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Sectoral value added of registered firms is compared with national GDP contributions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001558:10:0", "occurrence_id": "b514fc8c98a2606a49f1abb0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: obtain shapefiles with land use typologies from the municipal government, which allows us\n\n\nto approximately classify floor space in each location as either commercial or residential. We\n\n\nfurther obtain the Amman zoning restriction rules, which assign to each building type the\n\n\npercentage of the plot that can be used for construction and the maximum number of floors\n\n\nthat can be constructed (among other constraints not modeled here). We merge the land use\n\n\ntypology of each plot to the relevant building categories from the zoning rules to calculate\n\n\nthe maximum floor space that can legally be developed in each location. We perform this\n\n\nstep separately for commercial and residential building types.\n\n\n**Travel Times with Road Travel.** To compute average travel times between all locations,\n\n\nwe use the road network for Amman and assign average travel speeds to different types of\n\n\nroad segments based on estimates from the World Bank Transportation Unit. Specifically,\n\n\nwe assign a speed of 35 km/hour for primary roads, 20 km/h for secondary roads, 10 km/h\n\n\nfor tertiary roads, 8 km/h for other road types, and 5 km/h for walking. For metro lines we\n\n\nassign a speed of 35 km/h and for monorail 45 km/h. In the counterfactual, we compute\n\n\ncommuting times with the new BRT routes to which we assign a traveling speed of 25 km/h.\n\n\nFigure 4 shows average commuting times for each residency neighborhood to all possible\n\n\nwork locations where we weigh travel times of each work destination by the number of jobs\n\n\navailable in this destination. Aligned with the previous results on employment density, we\n\n\nsee that central neighborhoods are best connected to jobs while the outskirts have larger\n\n\naverage commuting costs to available employment opportunities. Commuting times are par\n\nticularly large for the south and some of the most western and northern neighborhoods.\n\n### **5 Calibration Results**\n\n\n9\n\nMention: \"shapefiles with land use typologies\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"These municipal layers allow floor space to be classified as commercial or residential.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001559:10:0", "occurrence_id": "3bbadc132235c60ee04ba9fb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ### **3 Data**\n\nThe goal of the paper is to assess the direct and indirect effects of export expansion on local labor\n\n\nmarket outcomes in Viet Nam while accounting for supply chain linkages. To do this, we exploit\n\n\nvariation in export expansion across provinces and industries between 2010 and 2019 and combine\n\n\nexport data from the United Nations Commodity Trade Statistics (UNCOMTRADE) data, input\n\noutput coefficient matrix from Global Trade Analysis Project (GTAP) data, and information on\n\n\nlocal labor market outcomes from Viet Nam’s Labor Force Survey (LFS) data. Details on each\n\n\ndataset and cleaning techniques are described below.\n\n##### **3.1 Labor Force Data**\n\n\nOur main source of labor market data is the LFS provided by General Statistics office of Viet Nam\n\n\n(GSO) between 2010 and 2019, a period during which it was implemented every year. The LFS\n\n\nobservations collect information in a host of areas including key labor market, household, and\n\n\nindividual demographic characteristics.\n\n\nOur analysis looks at two main sets of outcome variables: wage outcomes and employment\n\n\noutcomes. The wage outcome data sets include real annual wages, real annual income, college\n\n\ndegree wage premium and gender wage premium. The employment outcome sets include em\n\nployment rate, inactive rate, informality status, and female labor force participation. All of the\n\n\noutcomes are constructed from survey questionnaires, of which the wage outcomes are calculated\n\n\nat the province x sector level, while the employment outcomes are aggregated at the province\n\n\nlevel because we do not have sector information for those who are not employed. Over the period\n\n\n2010 to 2019, several changes were introduced in the Viet Namese LFS, together with updates\n\n\nin concepts and definitions. These have been standardized to make key labor market outcomes,\n\n\nadministrative geographies, as well as industry classifications, comparable over time.\n\n\n8\n\nMention: \"United Nations Commodity Trade Statistics\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Province-industry export expansion over the decade is measured from these flows.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001559:10:0", "occurrence_id": "d5e2a1a4f8e0300835ca41f2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ### **3 Data**\n\nThe goal of the paper is to assess the direct and indirect effects of export expansion on local labor\n\n\nmarket outcomes in Viet Nam while accounting for supply chain linkages. To do this, we exploit\n\n\nvariation in export expansion across provinces and industries between 2010 and 2019 and combine\n\n\nexport data from the United Nations Commodity Trade Statistics (UNCOMTRADE) data, input\n\noutput coefficient matrix from Global Trade Analysis Project (GTAP) data, and information on\n\n\nlocal labor market outcomes from Viet Nam’s Labor Force Survey (LFS) data. Details on each\n\n\ndataset and cleaning techniques are described below.\n\n##### **3.1 Labor Force Data**\n\n\nOur main source of labor market data is the LFS provided by General Statistics office of Viet Nam\n\n\n(GSO) between 2010 and 2019, a period during which it was implemented every year. The LFS\n\n\nobservations collect information in a host of areas including key labor market, household, and\n\n\nindividual demographic characteristics.\n\n\nOur analysis looks at two main sets of outcome variables: wage outcomes and employment\n\n\noutcomes. The wage outcome data sets include real annual wages, real annual income, college\n\n\ndegree wage premium and gender wage premium. The employment outcome sets include em\n\nployment rate, inactive rate, informality status, and female labor force participation. All of the\n\n\noutcomes are constructed from survey questionnaires, of which the wage outcomes are calculated\n\n\nat the province x sector level, while the employment outcomes are aggregated at the province\n\n\nlevel because we do not have sector information for those who are not employed. Over the period\n\n\n2010 to 2019, several changes were introduced in the Viet Namese LFS, together with updates\n\n\nin concepts and definitions. These have been standardized to make key labor market outcomes,\n\n\nadministrative geographies, as well as industry classifications, comparable over time.\n\n\n8\n\nMention: \"Global Trade Analysis Project\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its input-output coefficients translate export shocks into local demand.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001559:10:0", "occurrence_id": "77ecb898fd1cf1bcd373e5c6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ### **3 Data**\n\nThe goal of the paper is to assess the direct and indirect effects of export expansion on local labor\n\n\nmarket outcomes in Viet Nam while accounting for supply chain linkages. To do this, we exploit\n\n\nvariation in export expansion across provinces and industries between 2010 and 2019 and combine\n\n\nexport data from the United Nations Commodity Trade Statistics (UNCOMTRADE) data, input\n\noutput coefficient matrix from Global Trade Analysis Project (GTAP) data, and information on\n\n\nlocal labor market outcomes from Viet Nam’s Labor Force Survey (LFS) data. Details on each\n\n\ndataset and cleaning techniques are described below.\n\n##### **3.1 Labor Force Data**\n\n\nOur main source of labor market data is the LFS provided by General Statistics office of Viet Nam\n\n\n(GSO) between 2010 and 2019, a period during which it was implemented every year. The LFS\n\n\nobservations collect information in a host of areas including key labor market, household, and\n\n\nindividual demographic characteristics.\n\n\nOur analysis looks at two main sets of outcome variables: wage outcomes and employment\n\n\noutcomes. The wage outcome data sets include real annual wages, real annual income, college\n\n\ndegree wage premium and gender wage premium. The employment outcome sets include em\n\nployment rate, inactive rate, informality status, and female labor force participation. All of the\n\n\noutcomes are constructed from survey questionnaires, of which the wage outcomes are calculated\n\n\nat the province x sector level, while the employment outcomes are aggregated at the province\n\n\nlevel because we do not have sector information for those who are not employed. Over the period\n\n\n2010 to 2019, several changes were introduced in the Viet Namese LFS, together with updates\n\n\nin concepts and definitions. These have been standardized to make key labor market outcomes,\n\n\nadministrative geographies, as well as industry classifications, comparable over time.\n\n\n8\n\nMention: \"Labor Force Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Annual rounds provide the local labour market outcomes.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001559:10:0", "occurrence_id": "cab1b74871963bc24e32b1f8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ### **3 Data**\n\nThe goal of the paper is to assess the direct and indirect effects of export expansion on local labor\n\n\nmarket outcomes in Viet Nam while accounting for supply chain linkages. To do this, we exploit\n\n\nvariation in export expansion across provinces and industries between 2010 and 2019 and combine\n\n\nexport data from the United Nations Commodity Trade Statistics (UNCOMTRADE) data, input\n\noutput coefficient matrix from Global Trade Analysis Project (GTAP) data, and information on\n\n\nlocal labor market outcomes from Viet Nam’s Labor Force Survey (LFS) data. Details on each\n\n\ndataset and cleaning techniques are described below.\n\n##### **3.1 Labor Force Data**\n\n\nOur main source of labor market data is the LFS provided by General Statistics office of Viet Nam\n\n\n(GSO) between 2010 and 2019, a period during which it was implemented every year. The LFS\n\n\nobservations collect information in a host of areas including key labor market, household, and\n\n\nindividual demographic characteristics.\n\n\nOur analysis looks at two main sets of outcome variables: wage outcomes and employment\n\n\noutcomes. The wage outcome data sets include real annual wages, real annual income, college\n\n\ndegree wage premium and gender wage premium. The employment outcome sets include em\n\nployment rate, inactive rate, informality status, and female labor force participation. All of the\n\n\noutcomes are constructed from survey questionnaires, of which the wage outcomes are calculated\n\n\nat the province x sector level, while the employment outcomes are aggregated at the province\n\n\nlevel because we do not have sector information for those who are not employed. Over the period\n\n\n2010 to 2019, several changes were introduced in the Viet Namese LFS, together with updates\n\n\nin concepts and definitions. These have been standardized to make key labor market outcomes,\n\n\nadministrative geographies, as well as industry classifications, comparable over time.\n\n\n8\n\nMention: \"LFS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"It is the main source of labour market data, fielded every year across the period.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001559:10:0", "occurrence_id": "6bec56522c0740ec7ab4dcd4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ### **3 Data**\n\nThe goal of the paper is to assess the direct and indirect effects of export expansion on local labor\n\n\nmarket outcomes in Viet Nam while accounting for supply chain linkages. To do this, we exploit\n\n\nvariation in export expansion across provinces and industries between 2010 and 2019 and combine\n\n\nexport data from the United Nations Commodity Trade Statistics (UNCOMTRADE) data, input\n\noutput coefficient matrix from Global Trade Analysis Project (GTAP) data, and information on\n\n\nlocal labor market outcomes from Viet Nam’s Labor Force Survey (LFS) data. Details on each\n\n\ndataset and cleaning techniques are described below.\n\n##### **3.1 Labor Force Data**\n\n\nOur main source of labor market data is the LFS provided by General Statistics office of Viet Nam\n\n\n(GSO) between 2010 and 2019, a period during which it was implemented every year. The LFS\n\n\nobservations collect information in a host of areas including key labor market, household, and\n\n\nindividual demographic characteristics.\n\n\nOur analysis looks at two main sets of outcome variables: wage outcomes and employment\n\n\noutcomes. The wage outcome data sets include real annual wages, real annual income, college\n\n\ndegree wage premium and gender wage premium. The employment outcome sets include em\n\nployment rate, inactive rate, informality status, and female labor force participation. All of the\n\n\noutcomes are constructed from survey questionnaires, of which the wage outcomes are calculated\n\n\nat the province x sector level, while the employment outcomes are aggregated at the province\n\n\nlevel because we do not have sector information for those who are not employed. Over the period\n\n\n2010 to 2019, several changes were introduced in the Viet Namese LFS, together with updates\n\n\nin concepts and definitions. These have been standardized to make key labor market outcomes,\n\n\nadministrative geographies, as well as industry classifications, comparable over time.\n\n\n8\n\nMention: \"Viet Namese LFS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Design and definitional changes over the period are standardised so outcomes and geographies are comparable.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001560:16:0", "occurrence_id": "8c50031629c053aaab404383", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: risk, versus a 0.86 reduction in risk when concentrating on the areas where the top 25% of sewer\nbursts are concentrated. The efficiency and cost effectiveness of reducing cholera risk is therefore\nmuch greater when concentrating areas that fall within the top 25th percentile of sewer bursts. It\nis also more efficient to provide piped water to within 500 meters of the existing network (7.55),\nrather than to extend access to the whole city (8.2) when considering their relative impact on\nreducing the number of cholera cases.\n\nFinally, we also see how extending access to the sewer network in its state at the time of the data\ncollection in fact seems to increase cholera risk. This finding may seem perverse given that\naccording to the Joint Monitoring Program of the WHO, which set the standards for sanitation\nglobally, sewered sanitation is at the top of the improved sanitation ladder. However, it serves to\nhighlight the importance of proper maintenance of existing sanitation networks rather than\nnecessarily emphasizing new construction. This is particularly important in a context where the\ninstitutional capacity and financial resources to maintain large-scale sanitation systems may be\nlacking.\n\nAn important limitation of the study is that the analysis presented relies on observational data and\nthus we draw correlations and associations, rather than drawing causal inferences. This clearly has\nimportant implications when interpreting the scenario analysis. Additionally, there are likely to be\nsome errors in case data such as the locations and times of case identification. Similarly, a number\nof layers used in the analysis, such as proportion of households with access to piped water, were\ngenerated using DHS data, which have been spatially obfuscated. As a result, this may have\nintroduced error into the layer, which could obscure, or falsely lead to, associations with cases. In\naddition, sewer and water burst density was based on spatiotemporally referenced complaints data,\nwhich is obviously a proxy measure with a number of reported and recorded issues. Data from\ndirect inspections of sewer and water networks would have\n\nMention: \"DHS data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Spatially obfuscated locations may have introduced error into the piped water access layer.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001561:8:0", "occurrence_id": "5dc2441b4670becd1e2255d4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: If a regression approach were used, the additional consumption in 2016 would be completed by adding\n\nused to find the most similar household heads in both surveys (within the set of observations that share\nsome residuals to 𝑦𝑦�𝑖𝑖2016. That differs from the method presented in this paper since 𝑦𝑦�𝑖𝑖2016 and 𝑦𝑦�𝑖𝑖2022 are\ncommon characteristics). For every observation in the 2016 survey, a set of neighbors (in terms of 𝑦𝑦�𝑖𝑖2016\ndifference between the two linear predictions. Among these neighbors, one is randomly selected using\nand 𝑦𝑦�𝑖𝑖2022 closeness) is found in the 2022 HIES by looking for observations with the smallest absolute\nuniform selection probabilities within the subset of closest neighbors. In addition, the matching method\nrequires that the closest neighbors are searched only within a pre-defined subset of households having\nthe same stratum and per capita consumption decile. The observed share of additional consumption with 𝑎 **𝑎** 𝑎𝑎 ) from that neighbor in 2022 is\nrespect to the comparable consumption (𝑠𝑠ℎ𝑓 **𝑓** 𝑓 **𝑓** 𝑖𝑖2022𝑎 **𝑎** 𝑎𝑎 and 𝑠𝑠ℎ𝑛 **𝑛** 𝑛 **𝑛** 𝑛𝑛𝑖𝑖2022\n\n𝑎 **𝑎** 𝑎𝑎,\n\nthen imputed to the corresponding observation in 2016 (obtaining 𝑠𝑠ℎ𝑓 **𝑓** 𝑓 **𝑓** 𝑖𝑖2016𝑎 **𝑎** 𝑎𝑎 and ��𝑠ℎ𝑛 **𝑛** 𝑛 **𝑛** 𝑛𝑛𝑖𝑖2016\n\n\n\n𝑎 **𝑎** 𝑎𝑎\n\n**𝑓** **𝑓** 𝑖𝑖2022𝑎 **𝑎** 𝑎𝑎 and 𝑠𝑠ℎ𝑛 **𝑛** 𝑛 **𝑛** 𝑛𝑛𝑖𝑖2022\n\n**𝑓** **𝑓**\n\n\n\n**𝑎** ) from that neighbor in 2022 is\n\n𝑎 **𝑎** 𝑎𝑎\n\n**𝑓**\n\nMention: \"2022 HIES\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Closest neighbours are matched by selecting observations with the smallest absolute differences.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001566:21:0", "occurrence_id": "30d0d8480b1e3c6722c13cf3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Evidence** **from** **the** **Delhi** **Primary** **Survey** The results on female time use so far are\n\nbased on the CPHS data which does not separate travel times into purpose of travel (i.e.,\n\ncommuting to work vs non-work related trips). The Delhi Survey, on the other hand,\n\nreports the purpose of travel and the average travel distance for each respondent. While\nthis data does not report employment status, the survey answer ‘ _travel for work_ ’ is meant\nto capture travel _to work_ in the case of employed women and travel _to search_ for work in\n\nthe case of unemployed women.\n\n\nWe use the Delhi data to compare the purpose of travel and the average travel distance\n\nbetween new and continuous users of bus transit in the post-implementation periods. We\n\nacknowledge that the new users are induced into using the buses because they are free\n\nand hence are selected relative to incumbent users. The comparison group of continuous\n\nusers is matched on observable characteristics to the new users as discussed in Section\n\nE.2. While this does not address the selection issue completely, this comparison is still\n\nrevealing and useful. With this disclaimer in mind, we estimate the following regression:\n\n\n_Yi_ = _β_ _I_ new users + _Xi_ + _ϵi_, (3)\n\n\nwhere _Yi_ is a dummy variable for travel purposes which is equal to one if it is for work\nor a dummy variable for average travel distance via buses which is equal to one if the\naverage distance exceeds 10 kilometers. _I_ new users is a dummy variable set to one for\nnew users and zero for continuous users. _X_ is a vector of control variables including an\n\nindividual’s marital status and education level. The results are reported in Table A8.\n\n\nIn column 1, we find that new bus users are\n\nMention: \"CPHS data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Its time use measures do not separate travel purpose, limiting the commuting analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001566:21:0", "occurrence_id": "102267be77c773f3a88a5365", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Evidence** **from** **the** **Delhi** **Primary** **Survey** The results on female time use so far are\n\nbased on the CPHS data which does not separate travel times into purpose of travel (i.e.,\n\ncommuting to work vs non-work related trips). The Delhi Survey, on the other hand,\n\nreports the purpose of travel and the average travel distance for each respondent. While\nthis data does not report employment status, the survey answer ‘ _travel for work_ ’ is meant\nto capture travel _to work_ in the case of employed women and travel _to search_ for work in\n\nthe case of unemployed women.\n\n\nWe use the Delhi data to compare the purpose of travel and the average travel distance\n\nbetween new and continuous users of bus transit in the post-implementation periods. We\n\nacknowledge that the new users are induced into using the buses because they are free\n\nand hence are selected relative to incumbent users. The comparison group of continuous\n\nusers is matched on observable characteristics to the new users as discussed in Section\n\nE.2. While this does not address the selection issue completely, this comparison is still\n\nrevealing and useful. With this disclaimer in mind, we estimate the following regression:\n\n\n_Yi_ = _β_ _I_ new users + _Xi_ + _ϵi_, (3)\n\n\nwhere _Yi_ is a dummy variable for travel purposes which is equal to one if it is for work\nor a dummy variable for average travel distance via buses which is equal to one if the\naverage distance exceeds 10 kilometers. _I_ new users is a dummy variable set to one for\nnew users and zero for continuous users. _X_ is a vector of control variables including an\n\nindividual’s marital status and education level. The results are reported in Table A8.\n\n\nIn column 1, we find that new bus users are\n\nMention: \"Delhi Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It reports travel purpose and distance but not employment status.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001568:13:0", "occurrence_id": "83bb2df02a76c5ee1708a8f9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: For countries outside Europe and Central Asia, we use the GDIM dataset, which provides estimates\n\nof intergenerational educational mobility for four birth cohorts (individuals born in the 1950s, the\n\n1960s, the 1970s, and the 1980s) across 114 countries (van der Weide and others 2023). It draws\n\non microdata from social and household surveys, which contain retrospective information on\n\nparental education. Similarly to the LiTS, the GDIM relies on a correspondence between\n\ncategorical information on attained levels of education and the equivalent number of years of\n\nschooling.\n\n\nOur empirical analysis includes variables from other sources. GDP per capita and population come\n\nfrom the World Development Indicators <u>(World Bank 2023). Historical GDP per capita and</u>\n\npopulation (used to estimate GDP per capita and population at the decade of birth of the different\n\ncohorts) are from the 2020 edition of the Maddison Project Database (Maddison Project Database\n\n<u>2020; Bolt and van Zanden 2020). Historical infant mortality rates from 1950 onward are estimated</u>\n\nby the United Nations’ World Population Prospects Population Division (UNDESA 2022); rates\n\nfor 1930–50 are from various historical sources.\n\n\nOur cross-country panel of educational mobility and economic outcomes includes 1,419 country\nyear observations in the period 2000-2020, covering 108 countries. However, we have 10 or more\n\nnon-missing country-year observations only for 68 countries. <sup>14</sup> [^14: These are 18 countries in Europe and Central Asia, 16 countries in Latin America and the Caribbean, 24 highincome countries not elsewhere included, and 10 other developing countries in Asia and Africa.] This sub-panel of 1,244 country\nyear observations is the primary sample of our empirical analysis. <u>Table 1</u> presents the descriptive\n\nstatistics of the main variables of our analysis for the cross-country sample.\n\n\n**6.** **Educational mobility across countries and generations**\n\n\nThis section describes the patterns of intergenerational educational mobility observed across\n\ncountries and\n\nMention: \"GDIM dataset\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Intergenerational educational mobility for four birth cohorts is taken from this dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001568:13:0", "occurrence_id": "dcac90e1144f35d0a8e2c6d8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: For countries outside Europe and Central Asia, we use the GDIM dataset, which provides estimates\n\nof intergenerational educational mobility for four birth cohorts (individuals born in the 1950s, the\n\n1960s, the 1970s, and the 1980s) across 114 countries (van der Weide and others 2023). It draws\n\non microdata from social and household surveys, which contain retrospective information on\n\nparental education. Similarly to the LiTS, the GDIM relies on a correspondence between\n\ncategorical information on attained levels of education and the equivalent number of years of\n\nschooling.\n\n\nOur empirical analysis includes variables from other sources. GDP per capita and population come\n\nfrom the World Development Indicators <u>(World Bank 2023). Historical GDP per capita and</u>\n\npopulation (used to estimate GDP per capita and population at the decade of birth of the different\n\ncohorts) are from the 2020 edition of the Maddison Project Database (Maddison Project Database\n\n<u>2020; Bolt and van Zanden 2020). Historical infant mortality rates from 1950 onward are estimated</u>\n\nby the United Nations’ World Population Prospects Population Division (UNDESA 2022); rates\n\nfor 1930–50 are from various historical sources.\n\n\nOur cross-country panel of educational mobility and economic outcomes includes 1,419 country\nyear observations in the period 2000-2020, covering 108 countries. However, we have 10 or more\n\nnon-missing country-year observations only for 68 countries. <sup>14</sup> [^14: These are 18 countries in Europe and Central Asia, 16 countries in Latin America and the Caribbean, 24 highincome countries not elsewhere included, and 10 other developing countries in Asia and Africa.] This sub-panel of 1,244 country\nyear observations is the primary sample of our empirical analysis. <u>Table 1</u> presents the descriptive\n\nstatistics of the main variables of our analysis for the cross-country sample.\n\n\n**6.** **Educational mobility across countries and generations**\n\n\nThis section describes the patterns of intergenerational educational mobility observed across\n\ncountries and\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"GDP per capita and population provide the contemporary economic controls.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001568:13:0", "occurrence_id": "682896dac72a53f2f5688d39", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: For countries outside Europe and Central Asia, we use the GDIM dataset, which provides estimates\n\nof intergenerational educational mobility for four birth cohorts (individuals born in the 1950s, the\n\n1960s, the 1970s, and the 1980s) across 114 countries (van der Weide and others 2023). It draws\n\non microdata from social and household surveys, which contain retrospective information on\n\nparental education. Similarly to the LiTS, the GDIM relies on a correspondence between\n\ncategorical information on attained levels of education and the equivalent number of years of\n\nschooling.\n\n\nOur empirical analysis includes variables from other sources. GDP per capita and population come\n\nfrom the World Development Indicators <u>(World Bank 2023). Historical GDP per capita and</u>\n\npopulation (used to estimate GDP per capita and population at the decade of birth of the different\n\ncohorts) are from the 2020 edition of the Maddison Project Database (Maddison Project Database\n\n<u>2020; Bolt and van Zanden 2020). Historical infant mortality rates from 1950 onward are estimated</u>\n\nby the United Nations’ World Population Prospects Population Division (UNDESA 2022); rates\n\nfor 1930–50 are from various historical sources.\n\n\nOur cross-country panel of educational mobility and economic outcomes includes 1,419 country\nyear observations in the period 2000-2020, covering 108 countries. However, we have 10 or more\n\nnon-missing country-year observations only for 68 countries. <sup>14</sup> [^14: These are 18 countries in Europe and Central Asia, 16 countries in Latin America and the Caribbean, 24 highincome countries not elsewhere included, and 10 other developing countries in Asia and Africa.] This sub-panel of 1,244 country\nyear observations is the primary sample of our empirical analysis. <u>Table 1</u> presents the descriptive\n\nstatistics of the main variables of our analysis for the cross-country sample.\n\n\n**6.** **Educational mobility across countries and generations**\n\n\nThis section describes the patterns of intergenerational educational mobility observed across\n\ncountries and\n\nMention: \"Maddison Project Database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Historical GDP per capita and population are taken from this long-run database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001568:13:0", "occurrence_id": "c6fb95a7fd656f8997153240", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: For countries outside Europe and Central Asia, we use the GDIM dataset, which provides estimates\n\nof intergenerational educational mobility for four birth cohorts (individuals born in the 1950s, the\n\n1960s, the 1970s, and the 1980s) across 114 countries (van der Weide and others 2023). It draws\n\non microdata from social and household surveys, which contain retrospective information on\n\nparental education. Similarly to the LiTS, the GDIM relies on a correspondence between\n\ncategorical information on attained levels of education and the equivalent number of years of\n\nschooling.\n\n\nOur empirical analysis includes variables from other sources. GDP per capita and population come\n\nfrom the World Development Indicators <u>(World Bank 2023). Historical GDP per capita and</u>\n\npopulation (used to estimate GDP per capita and population at the decade of birth of the different\n\ncohorts) are from the 2020 edition of the Maddison Project Database (Maddison Project Database\n\n<u>2020; Bolt and van Zanden 2020). Historical infant mortality rates from 1950 onward are estimated</u>\n\nby the United Nations’ World Population Prospects Population Division (UNDESA 2022); rates\n\nfor 1930–50 are from various historical sources.\n\n\nOur cross-country panel of educational mobility and economic outcomes includes 1,419 country\nyear observations in the period 2000-2020, covering 108 countries. However, we have 10 or more\n\nnon-missing country-year observations only for 68 countries. <sup>14</sup> [^14: These are 18 countries in Europe and Central Asia, 16 countries in Latin America and the Caribbean, 24 highincome countries not elsewhere included, and 10 other developing countries in Asia and Africa.] This sub-panel of 1,244 country\nyear observations is the primary sample of our empirical analysis. <u>Table 1</u> presents the descriptive\n\nstatistics of the main variables of our analysis for the cross-country sample.\n\n\n**6.** **Educational mobility across countries and generations**\n\n\nThis section describes the patterns of intergenerational educational mobility observed across\n\ncountries and\n\nMention: \"World Population Prospects Population Division\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Historical infant mortality rates from 1950 onward are estimated in these prospects.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001585:5:0", "occurrence_id": "001c05324fea3a1e1afd7de2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: propensity score matching methods. They report earnings losses during the first five years after\n\n\ndisplacement that range from 18 to 35 percent for workers whose firm closes, and from 14 to 25\n\n\npercent for workers who exit a firm that suffers a mass layoff.\n\n\nThe literature has poorly studied the impacts on labor market outcomes of job displacement in\n\n\ndeveloping countries. Eslava, Haltiwanger, Kugler and Kugler (2010) study the Colombian labor\n\n\nmarket and look at how trade liberalization affected workers displaced due to plant closings. The\n\n\nauthors document the job reallocation that followed trade liberalization in the country and analyze\n\n\nworkers´ outcomes after they reentered the workforce. They find that displaced workers experience\n\n\na reduction in wages and tenure (time in a job) and a lower probability of working in the formal\n\n\nsector within the first two years after their displacement. Yet they find that Colombian workers’\n\n\nwages fully recover three years after a plant closing.\n\n\nFor the case of Mexico, Kaplan, Martinez, and Robertson (2005) look at changes in formal wages\n\n\nfollowing job displacement. They use administrative data from Social Security records to study\n\n\nthe effect of job displacement on wages under different economic conditions. They find large wage\n\n\nlosses when displacement occurs in periods of high unemployment and in regions with less\n\n\neconomic activity. Their study is subject to two limitations. First, the Social Security records only\n\n\ncollect information on formal employment and do not differentiate between informal employment,\n\n\nunemployment, and individuals who have left the labor force. Second, the Social Security records\n\n\ndo not make it possible to identify the source of job separation, and as the authors acknowledge,\n\n\n“the cause of separation is important because workers who leave voluntarily are more likely to\n\n\nhave more positive economic prospects after they leave their current firm. Including voluntary\n\n\nseparation would, therefore, bias the estimated effects of displaced workers upwards.” (Kaplan _et_\n\n\n_al_ ., 2005).\n\n\n3\n\nMention: \"administrative data from Social Security records\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"A cited study uses these records to estimate wage losses after displacement.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001585:5:0", "occurrence_id": "7ea16ab1440704032f1fc329", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: propensity score matching methods. They report earnings losses during the first five years after\n\n\ndisplacement that range from 18 to 35 percent for workers whose firm closes, and from 14 to 25\n\n\npercent for workers who exit a firm that suffers a mass layoff.\n\n\nThe literature has poorly studied the impacts on labor market outcomes of job displacement in\n\n\ndeveloping countries. Eslava, Haltiwanger, Kugler and Kugler (2010) study the Colombian labor\n\n\nmarket and look at how trade liberalization affected workers displaced due to plant closings. The\n\n\nauthors document the job reallocation that followed trade liberalization in the country and analyze\n\n\nworkers´ outcomes after they reentered the workforce. They find that displaced workers experience\n\n\na reduction in wages and tenure (time in a job) and a lower probability of working in the formal\n\n\nsector within the first two years after their displacement. Yet they find that Colombian workers’\n\n\nwages fully recover three years after a plant closing.\n\n\nFor the case of Mexico, Kaplan, Martinez, and Robertson (2005) look at changes in formal wages\n\n\nfollowing job displacement. They use administrative data from Social Security records to study\n\n\nthe effect of job displacement on wages under different economic conditions. They find large wage\n\n\nlosses when displacement occurs in periods of high unemployment and in regions with less\n\n\neconomic activity. Their study is subject to two limitations. First, the Social Security records only\n\n\ncollect information on formal employment and do not differentiate between informal employment,\n\n\nunemployment, and individuals who have left the labor force. Second, the Social Security records\n\n\ndo not make it possible to identify the source of job separation, and as the authors acknowledge,\n\n\n“the cause of separation is important because workers who leave voluntarily are more likely to\n\n\nhave more positive economic prospects after they leave their current firm. Including voluntary\n\n\nseparation would, therefore, bias the estimated effects of displaced workers upwards.” (Kaplan _et_\n\n\n_al_ ., 2005).\n\n\n3\n\nMention: \"Social Security records\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"validate\", \"summary\": \"They capture only formal employment, so informal work, unemployment and labour force exit cannot be distinguished.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001587:60:0", "occurrence_id": "4c76434a915d08de93c870c8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ###### Table A8: Heterogeneity in the Joint Effect of Internet and Roads on Employment: Place of Residence\n\n(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12)\n**Worked** **during** **Currently**\n**past** **year** **working** **Hours** **worked**\n\n\n**Urban** **Rural** **high** **Low** **Urban** **Rural** **high** **Low** **Urban** **Rural** **high** **Low**\n**Sample** **Sample** **developed** **developed** **Sample** **Sample** **developed** **developed** **Sample** **Sample** **developed** **developed**\n**Sample** **Sample** **Sample** **Sample** **Sample** **Sample**\n**DHS** **LSMS**\n\n\nSubmarine internet 0.210* 0.0369 0.0891 0.0603** 0.190* 0.0165 0.0419 0.0514* 2.005 -0.390** -0.0893 -1.002***\n(0.101) (0.0239) (0.0516) (0.0209) (0.0920) (0.0240) (0.0443) (0.0209) (2.464) (0.189) (0.\n\nMention: \"DHS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its samples form the outcome data in the columns estimated on developing and developed samples.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001587:60:0", "occurrence_id": "1a0fd199c6928a880928175c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ###### Table A8: Heterogeneity in the Joint Effect of Internet and Roads on Employment: Place of Residence\n\n(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12)\n**Worked** **during** **Currently**\n**past** **year** **working** **Hours** **worked**\n\n\n**Urban** **Rural** **high** **Low** **Urban** **Rural** **high** **Low** **Urban** **Rural** **high** **Low**\n**Sample** **Sample** **developed** **developed** **Sample** **Sample** **developed** **developed** **Sample** **Sample** **developed** **developed**\n**Sample** **Sample** **Sample** **Sample** **Sample** **Sample**\n**DHS** **LSMS**\n\n\nSubmarine internet 0.210* 0.0369 0.0891 0.0603** 0.190* 0.0165 0.0419 0.0514* 2.005 -0.390** -0.0893 -1.002***\n(0.101) (0.0239) (0.0516) (0.0209) (0.0920) (0.0240) (0.0443) (0.0209) (2.464) (0.189) (0.\n\nMention: \"LSMS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its household surveys provide an alternative outcome sample to the health survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001591:18:0", "occurrence_id": "b37fe743759db328fa90f4a9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: (Lenzen et al., 2012). These data provide a 26-harmonized sector version for almost all\n\n\ncountries of the world, which we use to split the Rest of the World category in WIOD. We\n\n\ndo this by first aggregating all countries in EORA with a 2009 GDP below 5bn USD into\n\n\nan aggregate rest of the world (ROW) category. For all countries in EORA (including the\n\n\nnew ROW group), we then exclude all countries contained in WIOD and determine the\n\n\nshare of each country on the total supply and demand to and from all sectors (to all EORA\n\n\ncountries, including those contained in WIOD). We then use these shares to split the WIOD\n\n\nROW category into these countries from EORA. We adjust the WIOD sectoral coverage to\n\n\nmatch both that of EORA and that of the environmental satellite accounts of WIOD, which\n\n\nwe need to determine the production emission-related parameters. A list of the final sector\n\n\ncoverage can be found in Table C1 and an overview of the sector mappings in Table C2.\n\n\nLast, we enforce balanced trade between countries.\n\n\nWe then use this input output matrix to calibrate the input shares _γ_ and _β_ as well as\n\n\nthe trade matrices _Y_ .\n\n\nThe environmental satellite accounts are most recently available for all countries in WIOD\n\n\nonly in 2009, which is why we base all of our analysis on data from that year. We make use\n\n\nof CO2 (carbon dioxide), CO4 (methane) and N2O (nitrous oxide) data, where we convert\n\n\nthe latter two into CO2 equivalents based on their global warming potential. The conversion\n\n\nfactor for methane is hereby 28 and the one for nitrous oxide 265, based on their cumulative\n\n\nforcing over 100 years, taken from IPCC, 2014. For Luxembourg, Romania, Switzerland,\n\n\nCroatia, and Norway and all countries not included in WIOD we rely on the EORA\n\nMention: \"environmental satellite accounts of WIOD\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Sectoral coverage is matched to these accounts so production emission parameters can be set.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001591:18:0", "occurrence_id": "7383ecf32b511f034de02cc1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: (Lenzen et al., 2012). These data provide a 26-harmonized sector version for almost all\n\n\ncountries of the world, which we use to split the Rest of the World category in WIOD. We\n\n\ndo this by first aggregating all countries in EORA with a 2009 GDP below 5bn USD into\n\n\nan aggregate rest of the world (ROW) category. For all countries in EORA (including the\n\n\nnew ROW group), we then exclude all countries contained in WIOD and determine the\n\n\nshare of each country on the total supply and demand to and from all sectors (to all EORA\n\n\ncountries, including those contained in WIOD). We then use these shares to split the WIOD\n\n\nROW category into these countries from EORA. We adjust the WIOD sectoral coverage to\n\n\nmatch both that of EORA and that of the environmental satellite accounts of WIOD, which\n\n\nwe need to determine the production emission-related parameters. A list of the final sector\n\n\ncoverage can be found in Table C1 and an overview of the sector mappings in Table C2.\n\n\nLast, we enforce balanced trade between countries.\n\n\nWe then use this input output matrix to calibrate the input shares _γ_ and _β_ as well as\n\n\nthe trade matrices _Y_ .\n\n\nThe environmental satellite accounts are most recently available for all countries in WIOD\n\n\nonly in 2009, which is why we base all of our analysis on data from that year. We make use\n\n\nof CO2 (carbon dioxide), CO4 (methane) and N2O (nitrous oxide) data, where we convert\n\n\nthe latter two into CO2 equivalents based on their global warming potential. The conversion\n\n\nfactor for methane is hereby 28 and the one for nitrous oxide 265, based on their cumulative\n\n\nforcing over 100 years, taken from IPCC, 2014. For Luxembourg, Romania, Switzerland,\n\n\nCroatia, and Norway and all countries not included in WIOD we rely on the EORA\n\nMention: \"environmental satellite accounts\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Full country coverage is available only for 2009, which fixes the analysis year.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001595:28:2", "occurrence_id": "617e5eaf2483e0eb0522d5d2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: . Indicators were collected across different data collection rounds\nand, due to attrition between rounds, the sample for psychological well-being outcomes is reduced to\nhouseholds surveyed in all rounds in order to have a complete set of indicators. Each psychological\nwell-being measure is estimated as a score on the variable’s scale, normalized to the unit SD of\ncontrol HHs, with mean equal to zero. Impacts on the normalized measures can be found in Table\nA9 of the Online [Appendix.](https://www.dropbox.com/scl/fi/bioyi778lybxemivdq8g4/Long-term-impacts-of-the-TUP-program_OnlineAppendix.pdf?rlkey=2xjb0z4fygh9czw406av7gqlm&dl=0) The psychological well-being index is an index of the three previous\nmeasures, normalized in the same way. School enrolment is estimated for school-age children (ages\n6 to 18 years), with the unit of observation being the child, hence all SEs are clustered at the\nhousehold level. PW = primary woman; SE = standard error; SD = standard deviation; PRA =\nparticipatory rural appraisal; WVS = World Values Survey; CES-D = Center for Epidemiological\nStudies Depression.\n\n\n27\n\nMention: \"World Values Survey\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"The table's abbreviation list expands the survey whose values appear in the notes.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001596:41:0", "occurrence_id": "53bbca7105de4f31d2d104e5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure AF.1: Historical Roads of India, 1872 vs Golden Quadrilateral Roads in 1992**\n\n\nNote: Schwartzenberg roads in red are from 1872. Digital Chart of the World in black depict roads in 1992.\nSource: Schwartzberg Atlas, Growth of Road Network, p. 125, available from the Digital South Asia Library. Digital Chart\nof the World.\n\nMention: \"Digital Chart of the World\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its mapped roads depict the network as it stood in 1992.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001596:41:0", "occurrence_id": "994de64f575328a4c0440422", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure AF.1: Historical Roads of India, 1872 vs Golden Quadrilateral Roads in 1992**\n\n\nNote: Schwartzenberg roads in red are from 1872. Digital Chart of the World in black depict roads in 1992.\nSource: Schwartzberg Atlas, Growth of Road Network, p. 125, available from the Digital South Asia Library. Digital Chart\nof the World.\n\nMention: \"Schwartzberg Atlas\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The historical road network from 1872 is digitised from this atlas plate.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001597:44:1", "occurrence_id": "c630469f0885c27d546c8e26", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: -83\nOptical Equipment 90-92\n\n\n\nPlastic and Rubber 39-40\nPlastic, Rubber and Toys\nToys 95\n\n\nWood Products 44-49\nFurniture and Wood Products\nFurniture 94\n\n\n\nMiscellaneous\n\n\n\nFoodstuffs and Beverage 16-24\nRaw hides and Leather 41-43\nFootwear and Headgear 64-67\nStone and Glass 68-71\n\n\n\n²⁵Please refer to Brandt et al. (2023) for a detailed and comprehensive discussion about the reporting and sample issues\nof both the ASIF data and the NTSD data and a review of the literature using these two datasets.\n\n\n42\n\nMention: \"ASIF data\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"validate\", \"summary\": \"A cited reference documents its reporting and sample problems.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001597:44:1", "occurrence_id": "742d31e8abcf80dbcb572f23", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: -83\nOptical Equipment 90-92\n\n\n\nPlastic and Rubber 39-40\nPlastic, Rubber and Toys\nToys 95\n\n\nWood Products 44-49\nFurniture and Wood Products\nFurniture 94\n\n\n\nMiscellaneous\n\n\n\nFoodstuffs and Beverage 16-24\nRaw hides and Leather 41-43\nFootwear and Headgear 64-67\nStone and Glass 68-71\n\n\n\n²⁵Please refer to Brandt et al. (2023) for a detailed and comprehensive discussion about the reporting and sample issues\nof both the ASIF data and the NTSD data and a review of the literature using these two datasets.\n\n\n42\n\nMention: \"NTSD data\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"validate\", \"summary\": \"Its reporting and sample issues are discussed alongside the other firm dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001598:34:2", "occurrence_id": "508d3548386c541e01a8c25d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  country fixed effects and (world) region by year\nfixed effects. Countries are classified into 7 regions based on World Bank classifications.\nDestination-specific control variables include the logarithms of population and GDP (source:\nConte et al. (2022)) and the average exchange rate over the three years preceding the\ninvestment (national currency/USD, Penn World Tables, mark 10.1, Feenstra et al. (2015)).\nStandard errors are clustered at the destination country level. Significance levels: ***\n_p <_ 0 _._ 01, ** _p <_ 0 _._ 05,  - _p <_ 0 _._ 1.\n\n\n33\n\nMention: \"Penn World Tables\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Three-year average exchange rates against the dollar are taken from this release.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001599:13:0", "occurrence_id": "602c2564adf16b21f1b4297d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: This version: 15 Feb 2024\n\n\n**Figure 2a:** Poverty rates (headcount ratio) by ethnic group, 1970-2019\n\n\n\n12\n\n\n\nSource: Department of Statistics, Malaysia (2020). Notes: Based on old national PLI. Ethnicity\nis based on the identity of the household head. This is an updated version of the figure\nreported in Ravallion (2019).\n\n**Figure 2b:** Mean real household income, national and by ethnic group, 1970-2019\n\n\nSources: Authors’ calculations, based on nominal income from DOSM (2020) and CPI from\nIMF (2024). Notes: Mean monthly gross income data corresponds to households; ethnicity\n\n\n12\n\nMention: \"DOSM\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Nominal household income by ethnic group over five decades is drawn from this office.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001600:1:0", "occurrence_id": "cc280db51beb7bebce47795d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Policy Research Working Paper 10659\n\n### **Abstract**\n\nInnovations to date in detecting women’s employment have\nfocused primarily on improving individual-level questions.\nThis paper explores an alternative approach, using data\n\non household enterprises and asking who participates in\nthese activities. This research uses the latest waves of the\nLabor Market Panel Surveys for the Arab Republic of Egypt\n(2018) and Tunisia (2014). The research questions are (1)\nHow do men’s and women’s employment rates change when\nadding enterprise-based detection questions to standard\nindividual-level questions? (2) Was the additional market\n\n\n\nemployment detected with project-based approaches classified as subsistence work with individual measurement\napproaches? (3) For which women is additional employment detected using project-based approaches? The paper\npresents descriptive results on work based on the different\napproaches. It also estimates changes in state (being reclassified as working) from adding enterprise-level data. The\nfindings show large increases in employment rates for rural\nwomen in both countries when including enterprise-based\ndetection questions.\n\n\n\nThis paper is a product of the Office of the Chief Economist, Middle East and North Africa Region. It is part of a larger\n\neffort by the World Bank to provide open access to its research and make a contribution to development policy discussions\naround the world. Policy Research Working Papers are also posted on the Web at http://www.worldbank.org/prwp. The\nauthors may be contacted at assaad@umn.edu or cgkrafft@stkate.edu.\n\n\n_The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development_\n_issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the_\n_names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are\n\nMention: \"waves of the\nLabor Market Panel Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Latest rounds for Egypt and Tunisia measure who participates in household enterprises.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001602:11:1", "occurrence_id": "c49032afeb945375a9b21506", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Table\n\nA.6 details the common variables we identify in both surveys and their respective coding. <sup>5</sup>\n\n\n4 Though official poverty is measured with the EU-SILC, income from the HBS is officially reported by the Romania Institute\n\nof Statistics, together with consumption, from the HBS (Romania NSI, 2022).\n\n\n5 For the detailed codebook on the EU-SILC see: <u>[https://www.gesis.org/missy/files/documents/EU-SILC/Codebook_EU-](https://www.gesis.org/missy/files/documents/EU-SILC/Codebook_EU-SILC-2019-cross-sec.pdf)</u>\n\n<u>[SILC-2019-cross-sec.pdf](https://www.gesis.org/missy/files/documents/EU-SILC/Codebook_EU-SILC-2019-cross-sec.pdf)</u>\n\n\n10\n\nMention: \"EU-SILC\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"It is the instrument behind official poverty, against which the other survey's income is set.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001602:11:1", "occurrence_id": "1ce66f892f40870422e38595", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Table\n\nA.6 details the common variables we identify in both surveys and their respective coding. <sup>5</sup>\n\n\n4 Though official poverty is measured with the EU-SILC, income from the HBS is officially reported by the Romania Institute\n\nof Statistics, together with consumption, from the HBS (Romania NSI, 2022).\n\n\n5 For the detailed codebook on the EU-SILC see: <u>[https://www.gesis.org/missy/files/documents/EU-SILC/Codebook_EU-](https://www.gesis.org/missy/files/documents/EU-SILC/Codebook_EU-SILC-2019-cross-sec.pdf)</u>\n\n<u>[SILC-2019-cross-sec.pdf](https://www.gesis.org/missy/files/documents/EU-SILC/Codebook_EU-SILC-2019-cross-sec.pdf)</u>\n\n\n10\n\nMention: \"HBS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its income and consumption figures are officially published by the national statistics institute.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001605:4:0", "occurrence_id": "985cd81ef4afba6bedfcb1ce", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: which crop cutting and self-reported yields are both available. This approach allows the authors\n\n\nto replicate the relationship between yields and input obtained with crop-cut yields using the\n\n\npredicted yield showcasing the validity of this imputation framework. However, their exercise\n\n\nis conducted with one crop (sorghum) and only considers one agricultural season limiting the\n\n\nexternal validity of the framework they develop.\n\n\nAgainst this background, our analysis attempts to address these limitations. We take\n\n\nadvantage of the availability of two consecutive rounds of the nationally representative agricul\n\nture survey of Mali collected as part of the Living Standards Measurement Study    - Integrated\n\nSurveys on Agriculture (LSMS-ISA) <sup>1</sup> [^1: provides an overview of data situations as it pertains to crop-cut yield and the potential] project to validate their approach when used in a survey\n\nto-survey imputation framework. Moreover we extend the imputation exercise to more crops:\n\n\nin addition to sorghum, we work with millet, maize, rice, groundnut and cowpea allowing us\n\n\nto generalize the conditions on which the approach is valid. For reasons of computational\n\n\nease and tractability, we focus on the estimation of the mean value of yields. We go beyond\n\n\nwithin-survey imputations exercises to look at survey-to-survey imputations. We also explore\n\n\nwhich set of covariates are more potent for the imputations exercises. Finally we also explore\n\n\nthe effect of training sample sizes and strategies. By varying the size of the training sample,\n\n\nwe investigate whether reducing the sample size can still yield reliable imputation results, with\n\n\nthe aim of balancing cost efficiency and the accuracy of imputation methods.\n\n\nOur main findings are threefold: (i) farmers’ reported yields are good predictors of crop-cut\n\n\nyields but the results point to a greater predictive power of integrated geo-spatial variables; (ii)\n\n\non average the imputation exercises work better for crops that have low intercropping rate, and\nare more commercialized, which can be related to the accuracy in standard units of the farmer\n\nreported yields; (iii) in most of the\n\nMention: \"agricul\n\nture survey of Mali\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Two consecutive nationally representative rounds support the analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001607:27:0", "occurrence_id": "5065621b7b82df86ea5cff48", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **4. Summary and Conclusions**\n\nIn a previous paper (Dasgupta et al. 2023), we developed occurrence maps for more than 600,000\nspecies from georeferenced data provided by the Global Biodiversity Informa�on Facility (GBIF).\nIn this paper, we have combined those maps with ex�nc�on threat factors to produce composite\nglobal indicators of ex�nc�on threat that provide new insights into the spa�al domain for\nconserva�on planning. The paper develops threat factors for terrestrial species that include their\nspecies occurrence region sizes, their sensi�vity to human encroachment, the density of\nneighboring popula�ons, and the degree of formal protec�on. We also develop threat factors for\nmarine species that include species occurrence region size, degree of formal protec�on, degree\nof coverage by EEZs (Exclusive Economic Zones), the impact of coastal popula�on, and two\nmeasures of commercial fishing ac�vity.\n\nThe paper explores three approaches to composite threat indicator construc�on. The first comes\nfrom econometric es�mates of the rela�onship between our threat indicators and expert-based\nRed List ra�ngs from IUCN. Our approach extends numerous pilot exercises in the published\nliterature to an econometric model that can be applied to all mapped species in the IUCN\ndatabase. The model uses our threat indicators to es�mate the probability that a species is\nassigned a threatened status (NT, VU, EN, CR) by IUCN. We find a close sta�s�cal associa�on\nbetween the threat probabili�es predicted by our model and IUCN threat status assignments. We\nalso find a close spa�al associa�on between the two variables. We use our econometric results\nto predict threat probabili�es for mapped GBIF species that have not been rated by IUCN. We\nadd these to the compara�ve mapping\n\nMention: \"IUCN\ndatabase\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Threat indicators are used to estimate the probability a mapped species is classed as threatened.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001610:45:1", "occurrence_id": "4e461d7a985601208cc52164", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  for no more than five years. Their profits are double those of other business types,\ntheir profits per worker, a proxy for labor productivity, is higher and more concentrated than\nthe other firm types. They are mostly in retail trade and operate outsides the household\npremises, operating in less poor neighborhoods. Though using few modern business\npractices, they are more connected to the financial sector than other firm types. In addition\nto a not seeing the benefit of formalization, HE-SE are not registered due to lack of\ninformation on how to register.\n\nEmployers are older firms that managed to grow and create jobs though firm\n\n\n\ninformation on how to register.\n\nEmployers are older firms that managed to grow and create jobs though firm\nproductivity is modest. By definition, the key defining characteristic of employers is that they\nemploy an average of 2.5 paid workers, as compared to no paid workers in other firm types.\nA majority of employers are low-educated adult men (55.5 percent). They are more active in\nservices or manufacturing than are the self-employed. They have the lowest use of good\nbusiness practices, such as not mixing business accounts with household expenses, but\nslighter better use of financial services, including mobile money, than the self-employed.\nThey tend to live in less dense areas that are accessible to other urban neighborhoods. While\nemployers have the highest total revenues and profits, their per capital revenues and profits\nare on par with LE-SE enterprises.\n\nInformality strongly shapes Sub-Saharan Africa’s labor markets. This paper uses data from **9. Conclusions**\n26 harmonized household and labor surveys (GMD) and three Informal Enterprises (IES) to\n\nPage **44** of **60**\n\nsketch a profile of the urban informal sector in Sub-Saharan Africa.\n\n\n\n**9. Conclusions**\n\n\n\nPage **44** of **60**\n\nMention: \"three Informal Enterprises\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Together with harmonised household surveys they profile the urban informal sector.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001614:1:0", "occurrence_id": "0365b583434f22c3f6e88b7a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Policy Research Working Paper 10888\n\n###### **Abstract**\n\nThis paper uses microdata from the Labor Force and\n\nHousehold Surveys conducted in Afghanistan to show the\nwage premium differences for education between men and\nwomen, documenting a significantly larger premium for\nwomen. This sharp distinction is causal as demonstrated by\n\n\n\nanalysis of the compulsory schooling law. Recent bans on\nwomen’s education and employment are projected to have\nsignificant negative impacts on women’s future schooling,\nwage growth, and national income growth.\n\n\n\nThis paper is a product of the Education Global Practice. It is part of a larger effort by the World Bank to provide open access\n\nto its research and make a contribution to development policy discussions around the world. Policy Research Working Papers\nare also posted on the Web at http://www.worldbank.org/prwp. The authors may be contacted at rbkattan@worldbank.org.\n\n\n_The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development_\n_issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the_\n_names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those_\n_of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and_\n_its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent._\n\n\nProduced by the Research Support Team\n\nMention: \"Labor Force and\n\nHousehold Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Microdata are used to estimate returns to education separately for men and women.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001615:11:0", "occurrence_id": "f9f079dd35526d28ade265db", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ### 3 DATA AND CONSTRUCTION OF VARIABLES\n\n##### 3.1 DEPENDENT VARIABLES\n\nWe use the World Bank World Development Indicators (WDI) as a data source for\n\n\nour dependent variables spanning the last four decades and more than a hundred developing\n\n\ncountries. Appendix Table A.1 provides a list of all the countries included in this study and\n\n\ntheir regional classification.\n\n\nFor robustness purposes, we consider five relevant dependent variables (Table 1). We\n\n\nexpect renewable energy policy packages to negatively affect the first three energy mix\n\n\nmeasures in Table 1. In contrast, we expect renewable energy policy packages to positively\n\n\naffect the remaining two energy mix measures in Table 1.\n\n\nTo standardize our estimation results across different specifications, we multiply the\n\n\nestimated coefficients of interest for the first three energy mix measures by minus one to\n\n\nindicate that a positive coefficient leads to impacts aligned with decarbonization. Conversely,\n\n\na negative coefficient denotes impacts that are not aligned with decarbonization.\n\n\nAppendix Figures A.1 and A.2 show the times series of each dependent variable over\n\n\ntime, aggregated over regions to preserve space. Temporal patterns indicate clear differences\n\n\nacross regions in all outcome variables. For example, the share of fossil fuels in the energy\n\n\nmix is consistently higher in relatively oil-dominated regions (e.g., the Middle East and North\n\n\nAfrica) than hydro-dominated regions (Latin America). These patterns reinforce our rationale\n\n\nto estimate the regressions separately across regions.\n\n\n10\n\nMention: \"World Bank World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The dependent variables across four decades and over a hundred countries are taken from these indicators.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001616:17:1", "occurrence_id": "bb1d360acd4cbcd524722912", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  For respondents in a couple in which the woman does not work (Column\n\n2), ten more seconds spent reading the intervention material increases the probability of\n\nthe respondent choosing the career mentoring course by 1.6 percentage points (5.2%).\n\nReading time has no effect for respondents in a couple in which the woman already works\n\n(Column 3).\n\n\n**5.2.2** **Heterogeneity** **Analysis**\n\n\nWe now explore the heterogeneity of treatment effects. One might expect larger treatment\n\neffects in households which are more open to attitudinal change, possibly through greater\n\nexposure to role models of working women, and who have real prospects of being able to\n\n\nsignificant. We also estimated specifications including controls for whether the respondent is supportive\nof women working outside the home for pay. The coefficients are similar to the ones reported here as the\nlevel of support prior to treatment is balanced across control and treatment groups. Results available\nupon request.\n28Weights for (non-)working women were calculated as the number of (non-)working women in the population over the number of (non-)working women in the sample. The population numbers were calculated\nfrom the 2020 National Socioeconomic Survey (SUSENAS).\n29On average people spent 19 seconds reading the information in treatment 1, 26 seconds in treatment 2\nand 35 seconds in treatment 3 (as the format of the information is similar in each treatment, the time\nspent accelerates). No one spent more than 2.5 minutes reading the information. We also estimated a\nspecification with a quadratic term but its coefficient was not significant.\n\n\n15\n\nMention: \"2020 National Socioeconomic Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Population counts of working and non-working women are calculated from this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001616:17:1", "occurrence_id": "2ec8ec6968d6d4e699f7b40f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  For respondents in a couple in which the woman does not work (Column\n\n2), ten more seconds spent reading the intervention material increases the probability of\n\nthe respondent choosing the career mentoring course by 1.6 percentage points (5.2%).\n\nReading time has no effect for respondents in a couple in which the woman already works\n\n(Column 3).\n\n\n**5.2.2** **Heterogeneity** **Analysis**\n\n\nWe now explore the heterogeneity of treatment effects. One might expect larger treatment\n\neffects in households which are more open to attitudinal change, possibly through greater\n\nexposure to role models of working women, and who have real prospects of being able to\n\n\nsignificant. We also estimated specifications including controls for whether the respondent is supportive\nof women working outside the home for pay. The coefficients are similar to the ones reported here as the\nlevel of support prior to treatment is balanced across control and treatment groups. Results available\nupon request.\n28Weights for (non-)working women were calculated as the number of (non-)working women in the population over the number of (non-)working women in the sample. The population numbers were calculated\nfrom the 2020 National Socioeconomic Survey (SUSENAS).\n29On average people spent 19 seconds reading the information in treatment 1, 26 seconds in treatment 2\nand 35 seconds in treatment 3 (as the format of the information is similar in each treatment, the time\nspent accelerates). No one spent more than 2.5 minutes reading the information. We also estimated a\nspecification with a quadratic term but its coefficient was not significant.\n\n\n15\n\nMention: \"SUSENAS\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"The acronym is given for the survey named in the preceding note.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001618:33:1", "occurrence_id": "d376974c55e7d04a72c08688", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 28\nLAC 39.77 25.59 12.98\nMENA 11.03 17.55 5.10\n<u>SAR</u> <u>9.33</u> <u>42</u> <u>4.25</u>\n<u>Source: Constructed by the authors using the WBES.</u>\nNote: SAR stands for South Asia, MENA Middle East and North Africa, EAP East Asia and Pacific, SSA Sub-Saharan\nAfrica, LAC Latin America and the Caribbean and ECA Europe and Central Asia.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nNotes: (i) SAR stands for South Asia, MENA Middle East and North Africa, EAP East Asia and Pacific, SSA SubSaharan Africa, LAC Latin America and the Caribbean and ECA Europe and Central Asia. (ii) GVC1 refers to firms\nthat export and import simultaneously, GVC2 = GVC1 + international certification, GVC3= GVC1+ share of its\ncapital owned by a foreign firm, GVC4 combines the four criteria altogether.\n\n\n\n11 Given that some firms might have one of the variables missing, the sum of production and non-production does\nnot necessarily add to the total number of employees.\n\n32\n\nMention: \"WBES\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Regional shares are constructed from the enterprise survey across regions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001619:11:0", "occurrence_id": "72d03b5827912b82cc9b8c35", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 10\n\n\nfor the months in between these observed postings for which we assume the establishment was\n\n\noperational, but not hiring. In total, 23.5% of all green technology-related postings in our sample\n\n\ncountries in this period are matched to a firm with both CDP energy consumption and Factset\n\n\nsales data.\n\n#### **3. Differential impact of the energy crisis**\n\n###### **3.1 Background on the energy crisis**\n\n\nThe 2022 global energy crisis began in the aftermath of the COVID-19 pandemic, when natu\n\nral gas prices in the EU and Asia started to increase in response to a supply-demand gap and a\n\n\ntight liquefied natural gas (LNG) market. Rising prices subsequently escalated into a widespread\n\n\nglobal energy crisis following the Russian invasion of Ukraine in February 2022. The invasion,\n\n\nand the international sanctions that followed, resulted in a reduction in natural gas imports from\n\n\nRussia, the suspension of the Nord Stream 2 pipeline’s certification and the halting of gas deliver\n\nies through the Nord Stream 1 pipeline. In the first half of 2022, Russia’s gas exports via pipeline\n\n\nto the European Union and the United Kingdom fell by nearly 50 percent compared with the\n\n\nprevious five-year average (EIA, 2022). These supply shortages were exacerbated by uncertainty\n\n\nand concerns of further reductions as well as reduced hydroelectric and nuclear electricity output\n\n\nthat pushed gas consumption for electricity (Tertre, 2023).\n\n\nThis led to soaring natural gas prices in parts of the world dependent on Russian imports and\n\n\nin other countries that competed in the global LNG market. In Europe the benchmark natural\n\n\ngas price peaked in August 2022 at a record high of $69.98 USD per MMBtu, a $42.09 increase\n\n\nfrom the January 2022 price, while the benchmark LNG price in Asia increased by $27.33 over\n\n\nthe same period. Natural gas prices were comparatively stable in countries not reliant on im\n\nports,\n\nMention: \"CDP energy consumption\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Firm energy use is matched to job postings to identify green hiring.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001619:11:0", "occurrence_id": "aac6d6915655085ac1616098", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 10\n\n\nfor the months in between these observed postings for which we assume the establishment was\n\n\noperational, but not hiring. In total, 23.5% of all green technology-related postings in our sample\n\n\ncountries in this period are matched to a firm with both CDP energy consumption and Factset\n\n\nsales data.\n\n#### **3. Differential impact of the energy crisis**\n\n###### **3.1 Background on the energy crisis**\n\n\nThe 2022 global energy crisis began in the aftermath of the COVID-19 pandemic, when natu\n\nral gas prices in the EU and Asia started to increase in response to a supply-demand gap and a\n\n\ntight liquefied natural gas (LNG) market. Rising prices subsequently escalated into a widespread\n\n\nglobal energy crisis following the Russian invasion of Ukraine in February 2022. The invasion,\n\n\nand the international sanctions that followed, resulted in a reduction in natural gas imports from\n\n\nRussia, the suspension of the Nord Stream 2 pipeline’s certification and the halting of gas deliver\n\nies through the Nord Stream 1 pipeline. In the first half of 2022, Russia’s gas exports via pipeline\n\n\nto the European Union and the United Kingdom fell by nearly 50 percent compared with the\n\n\nprevious five-year average (EIA, 2022). These supply shortages were exacerbated by uncertainty\n\n\nand concerns of further reductions as well as reduced hydroelectric and nuclear electricity output\n\n\nthat pushed gas consumption for electricity (Tertre, 2023).\n\n\nThis led to soaring natural gas prices in parts of the world dependent on Russian imports and\n\n\nin other countries that competed in the global LNG market. In Europe the benchmark natural\n\n\ngas price peaked in August 2022 at a record high of $69.98 USD per MMBtu, a $42.09 increase\n\n\nfrom the January 2022 price, while the benchmark LNG price in Asia increased by $27.33 over\n\n\nthe same period. Natural gas prices were comparatively stable in countries not reliant on im\n\nports,\n\nMention: \"Factset\n\n\nsales data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Firm sales from this commercial provider are matched alongside energy use.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001625:10:1", "occurrence_id": "d0e424d98abaff88249a1df4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <br>2013<br>2014<br>2018|2006<br>2008<br>2011<br>2012<br>2013<br>2014<br>2018|2006<br>2008<br>2011<br>2012<br>2013<br>2014<br>2018|2006<br>2008<br>2011<br>2012<br>2013<br>2014<br>2018|2006<br>2008<br>2011<br>2012<br>2013<br>2014<br>2018|2006<br>2008<br>2011<br>2012<br>2013<br>2014<br>2018|2006<br>2008<br>2011<br>2012<br>2013<br>2014<br>2018|\n\n\n\n**Announcement**\n10% CIT if incorporated after 2007\nor incorporated before 2008\n\nand older than 10 years\n\n\n\nEnforcement\n\n\n\nNote: This timeline depicts the subsequent budget law postponements of the enforcement of the Corporate Income Tax (CIT) reform for\n\n\nOffshore firms, thus leaving the full exemption in place until 2013, when the law 2006-80 was finally applied.\n\n#### **3 Data and Descriptive Statistics**\n\n\nThe main source of data we use are administrative records on Tunisian registered firms,\n\n\nthe Repertoire National des Entreprises (RNE). This data spans the universe of all private\n\n\nand public sector firms and includes information such as firms’ 4-digit economic activity\n\n\nsector, location, age and whether owners are foreign. <sup>9</sup> In joint collaboration with the\n\n\nTunisian Statistics Institute (INS), the RNE microdata has been merged with tax returns\n\n\nto include yearly measures of firms’ declared turnover, profits, and revenue from exports;\n\n\nsocial security data on total employment and wage bill; and customs data on total value\n\nMention: \"Repertoire National des Entreprises\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The universe of private and public firms with activity, location, age and ownership information.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001625:10:1", "occurrence_id": "b0f91ac095d0746d65a83588", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <br>2013<br>2014<br>2018|2006<br>2008<br>2011<br>2012<br>2013<br>2014<br>2018|2006<br>2008<br>2011<br>2012<br>2013<br>2014<br>2018|2006<br>2008<br>2011<br>2012<br>2013<br>2014<br>2018|2006<br>2008<br>2011<br>2012<br>2013<br>2014<br>2018|2006<br>2008<br>2011<br>2012<br>2013<br>2014<br>2018|2006<br>2008<br>2011<br>2012<br>2013<br>2014<br>2018|\n\n\n\n**Announcement**\n10% CIT if incorporated after 2007\nor incorporated before 2008\n\nand older than 10 years\n\n\n\nEnforcement\n\n\n\nNote: This timeline depicts the subsequent budget law postponements of the enforcement of the Corporate Income Tax (CIT) reform for\n\n\nOffshore firms, thus leaving the full exemption in place until 2013, when the law 2006-80 was finally applied.\n\n#### **3 Data and Descriptive Statistics**\n\n\nThe main source of data we use are administrative records on Tunisian registered firms,\n\n\nthe Repertoire National des Entreprises (RNE). This data spans the universe of all private\n\n\nand public sector firms and includes information such as firms’ 4-digit economic activity\n\n\nsector, location, age and whether owners are foreign. <sup>9</sup> In joint collaboration with the\n\n\nTunisian Statistics Institute (INS), the RNE microdata has been merged with tax returns\n\n\nto include yearly measures of firms’ declared turnover, profits, and revenue from exports;\n\n\nsocial security data on total employment and wage bill; and customs data on total value\n\nMention: \"Tunisian Statistics Institute\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"It collaborated in merging the firm register microdata with tax returns.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001625:10:1", "occurrence_id": "16fb7c5e4dab9d1c688a660c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <br>2013<br>2014<br>2018|2006<br>2008<br>2011<br>2012<br>2013<br>2014<br>2018|2006<br>2008<br>2011<br>2012<br>2013<br>2014<br>2018|2006<br>2008<br>2011<br>2012<br>2013<br>2014<br>2018|2006<br>2008<br>2011<br>2012<br>2013<br>2014<br>2018|2006<br>2008<br>2011<br>2012<br>2013<br>2014<br>2018|2006<br>2008<br>2011<br>2012<br>2013<br>2014<br>2018|\n\n\n\n**Announcement**\n10% CIT if incorporated after 2007\nor incorporated before 2008\n\nand older than 10 years\n\n\n\nEnforcement\n\n\n\nNote: This timeline depicts the subsequent budget law postponements of the enforcement of the Corporate Income Tax (CIT) reform for\n\n\nOffshore firms, thus leaving the full exemption in place until 2013, when the law 2006-80 was finally applied.\n\n#### **3 Data and Descriptive Statistics**\n\n\nThe main source of data we use are administrative records on Tunisian registered firms,\n\n\nthe Repertoire National des Entreprises (RNE). This data spans the universe of all private\n\n\nand public sector firms and includes information such as firms’ 4-digit economic activity\n\n\nsector, location, age and whether owners are foreign. <sup>9</sup> In joint collaboration with the\n\n\nTunisian Statistics Institute (INS), the RNE microdata has been merged with tax returns\n\n\nto include yearly measures of firms’ declared turnover, profits, and revenue from exports;\n\n\nsocial security data on total employment and wage bill; and customs data on total value\n\nMention: \"social security data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Total employment and the wage bill are added to the merged firm panel.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001627:7:1", "occurrence_id": "a7a91a89d5f1c2edd2040151", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: area**\n\n\nWe collect data on area planted in winter crops from NASA’s Earth Observing System Data and\n\n\nInformation System provided by the SEDAC center at Columbia University. This data provides\n\n\na raster layer that indicates whether a pixel is planted in crops based on the peak Enhanced\n\n\nVegetation Index value detected by the MODIS platform over the period October to March (Jain\n\n\net al., 2017a,b). We aggregate the underlying 1-km pixel-level data to district level to determine\n\n\nthe area of each district planted in winter crops in each year in the sample.\n\n\n**3.3** **Meteorological** **data**\n\n\nWe collect weather re-analysis data from ERA5. ERA5 is a weather re-analysis product produced\n\nby the European Commission’s Copernicus Climate Change Service. <sup>7</sup> We use data from the ERA5\n\nLand hourly product. This provides data at an hourly level on a grid of 0.1 <sup>_◦_</sup> _×_ 0.1 <sup>_◦_</sup>, which translates\n\n\nto a 9km resolution. We use data on temperature and precipitation over the full sample from 2003\n\n\nto 2014. We aggregate these weather variables to the district level by averaging over all grid points\n\n\nthat fall within a district boundary.\n\n\n6Data accessed from: http://cgwb.gov.in/GW-data-access.html\n7Data available here: https://cds.climate.copernicus.eu/cdsapp!/dataset/reanalysis-era5-land?tab=overview\n\n\n6\n\nMention: \"MODIS platform\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Peak vegetation index detections define which pixels are cropped, aggregated to districts.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001627:7:1", "occurrence_id": "c58b17e2225cad695a40925c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: area**\n\n\nWe collect data on area planted in winter crops from NASA’s Earth Observing System Data and\n\n\nInformation System provided by the SEDAC center at Columbia University. This data provides\n\n\na raster layer that indicates whether a pixel is planted in crops based on the peak Enhanced\n\n\nVegetation Index value detected by the MODIS platform over the period October to March (Jain\n\n\net al., 2017a,b). We aggregate the underlying 1-km pixel-level data to district level to determine\n\n\nthe area of each district planted in winter crops in each year in the sample.\n\n\n**3.3** **Meteorological** **data**\n\n\nWe collect weather re-analysis data from ERA5. ERA5 is a weather re-analysis product produced\n\nby the European Commission’s Copernicus Climate Change Service. <sup>7</sup> We use data from the ERA5\n\nLand hourly product. This provides data at an hourly level on a grid of 0.1 <sup>_◦_</sup> _×_ 0.1 <sup>_◦_</sup>, which translates\n\n\nto a 9km resolution. We use data on temperature and precipitation over the full sample from 2003\n\n\nto 2014. We aggregate these weather variables to the district level by averaging over all grid points\n\n\nthat fall within a district boundary.\n\n\n6Data accessed from: http://cgwb.gov.in/GW-data-access.html\n7Data available here: https://cds.climate.copernicus.eu/cdsapp!/dataset/reanalysis-era5-land?tab=overview\n\n\n6\n\nMention: \"ERA5\n\nLand hourly product\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Hourly weather on a 0.1 degree grid provides the climatic controls.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001629:12:0", "occurrence_id": "019c511c5fb3812c1716b490", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: After applying the basic background noise filters, the total luminosity mask contains 76,754 pixels. To\n\nconstruct the mask of civilian nighttime lights, we subtract 1,324 pixels identified as military bases,\n\nwhereas 1,110 pixels come from the OSM-military dataset. The paper results use the sum of all the\n\nlight captured in each pixel of the civilian and military masks, which, in the case of the VIIRS, data, is\n\navailable at monthly intervals and yearly for the VIIRS-like series. This pixel-level processing allows\n\nus to produce civilian and military luminosity series at any level of geographical aggregation.\n\n\n**4.** **Results**\n\n\nOur analysis is divided into three main components. As a first step, we show that civilian night lights\n\ncorrelation with GDP is higher than for total luminosity (including light from military bases).\n\nMoreover, we show that the spatial distribution of civilian and total nighttime lights differs due to the\n\nconcentration of military bases in strategically important areas. Using these two results, we argue\n\ncivilian lights provide a better benchmark for local economic activity, especially after the withdrawal\n\nof the international military presence began in 2014.\n\n\nOur second set of results compares the contraction of luminosity and GDP in 2021 and its evolution\n\nin the following years. We show that most of the reduction in total luminosity would have occurred\n\nin the absence of any economic dislocation since two-thirds of the contraction was driven by the\n\nshutdown of foreign military installations. Tracking the evolution in civilian nighttime light post-2021,\n\nwe find that civilian luminosity by 2023 is above its 2020 levels, a result we interpret as local economic\n\nactivity also displaying a significant recovery. We further validate this result using a synthetic control\n\napproach that benchmarks civilian luminosity in Afghanistan with that of neighboring countries.\n\nFinally, we explore the spatial changes in the distribution of economic activity. Results highlight that\n\nall regions experienced the 2021 shock (and its latter recovery) equally. We conclude by arguing for\n\nthe need to\n\nMention: \"VIIRS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Monthly pixel radiance is summed within civilian and military masks to build lit surfaces.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001631:14:0", "occurrence_id": "41bd50cd72923c5d874b1d98", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: the coefficient is significant at the 1% level. The coefficient is also economically significant as household per capita consumption when households do not receive a positive precipitation\nshock is INR 2,211 (USD 33.5), the coefficient implies that a positive rainfall shock increases\nper capita monthly household consumption by INR 133. As the average household size is 5,\nthis implies an increase of INR 8,000 (USD 121) per household on an annual basis.\n\nColumns (2)-(6) use the occupation classifications in the CP data to identify which households drive the consumption response to positive rainfall shocks. We interact a dummy for\neach occupational classification of interest with the indicator of positive rainfall shock to test\nfor differential responses to positive rainfall shocks. We see that except for households where\nat least one adult is self-employed or engaged in small business, the remaining households do\nnot exhibit any differential consumption response to positive rainfall shocks. As the sum of\nthe coefficients is not statistically significant in columns (4) and (5), it suggests consumption\nin households with individuals engaged as qualified self-employed professionals, or business\nowners, do not respond to precipitation shocks. The sum of the coefficients in the remaining\ncolumns are all statistically significant at the 5% level, confirming an aggregate increase in\nper capita monthly household expenditures for these households.\n\nTable 7 shows the corresponding impact for urban households. We note that while the\ncoefficient on positive rainfall shock is positive, it does not achieve statistical significance in\nany of the specifications. The only category of urban households for which a positive rainfall\nshock disproportionately increases household spending is households with at least 1 member\nemployed in the informal sector. This signifies that positive rainfall shocks in our sample\naffect household consumption in rural areas, but have a muted impact for urban households.\n\nTables 8, 9 and 10 focus solely on rural households and examine the source of increase in\naverage monthly per capita expenditures identified in Table 6. We note that consistent with\nthe evidence shown\n\nMention: \"CP data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Occupational classifications from this panel identify which households drive the consumption response to rainfall.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001620:3:1", "occurrence_id": "9cba123ee735af295614b9d5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: , 2016). In fact, some studies suggest that during the riots\n\nin Gujarat in 2002, women were victims of most extreme forms of sexual cruelty, including\n\nrape, gang rape, stripping, insertion of objects into their body, and molestation (Hameed\n\net al., 2002). This paper examines how women’s marriage age and postmarital socioeconomic\n\nstatus evolved in Gujarat after the 2002 riots. Even though several studies have examined the\n\nimpact of Hindu-Muslim riots on several social, economic and political outcomes of interest,\n\nwe specifically contribute to the literature by assessing the effect of the riots on women’s\n\nmarital outcomes.\n\n\nTo examine the effects of the Gujarat riots of 2002 on women’s marital outcomes, we use\n\na retrospective state-year level dataset from the fourth round of the National Family Health\n\nSurveys (2015-2016) and a difference-in-differences strategy. Our empirical strategy relies\n\non variation over time in marriages that took place before and after 2002 and the state\nwise variation in the occurrence of riots. We also use the event study approach to observe\n\ndifferential trends in women’s age at marriage between Gujarat and the other Indian states.\n\n\nThe difference-in-differences estimates suggest that age at marriage declined for women\n\n\n2\n\nMention: \"National Family Health\n\nSurveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Retrospective state-year marital histories from the fourth round support the difference-in-differences design.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001635:26:0", "occurrence_id": "5046843533076ef42a72777f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: #### Appendix A. Variables: the 2013 census and 2018/19 household survey\n\n**Table A.1. Observed poverty rate and small area estimation predictions at the regional level**\n\n\n**<u>ECVHM 2018/19</u>** **<u>Census 2013</u>**\n**<u>Mean</u>** **<u>Variance</u>** **<u>Mean</u>** **<u>Variance</u>**\nHH age <u>51.58</u> <u>199.17</u> <u>48.28</u> <u>209.47</u>\nHH size 8.93 33.81 8.20 35.36\nHH male 0.72 0.20 0.78 0.17\nDependency ratio 0.47 0.05 0.42 0.06\nNo education 0.66 0.23 0.64 0.23\nPrimary education 0.15 0.13 0.15 0.13\nSecondary education 0.12 0.11 0.15 0.12\nTertiary education 0.07 0.06 0.06 0.06\nRural 0.47 0.25 0.46 0.25\nTV 0.57 0.24 0.53 0.25\nIron 0.04 0.04 0.03 0.03\nCar 0.05 0.05 0.08 0.07\nComputer 0.11 0.10 0.12 0.11\nRefrigerator 0.32 0.22 0.23 0.18\nElectricity (grid) 0.59 0.24 0.58 0.24\nWater access: Yard 0.22 0.17 0.20 0.16\nWater access: Tap 0.45 0.25 0\n\nMention: \"2013 census\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Regional poverty rates observed in the enumeration anchor the small area estimation predictions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001635:26:0", "occurrence_id": "262ef87715efe4087a344625", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: #### Appendix A. Variables: the 2013 census and 2018/19 household survey\n\n**Table A.1. Observed poverty rate and small area estimation predictions at the regional level**\n\n\n**<u>ECVHM 2018/19</u>** **<u>Census 2013</u>**\n**<u>Mean</u>** **<u>Variance</u>** **<u>Mean</u>** **<u>Variance</u>**\nHH age <u>51.58</u> <u>199.17</u> <u>48.28</u> <u>209.47</u>\nHH size 8.93 33.81 8.20 35.36\nHH male 0.72 0.20 0.78 0.17\nDependency ratio 0.47 0.05 0.42 0.06\nNo education 0.66 0.23 0.64 0.23\nPrimary education 0.15 0.13 0.15 0.13\nSecondary education 0.12 0.11 0.15 0.12\nTertiary education 0.07 0.06 0.06 0.06\nRural 0.47 0.25 0.46 0.25\nTV 0.57 0.24 0.53 0.25\nIron 0.04 0.04 0.03 0.03\nCar 0.05 0.05 0.08 0.07\nComputer 0.11 0.10 0.12 0.11\nRefrigerator 0.32 0.22 0.23 0.18\nElectricity (grid) 0.59 0.24 0.58 0.24\nWater access: Yard 0.22 0.17 0.20 0.16\nWater access: Tap 0.45 0.25 0\n\nMention: \"2018/19 household survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its variables supply the predictors for the small area estimates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001636:10:0", "occurrence_id": "ed563072ec58cbcb162cddcf", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: worker. In the following, we describe the identification of deceased workers and incumbent\n\n\nworkers in the administrative data. We also provide evidence that, when comparing peer groups\n\n\nwith the same pre-death racial composition of employees, peer group characteristics do not\n\n\npredict whether a white or non-white worker dies. In addition, we verify that the death of a\n\n\nnon-white worker, compared to the death of a white worker, leads to a lasting drop in the non\n\nwhite share of the peer group. For these reasons, we can treat the race of the deceased worker\n\n\nas a conditionally exogenous and significant shock to the peer group’s racial composition and\n\n\ncan estimate its effect on incumbent workers’ retention at the firm.\n\n\n**3.1** **Sample** **Construction**\n\n\n**Unexpected** **worker** **deaths.** We identify worker deaths based on the employer notifications\n\n\nin the RAIS data. When employers report contracts that were terminated because the employee\n\n\ndied, they distinguish between deaths due to a work accident, deaths while commuting to work,\n\n\nand all other deaths. We exclude work accident and commuting deaths in order to avoid concerns\n\n\nthat safety conditions at and on the way to work may affect both the likelihood of worker deaths\n\n\nand the retention decisions of workers. Moreover, to rule out spurious death notifications, we\n\n\ndrop deaths of individuals for whom we observe an employment contract at least 30 days the\n\n\ndate of death (less than 3% of all deaths).\n\n\nIn order to identify separations that are likely unexpected for coworkers, we focus on deaths\n\n\nof workers younger than 65 who have a full-time (with at least 30 hours) and permanent job.\n\n\nMoreover, we only include individuals with a minimum tenure of three months at the time of\n\n\ndeath, as a large share of contracts are terminated in the first three months when firms can\n\n\ndismiss workers without cost. We also exploit employer reports on workers\n\nMention: \"RAIS data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Employer notifications of contracts terminated by death distinguish accident from other causes.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001642:11:15", "occurrence_id": "0d04e34ca1de1d56f20ee79d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <br>46.0<br>13.4<br>2.7<br>5.5<br>100.0<br>Free of charge<br>51.4<br>13.1<br>17.2<br>10.1<br>2.3<br>100.0<br>48.2<br>12.7<br>16.3<br>13.9<br>2.7<br>100.0<br>49.8<br>12.9<br>16.8<br>12.0<br>2.5<br>100.0|\n\n\nSource: Our elaboration on SHIW data. Data for 2020 and before than 2000 are missing.\n\nMention: \"SHIW data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Sample composition weights are elaborated from this household survey, with no data for 2020 or before 2000.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001643:4:1", "occurrence_id": "ea8fcea57c55f59f49facf33", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  empirical evidence that\n\nsuggests the presence of climate-induced immobility traps.\n\n\nAmong recent works related to this paper, <mark>Martinez Flores, Milusheva and Reichert</mark> (2021)\n\nleverage high-frequency data collected from the International Organization for Migration in 17\n\nWest and Central African countries over the period 2018-2019, and estimate that droughts\n\noccurring during the growing season reduce international migration. As they find these effects only\n\nin middle-income areas (and not in rich nor poor ones), they infer that liquidity constraints and\n\nincome losses are the key channels. Mueller et al. (2020) use data from LSMS-supported\n\nlongitudinal surveys conducted in Ethiopia, Malawi, Tanzania, and Uganda over the period 2009\n2014 and find that climate variability and weather anomalies cause a reduction in temporary urban\n\n\n3\n\nMention: \"LSMS-supported\n\nlongitudinal surveys\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"A cited study uses four countries' panels to estimate the effect of climate variability on employment.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001644:44:0", "occurrence_id": "2caf53f50fef06a6a87d6c1a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Appendix Figure 2\nAssets under management benchmarked to emerging economies bond indexes\n\n\n\n\n\n\n\nNote: The figure shows assets under management, in billions of U.S. dollars, benchmarked to emerging economies bond\nindexes. Sources: J.P. Morgan Markets, and authors’ calculations.\n\n\nAppendix Figure 3\nShare of U.S. dollar-denominated emerging economies sovereign debt\n\n\n\n\n\n\n\nNote: The bars show the U.S. dollar-denominated sovereign debt in the EMBI Global index as a percentage of each\ncountry’s general government debt securities issued in international markets. Averages are derived by calculating\nthis percentage for each country and year, and then averaging these values annually across countries. Each country’s\npercentage is weighted by its debt amount outstanding included in the EMBI Global indexes. Sources: BIS, J.P. Morgan\nMarkets, and authors’ calculations.\n\n\n42\n\nMention: \"EMBI Global index\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Dollar-denominated sovereign debt in the index is expressed as a share of each country's international debt securities.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001647:15:2", "occurrence_id": "3589c54c8e824a7719c4de06", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: /sup> _i_ <sup>_v_</sup>\n\n\n\n_i_ <sup>_loss_</sup> ) and smuggling ( _Ei_ <sup>_loss_</sup>\n\n\n\n_i_ <sup>_loss_</sup> ) activities.\n\n\n\n\n- _Fh,j_ + _Fh,j_ <sup>_loss_</sup>\n\n\n\n\n(30)\n\n\n\n\n\n\n_l_\n\n\n\n_FFh,l_ = \n\n_j_\n\n\n\nNote that firms’ total factor demand is the sum of standard factor demands for\nproduction ( _Fh,j_ ) and factors used for the purpose of tax evasion activities (e.g.\nlabour), denoted _Fh,j_ <sup>_loss_</sup> <sup>.</sup>\n\n\n**6.** **Data**\n\n\n_6.1._ _Economic_ _variables_\n\n\nThe baseline values for macro-economic parameters have been obtained from\nthe GTAP 9 database – in particular, Nigeria’s social accounting matrix (SAM)\nfor the 2011 reference year. These macro-economic parameters are the size of\neconomic sectors (i.e. output), intermediate inputs, capital and labour inputs,\ntaxes, government expenditure, household consumption, imports, exports, and\nthe current account balance. Four sectors are distinguished: (i) the (subsidised)\npetrol sector, (ii) the (unsubsidised) energy sector, which excludes petrol, (iii)\nthe formal (non-energy) sector, and (iv) the informal (non-energy) sector. The\nparameters and coefficients _αi,l_, _βh,j_, _bi_, _γi_, _µi_, _\n\nMention: \"GTAP 9 database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The social accounting matrix for the reference year sets the macroeconomic baseline.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001648:12:0", "occurrence_id": "e9b2221395acba29caabc2c9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: whether they work in the formal sector, their labor income, their occupation and the so\ncioeconomic status of their household. ENAHO collects very similar data from Peruvians.\n\nNeither LAPOP nor Gallup collect detailed data on employment and occupation.\n\n\nWe first show the information for our main explanatory variable.\n\n\nIn the first analysis, this is the informal employment rate measured in the 2017 census\nin the centro poblado in which Venezuelans reside which has a mean of 31.2%. <sup>5</sup> This is\n\nnoticeably below the overall informal employment rate of 59.9% among Peruvians surveyed\n\nin ENAHO, indicating the Venezuelans are generally settling in areas in Peru with less formal\n\nunemployment. Only 8.0% of Venezuelan immigrants surveyed in ENPOVE are employed\n\nin the formal sector (i.e., they have an employment contract with social security benefits),\n\nhence the employment rate in the informal sector among Peruvians in a particular location is\n\na good measure of the availability of job opportunities for Venezuelans and the competition\n\nwith Peruvians for these jobs. We hypothesize that locations with higher informal sector\n\nemployment rates have more opportunities and less competition with Peruvians for jobs. For\n\nthis reason, we sometimes describe locations with high informal employment rates as having\n\n’strong’ informal labor markets.\n\n\nIn the second analysis, this is the number of Venezuelan immigrants in a particular month\n\nand province (measured in the administrative data) as a share of the total local population\nmeasured in the 2017 census. <sup>6</sup> Over the full sample period of our analysis, the mean share of\n\nVenezuelans in the population is very low, 0.4% in the ENAHO sample, 0.3% in the LAPOP\n\nsample and 0.6% in the LAPOP sample. However, if we just look at ENAHO in December\n\n2018, we see that the share has risen\n\nMention: \"2017 census\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Informal employment rates at the settlement level where Venezuelans reside are measured here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001648:12:0", "occurrence_id": "d412c7e4f93c9af28ecb53e5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: whether they work in the formal sector, their labor income, their occupation and the so\ncioeconomic status of their household. ENAHO collects very similar data from Peruvians.\n\nNeither LAPOP nor Gallup collect detailed data on employment and occupation.\n\n\nWe first show the information for our main explanatory variable.\n\n\nIn the first analysis, this is the informal employment rate measured in the 2017 census\nin the centro poblado in which Venezuelans reside which has a mean of 31.2%. <sup>5</sup> This is\n\nnoticeably below the overall informal employment rate of 59.9% among Peruvians surveyed\n\nin ENAHO, indicating the Venezuelans are generally settling in areas in Peru with less formal\n\nunemployment. Only 8.0% of Venezuelan immigrants surveyed in ENPOVE are employed\n\nin the formal sector (i.e., they have an employment contract with social security benefits),\n\nhence the employment rate in the informal sector among Peruvians in a particular location is\n\na good measure of the availability of job opportunities for Venezuelans and the competition\n\nwith Peruvians for these jobs. We hypothesize that locations with higher informal sector\n\nemployment rates have more opportunities and less competition with Peruvians for jobs. For\n\nthis reason, we sometimes describe locations with high informal employment rates as having\n\n’strong’ informal labor markets.\n\n\nIn the second analysis, this is the number of Venezuelan immigrants in a particular month\n\nand province (measured in the administrative data) as a share of the total local population\nmeasured in the 2017 census. <sup>6</sup> Over the full sample period of our analysis, the mean share of\n\nVenezuelans in the population is very low, 0.4% in the ENAHO sample, 0.3% in the LAPOP\n\nsample and 0.6% in the LAPOP sample. However, if we just look at ENAHO in December\n\n2018, we see that the share has risen\n\nMention: \"ENAHO\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The comparable rate for Peruvians provides the reference point.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001648:12:0", "occurrence_id": "f07fe58a13d92a831fc8ba75", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: whether they work in the formal sector, their labor income, their occupation and the so\ncioeconomic status of their household. ENAHO collects very similar data from Peruvians.\n\nNeither LAPOP nor Gallup collect detailed data on employment and occupation.\n\n\nWe first show the information for our main explanatory variable.\n\n\nIn the first analysis, this is the informal employment rate measured in the 2017 census\nin the centro poblado in which Venezuelans reside which has a mean of 31.2%. <sup>5</sup> This is\n\nnoticeably below the overall informal employment rate of 59.9% among Peruvians surveyed\n\nin ENAHO, indicating the Venezuelans are generally settling in areas in Peru with less formal\n\nunemployment. Only 8.0% of Venezuelan immigrants surveyed in ENPOVE are employed\n\nin the formal sector (i.e., they have an employment contract with social security benefits),\n\nhence the employment rate in the informal sector among Peruvians in a particular location is\n\na good measure of the availability of job opportunities for Venezuelans and the competition\n\nwith Peruvians for these jobs. We hypothesize that locations with higher informal sector\n\nemployment rates have more opportunities and less competition with Peruvians for jobs. For\n\nthis reason, we sometimes describe locations with high informal employment rates as having\n\n’strong’ informal labor markets.\n\n\nIn the second analysis, this is the number of Venezuelan immigrants in a particular month\n\nand province (measured in the administrative data) as a share of the total local population\nmeasured in the 2017 census. <sup>6</sup> Over the full sample period of our analysis, the mean share of\n\nVenezuelans in the population is very low, 0.4% in the ENAHO sample, 0.3% in the LAPOP\n\nsample and 0.6% in the LAPOP sample. However, if we just look at ENAHO in December\n\n2018, we see that the share has risen\n\nMention: \"ENPOVE\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Only eight percent of surveyed Venezuelan immigrants hold formal contracts.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001652:18:1", "occurrence_id": "e1dc91e916c57049bb323cc9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: /sup> and finally from pensions (whether private or public). The _h_ subscript indexes a set of\nhouseholds (but equally could index individuals).\n\n\nTo note is that government policies may appear to have greater or lesser impacts depending on what is\nincluded in the variable describing the prefiscal income state. Egypt provides a good test case as rental\nmarket regulations create two different—and plausible—valuations of the value of housing services\nconsumed; therefore two different distributions of prefiscal income over which fiscal policy can then be\nallocated; and finally two different distributions of postfiscal income (at each postfiscal income concept).\n\n\nNext, the set of taxes and transfers _Ti_ to be examined is defined: for example, _Ti_ in Egypt might include\nthe sales tax and the Takaful and Karama social assistance programs. For each household _h_ found in the\nmicrodata _shares_ ( _Sih_ ) of each program, _i_ = 1,…. _I_ in _Ti_ are allocated.\n\n\nWith the estimated shares, an estimate of _postfiscal income_ is created at the household level _Yh_ such that\n\n\n𝑌� �𝐼� �Σ�𝑇�𝑆�� (1)\n\n\n20 A CEQ assessment typically includes two scenarios for contributions to, and income received from, the public\ncontributory pension system. The first, which we call the Pensions as Deferred Income (PDI) scenario, treats\ncontributions to and income from the public contributory pension system as compulsory savings and deferred\nincome (respectively). The alternative, which we call the Pensions as Government Transfer (PGT) scenario, treats\ncontributions and income as a tax and a transfer (from the fisc) respectively. We could not identify public\ncontributory pension income received in the HIECS 2015 nor did we have access to budget or administrative\nreporting that summarized total public contributory pension income distributed, so we did not estimate the PGT\nscenario.\n\n\n17\n\nMention: \"HIECS 2015\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Public contributory pension income cannot be identified in it, so no estimate is produced.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001657:7:0", "occurrence_id": "88ef884b104922ade27ac8bf", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: on rice farming. Since HIES 2014 does not include a comparable agricultural module, we\n\n\ncannot construct a 2014-16 panel of EAs for the agricultural production data along the lines\n\n\nof our consumption expenditure data. Instead, our analysis of the impact of the Ebola\n\n\noutbreak on farm labor and production outcomes is purely cross-sectional.\n\n\n2.2 Mortality during the Ebola Outbreak\n\nOn 30 March 2014, Liberia confirmed its first two cases of Ebola virus disease in Foya,\n\n\nLofa County. By June the first deaths from Ebola occurred in Monrovia when seven\n\n\npatients died from the disease. Although, at the time, only 16 cases were reported in Liberia\n\n\nin total, with the urbanization of the disease the outbreak quickly spiraled out of control,\n\n\npeaking in the first quarter of 2015. By June 2016, when Liberia was (provisionally)\n\n\ndeclared Ebola-free, 10,678 people had been infected with Ebola, and over 4,800 had died.\n\n\nEbola-related burial data obtained from the Ministry of Health up to December\n\n\n2015 indicate 4,621 total casualties, corresponding reasonably closely to the official\n\n\nfigures.  Of these deaths, 2,221 were matched to rural areas. <sup>8</sup> In rural Liberia, adult males\n\n\nwere the main group of mortal victims from Ebola, but close to a third of the fatalities were\n\n\nunder-fives (8%) and adults over 65 (24%) (See table A1.2 in Appendix 1).\n\n\n8 The rural classification was done for georeferenced burials (4257/4621) at the locality level using the\nofficial cutoff population of < 2,000 as used in the HIES.\n\n\n6\n\nMention: \"HIES 2014\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"It lacks a comparable agricultural module, so an agricultural production panel cannot be built.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001657:7:0", "occurrence_id": "cae5ea163cad401c2f4b35a0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: on rice farming. Since HIES 2014 does not include a comparable agricultural module, we\n\n\ncannot construct a 2014-16 panel of EAs for the agricultural production data along the lines\n\n\nof our consumption expenditure data. Instead, our analysis of the impact of the Ebola\n\n\noutbreak on farm labor and production outcomes is purely cross-sectional.\n\n\n2.2 Mortality during the Ebola Outbreak\n\nOn 30 March 2014, Liberia confirmed its first two cases of Ebola virus disease in Foya,\n\n\nLofa County. By June the first deaths from Ebola occurred in Monrovia when seven\n\n\npatients died from the disease. Although, at the time, only 16 cases were reported in Liberia\n\n\nin total, with the urbanization of the disease the outbreak quickly spiraled out of control,\n\n\npeaking in the first quarter of 2015. By June 2016, when Liberia was (provisionally)\n\n\ndeclared Ebola-free, 10,678 people had been infected with Ebola, and over 4,800 had died.\n\n\nEbola-related burial data obtained from the Ministry of Health up to December\n\n\n2015 indicate 4,621 total casualties, corresponding reasonably closely to the official\n\n\nfigures.  Of these deaths, 2,221 were matched to rural areas. <sup>8</sup> In rural Liberia, adult males\n\n\nwere the main group of mortal victims from Ebola, but close to a third of the fatalities were\n\n\nunder-fives (8%) and adults over 65 (24%) (See table A1.2 in Appendix 1).\n\n\n8 The rural classification was done for georeferenced burials (4257/4621) at the locality level using the\nofficial cutoff population of < 2,000 as used in the HIES.\n\n\n6\n\nMention: \"HIES\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its official population cutoff below two thousand defines the urban-rural classification used for burials.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001658:5:2", "occurrence_id": "fdbba438c4a3bf0c22198b30", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \n_Ménages_ (henceforth referred to as EBCNV 2015). The objective of this analysis is three-fold: first, to\nassess the pre-COVID situation in order to identify and profile the subgroups of population that are most\nvulnerable to getting infected by COVID-19 and impacted by the associated government measures;\nsecond, to simulate the impacts of COVID-19 on consumption, poverty and inequality based on sectoral\ngrowth performances over the whole year of 2020 and analyze the distribution of individuals expected to\nfall into poverty due to COVID-19; and finally, to simulate how the impacts would vary with or without\nmitigation measures adopted by the Tunisian government. <sup>3</sup>\n\n\n2 Refer to Ajwad, Aran, Azam and Hentschel (2013) for a detailed review.\n3 In terms of scope, we seek to estimate the impact of COVID and not the determinants of contamination in Tunisia by COVID-19, which remains\nout of scope.\n\n4\n\nMention: \"EBCNV 2015\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"This pre-pandemic household survey is used to profile subgroups most vulnerable to infection.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001659:11:0", "occurrence_id": "d5ab9e4b797c1f09c5827d05", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 4** : Pre-program trends in project areas (treatment) and other areas (control)\n\n\n_Regular blackouts_ _Use electricity for lighting_\n\n\n_Use electricity for cooking_\n\n\n_Notes:_ The vertical (red) line indicates the start of the program (2015). The values of “Amount paid for electricity”, and\n“Frequent blackouts” are based on the subset of households that have electricity in their dwelling.\n_Source:_ Authors’ analysis using data from the Living Standards Measurement Study (LSMS).\n\n\n9\n\nMention: \"the Living Standards Measurement Study\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Households with electricity in their dwelling are used to measure blackout frequency.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001668:14:4", "occurrence_id": "a06cf71f35a45136ae639baa", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: >obstacle</u> <u>(%)</u> <u>-2.079</u> <u>-1.976</u> <u>1.031</u> <u>-5.105</u> <u>0</u> <u>10,835</u>\n\nNotes: Authors’ calculations based on WBES, CEPII, Jeffrey Bergtrand’s website, WITS, Doing\nBusiness and LPI. Descriptive statistics computed from the sample of Column 1 of Table 3.\nVariables from WBES calculated over the entire population of firms (Panel A) or the sample of\nexporting firms only (Panel B). The description and source of variables are provided in Table\nA3 in the Appendix.\n\n\n13\n\nMention: \"WBES\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Firm-level obstacle responses enter regressions combined with several external datasets.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001671:72:0", "occurrence_id": "e146c9ef2841e37554cdbcc7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: We use the PILA data to construct for each firm the long difference between their mean\nemployment in 2013, and their mean employment in 2017. Figure A8.2 then shows the quantile\ntreatment effects for this change in employment at different quantiles. The left panel shows the\nfull range of quantiles. The confidence intervals are incredibly wide at the bottom quantile,\nreflecting the effect of the long left tail seen in Figure 4. The right panel zooms in to the subset of\nquantiles from 20 to 90. We see the group treatment has positive quantile treatment effects at all\nquantiles, which are statistically significant at the 10 percent level between the 60 <sup>th</sup> and 70 <sup>th</sup>\npercentiles.\n\n\n**Figure A8.2 Quantile Treatment Effects on Long Difference in Formal Employment 2013-17**\n\n\nNote: Formal employment data taken from the PILA.90 percent pointwise confidence intervals shown from crosssectional estimation of the quantile treatment effect on the long difference in employment.\n\n\n71\n\nMention: \"PILA data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Mean employment in each year is differenced across firms to measure formal job changes.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001671:72:0", "occurrence_id": "6c8aabfe240b16c5e3c4030d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: We use the PILA data to construct for each firm the long difference between their mean\nemployment in 2013, and their mean employment in 2017. Figure A8.2 then shows the quantile\ntreatment effects for this change in employment at different quantiles. The left panel shows the\nfull range of quantiles. The confidence intervals are incredibly wide at the bottom quantile,\nreflecting the effect of the long left tail seen in Figure 4. The right panel zooms in to the subset of\nquantiles from 20 to 90. We see the group treatment has positive quantile treatment effects at all\nquantiles, which are statistically significant at the 10 percent level between the 60 <sup>th</sup> and 70 <sup>th</sup>\npercentiles.\n\n\n**Figure A8.2 Quantile Treatment Effects on Long Difference in Formal Employment 2013-17**\n\n\nNote: Formal employment data taken from the PILA.90 percent pointwise confidence intervals shown from crosssectional estimation of the quantile treatment effect on the long difference in employment.\n\n\n71\n\nMention: \"PILA\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Formal employment underlying the quantile treatment effect figure is taken from this register.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001676:16:0", "occurrence_id": "e0a4434219b47f9fd97bf8e3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ### IV.  Earnings and decision-making roles\n\n###### Complementing missing data on earnings\n\nIn surveys, earnings are measured through (a) individual wage/salary earnings; (b) profit from the farm\n(measured at the household or plot level); and (c) profit from the nonfarm enterprise (measured at the\nenterprise level, which if own-account could be assigned to an individual household member). The quality\nof rural earnings data can vary substantially across country surveys, often resulting in missing and outlier\nvalues for occupations where rural women also tend to be overrepresented. In wage employment, for\nexample, this includes informal and seasonal jobs that can be paid out irregularly and/or in-kind; and in\nself-employment, in family farms/businesses where the value of household members’ labor is not recorded\n(if, as is common in rural areas, multiple family members are involved).\n\nFor individual wage/salary earnings, carefully collected community-level data on average wages or earnings\nfor men and women, for formal/casual work across different sectors, could be useful as complementary\ndata to understand expected labor market returns — particularly in contexts when many women are out\nof the labor force. For self-employment, within farming, several recent studies have discussed the use of\nindividual-level data on productivity that is associated with farm income (see, for example, studies from\nSub-Saharan Africa including Doss, 2018; Kilic et al, 2015). And as discussed in more detail below, for cases\nwhere income is pooled (across individual earnings, and farm/non-farm enterprises) the LSMS and other\nhousehold surveys are exploring alternative questions about decision making over earnings to better\nunderstand individuals’ control over income. A joint pilot study in Sri Lanka by the World Bank and ILO\nunder the WWEP, to coordinate survey questions across the LFS and LSMS, will also test questions on\ndecision\n\nMention: \"LSMS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"It is exploring new questions on decision making over pooled earnings.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001682:2:0", "occurrence_id": "5fa5982879dfb88b33ec321f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ###### **Incentives and Teacher Effort: Further Evidence from a Developing Country**\n\nHai-Anh H. Dang and Elizabeth M. King <sup>*</sup>\n\n\nJEL: C3, I2, O1\nKeywords: teacher incentives, teacher effort, private tutoring, joint probit-OLS, random effects,\nLao PDR\nSector Board: EDU, POV\n\n\n[* Dang (hdang@worldbank.org) and King (eking@worldbank.org) are economists respectively at the Poverty and](mailto:hdang@worldbank.org)\nInequality Unit, Development Research Group and Education Sector, Human Development Network at the World\nBank, Washington, DC. We thank our colleagues Keiko Miwa, Dominique van de Walle and Jeffrey Waite for their\ncontributions to the design and application of the school survey in Lao PDR and for helping us interpret some of our\nresults; and Harold Alderman, Eric Gronqvist, Yang Huang, Dean Jolliffe, Phil Keefer, Xubei Luo, Salvador\nNavarro, Ha Nguyen, Daniel O’Brien, Forhad Shilpi, Trang Tran, Futoshi Yamauchi, seminar participants at the\nNorth American Econometric Society Winter Meeting (San Francisco), and International Food Policy Research\nInstitute for useful comments on earlier drafts. We also thank the National Statistical Center, Lao PDR for\npermission to use the data. The findings, interpretations, and conclusions expressed in this paper are entirely those of\nthe author and should not be attributed to the World Bank, its Executive Directors or the countries they represent.\n\nMention: \"school survey in Lao PDR\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"Colleagues are credited with designing and applying this survey and interpreting results.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001683:40:3", "occurrence_id": "991302d79e06e2f1b18f869c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ity** **Observations** 4,163 4,160 4,256 4,256 4,256 4,163\n**Adj.** **R** _−_ **Squared** 0.702 0.223 0.065 0.13 0.078 0.381\n\n\n**Notes.** _∗∗∗_ denotes significance at 1 percent, _∗∗_ at 5 percent, and _∗_ at 10 percent level. The dependent variables come from National\nTreasury Bureau ( _Secretaria_ _do_ _Tesouro_ _Nacional_ (STN). All regressions restrict the sample to the 330 municipalities at the border\nbetween Ceará and adjacent states. The definition of border considers the three imediate municipalities at each side of _Ceará_ and\n_Piaúi_, _Rio_ _Grande_ _do_ _Norte_, _Pernambuco_ and _Paraíba_ . municipality controls include the number of births in the municipality _−_ year\naccording to the _Sistema_ _de_ _Nascidos_ _Vivos_ (SINASC), and GDP per capita as provided by IBGE. The vector of dependent variables\nincludes: (1) Natural log of education spending per student; (2) Natural log of health spending per capita; (3) Natural log of spending\non sanitation, housing and urbanism per capita; (4) Natural log of spending on social assistance and social security per capita; (5)\nNatural log of spending on sport, leisure and culture per capita; (6) Natural log of total spending per capita. Public spending\nvariables are deflated by the General Price Index (2017 = 100) of the _Fundação_ _Getúlio_ _Vargas_ . Time trends and Municipal specific\ntime trends are included. To\n\nMention: \"General Price Index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Public spending variables are deflated by this index before estimation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001689:10:0", "occurrence_id": "0742a4723e75a5c33a2855da", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: produce correlations by construction with macroeconomic variables, in this paper, we work\n\nonly with the non-imputed ILO data. Table A1 in the Appendix contains the descriptive\n\n\n\nproduce correlations by construction with macroeconomic variables, in this paper, we work\n\nonly with the non-imputed ILO data. Table A1 in the Appendix contains the descriptive\n\nstatistics of the resulting data sets, including for the unemployment variable.\n\n\n\nstatistics of the resulting data sets, including for the unemployment variable. _B._ _Proxies for the Incidence of the Digital Economy_\n\n\n\n_B._ _Proxies for the Incidence of the Digital Economy_\n\n\n\n_B._ _Proxies for the Incidence of the Digital Economy_\n\nTo study the correlation between the digital economy and unemployment, we rely on\n\ntwo alternative proxies for the incidence of the digital economy. The first, motivated by Hjort\n\n\n\nTo study the correlation between the digital economy and unemployment, we rely on\n\ntwo alternative proxies for the incidence of the digital economy. The first, motivated by Hjort\n\nand Paulsen (2019), is internet users as a percentage of the population. It is defined as the\n\n\n\ntwo alternative proxies for the incidence of the digital economy. The first, motivated by Hjort\n\nand Paulsen (2019), is internet users as a percentage of the population. It is defined as the\n\nproportion of individuals using the internet, based on results from national household\n\n\n\nand Paulsen (2019), is internet users as a percentage of the population. It is defined as the\n\nproportion of individuals using the internet, based on results from national household\n\nsurveys implemented by national authorities and then reported back to the United Nations.\n\n\n\nproportion of individuals using the internet, based on results from national household\n\nsurveys implemented by national authorities and then reported back to the United Nations.\n\nThe survey questionnaires were designed by experts under the aegis of the United Nations.\n\n\n\nsurveys implemented by national authorities and then reported back to the United Nations.\n\nMention: \"ILO data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"Imputed values are excluded because they induce correlations with macroeconomic variables.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001697:34:1", "occurrence_id": "9e1b74ed88fc6a6e3c1b370c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: , according to the tax rules, such tax is not\nrecovered by the sector. Making this calculation for all other costs and expenses of the sector, we\nconclude that 4.41% of the price charged by the sector refers to the “hidden taxes”. The Table A6\ndemonstrates the effect of this reduction on the supermarket PL.\n\n**Table A6: Estimate of the hidden taxes for the supermarket sector**\n\n\n\n\n\n\n\nGross Revenue 441,252,542\nTax on Sales (24,212,363)\nOther Deductions (11,787,702)\nNet Revenue 405,252,477\n% of Local Operation 94%\nTotal Costs (361,525,213) (361,259,573) (356,151,921)\nCost of Goods Sold (320,583,293) (320,583,293) 0.00% (315,531,024) 1.58% Industry\nTotal Expenses (33,697,850) (30,242,377) (28,821,770)\n<u>Marketing</u> <u>(2,390,173)</u> <u>(2,226,207)</u> <u>6.86%</u> <u>(2,120,815)</u> <u>4.41%</u> <u>Marketing</u>\nEbitda 10,029,414 (384,973,691)\nEbitda %\n(onNetRevenue) 2.47% 2.47%\n\n\n\n\n\n\n\nSource: Adaptation from IBGE tables.\n\n\n\n33\n\nMention: \"IBGE tables\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Firm financial aggregates such as EBITDA and net revenue are adapted from these tables.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001663:5:0", "occurrence_id": "0bac665cc882a7fdd08ff89d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Asia from 2000 to 2003, the AICD Database applied the same methodology for Sub-Saharan Africa\nand collected data from 2004 to 2009.\n\n**(b) Have energy subsidy reforms succeeded in restraining energy consumption?**\nTo provide an answer to this question the study uses both indicators of energy prices and related\nconsumption. In the case of gasoline and diesel, road sector consumption of fuels is reported, as\nthe transport sector is one of the most intensive users of such fuels. In the case of electricity power\nconsumption is provided:\n\n **Fossil Fuel Prices** (US$ per liter): Gasoline, diesel and kerosene retail tariff (US$ per liter).\n\n **Electricity Prices** (US$ cent per kWh) – Residential Average Electricity Tariff level (US$ cent per\n\nkWh)\n\n **Road Fuel Consumption** (kt of oil equivalent per capita): Gasoline and Diesel Fuel Consumption (kt\n\nof oil equivalent per capita) by the road sector\n\n **Power Consumption** (kWh per capita) - Electric power consumption measures the production of\n\npower plants and combined heat and power plants less transmission, distribution, and\ntransformation losses and own use by heat and power plants. (kWh per capita)\n\n\n**(c) Have energy subsidy reforms managed to avoid adverse social impacts?**\n\nTo address this final question, evidence from household surveys on the distributional impact of\nsubsidies by income quintile (or deciles, if available) as well as the materiality of energy subsidies,\nreporting for each fuel how much income is spent by each quintile of the population is used in the\ncase studies.\n\n\n **Expenditure of different energy sources** (what proportion of income poor households spend on\n\neach fuel), by quintiles or deciles where available divided by rural/urban areas.\n\n **Welfare impact of removing energy subsidies, by fuels\n\nMention: \"AICD Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"It extends the Asian methodology to African countries over a later period.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001690:15:1", "occurrence_id": "a2933990493335bea91a6b76", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: >\n\nSource: Susenas, 2017.\n\nThe same data source also shows that economic status is positively correlated with length of schooling.\nThis applies for both males and females (Table 4). The higher gender gap in years of schooling, while\ninsignificant, is faced by girls from the poorest (0.67 years) and the richest (0.66 years) quintiles. This\ngender gap includes those 15 years and older, and so captures the structure of the education system many\ndecades ago which at that time was biased in favor of males. This gap is expected to decline and then\nreverse as children who have attended schools in the past two decades become adults and replace those\nwho were educated under the former system.\n\n**Table 4. Average years of schooling for population aged 15 and older by economic status and sex**\n\n\n<u>Quintile Economic Status</u> <u>Male</u> <u>Female</u> <u>Gender gap</u>\n<u>Quintile 1</u> <u>6.96</u> <u>6.29</u> <u>0.67</u>\nQuintile 2 7.77 7.21 0.56\nQuintile 3 8.39 7.80 0.59\nQuintile 4 9.19 8.57 0.62\n<u>Quintile 5</u> <u>11.30</u> <u>10.64</u> <u>0.66</u>\n\nSource: MoWECP and BPS (2018) Profil Perempuan Indonesia.\n\nOne of the main challenges in addressing the issue of out-of-school children (OOSC) is a lack of accurate\ndata that could specifically identify who the OOSC are, where they live and why they do not go to school.\nSince 2016, UNICEF\n\nMention: \"Profil Perempuan Indonesia\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Years of schooling by wealth quintile are tabulated in this publication.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001698:11:1", "occurrence_id": "730209fe84b39b069e4b2735", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: b) NFE 0.22 0.22 93.6 89.0 426.9 410.6\n\n\n(2c) Agriculture 0.36 0.47 70.4*** 124.9*** 194.2*** 263.2***\n\n\n(3) Leisure 0.96 0.97 151.2*** 171.7*** 156.8*** 176.2***\n\n\n(4) Traveling 0.57 0.67 41.6*** 57.7*** 73.1*** 86.4***\n\n\n(5) School 0.12 0.04 9.6** 5.1** 82.0 114.9\n\n\n(6) Sleep 1.00 1.00 658.0 658.1 658.3 658.1\n\n\n(7) Eating/drinking 1.00 1.00 75.1 75.7 75.9 76.3\n\n\n(8) Other 0.10 0.10 19.4 19.9 196.2 205.7\n\n\nNumber of respondents 1,977 1,648 1,977 1,648\n\n\n<u>Notes:</u>\n(1) Sample includes men and women self-reporting in the LSMS+ time diary module, aged 18 and older. All estimates weighted\nby household sampling weight.\n(2) Significant differences in minutes spent across men and women (separately for full sample and among t>0) are indicated by\nasterisks; ***p<0.01 **p<0.05 *p<0.10\n(3) As seen in Table 1, the time use module also included categories for (a) making goods (furniture, pottery, baskets, clothing), as\nwell as (b) work for other household free of charge as an exchange laborer, but nearly no respondent reported time in\n\nMention: \"LSMS+ time diary module\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Self-reported minutes for adults are weighted by household sampling weights to compare men and women.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001699:3:1", "occurrence_id": "47fa7380dd62bfc586f92b9c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: , Vietnam). I use harmonized\n\n\ndata from the World Bank Enterprise Survey, which enables me to construct aggregates\n\n\nfigures for sales for worker (a measure of productivity), total labor costs, capital stock,\n\n\nfirm size, and a somewhat noisy measure of value added. I compute these figures for\n\n\nthe whole manufacturing sector and for the garment sector. Further, I use data from\n\n\nthe World Management Survey to study the quality of management.\n\n\nMy main finding is that Ethiopia’s labor cost advantage is more than offset by low\n\n\nproductivity with respect to all but one of the comparison countries (Bangladesh). Ethiopia’s\n\n\nindustrialization is thus on a knife’s edge. Its fundamentals make it competitive with\n\n\nrespect to the countries at the bottom of the productivity distribution. In recent years,\n\n\nthe increase in labour costs has also been modest compared to these countries. How\n\never, given the likely pressure on wages that will result from further expansion of the\n\n\nmanufacturing sector, additional gains in productivity will be crucial in order to secure\n\n\nEthiopia’s position as an attractive investment destination.\n\n\nThe low productivity of Ethiopian firms is not explained by differences in size, cap\n\n2\n\nMention: \"World Management Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It supplies the management quality measure used alongside productivity.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001702:11:0", "occurrence_id": "ca5f4f8c377648e7404d3d59", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Note: The mobility index in an average of the percentage change in mobility to retail and recreation locations, transit stations,\nparks, and workplaces. The 7-day rolling average is shown. The change is calculated relative to a pre-pandemic baseline. Grey\nareas indicate when each round of the HFPS was fielded in the specified country. Indonesia had five rounds, but only the first,\nthird, and fifth included information on employment and income. Source: Google Mobility Reports; HFPS\n\n###### 2. Work stoppages: Employment impacts of the pandemic in developing EAP countries\n\nThe HFPS provides insight into the employment impacts of the pandemic at varying points during 20202021 in each country. In particular, we look at employment impacts at the extensive margin – namely, work\nstoppages <sup>14</sup> - rather than at the intensive margin due to limitations in the data. The HFPS suggests that the\ninitial impact of the pandemic on work stoppages was significant in most countries <sup>15</sup> but particularly in\n\n\n14 As described in the previous section, work stoppages are defined as the share of respondents working pre-pandemic who are\nnot working in the week preceding the survey. A worker may be not working even though she is still employed.\n15 The initial impact of the pandemic on work stoppages may even be underestimated, given that most round one surveys\noccurred after the steep drop in mobility between March and May 2020 (see Figure 1).\n\n\n10\n\nMention: \"Google Mobility Reports\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Mobility indices accompany the household panel in describing pandemic restrictions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001703:4:1", "occurrence_id": "5216c43d9c99969f51c8efba", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  effects of a persistent\ninflation shock that raises inflation to 6 percent for the next five years on the public debt-to-GDP\nratio is simulated using the IMF’s _World Economic Outlook_ (WEO) data for G7 countries.\nCompared to them, we use updated WEO data and expand sample countries. We also take account\nof the debt rollover ratio more explicitly and examine the effects of a temporary inflation shock as\nwell as a persistent shock. Hilscher et al. (2017) and Equiza-Goñi (2016) also simulate the effects\nof inflation on public debt-to-GDP ratio based on the debt dynamics equation, but they use more\ndetailed government bonds data (the former also uses option price data to estimate the distribution\nof inflation risks) focusing on the US and a few euro area countries, respectively. Aizenman and\nMarion (2011) simply use the US debt maturity in 2009 to approximate the effects of persistently\nhigher inflation. End et al. (2015) uses a model of fiscal account dynamics for the euro area to\nsimulate the effects of persistent inflation/disinflation shocks. Krause and Moyen (2016) uses a\nNew Keynesian dynamic general equilibrium model with an imperfectly observed inflation target\nand simulate the effects of shocks to the inflation target. While our simulation approach is different\n\n\n3 In some countries (most notably Japan but also euro area and the U.S.), central banks hold substantial amount of\nlong-term government debt, and thus the debt-reducing effect of an inflation shock is partially offset by the loss of\ncentral banks. We assume that the loss of central banks caused by an inflation shock in our analysis is not so substantial\nto affect their credibility and monetary policy frameworks.\n\n\n3\n\nMention: \"updated WEO data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"A more recent vintage and wider country set than the cited study are used to simulate debt effects.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001715:9:0", "occurrence_id": "24f9f64a0f145b044ff8e1f9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 3.1 Financial Banking Data\n\n\nThe essential ingredient for this project is a micro data set that we built based upon the\n\nBasic Statistical Return (BSR)1A, maintained by the Central Bank (RBI). BSR-1A has\n\ndetails of each loan outstanding (above a threshold), reported annually by every branch of\n\nevery scheduled commercial bank in India. The threshold over which individual account\n\ndata are reported was Rs. 25,000 until 1998 and Rs. 2 lakh from 1999 onwards (the\n\nlatter is about $4,000 using historical exchange rates). The universal and comprehensive\n\nnature of these financial data is substantially stronger than what can be assembled for\n\nmost countries, including the United States for example. The BSR data have been used in\n\nrecent research by Cole (2009), Das et al. (2016), Kumar (2016), and Das et al. (2018).\n\nWhile the micro-data can only be accessed at the RBI by their staff, we were allowed\n\nto aggregate these data for external use. Our aggregations are at the level of the district\n\nx industry x year. Districts are administrative subdivisions of Indian states or union\n\nterritories that provide more-granular distances from the various highway networks. We\n\nprepare our platform to resemble those used in prior studies on India's manufacturing\n\nsector to facilitate comparability. <sup>10</sup> [^10: See Ghani et al. (2016). For additional detail on the manufacturing survey data, see Fernandes and\nPakes (2008), Hasan and Jandoc (2010), Kathuria et al. (2010), Nataraj (2011), and Ghani et al. (2014).] Accordingly, the core sample contains 311 districts\n\nthat account for over 90% of manufacturing activity in India. The excluded districts from\n\nthe full set of 630 districts make very limited contributions to organized manufacturing.\n\nIndustry categories are 2-digit NIC for manufacturing and 1-digit for all other indus\ntry groups. We have invested substantial time in cleaning and validating the data\n\nMention: \"Basic Statistical Return (BSR)1A\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Details of each loan above a threshold are reported annually by every branch of every scheduled commercial bank.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001715:9:0", "occurrence_id": "01df1f243c49d049f4c53277", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 3.1 Financial Banking Data\n\n\nThe essential ingredient for this project is a micro data set that we built based upon the\n\nBasic Statistical Return (BSR)1A, maintained by the Central Bank (RBI). BSR-1A has\n\ndetails of each loan outstanding (above a threshold), reported annually by every branch of\n\nevery scheduled commercial bank in India. The threshold over which individual account\n\ndata are reported was Rs. 25,000 until 1998 and Rs. 2 lakh from 1999 onwards (the\n\nlatter is about $4,000 using historical exchange rates). The universal and comprehensive\n\nnature of these financial data is substantially stronger than what can be assembled for\n\nmost countries, including the United States for example. The BSR data have been used in\n\nrecent research by Cole (2009), Das et al. (2016), Kumar (2016), and Das et al. (2018).\n\nWhile the micro-data can only be accessed at the RBI by their staff, we were allowed\n\nto aggregate these data for external use. Our aggregations are at the level of the district\n\nx industry x year. Districts are administrative subdivisions of Indian states or union\n\nterritories that provide more-granular distances from the various highway networks. We\n\nprepare our platform to resemble those used in prior studies on India's manufacturing\n\nsector to facilitate comparability. <sup>10</sup> [^10: See Ghani et al. (2016). For additional detail on the manufacturing survey data, see Fernandes and\nPakes (2008), Hasan and Jandoc (2010), Kathuria et al. (2010), Nataraj (2011), and Ghani et al. (2014).] Accordingly, the core sample contains 311 districts\n\nthat account for over 90% of manufacturing activity in India. The excluded districts from\n\nthe full set of 630 districts make very limited contributions to organized manufacturing.\n\nIndustry categories are 2-digit NIC for manufacturing and 1-digit for all other indus\ntry groups. We have invested substantial time in cleaning and validating the data\n\nMention: \"RBI\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"The acronym identifies the central bank that maintains the return.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001716:13:0", "occurrence_id": "f50cd085217da230f17c32b1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: that are identified are the main sources of electricity generation. Third, the locations of the sup\n\nply sources are exogenous to economic geographic development. Finally, the locations between\n\n\nthe generation sources and the main demand centers are more likely to be electrified.\n\n\nWe identified two sources of energy generation that can be used for the IV strategy: dams for\n\n\nhydroelectricity and wind farms. The main sources of energy supply are hydropower in Ethiopia\n\n\nand hydropower and wind in Kenya (Table A9). Similar to Moneke (2020), we developed an\n\n\nIV that yields a hypothetical electrification status based on a location’s proximity to a straight\n\nline corridor from electricity generators to the main cities. We identified the locations of the\n\n\nelectricity generators using two databases, one on the opening year of dams and another on\n\n\nthe locations of power plants (Platts database). For Ethiopia, we used a database including all\n\n\ndams in Africa, their location, and their year of opening. For Kenya, we used the global power\n\n\nplant database, which includes the capacity and year of commissioning of all power plants by\n\n\ntype of energy (hydro, wind, gas, and geothermal). From the year of the dam opening or power\n\n\nplant commissioning onward, all districts lying along the straight lines connecting the dams or\n\n\npower plants to the main demand centers are considered as having access to electricity. We then\n\n\nidentified the main sources of demand in each country (Addis Ababa in Ethiopia and Nairobi\n\n\nand Mombasa in Kenya). Map A8 in the appendix shows the results of the construction of the\n\n\nelectrification instrument around 1990 and 2015.\n\n\nOur IV satisfies the main assumptions of an IV strategy. The choice of location of hydro\n\n\nand wind generators can be assumed to be driven by geographic and climatic characteristics of\n\n\nthe locations and not by economic activity in the area. The timing of opening can be considered\n\n\nexogenous, as years of delay are common for such projects. The random assignment assumption\n\n\nof the IV would imply that a\n\nMention: \"Platts database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"It provides the locations of power plants used to geocode generation capacity.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001717:29:0", "occurrence_id": "b3126d35dcbce0d8478f7f9b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 6: Planned road and rail investments\n\n\nSource: forthcoming World Bank Project Appraisal Document\n\n\nWe create counterfactuals using the transport networks from each country described in Sec\n\ntion 2. We assume a new speed for the new road corridor and reduce border delays in some\n\n\nscenarios. For the rail corridor, we keep the road network as it is and assume a new direct\n\n\ntransport line between Ngaoundere, Yaounde and Douala. New transport times are computed\n\n\nassuming that the previous roads can be used as well as the new rail line, which is more effi\n\ncient. Stops between the main cities are not permitted. The first section between Ngaoundere\n\nand Yaounde is assumed to be 627 km with an average speed of 70km/h. <sup>22</sup> The second section\n\n\nbetween Yaounde and Douala is assumed to be 261 km with an average speed of 70km/h.\n\n\n**4.3.3** **Aggregate Impacts**\n\n\n**Employment impacts** Table 7 reports the change in employment share for all sectors at the\n\n\nnational level from the combined road and rail investments, complemented with universal ac\n\ncess to electricity and lower border delays. The proposed transport investments are expected\n\n\nto have a marginal impact on structural change at the national level for most countries. Rural\n\n\nelectrification tends to increase employment in sectors of tradable goods (agriculture and man\n\nufacturing). Better regional integration through lower transport and border costs will increase\n\n\nmanufacturing employment in Cameroon, and agricultural employment in Chad. Nigeria bene\n\nfits from a better access to its neighbors to specialize slightly more in the services sector. When\n\n\ncombining all interventions, structural changes are marginal in Nigeria but larger for Chad and\n\n\nCameroon. Cameroon specializes more in traded goods, both agricultural and manufacturing\n\n\n22Distance assumptions come from the website rome2rio.com which reports distance per transport mode.\n\n\n28\n\nMention: \"rome2rio.com\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Distance per transport mode from this website sets the trade cost assumptions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001718:18:0", "occurrence_id": "82c5677126920f017d164493", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 6.b, depicts the partial association between income per capita and the WEF perception of road\nquality during 2008–12. The figure confirms the findings of figure 6.a, that countries with higher income per\ncapita tend to have better road quality. Perceived quality of the road network is the highest in South Africa (and\nabove the international benchmark) for 2008–12—thus, outperforming countries in South America, like Brazil\nand Colombia. Other MICs in the region (for instance, Nigeria and Gabon) clearly underperform for their level\nof income per capita. Among LICs in the region, Guinea appears to have the lowest level of road quality,\nwhereas the quality of roads in Rwanda and Ethiopia is far better than the one predicted by the international\nnorm.\n\n\nRailroad quality also exhibits a direct and positive relationship with income per capita after controlling\nfor demographic and geographic indicators. The perception of railroad quality is the highest in Gabon, closely\nfollowed by South Africa. Countries with lower income per capita (say, Sri Lanka and Morocco) display similar\nlevels of rail quality. Among MICs in the region, the perception of rail quality in Nigeria and Angola is lower\nthan that predicted by the international norm. In the group of LICs in Sub-Saharan Africa, Uganda exhibits the\nlargest distance to the international norm.\n\n\n**Figure 6. Transport Quality and Income per Capita**\n\n\n**a. Share of paved roads** **b. WEF perception of road quality**\n\n\n\n2008 - 2012 Benchmark (average)\n\n\n\n\n\n2008 - 2012 Benchmark (average)\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n|Col1|Col2|\n|---|---|\n|MUS|MUS|\n||BFA<br>MDG<br>MLI<br>MOZ<br>NER<br>SEN|\n|CMR<br>GHA<br>KEN<br>TZA|CMR<br>GHA<br>KEN<br>\n\nMention: \"WEF perception of road\nquality\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Its survey-based road quality rating is correlated with income per capita.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001723:20:0", "occurrence_id": "cb2d6319e28bfc429a74c974", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Lack of innovativeness is also reflected in the small proportion of innovating firms introducing new\ninnovation outputs such as patents (Table 11), although regional innovation leaders present values above\nthe average for the ECA and Africa regions. However, laggards are below international standards. Other\ninnovation outputs such as trademarks or licensing are more important than patents for all the countries\nconsidered.\n\n\n<u>Table 11. Innovation outputs (% of innovating firms)</u>\n\n**<u>Bangladesh</u>** **<u>Pakistan</u>** **<u>India</u>** **<u>Nepal</u>** **<u>SAR average</u>** **<u>ECA</u>** **<u>Africa</u>**\nPatent 8% 2.2% 8% 1.3% 5% 2.5% 6%\n<u>Other than patents</u> <u>18%</u> <u>5%</u> <u>32%</u> <u>1.7%</u> <u>14%</u> <u>-</u> <u>17%</u>\nSource: Author’s elaboration from Enterprise Survey (2014)\n\n<u>Table 12. Links between innovativeness and firm characteristics</u>\n\n**<u>Bangladesh</u>** **<u>India</u>** **<u>Nepal</u>** **<u>Pakistan</u>**\nNo innovators age 18.70 22.54 13.65 22.15\nImitators age 18.42 20.39 18.21 34.24\n<u>radical</u>\n\nMention: \"Enterprise Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Innovation and patenting behaviour across South Asian countries is drawn from this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001728:4:0", "occurrence_id": "853fd026630f5013d2b110df", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Table 1: Venezuelan Net Population Displaced Abroad Over the Years\n\n\n<u>2010</u> <u>2015</u> <u>2019</u>\n\n\nWorldwide 556,641 695,551 4,326,330\nSouth America 62,240 86,964 3,239,730\nNorth America (US+Canada) 196,910 273,418 371,919\nEurope (Portugal, Spain, Italy) 203,117 224,328 309,170\nCentral America & Mexico 21,260 33,065 166,974\nCarribbean 19,629 21,74 119,333\n<u>Others</u> <u>53,485</u> <u>56,702</u> <u>119,204</u>\n<u>South</u> <u>America</u>\n\n<u>2010</u> <u>2015</u> <u>2020</u>\nColombia 43,511 48,714 1,717,352\nPeru 3,504 4,129 1,043,460\nChile 8,095 54,787 475,702\nEcuador 6,120 8,901 415,835\nBrazil 2,844 3,425 261,441\n<u>Argentina</u> <u>1,236</u> <u>1,240</u> <u>217,562</u>\n\n\nSource: Author’s calculation from UN Data\n\n\nThis paper attempts to fill this gap by investigating how integrated the Venezuelan refugees and\nmigrants are in the education, formal labor market and social protection sectors of Brazil <sup>4</sup> and how\ndi↵erent economic and social factors accelerate or hinder the process of integration. By integration,\nthis paper refers to the definition advocated by OECD (2011) that defines integration as a two-way\nprocess of adaptation by migrants\n\nMention: \"UN Data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Refugee and migrant stocks by host country are compiled from this portal.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001729:7:1", "occurrence_id": "bc9189208fbcd789321b4007", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  interest rates via explicit regulation. Specifically,\nwe focus on IRCs in lending and deposit markets, acknowledging that _de jure_ restrictions can\nalso apply to interbank and bond markets.\n\n\nThe formulation and completion of the survey entailed a number of inter-related steps.\nLeveraging existing contacts at central banks and financial supervisory authorities, the initial\noutreach effort targeted the pertinent staff within local regulators and supervisors. However,\ngiven the sometimes-elusive nature of IRCs, with implementing authorities varying from\ncentral banks to different ministries, we also decided to contact the local banking associations,\nas these are in a prime position to provide specific references to the domestic or regional\ninstitutional arrangements. With the aim to cover as many countries as possible, the outreach\neffort included more than 150 jurisdictions.\n\n\nDespite our best efforts, our survey faced several challenges. Not every country responded to\nthe survey. Also, officials from the same country or even the same agency sometimes provided\nconflicting answers. Therefore, we had a second outreach effort to non-respondent local\nauthorities relying on the contacts of World Bank Group financial sector experts, while we also\nfollowed up with the authorities that had provided inconsistent or incomplete responses. In a\nfew countries where we still faced difficulties getting a response from the local authorities, we\nrelied on the answers of World Bank Group financial sector experts. <sup>6</sup> In addition, in an attempt\nto reconcile any inconsistencies, we checked our responses with Ferrari et al. (2018), who\nrecently took stock of lending interest rate caps collecting data from private and public sources;\nwith the 2017 Global Financial Inclusion and Consumer Protection Survey, which included\n\n\n6 These countries are Algeria, Argentina, Bangladesh, China, Ethiopia, Ghana, Kenya, South Africa and the United\nStates.\n\n\n6\n\nMention: \"2017 Global Financial Inclusion and Consumer Protection Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its responses are compared with a separate stocktake of interest rate caps.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001730:29:0", "occurrence_id": "8a6d399c23d2d75efb1fdc58", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **<mark>Table 1: Children’s anthropometrics prior to the influx of refugees (refugees Vs. non-refugees receiving areas)</mark>**\n\n\n\n\n\n\n\n\n\n|Col1|(1)<br>Refugees<br>Areas*|(2)<br>Non-refugees<br>Areas|Overall|(1) vs. (2)|\n|---|---|---|---|---|\n|**1991**|||||\n|Height-for-age Z-score|-1.986|-2.028|-1.991|0.042|\n||(0.028)|(0.069)|(0.026)|(0.082)|\n|Stunting Prob (HAZ<-2SD)|0.513|0.542|0.516|-0.029|\n||(0.009)|(0.024)|(0.008)|(0.026)|\n|**1992**|||||\n|Height-for-age Z-score|-1.890|-2.017|-1.902|0.127|\n||(0.031)|(0.088)|(0.030)|(0.102)|\n|Stunting Prob (HAZ<-2SD)|0.489|0.529|0.492|-0.041|\n||(0.010)|(0.031)|(0.009)|(0.033)|\n\n\n**<mark>Source:</mark>** <mark>1991/1992 Tanzania DHS.(*)</mark> Kagera, Kigoma\n\nMention: \"Tanzania DHS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Stunting probabilities by region are computed from this early round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001732:33:1", "occurrence_id": "a0f6924b76af0074abefef45", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .1% for males and 3.1% for\nfemales. The sign of the coefficient is as expected because, as previously mentioned, overweight\nand obesity consume health capital which is in this case reflected in the expected returns.\n\nControlling for accumulated capital, measured by height and years of schooling, and for the\ncontext variables that determine labor participation, no statistically significant differences are\nobserved in the expected income for indigenous persons.\n\nOther results, consistent with the results of the mid-term evaluation, indicate that individuals\nwho migrated from their villages of origin have on average higher incomes: 47.3% higher for\nmales and 20.3% for females, in comparison to those who remained in their villages of origin.\nWe have no data on the population that migrated to the United States or to parts of Mexico other\nthan major cities. Thus, this estimate reflects only the migration to places captured in the ENCEL\n2017 data. We thus expect this estimate to underestimate the effect of migrating on income, as\nthose who migrate to the United States likely have the highest income and are excluded from this\nanalysis.\n\nDespite this, the estimator reflects that the return to accumulated capital in health and\neducation can be enhanced in presumably more favorable contexts, that is, those with less\n_marginalization_ (poverty) and isolation than the villages of origin. This same logic is reflected in\nthe estimations of the selection model (columns 2 and 4), where having migrated from one’s place\nof origin reflects a greater probability of entering the labor market, especially for females.\n\n\n17 The mobility in additional height for male individuals of the group with greater exposure in critical periods is of 2.4\ncm. In an equivalent manner, the mobility in additional years of schooling is 3.53 years. For females the calculation\nis equivalent.\n18This considers the average income of the population and adds the average gains by height and schooling. Pesos\nreported in the collection period of ENCEL 2017.\n\nMention: \"ENCEL\n2017\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"It captures only destinations included in its frame, so the migration income effect is understated.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001734:37:0", "occurrence_id": "a20791856cc8ec36301dfa9b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: are aware, has taken this fairly obvious feature of the original Putnam argument seriously. It\n\n\ndid nothing to foster these types of interactions and indeed, as our network survey suggests,\n\n\nmay have even crowded some of them out.\n\n\nThese results will undoubtedly come as a disappointment to those who had hoped for\n\n\nincreases in the program communities’ stocks of social capital but there are several points to\n\n\nkeep in mind. First, CDF had several goals besides increases in social capital and this study\n\n\nhas not assessed the impact of the program on those other goals. Second, perhaps five years is\n\n\ntoo short a time to expect such fundamental changes in people’s attitudes. Third, regardless\n\n\nof the impacts of the program on social capital or other outcomes, CDF has already made\n\n\na contribution to our understanding of development programming by agreeing to a rigorous\n\n\nrandomized impact evaluation. The findings from this impact evaluation combined with\n\n\nfindings from other programs in other countries will together make development planners\n\n\nsmarter so that scarce development funds can be allocated more efficiently. In that way\n\n\nCDF has already had a positive impact on development programming not only in Sudan but\n\n\nalso around the world.\n\n\n36\n\nMention: \"network survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Its responses on social interactions indicate some interactions were crowded out.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001736:26:1", "occurrence_id": "85c60a4d56a27f3eeb179bad", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  commonly as a reference\n\nfor studies on reserves accumulation. <sup>28</sup> A period in the model refers to a year. I choose a subset of\n\n\nparameter values (summarized in table 2) that can be directly pinned down from the data, and then\n\n\nI choose a second subset of parameter values (summarized in table 3) such that model simulations\n\n\nmatch key aspects of the data.\n\n\nThe risk aversion parameter value ( _σ_ = 2) is standard in quantitative business cycle and sovereign\n\n\ndefault studies. The international risk-free interest rate is set to match the average real interest rate\n\n\nfor US Treasury Bills from 1980 to 2017, which is set to 1 _._ 1% ( _r_ <sup>_∗_</sup> = 0 _._ 011). I use spreads from\n\n\nthe EMBI+ to parameterize the lenders’ stochastic discount factor. I assume that a period with\n\n\nhigh risk aversion is one in which the global EMBI+ is one standard deviation above the median\n\n\nover the sample period. With this procedure, I obtain 6 episodes of high risk premium every 20\n\n\n28The time period starts in 1994 because the Mexican central bank independence reform was approved in 1993.\n\n\n24\n\nMention: \"EMBI+\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its spreads parameterise the lenders' stochastic discount factor and the risk-aversion state.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001738:5:1", "occurrence_id": "49a01696ee0863ec3e76f990", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: : Unemployment and FLFP rates for Egypt and other MENA countries**\n\n**Panel A. Problem of Unemployment is persistent                      Panel B. Female labor force participation rates remain low**\n\n\n\n19\n\n\n14\n\n\n9\n\n\n4\n\n\n\n40\n\n\n30\n\n\n20\n\n\n10\n\n\n0\n\n\n\nAlgeria Egypt Lebanon\n\n\nLibya Morocco MENA\n\n\nSaudi Arabia Tunisia Yemen\n\n\n\nAlgeria Egypt Lebanon\n\n\nLibya Morocco MENA\n\n\nSaudi Arabia Tunisia Yemen\n\n\n\n<mark>Source: World Bank Databank, ILO’s estimations.</mark>\n\n**_<mark>FLFP.</mark>_** <mark>In Egypt, FLFP rates are oscillating at about 22 percent, with a decline in recent years</mark>\n<mark>(Figure 1, panel B). This trend is similar to that of MENA, where on average, one out of five</mark>\n<mark>women are outside the labor force – a rate that has shown no significant change over the past 15</mark>\n<mark>years, except for the Republic of Yemen, which has seen a steady drop. As for FLFP among youth,</mark>\n<mark>similar to its neighboring country, Jordan, the rate is low, increasing during prime-age (peaking</mark>\n<mark>around ages 40-45) and decreasing again as retirement approaches (World Bank, 2020).</mark>\n\n**_<mark>Informality.</mark>_** <mark>In Egypt, informality rates have increased from 58 percent in 2008 to 66.7 percent in</mark>\n<mark>2017, similar to the estimates for a typical MENA country.</mark> <sup>4</sup> <mark>But there remains significant</mark>\n\n\n4 Comparing informality rates within MENA countries can be problematic, given that many of them have either lagged in their capacity to generate data or have\nprevented access to their data\n\nMention: \"World Bank Databank\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Regional female labour force participation rates are read from this databank alongside ILO estimates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001741:16:0", "occurrence_id": "fe4e0546249da6f47989131f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: in Table 15, I rely on a first-difference model in which I examine the effect of the future use of\nInternet job search on past labor market outcomes. To do so, I additionally rely on the 2006\nELMPS survey round and follow the same individuals included in my estimation sample (the 2012\nand 2018 panel) that were also surveyed in 2006. Focusing on working aged individuals, I examine\nthe effect of the change in Internet job search between 2018 and 2012 on changes in labor force\nparticipation and employment between 2012 and 2006. The results reported in Table 15 show that\nthe change in Internet job search between 2018 and 2012 is actually associated with lower labor\nforce participation in the pre-period between 2012 and 2006, while my results in Table 6 actually\nshow a positive association. This suggests that individuals who are using the Internet for job search\ndo not initially have better labor market outcomes compared to individuals who are not—if\nanything, they seem to have lower labor force participation rates in the pre-period. Analyzing the\nimpact on employment in Panel B actually shows no effect of the future use of Internet job search\non past changes in employment, suggesting that the results are not driven by pre-existing trends in\nemployment between individuals who rely on Internet job search and those who do not.\n\n\nThe second test consists in utilizing all the available ELMPS survey rounds conducted in Egypt\nsince 1988 (1988, 1998, 2006, 2012 and 2018). Since attrition is problematic and it is impossible\nto follow individuals who remain in the working age category over such a long period of time, I\nrely on sub-district ( _shyakha_ ) level regressions. <sup>7</sup> [^7: The _shyakha_ is the smallest administrative division in Egypt. Egypt is divided into 27 governorates that are further divided into\n_qism_ (the second administrative division) and _shyakha_ (the third administrative division). The number of sub-districts ( _shyakha_ ) is\n1,101.] In these regressions, I use the percent of\nmale/female Internet job searchers\n\nMention: \"ELMPS survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The earlier round lets individuals in the later panel be traced back to past outcomes.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001745:39:0", "occurrence_id": "8b91d7ef23fc073edc5836f8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Table S1.2: Equality in divorce access**\n\n\n_<u>Equal access to divorce</u>_ _<u>Inequality in access to divorce</u>_\nColombia Azerbaijan\nHaiti Bangladesh\n\nHonduras Burkina Faso\n\nKyrgyz Republic Cambodia\nMozambique Cameroon\nSao Tome and Principe Democratic Republic of Congo\nUkraine Dominican Republic\nGabon\nUK <sup>*</sup> Ghana\nUS <sup>*</sup> India\nJordan\nKenya\nLiberia\nMalawi\nMali\nPakistan\nPeru\nPhilippines\nRwanda\nTajikistan\nTanzania\nTimor Leste\nUganda\n<u>Zambia</u>\n\n_Source:_\n[OECD 2019 Social and Institutions Gender Index, https://www.genderindex.org/](https://www.genderindex.org/)\n\n_Notes:_\n\n1. Equality in access to divorce is defined in the SIGI as \"Women have both the same rights to initiate\ndivorce and the same requirements to finalise divorce or annulment as men, without negative\nrepercussions on their parental authority. This applies to all groups of women. Customary, religious\nand traditional laws or practices do not discriminate against women’s rights regarding divorce or\nparental authority after divorce\".\n2.   - Not in sample; included for comparative purposes.\n\nMention: \"OECD 2019 Social and Institutions Gender Index\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Legal and social restrictions on gender equality, such as access to divorce, are coded in this index.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001749:18:2", "occurrence_id": "01e84e5dff9a7951e5b28958", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: *𝚤** <mark>)</mark> <sup>−1/𝜃𝜃</sup>\n𝑇 **𝑖**\n\n\n\n<mark>with appropriate parameter constraints. Second, these estimates are transformed using a sectoral</mark>\n𝑑 **𝚤**\n<mark>elasticity of substitution (</mark> <mark>𝜃</mark> 𝜃 <mark>):</mark>\n\n<mark>𝑇</mark> 𝑖𝑇 **𝑖** <mark>= (𝑑</mark> <sup><mark>�</mark></sup> 𝚤𝑑 **𝚤** <mark>)</mark> <sup>−1/𝜃𝜃</sup>\n<mark>Database (WIOD) is used and an experimental data set from the Asian Development Bank (ADB-MRIO).</mark>\n<mark>To obtain the estimates of</mark> <mark>𝑇</mark> 𝑖𝑇 **𝑖** <mark>, international and domestic trade data from the World Input-Output</mark>\n<mark>reaction of demand to prices, and hence a higher responsiveness of import demand to trade frictions.</mark>\n<mark>The parameter</mark> <mark>𝜃</mark> 𝜃 <mark>is estimated for each sector in Egger</mark> _<mark>et al</mark>_ <mark>. (2018). A higher</mark> <mark>𝜃</mark> 𝜃 <mark>means a more elastic</mark>\n<mark>reacts less to changes in trade costs.</mark>\n<mark>Generally,</mark> <mark>𝜃</mark> 𝜃 <mark>takes on lower\n\nMention: \"Database (WIOD)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its input-output tables are one of two sources used to estimate trade elasticities.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001751:15:1", "occurrence_id": "231f7d08d9bc439882b0de36", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \nMost countries with ADL questions are in Europe/Central Asia and Sub-Saharan Africa. As\nexpected, surveys focused on older persons (e.g. CHARLS in China) have questions on ADL.\nWhile not shown in Appendix 1, there is heterogeneity in the ADL questions that are asked\nacross countries with most often questions around physical basic activities such as walking over\n100 meters, standing or sitting.\n\nAppendix 2 includes 36 countries, 53 data sets and 76 data set-years with broad activity\nlimitation questions. The most commonly found broad activity question asks about the person\nbeing limited in daily activities or in learning or in any economic activity due to a health problem\nin the past six months (not shown in Appendix 2). Some countries have several questions to get\nat broad activity limitations, including some recording the number of days the person was unable\nto carry out usual activities.\n\nAppendix 3 has 50 countries, 81 data sets and 113 data set-years with a general disability\nquestion such as “does [Name] have a disability?” Often such questions are followed up with a\nquestion on the type of disability or the impairment the person has.\n\nAppendix 4 has 43 countries, 54 data sets and 75 data set-years that have other types of disability\nquestions such as belonging to the so-called disabled social group for Armenia. <sup>4</sup> In addition,\nsome of the questions found in this list can be mixes of different types of questions such as\nimpairments mixed with chronic health conditions in Tunisia, Bosnia, and Bulgaria. We have\n\n\n4 Disability focused surveys (e.g. Philippines Model functioning Survey) were not covered in this table for other\ndisability questions, but were covered in Tables 1 and 2 and Appendices 1, 2 and 3. Disability focused surveys\ntypically have several types of other disability questions not beyond functional ADL, general and broad activity\nlimitation questions.\n\n3\n\nMention: \"Philippines Model functioning Survey\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"It is named as a disability-focused survey excluded from this table but covered elsewhere.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001760:6:0", "occurrence_id": "c63b86776a94817d8a3480bc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Determinants of empowerment_\n\nA relatively rich research identifies and describes which socio-economic variables tend to have a\n\nstrong association with women’s empowerment and agency. This research shows that women\n\nwho are older, have higher literacy, more children, and live in households of higher\n\nsocioeconomic status are more likely to participate in household decisions (Hameed et al. 2014).\n\nA study in Indonesia found that women’s bargaining power for decisions on use of prenatal care\n\nand delivery is highly influenced by a woman’s access to economic resources. Further, social\n\npower influences whether decisions are made jointly rather than by the husband alone. Women\n\nwho are from higher education or family status than their male spouse were more likely to\n\nexperience couple joint decision-making (Beegle, Frankenberg, and Thomas 2001). As will be\n\ndiscussed in more detail below, the quality of a woman’s relationship is also related to her\n\nparticipation in household decisions and her agency overall (Allendorf 2012). For example, a 2000\n\nstudy in Bangladesh found that the extent to which women negotiated employment depends on\n\nthe level of harmony in their relationships (Kabeer 2001).\n\n\n_Empirical methods to measure agency and the intra-household bargaining models_\n\nEmpowerment is not a new concept but its systematic measurement by researchers and\n\npractitioners is relatively new (World Bank 2005). This recent shift is related to the idea that\n\nempowering the poor is crucial to reduce poverty – since no one has a greater stake in the fight\n\nagainst poverty than the poor themselves. A gold standard for measuring women’s\n\nempowerment is not yet available. The methods to gauge women’s agency, a key component of\n\nempowerment, have been largely inconsistent (Donald et al., 2016). Questions on household\n\ndecision-making and resources allocations, such as the ones of the DHS surveys used in this\n\npaper, are increasingly added to household surveys. However, some issues remain on how to\n\nconstruct a\n\nMention: \"DHS surveys\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Its decision-making and resource allocation questions are now common, but construction issues remain.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001761:13:2", "occurrence_id": "a1e63b128d3636d6a54a65b2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ||||Mea<br> Ar 2<br> <br>|Mea<br> Ar 2<br> <br>|n for non commmuters<br>  58816<br> Mean for commmuters<br>|\n||||||||||||\n||||||||||||\n|||||||||<br>|<br>|<br> Ar 309428|\n||||||||||||\n||||||||||||\n||||||||||||\n||||||||||||\n||||||||||||\n||||||||||||\n||||||||||||\n||||||||||||\n\n\n\n0 200000 400000 600000 800000\nMonthly after-tax salary (cash or in-kind)\n\n\n\n\n\n\n\nSources: INSTAT; World Bank estimates. Source: World Bank estimates.\n\nMention: \"INSTAT\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Monthly after-tax salary figures are taken from this statistical office.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001765:18:0", "occurrence_id": "206d90b17d8c3201ec64fdff", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: for the level of electricity outages in the respective years. To demonstrate that this proxy\n\n\ncorrelates with actual lightning intensity in the African context, I use data on lightning\n\n\nflash rate from NASA’s OTD/LIS satellite and correlate it with the proxy ( _CAPE_ _× P_ )\n\n\nas shown in Figure A1, and Figure A2 in the online appendix. The scatter plot in Figure\n\n\nA1 reveals a high r-square: _CAPE_ _× P_ explains about 77% of the variations in lightning\n\n\nintensity in Africa, thus confirming the earlier findings of Romps et al. (2014). Figure A2\n\n\nin the online appendix also shows a high spatial correlation between the actual lightning\n\n\nflash rate and the proxy.\n\n\nIn addition, data on mean annual temperature and precipitation from the ERA5 Global\n\nReanalysis Database <sup>22</sup> are used in the analysis. Summary statistics are presented in Table\n\n\nA1 in the appendix.\n\n#### **4 Empirical Strategy and Results**\n\n###### **4.1 Identification Strategy**\n\n\nEmpirical estimation of the causal impact of electricity outages on outcomes such as em\n\nployment and firm productivity is often beset with methodological challenges. Notable\n\n\namong them is the issue of endogeneity resulting from the potential correlation between\n\n\noutage intensity and (observable and unobservable) factors that (in)directly influence\n\n\nthese outcomes. In other words, any assumption that variations in outage intensity are\n\n\northogonal to economic outcomes such as employment and firm productivity is unlikely\n\n\nto be valid. For instance, firm location, industry composition, and the prevailing economic\n\n\nand political conditions in a country can influence both outage intensity and the perfor\n\nmance of firms. Also, regions with high unemployment and hence low income are more\n\n\nlikely to suffer outages possibly due to the non-payment of electricity services leading\n\n\n[22https://cds.climate.copernicus\n\nMention: \"ERA5 Global\n\nReanalysis Database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Mean annual temperature and precipitation enter the analysis as climatic controls.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001770:8:0", "occurrence_id": "979f5421face73ab13b70174", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 3. Firm ownership**\n\n\n\n**Figure 4. Firm size**\n**(number of employees)**\n\n\n\n**Figure 5. Distribution of firm age**\n\n\n\n\n\n\n\n_Note_ : Business’s age is truncated at 35.\n\nBy industry, information and communication technology (ICT) is the largest industry in the\nonline job portal in terms of the number of employers and job postings (Figure 6). ICT\nconstitutes 23% of the employers, and each employer has posted four jobs on average. The\nsecond largest industry is administrative, social, and personal services (15% and 4 jobs),\nfollowed by professional, scientific, and technical services (14% and 3 jobs), manufacturing\n(12% and 3 jobs), and wholesale, retail, hotel, and restaurant (11% and 2 jobs).\n\n\n\n**Figure 6. Distribution of firms and the number of job postings per firm over industries**\n\n\n_Source_ : Authors’ calculation using the data from Rozee.pk.\n\nIn sum, the comparison of the characteristics of Rozee.pk data and the labor market data clearly\nshows that Rozee.pk data represent a high-pay labor market domain where better qualified\njobseekers and more productive employers participate. The labor force participants in Rozee.pk\nare younger and more educated and earn more than the average labor force in Pakistan. Females\n\n\n7\n\nMention: \"data from Rozee.pk\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The distribution of job postings per firm across industries is computed from this job portal.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001772:38:1", "occurrence_id": "7fabbf1d4d72399bc69be6d7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: *<br>(0.0231)<br>(0.0238)<br>-0.0857***<br>-0.0640**<br>(0.0205)<br>(0.0236)<br>-0.0481*<br>-0.0215<br>(0.0257)<br>(0.0264)<br>219,984<br>147,360<br>0.211<br>0.208<br>Yes<br>Yes<br>Yes<br>Yes<br>Yes<br>Yes<br>All Other<br>Restricted|\n\n\n\nStandard errors in parentheses, *** _p <_ 0 _._ 01, ** _p <_ 0 _._ 05,        - _p <_ 0 _._ 1\n\n\nNote: this table shows the quarterly impacts of Formosa for the rest of 2016, on fishing activity in the fishingrestricted region (i.e. 20nm near-shore of Formosa-affected provinces), using VIIRS’ monthly-aggregate boat\ndetection data. Each observation is a grid-month. Sample includes all monthly observations between 2013\nand 2016. The grid size is 10-mile-square. A grid is considered near-shore when it is located less than 20\nnautical miles away from the coast line. The reported outcome variables include log-transformed monthlyaggregate boat counts in each grid (columns 1-2), and 0/1 probability that the grid was detected with at\nleast a boat in that month (columns 3-4). The control group ”All Other” refers to all coastal provinces in\nthe country except for the four Formosa-affected central provinces (Ha Tinh, Quang Binh, Quang Tri, and\nHue).\n\nMention: \"monthly-aggregate boat\ndetection data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Fishing activity in the near-shore restricted zone is measured on a ten-mile grid by month.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001777:3:1", "occurrence_id": "419cb8cdad8440cfba349bf2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  also increase land productivity generally.\n\n\nThe second empirical regularity has received far less attention. We call it the inverse\n\n\n_plot-size_ productivity relationship (and refer to it as plot-size IR). This relationship, observed\n\n\nin many Sub-Saharan African countries, states that smaller plots are more productive than larger\n\n\nplots, even within a farm household (Ali and Deininger, 2015; Assunção and Braido, 2007;\n\n\nBarrett et al., 2010). Although rarely noted, such a relationship implies that land\n\n\nfragmentation—subdividing already small plots into even smaller units—would heighten\n\n\nproductivity. But more and smaller plots lead to increased travel time (in moving from plot to\n\n\nplot), more wasted space along borders, and fewer options for mechanization, which typically\n\n\nfunctions best on larger plots. For these reasons and others, this finding does not seem credible.\n\n\nA common policy recommendation to improve land productivity is not fragmentation but\n\n\n1 Estimates from PovcalNet ( _iresearch.worldbank.org/PovcalNet_ ), the World Bank data tool for\nestimating extreme poverty, indicate that most of the poor in the rest of the world also live in rural areas.\nFor example, in China essentially all the extreme poor (living on less than US$1.90 a day) are estimated\nto be in rural areas; and in India about 80 percent.\n\n\n2\n\nMention: \"PovcalNet\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It indicates where most of the world's extreme poor live.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001781:21:0", "occurrence_id": "2511c34ece917b16a19feb35", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _established democracies_ (hereafter “democracies” or DEM) if they have an average Polity IV score\n\n\n>7.5 (Marshall et al. 2016), and as _autocracies or transition countries_ (hereafter “autocracies” or\n\n\nAUT) otherwise. The purpose of the high polity cutoff is to exclude countries that transition in\n\n\nand out of democracy, since countries are categorized once as one regime or the other. <sup>30</sup> [^30: AUT includes autocracies as well as democratic-transition countries that spent much of the sample as autocracies] Combined,\n\n\nthere are around 125 countries, 7,000 observations, and 1,100 leaders for which we have growth,\n\n\nleader, and polity data. Of these, 23% are established democracies. (See the Online Appendix for\n\n\nfurther details and descriptive statistics.) For our results involving individual leaders’ growth\n\n\nestimates, we restricted the sample to leaders with tenure ≥ 3 years, and growth data available for\n\n\ntheir whole tenure, leaving a sample of approximately 650 leaders and 5,400 observations.\n\n\nThere are six regions (for region-by-year fixed effects), which are based on the World Bank\n\n\nclassification, with three modifications: (i) European offshoots (the United States, Canada,\n\n\nAustralia, and New Zealand) are added to Europe, as is common in the literature; (ii) South Asia\n\n\nand East Asia & Pacific are combined, due to the small number of countries in the former; and (iii)\n\n\nCommunist transition countries geographically close to the former USSR form their own region,\n\n\nas they experienced a distinct set of common shocks, especially in the early 1990s. <sup>31</sup> [^31: The six regions are: Sub-Saharan Africa (SSA); Middle East and North Africa (MENA); Latin America and the]\n\n\n**Outliers.** Per capita GDP growth rates are often very volatile—for example, around wars—\n\n\n\nand a small number of extreme observations can have a large\n\nMention: \"Polity IV\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"An average score above 7.5 classifies a country as an established democracy.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001759:37:1", "occurrence_id": "9aad12783ec156a5f603d946", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  #36\ngovernance attributes. Higher values indicate better governance.\n\nEffective board size score ranging from 1 to 3. Higher values\nindicate better governance. If the board size is between 9 and 12, it\ntakes a value of 3. If the board size is 6-8 or 13-15, then it takes a\nvalue of 2. If the board size is less than 6 or greater than 15, then it\ntakes a value of 1.\n\n\n\nISS\n\n\nISS\n\n\n\nCEO Chairman separation score ranging from 1 to 3. The score is\n_ceo_chair_ ISS\nbased on attribute #8. Higher values indicate better governance.\n\n\n\n_gov38_\n\n\n_nom_comm_\n\n\n_comp_comm_\n\n\n_audit_comm_\n\n\n**<u>Country Variables</u>**\n\n\n\nThe governance index using 38 governance attributes. The\nexcluded ones are anti-takeover attributes. Higher values indicate\nbetter governance.\n\nNominating committee independence score ranging from 1 to 5.\nThe score is based on attribute #9. Higher values indicate better\ngovernance.\n\nCompensation committee independence score ranging from 1 to 5.\nThe score is based on attribute #7. Higher values indicate better\ngovernance.\n\nAudit committee independence score ranging from 1 to 5. The\nscore is based on attribute #26. Higher values indicate better\ngovernance.\n\n\n\nMeasure of minority shareholder protection. It ranges from 0 to 10.\n_investor_\nHigher values indicate better protection.\n\n\n\nISS\n\n\nISS\n\n\nISS\n\n\nISS\n\n\nWorld Bank\nDoing Business\n\nSurvey\n\n\nWorld Bank\n\n\nWorld Bank\nnational accounts\n\ndata, OECD\n\nNational\nAccounts data\n\n\n\n_common law_\n\n\n_gdp_growth_\n\n\n_inflation_\n\n\n\nIt is a dummy variable that attains a value of 1 if the country’s\nlegal origin is common law (US, UK, Australia, New Zealand,\nCanada, Hong Kong and Ireland and Singapore), and 0 otherwise.\n\nGrowth rate of real GDP per capita. GDP per capita is gross\ndomestic product divided by midyear population. GDP is the sum\nof gross value\n\nMention: \"ISS\n\n\nWorld Bank\nDoing Business\n\nSurvey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its indicator of minority shareholder protection on a zero to ten scale is used as a legal variable.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001759:37:1", "occurrence_id": "6d543c406575ff309c3d6b14", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  #36\ngovernance attributes. Higher values indicate better governance.\n\nEffective board size score ranging from 1 to 3. Higher values\nindicate better governance. If the board size is between 9 and 12, it\ntakes a value of 3. If the board size is 6-8 or 13-15, then it takes a\nvalue of 2. If the board size is less than 6 or greater than 15, then it\ntakes a value of 1.\n\n\n\nISS\n\n\nISS\n\n\n\nCEO Chairman separation score ranging from 1 to 3. The score is\n_ceo_chair_ ISS\nbased on attribute #8. Higher values indicate better governance.\n\n\n\n_gov38_\n\n\n_nom_comm_\n\n\n_comp_comm_\n\n\n_audit_comm_\n\n\n**<u>Country Variables</u>**\n\n\n\nThe governance index using 38 governance attributes. The\nexcluded ones are anti-takeover attributes. Higher values indicate\nbetter governance.\n\nNominating committee independence score ranging from 1 to 5.\nThe score is based on attribute #9. Higher values indicate better\ngovernance.\n\nCompensation committee independence score ranging from 1 to 5.\nThe score is based on attribute #7. Higher values indicate better\ngovernance.\n\nAudit committee independence score ranging from 1 to 5. The\nscore is based on attribute #26. Higher values indicate better\ngovernance.\n\n\n\nMeasure of minority shareholder protection. It ranges from 0 to 10.\n_investor_\nHigher values indicate better protection.\n\n\n\nISS\n\n\nISS\n\n\nISS\n\n\nISS\n\n\nWorld Bank\nDoing Business\n\nSurvey\n\n\nWorld Bank\n\n\nWorld Bank\nnational accounts\n\ndata, OECD\n\nNational\nAccounts data\n\n\n\n_common law_\n\n\n_gdp_growth_\n\n\n_inflation_\n\n\n\nIt is a dummy variable that attains a value of 1 if the country’s\nlegal origin is common law (US, UK, Australia, New Zealand,\nCanada, Hong Kong and Ireland and Singapore), and 0 otherwise.\n\nGrowth rate of real GDP per capita. GDP per capita is gross\ndomestic product divided by midyear population. GDP is the sum\nof gross value\n\nMention: \"OECD\n\nNational\nAccounts data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Growth and inflation are taken from these accounts alongside the World Bank's.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001774:22:1", "occurrence_id": "2c9bee46b2d9fb0a9d787546", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Group. 2014. “World Development Indicators.”\n\nhttp://databank.worldbank.org/data/views/variableselection/selectvariables.aspx?source=wo\nrld-development-indicators.\n\n\nYoung, Alwyn. 2012. “The African Growth Miracle.” _National Bureau of Economic Research_\n\n_Working Paper Series_ No. 18490.\n\n\n20\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"A reference list entry pointing to the databank page for these indicators.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001778:28:1", "occurrence_id": "0cf84b7a19269e1533a2f879", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: -differences regression using the years\n\n1990 and 2013 (1991 and 2011 for demographic census variables) as the starting and ending\n\npoints. This approach is motivated by an attempt to capture the long-term effects of the IA\n\nprogram. As a robustness test, we conduct an additional analysis which also uses additional,\n\nintermediate rounds of the Economic and Demographic Censuses.\n\n\nWe use data from census rounds in the years 1990, 1998, 2005, and 2013 for outcomes\n\nobserved in the Economic Census; and data from 1991, 2001 and 2011 for outcomes observed\n\nin the Demographic Census. Treatment is assigned the value of 1 for every round of data\n\noccurring after the establishment of the IA. The results are given in Table 10, and are\n\nconsistent with out main findings.\n\n\nWe also estimate specifications that allow for spillovers that extend beyond 5 kms from\n\nthe IA. For this analysis, we include indicator variables for 1 km distance intervals extending\n\nup to 10 kms from the IA and construct the control group from villages located more than 10\n\nkms from the IA. The results are given in Table 11. There is little evidence that treatment\n\neffects extend beyond 5 kms, and the coefficients on nearer distances are largely unaffected\n\n\n26\n\nMention: \"Economic Census\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Outcomes in rounds after the intervention are compared with earlier ones.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001778:28:1", "occurrence_id": "3ee9b26860932afdebc3749b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: -differences regression using the years\n\n1990 and 2013 (1991 and 2011 for demographic census variables) as the starting and ending\n\npoints. This approach is motivated by an attempt to capture the long-term effects of the IA\n\nprogram. As a robustness test, we conduct an additional analysis which also uses additional,\n\nintermediate rounds of the Economic and Demographic Censuses.\n\n\nWe use data from census rounds in the years 1990, 1998, 2005, and 2013 for outcomes\n\nobserved in the Economic Census; and data from 1991, 2001 and 2011 for outcomes observed\n\nin the Demographic Census. Treatment is assigned the value of 1 for every round of data\n\noccurring after the establishment of the IA. The results are given in Table 10, and are\n\nconsistent with out main findings.\n\n\nWe also estimate specifications that allow for spillovers that extend beyond 5 kms from\n\nthe IA. For this analysis, we include indicator variables for 1 km distance intervals extending\n\nup to 10 kms from the IA and construct the control group from villages located more than 10\n\nkms from the IA. The results are given in Table 11. There is little evidence that treatment\n\neffects extend beyond 5 kms, and the coefficients on nearer distances are largely unaffected\n\n\n26\n\nMention: \"Demographic Census\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Its rounds supply the demographic outcomes in the same difference-in-differences design.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001788:5:0", "occurrence_id": "bccb65fc2af295576632131a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ###### Introduction\n\nEven before COVID-19 shut schools, the world was in the middle of a learning crisis that threatened efforts\nto build human capital—the skills and know-how needed for the jobs of the future. More than half of\nchildren (53 percent) in low- and middle-income countries either were out of school or failed to learn to\nread with comprehension by age 10 and so were in learning poverty. <sup>1</sup> [^1: World Bank 2019.] The most recent historical trends\nsuggest that it will take 50 years to halve learning poverty in the developing world. COVID-19 is likely to\nslow that rate by deepening learning gaps and so increasing the number of children in learning poverty in\nmany countries.\n\n<mark>Temporary school closures in more than 180 countries have kept nearly 1.6 billion students out of school,</mark>\n<mark>further complicating global efforts to reduce learning poverty. Although most countries have made heroic</mark>\n<mark>efforts to put remote and remedial learning strategies in place, learning losses are accumulating rapidly.</mark>\n<mark>A recent survey on national education responses to COVID-19 by the United Nations Children’s Fund</mark>\n<mark>(UNICEF), the United Nations Educational, Scientific, and Cultural Organization (UNESCO), and the World</mark>\n<mark>Bank shows that while countries and regions have responded in various ways, they have found it difficult</mark>\n<mark>to reach even half of students.</mark>\n\nCOVID-19-related school closures are forcing countries even further off track from achieving their learning\ngoals. Students currently in school stand to lose $10 trillion in labor earnings over their working lives. <sup>2</sup>\nThat is one-tenth of global GDP, or half the United States’ annual economic output\n\nMention: \"survey on national education responses to COVID-19\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its country responses support the claim that learning losses are accumulating.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001792:27:0", "occurrence_id": "b5669c7800230f95d17fd062", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Table 2: Regression Analysis of External Capital Factors**\n\n\nThis table reports results from OLS regressions with value ratios (market capitalization divided by total sales and total debt divided by total sales) as the dependent\nvariables. Coefficients and _p_ -values (calculated using two-tailed tests and heteroskedasticity consistent standard errors) are tabled. Coefficients significant at the 5%\nlevel are in bold face. Not all variables are available for all firms so regressions are estimated using all firms for which sufficient data are available. The first set of\nresults is for all firms. The second set of results is for firms in the SDC-WDR database without foreign currency debt and firms not in the SBC-WDR database which are\nassumed to not have foreign currency debt. The third set of results is for firms with foreign currency debt in the SBC-WDR database. Data for log sales, asset tangibility\n(fixed assets as a percent of total assets), and profitability (operating income divided by total sales) are from the Worldscope database. Dummy variables for 1-digit SIC\ncodes of 2, 3, 4, 5, 6, 7, and 8 are included (SIC<1000 is the base case).\n\n**<u>All Firms</u>** **<u>Without Foreign Currency Debt</u>** **<u>With Foreign Currency Debt</u>**\n<u>MktCap / Sales</u> <u>Total Debt / Sales</u> <u>MktCap / Sales</u> <u>Total Debt / Sales</u> <u>MktCap / Sales</u> <u>Total Debt / Sales</u>\n<u>Coef.</u> _<u>p</u>_ <u>\n\nMention: \"SDC-WDR database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Firms are grouped by whether they carry foreign currency debt using this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001792:27:0", "occurrence_id": "2b318c777c7855e5e364f10c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Table 2: Regression Analysis of External Capital Factors**\n\n\nThis table reports results from OLS regressions with value ratios (market capitalization divided by total sales and total debt divided by total sales) as the dependent\nvariables. Coefficients and _p_ -values (calculated using two-tailed tests and heteroskedasticity consistent standard errors) are tabled. Coefficients significant at the 5%\nlevel are in bold face. Not all variables are available for all firms so regressions are estimated using all firms for which sufficient data are available. The first set of\nresults is for all firms. The second set of results is for firms in the SDC-WDR database without foreign currency debt and firms not in the SBC-WDR database which are\nassumed to not have foreign currency debt. The third set of results is for firms with foreign currency debt in the SBC-WDR database. Data for log sales, asset tangibility\n(fixed assets as a percent of total assets), and profitability (operating income divided by total sales) are from the Worldscope database. Dummy variables for 1-digit SIC\ncodes of 2, 3, 4, 5, 6, 7, and 8 are included (SIC<1000 is the base case).\n\n**<u>All Firms</u>** **<u>Without Foreign Currency Debt</u>** **<u>With Foreign Currency Debt</u>**\n<u>MktCap / Sales</u> <u>Total Debt / Sales</u> <u>MktCap / Sales</u> <u>Total Debt / Sales</u> <u>MktCap / Sales</u> <u>Total Debt / Sales</u>\n<u>Coef.</u> _<u>p</u>_ <u>\n\nMention: \"SBC-WDR database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Firms absent from it are assumed to have no foreign currency debt.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001792:27:0", "occurrence_id": "820d7c0b16fe936e2f32ab1f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Table 2: Regression Analysis of External Capital Factors**\n\n\nThis table reports results from OLS regressions with value ratios (market capitalization divided by total sales and total debt divided by total sales) as the dependent\nvariables. Coefficients and _p_ -values (calculated using two-tailed tests and heteroskedasticity consistent standard errors) are tabled. Coefficients significant at the 5%\nlevel are in bold face. Not all variables are available for all firms so regressions are estimated using all firms for which sufficient data are available. The first set of\nresults is for all firms. The second set of results is for firms in the SDC-WDR database without foreign currency debt and firms not in the SBC-WDR database which are\nassumed to not have foreign currency debt. The third set of results is for firms with foreign currency debt in the SBC-WDR database. Data for log sales, asset tangibility\n(fixed assets as a percent of total assets), and profitability (operating income divided by total sales) are from the Worldscope database. Dummy variables for 1-digit SIC\ncodes of 2, 3, 4, 5, 6, 7, and 8 are included (SIC<1000 is the base case).\n\n**<u>All Firms</u>** **<u>Without Foreign Currency Debt</u>** **<u>With Foreign Currency Debt</u>**\n<u>MktCap / Sales</u> <u>Total Debt / Sales</u> <u>MktCap / Sales</u> <u>Total Debt / Sales</u> <u>MktCap / Sales</u> <u>Total Debt / Sales</u>\n<u>Coef.</u> _<u>p</u>_ <u>\n\nMention: \"Worldscope database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Sales, asset tangibility and profitability are drawn from this commercial source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001796:11:2", "occurrence_id": "eb36475327482878730e9fd8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: > <u>1 (JPN)</u> <u>32 (CAN)</u>\n<u>Source: OECD STRI; own calculations</u>\n\n\n5 There is, however, no variation in limitations on duration for these service professionals across sectors for each\ncountry according to the STRI data for the year 2017 (detailed analysis at the sector level is presented in Annex\nTable 3).\n\n\n10\n\nMention: \"the STRI data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"It shows no cross-sector variation in duration limits for these professionals.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001798:4:0", "occurrence_id": "0334464d17b567fc76ffee27", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: of forced migrants, the PEP became a benchmark for a development-based approach to the\n\n\ncurrent predicament.\n\n\nRegularization (or legalization) programs give migrants legal residence and thus access to\n\n\nthe rights and services enjoyed by citizens, including the right to work. These programs,\n\n\nwhich grant temporal benefits but typically exclude voting rights, have been implemented\n\n\nin the United States, Europe, Latin America, and LMICs in response to a sharp increase in ir\n\nregular migration. Despite their prevalence, these programs are highly controversial. Critics\n\n\nclaim they encourage irregular migration, lead to job displacement and unemployment for\n\n\nnatives, increase crime, and burden social welfare and public finances. Yet, prior research\n\n\nhas demonstrated negligible effects on employment opportunities for natives and shown\n\n\nreductions in migrants’ crime rates (Bahar et al. 2021; Pinotti 2017; Baker 2015). Further\n\nmore, from a theoretical perspective, regularization programs may produce positive effects\n\n\nfor both migrants and host societies. The right to work means migrants may rely less on\n\n\nsocial welfare and be more likely to pay taxes, more productive by working in occupations\n\n\nmore closely related to their skills, and more inclined to invest in their children’s human\n\n\ncapital. Research on long-term migration has established positive effects on host countries\n\n\nand shown that these effects are larger or faster if migrants are permitted to integrate eco\n\nnomically (Abramitzky and Boustan 2022).\n\n\nThe features of PEP facilitate causal identification of its effects. First, the program was in\n\ntroduced unexpectedly, thereby isolating anticipatory decisions or ex-ante behavioral re\n\nsponses. Unknown to both migrants and government officials, ex-post eligibility for the\n\n\nprogram was based solely on prior registration in a nationwide census of irregular forced mi\n\ngrants, the _Registro Administrativo de Migrantes Venezolanos_ (RAMV for its Spanish acronym),\n\n\nthat was administered between April and June of 2018. According to the government of\n\nficials who designed RAMV\n\nMention: \"nationwide census of irregular forced mi\n\ngrants\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Prior registration in this confidential census determined programme eligibility.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001799:34:1", "occurrence_id": "b35eb78c41dc880915c7ce2d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  reported (out of 7 days) 0.16 0.32 -0.16\n(1.75) (1.81) (0.86)\n\n\nOutcome Variable Average 77.75 48.71 29.04\n\n\n<u>Observations</u> <u>1,097</u> <u>1,097</u> <u>1,097</u>\n\n\nSource: Author calculations from The Hohoe Garment Makers Project Data\n\n\n33\n\nMention: \"The Hohoe Garment Makers Project Data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Outcome averages over 1,097 observations are computed from this project's data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001803:6:1", "occurrence_id": "98f4eb790cf7faf4cfa55656", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  well as pre- and posttax/transfer estimations. This coverage, however, comes at the cost of extensive imputations to fill gaps over time in\na country, to extend the data to countries with missing data, and to obtain post-tax/transfer estimates. We recognize\nthis limitation and point out that the World Bank PovcalNet is the unique comprehensive global dataset of nonimputed inequality (and poverty) data. For our purposes in writing this survey paper, the advantages of the SWIID\noutweigh its limitations.\n4 This section analyzes trends in poverty and inequality starting in the 1980s. Longer time series on wealth and\nincome inequality have been collected by Piketty (2014) and are restricted mostly to advanced economies. Piketty\nand Saez (2014) report sustained improvements in wealth and income distribution across Europe and the United\nStates from the 1930s to the 1970s, followed by a worsening of inequality starting in the 1970s to the 1980s. This\nsection captures the rise in inequality in advanced economies starting in the 1980s. Later sections examine several\nchannels that might account for this more recent trend.\n5 Fabrizio et al. (2017) provide an overview of income inequality trends in low-income countries.\n\n3\n\nMention: \"the World Bank PovcalNet\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"It is the only comprehensive global set of non-imputed inequality and poverty estimates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001811:24:1", "occurrence_id": "ccad1e030ea31ff78bb2892e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \nstandard deviation increase in private sector credit to GDP is 39 percentage points, approximately\n\n\nequal to the growth experienced by Brazil from 1990 to 2020, or double the growth experienced by\n\n\nKenya during the same 30-year period.\n\n\nOur interpretation of these results is that capital controls and limited banking system depth have\n\n\nprevented capital from flowing to viable projects in less financially open and developed economies.\n\nWere this not the case, we would not observe a change in returns as economies open and develop. <sup>22</sup>\n\n\nOverall, these results are consistent with the idea that misallocation creates profit opportunities\n\n\nfor private equity investors. As financial systems open and become more competitive, “home run”\n\n\ninvestments are harder to find. Table A11 tests whether the effect of financial openness can be\n\n\nexplained by changes in perceived corruption, which may also have changed during this reform.\n\n\nThe direction of the result in Column (6) of Table 8 is robust to this control.\n\n\nIt is worth relating these findings to what is known about flows of private equity investments\n\n\nacross countries. Lerner et al. (2009) use data from Capital IQ to construct a database of 76,398\n\n\nprivate equity investments made in 1984-2008 across 123 countries. Although they do not measure\n\n\nreturns, they find associations between country characteristics and investment volume as a share\n\n\nof GDP. The authors report that private equity investment flows to countries with higher GDP per\n\n\ncapita and less corruption, although not to markets with less private sector credit to GDP. Although\n\n\ntheir paper examines a different sample of investors with preferences that potentially differ from\n\n\nthe IFC’s, one way to reconcile their results with ours is that the greatest volume of investment\n\n\nneed not necessarily go to markets with the highest returns; indeed, returns could be highest in\n\n\nmarkets with lower private sector credit to GDP precisely because there is less investment in those\n\n\nmarkets.\n\n\n21Table A8 reports results using the IRR as measure of performance. Table A9 reports results\n\nMention: \"data from Capital IQ\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"A cited study built a database of over seventy-six thousand investments across 123 countries from it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001813:19:0", "occurrence_id": "57c2e77090a3797272732c91", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 6: Trends in Students’ average Academic Performance by Cohort: 2010–2014\n\n\n\n\n\n\n\n\n\n_Source_ : Own calculations using the NUL students records data for the academic years 2010, 2011, 2012,\n2013, and 2014. _Notes_ : OWM is the overall weighted mean.\n\n\nbetween the FPE-treated and FPE-untreated groups (Abadie, 2005). In such a case, therefore, the\n\n\nFPE effect identified by the standard DID would be biased. It is, therefore, important to check\n\n\nwhether the pre-PFE characteristics are balanced between the control and treatment groups.\n\n\nTable 2 presents the pre-FPE compositional differences between the control and treatment\n\n\ngroups. Specifically, the table shows pre-FPE mean differences, by treatment group, in the main\n\n\noutcome variable (OWM) and the individual characteristics (APS, age, gender, and location).\n\n\nColumn (3) shows the _p−_ value of the _t−_ test for differences in mean characteristics of the control\n\n\nand treatment groups pre-FPE. Looking at these _p−_ values, we can see that there are statistic\n\nally significant differences between the control and treatment group with respect to their age,\n\n\npre-university (i.e. COSC) academic performance, gender and location (e.g. Berea and Botha\nBothe district). <sup>12</sup> Undoubtedly, these characteristics are associated with the changes in student’s\n\n\nacademic performance over time. For example, because early and current investments in human\n\n\n\n12Column (5) of the table presents the normalized difference in characteristics between groups. The normalized\ndifference between means is a scale-free measure of the difference in the distribution of characteristics by treatment\nstatus, and is given as Norm-diff _≡△_ **X** = _~~√~~_ _<u>X</u>\n\nMention: \"NUL students records data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Cohort trends in average academic performance are computed from these institutional records.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001814:27:4", "occurrence_id": "5c1195cfb7c0d3b861c5c10f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  9.203\n\n\nThis table shows the long-term effects of the program on a set of variables that result from aggregating groups of durable assets and\nhousehold characteristics. The reweighting strategy employed is the inverse probability weight (DiNardo et al., 1996). In Panel A observations are restricted to households included in 1997 ENCASEH (original households), while in Panel B observations are restricted\nto newly formed households in 2017 (new households) that were identified by searching for members no longer present in their original\n1997 households. The reweighting strategy seeks to reproduce the original sample distribution. Two different target distributions are set\nfollowing the attrition analysis done in _Table 3_ . _Original Distribution_ means the 1997 sample, which corresponds to the original households in the 506 localities. The purpose is to make our results comparable to the vast literature that focuses on this group. _Localities_\n_2017 distribution_ is based on the 334 localities chosen for the sample and assumes that all the original 1997 households in those localities\nwould have been found. The outcomes employed are: (1) _Transportation_, which adds the number of cars and vans; (2) _Entertainment_\n_equipment_, which gathers under entertainment equipment number of TV, radio and sound equipment; (3) _Kitchen supplies_, which sums the\nfollowing appliances: blender, refrigerator, electric grill, gas stove or microwave oven; (4) _Basic services_, which consists of the availability\nof the following services: firm floor, electricity, drainage and water access in the property; (5) _IT_ _items_, which refer to the number of\ncomputers and cellphones; and (6) _IT services_, which sums the home phone, internet and pay-TV; (7) _ln(imp cons)_, is the imputed measure\nof non-durable expenditure on food personal products, and clothing. Data employed comes from\n\nMention: \"1997 ENCASEH\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Original households are retained and reweighted by inverse probability to comparability with later years.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001783:4:1", "occurrence_id": "79259102ca943fa311e2c21c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  the establishment of the Rural Electrification Authority (REA) in\n2007; a separate transmission company, Kenya Electricity Transmission Company (KETRACO), in\n2008; and the Geothermal Development Company (GDC), also in 2008. The main institutions of\nKenya’s power sector are described further in box 1.\n\nIn the wake of these reforms, Kenya Vision 2030 was launched in 2008. A 22-year development plan,\nit sets out a path to Kenya becoming an industrializing, middle-income country by 2030. Much of the\nplan depends on the continued development of the energy sector, and power subsector, in order to\nsupport industrialization, the advancement of modern infrastructure, and socioeconomic\ndevelopment through increasing the access, availability and quality of supply. Some notable\nadvances have been made in this regard, including increasing electricity access to around 75\npercent, <sup>3</sup> moving to a position of ‘surplus’ capacity through a more resilient technology mix <sup>4</sup> with\ntotal generation now at 2,370MW, improved operational efficiency of the sector, and the successful\nattraction of private sector investment into independent power projects (of which there were 12 in\n2017). While some of these outcomes can be linked to sector reforms, others—including in\nparticular electrification and geothermal power development—depended critically on concerted\naction and significant funding from the government.\n\n\n3 Based on the World Bank’s Multi-Tier Framework Survey.\n4 In the 1990s, hydroelectric power plants accounted for almost 70 percent of installed capacity which had\nsevere implications in the drought prone region, especially in the critical drought of 1999-2001 where the\nreserve margin had been reduced to a point that regular load shedding was necessary.\n\n\n3\n\nMention: \"Multi-Tier Framework Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It supplies the measured access levels underlying the energy history.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001820:8:2", "occurrence_id": "07091515593473bec85475e7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  present value (PV) of lifetime earnings, we assume a working life of forty-five\nyears and a three percent discount rate for all groups. We use data for the total number of students,\n_S_ [, from UNESCO.](http://data.uis.unesco.org/)\n\nWe assume that some of the learning loss will be mitigated by distance learning measures.\nTherefore, we include _β_, an adjustment factor to allow for some distance learning during the period\nof school closure. We assume an optimistic level _β_, at 0.9, indicating that only 10 percent of\nstudents suffer learning loss due to the distance learning opportunities, a conservative estimate that\nassumes that having distance learning opportunities equates with learning. The actual learning\nlosses might be considerably higher. <sup>45</sup> Our estimate of distance learning coverage comes from\n<u>[UNESCO. Finally, we use World Bank data on GDP by country. We estimate total losses within](https://en.unesco.org/covid19/educationresponse)</u>\n<u>a range by adjusting labor force participation between 0.7 and 1.0.</u>\n\n\n7\n\nMention: \"World Bank data on GDP by country\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Country output levels convert estimated learning losses into total losses.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001821:17:0", "occurrence_id": "c9b97ccd8c20cf9730e6a9ea", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 6.5 Do private schools want to partner with the government? Within existing literature on education PPPs, the private sector’s motivation and reservations for\n\npartnering with the government have not received much attention. However, this aspect is critical to\nthe success of these partnerships.\n\nFor the 2015 Morogoro school survey, private school providers were asked about potential\n\n\n\n6.5 Do private schools want to partner with the government?\n\n\n\nthe success of these partnerships.\n\nFor the 2015 Morogoro school survey, private school providers were asked about potential\npartnership with the government. We randomize the delivery of these questions to better elicit\nunderlying beliefs. Private schools were randomly selected to be asked in one of two ways: in a direct\nand personalized way (e.g. what are _your_ views?) or in a generalized way (e.g. what are the _average_\n_private school owner’s_ views?). Of the 87 private schools, 46 schools (53 percent) were given\npersonalized questions and 41 schools (47 percent) were given generalized questions.\n\nThe personalized group expressed a higher likelihood of partnering with the government over the\n\n\n\npersonalized questions and 41 schools (47 percent) were given generalized questions.\n\nThe personalized group expressed a higher likelihood of partnering with the government over the\ngeneralized group by 14 percentage points (53 percent in personalized vs. 39 percent in generalized).\nOn the whole, slightly less than half (47 percent) of the private school owners who were interviewed\nwere willing to partner with the government. Corruption was the most cited issue that would\nthreaten the likelihood of a partnership, with 51 percent of all respondents reporting it as a major\nconcern. Among the issues of most importance for a partnership, private school owners rated clear\ncommunication channels and clarity of accountability as the most important criteria for partnership.\n#### 7. Parental Demand\n\nIn 2017, as part of its annual Sauti za Wananchi survey, TWAWEZA collected information\n\nMention: \"2015 Morogoro school survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Private providers were asked about their willingness to partner with government.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001823:5:1", "occurrence_id": "dc346a61cd94f84f9bfa4636", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  useful in other contexts, especially when the available sample is relatively\nsmall and contains a large number of highly-correlated potential explanatory variables.\n\nFinally, we contribute to a small existing literature that has used machine learning\nand other related methods to study the effects of trade agreements in a gravity context.\nFor example, Regmi and Baier (2020) use an unsupervised learning method to group\nPTAs by textual similarity, so as to provide a more nuanced notion of PTA depth.\nFollowing from a similar motivation, Hofmann, Osnago, and Ruta (2017) propose an\nearlier depth measure for PTAs based on principal components analysis applied to their\nprovisions data. In contrast, Baier, Yotov, and Zylkin (2019) use a two-step methodology\nwhere pair-specific PTA effects are estimated in a first stage and then predicted out of\nsample using country- and pair-specific variables.\n\nThe rest of this paper is structured as follows. Section 2 presents the data on PTA\nprovisions and provides a descriptive analysis of these data, highlighting a number of\nstylized facts about the provisions present in recent trade agreements. Section 3 introduces the variable selection problem in the three-way gravity model context and explains\nhow we implement PPML-lasso estimation with high-dimensional fixed effects. Section 4\npresents the results of a simulation study comparing the relative performance of different\nlasso methods in a simplified setting with high correlation between regressors. Section\n5 applies our methods to our database on PTA provisions and shows which individual\nprovisions are the strongest predictors of trade flows. Section 6 concludes and technical\ndetails are gathered in an Appendix.\n\n\n4\n\nMention: \"data on PTA\nprovisions\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Coded provisions are described and used as predictors of trade flows.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001823:5:1", "occurrence_id": "1bf41b74e778db193ed1361e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  useful in other contexts, especially when the available sample is relatively\nsmall and contains a large number of highly-correlated potential explanatory variables.\n\nFinally, we contribute to a small existing literature that has used machine learning\nand other related methods to study the effects of trade agreements in a gravity context.\nFor example, Regmi and Baier (2020) use an unsupervised learning method to group\nPTAs by textual similarity, so as to provide a more nuanced notion of PTA depth.\nFollowing from a similar motivation, Hofmann, Osnago, and Ruta (2017) propose an\nearlier depth measure for PTAs based on principal components analysis applied to their\nprovisions data. In contrast, Baier, Yotov, and Zylkin (2019) use a two-step methodology\nwhere pair-specific PTA effects are estimated in a first stage and then predicted out of\nsample using country- and pair-specific variables.\n\nThe rest of this paper is structured as follows. Section 2 presents the data on PTA\nprovisions and provides a descriptive analysis of these data, highlighting a number of\nstylized facts about the provisions present in recent trade agreements. Section 3 introduces the variable selection problem in the three-way gravity model context and explains\nhow we implement PPML-lasso estimation with high-dimensional fixed effects. Section 4\npresents the results of a simulation study comparing the relative performance of different\nlasso methods in a simplified setting with high correlation between regressors. Section\n5 applies our methods to our database on PTA provisions and shows which individual\nprovisions are the strongest predictors of trade flows. Section 6 concludes and technical\ndetails are gathered in an Appendix.\n\n\n4\n\nMention: \"database on PTA provisions\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Lasso methods are applied to identify which provisions predict trade.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001824:52:3", "occurrence_id": "623850fa0ead7252c7a2849b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: br>1,063<br>1,037<br>903<br>903<br>903<br>903<br>Y<br>Y<br>Y<br>Y<br>Y<br>Y<br>Y<br>Y<br>Y<br>Y<br>Y<br>Y<br>Y<br>Y<br>Y<br>Y<br>177<br>135<br>136<br>134<br>123<br>123<br>123<br>123|\n\n\nNote: Panel data estimation with country and year fixed effects. Annual information 2005‐2017. Dummy disaster loss is equal 1 if disaster loss is 1% of GDP or higher, 0 otherwise.\nData of disaster losses from GermanWatch. Controlled by GDP percapita\nRobust t‐statistics in parentheses\n*** p<0.01, ** p<0.05, * p<0.1\n\n\n51\n\nMention: \"GermanWatch\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Annual disaster losses define the dummy for losses of one percent of GDP or more.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001832:15:1", "occurrence_id": "5abfced85077315d92966409", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  was in place.\n\n\nData on tenure insecurity suggest that, with tenure of some 43% or 25% of agricultural parcels perceived to\n\n\nbe insecure or very insecure, levels of tenure insecurity on agricultural land are higher on state (52%) rather\n\n\nthan customary (42%) land but that the depth of insecurity is higher on the latter (25% on customary and\n\n\n23% on state land), i.e., insecurity may be more acute. Demand for title is 47% overall (47% on agricultural\n\n\nand 46.4% on residential parcels) among those who perceive their land rights as secure, in line with the\n\n\nnotion that, although titles can reduce insecurity, they perform other functions (e.g., reducing the cost of\n\n\ntransfers among strangers). Of those who perceive their rights as insecure, 51% of agricultural and 40% of\n\n\nresidential parcels would like to acquire title, suggesting that formal title is not perceived as a panacea to\n\n\ndeal with tenure insecurity. Substantively, this suggests that other measures to secure tenure might need to\n\n\nbe explored. Methodologically, it implies that, in situations where access to formal documents is incomplete,\n\n\nadding questions on title demand could greatly increase household surveys’ policy relevance.\n\n\n**4. Regression analysis and comparison to the Prindex global poll**\n\n\nRegressions for determinants of transferability, perceived tenure insecurity, and demand for title using LFS\n\n\ndata allow us to explore heterogeneity of outcomes, including whether having title is associated with greater\n\n\nfemale decision-making. As similar regressions for Prindex data produce very different results, we discuss\n\n\npotential reasons in terms of data quality and use these to identify ways to improve this approach to ensure\n\n\nbetter complementarity and avoid confusion.\n\n\n14\n\nMention: \"Prindex global poll\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Comparable regressions on it give very different results, raising data quality questions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001832:15:1", "occurrence_id": "e3e0fc14f660d2f5eaffbe8c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  was in place.\n\n\nData on tenure insecurity suggest that, with tenure of some 43% or 25% of agricultural parcels perceived to\n\n\nbe insecure or very insecure, levels of tenure insecurity on agricultural land are higher on state (52%) rather\n\n\nthan customary (42%) land but that the depth of insecurity is higher on the latter (25% on customary and\n\n\n23% on state land), i.e., insecurity may be more acute. Demand for title is 47% overall (47% on agricultural\n\n\nand 46.4% on residential parcels) among those who perceive their land rights as secure, in line with the\n\n\nnotion that, although titles can reduce insecurity, they perform other functions (e.g., reducing the cost of\n\n\ntransfers among strangers). Of those who perceive their rights as insecure, 51% of agricultural and 40% of\n\n\nresidential parcels would like to acquire title, suggesting that formal title is not perceived as a panacea to\n\n\ndeal with tenure insecurity. Substantively, this suggests that other measures to secure tenure might need to\n\n\nbe explored. Methodologically, it implies that, in situations where access to formal documents is incomplete,\n\n\nadding questions on title demand could greatly increase household surveys’ policy relevance.\n\n\n**4. Regression analysis and comparison to the Prindex global poll**\n\n\nRegressions for determinants of transferability, perceived tenure insecurity, and demand for title using LFS\n\n\ndata allow us to explore heterogeneity of outcomes, including whether having title is associated with greater\n\n\nfemale decision-making. As similar regressions for Prindex data produce very different results, we discuss\n\n\npotential reasons in terms of data quality and use these to identify ways to improve this approach to ensure\n\n\nbetter complementarity and avoid confusion.\n\n\n14\n\nMention: \"Prindex data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Divergent findings prompt discussion of how to improve this approach to measuring tenure insecurity.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001833:37:2", "occurrence_id": "b579c7dbb8a16857dc38a0d5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ** 𝑚 **𝑚** 𝑚 **𝑚** 𝑒𝑒𝑐𝑐,𝑡𝑡�, severance pay for\n\nredundancy dismissals (weeks of salary) �𝑠 **𝑠** 𝑠 **𝑠** 𝑠 **𝑠** 𝑠 **𝑠** 𝑒𝑒𝑐𝑐,𝑡𝑡�, and the share of women in wage\n\n**𝑤** **𝑤**\n\n\n\nindicators of labor market efficiency, we construct a composite index, using factor analysis, consisting of minimum wage (% of value added per worker) �𝑚 **𝑚** 𝑚 **𝑚** 𝑚 **𝑚** 𝑒𝑒𝑐𝑐,𝑡𝑡�, severance pay for\n\nredundancy dismissals (weeks of salary) �𝑠 **𝑠** 𝑠 **𝑠** 𝑠 **𝑠** 𝑠 **𝑠** 𝑒𝑒𝑐𝑐,𝑡𝑡�, and the share of women in wage\n\nemployment in the nonagricultural sector (𝑤 **𝑤** 𝑤 **𝑤** 𝑛𝑛𝑐𝑐,𝑡𝑡) from World Bank databases. The\n\n\n\n**𝑚** **𝑚** **𝑚**\n\nredundancy dismissals (weeks of salary) �𝑠 **𝑠** 𝑠 **𝑠** 𝑠 **𝑠** 𝑠 **𝑠** 𝑒𝑒𝑐𝑐,𝑡𝑡�, and the share of women in wage\n\nconstructed index is shown below.\n\nemployment in the nonagricultural sector (𝑤 **𝑤** 𝑤 **𝑤** 𝑛𝑛𝑐𝑐,𝑡𝑡) from World Bank databases. The\n\n**𝐸** **𝑏** **𝑏** **𝑏** **𝑓** **𝑓** **𝑓** **𝑓** **𝑙** **𝑙**\n\n\n\nconstructed index is shown below.\n\n**𝑤** **𝑤**\n\nMention: \"World Bank databases\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Female wage employment shares and dismissal cost measures are taken from these compilations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001835:4:1", "occurrence_id": "ab48243ac40b9951da49bdf7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  core management areas, explain relatively little in terms of\nvariation in overall firm performance, however, it does explain some of the regional differences between\nfirms in Russia’s Far East and the rest of Russia. Furthermore, they argue that exposure to foreign\ncompetition, both in the product market and the labor market, explains the variation in adoption of good\nmanagement practices across firms and broad regions.\n\n\nTo quantify the importance of management on the performance of Russian firms and identify potential\navenues for boosting firm productivity, during July-October 2018 we conducted a survey on managerial\n[practices in Russia using a slightly modified version of the US Census Management and Organizational](https://www.census.gov/programs-surveys/mops/technical-documentation/questionnaires.html)\n<u>[Practices Survey (MOPS). The survey was deliberately designed to replicate the US Census MOPS to](https://www.census.gov/programs-surveys/mops/technical-documentation/questionnaires.html)</u>\nmaximize comparability with the US as well as other countries where this survey has been conducted (e.g.\nMexico and Punjab province in Pakistan, UK, Canada, Japan, China and Croatia). In consultations with the\nRussian authorities, the survey collects data on 978 small-medium sized manufacturing firms ranging from\n\n\n1 Despite problems in identifying the actual role of the state sector in the economy, even by moderate assessment state-owned\nenterprises and state corporations present much more than 20% of Russia`s GDP (by some expert assessment up to 50%\nconsidering subsidiaries, affiliates, minority shares in different assets, etc.) and almost half of all internal procurements (Danilin,\n2015).\n2 The other 9 countries include - Belarus, Bulgaria, Kazakhstan, Lithuania, Poland, Romania, Serbia, Ukraine and Uzbekistan. The\nsurvey was modeled in a way that is similar\n\nMention: \"US Census Management and Organizational\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its questionnaire was modified to build the Russian manager survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001835:4:1", "occurrence_id": "c83a966e77142f80182f49c4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  core management areas, explain relatively little in terms of\nvariation in overall firm performance, however, it does explain some of the regional differences between\nfirms in Russia’s Far East and the rest of Russia. Furthermore, they argue that exposure to foreign\ncompetition, both in the product market and the labor market, explains the variation in adoption of good\nmanagement practices across firms and broad regions.\n\n\nTo quantify the importance of management on the performance of Russian firms and identify potential\navenues for boosting firm productivity, during July-October 2018 we conducted a survey on managerial\n[practices in Russia using a slightly modified version of the US Census Management and Organizational](https://www.census.gov/programs-surveys/mops/technical-documentation/questionnaires.html)\n<u>[Practices Survey (MOPS). The survey was deliberately designed to replicate the US Census MOPS to](https://www.census.gov/programs-surveys/mops/technical-documentation/questionnaires.html)</u>\nmaximize comparability with the US as well as other countries where this survey has been conducted (e.g.\nMexico and Punjab province in Pakistan, UK, Canada, Japan, China and Croatia). In consultations with the\nRussian authorities, the survey collects data on 978 small-medium sized manufacturing firms ranging from\n\n\n1 Despite problems in identifying the actual role of the state sector in the economy, even by moderate assessment state-owned\nenterprises and state corporations present much more than 20% of Russia`s GDP (by some expert assessment up to 50%\nconsidering subsidiaries, affiliates, minority shares in different assets, etc.) and almost half of all internal procurements (Danilin,\n2015).\n2 The other 9 countries include - Belarus, Bulgaria, Kazakhstan, Lithuania, Poland, Romania, Serbia, Ukraine and Uzbekistan. The\nsurvey was modeled in a way that is similar\n\nMention: \"US Census MOPS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Deliberate replication of it maximises comparability with the US and other countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001835:4:1", "occurrence_id": "4c2e0f6cdb3edd56157326a8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  core management areas, explain relatively little in terms of\nvariation in overall firm performance, however, it does explain some of the regional differences between\nfirms in Russia’s Far East and the rest of Russia. Furthermore, they argue that exposure to foreign\ncompetition, both in the product market and the labor market, explains the variation in adoption of good\nmanagement practices across firms and broad regions.\n\n\nTo quantify the importance of management on the performance of Russian firms and identify potential\navenues for boosting firm productivity, during July-October 2018 we conducted a survey on managerial\n[practices in Russia using a slightly modified version of the US Census Management and Organizational](https://www.census.gov/programs-surveys/mops/technical-documentation/questionnaires.html)\n<u>[Practices Survey (MOPS). The survey was deliberately designed to replicate the US Census MOPS to](https://www.census.gov/programs-surveys/mops/technical-documentation/questionnaires.html)</u>\nmaximize comparability with the US as well as other countries where this survey has been conducted (e.g.\nMexico and Punjab province in Pakistan, UK, Canada, Japan, China and Croatia). In consultations with the\nRussian authorities, the survey collects data on 978 small-medium sized manufacturing firms ranging from\n\n\n1 Despite problems in identifying the actual role of the state sector in the economy, even by moderate assessment state-owned\nenterprises and state corporations present much more than 20% of Russia`s GDP (by some expert assessment up to 50%\nconsidering subsidiaries, affiliates, minority shares in different assets, etc.) and almost half of all internal procurements (Danilin,\n2015).\n2 The other 9 countries include - Belarus, Bulgaria, Kazakhstan, Lithuania, Poland, Romania, Serbia, Ukraine and Uzbekistan. The\nsurvey was modeled in a way that is similar\n\nMention: \"data on 978 small-medium sized manufacturing firms\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Agreed with the authorities, the survey covers this sample of manufacturing plants.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001840:25:1", "occurrence_id": "5e89406168e9c3e3cfd2e631", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  allows private sector actors to digitally verify the identity of\nusers against the ID system.\n\n_4.4 Existence of laws vs. robustness of laws_\n\nThe details presented above on data regulation frameworks seek to capture the existence of foundational\nlaws, the robustness of these laws, and data governance administration. These additional details comprise\nthe value this diagnostic provides compared with efforts like UNCTAD’s e-transaction database and DLA\nPiper’s data protection law tracker. Besides collecting information on the existence of any e-commerce and\ndata protection laws, ATI, or open data legislation, the diagnostic also measures the robustness of such laws\nto assess whether they contain specific provisions or regulatory attributes. Moreover, in order to capture\ninformation on the implementation of those regulatory frameworks, the diagnostic also includes a few\nquestions on administrative aspects of data governance, such as any digital ID systems for e-services, key\ninstitutions such as CSIRTs for cybersecurity and data protection authorities, common technical standards\nto ensure government system interoperability, mandatory data classifications used across government\nsystems, and so on.\n\nThe diagnostic results show that although many countries have established various data regulations, those\nregulatory frameworks are not extensive enough or sufficiently aligned with good regulatory practices to\nensure that the legal framework can create an enabling environment for the data economy (figure 10). Lowincome countries in particular are weak on technical infrastructure or institutional settings that enable robust\ndata governance, such as the existence of digital ID system for e-services, mandatory use of common\ntechnical standards to ensure government system interoperability. In comparison, high-income countries\nshow a high score on the administrative aspects of data governance even though some legislation is not\nrobust enough with a full range of regulatory good practice incorporated. This reflects quite a pragmatic\napproach in exploring the most appropriate regulatory framework for their countries.\n\n\n[23 See art. 438 of the Digital Code Act, accessible at: https://sgg.gouv.bj/doc/loi-2017\n\nMention: \"data protection law tracker\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"validate\", \"summary\": \"It records only whether laws exist, so the diagnostic's added detail is greater.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001842:45:1", "occurrence_id": "9bb3cc280c3b03414e88f71c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  allocation proportional to the size of each stratum and an estimate\nof the within-strata variance of the indicators estimated in the survey. The number of\nobservations per stratum, _ni_, is then rounded up to the nearest integer. A cluster is defined\nas the combination of PCBS area code and economic sector. Clusters are randomly selected.\nWithin each cluster selected, we interviewed 1% or 10% of the establishments.\n\nData collection has been managed by Alpha International. In order to find establishments\nin the chosen cluster, enumerators were sent by field supervisors to areas believed to have a\nnumber of companies within the required sector. The field workers then searched for those\ncompanies on foot and interviewed whomever agreed. Furthermore, field supervisors also\nhelped enumerators by researching companies within the area that is part of the required\nsectors, made appointments for them, and sent them directly there to fill the questionnaire\non the first trial.\n\nThe targeted number of interviews is 90 for the ICT establishments (the number of\nestablishments in the 2017 census operating in the ICT is 1401) and 410 for the non-ICT, for\na total of 500 establishments, corresponding to 5% of all operating establishments interviewed\nby the PCBS as part of the Population Census 2017.\n\n\n4\n\nMention: \"2017 census\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its count of establishments operating in ICT sets the frame for the target sample.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001842:45:1", "occurrence_id": "b1be391603a3eb122a7b1f04", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  allocation proportional to the size of each stratum and an estimate\nof the within-strata variance of the indicators estimated in the survey. The number of\nobservations per stratum, _ni_, is then rounded up to the nearest integer. A cluster is defined\nas the combination of PCBS area code and economic sector. Clusters are randomly selected.\nWithin each cluster selected, we interviewed 1% or 10% of the establishments.\n\nData collection has been managed by Alpha International. In order to find establishments\nin the chosen cluster, enumerators were sent by field supervisors to areas believed to have a\nnumber of companies within the required sector. The field workers then searched for those\ncompanies on foot and interviewed whomever agreed. Furthermore, field supervisors also\nhelped enumerators by researching companies within the area that is part of the required\nsectors, made appointments for them, and sent them directly there to fill the questionnaire\non the first trial.\n\nThe targeted number of interviews is 90 for the ICT establishments (the number of\nestablishments in the 2017 census operating in the ICT is 1401) and 410 for the non-ICT, for\na total of 500 establishments, corresponding to 5% of all operating establishments interviewed\nby the PCBS as part of the Population Census 2017.\n\n\n4\n\nMention: \"Population Census 2017\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"The target establishment sample is expressed as a share of those it interviewed.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001838:23:2", "occurrence_id": "69268e84f9e88be82f7ce8d3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: /s), _f_ is the Coriolis parameter (1/sec),\n\nand _ϕ_ is the angle between the north and the storm heading.\n\n\nThe data source for the terrain roughness is the Global Land Cover Dataset 2000 and\n\nfor the topography characteristics is the Shuttle Radar Topography Mission database.\n\nThe method to estimate speed-up occurring in escarpments and ridges comes from the\n\nAmerican Society of Civil Engineers in 1994 and to retrieve the wind gusts factors comes\n\nfrom Vickery and Skerlj (2005). For more detailed information about the hurricane\n\nwindstorm model, see Pita et al. (2015).\n\n###### **S2 Implied monetary losses**\n\n\nTo convert the benefit of mangrove protection (mitigation of hurricane damage), mea\nsured in nightlights, into rough monetary values, we begin by calculating a conversion\n\nfactor between nightlights and economic activity as measured by Gross Domestic Prod\nuct (GDP). Using World Bank data on GDP in Parity Purchasing Power (PPP) in 2011\n\n\n22\n\nMention: \"Global Land Cover Dataset 2000\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It supplies the terrain roughness input to the wind speed-up calculation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001838:23:2", "occurrence_id": "9872bc3a73ee8081633a03bc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: /s), _f_ is the Coriolis parameter (1/sec),\n\nand _ϕ_ is the angle between the north and the storm heading.\n\n\nThe data source for the terrain roughness is the Global Land Cover Dataset 2000 and\n\nfor the topography characteristics is the Shuttle Radar Topography Mission database.\n\nThe method to estimate speed-up occurring in escarpments and ridges comes from the\n\nAmerican Society of Civil Engineers in 1994 and to retrieve the wind gusts factors comes\n\nfrom Vickery and Skerlj (2005). For more detailed information about the hurricane\n\nwindstorm model, see Pita et al. (2015).\n\n###### **S2 Implied monetary losses**\n\n\nTo convert the benefit of mangrove protection (mitigation of hurricane damage), mea\nsured in nightlights, into rough monetary values, we begin by calculating a conversion\n\nfactor between nightlights and economic activity as measured by Gross Domestic Prod\nuct (GDP). Using World Bank data on GDP in Parity Purchasing Power (PPP) in 2011\n\n\n22\n\nMention: \"Shuttle Radar Topography Mission database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Topographic characteristics are taken from this elevation dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001838:23:2", "occurrence_id": "779cc3cb674a0d7fd13db85a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: /s), _f_ is the Coriolis parameter (1/sec),\n\nand _ϕ_ is the angle between the north and the storm heading.\n\n\nThe data source for the terrain roughness is the Global Land Cover Dataset 2000 and\n\nfor the topography characteristics is the Shuttle Radar Topography Mission database.\n\nThe method to estimate speed-up occurring in escarpments and ridges comes from the\n\nAmerican Society of Civil Engineers in 1994 and to retrieve the wind gusts factors comes\n\nfrom Vickery and Skerlj (2005). For more detailed information about the hurricane\n\nwindstorm model, see Pita et al. (2015).\n\n###### **S2 Implied monetary losses**\n\n\nTo convert the benefit of mangrove protection (mitigation of hurricane damage), mea\nsured in nightlights, into rough monetary values, we begin by calculating a conversion\n\nfactor between nightlights and economic activity as measured by Gross Domestic Prod\nuct (GDP). Using World Bank data on GDP in Parity Purchasing Power (PPP) in 2011\n\n\n22\n\nMention: \"World Bank data on GDP\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"GDP in purchasing power parity is used to derive a conversion factor between nightlights and activity.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001844:4:0", "occurrence_id": "4b72b2c323bb4f5aa53e5635", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: consumers from costlessly shifting demand from one producer to another. As a result, “more\n\n\nefficient (lower cost) plants cannot lure away all demand from their less efficient industry rivals\n\n\nsimply with lower prices” (p. 534). This enables lower productivity plants to survive, even in long\n\n\nrun equilibrium, despite their lower productivity. Similarly, an implication of the Melitz‐\n\n\nOttaviano (2008) model is that high transport costs can reduce the scale of market selection via\n\n\nfirm entry and exit.\n\n\nAs outlined by Melitz‐Ottaviano (2008), the more differentiated a product is (in the sense of lower\n\n\nproduct substitutability), the lower is its industry (average and minimum) productivity and the\n\n\ngreater is the within‐industry dispersion of productivity. Any exogenous increase in market size\n\n\nwould therefore lead to growth in the average (or expected) plant‐level productivity and alter its\n\n\ndispersion relative to the industry average. The lower the degree of product differentiation, the\n\n\nlarger these changes will be because they are the outcome of market expansion that speeds up\n\n\nthe reallocation of market shares and factor inputs (both within the industry itself and between\n\n\nthe industry and the rest of the economy) by raising producers’ entry and exit rates to and from\n\n\nthat industry.\n\n\nWe test these predictions using a 10‐year panel of manufacturing firms from Ethiopia. <sup>1</sup> [^1: This panel is compiled from 10 years of the Ethiopian Survey of Large and Medium Scale Manufacturing Industries\nwhich, despite its name, is a census of all manufacturing firms with 10+ employees.] This panel\n\n\ncontains annual production data for the period 2000 to 2010 and covers all manufacturing firms\n\n\nemploying 10 or more employees. <sup>2</sup> [^2: Data from 2005 are dropped because a survey was conducted during that year rather than a census.] It is arguably the most comprehensive longitudinal data set\n\n\non manufacturing firms in Sub‐Saharan Africa at this moment. Industries are classified in\n\nMention: \"10‐year panel of manufacturing firms\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Entry and exit rates into the industry are measured to test the predictions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001848:24:0", "occurrence_id": "ae2ca63f8f9acb988711d9ca", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: where y <sup>c</sup> has been replaced by a function of the realized yields on other plots that share similar\ncharacteristics to the plot in question but did not incur any losses. <sup>32</sup> [^32: As a reminder, Yr and hac are always observed in the LSMS-ISA data. They are the realized production and GPS\nmeasured area, respectively.] Multiplying this by the planted\narea ha <sup>c</sup> thus gives the attainable production. Losses then constitute a deviation of the realized\nharvest Y <sup>r</sup> from what would have been expected based on past years or the performance of yields\non plots in unaffected areas, respectively.\n\nThe remainder of this section looks at the data requirements of both approaches as well as their\nlikely accuracy.\n\n\n<u>What do we see in the data?</u>\n\nWe start by regarding the suitability of using past yield data to proxy for attainable yields from a\ndata availability perspective. In order to ascertain a robust trend or average in yields, we require\nhigh frequency (e.g. annual) panel data reaching back several years. Micro-level panels of this\nfrequency are exceedingly rare and, in our case, the LSMS-ISA panel data only cover recurring\nsurvey rounds every two to three years. While such high frequency data is currently unavailable at\na sufficiently disaggregated level <sup>33</sup> [^33: Data sources such as FAOSTAT’s yield database, for example, contain yearly data but cover the country-level\nwhich may be too coarse to accurately accommodate sub-national, sub sub-national, or even micro-level variation.] it may become available in the future. <sup>34</sup> [^34: Projects such as the 50x2030 initiative are planning on collecting agricultural statistics in recurring patterns\nincluding a core of agricultural data at yearly frequency (50x2030 Initiative, 2020).]\n\nAnother challenge in this regard relates to split-off households and those\n\nMention: \"LSMS-ISA data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Realised production and GPS-measured plot area are always observed in it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001851:21:0", "occurrence_id": "7b91e5e1491c65d2e13a4698", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: component, there is no clear correlation between income and the contribution of health status to relative\n\nproductivity.\n\n\n**Figure 6. Contribution of health to relative productivity in Europe and Central Asia**\n\n\n_Note:_ Productivity figures are from the Europe and Central Asia extension of the Human Capital Index (ECA-HCI).\n\n\n**Table 7 Health component of the Europe and Central Asia extension of the Human Capital Index**\n**(ECA-HCI)**\n\n\n\n_Heavy_\n_episodic_\n_drinkers (%)_\n\n\n\n_Current_\n_smokers_\n\n_(%)_\n\n\n\n_Adult_\n_survival_\n\n_rate_\n\n\n\n_Children under_\n\n_5 not stunted_\n_(%)_\n\n\n\n_Health_\n_component,_\n\n_ECA-HCI_\n\n\n\n_Health_\n_component,_\n\n_HCI_\n\n\n\n_Subregion/country_\n\n\n\n_Obese adult_\n\n_population_\n\n_(%)_\n\n\n\n**Central Asia** **16.6** **11.5** **17.8** **0.859** **88.9** **0.941** **0.937**\n\nKazakhstan 21.3 19.9 24.3 0.845 92.0 0.928 0.937\n\nKyrgyz Republic 15.4 11.1 26.4 0.849 88.2 0.936 0.933\n\nTajikistan 12.6 7.9 18.8 0.871 82.5 0.942 0.930\n\nUzbekistan 15.3 7.9 12.3 0.866 89.2 0.949 0.939\n**Central Europe and Baltic**\n**15.9** **19.3** **27.3** **0.890**                          - **0.928** **0.933**\n**Countries**\n\nBulgaria 14\n\nMention: \"the Europe and Central Asia extension of the Human Capital Index (ECA-HCI)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its productivity figures and health component underlie the contribution analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001852:30:0", "occurrence_id": "cc2ddc7a8e22f4c66ffa6c69", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: SUPPLEMENT FIGURE III\nComparison to Item Response Theory (IRT) Linking (2000-2010)\n\n_Notes:_ We compare our data to the _Linking International Comparative Student Assessment_ (LINCS) project which uses Item-Response Theory\n(IRT) methods to link test score data. IRT methods are used to create scores for the underlying international and regional assessments used.\nHowever, to compare across assessments, IRT would require enough overlap in the test items. This is not true for a significant enough set of tests\nand time intervals to create a globally comparable panel data set. However, for a subset of tests, this is the case, such as a series of international\nstudies focused on reading at primary school. The LINCS project produces scores on a globally comparable scale using this subset of data\n(Steinmann, Strietholt **,** and Bos 2014). We compare results on the HLO as well as the LINCS data for primary school reading on average between\n2000-2010.\n\nSupplement Figure III compares scores for the same subject (reading), schooling level (primary)\nand time period (2000-2010), with a correlation of .984. This comparison indicates that as we\nexpand coverage to 164 countries, we maintain high consistency with alternative measures where\nthere is overlap.\n\nFinally, we compare our primary linking approach using regression and linear linking with two\nalternative approaches and compare robustness across them. First, we use simple mean linking\nwhich introduces a constant adjustment between tests matched with rounds, and which we average\nacross testing rounds. This approach assumes constant standard deviations across tests. Second,\nwe use a ratio between test means and also take an average across rounds. This approach assumes\na constant scalar adjustment 𝜆 between means and standard deviations across tests. The ratio\napproach is salient and intuitive for policymakers. However, a potential challenge in applying\nratios is that they are in principle sensitive to the scale of the test. For\n\nMention: \"HLO\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Reading scores are placed on a globally comparable scale and averaged across the decade.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001852:30:0", "occurrence_id": "e24de1863f1c2d10fbd59bcb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: SUPPLEMENT FIGURE III\nComparison to Item Response Theory (IRT) Linking (2000-2010)\n\n_Notes:_ We compare our data to the _Linking International Comparative Student Assessment_ (LINCS) project which uses Item-Response Theory\n(IRT) methods to link test score data. IRT methods are used to create scores for the underlying international and regional assessments used.\nHowever, to compare across assessments, IRT would require enough overlap in the test items. This is not true for a significant enough set of tests\nand time intervals to create a globally comparable panel data set. However, for a subset of tests, this is the case, such as a series of international\nstudies focused on reading at primary school. The LINCS project produces scores on a globally comparable scale using this subset of data\n(Steinmann, Strietholt **,** and Bos 2014). We compare results on the HLO as well as the LINCS data for primary school reading on average between\n2000-2010.\n\nSupplement Figure III compares scores for the same subject (reading), schooling level (primary)\nand time period (2000-2010), with a correlation of .984. This comparison indicates that as we\nexpand coverage to 164 countries, we maintain high consistency with alternative measures where\nthere is overlap.\n\nFinally, we compare our primary linking approach using regression and linear linking with two\nalternative approaches and compare robustness across them. First, we use simple mean linking\nwhich introduces a constant adjustment between tests matched with rounds, and which we average\nacross testing rounds. This approach assumes constant standard deviations across tests. Second,\nwe use a ratio between test means and also take an average across rounds. This approach assumes\na constant scalar adjustment 𝜆 between means and standard deviations across tests. The ratio\napproach is salient and intuitive for policymakers. However, a potential challenge in applying\nratios is that they are in principle sensitive to the scale of the test. For\n\nMention: \"LINCS data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its primary reading scores provide a cross-check on the harmonised estimates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001853:19:1", "occurrence_id": "776dcf21118002bb3f411c92", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  we set the speed for Class 3 to half of Class 2 for all slope categories. These speeds\nwere then confirmed via consultations with partners.\n\n\n**_Equation 3_**\n\n\n\nDb: Distance to bank\nDr: Distance to road\nTb: Time to bank\nTr: Time to road\n\n\n\n𝑉= <sup>Db −Dr</sup>\n\nTb −Tr\n\n\n\nWhere:\nV: estimated vehicle travel speed of\nthe household\n\n\n\nHowever, DRCN / SUR / UR / VR road types can mislead as they indicate the relative importance, not the actual\nsize or speediness, of a given highway segment. They may also reflect road future road development plans\ninstead of the current road state. For example, DRCNs in some areas are paved whilst in others they are\nimproved walking paths. The latter case is typical of mountainous areas and the hills and mountains of the Far\nWest region in particular.\n\n\nTherefore, some manual adjustments were made:\n\n\n  - All road segments above 2400m and all Far Western DRCNs in the Hills were assigned\nto class 4, walking paths, unless explicitly marked as paved within the RAI data set.\n\n  - Planned roads, VR and unpaved OSM highway segments tagged as smaller than\n“tertiary” were also assigned to class 4, paths.\n\n  - Similarly, SRNs in the Hill and Mountain areas of the Far West are usually rough dirt\nroads instead of paved highways. Therefore, they were assigned to class 3 instead of\nclass 1.\n\nMention: \"RAI data set\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Pavement markings in it determine whether a segment is classed as a path.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001854:14:0", "occurrence_id": "40507ed2387f50123063fde9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <u>Closing the data gap when surveys cannot be collected</u>\n\nMeasuring monetary poverty by survey-to-survey imputation <sup>12</sup> [^12: Here we focus on the methods which allow getting missing consumption at the household level, but there is\nanother method to get missing poverty rate at the aggregate level by using forecasting techniques. This method is\nused by the World Bank to track the corporate goal of reducing extreme poverty. The method and its shortcomings\nare discussed in the annex.]\n\nMany countries operate in an environment where household consumption data are partially missing. This\nincludes situations when: i) consumption data are available at one point in time but are not available in\nthe second point in time, or ii) situations when consumption data exist in all points in time but are not\ncomparable. “Imputation” is the key method to provide poverty estimates in the absence of consumption\ndata.\n\n\n<u>Method</u>\n\n\nMeasuring monetary poverty is possible only when information about consumption or income is available.\nWhen consumption data are not comparable across two survey rounds or do not exist in one of the\nperiods, (survey-to-survey) imputation methods can be used. The necessary condition is availability of at\nleast one HBS that includes consumption data. Key to the survey-to-survey imputation is the estimation\nof a model that explains consumption or income as a function of different social-economic and geographic\ncharacteristics for the same year consumption or income data are available. This model is then used to\npredict consumption in the second year of the survey, using the same explanatory variables. As a result,\neach household in the second year will get “imputed” consumption data, making it possible to construct\na poverty rate. Detailed technical reviews of the survey-to-survey imputation methodology can be found\nin Dang et al. (2014) and Dang et al. (2019).\n\n\nSeveral countries applied this methodology in the MENA region. Dang et al. (2004) used Jordan’s\nHousehold Income and Expenditure (HIES)\n\nMention: \"Household Income and Expenditure\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"A cited study applied the welfare comparison methodology to this Jordanian survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001858:3:0", "occurrence_id": "67882d7bd63af389ccbf8838", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ###### **1 Introduction**\n\nSatellite-recorded nighttime light data are used extensively as a proxy for economic\n\n\nactivity. However, surprisingly little economic analysis employs data from the Visible\n\nInfrared Imaging Radiometer Suite (VIIRS). <sup>1</sup> These new nighttime light data, with a\n\n\nbetter resolution and a higher frequency than the previous generation of data, have the\n\n\npotential to facilitate our understanding of rapid and spatially heterogeneous economic\n\nchanges, such as those during the COVID-19 pandemic. <sup>2</sup> One reason for the hesitancy\n\n\nto use these data in the economic literature may be the difficulty of converting changes\n\n\nin nighttime light intensity into changes in economic activity. To the best of our\n\n\nknowledge, no properly estimated and widely accepted quarterly elasticity between\n\nVIIRS nighttime lights and economic activity exists to date. <sup>3</sup> In this paper, we attempt\n\n\nto fill this gap.\n\n\nVIIRS nighttime light data are an imprecise measure of man-made lights. Even\n\n\nafter aggregating the data to the country level and to quarterly frequency, substantial\n\n\nstatistical noise remains. It mainly stems from atmospheric conditions like cloud cover\n\n\nthat impact the effective number of observations. For instance, there are only five\n\n\neffective observations at the pixel level on average each month for a median devel\n\noping country. Equally important, missing observations in the summer months and\n\n\noccasional satellite sensor recalibrations contribute to the noise as well.\n\n\nIn this paper, we provide a novel framework to estimate the elasticity between\n\n\nnighttime light intensity and gross domestic product (GDP) at quarterly frequency\n\n\nthat takes measurement errors of nighttime lights explicitly into account. The elas\n\nticity can be identified because countries exhibit varying noise in their nighttime light\n\n\ngrowth. Using information from the average number of effective observations, we pro\n\nvide a regression equation that estimates the elasticity precisely. In emerging markets\n\n\nand developing economies (EMDEs)\n\nMention: \"VIIRS nighttime light data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Even aggregated to country quarters it remains an imprecise measure with substantial statistical noise.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001859:7:0", "occurrence_id": "4429160c932d1949b0524f5e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: measurements and analysis of personality traits in empirical studies (Dal Bo, Finan, and Rossi,\n\n\n2013; Callen et al., 2015).\n\n\nThe insights from this paper are also relevant for various strands of the wider literature. In light of\n\n\nthe large debate about whether the worldwide increase in schooling is leading to measurable and\n\n\nsustained gains in learning (Pritchett and Beatty, 2015), having widely comparable measures of\n\n\ncognitive abilities, that can be measured for adults, outside of the classroom, and in a variety of\n\n\nsettings, is arguably key. Certain large data collection efforts covering wide and heterogeneous\n\n\npopulations, such as the Young Lives Surveys or the World Bank STEPS surveys (Pierre et al.,\n\n\n2014), are now including measures for noncognitive abilities and personality traits. Increasingly\n\n\nsuch measures are also included in targeted surveys for impact evaluation purposes, most directly\n\n\nwhen interventions aim to change noncognitive traits (Bernard et al., 2014; Groh, McKenzie, and\n\n\nVishwanath, 2015; Blattman, Jamison, and Sheridan, 2016; Ghosal et al., 2016; Adhvaryu, Kala,\n\n\nand Nyshadnam, 2016) but also when changes in noncognitive abilities are seen as potential\n\n\nmechanisms to explain changes in final outcomes (Blattman and Dercon, 2016). Other recent\n\n\nwork focuses on the long-term impact of external factors during childhood on noncognitive\n\n\noutcomes of adults (Leigh, Glewwe, and Park, 2015; Krutikova and Lilleor, 2015). Finally, there\n\n\nis a renewed interest and widening literature on learning, technology adoption and agricultural\n\n\nproductivity in developing countries (Jack, 2011; de Janvry, Sadoulet, and Suri, 2016), for which\n\n\nhaving reliable measures of agricultural knowledge and learning is key.\n\n\nThe paper is organized as follows: the next section provides more information about the context,\n\n\nthe instrument and the implementation of the survey experiment.\n\nMention: \"Young Lives Surveys\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"It is cited as a large heterogeneous-panel effort now including non-cognitive measures.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001859:7:0", "occurrence_id": "429af2b3836f59c942f0efac", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: measurements and analysis of personality traits in empirical studies (Dal Bo, Finan, and Rossi,\n\n\n2013; Callen et al., 2015).\n\n\nThe insights from this paper are also relevant for various strands of the wider literature. In light of\n\n\nthe large debate about whether the worldwide increase in schooling is leading to measurable and\n\n\nsustained gains in learning (Pritchett and Beatty, 2015), having widely comparable measures of\n\n\ncognitive abilities, that can be measured for adults, outside of the classroom, and in a variety of\n\n\nsettings, is arguably key. Certain large data collection efforts covering wide and heterogeneous\n\n\npopulations, such as the Young Lives Surveys or the World Bank STEPS surveys (Pierre et al.,\n\n\n2014), are now including measures for noncognitive abilities and personality traits. Increasingly\n\n\nsuch measures are also included in targeted surveys for impact evaluation purposes, most directly\n\n\nwhen interventions aim to change noncognitive traits (Bernard et al., 2014; Groh, McKenzie, and\n\n\nVishwanath, 2015; Blattman, Jamison, and Sheridan, 2016; Ghosal et al., 2016; Adhvaryu, Kala,\n\n\nand Nyshadnam, 2016) but also when changes in noncognitive abilities are seen as potential\n\n\nmechanisms to explain changes in final outcomes (Blattman and Dercon, 2016). Other recent\n\n\nwork focuses on the long-term impact of external factors during childhood on noncognitive\n\n\noutcomes of adults (Leigh, Glewwe, and Park, 2015; Krutikova and Lilleor, 2015). Finally, there\n\n\nis a renewed interest and widening literature on learning, technology adoption and agricultural\n\n\nproductivity in developing countries (Jack, 2011; de Janvry, Sadoulet, and Suri, 2016), for which\n\n\nhaving reliable measures of agricultural knowledge and learning is key.\n\n\nThe paper is organized as follows: the next section provides more information about the context,\n\n\nthe instrument and the implementation of the survey experiment.\n\nMention: \"World Bank STEPS surveys\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"Named alongside another effort as a source of personality and non-cognitive measures.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001861:11:0", "occurrence_id": "5bbbe2f4e5a515fe43f35a7c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Banking sector vulnerabilities could be compounded by a sluggish economy. Some countries have a\nnegative forecasted real GDP growth for year 2021 of more than 5 percent which may have repercussions\nfor bank resilience (Figure 2, Panel C). A weak macroeconomy may weigh negatively on the banking sector,\nsetting off a negative feedback loop between lackluster financial sector performance and a weakening real\neconomy.\n\nThe government fiscal position also has implications for the stability of banking systems, and vice versa.\nEconomies with weak public balance sheets may find it difficult to provide policy support and maintain a\ncredible backstop for the banking sector, which could also increase funding costs. In turn, weak banks may\nput sovereign debt sustainability in doubt and contribute to the bank-sovereign “doom loop” (Feyen &\nZuccardi, 2019). Some countries are forecasted to approach 100 percent general gross government debt to\nGDP (Figure 2, Panel D), suggesting potential future vulnerabilities arising from the bank-sovereign nexus.\n\nTo assess whether the CDBP correlates with other indicators of banking systems' stability, we run crosssectional regressions and relate the CDBP with a set of core financial soundness indicators (FSIs). <sup>13</sup> [^13: For an overview of the FSIs, see https://data.imf.org/?sk=51B096FA-2CD2-40C2-8D09-0699CC1764DA.] A\nsummary of the values of CDBP and FSIs across regions appears in Table 2. The Sub-Saharan Africa region\n(SSA) shows the highest un-weighted values of the CDBP, the regulatory Tier1 capital ratio, and liquid\nassets to short-term liabilities. South Asia (SAR) is the worst performing in terms of CDBP, non-performing\nloans ratio and exposure to changes in foreign exchange. Latin America and the Caribbean (LAC) has the\nlowest capital ratios, and MENA has the lowest liquidity ratios.\n\n**Table 2. Consolidated distance to breakpoint and selected FSI indicators (2020; unweighted median**\n**and median weighted by GDP at market prices)**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n**Un-**\n*\n\nMention: \"core financial soundness indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"The new index is related to these standard stability indicators in cross-section.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001862:7:0", "occurrence_id": "6a2e473fcb0c45e8965c2f3c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: expectations around education, marriage, employment, and occupation. Moreover, when it comes to actively\n\nparticipating in community-level decision making, women enjoy very limited agency and they are also less likely to\n\nhold positions of leadership. This can be mainly connected to the widespread belief that men’s opinions matter more\n\nthan women’s, the overwhelming weight of household work and family duties, the strength of the status quo and\n\nthe fear of transgressing gender norms. In addition, there is evidence that elected women officials continue to face\n\ndiscrimination in access to training, information, day-to-day interactions and how they are treated by male\n\ncolleagues (World Bank, 2020).\n\n\n**Ensuring that women entrepreneurs and employees gain from the opportunities presented by economic**\n\n**development must be a priority.** Development of gender-related support interventions necessitates an\n\nunderstanding of the differences between male and female entrepreneurs. This should include the sectors in which\n\nthey operate, location, the experience of social and physical barriers, flexible workplace policies, and whether\n\nperformance gaps exist between male and female entrepreneurial firms, including the drivers behind these gaps.\n\n\nThe aim of the note is to gain a better understanding of the barriers to gender-inclusive Private Sector Development\n\n(PSD) in Myanmar, explore the differential experience female-owned enterprises have in comparison to their male\nowned counterparts, and finally, to provide recommendations for inclusive PSD based on the findings. To achieve\n\nthis, the note uses data from the 2019 Myanmar Gender and Entrepreneurship Survey (MGES) to explore the\n\ndifferential experiences of female-owned micro, small and medium sized enterprises in comparison to their male\n\ncounterparts.\n\n\nTaking the above into account, this note covers and highlights the following issues:\n\n\n  - Characteristics of female and male owned enterprises\n\n  - Barriers faced by women entrepreneurs\n\n  - Determinants of Human Resource Management (HRM) policies\n\n  - Impact of HRM policies on business development\n\n  - Impact of HRM policies on workplace attitude/employee satisfaction and other areas\n\nMention: \"the 2019 Myanmar Gender and Entrepreneurship Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Differential experiences of female-owned micro, small and medium enterprises are explored from it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001867:5:2", "occurrence_id": "713779c9a74384abf7bd0903", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: DEA panel regression analysis). In\naddition, our analysis builds on the findings presented in this literature to inform our selection of\nvariables included as input factors in the DEA. Among the many variables that have been considered in\nthis literature, four economic factors are commonly identified as important determinants of revenue\npotential (see the survey in Gupta 2007): (i) per-capita income levels, (ii) sectoral composition of output,\n\n\n3 Specifically, the analysis is based on data from the UNU-WIDER Government Revenue Dataset which was launched in 2014 and\ncombines government revenue data from various sources (see Section IV for more details).\n4 DEA and regression analysis represent alternative approaches to performance assessments which each feature their own\nbenefits and weaknesses (for a detailed discussion, we refer to Thanassoulis 1993; Sickles and Zelenyuk 2019). Regression\nanalysis features more tools to identify causal relationships in observational data. However, it also involves stronger\nassumptions on the structure of the underlying data than non-parametric methods, making the results sensitive to\neconometric challenges such as endogeneity and omitted variable bias.\n\n\n4\n\nMention: \"UNU-WIDER Government Revenue Dataset\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The analysis rests on this harmonised compilation of revenue series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001867:5:2", "occurrence_id": "7d9d9be9b9826852e06cf589", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: DEA panel regression analysis). In\naddition, our analysis builds on the findings presented in this literature to inform our selection of\nvariables included as input factors in the DEA. Among the many variables that have been considered in\nthis literature, four economic factors are commonly identified as important determinants of revenue\npotential (see the survey in Gupta 2007): (i) per-capita income levels, (ii) sectoral composition of output,\n\n\n3 Specifically, the analysis is based on data from the UNU-WIDER Government Revenue Dataset which was launched in 2014 and\ncombines government revenue data from various sources (see Section IV for more details).\n4 DEA and regression analysis represent alternative approaches to performance assessments which each feature their own\nbenefits and weaknesses (for a detailed discussion, we refer to Thanassoulis 1993; Sickles and Zelenyuk 2019). Regression\nanalysis features more tools to identify causal relationships in observational data. However, it also involves stronger\nassumptions on the structure of the underlying data than non-parametric methods, making the results sensitive to\neconometric challenges such as endogeneity and omitted variable bias.\n\n\n4\n\nMention: \"government revenue data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Series from many different origins are combined into one comparable set.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001873:32:0", "occurrence_id": "36e4a655b3b8cb40a4e298b5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: epidemic, during which face-to-face data collection came to a standstill, we used 2019 as\nbenchmark year, implying that any censuses done after 2009 and surveys done after 2014 were\nconsidered up to date.\n\nOur findings are threefold. First price data are typically collected (often at a monthly basis), but\ncensus and survey data are often out of date. Only 14 out of 20 countries are current on their\npopulation census; 9 out of 20 are up to date on their economic census. Only 5 out of the 20\ncountries carried out an establishment survey recently and about half the countries are up to\ndate with respect to their health, labor force and consumption surveys (having been completed\nin 9, 11, and 14 countries respectively). The implication is that in almost half the cases, no or\noutdated microdata are used to produce core statistics including National Accounts and SDG\nreporting.\n\nOur second finding is that only in few instances where microdata has been collected, they are\nmade publicly accessible. Of the 140 potential microdata sets we looked for (7 data categories in\n20 countries) 81 had been collected and as few as 25 were accessible. Remarkably, of these 25\nabout a third are not accessible through the website of the NSO; they can only be downloaded\nfrom international microdata repositories. Our third finding is that recent microdata is scarce in\nMENA. Summary statistics are generally available –as evidenced by the Open Data Inventory\n(ODIN). However, many of these statistics are necessarily based on outdated microdata and\ndecision makers relying on such information would need to consider them with care.\n\nHence, we present a new indicator to complement the ODIN. The new indicator assesses the\ndegree of openness of recent microdata as well as the ease with which microdata users can\naccess this data on NSOs’ websites. Together with the ODIN, it gives a robust picture of data\nopenness in MENA. It is cost effective, takes less\n\nMention: \"Open Data Inventory\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It evidences that summary statistics are generally available even where recent microdata are scarce.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001876:13:1", "occurrence_id": "e57966876e9834b09d3f5337", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |‐19.5|‐8.4|‐5.7|\n|Tajikistan|‐32.2|‐13.1|‐2.7|‐23.2|\n|Turkmenistan|21.9|23.1|‐11.5|‐23.7|\n|Uzbekistan|‐37.1|‐38.3|‐20.1|‐14.1|\n||||||\n|Belarus|‐52.2|‐33.9|‐37.9|‐44.5|\n|Moldova|‐50.4|‐31.0|‐4.6|‐3.9|\n|Russian Federation|‐33.0|‐17.7|‐13.8|‐28.4|\n|Ukraine|‐55.6|‐25.0|‐20.7|‐34.7|\n||||||\n|FSU countries*|‐37.6|‐21.8|‐15.2|‐24.9|\n|Remittance receiving FSU countries|‐30.0|‐14.7|‐6.3|‐10.7|\n|FSU Hydrocarbon exporters|‐38.8|‐25.8|‐18.6|‐33.1|\n||||||\n\n\n\n*Excludes the three Baltic States\n\n\nA positive number denotes an under-valuation of the RER, a negative number an over-valuation\n\n\nSource: authors’ calculations using data from WDI\n\n\nThe estimated BSA RER over or under-valuations for each of the eight FSU countries in CASC\nalong with other FSU countries are reported in table 8. <sup>5</sup> Each of the four remittance receiving\ncountries in CAS\n\nMention: \"WDI\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Real exchange rate over- and under-valuations are calculated from these indicators.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001889:24:1", "occurrence_id": "c8a2f8c1d6bd73cc1396a0da", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  interventions improved teachers’ giving attitudes toward parents._\n\n\n**7.4** **Indirect effects on student performance**\n\nTo complete our analysis of the indirect impacts of our interventions, we estimate the treatment\n\neffects on various measures of student performance. We employ four sets of outcome variables. First,\n\nwe look at parents’ satisfaction with their children’s education (column 1). Then, we examine the\n\npercentage of students repeating a grade within each school, using data provided by the ministry of\n\neducation (column 2). Subsequently, we present results for student absenteeism (column 3), an outcome\n\nvariable that is based on our recording of the absence of a randomly selected sample of students at\n\nendline during an unannounced school visit. Finally, we examine student performance in standardized\n\nPortuguese and Math tests that we administered in all schools to students enrolled in grades 3 and 4. <sup>40</sup>\n\nWe employ specifications with full controls like in the previous tables. <sup>41</sup>\n\n\n38 Figure A5 in the Appendix shows the full distribution of teachers’ giving in the dictator game across treatments.\n39 Similar results are obtained when including block fixed effects, as shown in Table A11 in the Appendix.\n40 In schools offering grade 5, we also administered the tests to grade 5 students. However, since 25 percent of our schools did\nnot offer grade 5, we restrict the analysis to the performance of students in grades 3 and 4.\n41 For the standardized tests, since we have student performance at baseline (even though not for the same students), we also\nestimate difference-in-difference regressions. The results of these estimations provide similar results to the ones presented in\nTable 6, i.e., a null impact of the treatments of student performance in the tests.\n\n\n23\n\nMention: \"the ministry of\n\neducation\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Grade repetition rates within each school are taken from administrative records.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001895:3:0", "occurrence_id": "24caf679ecadcb463a6ff8a4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ##### **I. INTRODUCTION**\n\nFossil fuel exporters are exposed to the vagaries of fossil fuel markets. The collapse in oil prices\n\n\nthat started in June 2014 is a stark reminder about the challenges posed by the dependence on oil\n\n\nand other fossil fuels (Chart 1). While the literature on appropriate macroeconomic policies for\n\n\nfossil fuel exporters is extensive, much more attention has been paid to the role of fiscal policy.\n\n\nPart of the reason why monetary policy has been less subject to attention may have to do with the\n\n\nfact that most fossil fuel exporters have pegs or relatively fixed exchange rate regimes and hence\n\n\nhave no independent monetary policy. There are however good reasons to take a fresh look at the\n\n\nissue of monetary policy in fossil fuel exporters. Traditionally, the horizon of monetary policy has\n\n\nbeen limited to that of the business cycle, typically 2-6 years. Considering the degree of wealth\n\n\nconcentration, the strong complementarity between fiscal and monetary policies, and the\n\n\nemergence of new risks to fossil fuel assets, there is a need to rethink monetary policy in fossil\n\n\nfuel exporters. In the present paper, I specifically examine the role monetary policy should play at\n\n\ndifferent horizons.\n\n\n**Chart 1 – 2014‐2016 Oil Price Slump**\n\n\n**Crude Oil Price (APSP)**\n_U.S. dollars a barrel_\n\n\n120\n\n\n100\n\n\n80\n\n\n60\n\n\n40\n\n\n20\n\n\n0\n\nJan-14 Apr-14 Jul-14 Oct-14 Jan-15 Apr-15 Jul-15 Oct-15 Jan-16 Apr-16 Jul-16 Oct-16\n\n\nSource: IMF, Primary Commodity Price System.\nNote: APSP = average petroleum spot price—average of U.K. Brent, Dubai, and West Texas\nIntermediate, equally weighted.\n\n\n2\n\nMention: \"Primary Commodity Price System\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Daily petroleum spot prices averaged across three benchmarks are plotted from this system.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001897:32:1", "occurrence_id": "bbbb4b5e9b9447602060ae9e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  buildings and building materials) as a function of the number of floors of a building.\n\n\nFinally, buildings are supposed to be built for a century which gives us a depreciation rate 𝜌 of 0.01.\nThe interest rate 𝑖 which provides the cost of capital is set at a value of 5%.\n\n\n16 This figure is very close to the increase in the yearly income by inhabitant of the Ile de France compared to the\naverage in France (INSEE).\n17 The net budget of a household must be understood as the income minus transport costs.\n\n\n31\n\nMention: \"INSEE\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its figures on per-capita income in the region relative to the national average are cited.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001866:2:0", "occurrence_id": "4e98cecc7dc27ff16676354f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ### Measuring Total Factor Productivity Using the Enterprise Surveys: A Methodological Note ∗\n\nDavid C. Francis Nona Karalashvili Hibret Maemir\n\nJorge Rodriguez Meza <sup>_†_</sup>\n\n\n_Keywords_ : TFP, Enterprise Surveys.\n\n\n_JEL_ _Classification_ : 012\n\n\n_∗_ This note uses the dataset “Firm Level TFP Estimates and Factor Ratios ~~S~~ eptember ~~1~~ 0 ~~2~~ 020.dta” which is published\nat `www.enterprisesurvyes.org` in the Firm-Level Datasets for Researchers section. The estimation procedure of this note\nuses the full cross-country ES data, which is continuously being updated as new data becomes available. Every month\nestimates are updated, and older datasets are archived and available upon request.\n\n_†_ The authors are from the Enterprise Analysis Unit of the World Bank. We would like to thank Arvind Jain and Rita\nRamalho for very helpful comments. The findings, interpretations, and conclusions expressed in this paper are entirely\nthose of the authors. They do not necessarily represent the views of the World Bank and its affiliated organizations, or\nthose of the Executive Directors of the World Bank or the governments they represent.\n\nMention: \"Enterprise Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The note's total factor productivity method is built on these firm surveys.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001866:2:0", "occurrence_id": "d17f7b71fbfe66bc6119ef1d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ### Measuring Total Factor Productivity Using the Enterprise Surveys: A Methodological Note ∗\n\nDavid C. Francis Nona Karalashvili Hibret Maemir\n\nJorge Rodriguez Meza <sup>_†_</sup>\n\n\n_Keywords_ : TFP, Enterprise Surveys.\n\n\n_JEL_ _Classification_ : 012\n\n\n_∗_ This note uses the dataset “Firm Level TFP Estimates and Factor Ratios ~~S~~ eptember ~~1~~ 0 ~~2~~ 020.dta” which is published\nat `www.enterprisesurvyes.org` in the Firm-Level Datasets for Researchers section. The estimation procedure of this note\nuses the full cross-country ES data, which is continuously being updated as new data becomes available. Every month\nestimates are updated, and older datasets are archived and available upon request.\n\n_†_ The authors are from the Enterprise Analysis Unit of the World Bank. We would like to thank Arvind Jain and Rita\nRamalho for very helpful comments. The findings, interpretations, and conclusions expressed in this paper are entirely\nthose of the authors. They do not necessarily represent the views of the World Bank and its affiliated organizations, or\nthose of the Executive Directors of the World Bank or the governments they represent.\n\nMention: \"Firm Level TFP Estimates and Factor Ratios\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"This published firm-level file is the dataset the note runs on.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001866:2:0", "occurrence_id": "e5fa4d8cad1f6a971c4b5e98", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ### Measuring Total Factor Productivity Using the Enterprise Surveys: A Methodological Note ∗\n\nDavid C. Francis Nona Karalashvili Hibret Maemir\n\nJorge Rodriguez Meza <sup>_†_</sup>\n\n\n_Keywords_ : TFP, Enterprise Surveys.\n\n\n_JEL_ _Classification_ : 012\n\n\n_∗_ This note uses the dataset “Firm Level TFP Estimates and Factor Ratios ~~S~~ eptember ~~1~~ 0 ~~2~~ 020.dta” which is published\nat `www.enterprisesurvyes.org` in the Firm-Level Datasets for Researchers section. The estimation procedure of this note\nuses the full cross-country ES data, which is continuously being updated as new data becomes available. Every month\nestimates are updated, and older datasets are archived and available upon request.\n\n_†_ The authors are from the Enterprise Analysis Unit of the World Bank. We would like to thank Arvind Jain and Rita\nRamalho for very helpful comments. The findings, interpretations, and conclusions expressed in this paper are entirely\nthose of the authors. They do not necessarily represent the views of the World Bank and its affiliated organizations, or\nthose of the Executive Directors of the World Bank or the governments they represent.\n\nMention: \"cross-country ES data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"It is continuously updated monthly, with older datasets archived.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001868:7:1", "occurrence_id": "977b495575561e537d45af63", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: up>_−_</sup> <sup>1</sup> **E** _sr/_ **u** _N_ **E** _sr,_ (1)\n\n\n\nwhere **u** _N_ is the 1 _×N_ unit row vector and ( **I** _−_ **A** _ss_ ) <sup>_−_</sup> <sup>1</sup> represents the domestic\n\ninverse Leontief matrix.\n\n\nAlthough the **VS** indicator provides valuable information for several empirical applications, <sup>4</sup> there are relevant aspects that are not taken into account. In\n\nparticular, imported inputs in exports are considered as a single category, with\nout distinguishing between the part that originated abroad and the part that was\n\noriginally produced by _s_ itself and then re-imported.\n\n\nIndeed, by exploiting Inter-Country Input-Output tables, gross exports can\n\nbe broken down according to the country that initially produced each component.\n\nKoopman et al. (2010) use the global Leontief inverse matrix to trace back the total\n\ngross output produced by each country _j_ to deliver one unit of country _s_ exports\n\n( **B** _js_ ), and the related value-added shares ( **V** _j_ ). The part that originated in country\n\n_s_ is referred to as the ‘domestic content of exports’ ( **DC** _sr_ ), whereas the remaining\n\n\n3Hummels et al. (2001) measures can be computed without resorting to an Inter-Country InputOutput database since they were developed for national Input-Output tables that distinguish only\nbetween imported and domestic inputs.\n\n4In particular, later in the paper we\n\nMention: \"Inter-Country Input-Output tables\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Gross exports are decomposed by the country that initially produced each component.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001881:12:1", "occurrence_id": "e3d48e9e012a57b3338bcfb4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  over the tax system across\n\ngovernment tiers. It was constructed through in-depth reviews of legal and policy documents,\n\nscientific and grey literature, as well as fiscal archives from the International Bureau of Fiscal\n\nDocumentation (IBRD, Access: 2015-2017), which inform on the discretionary power of central\n\nand local governments over deciding on tax-related matters. Based on the legal and policy\n\ninformation and using a pre-defined matrix, the discretionary power of all government tiers is\n\ncoded for each significant revenue instrument and across four decision components which refer\n\nto the legal ability of local and central authorities to _(i)_ introduce new instruments or altering\n\nexisting ones, _(ii)_ define the base, _(iii)_ set the rates and _(iv)_ collect and administer the revenues.\n\n\nFor each decision component, a score is calculated by taking the ratio of subnational\n\ngovernments’ involvement across the number of identified revenue instruments. An overall\n\nindex is then computed by taking the average of scores on the four decision components. This\n\naggregated indicator reflects the overall discretionary power of subnational governments over\n\nthe fiscal space or their level of taxing rights in each country. Given that natural resources\n\nextraction generates revenues for local authorities, either directly through extraction royalties or\n\n\n1 See Vincent (2020).\n\n\n11\n\nMention: \"fiscal archives from the International Bureau of Fiscal\n\nDocumentation\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its archives inform the coding of discretionary power over tax matters across government tiers.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001890:28:0", "occurrence_id": "d37e5bcec6ee3f66207b78f2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 10: Association Between Spending and Enrollment-Adjusted Learning**\n_Expenditure Per Child and Learning-Adjusted Years of Schooling (LAYS), 1991_ **_–_** _2015_\n\n\n\n_b. Countries spending less than $2,500 per child_\n\n_(constant 2015 PPP dollars)_\n\n14\n\n\n\n12\n\n\n10\n\n\n8\n\n\n6\n\n\n4\n\n\n2\n\n\n0\n\n\n\n14\n\n\n12\n\n\n10\n\n\n8\n\n\n6\n\n\n4\n\n\n2\n\n\n0\n\n\n\n_a. All countries_\n_(constant 2015 PPP dollars)_\n\n\n\nSpending per child ($)\n\n\n\n\n\n\n\n\n\n\n\nSpending per child ($)\n\n\n\n_Source:_ World Bank calculations based on HCI, UIS and IMF data.\n_Note:_ Calculations only incorporate countries with available data.\n\nThere is a clear relationship between our measures of spending and outcomes, which change with the level\nof spending. Figure 10 plots our measures of spending and outcomes averaged for each country within fiveyear periods (1991 **–** 1995, 1996 **–** 2000, 2001 **–** 05, 2006 **–** 10, and 2011 **–** 15) and includes a stochastic frontier\ncurve that estimates the maximum learning-adjusted years of schooling that is expected for a given level of\nspending per child. <sup>14</sup> [^14: We use Stata’s Frontier command to estimate the stochastic frontier and assume that the production frontier is\nlinear in logs.] The frontier does not estimate the relationship between learning-adjusted years of\nschooling and spending per child, but the vertical distance from a particular country observation to the\nfrontier provides a measure of how well the country is able to translate spending per child into education\noutcomes. <sup>15</sup> [^15: We also conducted Ordinary Least Squares regressions of LAYS on public education spending per child and\nfound a positive and statistically significant relationship, both with and without a control for gross domestic product\nper capita.] However, there are large differences among countries at similar levels of spending, with some\ncountries getting significantly better outcomes for the same (\n\nMention: \"HCI\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Its outcome measures enter the spending-per-child comparison.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001890:28:0", "occurrence_id": "b90d7f3b3a1a121df0bd4569", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 10: Association Between Spending and Enrollment-Adjusted Learning**\n_Expenditure Per Child and Learning-Adjusted Years of Schooling (LAYS), 1991_ **_–_** _2015_\n\n\n\n_b. Countries spending less than $2,500 per child_\n\n_(constant 2015 PPP dollars)_\n\n14\n\n\n\n12\n\n\n10\n\n\n8\n\n\n6\n\n\n4\n\n\n2\n\n\n0\n\n\n\n14\n\n\n12\n\n\n10\n\n\n8\n\n\n6\n\n\n4\n\n\n2\n\n\n0\n\n\n\n_a. All countries_\n_(constant 2015 PPP dollars)_\n\n\n\nSpending per child ($)\n\n\n\n\n\n\n\n\n\n\n\nSpending per child ($)\n\n\n\n_Source:_ World Bank calculations based on HCI, UIS and IMF data.\n_Note:_ Calculations only incorporate countries with available data.\n\nThere is a clear relationship between our measures of spending and outcomes, which change with the level\nof spending. Figure 10 plots our measures of spending and outcomes averaged for each country within fiveyear periods (1991 **–** 1995, 1996 **–** 2000, 2001 **–** 05, 2006 **–** 10, and 2011 **–** 15) and includes a stochastic frontier\ncurve that estimates the maximum learning-adjusted years of schooling that is expected for a given level of\nspending per child. <sup>14</sup> [^14: We use Stata’s Frontier command to estimate the stochastic frontier and assume that the production frontier is\nlinear in logs.] The frontier does not estimate the relationship between learning-adjusted years of\nschooling and spending per child, but the vertical distance from a particular country observation to the\nfrontier provides a measure of how well the country is able to translate spending per child into education\noutcomes. <sup>15</sup> [^15: We also conducted Ordinary Least Squares regressions of LAYS on public education spending per child and\nfound a positive and statistically significant relationship, both with and without a control for gross domestic product\nper capita.] However, there are large differences among countries at similar levels of spending, with some\ncountries getting significantly better outcomes for the same (\n\nMention: \"UIS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its education statistics complement the index in the calculation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001890:28:0", "occurrence_id": "c2ed71dd2b75190260c824c1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 10: Association Between Spending and Enrollment-Adjusted Learning**\n_Expenditure Per Child and Learning-Adjusted Years of Schooling (LAYS), 1991_ **_–_** _2015_\n\n\n\n_b. Countries spending less than $2,500 per child_\n\n_(constant 2015 PPP dollars)_\n\n14\n\n\n\n12\n\n\n10\n\n\n8\n\n\n6\n\n\n4\n\n\n2\n\n\n0\n\n\n\n14\n\n\n12\n\n\n10\n\n\n8\n\n\n6\n\n\n4\n\n\n2\n\n\n0\n\n\n\n_a. All countries_\n_(constant 2015 PPP dollars)_\n\n\n\nSpending per child ($)\n\n\n\n\n\n\n\n\n\n\n\nSpending per child ($)\n\n\n\n_Source:_ World Bank calculations based on HCI, UIS and IMF data.\n_Note:_ Calculations only incorporate countries with available data.\n\nThere is a clear relationship between our measures of spending and outcomes, which change with the level\nof spending. Figure 10 plots our measures of spending and outcomes averaged for each country within fiveyear periods (1991 **–** 1995, 1996 **–** 2000, 2001 **–** 05, 2006 **–** 10, and 2011 **–** 15) and includes a stochastic frontier\ncurve that estimates the maximum learning-adjusted years of schooling that is expected for a given level of\nspending per child. <sup>14</sup> [^14: We use Stata’s Frontier command to estimate the stochastic frontier and assume that the production frontier is\nlinear in logs.] The frontier does not estimate the relationship between learning-adjusted years of\nschooling and spending per child, but the vertical distance from a particular country observation to the\nfrontier provides a measure of how well the country is able to translate spending per child into education\noutcomes. <sup>15</sup> [^15: We also conducted Ordinary Least Squares regressions of LAYS on public education spending per child and\nfound a positive and statistically significant relationship, both with and without a control for gross domestic product\nper capita.] However, there are large differences among countries at similar levels of spending, with some\ncountries getting significantly better outcomes for the same (\n\nMention: \"IMF data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Public finance figures complete the spending side of the calculation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001905:7:0", "occurrence_id": "a2f0b680d22a4c6b7b571a30", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **2. LSMS+: Land Rights and Decision-Making Measures in Defining Ownership**\n\n**2.1 Multidimensionality of Ownership**\n\nThis paper uses three nationally representative, multi-topic household surveys supported by the\nLSMS+ program, namely the Tanzania National Panel Survey (NPS4 2018/2019), Ethiopia\nSocioeconomic Survey (ESS4 2018/2019), and the Malawi Integrated Household Panel Survey\n(IHPS 2016). The LSMS+ modules on asset ownership and rights are comparable across countries\nand were administered directly to all household members 18 and older on different asset classes,\nincluding ownership and rights to residential and non-residential land (Hasanbasri et al., 2021). <sup>10</sup> [^10: Other types of assets were also covered, including financial accounts, and mobile phones. Ownership of livestock\nwas covered in Ethiopia.]\nA specific emphasis of the survey data collection was on self-reporting and interviewing\nrespondents privately — and hence conducting interviews within the household simultaneously,\nwhen possible. <sup>11</sup> [^11: The module on land specifically covers all land owned or accessed via use rights and follows the recommendations\nof FAO, World Bank and UN Habitat (2019).]\n\nThis paper focuses on characterizing landowners’ rights and decision-making over non-residential\nland parcels. A respondent is defined as the reported landowner if they answer upfront that he/she\nowns the parcel. A _parcel_ is defined as a continuous piece of land which can have more than one\nparcel. Parcels were first identified and rostered through the household questionnaire and then\ncarried forward to individual interviews. Non-residential land in these contexts is mainly used for\nagriculture. In Ethiopia, for example, 87 percent of non-residential parcels have been used in\nagriculture in the last 12 months. <sup>12</sup> [^12: Ethiopia and Malawi looked at all non-residential land used for agriculture. Tanzania was not restricted to\nagricultural use only due to missing data on current use.]\n\nRights and decision-making questions were\n\nMention: \"Tanzania National Panel Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Its LSMS+ ownership and rights modules, asked of all household members, measure ownership extensionality.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001905:7:0", "occurrence_id": "47f37d7cd35192694f125984", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **2. LSMS+: Land Rights and Decision-Making Measures in Defining Ownership**\n\n**2.1 Multidimensionality of Ownership**\n\nThis paper uses three nationally representative, multi-topic household surveys supported by the\nLSMS+ program, namely the Tanzania National Panel Survey (NPS4 2018/2019), Ethiopia\nSocioeconomic Survey (ESS4 2018/2019), and the Malawi Integrated Household Panel Survey\n(IHPS 2016). The LSMS+ modules on asset ownership and rights are comparable across countries\nand were administered directly to all household members 18 and older on different asset classes,\nincluding ownership and rights to residential and non-residential land (Hasanbasri et al., 2021). <sup>10</sup> [^10: Other types of assets were also covered, including financial accounts, and mobile phones. Ownership of livestock\nwas covered in Ethiopia.]\nA specific emphasis of the survey data collection was on self-reporting and interviewing\nrespondents privately — and hence conducting interviews within the household simultaneously,\nwhen possible. <sup>11</sup> [^11: The module on land specifically covers all land owned or accessed via use rights and follows the recommendations\nof FAO, World Bank and UN Habitat (2019).]\n\nThis paper focuses on characterizing landowners’ rights and decision-making over non-residential\nland parcels. A respondent is defined as the reported landowner if they answer upfront that he/she\nowns the parcel. A _parcel_ is defined as a continuous piece of land which can have more than one\nparcel. Parcels were first identified and rostered through the household questionnaire and then\ncarried forward to individual interviews. Non-residential land in these contexts is mainly used for\nagriculture. In Ethiopia, for example, 87 percent of non-residential parcels have been used in\nagriculture in the last 12 months. <sup>12</sup> [^12: Ethiopia and Malawi looked at all non-residential land used for agriculture. Tanzania was not restricted to\nagricultural use only due to missing data on current use.]\n\nRights and decision-making questions were\n\nMention: \"Ethiopia\nSocioeconomic Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Its comparable ownership module supplies the Ethiopian estimates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001905:7:0", "occurrence_id": "268a9a5925faf88682a37ade", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **2. LSMS+: Land Rights and Decision-Making Measures in Defining Ownership**\n\n**2.1 Multidimensionality of Ownership**\n\nThis paper uses three nationally representative, multi-topic household surveys supported by the\nLSMS+ program, namely the Tanzania National Panel Survey (NPS4 2018/2019), Ethiopia\nSocioeconomic Survey (ESS4 2018/2019), and the Malawi Integrated Household Panel Survey\n(IHPS 2016). The LSMS+ modules on asset ownership and rights are comparable across countries\nand were administered directly to all household members 18 and older on different asset classes,\nincluding ownership and rights to residential and non-residential land (Hasanbasri et al., 2021). <sup>10</sup> [^10: Other types of assets were also covered, including financial accounts, and mobile phones. Ownership of livestock\nwas covered in Ethiopia.]\nA specific emphasis of the survey data collection was on self-reporting and interviewing\nrespondents privately — and hence conducting interviews within the household simultaneously,\nwhen possible. <sup>11</sup> [^11: The module on land specifically covers all land owned or accessed via use rights and follows the recommendations\nof FAO, World Bank and UN Habitat (2019).]\n\nThis paper focuses on characterizing landowners’ rights and decision-making over non-residential\nland parcels. A respondent is defined as the reported landowner if they answer upfront that he/she\nowns the parcel. A _parcel_ is defined as a continuous piece of land which can have more than one\nparcel. Parcels were first identified and rostered through the household questionnaire and then\ncarried forward to individual interviews. Non-residential land in these contexts is mainly used for\nagriculture. In Ethiopia, for example, 87 percent of non-residential parcels have been used in\nagriculture in the last 12 months. <sup>12</sup> [^12: Ethiopia and Malawi looked at all non-residential land used for agriculture. Tanzania was not restricted to\nagricultural use only due to missing data on current use.]\n\nRights and decision-making questions were\n\nMention: \"Malawi Integrated Household Panel Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Its ownership module completes the cross-country comparison.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001908:49:0", "occurrence_id": "55870574541da3b280851497", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure A.1. Did the Governing Coalition Manipulate Elections?\n\n\n\n\n\n_Source:_ Data for the study comes from field reports of the Armed Forces of the Philippines (AFP) and election returns from\nthe Philippines Commission on Elections (COMELEC).\n_Note:_ The figure presents the probability density of coalition candidates’ margin of victory in the 2007 election. Municipalities\nwere grouped into 20 bins of equal width according to the incumbent’s margin of victory. Each scatter dot represents one bin.\nIts horizontal coordinate represents the midpoint of the bin, and its vertical coordinate represents the fraction of municipalities\nfor which the incumbent’s margin of victory was within the bin. Dashed lines are linear fits, separately estimated on both sides\nof the eligibility threshold. Solid lines are nonparametric fits from a local linear regression that uses triangular kernels with a\nbandwidth of 5 percentage points, separately estimated on both sides of the eligibility threshold.\n\n\n47\n\nMention: \"the Philippines Commission on Elections (COMELEC)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Official election returns give coalition candidates' margins of victory by municipality.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001923:5:1", "occurrence_id": "d623c5fc1b3de440b677130a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  labor input is not differentiated by plot,\nby crop, or by farm activity (such as weeding, harvesting, and so on). Instead, in the agricultural\nmodule outlined by Reardon and Glewwe (2000), the total days of labor at the household level\nover the last completed season are collected for each plot and by specific activity. An expanded\nagricultural module would have the same questions for each household member (as in the LSMSISA). <sup>5</sup> A common feature in these surveys is that labor information is collected from a single\ninterview.\n\nThough it is considered an improvement over surveys with more general labor force\nquestions, the expanded LSMS-ISA agricultural module has several potential drawbacks. First, it\nis time-consuming to collect this detailed information. Second, the burden on respondents is\nsubstantial: respondents are asked to provide information that they may never have considered (for\ninstance, about labor by activity for each plot). Third, there is potential for problems in recall and\nmemory.\n\n##### _2.2. What complicates the measurement of smallholder farm labor?_\n\n\n**Features of smallholder farming**\n\n\n4 Apart from multitopic household surveys, smallholder information can be collected through specialized farm\nsurveys. These often entail visiting the household at multiple times, particularly those surveys utilizing resident\nenumerators (for example, agricultural extension agents or other ministry of agriculture staff). However, these surveys\ntypically do not collect details on household farm labor.\n5 The LSMS-ISA program has been conducted in Burkina Faso, Ethiopia, Malawi, Mali, Niger, Nigeria, Tanzania,\nand Uganda. See LSMS (Living Standards Measurement Study) (database), World Bank, Washington, DC,\nhttp://www.worldbank.org/lsms.\n\n\n4\n\nMention: \"Living Standards Measurement Study\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"A footnote lists the eight countries where the programme has run and gives the database reference.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001927:6:1", "occurrence_id": "43d749f97418ec0ecb422ecb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \npublished as the continuous demographic and health surveys; (2) the Survey to\nMeasure Nutritional Composition of Meals Most Frequently Consumed away from\nHome ( _Encuesta para Medir la Composición Nutricional_ _de los Principales Alimentos_\n_Consumidos Fuera del Hogar_, ENCONUT); (3) ENAHO; and (4) the National Survey of\nFamily Budgets (ENAPREF).\n\n\nThe continuous demographic and health survey is part of the traditional demographic\nand health survey (DHS), which provides nationally representative information on\nvarious aspects of maternal and child health. While the standard demographic and\nhealth survey had been carried out in Peru at five-year intervals since 1986, the\ncontinuous design replaced it in 2004 and is being conducted annually. This allows\nhealth information to be monitored on women of reproductive age (15–49) and their\nchildren under age 5. The traditional questionnaire covers fertility, mortality,\nnutrition, and anthropometric information measured by health professionals, as well\nas individual and household characteristics. For 2013, the data set contains\ninformation on 2,106 non-pregnant women of reproductive age living in 40 districts\nand neighborhoods in the Lima Metropolitan Area (Lima and Callao).\n\n\nIn 2013, the Peruvian National Institute of Statistics and Informatics ( _Instituto_\n_Nacional de Estadística e Informática_ ), with technical support from the National Food\nand Nutrition Center ( _Centro Nacional de Alimentacion y Nutricion_ in the Ministry of\nHealth) and the support of the World Bank, conducted a unique, representative\n\n\n5\n\nMention: \"continuous demographic and health surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"One of four sources assembled for the nutrition analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001927:6:1", "occurrence_id": "9a72a3ad925a723a0e6f9a9f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \npublished as the continuous demographic and health surveys; (2) the Survey to\nMeasure Nutritional Composition of Meals Most Frequently Consumed away from\nHome ( _Encuesta para Medir la Composición Nutricional_ _de los Principales Alimentos_\n_Consumidos Fuera del Hogar_, ENCONUT); (3) ENAHO; and (4) the National Survey of\nFamily Budgets (ENAPREF).\n\n\nThe continuous demographic and health survey is part of the traditional demographic\nand health survey (DHS), which provides nationally representative information on\nvarious aspects of maternal and child health. While the standard demographic and\nhealth survey had been carried out in Peru at five-year intervals since 1986, the\ncontinuous design replaced it in 2004 and is being conducted annually. This allows\nhealth information to be monitored on women of reproductive age (15–49) and their\nchildren under age 5. The traditional questionnaire covers fertility, mortality,\nnutrition, and anthropometric information measured by health professionals, as well\nas individual and household characteristics. For 2013, the data set contains\ninformation on 2,106 non-pregnant women of reproductive age living in 40 districts\nand neighborhoods in the Lima Metropolitan Area (Lima and Callao).\n\n\nIn 2013, the Peruvian National Institute of Statistics and Informatics ( _Instituto_\n_Nacional de Estadística e Informática_ ), with technical support from the National Food\nand Nutrition Center ( _Centro Nacional de Alimentacion y Nutricion_ in the Ministry of\nHealth) and the support of the World Bank, conducted a unique, representative\n\n\n5\n\nMention: \"ENAHO\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"A household survey included in the multi-source compilation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001927:6:1", "occurrence_id": "ec901a6b421969b2f1f165d0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \npublished as the continuous demographic and health surveys; (2) the Survey to\nMeasure Nutritional Composition of Meals Most Frequently Consumed away from\nHome ( _Encuesta para Medir la Composición Nutricional_ _de los Principales Alimentos_\n_Consumidos Fuera del Hogar_, ENCONUT); (3) ENAHO; and (4) the National Survey of\nFamily Budgets (ENAPREF).\n\n\nThe continuous demographic and health survey is part of the traditional demographic\nand health survey (DHS), which provides nationally representative information on\nvarious aspects of maternal and child health. While the standard demographic and\nhealth survey had been carried out in Peru at five-year intervals since 1986, the\ncontinuous design replaced it in 2004 and is being conducted annually. This allows\nhealth information to be monitored on women of reproductive age (15–49) and their\nchildren under age 5. The traditional questionnaire covers fertility, mortality,\nnutrition, and anthropometric information measured by health professionals, as well\nas individual and household characteristics. For 2013, the data set contains\ninformation on 2,106 non-pregnant women of reproductive age living in 40 districts\nand neighborhoods in the Lima Metropolitan Area (Lima and Callao).\n\n\nIn 2013, the Peruvian National Institute of Statistics and Informatics ( _Instituto_\n_Nacional de Estadística e Informática_ ), with technical support from the National Food\nand Nutrition Center ( _Centro Nacional de Alimentacion y Nutricion_ in the Ministry of\nHealth) and the support of the World Bank, conducted a unique, representative\n\n\n5\n\nMention: \"National Survey of\nFamily Budgets\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The family budget survey contributes consumption information.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001927:6:1", "occurrence_id": "68c34d704381e55f5216b607", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \npublished as the continuous demographic and health surveys; (2) the Survey to\nMeasure Nutritional Composition of Meals Most Frequently Consumed away from\nHome ( _Encuesta para Medir la Composición Nutricional_ _de los Principales Alimentos_\n_Consumidos Fuera del Hogar_, ENCONUT); (3) ENAHO; and (4) the National Survey of\nFamily Budgets (ENAPREF).\n\n\nThe continuous demographic and health survey is part of the traditional demographic\nand health survey (DHS), which provides nationally representative information on\nvarious aspects of maternal and child health. While the standard demographic and\nhealth survey had been carried out in Peru at five-year intervals since 1986, the\ncontinuous design replaced it in 2004 and is being conducted annually. This allows\nhealth information to be monitored on women of reproductive age (15–49) and their\nchildren under age 5. The traditional questionnaire covers fertility, mortality,\nnutrition, and anthropometric information measured by health professionals, as well\nas individual and household characteristics. For 2013, the data set contains\ninformation on 2,106 non-pregnant women of reproductive age living in 40 districts\nand neighborhoods in the Lima Metropolitan Area (Lima and Callao).\n\n\nIn 2013, the Peruvian National Institute of Statistics and Informatics ( _Instituto_\n_Nacional de Estadística e Informática_ ), with technical support from the National Food\nand Nutrition Center ( _Centro Nacional de Alimentacion y Nutricion_ in the Ministry of\nHealth) and the support of the World Bank, conducted a unique, representative\n\n\n5\n\nMention: \"continuous demographic and health survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"It is described as belonging to the traditional DHS family.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001927:6:1", "occurrence_id": "3c3980dc085f1ba16567e505", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \npublished as the continuous demographic and health surveys; (2) the Survey to\nMeasure Nutritional Composition of Meals Most Frequently Consumed away from\nHome ( _Encuesta para Medir la Composición Nutricional_ _de los Principales Alimentos_\n_Consumidos Fuera del Hogar_, ENCONUT); (3) ENAHO; and (4) the National Survey of\nFamily Budgets (ENAPREF).\n\n\nThe continuous demographic and health survey is part of the traditional demographic\nand health survey (DHS), which provides nationally representative information on\nvarious aspects of maternal and child health. While the standard demographic and\nhealth survey had been carried out in Peru at five-year intervals since 1986, the\ncontinuous design replaced it in 2004 and is being conducted annually. This allows\nhealth information to be monitored on women of reproductive age (15–49) and their\nchildren under age 5. The traditional questionnaire covers fertility, mortality,\nnutrition, and anthropometric information measured by health professionals, as well\nas individual and household characteristics. For 2013, the data set contains\ninformation on 2,106 non-pregnant women of reproductive age living in 40 districts\nand neighborhoods in the Lima Metropolitan Area (Lima and Callao).\n\n\nIn 2013, the Peruvian National Institute of Statistics and Informatics ( _Instituto_\n_Nacional de Estadística e Informática_ ), with technical support from the National Food\nand Nutrition Center ( _Centro Nacional de Alimentacion y Nutricion_ in the Ministry of\nHealth) and the support of the World Bank, conducted a unique, representative\n\n\n5\n\nMention: \"demographic\nand health survey\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"It is characterised as providing nationally representative maternal and child health information.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001928:25:1", "occurrence_id": "53cc640b3feac95393e03587", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: . We\n\n\nthus take household expenditure reported in the National Sample Survey as our measure\n\n\nof household permanent income.\n\nTable 4 (panel A) provides estimates of household-level returns to education, _R_ <sup>_pn_</sup> and\n\n_R_ <sup>_pf_</sup>, in rural China and tests the null hypothesis that _R_ <sup>_pn_</sup> = _R_ <sup>_pf_</sup> using data from two\n\n\nrounds (1995 and 2002) of the Chinese Household Income Project (CHIP) survey. The\n\n\nstandard errors are clustered at the primary sampling unit (county). The estimates based\n\n\non the 5-year average income of a household in CHIP 2002 data in the last two columns\n\n\nof Table 4 show that the null hypothesis cannot be rejected with a P-value equal to 0.58. <sup>37</sup>\n\n\nThe evidence from the 1995 data (three year average income) also delivers a similar conclu\nsion: the null hypothesis cannot be rejected with a P-value of 0.33. <sup>38</sup> The conclusion that\n\n\n34To see this clearly, consider the polar case where schooling is provided only by the government free of\ncharge and there is no private schools (or private tutoring). In this case, the scope for parental financial\ninvestment to improve a child’s educational attainment is effectively nonexistent, making _θ_ 2 ≈ 0.\n\n35The available estimates on Mincerian labor market returns to education at the individual level in China\nshow low returns in the early years after the reform, but there is evidence of increasing returns in the later\nyears, as one would expect with the deepening of the labor market. The evidence also suggests higher labor\nmarket returns in nonfarm occupations (DeBrauw and Rozelle (2008)\n\nMention: \"Chinese Household Income Project\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Two rounds test the hypothesis on price ratios, with errors clustered at county level.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001928:25:1", "occurrence_id": "b87b50a9c3fc7404f437ed6f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: . We\n\n\nthus take household expenditure reported in the National Sample Survey as our measure\n\n\nof household permanent income.\n\nTable 4 (panel A) provides estimates of household-level returns to education, _R_ <sup>_pn_</sup> and\n\n_R_ <sup>_pf_</sup>, in rural China and tests the null hypothesis that _R_ <sup>_pn_</sup> = _R_ <sup>_pf_</sup> using data from two\n\n\nrounds (1995 and 2002) of the Chinese Household Income Project (CHIP) survey. The\n\n\nstandard errors are clustered at the primary sampling unit (county). The estimates based\n\n\non the 5-year average income of a household in CHIP 2002 data in the last two columns\n\n\nof Table 4 show that the null hypothesis cannot be rejected with a P-value equal to 0.58. <sup>37</sup>\n\n\nThe evidence from the 1995 data (three year average income) also delivers a similar conclu\nsion: the null hypothesis cannot be rejected with a P-value of 0.33. <sup>38</sup> The conclusion that\n\n\n34To see this clearly, consider the polar case where schooling is provided only by the government free of\ncharge and there is no private schools (or private tutoring). In this case, the scope for parental financial\ninvestment to improve a child’s educational attainment is effectively nonexistent, making _θ_ 2 ≈ 0.\n\n35The available estimates on Mincerian labor market returns to education at the individual level in China\nshow low returns in the early years after the reform, but there is evidence of increasing returns in the later\nyears, as one would expect with the deepening of the labor market. The evidence also suggests higher labor\nmarket returns in nonfarm occupations (DeBrauw and Rozelle (2008)\n\nMention: \"CHIP 2002 data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Five-year average household income is used in the robustness columns.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001934:26:0", "occurrence_id": "e31f1174b92206d5455f7b61", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 25\n\n\nFigure 5. Shares of global consumption of metals\n\n_(in percent)_\n\n80\n\n\n60\n\n\n40\n\n\n20\n\n\n0\n\n\nEM Share AE Share\n\n\nSource: World Bureau of Metal Statistics; IMF Primary Commodity Price System.\n\nFigure 6. Shares of global consumption of energy\n\n_(in percent)_\n\n70\n60\n50\n40\n30\n20\n10\n0\n\n\nEM Share AE Share\n\n\nSource: International Energy Agency, BP Statistical Review of World Energy June 2013; IMF\nPrimary Commodity Price System.\n\nMention: \"IMF Primary Commodity Price System\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Commodity price series accompany the metal consumption shares in the figures.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001935:66:0", "occurrence_id": "3c278e8d7afbeab2941c7273", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Table A.17: Number of births: Hindus vs Muslims\n\n\nHindus Muslims\n(1) (2)\n\n\n_Firstborn_ _girl_ 0.131*** 0.161***\n(0.016) (0.056)\n\n\n_Firstborn_ _girl_                - _Post1_ -0.077*** -0.054\n(0.019) (0.058)\n\n\n_Firstborn_ _girl_                - _Post2_ -0.093*** -0.046\n(0.026) (0.065)\n\n\n_Ideal_ _no._ _of_ _children_ 0.347*** 0.260***\n(0.017) (0.032)\n\n\n_Ideal_ _fraction_ _of_ _sons_ 0.039** 0.011\n(0.018) (0.026)\n\n\nN 67,433 9,038\nBaseline mean 2.997 3.200\n\n\nNOTES: Estimates of specification (3) estimated using OLS. The dependent variable is the number of births\nat the time of interview. Baseline means are average of the outcome variable in each column for mothers who\nhad both their first and last birth within 1973-1984. Standard errors in parentheses are clustered by state.\nData: NFHS. *** 1%, ** 5%, - 10%.\n\n\n64\n\nMention: \"NFHS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Birth histories of mothers whose first and last births fall in the window are used for the estimates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001940:20:0", "occurrence_id": "54774064a4125b390c1191b9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The third column of Table 4 extends our main analysis by controlling for the effect of fear in\n\ncitizens’ perception of COVID-19 on economic outcomes. The coefficient on the fear variable is\n\nnegative and significant, indicating that a higher perception of fear among the population\n\nnegatively affects economic activity. A 1 percent increase in the “fear” of the pandemic (search\n\nfrequencies) reduces the electricity consumption by about 0.02 percent. Compared to the baseline\n\nspecification shown in column (1), the introduction of the fear variable reduces slightly the\n\nmagnitude of the wait interaction effects without affecting their significance. All other coefficients\n\nare quantitatively similar to the coefficients in the baseline specification. Similar results are\n\nobserved in regressions when we used the terms “infection” and “COVID.”\n\n\nOne of the problems with using Google trends data is the issues of attribution (e.g., Mavragani and\n\nOchoa 2019). What if the searches of the term “death” are not connected with perceptions but\n\nrather reflect a time trend when people use Google more frequently for all searchers? To rule-out\n\nsuch possibility, we use searches for “neutral” terms “piano” and “milk.” We believe these terms\n\nnot to be correlated with perceptions of fear. Our falsification test confirms that these terms have\n\nno significant impact on the consumption of electricity.\n\n\n**6.** **Assessing the role of press coverage**\n\n\nThe results presented in the last specification of Table 4 show that individual perceptions of the\n\npandemic may have a role in explaining how economic activity varies along the reopening process.\n\nFetzer et al. (2020) show that economic anxiety driven by the COVID-19 pandemic can be detected\n\nin the increase in the volume of searches for “recession” and “stock market crash” as soon as the\n\nfirst cases of the disease were detected in each location. This points to the importance of diffusion\n\nof information as a factor driving individual behaviors.\n\nMention: \"Google trends data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Search terms cannot be attributed to perceptions, since searches may reflect other motives.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001944:10:1", "occurrence_id": "f493583b9d121e440cbf55c1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: enditures – data**\n\n\nIn the rest of the paper we present new global estimates of out-of-pocket health\nexpenditures for each of the measures listed in Table 1. We present in this section\ndetails of the data we use.\n\n\nOur data are drawn from the 2019 version of the World Bank’s Health Equity and\nFinancial Protection Indicators (HEFPI) database (Wagstaff et al. 2018a; Wagstaff\net al. 2019). In generating the database, we identified potentially suitable surveys\nby searching the above-mentioned microdata catalogs and multi-country survey\ncollections. In some cases, the surveys were not accessible to us. In others, the data\nwere accessible, but the surveys turned out to lack key information. Sometimes the\ndata sets we analyzed were different adaptations of the same survey – for example,\nas mentioned above, Romania’s HBS has been harmonized by Eurostat and the\nWorld Bank’s ECAPOV exercise. We identified 1,948 potentially suitable surveys or\nadaptations thereof from 182 countries. We analyzed and obtained out-of-pocket\nexpenditure estimates for 1,000 of these surveys, covering 154 countries.\n\n\nOur out-of-pocket expenditure estimates vary in how they are computed. In some\ncases, the survey asks only in the consumption or expenditure section about out-of\n\nMention: \"Health Equity and\nFinancial Protection Indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its measures, assembled from identified suitable surveys, are the study's data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001938:36:1", "occurrence_id": "3f75dbf56587f830aa5a8d45", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  **𝐺** 𝐺𝐺𝑌𝑌𝑖 **𝑖** −1 x 𝑁 **𝑁** 𝑁𝑁𝑃𝑃𝑖 **𝑖** −1]⁄𝑀 **𝑀** 𝑀 **𝑀** 𝑀𝑀𝑃𝑃𝑖\n\nwhere 𝑀 **𝑀** 𝑀 **𝑀** 𝑀𝑀𝑃𝑃𝑖 **𝑖** is the 3-year moving average of NGDP and 𝜋𝜋𝑡𝑡 =\n\n𝐶 **𝐶** 𝐼𝐼𝑡𝑡⁄𝐶 **𝐶** 𝐼𝐼𝑡𝑡−1 −1 is the inflation rate.\n\n**𝐶** **𝐶**\n\n\n\n**𝐺** **𝑖** **𝑁** **𝑖** **𝐺** **𝑖** **𝑁** **𝑖** **𝑀** **𝑀**\n\nwhere 𝑀 **𝑀** 𝑀 **𝑀** 𝑀𝑀𝑃𝑃𝑖 **𝑖**\n\n𝐶 **𝐶** 𝐼𝐼𝑡𝑡⁄𝐶 **𝐶** 𝐼𝐼𝑡𝑡−1 −1 is the inflation rate.\n\n**𝐶** **𝐶**\n\n\n\n**𝑀** **𝑀** **𝑖**\n\n**𝐶** **𝐶**\n\nΔ𝐶 **𝐶** 𝐶 **𝐶** 𝑖\n\n\n\n**𝑀** **𝑀** **𝑖**\n\nThe primary data source to construct the commodity resource revenue variable,\n\n**𝐶** **𝐶**\n\n**𝐶** **𝐶**\n\n\n\nThe primary data source to construct the commodity resource revenue variable,\n\n**𝐶** **𝐶**\n\n𝑦𝑦, is the ICTD/UNU-WIDER Government Revenue Dataset 2018 (GRD),\n\n**𝐶** **𝐶**\n\n\n\nwhich provides data on “Total Natural Resource Revenue, General Government,\n\n**𝐶**\n\nMention: \"ICTD/UNU-WIDER Government Revenue Dataset 2018\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Total natural resource revenue of general government is taken from this dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001948:27:1", "occurrence_id": "1c1c1dac512830df86f61693", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: . Village exposure to _JS_ increased the\n\n\nratio of female-to-male literacy rates from 56 percent to 77 percent ( _p_ < .01).\n\n\nWe do not use official records of police stations as an additional source of data\n\n\non domestic violence because these vastly underreport it. The largest multi-country\n\n\ncomparison to date of reporting gender-based violence to formal sources used\n\n\nDemographic Health Surveys to assess bounds on formal sources of data on gender\n\nbased violence (Palermo, Bleck, and Peterman 2014). The study found that in India\n\n\nand East Asia, only 2 percent of women of reproductive age who had experienced\n\n\nphysical or sexual violence reported it to a formal source.\n\n\n18 This estimation uses 59 instead of 60 control villages from the original sample\n\n\nbecause for one of the control villages data on distance to nearest town was not\n\n\navailable for 2011 from the Census.\n\n\n25\n\nMention: \"Demographic Health Surveys\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"A cited comparison used them to bound reporting of gender-based violence to formal sources.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001954:5:1", "occurrence_id": "3edec448c3d66023e2550cf8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ). While this is the case, it can make sense for allocations to these assets to remain stable. However, it\nwould be expected that these would decline as macroeconomic conditions and the pension sector develop. Whereas this has\nbeen observed in some countries (such as Mexico), others lag in their portfolio diversification trends.\n8 Investment returns are taken from OECD pension statistics. Geometric annual returns were calculated over a period of 7\nyears between 2009 and 2016 in order to have a sample of non-OECD countries as large as possible. Returns vs. wage growth\nare shown from the period 2006-2013 based on data available from the ILO wage database.\n9 The return period was selected based on the data available for the largest number of countries to allow for a broad\ncomparison.\n\n4\n\nMention: \"OECD pension statistics\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Seven-year geometric annual returns are computed from these statistics over a wide country set.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001954:5:1", "occurrence_id": "f790dc39b4c5fc85ef5d3d0a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ). While this is the case, it can make sense for allocations to these assets to remain stable. However, it\nwould be expected that these would decline as macroeconomic conditions and the pension sector develop. Whereas this has\nbeen observed in some countries (such as Mexico), others lag in their portfolio diversification trends.\n8 Investment returns are taken from OECD pension statistics. Geometric annual returns were calculated over a period of 7\nyears between 2009 and 2016 in order to have a sample of non-OECD countries as large as possible. Returns vs. wage growth\nare shown from the period 2006-2013 based on data available from the ILO wage database.\n9 The return period was selected based on the data available for the largest number of countries to allow for a broad\ncomparison.\n\n4\n\nMention: \"ILO wage database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Wage growth over the selected period is taken from this database to compare with returns.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001959:23:0", "occurrence_id": "bea6605b4c79198edf956fc3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 7:** Relative risk rates obtained from multinomial logit regressions: improved (above) and worsened\n(below)\n\n\n**_(a) By classes_** **_(b) By quintiles_**\n\n\n\n\n\n\n\n\n\n\n\n\n\n_Source:_ Authors’ calculations based on Morocco’s household surveys.\n\n\nAs opposed to results for “Worsened”, for which most of the coefficients are not statistically\n\n\nsignificant, the results for “Improved” relative to “Not changed” are particularly insightful with\n\n\nrespect to the relation between consumption polarization and household self-reported welfare.\n\n\nKeeping all the other variables constant, if polarization were to increase, the relative probability\n\n\nof declaring that a household’s welfare improved, rather than not changed, is\n\n 1  0.431  100  56.9% lower for the poor,  1  0.574  100  42.6% lower for the vulnerable,\n\nand  1  0.872  100  12.8% lower for households in the lower-middle class—albeit the 95\n\n\npercent confidence interval of the corresponding relative risk rate overlaps with 1 (0.713—1.068);\n\n\nhence, this decrease in relative probability is not significant. For middle-class and upper-middle\nclass households, the same relative probabilities are, respectively,  3.217  1  100  221.7% and\n\n 3.425  1  100  242.5% higher.\n\n\nSimilar results hold _vis-à-vis_ a robustness check where model (6) was re-estimated by replacing\n\nthe factor variable _CLASS_ with a new indicator variable _QGP_ identifying quintiles of expenditure,\n\n\ni.e.:\n\n\n\n5\n### (9) ln  Pr  Yi  j \n\nMention: \"Morocco’s household surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Distributional changes by class and quintile are computed from these surveys.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001960:3:0", "occurrence_id": "0e62a880b579d052b015a184", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **1.** **Introduction**\n\n\nThe economic impact of the COVID-19 pandemic has been severe and persistent. In 2020, the\n\n\nglobal economy is expected to experience its worst recession since the great depression of the 1930s,\n\n\nwith the highest share of countries experiencing contractions in per capita GDP since 1870 (WBG,\n2020). <sup>1</sup> Estimates by Apedo-Amah et al. (2020) show that firm sales declined by more than 70%\n\naround the peak of the crisis (as measured by the steepest drop in Google Mobility), and have\n\n\nremained more than 40% below last year’s levels even several months later. Two-thirds of firms have\n\n\neither fired employees, reduced worker hours or wages, or asked workers to take leave. More than\n\nhalf of micro and small businesses (those with fewer than 20 employees) are in arrears or expect\n\nto fall into arrears in the next six months. And while there is welcome news of vaccines and new\n\ntreatments for COVID-19, these are unlikely to be available everywhere at once meaning that the\n\n\neconomic pain is likely to persist for some time.\n\n\nIn response to the crisis, governments around the world have relied on a wide range of policy\n\n\nmeasures to support firms and households. While these responses have been uneven across countries,\n\n\nthey are unprecedented in their magnitude. Fiscal stimulus in high-income economies has reached\n\n\n10% of GDP or more, with around 40% going to firm support. In developing countries, faced with\n\n\nmore limited fiscal space, governments have allocated between 1% and 3% of GDP to this purpose,\n\n\nwith about a quarter of this amount dedicated to supporting businesses. Irrespective of the amount\n\n\nof the resources invested, data on whether these are reaching the intended beneficiaries, addressing\n\n\ntheir needs, and helping firms adjust to COVID-19 are desperately needed to inform policy-making.\n\n\nThis paper provides the first such assessment\n\nMention: \"Google Mobility\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its steepest drop marks the peak of the crisis against which sales declines are dated.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001961:23:0", "occurrence_id": "f458c44f9f9532ba3986b7ca", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Source:_ Universal Postal Union database.\n_Note:_ Data are from 2015–17. No data were available for Bangladesh, Kyrgyzstan, and Senegal.\n\nIn particular, the emergence of cross-border e-commerce calls for reexamining the efficiency of customs\nprocesses. Cross-border e-commerce now accounts for 10–15 percent of the volume of global e-commerce,\nand it is expected to grow at twice the rate of domestic e-commerce until 2020. <sup>84</sup> A simplified customs\nprocedure such as prearrival risk assessment and digitizing the process through electronic data exchange\nand payment could all contribute to a more cost-efficient customs process.\n\nGiven the growing flow of low value cross-border shipments, customs authorities face more severe\nchallenges on the long-standing trade-off between foregone revenue and the cost of collection and longer\nclearance times. Many customs authorities around the world have adopted a de minimis threshold—that is,\na “valuation ceiling for goods, including documents and trade samples, below which no duty or tax is\ncharged and clearance procedures, including data requirements, are minimal.” <sup>85</sup> How to set a reasonable de\nminimis threshold has become of special importance. The calculation and collection of customs duties and\ntaxes on low-value shipments disproportionally burden small e-commerce exporters because they are\nunable to streamline these tasks. Research has shown that a reasonable de minimis threshold can be vital\nfor small business firms that depend on cross-border sales. <sup>86</sup>\n\nAmong the 21 countries covered under Digital Business Indicators project, Bangladesh, Burkina Faso,\nPakistan, Senegal, and Tunisia do not have a de minimis threshold in place. Thresholds are often set based\non the value of the consignment, ranging from zero up to €1,000 (figure 10). Kenya and Tanzania set a low\nconsignment value threshold\n\nMention: \"Universal Postal Union database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Postal indicators for the covered countries are taken from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001961:23:0", "occurrence_id": "f033205ae49b9ee8a782c30f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Source:_ Universal Postal Union database.\n_Note:_ Data are from 2015–17. No data were available for Bangladesh, Kyrgyzstan, and Senegal.\n\nIn particular, the emergence of cross-border e-commerce calls for reexamining the efficiency of customs\nprocesses. Cross-border e-commerce now accounts for 10–15 percent of the volume of global e-commerce,\nand it is expected to grow at twice the rate of domestic e-commerce until 2020. <sup>84</sup> A simplified customs\nprocedure such as prearrival risk assessment and digitizing the process through electronic data exchange\nand payment could all contribute to a more cost-efficient customs process.\n\nGiven the growing flow of low value cross-border shipments, customs authorities face more severe\nchallenges on the long-standing trade-off between foregone revenue and the cost of collection and longer\nclearance times. Many customs authorities around the world have adopted a de minimis threshold—that is,\na “valuation ceiling for goods, including documents and trade samples, below which no duty or tax is\ncharged and clearance procedures, including data requirements, are minimal.” <sup>85</sup> How to set a reasonable de\nminimis threshold has become of special importance. The calculation and collection of customs duties and\ntaxes on low-value shipments disproportionally burden small e-commerce exporters because they are\nunable to streamline these tasks. Research has shown that a reasonable de minimis threshold can be vital\nfor small business firms that depend on cross-border sales. <sup>86</sup>\n\nAmong the 21 countries covered under Digital Business Indicators project, Bangladesh, Burkina Faso,\nPakistan, Senegal, and Tunisia do not have a de minimis threshold in place. Thresholds are often set based\non the value of the consignment, ranging from zero up to €1,000 (figure 10). Kenya and Tanzania set a low\nconsignment value threshold\n\nMention: \"Digital Business Indicators project\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its country coverage identifies which economies lack a de minimis threshold.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001964:3:0", "occurrence_id": "b0e5af802564e2f5b1809134", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **1.** **Introduction**\n\n\nGlobal trade has grown slowly after the Great Recession, and 2019 is projected to continue the\ntrend as one of the years with the weakest trade performance since the 2008-2009 global financial\ncrisis. A growing body of work analyzes the determinants of the current slowdown in global trade\n(Constantinescu et al. 2018; Hoekman 2015; Haugh et al. 2016; IMF 2016). Recent analyses\nattribute the trade slowdown, in varying degrees, to such factors as changes in the composition of\neconomic activity away from import-intensive investment, a slowing pace of global value chain\ngrowth and trade liberalization, and an increase in trade protectionism.\n\n\nThis paper draws upon recent pioneering work on the measurement and impact of economic policy\nuncertainty (Baker et al. 2016) to examine whether policy uncertainty can help explain the\nslowdown in world trade growth. To motivate the analysis, Figure 1 juxtaposes the growth in\nvolume of world merchandise imports against an index of global economic policy uncertainty.\nThe figure does suggest that there is a negative association between the two, and the particularly\nweak trade performance in 2019 has coincided with unusually high levels of economic policy\nuncertainty in that year.\n\n\n**Figure 1.** **_World import growth and policy uncertainty, January 2012 to June 2019_**\n\n\n\n350\n\n\n300\n\n\n250\n\n\n200\n\n\n150\n\n\n100\n\n\n50\n\n\n0\n\n\n\n\n\n6\n\n\n5\n\n\n4\n\n\n3\n\n\n2\n\n\n1\n\n\n0\n\n\n‐1\n\n\n‐2\n\n\n\n|Monthly global economic policy uncertainty<br>index<br>World merchandise trade volume (year‐<br>over‐year percent change of 3 month<br>moving average, right axis)|Col2|Col3|Col4|Col5|Col6|Col7|Col8|\n|---|---|---|---|---|---|---|---|\n|Mar<br>Jun<br>Sep<\n\nMention: \"an index of global economic policy uncertainty\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Its path is juxtaposed with merchandise import volumes to motivate the analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001965:4:1", "occurrence_id": "0ed17dfb868c7d885442d0ce", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  undertake due diligence. In the PEP data set, the\n\n\ndefinition of politically exposed person depends on the classifications by intergovernmental bodies\n\n\nsuch as Financial Action Task Force (FATF) or legislation such as the European Union’s Anti-Money\n\n\nLaundering Directive. More importantly, for our purposes, a wide range of information is collected\n\n\nfor each politically exposed person. For example, the data set reports association with specific\n\n\ncompanies, and whether these linkages are established directly, or through family members, close\n\n\nbusiness associates or advisers of each individual PEP. Each politically exposed person is further\n\n\nclassified into four sub-categories based on their primary affiliation with Local Government, State\n\n\n3\n\nMention: \"PEP data set\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Its definition of politically exposed person follows intergovernmental classifications rather than being fixed.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001966:11:0", "occurrence_id": "8830a109076d0ba9a0596bdf", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: application _Rosyama_ (rosyama.ru) as a means to address the problem of potholes in the city and the\n\nmechanism of keeping the local government accountable. <sup>6</sup> [^6: _Rosyama_ was not the only mobile app created by the Foundation Against Corruption. Other applications included\nRosZKH, helping residents write petitions to the Housing Inspection Committees on problems of housing\ninfrastructure services.] This mobile application allows its users to\n\ngeolocate a road problem (besides potholes, it included broken rails, snow accumulation, open\n\nmanholes, and persistent puddles), take a picture, and place it on the map. The application then\n\ngenerates a text complaint forwarded to the local Road Inspectorate (Ermoshina 2014). Ermoshina\n\n(2014) observes that between May 2011 and May 2014, 56,919 potholes were reported using _Rosyama_\n\ndigital application, and 15,966 of those were repaired. Researchers referred to _Rosyama_ and other\n\nrelated web applications, as ‘civic applications’ and as instruments of citizens’ legal empowerment.\n\nThey were conceived as instruments for keeping the local administrators accountable for their\n\nresponsibilities. Navalny’s team intended these applications as a means for generating collective\n\naction in a wider fight against corruption in Russia. Not all of these opposition plans achieved their\n\ndesired ends. Some of these initiatives in the digital sphere were ‘highjacked’ by the city government.\n\nThe Russian government invested heavily in digital tools of government since Dmitry\n\nMedvedev’s presidency, during 2008-2012. The Moscow city authorities also actively developed\n\ndigital instruments of governance and have invested massive resources into promoting a ‘smart city’\n\nagenda. Its most recent strategy document, “the Smart City – 2030,” outlined the main goals and\n\nobjectives of a continuing urban transformation (City of Moscow 2018). The 2018 United Nations\n\ne-government survey of 40 cities across the world ranked Moscow as a leading city (along with Cape\n\nTown and Tallinn) in using information and communication technologies to deliver public services\n\n(United Nations 2018). The major city redevelopment plans\n\nMention: \"The 2018 United Nations\n\ne-government survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its ranking of forty cities places Moscow among the leaders in ICT use.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001968:27:1", "occurrence_id": "fb86363ef85760b5ae6a74e6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: uza, 2014), but the cabinet ultimately rejected the policy because it\n\nwas deemed duplicative with the existing Child Development Policy. More recently, in 2018, a national\n\nstakeholder forum brought together government, donors and implementers to discuss ECD. However,\n\nwithout a formal coordinating structure or budget, it has been difficult for the government to follow\n\nthrough on these discussions.\n\n\n_Issue characteristics_\n\n\nEvidence of poor learning outcomes has raised public and political awareness of challenges facing the\n\neducation system while also increasing understanding of the potential of quality ECE programs to address\n\nthem (see Table 6). The first factor related to “issue characteristics” -- the availability and use of evidence\n\n( _credible indicators_ ) -- is most prominent in Ethiopia. The first EGRA results showing poor learning\n\noutcomes in rural areas were released at same time as the National Policy Framework (2010), and this\n\nfavorable timing drew attention to the new policy and connected investment in ECE to improved school\n\n\n13 The “Organisation Mondiale pour l'Éducation Préscolaire” or “World Organisation for Early Childhood Education” is\nabbreviated as OMEP in all languages.\n14 https://www.opensocietyfoundations.org/voices/president-johnson-sirleaf-backs-early-childhood-development-liberia\n\n26\n\nMention: \"EGRA results\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Poor rural learning outcomes released alongside the policy framework drew attention to the issue.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001962:32:0", "occurrence_id": "d43126a65ecce4f20e5954c6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 4. Financial sector developments\n\nA. Number of EMDE in post-crisis booms\nand credit crunches\n\nNumber of EMDEs\n\n25\n\n\n20\n\n\n15\n\n\n10\n\n\n5\n\n\n0\n\n\n\nB. EMDEs: Bank credit to\nprivate sector\n\n\n\nPercent of GDP\n\n\n70\n\n60\n\n50\n\n40\n\n30\n\n20\n\n10\n\n0\n\n\n\n2000 2007 2016\n\n\n\n\n\n\n\nC. Net capital inflows D. Banks’ profitability\n\n\n\nPercent of GDP Percentage points\n\n2.0 Equity Debt FX volatility (RHS)\n\n\n\nPercent Percent\n\nReturn on equity\n\n\n\n3\n\n\n2\n\n\n1\n\n\n\nReturn on equity\nReturn on assets (RHS)\n\n\n\n25\n\n\n20\n\n\n15\n\n\n10\n\n\n5\n\n\n0\n\n\n\n1.5\n\n\n1.0\n\n\n0.5\n\n\n0.0\n\n\n-0.5\n\n\n\n24\n\n\n18\n\n\n12\n\n\n6\n\n\n0\n\n\n\nE. Bank assets F. EMDE-based banks operating\nin EMDEs\n\n\n\nPercent of GDP\n\n\n120\n100\n80\n60\n40\n20\n0\n\n\n\n\n\n2000 2007 2016\nAdvanced economy 2016 average\n\n\n\nNumber of branches and subsidaries\n\n\n\n600\n\n\n400\n\n\n200\n\n\n0\n\n\n\nWithin region\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nSource: Institute of International Finance, International Monetary Fund, World Bank.\nNote: EMDEs = Emerging market and developing economies, EAP = East Asia and Pacific, ECA = Europe and\nCentral Asia, LAC = Latin America and the Caribbean, MNA = Middle East and North Africa, SAR = South Asia,\nSSA = Sub-Saharan Africa, and GFC = global financial crisis, 2008/09.\nA. Credit booms (crunches) are episodes when private credit to GDP ratio exceeds (falls below) its long-term trend\nby 1.65 times one standard deviation of a cyclical component obtained with the HP filter. Sample includes about 140\nEMDEs with private sector credit data. Weights are based on nominal GDP measured in U.S. dollars at market\nexchange rates.\nB. Unweighted averages. Based on the private credit by deposit money banks and other financial institutions to\nGDP (%) from the World Bank’s Financial Development and Structure Dataset.\n\nMention: \"Financial Development and Structure Dataset\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Private credit by banks and other institutions relative to GDP is taken from this dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001977:34:5", "occurrence_id": "5b0f3cc325cc19b096301eeb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: </u> <u>8.00</u> <u>2.18</u> <u>2.31</u> <u>8.43</u> <u>5.64</u> <u>6.18</u> <u>14.42</u> <u>21.15</u> <u>17.75</u> <u>23.15</u>\nSource: Authors’ estimates based on HSS data.\n\n\n33\n\nMention: \"HSS data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Household survey figures are used to estimate the reported quantities.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001993:20:1", "occurrence_id": "329eb2cd615a168864f54b94", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Hidalgo and Hausmann (2009).\nD. Exports and imports in percent of GDP.\nE. Government effectiveness survey from the World Bank’s Worldwide Governance indicators, defined as: perceptions of the quality of public\nservices, the quality of the civil service and the degree of its independence from political pressures, the quality of policy formulation and\nimplementation, and the credibility of the government’s commitment to such policies.\nF. Gross fixed capital formation in percent of GDP.\n\n\n1980-90 covariates. Higher initial education, greater economic complexity, institutional quality\n(measured using the WGI government effectiveness index), and lower initial productivity levels\nare consistently associated with a higher probability of switching into a higher-productivity club\n\n\n18\n\nMention: \"Worldwide Governance indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its government effectiveness perceptions of public service and civil service quality are used.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002001:25:0", "occurrence_id": "2887514b05a29f5102b17751", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ##### **9 Tables & Figures**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n**Figure** **1:** States of Mexico\n\n\nNote: The figure shows the percent of indigenous population in each state in Mexico. The parental involvement interventions were carried\nout in Guerrero, Puebla, Chiapas and Yucatan (labeled in white font). Data source: World Bank Open Data.\n\nMention: \"World Bank Open Data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It supplies the data shown on the map of intervention states.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002003:20:0", "occurrence_id": "2e77c3a736aceeae36059e82", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 4. Long-term growth prospects of EMDEs\n\nA. Long-term consensus forecasts: output B. Long-term consensus forecasts:\ngrowth Investment growth\n\n\n\nPercent\n\n4\n\n\n3\n\n\n2\n\n\n1\n\n\n0\n\n\n\n2000 2007 2019\n\n\n\nPercent\n\n8\n\n\n6\n\n\n4\n\n\n2\n\n\n0\n\n\n\nPercent\nEMDEs EMDEs excl. China World\n\n12\n\n\n10\n\n\n8\n\n\n6\n\n\n4\n\n\n2\n\n\n\nWorld Advanced\nEconomies\n\n\n\nEMDEs (RHS)\n\n\n\nC. Potential output growth D. Potential per capita output growth\n\n\n\nPercent\n\n7\n6\n5\n4\n3\n2\n1\n0\n\n\n\nPotential growth\n\n\n\nPercent\n\n6\n\n\n4\n\n\n2\n\n\n0\n\n\n\nPotential growth\n\n\n\n\n\n\n\nSource: Consensus Economics, Haver Analytics, Penn World Tables, UN Population Prospects, World Bank.\nA. Bars show long-term (10 years ahead) average annual growth forecasts surveyed in respective years. Sample\ncomprises 38 countries—20 advanced economies and 18 EMDEs—for which consensus forecasts are consistently\navailable during 1998-2019. Aggregate growth rates calculated using constant 2010 U.S. dollar GDP weights.\nB. 10-year-ahead forecasts surveyed in indicated year. Aggregate growth rates are calculated using constant 2010\nU.S. dollar investment weights. Sample comprises 23 advanced economies and 20 EMDEs.\nC.D. Period average of annual GDP-weighted averages. Estimates based on production function approach. World\nsample comprises 50 EMDEs and 30 advanced economies.\n\nMention: \"Penn World Tables\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its national accounts inputs feed the potential growth projections.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002003:20:0", "occurrence_id": "6135e8a6bdfbf0d8364030b7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 4. Long-term growth prospects of EMDEs\n\nA. Long-term consensus forecasts: output B. Long-term consensus forecasts:\ngrowth Investment growth\n\n\n\nPercent\n\n4\n\n\n3\n\n\n2\n\n\n1\n\n\n0\n\n\n\n2000 2007 2019\n\n\n\nPercent\n\n8\n\n\n6\n\n\n4\n\n\n2\n\n\n0\n\n\n\nPercent\nEMDEs EMDEs excl. China World\n\n12\n\n\n10\n\n\n8\n\n\n6\n\n\n4\n\n\n2\n\n\n\nWorld Advanced\nEconomies\n\n\n\nEMDEs (RHS)\n\n\n\nC. Potential output growth D. Potential per capita output growth\n\n\n\nPercent\n\n7\n6\n5\n4\n3\n2\n1\n0\n\n\n\nPotential growth\n\n\n\nPercent\n\n6\n\n\n4\n\n\n2\n\n\n0\n\n\n\nPotential growth\n\n\n\n\n\n\n\nSource: Consensus Economics, Haver Analytics, Penn World Tables, UN Population Prospects, World Bank.\nA. Bars show long-term (10 years ahead) average annual growth forecasts surveyed in respective years. Sample\ncomprises 38 countries—20 advanced economies and 18 EMDEs—for which consensus forecasts are consistently\navailable during 1998-2019. Aggregate growth rates calculated using constant 2010 U.S. dollar GDP weights.\nB. 10-year-ahead forecasts surveyed in indicated year. Aggregate growth rates are calculated using constant 2010\nU.S. dollar investment weights. Sample comprises 23 advanced economies and 20 EMDEs.\nC.D. Period average of annual GDP-weighted averages. Estimates based on production function approach. World\nsample comprises 50 EMDEs and 30 advanced economies.\n\nMention: \"UN Population Prospects\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its demographic projections supply the labour input to potential growth.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002004:10:0", "occurrence_id": "0b6f8c99cedcbdf71966bfee", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: and 4 countries in the Latin American and Caribbean region.\n\n\nWe use information on the gender of the household head to classify households as\n\n\nmale-headed and female-headed households. Table 2 provides summary statistics for each\n\n\nhousehold survey. We report the average log per capita expenditure and average household\n\n\nsize for the whole survey and by gender of the head of the household. It is interesting to\n\n\nnote that the average level of livelihood for male-headed household is not systematically\n\n\nhigher than for female-headed households. In addition, outside the Sub-Saharan Africa\n\n\n(SSA) region, male-headed household are on average larger than female-headed households\n\n\nin all countries. In SSA, male-headed households are larger in 22 of the 28 countries.\n\n\nThe import tariff data is summarized in Table 3 and in Figure 1. For our analysis we have\n\n\ngrouped the import tariff data in three categories of goods: staple agricultural products,\n\n\nnon-staple agricultural products and manufactured imported products. Table 3 reveals a\n\n\nlarge dispersion in the average import tariffs. For instance in Sub-Saharan Africa, for staple\n\n\nagricultural products the average tariff applied in Burundi is 23.8 percent but only 1.8\n\n\npercent in the Comoros. The same applies for non-staple agricultural products where Rwanda\n\n\napplies the highest average import tariff at 30.1 percent while Liberia’s average tariff for this\n\n\ncategory is 5.6 percent. In manufacturing, tariffs in Sub-Saharan African countries vary\n\n\nfrom 6.8 percent (Zambia) up to 23 percent (Cameroon). Tariff dispersion across countries\n\n\nand within regions is also observed in other developing regions. For instance, in South Asia,\n\n\nimport tariffs are very high for agricultural products in Bhutan (43.7 percent) and very low in\n\n\nPakistan (3.7 percent), while there is far less dispersion of tariffs for manufactured products\n\n\n(ranging from 15.3 to 23.5 percent). In Latin America, Ecuador applies lower average tariff\n\nMention: \"import tariff data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Rates are grouped into three goods categories for the analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002010:8:2", "occurrence_id": "3df070544963a7f76ff8cb87", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  country cases) and redistributed “1B1” emissions between not only\ncoal, but also other fossil fuels and electricity, assuming that in the initial country reporting some misallocation\nof emissions could have occurred. Such treatment has impacted mostly small fossil fuel producers in Africa.\nSecond, in the case of non-CO2 GHG emissions only, it was identified that emission intensity of the raw milk\nproduction activity (‘rmk’) in the GTAP database is much higher in some countries compared to the\ncorresponding region’s average intensity. As in the case of fugitive emissions discussed above, such large\ndeviations for the raw milk activity were observed mostly in some African countries. To address this issue, we\nhave revised the treatment of emissions reallocation and assumed that emissions from dairy animal care\nattributed to both raw milk and cattle sectors (‘ctl’) in the GTAP database.\n\n\n7\n\nMention: \"the GTAP database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its non-CO2 emission intensity for raw milk production is implausibly high in some countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001978:14:1", "occurrence_id": "724e1896f7cfa96355424019", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: *𝜃** 𝜃 **𝜃** 𝜃𝜃𝑖 **𝑖** + 𝛽𝛽x𝑋𝑋𝑖 **𝑖** + μ𝑖𝑖 + 𝜀𝜀𝑖 **𝑖** (1)\n\n**𝑖** 𝑖𝑖 at time 𝑡𝑡\n\n**𝑖**\n\n\n\n𝑌𝑌𝑖 **𝑖** 𝑖𝑖 = α + 𝜃 **𝜃** 𝜃 **𝜃** 𝜃𝜃𝑖 **𝑖** + 𝛽𝛽x𝑋𝑋𝑖 **𝑖** + μ𝑖𝑖 + 𝜀𝜀𝑖 **𝑖** (1)\n\nwhere 𝑌𝑌𝑖 **𝑖** denotes a vector of sector outcomes for country 𝑖𝑖 at time 𝑡𝑡 as described in section\n\n3.1; 𝑋𝑋𝑖 **𝑖** is a set of controls for a country’s economic and political characteristics and sector\n\n\n**𝑖** **𝜃** **𝜃** **𝑖** **𝑖** **𝑖**\n\nwhere 𝑌𝑌𝑖 **𝑖** denotes a vector of sector outcomes for country 𝑖𝑖 at time 𝑡𝑡 as described in section\n\nspecific characteristics, including GDP per capita; and _polity2_ is a regime-type indicator that\n\n3.1; 𝑋𝑋𝑖 **𝑖** is a set of controls for a country’s economic and political characteristics and sector\n\n\n**𝑖**\n\nspecific characteristics, including GDP per capita; and _polity2_ is a regime-type indicator that\n\n**𝑖**\n\ncaptures a country’s political regime authority, and power system size (installed electricity\n\ncapacity in gigawatts). To test the robustness of the results, other control variables, including\n\ndemocratic accountability (from the International Country Risk Guide Database) and\n\n\nTable A1 in the appendix summarizes the description and source of the variables used in the\n\npopulation density (from WDI) are used. μ𝑖𝑖 are country fixed effects\n\nMention: \"the International Country Risk Guide Database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its democratic accountability measure enters as a robustness control.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:001987:4:0", "occurrence_id": "8e46e0fedc4ad6a58f1bf803", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Against this backdrop, this paper seeks to determine the prevalence of price controls across\nEMDEs. To help answer the first question, this research contributes to the literature on price\ncontrols by presenting the findings from a new data set covering an almost complete set of EMDEs.\nThis data set extracts the list of products subject to price controls from the latest available Trade\nPolicy Reviews for each EMDE member country of the World Trade Organization. This list of\nproducts is compiled using existing legislation and additional material provided by country\nauthorities. The data set provides a rough view of the prevalence of price control measures across\ncountries. When combined with detailed information on country-level trade flows, the data set also\nprovides an indication of imports and exports potentially subject to controls. <sup>2</sup> [^2: Data on price controls on tradable goods combines the information on controlled prices from the World Trade\nOrganization’s Trade Policies Reviews with 4-digit HS trade values from the World Bank’s World Integrated Trade\nSolutions database.] Armed with this data\nset, this study finds that while price controls are seldom imposed on goods in advanced economies,\nthey are near ubiquitous in EMDEs and LICs in particular.\n\nSecond, this study seeks to enumerate the challenges that price controls impose for growth and\ndevelopment and government policies. While they may be introduced with the best intentions to\nimprove social outcomes, available evidence suggests that price controls often undermine growth\nand development, impose fiscal burdens and can weaken the effectiveness of monetary policy. At\nleast in part, this is because price controls cause a shift in consumption towards the subsidized\ngood, and away from other non-subsidized goods. Moreover, when there are trend increases in\ninternational prices, or when they interact with barriers to entry, price control measures frequently\nmorph into distortive subsidy regimes. Important social, fiscal and environmental costs are likely\nto follow, as well as adverse consequences for investment and employment, and productivity\ngrowth.\n\n**2.** **Use of price controls**\n\n**\n\nMention: \"World Integrated Trade\nSolutions database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Four-digit trade values are matched to controlled prices identified in trade policy reviews.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002007:22:1", "occurrence_id": "d32719b9b189b006d647dc4a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  these efforts. The results\nalso show that high levels of external scrutiny in the form of ratings and Eurobond issuances have a high\nimpact on LICDs’ transparency as a result of the data disclosure and investor relations practices required\nby them. Finally, increasing the numbers of highly skilled staff in debt offices has also been identified as\na key part of improving transparency.\n\n\nIn the absence of reliable and comprehensive data disclosed by the authorities, indirect reporting is\ncurrently integrating and, in some cases, replacing direct reporting channels. Indirect reporting has been\nfound to expand the coverage of debt statistics in most LIDCs, and country-specific comparisons of debt\nstocks from direct sources versus those recorded in the WB-IMF’s DSAs show that deviations therein can\nrepresent up to 30 percent of national GDP, confirming that differences in coverage are a key driver of\nthis gap. Even when direct and indirect reporting coincide in terms of sectoral coverage, discrepancies can\nbe as high as 7 percent of national GDP. This additional mismatch can be attributed to the use of\ndecentralized and non-automated recording processes, data manipulation, and different debt definitions.\n\n\nGiven these challenges, this paper puts forward a call for action to i) improve debt transparency by\nfocusing on those factors that are most likely to promote transparency; (ii) shift the focus of MDBs’\noperations and technical assistance from indirect to direct reporting; (iii) introduce a core of enforceable\n\nMention: \"WB-IMF’s DSAs\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Debt stocks recorded here deviate from direct sources by up to thirty percent of GDP, revealing coverage gaps.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002016:12:5", "occurrence_id": "15a03bde0ec719ce1ace98cb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .8 percentage points higher. In our sample the standard deviation of cumulative sales is\n0.193, so one standard deviation increase in cumulative sales would be associated with a 0.7\npercentage point (=0.193*0.038) increase in ROE. Finally, regressions 7-9 in Table 3 suggest that\nthere are no statistically significant short-term differences in cost-to-income ratios a few years\n\n\n20 We control for lagged ROA in the specifications except where the dependent variable was ROA, ROE or NIM.\n21 All bank-level variables come from the Bankscope (1995-2014) and Bank Focus Orbis (2015-2017) databases and\nall macro variables come from the Global Financial Development Database.\n\n\n11\n\nMention: \"Global Financial Development Database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"All macroeconomic variables in the bank panel are drawn from this compilation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002018:15:0", "occurrence_id": "26b9498dc33fcbb628732f37", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: MVAC forecasts of the number of people who are expected to suffer from food deficits in the\n\n\nupcoming lean season using data from a wide range of primary and secondary data collected by dedicated\n\n\nteam, which feed into a Household Economic Analysis model, and model results form the basis for the\n\n\nfirst-level of geographic targeting (MVAC, 2013; Svesve, 2015; FEWSNET, 2016; Babu et al., 2018). Though\n\n\nit is unclear exactly which rainfall estimate data was used to quantify hazards in the MVAC model – or that\n\n\nare subsequently taken into consideration by FARP – we tested a number of rainfall variables constructed\n\n\nusing different sources and covering different periods noted in those documents. We chose to retain\n\n\nthose variables that had the best predictive power in explaining aid distribution and that included:\n\n\n(i) the percent difference in total crop season (last dekad of November through the final\n\n\ndekad of March) rainfall realizations when rainfall was below the long- mean using the\n\n\ndecadal rainfall estimates from the University of California at Santa Barbara’s Climate\n\n\nHazards Group InfraRed Precipitation with Station (CHIRPS) data set (covering 1981\n\n2015),\n\n\n(ii) the percent difference in total season rainfall realizations when rainfall was above the\n\n\nsame long-term mean derived from the CHIRPS data set,\n\n\n(iii) the median 3-month Standardized Precipitation Index (SPI) values for the period\n\n\nDecember-January, as provided by the Instituto Pirenaico de Ecología, and\n\n\n(iv) the median total crop season Normalized Difference Vegetation Index (NDVI) constructed\n\n\nfrom Climate Data Record of Advanced Very High-Resolution Radiometer (AVHRR)\n\n\nSurface Reflectance, as provided by the National Oceanic and Atmospheric\n\n\nAdministration (NOAA).\n\n\nMVAC also bases its needs assessment on extension planning area (EPA)-level agricultural\n\n\nproduction estimates. Given data quality issues that a range of specific crops – and not\n\nMention: \"CHIRPS data set\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Its long-term mean defines the percent deviation of season rainfall.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002018:15:0", "occurrence_id": "4bd00520ddca273881ad0b8e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: MVAC forecasts of the number of people who are expected to suffer from food deficits in the\n\n\nupcoming lean season using data from a wide range of primary and secondary data collected by dedicated\n\n\nteam, which feed into a Household Economic Analysis model, and model results form the basis for the\n\n\nfirst-level of geographic targeting (MVAC, 2013; Svesve, 2015; FEWSNET, 2016; Babu et al., 2018). Though\n\n\nit is unclear exactly which rainfall estimate data was used to quantify hazards in the MVAC model – or that\n\n\nare subsequently taken into consideration by FARP – we tested a number of rainfall variables constructed\n\n\nusing different sources and covering different periods noted in those documents. We chose to retain\n\n\nthose variables that had the best predictive power in explaining aid distribution and that included:\n\n\n(i) the percent difference in total crop season (last dekad of November through the final\n\n\ndekad of March) rainfall realizations when rainfall was below the long- mean using the\n\n\ndecadal rainfall estimates from the University of California at Santa Barbara’s Climate\n\n\nHazards Group InfraRed Precipitation with Station (CHIRPS) data set (covering 1981\n\n2015),\n\n\n(ii) the percent difference in total season rainfall realizations when rainfall was above the\n\n\nsame long-term mean derived from the CHIRPS data set,\n\n\n(iii) the median 3-month Standardized Precipitation Index (SPI) values for the period\n\n\nDecember-January, as provided by the Instituto Pirenaico de Ecología, and\n\n\n(iv) the median total crop season Normalized Difference Vegetation Index (NDVI) constructed\n\n\nfrom Climate Data Record of Advanced Very High-Resolution Radiometer (AVHRR)\n\n\nSurface Reflectance, as provided by the National Oceanic and Atmospheric\n\n\nAdministration (NOAA).\n\n\nMVAC also bases its needs assessment on extension planning area (EPA)-level agricultural\n\n\nproduction estimates. Given data quality issues that a range of specific crops – and not\n\nMention: \"Instituto Pirenaico de Ecología\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It provides the median three-month standardised precipitation values for December-January.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002018:15:0", "occurrence_id": "170543a606928a9d73fcb580", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: MVAC forecasts of the number of people who are expected to suffer from food deficits in the\n\n\nupcoming lean season using data from a wide range of primary and secondary data collected by dedicated\n\n\nteam, which feed into a Household Economic Analysis model, and model results form the basis for the\n\n\nfirst-level of geographic targeting (MVAC, 2013; Svesve, 2015; FEWSNET, 2016; Babu et al., 2018). Though\n\n\nit is unclear exactly which rainfall estimate data was used to quantify hazards in the MVAC model – or that\n\n\nare subsequently taken into consideration by FARP – we tested a number of rainfall variables constructed\n\n\nusing different sources and covering different periods noted in those documents. We chose to retain\n\n\nthose variables that had the best predictive power in explaining aid distribution and that included:\n\n\n(i) the percent difference in total crop season (last dekad of November through the final\n\n\ndekad of March) rainfall realizations when rainfall was below the long- mean using the\n\n\ndecadal rainfall estimates from the University of California at Santa Barbara’s Climate\n\n\nHazards Group InfraRed Precipitation with Station (CHIRPS) data set (covering 1981\n\n2015),\n\n\n(ii) the percent difference in total season rainfall realizations when rainfall was above the\n\n\nsame long-term mean derived from the CHIRPS data set,\n\n\n(iii) the median 3-month Standardized Precipitation Index (SPI) values for the period\n\n\nDecember-January, as provided by the Instituto Pirenaico de Ecología, and\n\n\n(iv) the median total crop season Normalized Difference Vegetation Index (NDVI) constructed\n\n\nfrom Climate Data Record of Advanced Very High-Resolution Radiometer (AVHRR)\n\n\nSurface Reflectance, as provided by the National Oceanic and Atmospheric\n\n\nAdministration (NOAA).\n\n\nMVAC also bases its needs assessment on extension planning area (EPA)-level agricultural\n\n\nproduction estimates. Given data quality issues that a range of specific crops – and not\n\nMention: \"Climate Data Record of Advanced Very High-Resolution Radiometer\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Crop season vegetation index medians are constructed from this surface reflectance record.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002021:9:0", "occurrence_id": "177492ac86d565ebcd40b5ae", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: would be willing and able to engage in these activities. In total, 630 individu\n\nals were recommended by the CGFs (about 10 people per reforestation plot),\n\n\nof which half were randomly selected to participate in the program. Partic\n\nipants employed in the PES contract were randomly grouped into teams of\n\n\nfive and informed that they could collectively earn money based on mainte\n\nnance and survival of the newly planted trees on their assigned reforestation\n\n\nplot. PES contracts assured the groups that they would receive about $0.70\n\n\nfor every tree that was still alive at verification and that every group member\n\n\nwould receive an equal share (20%) of the total amount. About 9 months\n\n\nlater, in June/July 2018 (just before the start of the next rainy season), all\n\n\nreforestation areas were visited by independent monitors and participants\n\n\nreceived their payments based on the number of trees that were still alive.\n\n\nFrom those selected to be part of the PES scheme, more than 95% received\n\n\na positive payment. The individual payments ranged from about 840 FCFA\n\n\n($1.50) to 25,620 FCFA (about $45.6) with an average of 8,300 FCFA ($14.8),\n\n\nwhich represents approximately a week of food consumption for the median\n\n\nrural household in Burkina Faso <sup>3</sup>, and about half the monthly agricultural\n\n\nincome of households in our study sample. Figure 1 shows the distribution\n\n\nof the payment amounts received by the participants.\n\n\nOur baseline data suggest that the participants in the scheme come from\n\n\n3Authors’ calculations from the BF 2014 Living Standards Measurement Survey\n(LSMS) data estimates the median rural household’s daily consumption expenditures at\nabout 1580 FCFA ($3). In July 2018, 1000 FCFA = US$ 1.78.\n\n\n7\n\nMention: \"the BF 2014 Living Standards Measurement Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Median rural household daily consumption expenditure is estimated from it to contextualise payment amounts.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002025:26:1", "occurrence_id": "16145f7e028f59ed55d8baf4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: s_**, multiplied by the elasticity of substitution of imports between the\nsources **_σM,k_**, using the Armington elasticity for the sector as in the GTAP Data Base, (ii)\nimport-augmenting technical change, _amsk_ **_k,r,s_**, that lowers the effective price of a good in\nthe destination market, and (iii) the import penetration as captured by the change in\ncomposite imports of subsector commodity _k_, _qimk_ **_k,s_** . Note that the substitution effect for\na particular flow _(k,r,s)_ increases in divergence of import tariff for good _k_ from regions _r_\nto _s_, from the weighted-average tariff of _s_ . Since a higher weight means lower divergence,\nthis effect decreases in import-shares of region _r_ in the total imports by region _s_ of the good\n_k_ .\n\n\n**For all sectors k in SECT, regions r and s in REG:**\n\n\n**_qxskk,r,s = -amskk,r,s + qimkk,s -σM,k * [pmskk,r,s - amskk,r,s - pimkk,s]  …………… (1)_**\n\n\nGlobal transport margins are treated in the same manner as in the standard GTAP model,\nwith the quantity of international trade, transport and insurance services required being a\nfixed proportion of the volume of goods shipped. Technical change in this sector is\nrepresented with the variable atmfsdkk **_,r,s_** and is obtained by adding up the changes at\ndifferent levels, which are directly translated from the aggregate changes in the\ncorresponding variables. Trade and transport services\n\nMention: \"the GTAP Data Base\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its sectoral Armington elasticities set the substitution between import sources.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002026:35:1", "occurrence_id": "2fcbfb6a65198f694ccd788b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  many observations are missing for services, particularly\nfor the sectors of interest. The data reported is administrative data which in the case of TiVA may\nnot always be the case. It also includes internal trade, which is convenient for the gravity model.\nAnother downside however is that this database groups all digital services sectors together, as\nshown in Annex Table A3. The years covered for the panel analysis is 2005-2016.\n\n\nThe results using the ITPD-E data with respect to the baseline regression following equation (3) are\npresented in Table 8, which means that it covers a panel data set and includes the two sets of\ncontrol variables of the data restrictiveness index and digital enabling environment. As such, this\ntable is readily comparable with Table 7, the more so because the sectoral grouping for Digital+ and\nbusiness services are carefully cleaned and matched between the two data sources. The results are\nreported in Table 8. The results for In Table 8, the total number of observations is lower than using TiVA data which is due to missing GRAV𝑜 **𝑜** 𝑜𝑜 and C ∗intl𝑜 **𝑜** 𝑜𝑜 are reported in Annex Table A6.\n\nobservations. The list of countries in this dataset is also different and not balanced compared to the\nTiVA data. The results reported in columns 1-3 are without any control variables included. A first\nfinding is that none of the variables for the CB component have a significant coefficient outcome.\nThat stands in contrast to some of the significant findings in the bottom panel of Table 4, to which\nthese results need to be compared. When looking at the DR component, the results are largely in\nline with Table 4, except for the insignificance for the conditional model in column 2, even though\nthe sign remains positive. The difference in results may be due to the fact that even though the\nsector groupings are matched as closely as possible, some sub-sectoral differences remain\n\nMention: \"ITPD-E data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"It supplies the alternative panel for the baseline regression over the covered years.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002026:35:1", "occurrence_id": "76ffb47d8c0df0129db83288", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  many observations are missing for services, particularly\nfor the sectors of interest. The data reported is administrative data which in the case of TiVA may\nnot always be the case. It also includes internal trade, which is convenient for the gravity model.\nAnother downside however is that this database groups all digital services sectors together, as\nshown in Annex Table A3. The years covered for the panel analysis is 2005-2016.\n\n\nThe results using the ITPD-E data with respect to the baseline regression following equation (3) are\npresented in Table 8, which means that it covers a panel data set and includes the two sets of\ncontrol variables of the data restrictiveness index and digital enabling environment. As such, this\ntable is readily comparable with Table 7, the more so because the sectoral grouping for Digital+ and\nbusiness services are carefully cleaned and matched between the two data sources. The results are\nreported in Table 8. The results for In Table 8, the total number of observations is lower than using TiVA data which is due to missing GRAV𝑜 **𝑜** 𝑜𝑜 and C ∗intl𝑜 **𝑜** 𝑜𝑜 are reported in Annex Table A6.\n\nobservations. The list of countries in this dataset is also different and not balanced compared to the\nTiVA data. The results reported in columns 1-3 are without any control variables included. A first\nfinding is that none of the variables for the CB component have a significant coefficient outcome.\nThat stands in contrast to some of the significant findings in the bottom panel of Table 4, to which\nthese results need to be compared. When looking at the DR component, the results are largely in\nline with Table 4, except for the insignificance for the conditional model in column 2, even though\nthe sign remains positive. The difference in results may be due to the fact that even though the\nsector groupings are matched as closely as possible, some sub-sectoral differences remain\n\nMention: \"TiVA data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"It yields more observations than the alternative source, whose missing values reduce its sample.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002028:1:0", "occurrence_id": "c6fd5e19d071f9d1fca93fd7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Policy Research Working Paper 9323\n\n### **Abstract**\n\nThis study tests the weak form of the efficient capital mar\nkets theorem in five transition economies in Southeast\nEurope between 2005 and 2016. A panel pooled mean\ngroup estimator is used to examine the relationship between\nmacroeconomic indicators and the performance of stock\nmarket indexes. This is a suitable estimator for these young\nfrontier markets, given that they have yet to develop the\nbreadth and depth of an advanced market—such as ample\nliquidity and traders—to aggregate cross-country data and\nuse level series prime data instead of differentials of the same.\nThese frontier capital markets are found to be weak form\n\n\n\ninefficient, meaning that stock prices do not reflect available current public information. In other words, when a\nmarket is transparent and investor behavior is rational, the\nmacroeconomic data should be included in the value of the\nstock indexes. The five countries may benefit from bringing\ntheir capital markets legislation in line with those of developed countries and by improving corporate governance and\ntransparency. This would boost investor trust and liquidity.\nThe coverage of this research can be extended to find more\n\nstandardized data values and develop additional factors not\ncaptured by this model.\n\n\n\nThis paper is a product of Financial Institutions Group, International Finance Corporation. It is part of a larger effort by\n\nthe World Bank to provide open access to its research and make a contribution to development policy discussions around\nthe world. Policy Research Working Papers are also posted on the Web at http://www.worldbank.org/prwp. The author\nmay be contacted at adodig@ifc.org.\n\n\n_The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development_\n_issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the_\n_names of the authors and should be cited accordingly. The findings, interpretations,\n\nMention: \"cross-country data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Frontier markets lack the depth to aggregate it, so level series are used instead of differentials.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002028:1:0", "occurrence_id": "46f47a07beed2b3069d427a1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Policy Research Working Paper 9323\n\n### **Abstract**\n\nThis study tests the weak form of the efficient capital mar\nkets theorem in five transition economies in Southeast\nEurope between 2005 and 2016. A panel pooled mean\ngroup estimator is used to examine the relationship between\nmacroeconomic indicators and the performance of stock\nmarket indexes. This is a suitable estimator for these young\nfrontier markets, given that they have yet to develop the\nbreadth and depth of an advanced market—such as ample\nliquidity and traders—to aggregate cross-country data and\nuse level series prime data instead of differentials of the same.\nThese frontier capital markets are found to be weak form\n\n\n\ninefficient, meaning that stock prices do not reflect available current public information. In other words, when a\nmarket is transparent and investor behavior is rational, the\nmacroeconomic data should be included in the value of the\nstock indexes. The five countries may benefit from bringing\ntheir capital markets legislation in line with those of developed countries and by improving corporate governance and\ntransparency. This would boost investor trust and liquidity.\nThe coverage of this research can be extended to find more\n\nstandardized data values and develop additional factors not\ncaptured by this model.\n\n\n\nThis paper is a product of Financial Institutions Group, International Finance Corporation. It is part of a larger effort by\n\nthe World Bank to provide open access to its research and make a contribution to development policy discussions around\nthe world. Policy Research Working Papers are also posted on the Web at http://www.worldbank.org/prwp. The author\nmay be contacted at adodig@ifc.org.\n\n\n_The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development_\n_issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the_\n_names of the authors and should be cited accordingly. The findings, interpretations,\n\nMention: \"macroeconomic data\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"In transparent rational markets these series should be reflected in stock index values.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002031:12:0", "occurrence_id": "ce640c9bfc4cc66aa35df499", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Some specifications replace the location-specific characteristics ( _Xjd_ <sup>_l_</sup> <sup>) with fixed effects for</sup>\n\n\nhousehold primary sampling unit (PSU) that corresponds to the household’s village, town,\n\n\nor neighborhood. These fixed effects capture unobservable local factors that may influence\n\n\nsanitation practices in a geographic area. PSU fixed effects absorb the control variables for\n\n\nhousehold location, as the household location characteristics are homogeneous within a PSU.\n\n\nAll specifications limit the sample to households that are headed by a Hindu or Muslim.\n\n\nStandard errors are clustered at the PSU level (village, town, or neighborhood), which adjusts\n\n\nfor the spatial correlation in sanitation practices within nearby areas.\n\n\n**IV** **Results**\n\n\n**IV.A** **Main** **Estimates**\n\n\nTable 2 reports results from estimating equation (1), using the IHDS data set which observes\n\n\nthree sanitation practices. Panel A reports estimated differences in latrine ownership for\n\n\nHindus, relative to Muslims. Panel B reports estimated differences in handwashing. Panel\n\n\nC reports estimated differences in whether no fecal material was observed around the house.\n\n\nColumn 1, panel A, reports that Hindus are 17.9 percentage points less likely to own\n\n\na latrine than Muslims, without adjustment for other covariates (95% confidence interval:\n\n\n13.0pp less likely - 22.9pp less likely). Column 2 reports similar differences between Hindus\n\n\nand Muslims when controlling for household characteristics, such as household demographics\n\n\nand proxies for household wealth, though these controls do contribute substantial explanatory\n\n\npower (raising the adjusted R-squared from 0.014 to 0.341).\n\n\nColumn 3 reports that Hindus are 5.6 percentage points less likely to own a latrine than\n\n\nMuslims, after controlling for differences in household location (95% confidence interval:\n\n\n1.2pp less likely – 10.0pp\n\nMention: \"the IHDS data set\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"It observes three sanitation practices, enabling contrasts in latrine ownership by religious group.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002037:52:2", "occurrence_id": "a423042683a67269cc7e3d2e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: *Baseline**<br>**Repair**<br>**Unit**<br>**Cost**<br>**(USD**|N/A|$1.4 / mile<br>(PG&E 2018)|$1.4 / mile<br>(PG&E 2018)|$1.6 / mile<br>(PG&E 2018)|$1.6 / mile<br>(PG&E 2018)|$0.91 / mile<br>(Hall 2012)|$5.50/<br>substati<br>on<br>(PG&E<br>2018)|$7.50/<br>substati<br>on<br>(PG&E<br>2018)|$9.50/<br>Substati<br>on<br>(PG&E<br>2018)|\n\n\n6 Using 60/70 kV prices estimated from PG&E chart as 38 kV prices could not be found (PG&E 2018)\n7 From NY report, know that 75% of 1,200 distribution circuits need repair\n8 From NY report, know that 38 kV, 115 kV, and 230 kV substations were damaged; assume that an equal number of each were\ndamaged; from NY report (Table 3-4), know that 121/252 substations had minor damage and are in need of repair and 131/252\nsubstations have major damage and need to be replaced. Thus, assume half of substations need repair and half need replaced. Assume\nthis is true across 38 kV, 115 kV, and 230 kV substations. So, 42 38 kV need repair, 42 38 kV need replaced, 42 115 kV need repair, 42\n115 kV need replaced, 42 230 kV need repair,\n\nMention: \"PG&E\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its chart supplies substation connection prices, with estimated values used where a voltage class is unavailable.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002040:64:2", "occurrence_id": "0d3d7ca08e02e4bc00383415", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  prices (in US dollars). For identification of equation (2),\nwe use diplomatic exchange between Egypt and the countries of destination of Egyptian migrants at\ntime of migration. It is a dummy variable equal zero if there is no evidence of diplomatic exchange\nneither in Egypt, nor in the country of destination and is equal one if there is evidence of diplomatic\nexchange either in Egypt or in the country of destination at the level of charg´e d’affaires, minister,\nambassador or other. For identification of equation (3), we use the number of active armed conflicts\nthat is both country and year specific from the Uppsala Conflict Data Project (UCDP). Equation (4)\nestimates the effect of illegality on the logarithm of hourly wages, unconditional on return migration.\nThe predicted logarithm of hourly wages for non-migrants, illegal migrants and legal migrants are\nreported in the last row.\n\n\n63\n\nMention: \"Uppsala Conflict Data Project\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Country-year counts of active armed conflicts serve as the instrument for identification.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002044:3:0", "occurrence_id": "219af0fa5635c8d1950f97a1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: #### **1 Introduction**\n\nThe increase in gasoline prices stemming from the oil crisis overshadowed the\n\n\nUnited States presidential debate of October 1980. Ronald Reagan and then-President\n\n\nJimmy Carter were going head to head in the election. Carter’s loss that year coin\n\ncided with a peak in oil prices. Other modern US presidential incumbents - for\n\n\nexample, Presidents Ford and George H.W. Bush - also lost their reelection bids fol\n\nlowing oil price spikes. This anecdotal evidence points to a broader question about\n\n\nthe role of exogenous shocks in determining electoral outcomes.\n\n\nOur paper is the first effort to systematically examine the effect of oil price shocks\n\n\non the reelection probability of political incumbents in oil-importing countries. We\n\n\nuse a novel dataset of 207 national elections in 50 mostly oil-importing democracies,\n\n\nwhich crucially also includes polling data in the run-up to elections. The polling data\n\n\nallow us to explore the effect of exogenous shocks not only on the election outcome,\n\n\nbut also on voting intentions in the months leading up to the vote. Depending on the\n\n\npolitical system, the dataset includes elections of the chief executive in parliamentary\n\n\nor presidential systems. As large oil imports leave a country vulnerable to changes\n\n\nin _international_ crude oil prices, we rely on these prices as an exogenous source of\n\n\n_country-specific_ variation in terms of trade. Specifically, we create a measure of ’oil\n\n\nprice shock’ by interacting the fluctuations in international oil prices (over which\n\n\nindividual countries do not have explicit control) with the country-specific intensity\n\n\nof oil imports relative to GDP.\n\n\nAn oil price shock one year prior to an election is found to significantly reduce\n\n\nthe likelihood of reelection for the incumbent party. Specifically, a 1% increase in\n\n\nour oil shock measure decreases the likelihood of incumbent re-election by nearly six\n\n\npercentage points. Furthermore, novel polling data confirm that these shocks operate\n\n\nby reducing the popularity of the sitting government. We show that both\n\nMention: \"dataset of 207 national elections\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"It covers fifty mostly oil-importing democracies and includes pre-election polling.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002048:40:0", "occurrence_id": "771490ca1abd14706269404f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Table 5: Data to be calibrated\n\n\n<u>Frequency</u> <u>of</u> <u>lagged</u> <u>shipments</u> <u>in</u> <u>bin</u> <u>0</u> <u>1</u> <u>2</u> <u>3-5</u> <u>6-10</u> <u>11-100</u> <u>100+</u>\nShare of total 2016 shipments 0.189 0.051 0.038 0.084 0.089 0.374 0.176\nInspections per 100 shipments 14 6.7 6.4 6.0 5.6 5.7 6.0\nMean no. of lagged shipments 0 1 2 3.8 7.6 26.9 205\n<u>Total</u> <u>inspections</u> <u>per</u> <u>100</u> <u>shipments</u> <u>7.3</u>\n\nTable notes: Author calculations using Serbian customs data from 2016.\n\n\nmade up entirely of observations with exactly zero, one, or two lagged shipments, respectively.\n\n\nIn the last four bins, we average over _ofk_ triplets that have varying levels of lagged shipment\n\n\nfrequencies. The mean for each bin is reported in row 3 of Table 5.\n\n\n**Parameterization**\n\n\nIn order to emphasize the role of ambiguity, we shut down by assumption other possible\n\n\nsources of cross-bin variation that might otherwise affect search behavior and/or compliance.\n\n\nWe restrict the optimism parameter to be equal across bins, _αl_ = _α_, _∀l_ . Likewise the firm’s\n\n\nloss parameter, _L_, is assumed\n\nMention: \"Serbian customs data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Inspections per hundred shipments are calculated from these records.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002051:11:2", "occurrence_id": "658f08cc51a18d0c203c6290", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  the mean distance of households from the main roads is 66 km in the DHS and\n\n\n14.6 km in LSMS surveys.\n\n\n3The roads digitization effort of Jedwab and Storeygard (2021) benefited from funding under the DFID/World Bank’s\nStrategic Research Program on Transport Policies for Sustainable and Inclusive Growth.\n\n\n10\n\nMention: \"DHS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"It places households 66 km from main roads on average, far more distant than the alternative surveys imply.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002051:11:2", "occurrence_id": "20f69a91e81c97f400341120", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  the mean distance of households from the main roads is 66 km in the DHS and\n\n\n14.6 km in LSMS surveys.\n\n\n3The roads digitization effort of Jedwab and Storeygard (2021) benefited from funding under the DFID/World Bank’s\nStrategic Research Program on Transport Policies for Sustainable and Inclusive Growth.\n\n\n10\n\nMention: \"LSMS surveys\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its mean distance of 14.6 km contrasts sharply with the other programme's estimate.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002056:7:0", "occurrence_id": "3f8d3a70c230524c8d61fb0a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ##### **3. Data**\n\nThe road network data in the Ethiopia road database were collected from different sources. Federal\nRoad Network data were obtained from the Ethiopian Roads Authority, the main agency in the\nRoad Sector Development Program (RSDP) and Growth and Transformation Plan (GTP); and\ndistrict- and woreda-level road data were collected from the Regional Road Authorities. The\ncollected information includes years of new construction, upgrading, and rehabilitation, as well as\nthe volume and distribution of the network in the country. It is important to note that the database\nincludes rural roads that were motorable at the time of the national road inventory carried out in\n2015−16. Rural roads in some parts of Oromia and Amhara regions were not covered due to\nmissing bridges, as well as Somali region for security reasons.\n\n\nIn analyzing household-level impacts of the recent rural road development, this study primarily\nrelies on the ESS. The ESS, part of the World Bank supported Living Standards Measurement\nStudy-Integrated Surveys on Agriculture (LSMS-ISA), started as the Ethiopian Rural Socioeconomic Survey (ERSS) in 2012. A sample of about 3,470 households were interviewed in rural\nareas and re-interviewed in 2014 and 2016. Among the rural communities covered by the ESS, this\nstudy focuses on the communities that received rural roads between 2010 and 2014 (i.e., treated\ngroup) and the other communities that did not receive rural roads between 2010 and 2016 (i.e.,\ncontrol group). This results in 204 communities, including 59 communities in Amhara, 46\ncommunities in Oromia, 37 communities in SNNPR, 26 communities in Tigray, and 36\ncommunities in the other regions.\n\n\nThe recipient indicator used for the econometric analysis in this study is a binary indicator about\nwhether a rural road passes by within the community’s boundary regardless of the road length. <sup>\n\nMention: \"national road inventory\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"It includes only rural roads motorable at the time and omits regions lacking bridges or with security constraints.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002056:7:0", "occurrence_id": "ea0aa3c12a9b94c3fbdc9977", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ##### **3. Data**\n\nThe road network data in the Ethiopia road database were collected from different sources. Federal\nRoad Network data were obtained from the Ethiopian Roads Authority, the main agency in the\nRoad Sector Development Program (RSDP) and Growth and Transformation Plan (GTP); and\ndistrict- and woreda-level road data were collected from the Regional Road Authorities. The\ncollected information includes years of new construction, upgrading, and rehabilitation, as well as\nthe volume and distribution of the network in the country. It is important to note that the database\nincludes rural roads that were motorable at the time of the national road inventory carried out in\n2015−16. Rural roads in some parts of Oromia and Amhara regions were not covered due to\nmissing bridges, as well as Somali region for security reasons.\n\n\nIn analyzing household-level impacts of the recent rural road development, this study primarily\nrelies on the ESS. The ESS, part of the World Bank supported Living Standards Measurement\nStudy-Integrated Surveys on Agriculture (LSMS-ISA), started as the Ethiopian Rural Socioeconomic Survey (ERSS) in 2012. A sample of about 3,470 households were interviewed in rural\nareas and re-interviewed in 2014 and 2016. Among the rural communities covered by the ESS, this\nstudy focuses on the communities that received rural roads between 2010 and 2014 (i.e., treated\ngroup) and the other communities that did not receive rural roads between 2010 and 2016 (i.e.,\ncontrol group). This results in 204 communities, including 59 communities in Amhara, 46\ncommunities in Oromia, 37 communities in SNNPR, 26 communities in Tigray, and 36\ncommunities in the other regions.\n\n\nThe recipient indicator used for the econometric analysis in this study is a binary indicator about\nwhether a rural road passes by within the community’s boundary regardless of the road length. <sup>\n\nMention: \"ESS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Its rural household panel is the main source for analysing road development impacts.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002056:7:0", "occurrence_id": "e212adadb50fc21fa9774c15", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ##### **3. Data**\n\nThe road network data in the Ethiopia road database were collected from different sources. Federal\nRoad Network data were obtained from the Ethiopian Roads Authority, the main agency in the\nRoad Sector Development Program (RSDP) and Growth and Transformation Plan (GTP); and\ndistrict- and woreda-level road data were collected from the Regional Road Authorities. The\ncollected information includes years of new construction, upgrading, and rehabilitation, as well as\nthe volume and distribution of the network in the country. It is important to note that the database\nincludes rural roads that were motorable at the time of the national road inventory carried out in\n2015−16. Rural roads in some parts of Oromia and Amhara regions were not covered due to\nmissing bridges, as well as Somali region for security reasons.\n\n\nIn analyzing household-level impacts of the recent rural road development, this study primarily\nrelies on the ESS. The ESS, part of the World Bank supported Living Standards Measurement\nStudy-Integrated Surveys on Agriculture (LSMS-ISA), started as the Ethiopian Rural Socioeconomic Survey (ERSS) in 2012. A sample of about 3,470 households were interviewed in rural\nareas and re-interviewed in 2014 and 2016. Among the rural communities covered by the ESS, this\nstudy focuses on the communities that received rural roads between 2010 and 2014 (i.e., treated\ngroup) and the other communities that did not receive rural roads between 2010 and 2016 (i.e.,\ncontrol group). This results in 204 communities, including 59 communities in Amhara, 46\ncommunities in Oromia, 37 communities in SNNPR, 26 communities in Tigray, and 36\ncommunities in the other regions.\n\n\nThe recipient indicator used for the econometric analysis in this study is a binary indicator about\nwhether a rural road passes by within the community’s boundary regardless of the road length. <sup>\n\nMention: \"Living Standards Measurement\nStudy-Integrated Surveys on Agriculture\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"It is the programme of which the national survey forms part.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002056:7:0", "occurrence_id": "09a5dd3a0840707091cf2deb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ##### **3. Data**\n\nThe road network data in the Ethiopia road database were collected from different sources. Federal\nRoad Network data were obtained from the Ethiopian Roads Authority, the main agency in the\nRoad Sector Development Program (RSDP) and Growth and Transformation Plan (GTP); and\ndistrict- and woreda-level road data were collected from the Regional Road Authorities. The\ncollected information includes years of new construction, upgrading, and rehabilitation, as well as\nthe volume and distribution of the network in the country. It is important to note that the database\nincludes rural roads that were motorable at the time of the national road inventory carried out in\n2015−16. Rural roads in some parts of Oromia and Amhara regions were not covered due to\nmissing bridges, as well as Somali region for security reasons.\n\n\nIn analyzing household-level impacts of the recent rural road development, this study primarily\nrelies on the ESS. The ESS, part of the World Bank supported Living Standards Measurement\nStudy-Integrated Surveys on Agriculture (LSMS-ISA), started as the Ethiopian Rural Socioeconomic Survey (ERSS) in 2012. A sample of about 3,470 households were interviewed in rural\nareas and re-interviewed in 2014 and 2016. Among the rural communities covered by the ESS, this\nstudy focuses on the communities that received rural roads between 2010 and 2014 (i.e., treated\ngroup) and the other communities that did not receive rural roads between 2010 and 2016 (i.e.,\ncontrol group). This results in 204 communities, including 59 communities in Amhara, 46\ncommunities in Oromia, 37 communities in SNNPR, 26 communities in Tigray, and 36\ncommunities in the other regions.\n\n\nThe recipient indicator used for the econometric analysis in this study is a binary indicator about\nwhether a rural road passes by within the community’s boundary regardless of the road length. <sup>\n\nMention: \"Ethiopian Rural Socioeconomic Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"About 3,470 rural households were interviewed in this baseline and re-interviewed later.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002057:34:0", "occurrence_id": "7c93612808c6a3b58ad369d7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: marriage. <sup>22</sup> [^22: In the 2001 Indian census, 65 percent of women’s internal migration movement was motivated by marriage and\nonly 3 percent by employment (Food and Agricultural Organization, 2018).] With economic development, more women are joining the labor force and moving to\n\n\ncities—sometimes temporarily—to avail employment opportunities, for instance in garment\n\n\nfactories. Yet, the migration literature almost exclusively focuses on the migration of young\n\n\nworking-age men. With the increasing feminization of labor force and women accounting for a\n\n\ngreater share of labor migrants, the migration patterns, causes and consequences of women’s labor\n\n\nmigration should be another priority area for future research.\n\n\nFor many developing countries, both international migration and internal migration are of very\n\n\nsignificant economic importance. However, these two types of migration are typically analyzed\n\n\nin complete isolation of each other. How migrants sort themselves between different international\n\n\nand internal migration and the implications such sorting have for economic well-being and\n\n\ninequality should be yet another priority of future research.\n\n\nIn view of these research gaps, it is clear that furthering the research on internal migration will\n\n\ndemand much higher quality data than what is currently available. Much of the research gaps in\n\n\nthe literature on developing countries are in fact due to data unavailability. Currently, the main\n\n\nsources of data are censuses which have relatively good spatial resolution but insufficient\n\n\nfrequency (as data is typically collected once every 10 years), household surveys whose spatial\n\n\ncoverage is limited though they typically collect more detailed information, and administrative\n\n\ndata which are still scarce and incomplete in developing countries (see Kirchberger (2020) for\n\n\ndetailed discussion on the data issues). With the exception of a few RCTs and panel surveys,\n\n\nmigrants are not tracked over time, making it difficult to examine temporary, return, and circular\n\n\nmigration. Recent studies increasingly use cell phone call data records (tower pings) to trace\n\n\nmigration (Blumenstock, Chi and Tan 2019, Lai et al., 2020) and trace\n\nMention: \"2001 Indian census\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It records that most women's internal migration was motivated by marriage rather than employment.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002062:9:1", "occurrence_id": "d0338b5b49b2c4c53273fbe4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: </u> Our expected dowry variable is a sig\n\nnificant predictor of the actual dowry paid at the time of marriage. Using data on individuals whose\n\n\nyear of birth and year of marriage fall within our data, we run an ordinary least-squares (OLS)\n\n\nregression of the actual dowry paid at the time of marriage on the expected dowry in the year of\n\n\nbirth (using our definition). Table A.3 shows that the “first stage” relationship between our con\n\nstructed expected dowry measure and actual dowry that the household pays, including state fixed\n\neffects, year-of-birth fixed effects, caste fixed effects and their interactions, is highly significant. <sup>23</sup>\n\n\nFigure A.4 shows the scatter plot overlaid with the linear prediction plot without any controls.\n\n\nMoreover, the correlation between expected and actual dowry is positive and significant for all four\n\n\nof our caste categories.\n\n\n<u>Recall</u> <u>bias</u> <u>in</u> <u>dowry</u> <u>data:</u> Like most survey data, one may be concerned about the recall bias\n\n\nin the reported dowry variable, especially when the year of marriage is too far back in time. To\n\n\nexamine the extent of the recall error, we utilize data from the 1999 round of REDS and compare\n\n\naverage dowry paid by year of marriage for years that are available in both rounds. Note that the\n\n\nquestion about dowry was asked differently in both rounds. Nonetheless, Figure A.5 shows that,\n\n\nfor the sample period of our savings analysis (1986-2007), the two rounds report similar dowry\n\n\namounts. Thus, recall bias is unlikely to be a significant issue for our regression estimates. Given\n\n\n20The same issue prevents us from using district-level variation—the average number of marriages in a\ndistrict-caste-year cell is just four\n\nMention: \"1999 round of REDS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Dowry reported by year of marriage is compared across rounds to gauge recall error.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002063:10:0", "occurrence_id": "f5bb49f3731c9d88138e23a6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: recent phenomenon.\n\n\nTo establish that our mobile coverage measure significantly predicts actual mobile phone\n\n\nusage, we regress mobile phone ownership dummy and indicators for usage of specific mobile\n\n\nphone services on our mobile coverage variables. In Table 1, we present results of this exercise\n\n\n- the correlation between mobile coverage rates and uptake of mobile phone services – using\n\n\ndata from the Demographic Health Surveys (DHS). Across all panels, we observe a strong and\n\n\npositive correlation between the mobile coverage rates and ownership of a mobile phone, use of\n\n\nmobile money for financial transactions and mobile broadband internet usage. These findings\n\n\nprovide support to the use of our coverage data to evaluate the effect of mobile phones on health\n\n\noutcomes.\n\n\n**2.2** **Health Outcomes Data**\n\n\nThe health outcomes data come from the DHS and span 25 countries in Africa, of which 23 are\n\nin Sub-Saharan Africa. <sup>13</sup> The DHS datasets are collected using standardized questionnaires,\n\n\nwith some adaptations based on country specific needs. To allow comparability across coun\n\ntries and over time, the DHS program standardizes variable names and definitions, and cleaner\n\nand consistent datasets are released to users as “recodes\". <sup>14</sup> The Integrated Public Use Mi\n\ncrodata Series (IPUMS) compiles these country level datasets into consolidated multi-country\n\n\ndata. We extract key variables on infant mortality, health seeking behavior and health care uti\n\nlization at the child and the household level from the IPUMS repository.\n\n\nThe DHS surveys consist of three core questionnaires: the household, women’s and men’s\n\n\nquestionnaires. We rely on information from the first two in this paper. The household ques\n\ntionnaire covers household roster, age and gender of household members, relationship status\n\n\nwith household head, education and place of residence. The women’s questionnaire collects in\n\nformation on mother’s characteristics including age, marital status and education; reproductive\n\nMention: \"Demographic Health Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Mobile phone ownership and use are correlated with modelled coverage rates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002063:10:0", "occurrence_id": "b09c831bb5b8c013e948da88", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: recent phenomenon.\n\n\nTo establish that our mobile coverage measure significantly predicts actual mobile phone\n\n\nusage, we regress mobile phone ownership dummy and indicators for usage of specific mobile\n\n\nphone services on our mobile coverage variables. In Table 1, we present results of this exercise\n\n\n- the correlation between mobile coverage rates and uptake of mobile phone services – using\n\n\ndata from the Demographic Health Surveys (DHS). Across all panels, we observe a strong and\n\n\npositive correlation between the mobile coverage rates and ownership of a mobile phone, use of\n\n\nmobile money for financial transactions and mobile broadband internet usage. These findings\n\n\nprovide support to the use of our coverage data to evaluate the effect of mobile phones on health\n\n\noutcomes.\n\n\n**2.2** **Health Outcomes Data**\n\n\nThe health outcomes data come from the DHS and span 25 countries in Africa, of which 23 are\n\nin Sub-Saharan Africa. <sup>13</sup> The DHS datasets are collected using standardized questionnaires,\n\n\nwith some adaptations based on country specific needs. To allow comparability across coun\n\ntries and over time, the DHS program standardizes variable names and definitions, and cleaner\n\nand consistent datasets are released to users as “recodes\". <sup>14</sup> The Integrated Public Use Mi\n\ncrodata Series (IPUMS) compiles these country level datasets into consolidated multi-country\n\n\ndata. We extract key variables on infant mortality, health seeking behavior and health care uti\n\nlization at the child and the household level from the IPUMS repository.\n\n\nThe DHS surveys consist of three core questionnaires: the household, women’s and men’s\n\n\nquestionnaires. We rely on information from the first two in this paper. The household ques\n\ntionnaire covers household roster, age and gender of household members, relationship status\n\n\nwith household head, education and place of residence. The women’s questionnaire collects in\n\nformation on mother’s characteristics including age, marital status and education; reproductive\n\nMention: \"DHS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its standardised questionnaires supply health outcomes for twenty-five African countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002063:10:0", "occurrence_id": "a1639e0b4967ab5bee21d8e4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: recent phenomenon.\n\n\nTo establish that our mobile coverage measure significantly predicts actual mobile phone\n\n\nusage, we regress mobile phone ownership dummy and indicators for usage of specific mobile\n\n\nphone services on our mobile coverage variables. In Table 1, we present results of this exercise\n\n\n- the correlation between mobile coverage rates and uptake of mobile phone services – using\n\n\ndata from the Demographic Health Surveys (DHS). Across all panels, we observe a strong and\n\n\npositive correlation between the mobile coverage rates and ownership of a mobile phone, use of\n\n\nmobile money for financial transactions and mobile broadband internet usage. These findings\n\n\nprovide support to the use of our coverage data to evaluate the effect of mobile phones on health\n\n\noutcomes.\n\n\n**2.2** **Health Outcomes Data**\n\n\nThe health outcomes data come from the DHS and span 25 countries in Africa, of which 23 are\n\nin Sub-Saharan Africa. <sup>13</sup> The DHS datasets are collected using standardized questionnaires,\n\n\nwith some adaptations based on country specific needs. To allow comparability across coun\n\ntries and over time, the DHS program standardizes variable names and definitions, and cleaner\n\nand consistent datasets are released to users as “recodes\". <sup>14</sup> The Integrated Public Use Mi\n\ncrodata Series (IPUMS) compiles these country level datasets into consolidated multi-country\n\n\ndata. We extract key variables on infant mortality, health seeking behavior and health care uti\n\nlization at the child and the household level from the IPUMS repository.\n\n\nThe DHS surveys consist of three core questionnaires: the household, women’s and men’s\n\n\nquestionnaires. We rely on information from the first two in this paper. The household ques\n\ntionnaire covers household roster, age and gender of household members, relationship status\n\n\nwith household head, education and place of residence. The women’s questionnaire collects in\n\nformation on mother’s characteristics including age, marital status and education; reproductive\n\nMention: \"Integrated Public Use Mi\n\ncrodata Series\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It consolidates the standardised country recodes into multi-country files from which variables are extracted.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002072:1:0", "occurrence_id": "a2071b937d98b305f26c8391", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Policy Research Working Paper 7520\n\n##### **Abstract**\n\nThis paper revisits the extent of seasonality in African\n\nlivelihoods. It uses 19 years of monthly food prices from\n20 markets and three years of nationally representative\nhousehold panel surveys from Tanzania. Trigonometric\nspecifications are introduced to measure the seasonal gap.\nWhen samples are short and seasonality is poorly defined,\n\nthey produce less upward bias than the common dummy\nvariable approach. On average, the seasonal gap for maize\nprices is estimated to be 27 percent; it is 15 percent for rice.\n\n\n\nIn both cases it is two and a half to three times higher than\nin the international reference market. Food price seasonality\nis not a major contributor to food price volatility, but it does\ntranslate into seasonal variation in caloric intake of about 10\npercent among poor urban households and rural net food\nsellers. Rural net food-buying households appear able to\nsmooth their consumption. The disappearance of seasonality from Africa’s development debate seems premature.\n\n\n\nThis paper is a product of the “Agriculture in Africa—Telling Facts from Myths” project managed by the Office of the\nChief Economist, Africa Region and the Jobs Group of the World Bank, in collaboration with the Poverty and Inequality\nUnit, Development Economics Department of the World Bank, the African Development Bank, the Alliance for a Green\nRevolution in Africa, Cornell University, Food and Agriculture Organization, Maastricht School of Management, Trento\nUniversity, University of Pretoria, and the University of Rome Tor Vergata. It is part of a larger effort by the World Bank\nto provide open access to its research and make a contribution to development policy discussions around the world. Policy\nResearch Working Papers are also posted on the Web at http://econ.worldbank.org. The authors of the paper may be\ncontacted at lchristiaensen@worldbank.org.\n\n\n_The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about\n\nMention: \"nationally representative\nhousehold panel surveys from Tanzania\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Three years of panel data measure the seasonal gap in livelihoods alongside food prices.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002073:48:0", "occurrence_id": "1f742e43d48506c5b28e3f2d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure S3.2: Teenage parenthood by age of migration\n\n\n\n\n\n\n\n\n\n\n\n\n\n_a)  Female, 19 years old at census_ _b) Female, 24 years old at census_\n\n\n\n\n\n\n\n_c)_ _Female, 30 years old at census_\n\n\n\n\n\n\n\n\n\n\n\n\n\n_d)_ _Male, 19 years old at census_ _e) Male, 24 years old at census_\n\n\nNotes: Markers indicate mean of teenage pregnancies, bars indicate 95% confidence interval. Red\nhorizontal line is the estimated teenage birth rate for non-migrating population of same age.\n\n\nSource: School data come from the 1997–2009 waves of the Brazilian School Census; official\npopulation estimates from the Brazilian Census Bureau; births by age of conception from Brazilian\nVital Statistics.\n\n\n\n13\n\nMention: \"Brazilian School Census\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its waves supply the school data behind the teenage birth rate estimates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002073:48:0", "occurrence_id": "520b7ef0572c4bd16e4769e5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure S3.2: Teenage parenthood by age of migration\n\n\n\n\n\n\n\n\n\n\n\n\n\n_a)  Female, 19 years old at census_ _b) Female, 24 years old at census_\n\n\n\n\n\n\n\n_c)_ _Female, 30 years old at census_\n\n\n\n\n\n\n\n\n\n\n\n\n\n_d)_ _Male, 19 years old at census_ _e) Male, 24 years old at census_\n\n\nNotes: Markers indicate mean of teenage pregnancies, bars indicate 95% confidence interval. Red\nhorizontal line is the estimated teenage birth rate for non-migrating population of same age.\n\n\nSource: School data come from the 1997–2009 waves of the Brazilian School Census; official\npopulation estimates from the Brazilian Census Bureau; births by age of conception from Brazilian\nVital Statistics.\n\n\n\n13\n\nMention: \"Brazilian\nVital Statistics\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Births by age at conception are taken from these civil registration records.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002079:4:0", "occurrence_id": "7ab3e89d1e25b4c200ee5b60", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: and whether it is perceived in the same way by both parties. Joint decision making has been found to\n\n\ncorrelate negatively with IPV in several country studies, including Ebrahim and Atteraya (2019),\n\n\nFriedemann-Sánchez and Lovatón (2012), Gage (2005), Hindin and Adair (2002) and Svec and Andic\n\n\n(2018). This literature primarily focuses on the wife’s perception of the process of decision making and\n\n\nhow egalitarian she thinks this process is. While one analysis included both the husband’s and wife’s\n\n\nresponses (Zegenhagen et al. 2019), we go beyond this and include covariates that reflect not only their\n\n\nresponses, but whether or not they give the same responses. This allows us to explore the direction of\n\n\ndisagreement and its asymmetric effect on IPV.\n\n\nA separate body of interdisciplinary research offers a theoretical explanation for why joint decision making\n\n\nmay be associated with lower IPV. In the management, political science and conflict resolution literatures,\n\n\njoint decision making is highlighted as a tool for diffusing responsibility and reducing decisional conflict by\n\n\nsharing risk. For example, in her typology of political blame management strategies, McGraw (1991)\n\n\nfeatures ‘horizontal diffusion of responsibility’, whereby political leaders avoid conflict in the face of a\n\n\ndecision gone awry by arguing that their decision was a joint product of a group of individuals. Thompson\n\n\n(1980) discusses the blame-mitigation strategy of emphasizing ‘collective responsibility’. Shared decision\n\n\nmaking between doctors and patients is also highlighted as a tool for reducing decisional conflict—and\n\n\nhence, malpractice lawsuits—in the medical literature (Hoffmann et al. 2014, Kremer et al. 2007).\n\n\nOur paper builds on existing work by examining the relationship between joint decision making and\n\n\nphysical, sexual and emotional IPV across 12 Sub-Saharan African countries, using nationally\n\n\nrepresentative Demographic and Health Surveys (DHS) for 31,243 couples. We find\n\nMention: \"Demographic and Health Surveys (DHS)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Thirty-one thousand couples are used to relate joint decision making to physical, sexual and emotional violence.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002050:12:0", "occurrence_id": "bab96085ce04056a916058fa", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 5: Roads do not exist in isolation, but as part of networks, and some proposed roads may depend on the construction of\nother roads - resulting in internal hierarchies (tiers). Numerous possible combinations exist for such roads when selected with\nconsideration given to their tier rank, and the resulting combinations of roads are called sequences.\n\n\n3.4. Cost-time modeling\n\n\nThe first component of our methodology involves calculating the accessibility to services by means of\ncost-time models. The backbone of our cost-time models comes from a layer of on-road/on-path speeds. The\nprincipal data input is a transportation layer produced by merging roads and paths from the Nepal Survey\nDepartment (2019) and OpenStreetMap with roads surfaces and conditions data compiled in 2015 for the\nRural Access Index. Road segments were assigned speeds based on their surface type and Department of\nRoads classification (Department of Roads, 2013). These speeds were then modified according to the condition\nand gradient of the road. Using modifiers provided by a focus group of World Food Programme civil engineers,\nwe then calculated a separate monsoon speed for each segment based on the average deterioration in speeds\nover given road surface/condition combinations under monsoon conditions.\n\nA separate interim model of off-road, off-path walking speeds was then prepared. To do so, we first calculated the slope for each grid cell in the SRTM 1 arc-second (≈ 30 m <sup>2</sup> ) digital elevation model. These slopes\nwere input into Tobler’s hiking formula to derive a base walking speed for each grid cell. We multiplied this\nspeed by a fractional landcover walking speed modifier using a 1 km <sup>2</sup> gridded landcover model of Nepal from\nthe International Centre for Integrated Mountain Development (ICIMOD, 2013). A downward adjustment\nin walking speeds was applied to extreme altitudes (greater than 3500 m) where lower oxygen counts reduce\n\nMention: \"OpenStreetMap\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"Its roads and paths are merged with the survey department layer to build the transport network.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002050:12:0", "occurrence_id": "9d49650ee7376edb61956b59", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 5: Roads do not exist in isolation, but as part of networks, and some proposed roads may depend on the construction of\nother roads - resulting in internal hierarchies (tiers). Numerous possible combinations exist for such roads when selected with\nconsideration given to their tier rank, and the resulting combinations of roads are called sequences.\n\n\n3.4. Cost-time modeling\n\n\nThe first component of our methodology involves calculating the accessibility to services by means of\ncost-time models. The backbone of our cost-time models comes from a layer of on-road/on-path speeds. The\nprincipal data input is a transportation layer produced by merging roads and paths from the Nepal Survey\nDepartment (2019) and OpenStreetMap with roads surfaces and conditions data compiled in 2015 for the\nRural Access Index. Road segments were assigned speeds based on their surface type and Department of\nRoads classification (Department of Roads, 2013). These speeds were then modified according to the condition\nand gradient of the road. Using modifiers provided by a focus group of World Food Programme civil engineers,\nwe then calculated a separate monsoon speed for each segment based on the average deterioration in speeds\nover given road surface/condition combinations under monsoon conditions.\n\nA separate interim model of off-road, off-path walking speeds was then prepared. To do so, we first calculated the slope for each grid cell in the SRTM 1 arc-second (≈ 30 m <sup>2</sup> ) digital elevation model. These slopes\nwere input into Tobler’s hiking formula to derive a base walking speed for each grid cell. We multiplied this\nspeed by a fractional landcover walking speed modifier using a 1 km <sup>2</sup> gridded landcover model of Nepal from\nthe International Centre for Integrated Mountain Development (ICIMOD, 2013). A downward adjustment\nin walking speeds was applied to extreme altitudes (greater than 3500 m) where lower oxygen counts reduce\n\nMention: \"roads surfaces and conditions data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Surface type and classification determine the speed assigned to each segment.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002050:12:0", "occurrence_id": "9d80d32397ea427d5b44cfac", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 5: Roads do not exist in isolation, but as part of networks, and some proposed roads may depend on the construction of\nother roads - resulting in internal hierarchies (tiers). Numerous possible combinations exist for such roads when selected with\nconsideration given to their tier rank, and the resulting combinations of roads are called sequences.\n\n\n3.4. Cost-time modeling\n\n\nThe first component of our methodology involves calculating the accessibility to services by means of\ncost-time models. The backbone of our cost-time models comes from a layer of on-road/on-path speeds. The\nprincipal data input is a transportation layer produced by merging roads and paths from the Nepal Survey\nDepartment (2019) and OpenStreetMap with roads surfaces and conditions data compiled in 2015 for the\nRural Access Index. Road segments were assigned speeds based on their surface type and Department of\nRoads classification (Department of Roads, 2013). These speeds were then modified according to the condition\nand gradient of the road. Using modifiers provided by a focus group of World Food Programme civil engineers,\nwe then calculated a separate monsoon speed for each segment based on the average deterioration in speeds\nover given road surface/condition combinations under monsoon conditions.\n\nA separate interim model of off-road, off-path walking speeds was then prepared. To do so, we first calculated the slope for each grid cell in the SRTM 1 arc-second (≈ 30 m <sup>2</sup> ) digital elevation model. These slopes\nwere input into Tobler’s hiking formula to derive a base walking speed for each grid cell. We multiplied this\nspeed by a fractional landcover walking speed modifier using a 1 km <sup>2</sup> gridded landcover model of Nepal from\nthe International Centre for Integrated Mountain Development (ICIMOD, 2013). A downward adjustment\nin walking speeds was applied to extreme altitudes (greater than 3500 m) where lower oxygen counts reduce\n\nMention: \"SRTM\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Slopes from this elevation model feed the hiking formula for off-path walking speed.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002050:12:0", "occurrence_id": "a63425761c386cb4d62b6d2d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 5: Roads do not exist in isolation, but as part of networks, and some proposed roads may depend on the construction of\nother roads - resulting in internal hierarchies (tiers). Numerous possible combinations exist for such roads when selected with\nconsideration given to their tier rank, and the resulting combinations of roads are called sequences.\n\n\n3.4. Cost-time modeling\n\n\nThe first component of our methodology involves calculating the accessibility to services by means of\ncost-time models. The backbone of our cost-time models comes from a layer of on-road/on-path speeds. The\nprincipal data input is a transportation layer produced by merging roads and paths from the Nepal Survey\nDepartment (2019) and OpenStreetMap with roads surfaces and conditions data compiled in 2015 for the\nRural Access Index. Road segments were assigned speeds based on their surface type and Department of\nRoads classification (Department of Roads, 2013). These speeds were then modified according to the condition\nand gradient of the road. Using modifiers provided by a focus group of World Food Programme civil engineers,\nwe then calculated a separate monsoon speed for each segment based on the average deterioration in speeds\nover given road surface/condition combinations under monsoon conditions.\n\nA separate interim model of off-road, off-path walking speeds was then prepared. To do so, we first calculated the slope for each grid cell in the SRTM 1 arc-second (≈ 30 m <sup>2</sup> ) digital elevation model. These slopes\nwere input into Tobler’s hiking formula to derive a base walking speed for each grid cell. We multiplied this\nspeed by a fractional landcover walking speed modifier using a 1 km <sup>2</sup> gridded landcover model of Nepal from\nthe International Centre for Integrated Mountain Development (ICIMOD, 2013). A downward adjustment\nin walking speeds was applied to extreme altitudes (greater than 3500 m) where lower oxygen counts reduce\n\nMention: \"gridded landcover model of Nepal\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Its landcover classes provide fractional modifiers that adjust walking speed per cell.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002075:8:0", "occurrence_id": "2459af8cbe72fd0495171a2a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 4). The World Bank collects these sectoral value added data from national accounts and converts them\n\nfrom local currency units to constant prices in 2010 U.S. dollars. The disaggregation of the national value\n\nadded into three sectors accords with the sectoral definition of the ISIC 4 by the United Nations as\n\nfollows. <sup>7</sup> [^7: For a detailed description of these classifications see:\nUnited Nations Statistics Division. (2002). International Standard Industrial Classification of All Economic Activities\n(ISIC) Revision 3.1.\nUnited Nations Statistics Division. (2008). International Standard Industrial Classification of All Economic Activities\nRevision 4.] Agriculture value added contains ISIC 4 sectors 01-03: forestry and hunting, fishing, cultivation\n\nof crops and livestock production; manufacturing or industry value added consists of ISIC 4 sectors 05-43:\n\nmining, manufacturing, utilities, and construction; and services value added covers the remaining ISIC 4\n\nsectors 45-99: wholesale and retail trade, transport, government, financial, professional, personal\n\nservices, and miscellaneous value added, such as imputed bank services and import duties.\n\n\nThe IEA provides sector-level electricity consumption by using a slightly different composition for the\n\nmanufacturing and service sectors, while applying the identical definition for the agriculture sector. The\n\nIEA compiles the data on the country-sector level by using national sources. <sup>8</sup> [^8: The IEA’s World Energy Balances 2018 Edition – Database Documentation provides a country-by-country\ndescription of sources for energy and electricity generation and consumption.] It defines the agriculture\n\nsector according to the ISIC 4 and includes sub-sectors 01-03, where the electricity consumption for the\n\nfishing industry also contains the provision of electricity to ships refueling in the country. Furthermore,\n\nthe manufacturing sector comprises the ISIC 4 codes 07-18, 20-32, and 41-43, notably leaving out sub\nsectors 33-40 and including most of them in the services sector. From IEA data, the services sector\n\ncombines separately provided transport and service sectors with their respective ISIC 4 codes: 49-51\n\nMention: \"sectoral value added data from national accounts\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"The World Bank collects these data and converts them from local currency to constant 2010 dollars.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002075:8:0", "occurrence_id": "5e8bb7fd52636a92ac62f350", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 4). The World Bank collects these sectoral value added data from national accounts and converts them\n\nfrom local currency units to constant prices in 2010 U.S. dollars. The disaggregation of the national value\n\nadded into three sectors accords with the sectoral definition of the ISIC 4 by the United Nations as\n\nfollows. <sup>7</sup> [^7: For a detailed description of these classifications see:\nUnited Nations Statistics Division. (2002). International Standard Industrial Classification of All Economic Activities\n(ISIC) Revision 3.1.\nUnited Nations Statistics Division. (2008). International Standard Industrial Classification of All Economic Activities\nRevision 4.] Agriculture value added contains ISIC 4 sectors 01-03: forestry and hunting, fishing, cultivation\n\nof crops and livestock production; manufacturing or industry value added consists of ISIC 4 sectors 05-43:\n\nmining, manufacturing, utilities, and construction; and services value added covers the remaining ISIC 4\n\nsectors 45-99: wholesale and retail trade, transport, government, financial, professional, personal\n\nservices, and miscellaneous value added, such as imputed bank services and import duties.\n\n\nThe IEA provides sector-level electricity consumption by using a slightly different composition for the\n\nmanufacturing and service sectors, while applying the identical definition for the agriculture sector. The\n\nIEA compiles the data on the country-sector level by using national sources. <sup>8</sup> [^8: The IEA’s World Energy Balances 2018 Edition – Database Documentation provides a country-by-country\ndescription of sources for energy and electricity generation and consumption.] It defines the agriculture\n\nsector according to the ISIC 4 and includes sub-sectors 01-03, where the electricity consumption for the\n\nfishing industry also contains the provision of electricity to ships refueling in the country. Furthermore,\n\nthe manufacturing sector comprises the ISIC 4 codes 07-18, 20-32, and 41-43, notably leaving out sub\nsectors 33-40 and including most of them in the services sector. From IEA data, the services sector\n\ncombines separately provided transport and service sectors with their respective ISIC 4 codes: 49-51\n\nMention: \"IEA data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Its separately reported transport and services sectors are combined to match the classification.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002088:18:0", "occurrence_id": "27bce769b2f6fc0da0ee7be8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Early empirical evidence in favor of the inverse-yield relationship prompted criticism, since\n\nthe early analyses omitted potentially important information about natural endowments. Moreover,\n\nsystematic errors in the survey methods were thought to create biased results. As a consequence,\n\nempirical studies became increasingly focused on the variation in plot yields for a given farm and\n\nbased on detailed data with a limited geographic reach. This created tension, since what matters for\n\npolicy is the generalization that small farms are an appropriate and a natural outcome of the\n\ninterplay of endowments and markets. In contrast, researchers look for settings that limit the scope\n\nfor unmeasured influences that undermine the credibility of results, and thereby narrow the\n\ngeographic relevance of the results. <sup>17</sup> [^17: At least, this is the case in the narrow applications of individual studies. Collectively, we believe the accumulated\nevidence cited in the front of the paper is consistent with our findings.]\n\n\nIn this paper, we develop a model of endogenous technology choice in which farming\n\nmethods (applied technologies) are an outcome of the economic environment in which farmers\n\noperate. One consequence is that heterogeneity in farming households, farm endowments and\n\nmarkets generate heterogeneity in applied technologies. Econometrically, the underlying production\n\nfunction associated with a given applied technology is identified by the relevant state variables, and\n\nincompleteness in the observed state variables introduces a greater likelihood of mis-identification.\n\nWith this in mind, we develop an applied model to measure the inverse relationship between maize\n\nyields and production scale in Sub-Saharan Africa. To estimate the model, we progress from coarse\n\ndata that is geographically and temporally diverse to more granular and place-specific data, with the\n\npresumption that the opportunity for omitting relevant information about the state variables is\n\nreduced. We find empirical evidence that supports the hypothesis that maize yields fall with\n\nincreases in the scale of production in all settings. Moreover, once extreme yield values are expunged\n\nfrom the data, the estimated elasticities measuring the inverse relationship are remarkably similar\n\nacross the assorted data platforms.\n\n\nAt the same time,\n\nMention: \"detailed data\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"validate\", \"summary\": \"Its limited geographic reach makes generalisation for policy difficult.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002088:18:0", "occurrence_id": "1ea9a7551819f3c4f92102c2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Early empirical evidence in favor of the inverse-yield relationship prompted criticism, since\n\nthe early analyses omitted potentially important information about natural endowments. Moreover,\n\nsystematic errors in the survey methods were thought to create biased results. As a consequence,\n\nempirical studies became increasingly focused on the variation in plot yields for a given farm and\n\nbased on detailed data with a limited geographic reach. This created tension, since what matters for\n\npolicy is the generalization that small farms are an appropriate and a natural outcome of the\n\ninterplay of endowments and markets. In contrast, researchers look for settings that limit the scope\n\nfor unmeasured influences that undermine the credibility of results, and thereby narrow the\n\ngeographic relevance of the results. <sup>17</sup> [^17: At least, this is the case in the narrow applications of individual studies. Collectively, we believe the accumulated\nevidence cited in the front of the paper is consistent with our findings.]\n\n\nIn this paper, we develop a model of endogenous technology choice in which farming\n\nmethods (applied technologies) are an outcome of the economic environment in which farmers\n\noperate. One consequence is that heterogeneity in farming households, farm endowments and\n\nmarkets generate heterogeneity in applied technologies. Econometrically, the underlying production\n\nfunction associated with a given applied technology is identified by the relevant state variables, and\n\nincompleteness in the observed state variables introduces a greater likelihood of mis-identification.\n\nWith this in mind, we develop an applied model to measure the inverse relationship between maize\n\nyields and production scale in Sub-Saharan Africa. To estimate the model, we progress from coarse\n\ndata that is geographically and temporally diverse to more granular and place-specific data, with the\n\npresumption that the opportunity for omitting relevant information about the state variables is\n\nreduced. We find empirical evidence that supports the hypothesis that maize yields fall with\n\nincreases in the scale of production in all settings. Moreover, once extreme yield values are expunged\n\nfrom the data, the estimated elasticities measuring the inverse relationship are remarkably similar\n\nacross the assorted data platforms.\n\n\nAt the same time,\n\nMention: \"coarse\n\ndata\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Geographically and temporally diverse observations are the first stage of estimating the yield-scale relationship.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002106:43:0", "occurrence_id": "07e0e0a08bdffa7b595cec93", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: technical appendix        - methodology and model description 42\n\n\nTable 5 lists the aggregated household expenditures (AHE) for each district in Sri\n\n\nLanka.\n\n\n**AHE** **National**\n\n**accounts**\n**District** <u>(bil.</u> <u>LKR/year)</u>\n\n\nColombo 609.8          Gampaha 460.3          Kalutara 233.9            Kandy 218.3          Matale 69.0            Nuwara Eliya 88.3            Galle 172.8             Matara 116.1           Hambantota 109.4           Jaffna 66.8              Mannar 13.5          Vavuniya 25.4           Mullaitivu 9.0             Kilinochchi 9.8             Batticaloa 49.8              Ampara 81.2          Trincomalee 45.3            Kurunegala 288.8           Puttalam 135.4            Anuradhapura 129.1           Polonnaruwa 61.3           Badulla 102.8            Moneragala 52.1           Rathnapura 135.2           <u>Kegalle</u> 118.1 <u>–</u>\n\n\nTotal 3,401.6 11,400.0\n\n\n**Table** **5:** Aggregated household expenditures (AHE) at district level. At left, we sum total\n\nexpenditures as reported in the 2016 HIES; at right, we present the nominal GDP of\nSri Lanka. Values are expressed in millions of LKR per year.\n\n\nIn the following sections, we will trace the impacts of disasters on household assets\n\n\nand wellbeing through the following steps:\n\n\n1. Disasters result in losses to households’ effective capital stock (∆k <sup>eff</sup> h <sup>).</sup>\n\n\n2. The diminished asset base generates less income (∆ih).\n\nMention: \"the 2016 HIES\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Reported expenditures are aggregated to districts and compared with nominal GDP.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002112:23:2", "occurrence_id": "b787a9f83bbccb242f4e6297", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: , and Smart, J. 2018. “Ethiopia’s Spatial and Structural Transformation:\n\nPublic Policy and Drivers of Change.” IFPRI/EDRI Strategy Support Program Working Paper 119.\n\nWieser, C., A. A. Ambel, T. Bundervoet, and A. Haile. 2020. _Monitoring COVID-19 Impacts on Households_\n\n_in Ethiopia: Results from a High-Frequency Phone Survey of Households._ World Bank.\n\nWorld Bank. 2020a. _Poverty and Shared Prosperity 2020: Reversals of Fortune._ Washington, DC: World\n\nBank.\n\nWorld Bank. 2020b _Monetary Poverty and Shared Prosperity in Kinshasa_ . Washington, DC: World Bank.\n\nWorld Bank. 2020c. _Profiling Living Conditions of DRC Urban Population: Access to Housing and Services_\n\n_in Kinshasa Province._ Washington, DC: World Bank.\n\n\n22\n\nMention: \"High-Frequency Phone Survey of Households\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"A reference entry for the report monitoring pandemic impacts on households.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002115:15:0", "occurrence_id": "05cc18a61a39aa2d3fbd9372", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 2 Financial performance of SOEs in energy-rich economies, 2012–18**\n\n**a.** **EBITDA** **b.** **Adjusted EBITDA**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n**c.** **Capital expenditures** **d.** **Operating expenditures**\n\n\n\n\n\n\n\n\n\n\n\n_Source_ : World Bank Infrastructure SOEs Database and authors’ calculations.\n_Note_ : EBITDA: Earnings before interest, taxes, depreciation, and amortization. FSOE: Fully owned state-owned\nenterprise. PPSOE: partially privatized state-owned enterprise.\n\nFigure 3 shows the evolution of fiscal support for PPSOEs and FSOEs in energy-rich countries. The\naverage FSOE has received more operations subsidies as percent of assets than the average\nPPSOEs since 2013 (panel a.). Operational subsidies over assets increased sharply in 2012–14 for\nall SOEs, reaching almost 90 percent of average assets for the mean SOE, followed by a slight\ndecrease after 2014. In the case of overall fiscal injections, there is significant volatility in the\nperiod of analysis, particularly for PPSOEs, with the mean fiscal injections to average assets for\nall SOEs ranging between 2 and 6 (panel b.).\n\n\n14\n\nMention: \"World Bank Infrastructure SOEs Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Earnings, capital and operating expenditure of state-owned infrastructure enterprises are drawn from it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002122:26:0", "occurrence_id": "783751bd1ae28fea19c300f4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Finally, we investigate if the distance to the technological frontier also matters for the impacts of\nreforms on the within and structural change components. The results for both components are\nreported in tables A.5 and A.6 A in the appendix, respectively. Like the previous estimations, we\nfind that current account, domestic finance, banking, and securities reforms all have positive\neffects on the within component (Table A.5). Still, these effects decline as the productivity gap\nwith the technological frontier reduces. Table A.6 repeats the same exercise using the structural\nchange component as the dependent variable. Most of the estimated coefficients are not\nsignificant. Still, we find, in a few cases, a positive sign on the coefficient of reforms and a negative\nsign on the interaction term between product markets and trade reforms and technological distance\nto the US, although the level of significance is 10 percent.\n\n#### **6. Concluding remarks**\n\nIt is widely recognized that productivity is an important determinant of countries’ economic\nperformance in the long run. Understanding which policies raise productivity growth in developing\ncountries is essential, given the low level of productivity and a limited number of studies. This\npaper adds to the limited literature on the impacts of structural reforms on productivity growth in\nlow-income and lower-middle-income countries. It pays particular attention to how reforms affect\nthe movement of labor across sectors of the economy, i.e., structural change. The empirical\nanalysis combines the data set on structural reforms from Prati et al. (2013) with a sectoral database\nfrom the GGDC 10-sector database and EASD to compute productivity growth and structural\nchange. Our findings show that financial sector reforms have greater effects on the growth rate of\nlabor productivity than product market and trade reforms. However, reforms affect the within and\nthe structural change contributions differently. Financial sector reforms have positive effects on\nthe within component of the productivity growth rate.\n\n\nIn contrast, the different reforms\n\nMention: \"data set on structural reforms\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its reform dates are combined with sectoral data to study structural change.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002122:26:0", "occurrence_id": "327a666dde2ead4089e8a0be", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Finally, we investigate if the distance to the technological frontier also matters for the impacts of\nreforms on the within and structural change components. The results for both components are\nreported in tables A.5 and A.6 A in the appendix, respectively. Like the previous estimations, we\nfind that current account, domestic finance, banking, and securities reforms all have positive\neffects on the within component (Table A.5). Still, these effects decline as the productivity gap\nwith the technological frontier reduces. Table A.6 repeats the same exercise using the structural\nchange component as the dependent variable. Most of the estimated coefficients are not\nsignificant. Still, we find, in a few cases, a positive sign on the coefficient of reforms and a negative\nsign on the interaction term between product markets and trade reforms and technological distance\nto the US, although the level of significance is 10 percent.\n\n#### **6. Concluding remarks**\n\nIt is widely recognized that productivity is an important determinant of countries’ economic\nperformance in the long run. Understanding which policies raise productivity growth in developing\ncountries is essential, given the low level of productivity and a limited number of studies. This\npaper adds to the limited literature on the impacts of structural reforms on productivity growth in\nlow-income and lower-middle-income countries. It pays particular attention to how reforms affect\nthe movement of labor across sectors of the economy, i.e., structural change. The empirical\nanalysis combines the data set on structural reforms from Prati et al. (2013) with a sectoral database\nfrom the GGDC 10-sector database and EASD to compute productivity growth and structural\nchange. Our findings show that financial sector reforms have greater effects on the growth rate of\nlabor productivity than product market and trade reforms. However, reforms affect the within and\nthe structural change contributions differently. Financial sector reforms have positive effects on\nthe within component of the productivity growth rate.\n\n\nIn contrast, the different reforms\n\nMention: \"GGDC 10-sector database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its sectoral data supply the productivity growth measures.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002122:26:0", "occurrence_id": "00ecd1dcfaa683863f93bf20", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Finally, we investigate if the distance to the technological frontier also matters for the impacts of\nreforms on the within and structural change components. The results for both components are\nreported in tables A.5 and A.6 A in the appendix, respectively. Like the previous estimations, we\nfind that current account, domestic finance, banking, and securities reforms all have positive\neffects on the within component (Table A.5). Still, these effects decline as the productivity gap\nwith the technological frontier reduces. Table A.6 repeats the same exercise using the structural\nchange component as the dependent variable. Most of the estimated coefficients are not\nsignificant. Still, we find, in a few cases, a positive sign on the coefficient of reforms and a negative\nsign on the interaction term between product markets and trade reforms and technological distance\nto the US, although the level of significance is 10 percent.\n\n#### **6. Concluding remarks**\n\nIt is widely recognized that productivity is an important determinant of countries’ economic\nperformance in the long run. Understanding which policies raise productivity growth in developing\ncountries is essential, given the low level of productivity and a limited number of studies. This\npaper adds to the limited literature on the impacts of structural reforms on productivity growth in\nlow-income and lower-middle-income countries. It pays particular attention to how reforms affect\nthe movement of labor across sectors of the economy, i.e., structural change. The empirical\nanalysis combines the data set on structural reforms from Prati et al. (2013) with a sectoral database\nfrom the GGDC 10-sector database and EASD to compute productivity growth and structural\nchange. Our findings show that financial sector reforms have greater effects on the growth rate of\nlabor productivity than product market and trade reforms. However, reforms affect the within and\nthe structural change contributions differently. Financial sector reforms have positive effects on\nthe within component of the productivity growth rate.\n\n\nIn contrast, the different reforms\n\nMention: \"EASD\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It provides complementary sectoral data for East Asia.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002123:7:2", "occurrence_id": "4766f717df3497338747d333", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <br>|Output gap<br> <br>Change in terms of trade (log).<br>|WEO<br> <br>WEO<br>|\n|Pillar 5: Health<br>Pillar 6: Skills<br>|Human capitalc <br>|Index of human capital per person, based on years of schooling (Barro<br>and Lee 2013) and returns to education (Psacharopoulos 1994)<br>|Penn World Table<br>|\n|Pillar 7: Product<br>market|Taxesd <br> <br>|Top marginal income and payroll tax rate, percent<br>|Fraser (2016)<br>|\n|Pillar 7: Product<br>market|Regulatory qualitye|Measures perceptions of the ability of the government to formulate<br>and implement sound policies and regulations that permit and<br>promote private sector development.|WGI (2016)<br>|\n\n\n\n3The tenth pillar in the WEF GCI classification is market size (proxied by either GDP or volume of imports), a variable\nthat either coincides or is highly correlated with the dependent variable.\n\n6\n\nMention: \"Penn World Table\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It provides the human capital index based on schooling years and returns to education.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002093:5:1", "occurrence_id": "e446a11dead9860de44da7bb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: tho<br>becomes||||<br>AGOA wearing apparel<br>~~provisions~~|\n|<br>eligible for<br>~~AGOA~~|||||\n|<br>preferences|<br>preferences||||\n|<br>preferences|<br>preferences||||\n|<br>preferences|<br>preferences||||\n|<br>preferences|<br>preferences||||\n\n\n\nExports of goods and services ('000, BoP current US$) GDP per capita (current US$)\n\n\nSource: World Bank World Development Indicators\n\nHowever, over a decade of export-led growth has resulted in only limited poverty reduction, while\ninequality has worsened (World Bank 2017). <sup>3</sup> Moreover, the dynamism of Lesotho’s export sectors has in\nrecent years stagnated and contributed to a significant worsening of its macroeconomic situation. GDP\ngrowth has remained at or below 5% since 2008. In terms of sectoral contribution, services have made a far\nmore significant impact on growth than other sectors and following the discovery of significant diamond\n\n\n3 While Lesotho has seen some poverty reduction since the start of the century, there is little reliable data on the magnitude of this\ndecline, and it is assumed to have been rather modest. The headcount poverty rate ($1.90/ day PPP) fell from 61.3% in 2002 to\n59.7 in 2011 and estimates suggest that 56.3% were in poverty in 2016. According to the most recent World Bank Macro Poverty\nOutlook (World Bank 2017), the slow pace of poverty reduction is likely to continue in coming years due to fiscal consolidation,\ndeclining remittances and droughts.\n\n\n4\n\nMention: \"World Bank Macro Poverty\nOutlook\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its projections indicate the slow pace of poverty reduction will continue.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002099:17:1", "occurrence_id": "0f8b0e2b5661cccb3b0b9470", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  the Uganda national census by the\nUgandan Bureau of Statistics (UBOS). <sup>11</sup> We also control for a vector of socio-economic and\n\ndemographic variables _Xi_ to limit the risk of omitted variable bias. We use two approaches\nto select the variables included in _Xi_ . First, we selected a list of variables that could explain\n\nperceptions or interactions based on our knowledge of the literature (see Tables A.3 and A.4\n\nin the Appendix for descriptive statistics). We include enumeration areas fixed effects ( _µEA_ )\n\nor household fixed effects ( _µHH_ ) to control for unobserved characteristics that are constant\n\nwithin EAs/households. Second, we implement the double LASSO method of Chernozhukov\n\net al. (2017) using the original list of potential controls augmented with fourth-degree poly\nnomials along with all first-order interactions and the inverse hyperbolic sine transformation\n\nof continuous variables (Knaus et al. 2020). We also estimate specifications without control\n\nvariables.\n\n\nThree main methodological challenges could afflict this OLS regression  - and more gen\n\n10 _PHH_ and _PEA_ are standardised to facilitate interpretation. Before standardisation, the variable _PHH_ is\nset as equal to 0 (= the mean of the perception index _Pi_ ) in households with only one adult.\n\n11When describing host perceptions across all sites in the later OLS analysis without a measure for\ninteraction (Table 4), the neighbourhoods are defined by similar Enumeration Areas from the Kenyan\ncensus for respondents in Nairobi, by villages around Kakuma, by the five main towns close to each of the\nfive Dollo Ado camps, and by specific sublocations to the official woredas in Addis Ababa.\n\n\n15\n\nMention: \"Kenyan\ncensus\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its enumeration areas define comparable neighbourhoods for Nairobi respondents.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002110:12:1", "occurrence_id": "e565fbe369e34dd1a967633e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  direct investment, portfolio investment, and other investments\n\n(comprising largely bank lending) by non-residents net of their own disinvestment in the\n\nreporting economy. <sup>8</sup> Similarly, gross private outflows refer to flows by residents, net of their\n\nown disinvestment. We use quarterly net capital inflows as our main measure. Quarterly gross\n\ncapital inflows are used to check the robustness of our results to the alternative definition of\n\n\n6 The absolute ratings and relative ratings have the same cardinal measure but there is no strict correspondence in\nterms of the values.\n7 The data collection exercise for capital flows addresses two challenges: 1) Consistency of definitions for specific\ncategories of flows (for instance, there are differences in definitions and, therefore, values between the various\nsub-components of portfolio flows and debt reported by World Bank World Development Indicators, Global\nEconomic Prospects and the IMF Balance of Payment statistics); 2) Issues due to change in Balance of Payments\nstatistics guidelines. Historical capital flows data currently available from the IMF Balance of Payments statistics\n(on which the World Bank World Development Indicators data are largely based) are currently inconsistent\nbetween the pre-2005 and post-2005 periods, given the transition from BPM5 to BPM6 definitions.\n8 These are standard components used to aggregate capital flows in the literature (see Ahmed and Zlate, 2014).\n\n\n11\n\nMention: \"World Bank World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its definitions of portfolio flow and debt sub-components differ from the other compilations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002110:12:1", "occurrence_id": "63317a6b64bd42a04d42494c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  direct investment, portfolio investment, and other investments\n\n(comprising largely bank lending) by non-residents net of their own disinvestment in the\n\nreporting economy. <sup>8</sup> Similarly, gross private outflows refer to flows by residents, net of their\n\nown disinvestment. We use quarterly net capital inflows as our main measure. Quarterly gross\n\ncapital inflows are used to check the robustness of our results to the alternative definition of\n\n\n6 The absolute ratings and relative ratings have the same cardinal measure but there is no strict correspondence in\nterms of the values.\n7 The data collection exercise for capital flows addresses two challenges: 1) Consistency of definitions for specific\ncategories of flows (for instance, there are differences in definitions and, therefore, values between the various\nsub-components of portfolio flows and debt reported by World Bank World Development Indicators, Global\nEconomic Prospects and the IMF Balance of Payment statistics); 2) Issues due to change in Balance of Payments\nstatistics guidelines. Historical capital flows data currently available from the IMF Balance of Payments statistics\n(on which the World Bank World Development Indicators data are largely based) are currently inconsistent\nbetween the pre-2005 and post-2005 periods, given the transition from BPM5 to BPM6 definitions.\n8 These are standard components used to aggregate capital flows in the literature (see Ahmed and Zlate, 2014).\n\n\n11\n\nMention: \"Global\nEconomic Prospects\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its reported values differ from the other two compilations for the same flows.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002110:12:1", "occurrence_id": "ec71426750ef0874da66e418", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  direct investment, portfolio investment, and other investments\n\n(comprising largely bank lending) by non-residents net of their own disinvestment in the\n\nreporting economy. <sup>8</sup> Similarly, gross private outflows refer to flows by residents, net of their\n\nown disinvestment. We use quarterly net capital inflows as our main measure. Quarterly gross\n\ncapital inflows are used to check the robustness of our results to the alternative definition of\n\n\n6 The absolute ratings and relative ratings have the same cardinal measure but there is no strict correspondence in\nterms of the values.\n7 The data collection exercise for capital flows addresses two challenges: 1) Consistency of definitions for specific\ncategories of flows (for instance, there are differences in definitions and, therefore, values between the various\nsub-components of portfolio flows and debt reported by World Bank World Development Indicators, Global\nEconomic Prospects and the IMF Balance of Payment statistics); 2) Issues due to change in Balance of Payments\nstatistics guidelines. Historical capital flows data currently available from the IMF Balance of Payments statistics\n(on which the World Bank World Development Indicators data are largely based) are currently inconsistent\nbetween the pre-2005 and post-2005 periods, given the transition from BPM5 to BPM6 definitions.\n8 These are standard components used to aggregate capital flows in the literature (see Ahmed and Zlate, 2014).\n\n\n11\n\nMention: \"IMF Balance of Payment statistics\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its definitions changed, leaving historical capital flow series inconsistent across periods.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002110:12:1", "occurrence_id": "0df5e1ff82c838b2af88eb39", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  direct investment, portfolio investment, and other investments\n\n(comprising largely bank lending) by non-residents net of their own disinvestment in the\n\nreporting economy. <sup>8</sup> Similarly, gross private outflows refer to flows by residents, net of their\n\nown disinvestment. We use quarterly net capital inflows as our main measure. Quarterly gross\n\ncapital inflows are used to check the robustness of our results to the alternative definition of\n\n\n6 The absolute ratings and relative ratings have the same cardinal measure but there is no strict correspondence in\nterms of the values.\n7 The data collection exercise for capital flows addresses two challenges: 1) Consistency of definitions for specific\ncategories of flows (for instance, there are differences in definitions and, therefore, values between the various\nsub-components of portfolio flows and debt reported by World Bank World Development Indicators, Global\nEconomic Prospects and the IMF Balance of Payment statistics); 2) Issues due to change in Balance of Payments\nstatistics guidelines. Historical capital flows data currently available from the IMF Balance of Payments statistics\n(on which the World Bank World Development Indicators data are largely based) are currently inconsistent\nbetween the pre-2005 and post-2005 periods, given the transition from BPM5 to BPM6 definitions.\n8 These are standard components used to aggregate capital flows in the literature (see Ahmed and Zlate, 2014).\n\n\n11\n\nMention: \"IMF Balance of Payments statistics\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"It is the underlying provider for the other compilation, so both inherit the same break.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002110:12:1", "occurrence_id": "e84d3811d4385753d8494e98", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  direct investment, portfolio investment, and other investments\n\n(comprising largely bank lending) by non-residents net of their own disinvestment in the\n\nreporting economy. <sup>8</sup> Similarly, gross private outflows refer to flows by residents, net of their\n\nown disinvestment. We use quarterly net capital inflows as our main measure. Quarterly gross\n\ncapital inflows are used to check the robustness of our results to the alternative definition of\n\n\n6 The absolute ratings and relative ratings have the same cardinal measure but there is no strict correspondence in\nterms of the values.\n7 The data collection exercise for capital flows addresses two challenges: 1) Consistency of definitions for specific\ncategories of flows (for instance, there are differences in definitions and, therefore, values between the various\nsub-components of portfolio flows and debt reported by World Bank World Development Indicators, Global\nEconomic Prospects and the IMF Balance of Payment statistics); 2) Issues due to change in Balance of Payments\nstatistics guidelines. Historical capital flows data currently available from the IMF Balance of Payments statistics\n(on which the World Bank World Development Indicators data are largely based) are currently inconsistent\nbetween the pre-2005 and post-2005 periods, given the transition from BPM5 to BPM6 definitions.\n8 These are standard components used to aggregate capital flows in the literature (see Ahmed and Zlate, 2014).\n\n\n11\n\nMention: \"World Bank World Development Indicators data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"It largely reproduces the balance of payments source and so is inconsistent across the definitional transition.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002121:6:1", "occurrence_id": "b415930771ecc6bef9795bb1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: , they find that growth was driven by static structural\n\nchange and within-sector productivity growth. However, productivity growth was low in the sub\nregion over the period of analysis.\n\n\nde Vries _et al_ . (2015) analyze structural change in Africa and its implications for productivity\n\ngrowth during the past 50 years. According to this paper, during the early post-independence\n\n\n3 DHS are nationally representative surveys designed to collect detailed information on child mortality, health, and\nfertility, as well as on households’ durables and quality of dwellings. In addition, DHS data include information on\ngender, age, location, education, employment status, and occupation of women and their partners between the ages of\n15 and 59.\n\n5\n\nMention: \"3 DHS\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"Described as nationally representative surveys collecting mortality, health, fertility and dwelling information.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002121:6:1", "occurrence_id": "f4f33c9abea8a7cad7ff1a9d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: , they find that growth was driven by static structural\n\nchange and within-sector productivity growth. However, productivity growth was low in the sub\nregion over the period of analysis.\n\n\nde Vries _et al_ . (2015) analyze structural change in Africa and its implications for productivity\n\ngrowth during the past 50 years. According to this paper, during the early post-independence\n\n\n3 DHS are nationally representative surveys designed to collect detailed information on child mortality, health, and\nfertility, as well as on households’ durables and quality of dwellings. In addition, DHS data include information on\ngender, age, location, education, employment status, and occupation of women and their partners between the ages of\n15 and 59.\n\n5\n\nMention: \"DHS data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"It adds gender, age, location, education and occupation details for women and their partners.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002129:47:2", "occurrence_id": "dae233ef4d2153cf4cbcd0ab", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 01 0.10 0.00 0.00 1.00 3237\n\n\nHH head is literate ERIVaS 0.08 0.28 0.00 0.00 1.00 577\nEACI 2017 0.20 0.40 0.00 0.00 1.00 3237\n<u>***</u>\n\n\n_Notes:_ To ensure comparability of the variables, the EACI 2017 sample is limited to pure stand plots.\n\nThe stars below the sample mean values of the two surveys denotes the statistical significance\nof the difference between the two means: <sup>_∗_</sup> p _<_ 0.1; <sup>_∗∗_</sup> p _<_ 0.05; <sup>_∗∗∗_</sup> p _<_ 0.01\n\n\n46\n\nMention: \"EACI 2017\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Plot-level variables are restricted to pure stands so they are comparable with the other source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002131:3:0", "occurrence_id": "d2d29d7aa89cebe0a0018d86", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Introduction**\n\n\nEconomists explain the closure of firms during recessions with Schumpeter’s (1934) creative\n\ndestruction theory, where during downturns small and less efficient firms are the ones to exit the\n\nmarket. In times of extreme economic distress, however, firms in every country are reeling from\n\nthe inability to do business as usual. To make things worse, many sectors see collapsed demand\n\nand economic uncertainty stretching months, if not years. In the current pandemic, governments\n\nrightly focus on dealing with the health aspects first, and only then on the recovery of the economy\n\nonce the immediate danger of the pandemic is over.\n\n\nIn the meantime, businesses are rapidly running out of cash. In the United States, half of\n\nsmall firms – those with fewer than 500 employees – have cash reserves for less than a month, and\n\nanother quarter of businesses may run out of cash in two months. For service industries, the period\n\nto illiquidity is even shorter. Restaurants, for example, have less than a month of cash in hand\n\n(Didier et al., 2020).\n\n\nThis breathing period is extended with government programs already in place to support\n\nworker retention through subsidizing jobs, freezing interest payments on loans, and extending new\n\nbank credit. This extension differs across industries – it helps labor-intensive sectors more, and\n\nfirms with established lines of credit benefit more as well. Still, other payments – like rent and cost\n\nof materials – are weighing on businesses. Exporters are unable to ship goods due to disrupted\n\ntransport links. Even when transport is possible, new trade restrictions may apply or demand has\n\nsimply collapsed.\n\n\nNew analysis on 12 high-income and middle-income economies across Africa, Central Asia,\n\nEurope, Latin America, and the Middle East reflects similar patterns. The analysis uses a sample\n\nof 6,345 firms from the World Bank’s Enterprises Surveys in Colombia, Greece, Italy, Jordan,\n\nKazakhstan, Kenya, Morocco, Peru, Portugal, the\n\nMention: \"the World Bank’s Enterprises Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"A sample of 6,345 firms across many countries supplies the analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002133:6:0", "occurrence_id": "778d4ca49ff96e91d4f0b9b5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ##### **II. Data Description**\n\nFacebook, in partnership with the World Bank and the Organisation for Economic Co-Operation and\nDevelopment (OECD), conducts a bi-annual global survey on small and medium-sized enterprises (SMEs)\nthat use Facebook. This _Future of Business_ survey currently includes 97 countries across every geographic\nregion. <sup>5</sup> [^5: Map presented in Appendix A displays the geographic coverage of the December 2018 wave of the _Future of Business_ Survey\nand the table in Appendix B provides information for all the countries included in the analysis plus the percentage of female\nrespondents per each country.]\n\nThe _Future of Business_ (FoB) survey instrument was redesigned for the December 2018 wave to contain a\nset of questions capturing firm performance outcomes, as well as information from the business owner or\nmanager including his/her background, family history, education, skills, socioeconomic characteristics and\nother variables. <sup>6</sup> [^6: Facebook did not share any platform data that were not collected through this particular survey round. In the analysis for this\npaper, we use only anonymized survey responses.]\n\nThe survey was offered, by Facebook, to administrators <sup>7</sup> [^7: Note some categories of businesses were excluded (governments etc.) from the sampling frame.] of Facebook-designated SME pages in the\nspecified countries through Facebook for a period of two weeks in early to mid-December. The data were\ncollected on the Facebook platform and the final clean and aggregated, anonymized data are now published\non the World Bank Open Data portal. <sup>8</sup> [^8: World Bank Open Data portal: https://datacatalog.worldbank.org/dataset/future-business-survey-aggregated-data]\n\nThe online survey format – as opposed to the traditional in-person household survey – may be subject to\nhigher rates of attrition or non-completion and therefore the survey had to be especially cautious of long\nand/or sensitive questions\n\nMention: \"_Future of Business_ survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its bi-annual round covering ninety-seven countries provides the SME data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002133:6:0", "occurrence_id": "91dbfe5b33db4ea3ce480bfe", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ##### **II. Data Description**\n\nFacebook, in partnership with the World Bank and the Organisation for Economic Co-Operation and\nDevelopment (OECD), conducts a bi-annual global survey on small and medium-sized enterprises (SMEs)\nthat use Facebook. This _Future of Business_ survey currently includes 97 countries across every geographic\nregion. <sup>5</sup> [^5: Map presented in Appendix A displays the geographic coverage of the December 2018 wave of the _Future of Business_ Survey\nand the table in Appendix B provides information for all the countries included in the analysis plus the percentage of female\nrespondents per each country.]\n\nThe _Future of Business_ (FoB) survey instrument was redesigned for the December 2018 wave to contain a\nset of questions capturing firm performance outcomes, as well as information from the business owner or\nmanager including his/her background, family history, education, skills, socioeconomic characteristics and\nother variables. <sup>6</sup> [^6: Facebook did not share any platform data that were not collected through this particular survey round. In the analysis for this\npaper, we use only anonymized survey responses.]\n\nThe survey was offered, by Facebook, to administrators <sup>7</sup> [^7: Note some categories of businesses were excluded (governments etc.) from the sampling frame.] of Facebook-designated SME pages in the\nspecified countries through Facebook for a period of two weeks in early to mid-December. The data were\ncollected on the Facebook platform and the final clean and aggregated, anonymized data are now published\non the World Bank Open Data portal. <sup>8</sup> [^8: World Bank Open Data portal: https://datacatalog.worldbank.org/dataset/future-business-survey-aggregated-data]\n\nThe online survey format – as opposed to the traditional in-person household survey – may be subject to\nhigher rates of attrition or non-completion and therefore the survey had to be especially cautious of long\nand/or sensitive questions\n\nMention: \"_Future of Business_ Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"The appendix documents the geographic coverage and female respondent share of this wave.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002134:7:0", "occurrence_id": "8ea9f7662b01be744b3c1832", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: exchange rate brought the rates in the two-tier system much closer together, and in 1994, the\n\n\nexchange rate was unified, removing this disincentive for trade.\n\n\n_2.2 Liberalizing processing trade_\n\n\nImportantly, in 1979, China introduced zero tariffs and exempted from non-tariff barriers all\n\n\nimported intermediate inputs and capital goods used in the production of exports. These\n\n\nexemptions, described in detail in Ianchovichina (2007) and Hong Kong Trade Development\n\n\nCommittee (2018), were part of China’s export processing system. They provided incentives for\n\n\nboth the processing of imported raw materials and the assembly of imported parts and components\n\n\ninto _finished_ export products. Provisions for processing with supplied materials also allowed firms\n\n\nto deal with the capital market failures that were a serious constraint on export growth in the early\n\n\nreform era.\n\n\nThe effect of duty exemptions was sizable. Right after their introduction in 1979 in four coastal\n\n\ncities designated as special economic zones (SEZs) (Yu and Tian 2012), processing trade\n\n\ncomprised only 5% of Chinese trade flows (General Administration of Customs of the People’s\n\n\nRepublic of China, 2000), but after 1984, the authorities expanded the scope of the duty exemption\n\n\nsystem beyond the four SEZs to 14 ‘open’ cities and a range of other zones (Yu and Tian 2012).\n\n\nThis gave a significant boost to trade growth, well ahead of accession, with duty-free imports used\n\n\nfor export processing quadrupling as a share of total imports between 1981 and 1988\n\n\n(Ianchovichina 2007 based on data from General Administration of Customs of the People’s\n\n\nRepublic of China, 2000). By 1992 this open-door policy was extended from eastern to central and\n\n\nwestern China and foreign direct investment inflows surged from 1% of GDP in 1991 to more than\n\n\n6% of GDP in 1992 (Figure 1), speeding up the integration of China into global production\n\n\nnetworks.\n\n\n6\n\nMention: \"data from General Administration of Customs\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"A cited study used these records to track duty-free imports for export processing.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002138:49:1", "occurrence_id": "9d7f78b34e2e719a5925c3dc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .2 3.2 2.8 1.6 1.2 3.1 2.7\n\n\nCosta Rica -2.4 -0.4 2.0 1.3 -1.3 -2.4 0.4 -0.6\n**Panel B: Labor force participation rate, difference over 2015: men aged 25-34 (pp)** (𝝈𝝈= 𝟒𝟒)\nEl Salvador 0.2 0.1 0.2 0.0 7.4 6.5 7.5 6.5\nPanama -2.0 -1.4 2.5 2.1 0.7 0.9 2.4 2.1\n\n\n\n**References**\n\n**CEDLAS, and the World Bank.** “Socio-economic database for Latin America and the Caribbean _.” Centro de_\n_Estudios Distributivos, Laborales y Sociales,_ Facultad de Ciencias Economicas, Universidad Nacional de La\nPlata, La Plata, Argentina; _the World Bank,_ Washington DC (2019).\n\n**Ganzeboom, Harry BG, and Donald J. Treiman.** “Internationally comparable measures of occupational\nstatus for the 1988 International Standard Classification of Occupations.” _Social Science Research_ 25, no.\n3 (1996): 201-239.\n\n**Hsieh, Chang‐Tai, Erik Hurst, Charles I. Jones, and Peter J. Klenow.** “The allocation of talent and US\neconomic growth.” _Econometrica_ 87, no. 5 (2019): 1439-1474.\n\n**International Labor Organization.** “International standard classification of occupations: Structure, group\ndefinitions and correspondence tables.” _International Labor Office, Geneva_ (2012).\n\n**National Institute of Statistics and Censuses.** “Encuesta de Hog\n\nMention: \"Encuesta de Hog\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"A reference list entry for the household survey published by the statistical institute.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002141:31:0", "occurrence_id": "f4d113f23e6ff89395f7af0f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Table 7: Descriptive Statistics       - Forecast survey\n\n\nMean p50 SD Min p25 p75 Max N\n\n\n_Filing_ _probability_ _(p.p.)_\n\nPooled 3.9 4.0 3.4 -6.0 2.0 5.0 20.0 64\nSanctions 6.7 5.5 4.6 1.0 4.0 8.0 22.0 62\nProcedures 5.3 4.0 5.5 0.0 2.0 7.0 23.0 63\nTax morale 1.6 1.0 3.8 -5.0 0.0 3.0 20.0 61\n_Gross_ _revenue_ _(%)_\n\nPooled 6.7 5.0 6.2 -3.0 2.5 10.0 25.0 64\nSanctions 11.5 9.5 10.8 -4.0 5.0 13.0 50.0 64\nProcedures 9.3 5.0 12.6 -2.0 1.0 11.0 58.0 63\nTax morale 2.3 0.5 6.1 -18.0 0.0 3.0 36.0 62\nLow risk 6.3 1.0 9.4 -5.0 0.0 8.0 38.0 61\nMedium-low risk 6.1 4.0 8.0 -7.0 0.0 8.0 28.0 61\nMedium risk 7.0 4.0 8.8 -3.0 0.0 11.0 50.0 61\nMedium-high 8.8 7.0 10.6 -3.0 1.0 12.0 50.0 61\nHigh risk 9.1 6.0 11.5 -5.0 0.0 15.0 51.0 61\n_Taxable_ _income_ _(%)_\n\nPooled 5.7\n\nMention: \"Forecast survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Filing probability expectations are summarised for the pooled sample and the sanctions subset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002148:32:0", "occurrence_id": "cd0a6332b2ba296f27638aac", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Annex 3. Study by the NCDCPH et al. on tobacco taxation (2016)\n\n\nA study by the National Center for Disease Control and Public Health uses a top-down approach to model\nthe effect on tobacco consumption and revenue collection of introducing (a) the EU-tobacco taxation\ntargets, and (b) the national target to reduce smoking by 2 percent annually (NCDCPH et al 2016).\n\n\n\nData sources and assumptions\n\n- [Supply-side] Total quantity of cigarettes (2005-2014) . To calculate the quantity of cigarettes consumed,\nthe study uses (a) volumes of imported filtered cigarettes (as reported by customs authorities), plus (b)\nvolumes of domestically-produced tobacco sales (quantities are inferred from excise tax declarations).\n\n21 No smuggling (in or out of Georgia) and zero stocks are assumed. Possible smuggling of tobacco, and\ninformal purchases of domestic cigarettes (not subject to the excise tax) are not observed. The study\nrecognizes a possible overestimation bias of consumption under this methodology. The resulting\nquantities consumed exceed (by three- to six-times) data by Geostat.\n\n- Average retail price . Data on the retail prices of (local and imported) filtered cigarettes were provided\nby Geostat (2005-2014). The average retail price was weighted by the shares of local and imported\ncigarettes in consumption. In 2015, the average retail price was based on data from the MoF.\n\n- Average tax rates . Average excise rates were weighted by consumption volumes of imported and\nlocally-produced tobacco products. A _d valorem_ excise taxes were assumed at 10% after 2016.\n\n- Price elasticity. Lacking an empirical estimation of price-elasticities of tobacco in Georgia, the study\nassumes a single elasticity for the medium brand cigarette of -0.5, as recommended in the WHO\nTobacco Tax Simulation Model (WHO Tobacco Tax Simulation Model (June 2013).\n\n- Income elasticity . Using historical\n\nMention: \"data by Geostat\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Quantities implied by the methodology exceed these official figures three- to six-fold, indicating overestimation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002148:32:0", "occurrence_id": "6912894bea0da424cd37a247", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Annex 3. Study by the NCDCPH et al. on tobacco taxation (2016)\n\n\nA study by the National Center for Disease Control and Public Health uses a top-down approach to model\nthe effect on tobacco consumption and revenue collection of introducing (a) the EU-tobacco taxation\ntargets, and (b) the national target to reduce smoking by 2 percent annually (NCDCPH et al 2016).\n\n\n\nData sources and assumptions\n\n- [Supply-side] Total quantity of cigarettes (2005-2014) . To calculate the quantity of cigarettes consumed,\nthe study uses (a) volumes of imported filtered cigarettes (as reported by customs authorities), plus (b)\nvolumes of domestically-produced tobacco sales (quantities are inferred from excise tax declarations).\n\n21 No smuggling (in or out of Georgia) and zero stocks are assumed. Possible smuggling of tobacco, and\ninformal purchases of domestic cigarettes (not subject to the excise tax) are not observed. The study\nrecognizes a possible overestimation bias of consumption under this methodology. The resulting\nquantities consumed exceed (by three- to six-times) data by Geostat.\n\n- Average retail price . Data on the retail prices of (local and imported) filtered cigarettes were provided\nby Geostat (2005-2014). The average retail price was weighted by the shares of local and imported\ncigarettes in consumption. In 2015, the average retail price was based on data from the MoF.\n\n- Average tax rates . Average excise rates were weighted by consumption volumes of imported and\nlocally-produced tobacco products. A _d valorem_ excise taxes were assumed at 10% after 2016.\n\n- Price elasticity. Lacking an empirical estimation of price-elasticities of tobacco in Georgia, the study\nassumes a single elasticity for the medium brand cigarette of -0.5, as recommended in the WHO\nTobacco Tax Simulation Model (WHO Tobacco Tax Simulation Model (June 2013).\n\n- Income elasticity . Using historical\n\nMention: \"data from the MoF\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The average retail price for that year is taken from these finance ministry records.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002156:5:0", "occurrence_id": "c7c5b1d8c726df70cf85ee83", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: For the empirical analysis, we combine the satellite data on global human settlement\n\n\nwith data from two rounds of population census to create panel data sets at the commune\n\n\nlevel. To estimate the effects of the CIV trade shock on spatial reallocation across Abidjan\n\n\nand non-Abidjan routes, we adopt a triple-difference strategy with location fixed effects. A\n\n\ndifficult challenge in analyzing the effects of a large-scale trade cost shock is that, because of\n\n\ngeneral equilibrium effects, it may be difficult, if not impossible, to find a comparison area\n\n\nthat remains unaffected (Donaldson (2015)). This is especially challenging in developed\n\n\ncountries where the economy is spatially well-integrated by extensive transport infrastruc\n\nture. The fact that, in many developing countries, a significant part of the economy is\n\n\neffectively in an isolated autarkic equilibrium offers us a plausible strategy; we can use\n\n\nas comparison the part of the economy that remains autarkic both before and after the\n\n\ntrade shock. The changes in employment structure and built-up density observed in these\n\n\ncomparison areas cannot be due to the trade shock, and thus provide an estimate of the\n\n\nstructural changes due to the underlying long-term technological and institutional factors.\n\n\nThe core of our empirical strategy is to compare the DID estimate of the effect of the\n\n\nCIV crisis on the integrated segment of the non-Abidjan route with the corresponding\n\n\nDID estimate for the integrated segment of the Abidjan route, with the respective autarkic\n\n\nareas as the comparison. It is important to appreciate that the triple-difference set-up\n\n\nis used for understanding the spatial reallocation, not for estimating the causal effects of\n\n\nthe negative trade shock on a route. An important advantage of our focus on spatial and\n\n\nsectoral reallocation is that the intractable problem of double-counting one faces while\n\nestimating the causal effects is not relevant in our context. <sup>4</sup> Using data on a vector of\n\n\nrelevant characteristics\n\nMention: \"satellite data on global human settlement\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"It is combined with two census rounds to build commune-level panels.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002135:4:0", "occurrence_id": "123500dd6616bbe895fe6af6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: largest dividends for women, households, and communities – has become increasingly urgent.\n\n\nWhile this is globally relevant, it is particularly important in Sub-Saharan Africa, which has both\n\n\nthe lowest ranking on the Human Development Index and on the Gender Inequality Index (UNDP,\n\n\n2018). Moreover, as gender-related data collection efforts expand, there is growing demand for\n\n\nhow to capture empowerment in a meaningful way in cross-country surveys without being cost\n\nprohibitive.\n\n\nIn this paper, we build on a classical conceptualization of empowerment which has been discussed\n\n\n[across literatures (e.g., Batliwala, 1994; Freire, 1990; Rowlands, 1995), but not operationalized as](https://www.sciencedirect.com/science/article/pii/S0305750X18300408#b0050)\n\n\na common survey measure. Specifically, we argue that women “taking power” (declaring higher\n\n\ndecision-making ability for themselves than husbands report about them, when both are separately\n\n\nasked about intrahousehold decision-making) versus the case when they are “given power” (the\n\n\nhusband reports higher decision-making for his wife than she does for herself) captures an\n\n\nimportant dimension of empowerment, prominent in theoretical work but not as well addressed in\n\n\nthe empirical empowerment literature.\n\n\nUsing recent Demographic and Health Survey (DHS) data for more than 70,000 couples across 23\n\n\ncountries in Sub-Saharan Africa, we examine patterns of power and their relationship to\n\n\nempowerment proxies. We find that there is substantial contention about decision-making power\n\n\nacross households, and that this power is multifaceted. Grounding ourselves in empowerment\n\n\ntheory, we compare scenarios where the woman takes power versus is given power, and contrast\n\n\nthis with the most studied case of no intra-household contention regarding women’s agency.\n\n\n3\n\nMention: \"Human Development Index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its ranking shows the region lowest on human development.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002135:4:0", "occurrence_id": "4ca8e54557cfeefc3b67b624", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: largest dividends for women, households, and communities – has become increasingly urgent.\n\n\nWhile this is globally relevant, it is particularly important in Sub-Saharan Africa, which has both\n\n\nthe lowest ranking on the Human Development Index and on the Gender Inequality Index (UNDP,\n\n\n2018). Moreover, as gender-related data collection efforts expand, there is growing demand for\n\n\nhow to capture empowerment in a meaningful way in cross-country surveys without being cost\n\nprohibitive.\n\n\nIn this paper, we build on a classical conceptualization of empowerment which has been discussed\n\n\n[across literatures (e.g., Batliwala, 1994; Freire, 1990; Rowlands, 1995), but not operationalized as](https://www.sciencedirect.com/science/article/pii/S0305750X18300408#b0050)\n\n\na common survey measure. Specifically, we argue that women “taking power” (declaring higher\n\n\ndecision-making ability for themselves than husbands report about them, when both are separately\n\n\nasked about intrahousehold decision-making) versus the case when they are “given power” (the\n\n\nhusband reports higher decision-making for his wife than she does for herself) captures an\n\n\nimportant dimension of empowerment, prominent in theoretical work but not as well addressed in\n\n\nthe empirical empowerment literature.\n\n\nUsing recent Demographic and Health Survey (DHS) data for more than 70,000 couples across 23\n\n\ncountries in Sub-Saharan Africa, we examine patterns of power and their relationship to\n\n\nempowerment proxies. We find that there is substantial contention about decision-making power\n\n\nacross households, and that this power is multifaceted. Grounding ourselves in empowerment\n\n\ntheory, we compare scenarios where the woman takes power versus is given power, and contrast\n\n\nthis with the most studied case of no intra-household contention regarding women’s agency.\n\n\n3\n\nMention: \"Gender Inequality Index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its ranking shows the region lowest on gender inequality alongside human development.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002135:4:0", "occurrence_id": "a8655e41826dea63a0f4fb3b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: largest dividends for women, households, and communities – has become increasingly urgent.\n\n\nWhile this is globally relevant, it is particularly important in Sub-Saharan Africa, which has both\n\n\nthe lowest ranking on the Human Development Index and on the Gender Inequality Index (UNDP,\n\n\n2018). Moreover, as gender-related data collection efforts expand, there is growing demand for\n\n\nhow to capture empowerment in a meaningful way in cross-country surveys without being cost\n\nprohibitive.\n\n\nIn this paper, we build on a classical conceptualization of empowerment which has been discussed\n\n\n[across literatures (e.g., Batliwala, 1994; Freire, 1990; Rowlands, 1995), but not operationalized as](https://www.sciencedirect.com/science/article/pii/S0305750X18300408#b0050)\n\n\na common survey measure. Specifically, we argue that women “taking power” (declaring higher\n\n\ndecision-making ability for themselves than husbands report about them, when both are separately\n\n\nasked about intrahousehold decision-making) versus the case when they are “given power” (the\n\n\nhusband reports higher decision-making for his wife than she does for herself) captures an\n\n\nimportant dimension of empowerment, prominent in theoretical work but not as well addressed in\n\n\nthe empirical empowerment literature.\n\n\nUsing recent Demographic and Health Survey (DHS) data for more than 70,000 couples across 23\n\n\ncountries in Sub-Saharan Africa, we examine patterns of power and their relationship to\n\n\nempowerment proxies. We find that there is substantial contention about decision-making power\n\n\nacross households, and that this power is multifaceted. Grounding ourselves in empowerment\n\n\ntheory, we compare scenarios where the woman takes power versus is given power, and contrast\n\n\nthis with the most studied case of no intra-household contention regarding women’s agency.\n\n\n3\n\nMention: \"Demographic and Health Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Over seventy thousand couples are used to examine power relations and empowerment.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002152:18:0", "occurrence_id": "5ffbaf33bff599cb7e4d8e32", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: sales, in addition to technology sophistication, is the size of the firm (see Tables B1, B2, B3, and\n\n\nB4 in the appendix). Whether the manager has a post-graduate degree has an indirect effect on\n\n\nperformance during the pandemic, but the direct effect is not statistically significant. In contrast,\n\n\nthe dummy for whether the business is part of a multi-national has a direct effect but the indirect\n\n\neffect is not statistically significant. The z-score for management practices is not statistically sig\n\nnificant in the equation for additional use of digital technology during the pandemic (the mediator)\n\n\nnor in the equation for the shock to sales.\n\n#### **5 Robustness checks**\n\n\nOur estimations combine the four digital responses available in the COV-BPS data in a single\n\n\nbinary indicator. In this section we show results for each digital adjustment separately. We also\n\n\npresent results for the indirect effect using matching estimators.\n\n###### **5.1 The treatment effect under different mediators**\n\n\nOur estimation results are robust to different definitions of the mediator. Figure 3 shows esti\n\nmates for the direct and indirect effects using each potential digital response as an indirect channel\n\n\nfor the effect of technology sophistication. The total effect remains close to 6.5 pp when the re\n\nsponse is the additional use of digital platforms or the new investment in equipment, software, or\n\n\ndigital solutions. When the response is more home-based work the effect increases by 1 pp, and\n\n\nclose to 2 pp when we consider the higher share of online sales.\n\n\n17\n\nMention: \"COV-BPS data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Four digital adjustment responses are combined into one binary indicator.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002160:9:0", "occurrence_id": "6311ee5868be72cfc07fa3fa", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 7\n\n\nThis trend-check exercise is detailed further in section 3.1 on dropped data points. Section\n\n\n3.2 discussed data points which are included in both the 2018 and 2019 HEFPI versions but change\n\n\nin value due to changes in their underlying source or our indicator definitions, or because of coding\n\n\ncorrections. Section 3.3 introduces the data points we newly added to the HEFPI database in its 2019\n\n\nversion. Finally, section 3.4 gives an overview of the subnational surveys included on the health\n\n\nequity side of HEFPI.\n\n\n3.1. <u>Dropped data points</u>\n\n\nFor the 2019 HEFPI database, we quality-checked all points from the 2018 database and all\n\n\nnewly obtained points for the following eight health equity indicators: Inpatient care use, 4+\n\n\nantenatal care visits, skilled birth attendance, full child immunization, formal health care provider\n\n\nvisits for children with acute respiratory infections, receipt of oral rehydration salts among children\n\n\nwith diarrhea, adult inpatient care use, mammograms, and pap smears.\n\n\nThe quality check consisted in reviewing each point in a country’s time-series for the\n\n\nrespective indicator via visual inspection of trend charts. For outlying points that we computed from\n\n\nmicrodata we reviewed, and if necessary corrected our code. <sup>3</sup> If the data came from publications or if\n\n\nwe found our coding to be correct, we re-checked if the survey’s raw data indeed allowed us to\n\n\nconstruct the indicator in line with or sufficiently similarly to the HEFPI indicator definitions <sup>4</sup> [^4: See Tables 1-3 in Wagstaff et al. (2018) for detailed HEFPI indicator definitions.] - and\n\n\ndropped the point if we found this to not be possible. If definitions were sufficiently similar, we\n\n\n3 Changes in indicator values between the 2018 and 2019 HEFPI versions due to such coding corrections or for\nother reasons are discussed further in section 3.2.3 below.\n4 See Tables 1-3 in Wagstaff et\n\nMention: \"HEFPI database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"New data points and subnational health equity surveys are documented for this version.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002162:7:0", "occurrence_id": "6d35f51b25f7902a6b72ae0f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The BVC proprietary data consist of all the transactions in Colombian stocks\nduring the 2006-2017 period. The data contain detailed execution information,\nincluding a time stamp (to the second), price, direction, quantity, and the broker\nrouting the trade. <sup>9</sup> Each transaction record has a unique investor identification\nnumber. In addition, the BVC classifies investors into three broad groups - domestic individuals, domestic institutions, and foreign investors. <sup>10</sup> The stock exchange further classifies domestic institutions into mutual funds, pension funds,\nnon-financial corporations, brokers trading on their own accounts, and other institutions.\n\nTo obtain a more detailed classification of foreign investors we proceed as\nfollows. For the exclusive purpose of this project, the BVC discloses the name of\nforeign institutions involved in the transactions. We search manually on the web\nand cross-reference the names with Factset Ownership – a database with extensive\nglobal coverage on institutional investors - and classify each foreign investor into\none of the following groups: Active mutual funds, passive mutual funds, pension\nfunds, exchange-traded funds (ETFs), hedge funds, government institutions (e.g.,\nsovereign wealth funds and central banks), and investment firms. Other investors\ninclude foreign individuals, trust funds, and private endowments.\n\nThe BVC database has about 17 million records during our 12-year period.\nDomestic institutions represent over half of the traded value in domestic stocks.\nConsistent with international trends in cross-border portfolio investments (IMF,\n2017), the presence of foreign investors in Colombia increases over time. The\nnumber of foreign investors rose from 324 in 2006 to almost 4000 at the end of\n2017 and the share of total traded value by this group increased from 4.5% to\n31.8%. In addition to the generalized growth of foreign flows to emerging markets\nafter the 2008 global financial crisis, there are two key related\n\nMention: \"BVC proprietary data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"All stock transactions with execution timestamps are drawn from this exchange data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002162:7:0", "occurrence_id": "5b49e2d24049c07d4e075c57", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The BVC proprietary data consist of all the transactions in Colombian stocks\nduring the 2006-2017 period. The data contain detailed execution information,\nincluding a time stamp (to the second), price, direction, quantity, and the broker\nrouting the trade. <sup>9</sup> Each transaction record has a unique investor identification\nnumber. In addition, the BVC classifies investors into three broad groups - domestic individuals, domestic institutions, and foreign investors. <sup>10</sup> The stock exchange further classifies domestic institutions into mutual funds, pension funds,\nnon-financial corporations, brokers trading on their own accounts, and other institutions.\n\nTo obtain a more detailed classification of foreign investors we proceed as\nfollows. For the exclusive purpose of this project, the BVC discloses the name of\nforeign institutions involved in the transactions. We search manually on the web\nand cross-reference the names with Factset Ownership – a database with extensive\nglobal coverage on institutional investors - and classify each foreign investor into\none of the following groups: Active mutual funds, passive mutual funds, pension\nfunds, exchange-traded funds (ETFs), hedge funds, government institutions (e.g.,\nsovereign wealth funds and central banks), and investment firms. Other investors\ninclude foreign individuals, trust funds, and private endowments.\n\nThe BVC database has about 17 million records during our 12-year period.\nDomestic institutions represent over half of the traded value in domestic stocks.\nConsistent with international trends in cross-border portfolio investments (IMF,\n2017), the presence of foreign investors in Colombia increases over time. The\nnumber of foreign investors rose from 324 in 2006 to almost 4000 at the end of\n2017 and the share of total traded value by this group increased from 4.5% to\n31.8%. In addition to the generalized growth of foreign flows to emerging markets\nafter the 2008 global financial crisis, there are two key related\n\nMention: \"Factset Ownership\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Disclosed foreign institution names are cross-referenced with it to classify investor type.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002162:7:0", "occurrence_id": "6fea331e51b71da813d932ec", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The BVC proprietary data consist of all the transactions in Colombian stocks\nduring the 2006-2017 period. The data contain detailed execution information,\nincluding a time stamp (to the second), price, direction, quantity, and the broker\nrouting the trade. <sup>9</sup> Each transaction record has a unique investor identification\nnumber. In addition, the BVC classifies investors into three broad groups - domestic individuals, domestic institutions, and foreign investors. <sup>10</sup> The stock exchange further classifies domestic institutions into mutual funds, pension funds,\nnon-financial corporations, brokers trading on their own accounts, and other institutions.\n\nTo obtain a more detailed classification of foreign investors we proceed as\nfollows. For the exclusive purpose of this project, the BVC discloses the name of\nforeign institutions involved in the transactions. We search manually on the web\nand cross-reference the names with Factset Ownership – a database with extensive\nglobal coverage on institutional investors - and classify each foreign investor into\none of the following groups: Active mutual funds, passive mutual funds, pension\nfunds, exchange-traded funds (ETFs), hedge funds, government institutions (e.g.,\nsovereign wealth funds and central banks), and investment firms. Other investors\ninclude foreign individuals, trust funds, and private endowments.\n\nThe BVC database has about 17 million records during our 12-year period.\nDomestic institutions represent over half of the traded value in domestic stocks.\nConsistent with international trends in cross-border portfolio investments (IMF,\n2017), the presence of foreign investors in Colombia increases over time. The\nnumber of foreign investors rose from 324 in 2006 to almost 4000 at the end of\n2017 and the share of total traded value by this group increased from 4.5% to\n31.8%. In addition to the generalized growth of foreign flows to emerging markets\nafter the 2008 global financial crisis, there are two key related\n\nMention: \"BVC database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Its scale of about seventeen million records and domestic share are described.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002165:36:1", "occurrence_id": "7f39756425f0009f187992dd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: country origin‐destination shift parameters, 𝐴��� . The 𝑧�� are calibrated to match observed\n\nemployment levels in each sector‐cell and per capita income levels across countries. 𝐴��� are calibrated\n\n\n\nemployment levels in each sector‐cell and per capita income levels across countries. 𝐴��� are calibrated\n\nto match the international trade data.\n\nThe calibration proceeds in three main steps. First, we calibrate relative 𝑧�� within each country, so\nthat equilibrium employment in each sector and cell matches the employment data. Second, the\nabsolute levels of 𝑧�� in each country are scaled to match income per capita at a country level. Finally,\nwe calibrate preference parameters 𝐴��� such that the share of country _i_ output being sold in country\n\n_j_ is consistent with international trade data.\n\n\n\nThe calibration proceeds in three main steps. First, we calibrate relative 𝑧�� within each country, so\nthat equilibrium employment in each sector and cell matches the employment data. Second, the\nabsolute levels of 𝑧�� in each country are scaled to match income per capita at a country level. Finally,\nwe calibrate preference parameters 𝐴��� such that the share of country _i_ output being sold in country\n\n\n\n**_Simulations_**\n\n\nThe calibrated model is ‘shocked’ with new transport time matrices, and firms and households adjust\ntheir decisions to optimally respond to the new connectivity of the region. We look at four different\nlevels of economic response\n\n\n(1) The direct trade value of the transport cost reduction. We take the base value of exports of each\ncell according to the destination and calculate the reduction in costs in shipping the same amount of\ngoods and services that results from the decreased travel times. No demand or supply response\noccurs.\n\n\n(2) Armington assumption; the number of firms/ varieties of products produced in each cell is held\nconstant. Output per firm adjusts, as do equilibrium prices and demands.\n\n\n(3) Monopolistic competition in manufacturing production, where\n\nMention: \"international trade data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Bilateral trade values are the targets matched when calibrating productivity terms.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002165:36:1", "occurrence_id": "f74cbe0adec0570d1d7cf735", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: country origin‐destination shift parameters, 𝐴��� . The 𝑧�� are calibrated to match observed\n\nemployment levels in each sector‐cell and per capita income levels across countries. 𝐴��� are calibrated\n\n\n\nemployment levels in each sector‐cell and per capita income levels across countries. 𝐴��� are calibrated\n\nto match the international trade data.\n\nThe calibration proceeds in three main steps. First, we calibrate relative 𝑧�� within each country, so\nthat equilibrium employment in each sector and cell matches the employment data. Second, the\nabsolute levels of 𝑧�� in each country are scaled to match income per capita at a country level. Finally,\nwe calibrate preference parameters 𝐴��� such that the share of country _i_ output being sold in country\n\n_j_ is consistent with international trade data.\n\n\n\nThe calibration proceeds in three main steps. First, we calibrate relative 𝑧�� within each country, so\nthat equilibrium employment in each sector and cell matches the employment data. Second, the\nabsolute levels of 𝑧�� in each country are scaled to match income per capita at a country level. Finally,\nwe calibrate preference parameters 𝐴��� such that the share of country _i_ output being sold in country\n\n\n\n**_Simulations_**\n\n\nThe calibrated model is ‘shocked’ with new transport time matrices, and firms and households adjust\ntheir decisions to optimally respond to the new connectivity of the region. We look at four different\nlevels of economic response\n\n\n(1) The direct trade value of the transport cost reduction. We take the base value of exports of each\ncell according to the destination and calculate the reduction in costs in shipping the same amount of\ngoods and services that results from the decreased travel times. No demand or supply response\noccurs.\n\n\n(2) Armington assumption; the number of firms/ varieties of products produced in each cell is held\nconstant. Output per firm adjusts, as do equilibrium prices and demands.\n\n\n(3) Monopolistic competition in manufacturing production, where\n\nMention: \"employment data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Sector and cell employment levels are the first calibration target.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002166:14:0", "occurrence_id": "a297ab2065d1d0f579cf0c2b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: These two pressure levels are the closest to the ground available in the NCEP/NCAR data set.\n\n\nThermal inversions occur when a layer of warm air passes between two layers of cold air. To\n\n\nderive the instrument for long-term exposure to PM2.5, the mean temperature is first calculated\n\n\nfor each pressure level for the 10 years from 2006 to 2017 by district. Then, the paper identifies\n\n\nthermal inversions when the temperature difference D is negative by using the following\n\n\nformula: D = (temperature at 1000 hPa) – (temperature at 925 hPa). Here, the temperature data\n\n\nat midnight (00:00) is used in line with previous studies (Jans et al., 2018; Molina, 2020; Tsaneva\n\n\nand Balakrishnan, 2020) in order to hold the exogeneity because daytime temperatures are\n\n\ndeemed to be more susceptible to economic activities.\n\n\n_Control variables_\n\n\nAs additional controls, the paper uses wind velocity, humidity, precipitation,\n\n\ntemperature, humidity squared, and temperature squared in order to mitigate concerns about the\n\n\nexclusion restrictions of the IV approach given that they could potentially affect the occurrence\n\n\nof thermal inversions. The quadratic terms of humidity and temperature consider the potential\n\n\nnonlinearity between COVID-19 and explanatory variables. In each variable, either daily or\n\n\nmonthly mean values are used to compute the mean yearly values of 2007–2016.\n\n\nWind velocity data are from ERA5, the fifth-generation ECMWF reanalysis data set on\n\n\nglobal climate and weather for the past 4 to 7 decades. The data values show the wind velocity\n\n\nat a height of 10 m above the surface of the Earth with a resolution 0.5° × 0.5°. Humidity data\n\n\nare from the NCEP/NCAR data set Reanalysis 1: Surface, which is a grid-level data set from\n\n\ngrid-level (e.g., Garg et al. 2018; Hansen-Leiws, 2018; T\n\nMention: \"NCEP/NCAR data set\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Pressure at the two lowest levels available are used to identify thermal inversions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002166:14:0", "occurrence_id": "a92c4cdff9594c39900f6d15", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: These two pressure levels are the closest to the ground available in the NCEP/NCAR data set.\n\n\nThermal inversions occur when a layer of warm air passes between two layers of cold air. To\n\n\nderive the instrument for long-term exposure to PM2.5, the mean temperature is first calculated\n\n\nfor each pressure level for the 10 years from 2006 to 2017 by district. Then, the paper identifies\n\n\nthermal inversions when the temperature difference D is negative by using the following\n\n\nformula: D = (temperature at 1000 hPa) – (temperature at 925 hPa). Here, the temperature data\n\n\nat midnight (00:00) is used in line with previous studies (Jans et al., 2018; Molina, 2020; Tsaneva\n\n\nand Balakrishnan, 2020) in order to hold the exogeneity because daytime temperatures are\n\n\ndeemed to be more susceptible to economic activities.\n\n\n_Control variables_\n\n\nAs additional controls, the paper uses wind velocity, humidity, precipitation,\n\n\ntemperature, humidity squared, and temperature squared in order to mitigate concerns about the\n\n\nexclusion restrictions of the IV approach given that they could potentially affect the occurrence\n\n\nof thermal inversions. The quadratic terms of humidity and temperature consider the potential\n\n\nnonlinearity between COVID-19 and explanatory variables. In each variable, either daily or\n\n\nmonthly mean values are used to compute the mean yearly values of 2007–2016.\n\n\nWind velocity data are from ERA5, the fifth-generation ECMWF reanalysis data set on\n\n\nglobal climate and weather for the past 4 to 7 decades. The data values show the wind velocity\n\n\nat a height of 10 m above the surface of the Earth with a resolution 0.5° × 0.5°. Humidity data\n\n\nare from the NCEP/NCAR data set Reanalysis 1: Surface, which is a grid-level data set from\n\n\ngrid-level (e.g., Garg et al. 2018; Hansen-Leiws, 2018; T\n\nMention: \"ERA5\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Wind velocity ten metres above the surface enters the pollution dispersion controls.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002166:14:0", "occurrence_id": "eaaa15faca0b06fe9475dcaa", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: These two pressure levels are the closest to the ground available in the NCEP/NCAR data set.\n\n\nThermal inversions occur when a layer of warm air passes between two layers of cold air. To\n\n\nderive the instrument for long-term exposure to PM2.5, the mean temperature is first calculated\n\n\nfor each pressure level for the 10 years from 2006 to 2017 by district. Then, the paper identifies\n\n\nthermal inversions when the temperature difference D is negative by using the following\n\n\nformula: D = (temperature at 1000 hPa) – (temperature at 925 hPa). Here, the temperature data\n\n\nat midnight (00:00) is used in line with previous studies (Jans et al., 2018; Molina, 2020; Tsaneva\n\n\nand Balakrishnan, 2020) in order to hold the exogeneity because daytime temperatures are\n\n\ndeemed to be more susceptible to economic activities.\n\n\n_Control variables_\n\n\nAs additional controls, the paper uses wind velocity, humidity, precipitation,\n\n\ntemperature, humidity squared, and temperature squared in order to mitigate concerns about the\n\n\nexclusion restrictions of the IV approach given that they could potentially affect the occurrence\n\n\nof thermal inversions. The quadratic terms of humidity and temperature consider the potential\n\n\nnonlinearity between COVID-19 and explanatory variables. In each variable, either daily or\n\n\nmonthly mean values are used to compute the mean yearly values of 2007–2016.\n\n\nWind velocity data are from ERA5, the fifth-generation ECMWF reanalysis data set on\n\n\nglobal climate and weather for the past 4 to 7 decades. The data values show the wind velocity\n\n\nat a height of 10 m above the surface of the Earth with a resolution 0.5° × 0.5°. Humidity data\n\n\nare from the NCEP/NCAR data set Reanalysis 1: Surface, which is a grid-level data set from\n\n\ngrid-level (e.g., Garg et al. 2018; Hansen-Leiws, 2018; T\n\nMention: \"ECMWF reanalysis data set\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"Described as the fifth-generation product covering global climate and weather for several decades.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002178:5:2", "occurrence_id": "ba30dad8e16df0e3b393d2a0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  incidence analysis to comprehensively assess the impact of the fiscal system**\n**on poverty and inequality in Albania, in order to inform the dialogue around fiscal reform with an**\n**equity lens.** It also benchmarks Albania to other countries in which a similar methodology has been applied.\nIn particular, the analysis employs the Commitment to Equity Methodology (Lustig and Higgins, 2017),\nalready applied in many countries around the world and across regions, to explore: a) How much income\nredistribution and poverty reduction is achieved in Albania through the fiscal system (taxes, subsidies and\nsocial spending), b) Who bears the burden of taxes and receives the benefits?; c) How equalizing and propoor are particular fiscal interventions, and how equitable is the use of education and health services?, and\nd) What is the expected impact of changes in the size and progressivity of a particular tax (i.e. the personal\nincome tax) on poverty and inequality outcomes.\n\n**The analysis relies on the 2015 Household Budget Survey collected by INSTAT, together with**\n**macroeconomic and fiscal data from the national income accounts in Albania** . The analysis does not\nincorporate behavioral or general equilibrium effects. Moreover, using an input-output matrix it captures\nthe indirect effects of excise taxes on intermediate goods, since these can be important as they are passed\nthrough to prices of all other goods. Given the available data, the analysis incorporates the fiscal\ncomponents that capture around 70 percent of fiscal revenues, in particular direct taxes (personal income\ntax and social security contributions) and indirect taxes (household-paid VAT and excises). On the spending\nside, it captures social spending (education, health and social protection), which amounts to 45 percent of\n\n\n4\n\nMention: \"2015 Household Budget Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Combined with national accounts data to assess how income tax progressivity affects poverty and inequality.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002180:7:1", "occurrence_id": "4df9fc4206ec1a4e5371716b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  employment opportunities might increase (George et\nal., 2019), but institutional and safety challenges prevail in many places, which often leaves women\nstruggling to find decent employment and having to work informally, without any protection and\nsafety (Young et al., 2005; George et al., 2019). Thus, identifying and addressing gender-specific\nissues is considered an integral part of understanding and improving livelihoods, and ultimately of\nreaching durable solutions.\n\n**1.3** **Gender inequality in Sudan**\n\nGender inequality in Sudan is extremely high, even in comparison with most other low-income\ncountries. Sudan consistently ranks among the bottom 10 or 20 countries in common statistics, such\nas the Women, Peace and Security Index (GIWPS and PRIO, 2019) or the Gender Inequality Index. <sup>9</sup>\n\nSudan’s laws are based on Sharia and common law, and although women's advancement and\nawareness of gender issues are stated as formal objectives, only little progress has been made. <sup>10</sup>\nFor instance, forced (child-) marriage and rape are not (clearly) illegal (UNICEF, 2017). The burden\nof proof is on the victims, and seeking legal support raises concerns of stigma and risk of punishment\n(Tønnessen, 2019). Gender-based violence more generally remains a key concern and 88% of the\nfemale population are reported to have undergone genital mutilation/cutting, putting their health in\nextreme danger. <sup>11</sup> Furthermore, men continue to have a strong say in women’s decisions. For\ninstance, a man is entitled to denying his wife to work outside the home and to limiting the use of\ncontraceptives (UNDP, 2018). All these risks often affect IDPs in particular. For example, it has been\ndocumented that displaced women are at increased risk when\n\nMention: \"Women, Peace and Security Index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its ranking places the country among the bottom ten to twenty.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002180:7:1", "occurrence_id": "e3aec9c34357b416684425b9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  employment opportunities might increase (George et\nal., 2019), but institutional and safety challenges prevail in many places, which often leaves women\nstruggling to find decent employment and having to work informally, without any protection and\nsafety (Young et al., 2005; George et al., 2019). Thus, identifying and addressing gender-specific\nissues is considered an integral part of understanding and improving livelihoods, and ultimately of\nreaching durable solutions.\n\n**1.3** **Gender inequality in Sudan**\n\nGender inequality in Sudan is extremely high, even in comparison with most other low-income\ncountries. Sudan consistently ranks among the bottom 10 or 20 countries in common statistics, such\nas the Women, Peace and Security Index (GIWPS and PRIO, 2019) or the Gender Inequality Index. <sup>9</sup>\n\nSudan’s laws are based on Sharia and common law, and although women's advancement and\nawareness of gender issues are stated as formal objectives, only little progress has been made. <sup>10</sup>\nFor instance, forced (child-) marriage and rape are not (clearly) illegal (UNICEF, 2017). The burden\nof proof is on the victims, and seeking legal support raises concerns of stigma and risk of punishment\n(Tønnessen, 2019). Gender-based violence more generally remains a key concern and 88% of the\nfemale population are reported to have undergone genital mutilation/cutting, putting their health in\nextreme danger. <sup>11</sup> Furthermore, men continue to have a strong say in women’s decisions. For\ninstance, a man is entitled to denying his wife to work outside the home and to limiting the use of\ncontraceptives (UNDP, 2018). All these risks often affect IDPs in particular. For example, it has been\ndocumented that displaced women are at increased risk when\n\nMention: \"Gender Inequality Index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Another ranking in which the country performs poorly.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002183:27:0", "occurrence_id": "61956e6fbab63e5df93e9732", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: in developing countries. High-frequency phone surveys can shed light on aspects of the\n\n\neconomic impact of the pandemic that are difficult to capture using more traditional data.\n\n\nThey can be conducted more quickly than GDP statistics, which often rely on projections\n\n\nin real time and are significantly revised in subsequent years. Moreover, phone surveys\n\n\ncan uncover dimensions of heterogeneity, as well as information on coping mechanisms and\n\n\npotential implications on the lives of respondents that are not captured by official economic\n\n\nstatistics.\n\n\nOf course, HFPS survey data have limitations too. Particular care and effort must\n\n\nbe taken when comparing and harmonizing surveys from different countries. Further, the\n\n\nsurveys are not nationally representative, as they can only reach phone owners and in many\n\n\ncountries greatly over-represented heads. However, we see that measures of stopped work in\n\n\nthe phone surveys are more consistent with reductions in household wage income, estimates of\n\n\ndemand reduction in firms, and ILO employment estimates than GDP growth projections.\n\n\nAlso, the initial round of the phone surveys contained few variables found in traditional\n\n\nface-to-face household surveys implemented before the crisis. As a result, the reweighting\n\n\nprocedure could only utilize a few variables and did not have a large effect on the results.\n\n\nDespite these limitations, harmonized data from phone surveys appear to contribute valuable\n\n\nnew information on how households in a broad cross-section of developing countries were\n\n\naffected by this severe shock. This is in line with recent findings by Heath et al. (2021) that\n\n\ncomparing interviews conducted on the phone and in-person led to differences in measures\n\n\nof employment, hours and days worked for the self-employed in Ghana.\n\n\nThe rapid deployment of phone surveys to measure the socio-economic impacts of COVID\n\n19 was only possible because of an extraordinary effort around the globe. To be better pre\n\npared for such rapid deployment in future emergencies, National Statistics Offices can invest\n\n\nnow to improve the speed of deployment and quality of data. Investments in statistical infras\n\ntructure (e\n\nMention: \"High-frequency phone surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"inform\", \"summary\": \"They capture pandemic impacts that traditional data cannot and can be fielded quickly.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002189:10:0", "occurrence_id": "ccf15a45c1e9f348722ea128", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The survey is nationally representative of the formal private sector. It is built based on\n\n\na stratified random sampling frame designed by the WBES team. Three variables are used\n\n\nto construct the strata: firm size, sector, and geographic area within a country. Under the\n\n\nWBES sampling framework, firms are divided into three categories according to their size:\n\n\nsmall, medium-sized, and large. Small firms are those with 5-19 full-time employees; medium\n\nsized firms have 20-99 full-time employees; and the large ones have more than 99 full-time\n\n\nemployees. The industries are classified according to the ISIC Revision 3.1 classification at\n\n\n2-digits. The regions within a country are defined by the WBES team. The database also\n\n\nincludes sampling weights that can be used to mimic nationally representative samples in\n\n\nthe empirics.\n\n\nThe WBES collects data on a broad range of variables related to firm production, perfor\n\nmance, and the business environment in which firms operate. It has information on variables\n\n\nrelated to production such as sales, capital, labor, materials, investment, exports, and man\n\nager’s education, among others. Due to the lack of information on prices at the firm-level,\n\n\nwe use the consumer price index from the World Bank’s World Development Indicators to\n\n\ndeflate sales, capital, materials, and investment, thus transforming nominal values into 2010\n\ndollar values. Firms’ labor is equal to the number of permanent employees that work for the\n\n\nfirm. The survey collects data on the percentage of firms’ sales that are exported. Last, a\n\n\nfirm’s managerial capability is measured by the number of years of experience of the man\n\nager. The novelty of the WBES is that it also collects information on technology adoption\n\n\nat the firm-level. Thus, at every wave, firms are asked whether they use a business email to\n\n\ncommunicate with clients and suppliers and whether they have a business website in\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Its consumer price index deflates sales, capital, materials and investment to constant dollars.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002195:20:4", "occurrence_id": "a2e758a8c8069a807a291052", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: /u> **𝑎** <u>𝑎𝑖𝑎</u> **𝑖** and <u>𝑆𝑆ℎ𝑎</u> **𝑎** <u>𝑎𝑗𝑎</u> **𝑗** across all\noccupations, denoted as **Table 2. Summary statistics of the coagglomeration measures** <u>𝐿</u> **𝐿** <u>𝑖</u> **𝑖** . The values of the three measures are summarized in Table 2.\n\n\n\nThe third measure is on the similarity between sectors in terms of the types of workers employed. We\nmake use of the employment module of the 2011-2012 National Sample Survey of India (NSSO 2012),\nwhich reports detailed occupation and sector information for each worker. We construct 𝑆𝑆ℎ𝑎 **𝑎** 𝑎𝑎𝑖 **𝑖** as the\nSurvey data is limited, we calculate this variable at 2-digit NIC 2008 level. The similarity of worker types\nfraction of sector 𝑖𝑖 ’s employment in occupation 𝑜𝑜 . As the number of observations in the National Sample\nbetween industries <u>𝑖𝑖</u> and <u>𝑗𝑗</u> are proxied as the correlation between <u>𝑆𝑆ℎ𝑎</u> **𝑎** <u>𝑎𝑖𝑎</u> **𝑖** and <u>𝑆𝑆ℎ𝑎</u> **𝑎** <u>𝑎𝑗𝑎</u> **𝑗** across all\noccupations, denoted as **Table 2. Summary statistics of the coagglomeration measures** <u>𝐿</u> **𝐿** <u>𝑖</u> **𝑖** . The values of the three measures are summarized in Table 2.\n\n\n\n\n|Col1|Coagglomeration index|Input-Output linkages|Labor pooling linkages|\n\nMention: \"2011-2012 National Sample Survey of India\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its employment module gives occupation and sector for each worker, measuring workforce similarity across sectors.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002195:20:4", "occurrence_id": "b055f551140f75b16376436e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: /u> **𝑎** <u>𝑎𝑖𝑎</u> **𝑖** and <u>𝑆𝑆ℎ𝑎</u> **𝑎** <u>𝑎𝑗𝑎</u> **𝑗** across all\noccupations, denoted as **Table 2. Summary statistics of the coagglomeration measures** <u>𝐿</u> **𝐿** <u>𝑖</u> **𝑖** . The values of the three measures are summarized in Table 2.\n\n\n\nThe third measure is on the similarity between sectors in terms of the types of workers employed. We\nmake use of the employment module of the 2011-2012 National Sample Survey of India (NSSO 2012),\nwhich reports detailed occupation and sector information for each worker. We construct 𝑆𝑆ℎ𝑎 **𝑎** 𝑎𝑎𝑖 **𝑖** as the\nSurvey data is limited, we calculate this variable at 2-digit NIC 2008 level. The similarity of worker types\nfraction of sector 𝑖𝑖 ’s employment in occupation 𝑜𝑜 . As the number of observations in the National Sample\nbetween industries <u>𝑖𝑖</u> and <u>𝑗𝑗</u> are proxied as the correlation between <u>𝑆𝑆ℎ𝑎</u> **𝑎** <u>𝑎𝑖𝑎</u> **𝑖** and <u>𝑆𝑆ℎ𝑎</u> **𝑎** <u>𝑎𝑗𝑎</u> **𝑗** across all\noccupations, denoted as **Table 2. Summary statistics of the coagglomeration measures** <u>𝐿</u> **𝐿** <u>𝑖</u> **𝑖** . The values of the three measures are summarized in Table 2.\n\n\n\n\n|Col1|Coagglomeration index|Input-Output linkages|Labor pooling linkages|\n\nMention: \"NSSO 2012\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its limited sample size constrains the construction of the similarity measure.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002197:11:0", "occurrence_id": "4b147973d7b026a36bae584f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: managerial, and business environment measures. Within the WBES, the Green Economy\n\n\nmodule is rolled out by the World Bank Group, jointly with the EBRD-EIB.\n\n\n_Generic_ _and_ _Energy-centric_ _Management_ _Scores_ . General management scores are calculated\n\n\nusing 11 questions from the management practice module. These questions map to eight\n\n\nquestions from the U.S. Management and Organizational Practices Survey (U.S. Census\n\n\nBureau, 2015). They cover four categories of practices: target-setting, monitoring, incentives,\n\n\nand operations (see Table A1). Firms’ responses to questions on management practices are\n\n\naggregated into a single management score following Bloom et al. (2019a) in two steps.\n\n\n_First_, the responses to each of the management practice questions are normalized on a 0-1\n\n\nscale: the response associated with the most structure is normalized to one, and the one\n\n\nassociated with the least structured is normalized to zero. _Second_, overall management\n\n\nscore is calculated as the unweighted average of the individual question scores as in Bloom\n\n\net al. (2012c) in computing the average based on the z-scores of individual questions. We\n\n\nuse these scores to compute descriptive statistics. About 2% of the firms with non-missing\n\n\nentries report a zero average score, while the average for all firms is 0.48 (standard deviation\n\n\nof 0.21). For regressions, we take the z-scores of individual questions, and compute the\n\n\nunweighted average. We then once again take the z-score to obtain a composite management\n\n\nscore. General management score also separates the overall management index into four sub\n\nindices to separately assess the importance of monitoring, incentives, targets and operations\n\n\npractices.\n\n\nEnergy management scores use questions from the Green Economy Module of the WBES\n\n\nthat relate to monitoring, target setting, and relative emphasis on energy efficiency. Ques\n\ntions and scoring scales are\n\nMention: \"U.S. Management and Organizational Practices Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Eight of its questions map to the management practice module used for scoring.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002200:18:0", "occurrence_id": "78db22898f896e8ce8d592b7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: a commonly-used proxy for income, Nighttime Lights (NTL) per capita. <sup>5</sup> [^5: While Defense Meteorological Satellite Program/Operational Linescan System (DMSP-OLS) was originally designed\nfor military use, DMSP-OLS has been widely used in social science because of its low-light imaging capabilities to\ndetect nighttime artificial lighting (Elvidge et al., 1997). Henderson et al. (2012) empirically prove the concept of\nusing an annual composite of DMSP-OLS as a proxy measurement of economic growth across the national and/or\nlocal level. Many studies extensively use nighttime lights to estimate the national and sub-national GDP. These\nestimates are critically useful for developing countries and/or conflict-affected areas where official statistics are\nunavailable or limited.] We process DMSP-OLS data for\nthree years (1992, 2000, and 2012) for the areas matching with the GHS-UCDB urban areas used in the\nrest of the paper. Column (5) reports the results using Nighttime Lights per capita instead of the previous\nincome measure. Our preferred specification is given by the between and within estimator in Column (2).\nEstimates are similar when using instrumental variables and Nighttime Lights as a proxy for income.\nIncome and population elasticities can be inferred from the results of the regression. <sup>6</sup> [^6: Other possible specifications were considered but not reported in this note. A dynamic panel specification was\nconsidered but 3 years of data was not sufficient to report strong results for a dynamic panel.]\n\nAs a city’s population doubles, its built-up surface increases by 34.7% when keeping income constant and\ncontrolling for unobserved city and year characteristics. This number is lower than previous estimates\nreported in the literature (Angel, et al., 2016) and shows that cities have not extended their built-up as\nmuch as their population growth would have suggested. As a city’s income doubles, the built-up surface\nincreases by 10.4% when keeping population constant and controlling for unobserved city\n\nMention: \"DMSP-OLS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its annual composite has been shown to proxy economic growth at national and local levels.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002200:18:0", "occurrence_id": "dd0dae7bde743100f5c264da", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: a commonly-used proxy for income, Nighttime Lights (NTL) per capita. <sup>5</sup> [^5: While Defense Meteorological Satellite Program/Operational Linescan System (DMSP-OLS) was originally designed\nfor military use, DMSP-OLS has been widely used in social science because of its low-light imaging capabilities to\ndetect nighttime artificial lighting (Elvidge et al., 1997). Henderson et al. (2012) empirically prove the concept of\nusing an annual composite of DMSP-OLS as a proxy measurement of economic growth across the national and/or\nlocal level. Many studies extensively use nighttime lights to estimate the national and sub-national GDP. These\nestimates are critically useful for developing countries and/or conflict-affected areas where official statistics are\nunavailable or limited.] We process DMSP-OLS data for\nthree years (1992, 2000, and 2012) for the areas matching with the GHS-UCDB urban areas used in the\nrest of the paper. Column (5) reports the results using Nighttime Lights per capita instead of the previous\nincome measure. Our preferred specification is given by the between and within estimator in Column (2).\nEstimates are similar when using instrumental variables and Nighttime Lights as a proxy for income.\nIncome and population elasticities can be inferred from the results of the regression. <sup>6</sup> [^6: Other possible specifications were considered but not reported in this note. A dynamic panel specification was\nconsidered but 3 years of data was not sufficient to report strong results for a dynamic panel.]\n\nAs a city’s population doubles, its built-up surface increases by 34.7% when keeping income constant and\ncontrolling for unobserved city and year characteristics. This number is lower than previous estimates\nreported in the literature (Angel, et al., 2016) and shows that cities have not extended their built-up as\nmuch as their population growth would have suggested. As a city’s income doubles, the built-up surface\nincreases by 10.4% when keeping population constant and controlling for unobserved city\n\nMention: \"DMSP-OLS data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Three years are processed for urban areas to construct lights per capita.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002200:18:0", "occurrence_id": "9edd1458c3ca7950eed2abeb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: a commonly-used proxy for income, Nighttime Lights (NTL) per capita. <sup>5</sup> [^5: While Defense Meteorological Satellite Program/Operational Linescan System (DMSP-OLS) was originally designed\nfor military use, DMSP-OLS has been widely used in social science because of its low-light imaging capabilities to\ndetect nighttime artificial lighting (Elvidge et al., 1997). Henderson et al. (2012) empirically prove the concept of\nusing an annual composite of DMSP-OLS as a proxy measurement of economic growth across the national and/or\nlocal level. Many studies extensively use nighttime lights to estimate the national and sub-national GDP. These\nestimates are critically useful for developing countries and/or conflict-affected areas where official statistics are\nunavailable or limited.] We process DMSP-OLS data for\nthree years (1992, 2000, and 2012) for the areas matching with the GHS-UCDB urban areas used in the\nrest of the paper. Column (5) reports the results using Nighttime Lights per capita instead of the previous\nincome measure. Our preferred specification is given by the between and within estimator in Column (2).\nEstimates are similar when using instrumental variables and Nighttime Lights as a proxy for income.\nIncome and population elasticities can be inferred from the results of the regression. <sup>6</sup> [^6: Other possible specifications were considered but not reported in this note. A dynamic panel specification was\nconsidered but 3 years of data was not sufficient to report strong results for a dynamic panel.]\n\nAs a city’s population doubles, its built-up surface increases by 34.7% when keeping income constant and\ncontrolling for unobserved city and year characteristics. This number is lower than previous estimates\nreported in the literature (Angel, et al., 2016) and shows that cities have not extended their built-up as\nmuch as their population growth would have suggested. As a city’s income doubles, the built-up surface\nincreases by 10.4% when keeping population constant and controlling for unobserved city\n\nMention: \"GHS-UCDB\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its urban area boundaries define the zones over which lights are aggregated.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002201:19:1", "occurrence_id": "fdb90fdaf2af5fca55dae817", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  in the area of technical regulations. This type of information can be\nused in the estimation exercises we discuss below, as it allows the researcher to address problems\nassociated with large numbers of possible variables at hand, such as multicollinearity (i.e., the high\ncorrelation between the different provisions within and across policy areas).\n\n\n_Indirectly estimated measures_\n\n\nThese measures are obtained by estimating the impact of the provisions on a variable of interest. For\nexample, we could infer the value of individual provisions by estimating their impact on bilateral trade,\ncontrolling for other influences. In principle, each binary element in the relevant DTA areas could be\nincluded in a country-product import regression as a right-hand variable while controlling for applied\npolicies, including tariffs and non-tariff measures. Similar methods have been used to estimate the Overall\nTrade Restrictiveness Index. <sup>16</sup> However, even for trade in goods we have limited degrees of freedom, and\nin other areas (such as services), we do not have sufficiently fine outcome data. In these areas, it may be\nnecessary to take a hybrid approach, based on first constructing more aggregated indices.\n\n\nAnother approach to quantify the effects of DTAs and build indicators of depth is to use new statistical\nmethods. As a first example, we employ machine learning techniques to detect the influential\nvariables/provisions in DTAs for trade. <sup>17</sup> [^17: This exercise has been carried out in collaboration with Kazusa Yoshimura and Edith Laget. Parallel work by\nBreinlich et al. (2020) also uses machine learning techniques to precisely quantify the impact of individual provisions\nin trade agreements on trade flows.] Machine learning is a generic term referring to a wide variety of\nalgorithms which detect a certain pattern from a large data set, often referred to as “Big Data,” and make\npredictions based on that pattern. In this case, we use a method called Random Forest (RF) to calculate\n\n\n16 Kee, Nicita and Olarreaga, 2009.\n\nMention: \"large data set\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"Described generically as the material from which machine learning algorithms detect patterns.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002201:19:1", "occurrence_id": "2cfd60efbcc9b7e366e2f889", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  in the area of technical regulations. This type of information can be\nused in the estimation exercises we discuss below, as it allows the researcher to address problems\nassociated with large numbers of possible variables at hand, such as multicollinearity (i.e., the high\ncorrelation between the different provisions within and across policy areas).\n\n\n_Indirectly estimated measures_\n\n\nThese measures are obtained by estimating the impact of the provisions on a variable of interest. For\nexample, we could infer the value of individual provisions by estimating their impact on bilateral trade,\ncontrolling for other influences. In principle, each binary element in the relevant DTA areas could be\nincluded in a country-product import regression as a right-hand variable while controlling for applied\npolicies, including tariffs and non-tariff measures. Similar methods have been used to estimate the Overall\nTrade Restrictiveness Index. <sup>16</sup> However, even for trade in goods we have limited degrees of freedom, and\nin other areas (such as services), we do not have sufficiently fine outcome data. In these areas, it may be\nnecessary to take a hybrid approach, based on first constructing more aggregated indices.\n\n\nAnother approach to quantify the effects of DTAs and build indicators of depth is to use new statistical\nmethods. As a first example, we employ machine learning techniques to detect the influential\nvariables/provisions in DTAs for trade. <sup>17</sup> [^17: This exercise has been carried out in collaboration with Kazusa Yoshimura and Edith Laget. Parallel work by\nBreinlich et al. (2020) also uses machine learning techniques to precisely quantify the impact of individual provisions\nin trade agreements on trade flows.] Machine learning is a generic term referring to a wide variety of\nalgorithms which detect a certain pattern from a large data set, often referred to as “Big Data,” and make\npredictions based on that pattern. In this case, we use a method called Random Forest (RF) to calculate\n\n\n16 Kee, Nicita and Olarreaga, 2009.\n\nMention: \"Big Data\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"The term is introduced as the common name for such datasets.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002203:3:1", "occurrence_id": "e0c0603a57e4536d6ce888a6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Bank Microdata Library, under the High\nFrequency Phone Survey collection (World Bank 2020a,b,c,d).\n\n\nThe sample for these surveys is drawn from households that had been interviewed during the latest\n\nround of the national longitudinal household survey implemented by the respective national statistical\n\noffice, with assistance from the World Bank LSMS-Integrated Surveys on Agriculture (LSMS-ISA)\n\ninitiative. The pre-COVID-19 LSMS-ISA-supported surveys were designed to be representative at the\n\nnational, regional, and urban/rural levels. These surveys include the Ethiopia Socio-economic Survey (ESS)\n\n2018/19, Malawi Integrated Household Panel Survey (IHPS) 2019, Nigeria General Household Survey\n\n(GHS) - Panel 2018/19, and Uganda National Panel Survey (UNPS) 2019/20. Specific details on each\n\nsurvey are available through the World Bank Microdata Library.\n\n\n2\n\nMention: \"High\nFrequency Phone Survey collection\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The pandemic survey instruments are archived in this library collection.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002203:3:1", "occurrence_id": "57b9ace98f7131e12de52d21", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Bank Microdata Library, under the High\nFrequency Phone Survey collection (World Bank 2020a,b,c,d).\n\n\nThe sample for these surveys is drawn from households that had been interviewed during the latest\n\nround of the national longitudinal household survey implemented by the respective national statistical\n\noffice, with assistance from the World Bank LSMS-Integrated Surveys on Agriculture (LSMS-ISA)\n\ninitiative. The pre-COVID-19 LSMS-ISA-supported surveys were designed to be representative at the\n\nnational, regional, and urban/rural levels. These surveys include the Ethiopia Socio-economic Survey (ESS)\n\n2018/19, Malawi Integrated Household Panel Survey (IHPS) 2019, Nigeria General Household Survey\n\n(GHS) - Panel 2018/19, and Uganda National Panel Survey (UNPS) 2019/20. Specific details on each\n\nsurvey are available through the World Bank Microdata Library.\n\n\n2\n\nMention: \"Ethiopia Socio-economic Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its latest round supplies the sampling frame for the phone survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002203:3:1", "occurrence_id": "ede84691829ed9f5dd17af1c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Bank Microdata Library, under the High\nFrequency Phone Survey collection (World Bank 2020a,b,c,d).\n\n\nThe sample for these surveys is drawn from households that had been interviewed during the latest\n\nround of the national longitudinal household survey implemented by the respective national statistical\n\noffice, with assistance from the World Bank LSMS-Integrated Surveys on Agriculture (LSMS-ISA)\n\ninitiative. The pre-COVID-19 LSMS-ISA-supported surveys were designed to be representative at the\n\nnational, regional, and urban/rural levels. These surveys include the Ethiopia Socio-economic Survey (ESS)\n\n2018/19, Malawi Integrated Household Panel Survey (IHPS) 2019, Nigeria General Household Survey\n\n(GHS) - Panel 2018/19, and Uganda National Panel Survey (UNPS) 2019/20. Specific details on each\n\nsurvey are available through the World Bank Microdata Library.\n\n\n2\n\nMention: \"Malawi Integrated Household Panel Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its households form the frame for the Malawian phone rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002203:3:1", "occurrence_id": "219f8fac7bd040f79e577771", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Bank Microdata Library, under the High\nFrequency Phone Survey collection (World Bank 2020a,b,c,d).\n\n\nThe sample for these surveys is drawn from households that had been interviewed during the latest\n\nround of the national longitudinal household survey implemented by the respective national statistical\n\noffice, with assistance from the World Bank LSMS-Integrated Surveys on Agriculture (LSMS-ISA)\n\ninitiative. The pre-COVID-19 LSMS-ISA-supported surveys were designed to be representative at the\n\nnational, regional, and urban/rural levels. These surveys include the Ethiopia Socio-economic Survey (ESS)\n\n2018/19, Malawi Integrated Household Panel Survey (IHPS) 2019, Nigeria General Household Survey\n\n(GHS) - Panel 2018/19, and Uganda National Panel Survey (UNPS) 2019/20. Specific details on each\n\nsurvey are available through the World Bank Microdata Library.\n\n\n2\n\nMention: \"Nigeria General Household Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its panel households are the source of the Nigerian phone sample.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002203:3:1", "occurrence_id": "2401c475d82204fe70957907", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Bank Microdata Library, under the High\nFrequency Phone Survey collection (World Bank 2020a,b,c,d).\n\n\nThe sample for these surveys is drawn from households that had been interviewed during the latest\n\nround of the national longitudinal household survey implemented by the respective national statistical\n\noffice, with assistance from the World Bank LSMS-Integrated Surveys on Agriculture (LSMS-ISA)\n\ninitiative. The pre-COVID-19 LSMS-ISA-supported surveys were designed to be representative at the\n\nnational, regional, and urban/rural levels. These surveys include the Ethiopia Socio-economic Survey (ESS)\n\n2018/19, Malawi Integrated Household Panel Survey (IHPS) 2019, Nigeria General Household Survey\n\n(GHS) - Panel 2018/19, and Uganda National Panel Survey (UNPS) 2019/20. Specific details on each\n\nsurvey are available through the World Bank Microdata Library.\n\n\n2\n\nMention: \"Uganda National Panel Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its households provide the frame for the Ugandan phone survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002212:16:0", "occurrence_id": "d4c66a7ba5ab3346ad6b17ae", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: spending via consumer purchase subsidies. In contrast, the average government spending needed\n\n\nis only $1,587 by installing charging infrastructure to induce one additional sale. That is, it\n\n\nis more cost-effective for governments to invest in expanding the charging network rather than\n\n\nproviding purchase incentives for EVs. Governments can subsidize the private sector and share\n\n\nthe investment cost for charging network construction. This is qualitatively consistent with\n\n\nprevious findings in China (Li et al., 2021), US (Li et al., 2017a), and Norway (Springel, 2021).\n\n\nOne important caveat is that the calculated average government spending per induced sale in\n\n\nthis analysis only applies to central government incentives. Our data on incentives only include\n\n\nnational-level subsidies or tax reduction, but local subsidies account for 22% of total subsidies\n\n\non average in China while US states such as California offer rebates or tax credits equivalent to\n\n\n30% of the federal-level tax credit. The estimated impact from incentives might be larger if all\n\n\nlocal incentives were available in the data.\n\n##### **Conclusion**\n\n\nThis study provides a comprehensive analysis of global EV diffusion during the first decade\n\n\nof the technology being introduced to the mass market. The analysis shows that investing in\n\n\ncharging infrastructure is more cost-effective than providing consumer subsidies in promoting EV\n\n\nadoption, consistent with the finding from the literature based on data from individual countries.\n\n\nMore importantly, this study allows us to understand the significant variation in EV adoption\n\n\nacross countries, even among those with similar socio-economic characteristics. Our analysis\n\n\nshows that 17% of the cross-country variation is driven by observed differences in subsidy levels\n\n\nand 69% by differences in charging network size. These findings highlight the importance of\n\n\ninvesting in charging infrastructure in order to further electrify the transportation sector in the\n\n\nnext decade.\n\n\nWe conclude with a discussion on the limitations of our study and directions for future\n\n\nresearch. First, our data do not include gasoline models and do not allow\n\nMention: \"data on incentives\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"It covers only national subsidies and tax cuts, omitting the local subsidies that are a large share of the total.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002212:16:0", "occurrence_id": "fdcfcb739f04dc7d8f357598", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: spending via consumer purchase subsidies. In contrast, the average government spending needed\n\n\nis only $1,587 by installing charging infrastructure to induce one additional sale. That is, it\n\n\nis more cost-effective for governments to invest in expanding the charging network rather than\n\n\nproviding purchase incentives for EVs. Governments can subsidize the private sector and share\n\n\nthe investment cost for charging network construction. This is qualitatively consistent with\n\n\nprevious findings in China (Li et al., 2021), US (Li et al., 2017a), and Norway (Springel, 2021).\n\n\nOne important caveat is that the calculated average government spending per induced sale in\n\n\nthis analysis only applies to central government incentives. Our data on incentives only include\n\n\nnational-level subsidies or tax reduction, but local subsidies account for 22% of total subsidies\n\n\non average in China while US states such as California offer rebates or tax credits equivalent to\n\n\n30% of the federal-level tax credit. The estimated impact from incentives might be larger if all\n\n\nlocal incentives were available in the data.\n\n##### **Conclusion**\n\n\nThis study provides a comprehensive analysis of global EV diffusion during the first decade\n\n\nof the technology being introduced to the mass market. The analysis shows that investing in\n\n\ncharging infrastructure is more cost-effective than providing consumer subsidies in promoting EV\n\n\nadoption, consistent with the finding from the literature based on data from individual countries.\n\n\nMore importantly, this study allows us to understand the significant variation in EV adoption\n\n\nacross countries, even among those with similar socio-economic characteristics. Our analysis\n\n\nshows that 17% of the cross-country variation is driven by observed differences in subsidy levels\n\n\nand 69% by differences in charging network size. These findings highlight the importance of\n\n\ninvesting in charging infrastructure in order to further electrify the transportation sector in the\n\n\nnext decade.\n\n\nWe conclude with a discussion on the limitations of our study and directions for future\n\n\nresearch. First, our data do not include gasoline models and do not allow\n\nMention: \"data from individual countries\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"Earlier single-country studies using such data reach the same conclusion on adoption cost-effectiveness.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002214:13:0", "occurrence_id": "2e5fc9d556e0a4fc695e177f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: input-output table includes the code 562 (Waste management and remediation services). Note\n\n\nthat this could mean residential or commercial waste management. Our focus only includes\n\n\n562112 (Hazardous Waste Collection) and 526221 (Hazardous Waste Treatment and Disposal)\n\n\nfrom the broader 562 category, because we are more confident that Hazardous Waste Collection\n\n\nserves the tradable sectors rather than local residents.\n\n\nWe also use Mian and Sufi (2014)’s data for several of our regressions. Most notably, we\n\n\nuse several pre-crisis county characteristics in 2007 as controls: percentage white, the log of\n\n\nmedian household income, the log of median home value, percentage of homes occupied by\n\n\nowners, percentage of population with less than high school diplomas, percentage with only high\n\n\nschool diplomas, percentage with college degrees, unemployment rate, poverty rate, and\n\n\npercentage urban. We have also included pre-crisis household leverage in 2006 as it serves as a\n\n\nstrong proxy for household cuts in consumption (see Mian and Sufi, 2010). Aside from this, we\n\n\nalso use their data for the robustness checks we conduct later in the paper, in which we test the\n\n\ninfluence of credit constraints on the change in supporting sector employment.\n\n\nSummary statistics for the variables used in our regressions are provided in Table 4.1.\n\n\nSupporting sector employment makes up 3.5% of the total employment in 2007, while tradable\n\n\nemployment makes up around 14.5%. On average, a given county lost a staggering 16.1% of\n\n\ntradable employment between 2007 and 2010. In fact, tradable job losses driven by aggregate\n\n\nshocks (i.e. Bartik) for a county between 2007 and 2010 are 18.3% on average.\n\n\nMost of the variables have full coverage. The only exception is pre-crisis leverage. This\n\n\nvariable represent how indebted households in a county were before the crisis. Mian and Sufi\n\n\n(2010) show that pre-crisis household leverage is strongly connected to the\n\nMention: \"Mian and Sufi (2014)\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its pre-crisis county characteristics are used as regression controls.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002217:25:0", "occurrence_id": "ef0babc47f8e695796325789", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: with high vs. low quality of financial institutions is a mere proxy for the differential effect\n\n\nof corruption across countries with varying levels of the quality of the broader business\n\n\nenvironment. The additional interaction term controls include interaction terms between\n\n\ncorruption (incidence and depth of petty corruption, separately) and the following: Rule\n\n\nof Law (Worldwide Governance Indicators), Gross Primary Education Enrollment rate\n\n\n(WDI), Gross Secondary Education Enrollment rate (WDI), Freedom from Business\n\n\nRegulation (Economic Freedom of the World), Legal Systems and Property Rights\n\n\n(Economic Freedom of the World), and hours of power outages faced by the firm in a\n\n\ntypical month (ES). These interaction term controls are in addition the ones between\n\n\ncorruption and GDP per capita (logs, PPP adjusted and in constant 2011 international\n\n\ndollars) and between corruption and _Annual Sales_ (firm-size). Data source for GDP per\n\n\ncapita is WDI, World Bank. Regression results in table 5 and 6 confirm our prediction.\n\n\nThat is, while higher corruption increases the likelihood of a firm being financially\n\n\nconstrained, the increase is significantly smaller in countries that have better financial\n\n\ninstitutions.\n\n\n5.2 _Falsification test_\n\n\nNext, we provide results from the falsification test. That is, we check if like corruption,\n\n\nthe impact of more burdensome regulation and higher power outages on exacerbating\n\n\nfirm’s financial constraints is smaller (less positive or more negative) in countries with\n\n\nbetter quality of financial institutions as measured by Getting Credit sub-index and the\n\n\npresence of credit bureaus. A failure to find such an effect will raise our confidence\n\n\nagainst endogeneity concerns with our main results. For regulation and in separate\n\n\nregressions, we use two measures – a dummy variable equal to 1 if the firm was inspected\n\n\nby tax officials and 0 otherwise; and percentage of firm’s senior management’s time spent\n\n\n24\n\nMention: \"Worldwide Governance Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Its rule of law measure is interacted with corruption in the long-term controls.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002217:25:0", "occurrence_id": "3469bcf9e86cb24e84f0f6ac", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: with high vs. low quality of financial institutions is a mere proxy for the differential effect\n\n\nof corruption across countries with varying levels of the quality of the broader business\n\n\nenvironment. The additional interaction term controls include interaction terms between\n\n\ncorruption (incidence and depth of petty corruption, separately) and the following: Rule\n\n\nof Law (Worldwide Governance Indicators), Gross Primary Education Enrollment rate\n\n\n(WDI), Gross Secondary Education Enrollment rate (WDI), Freedom from Business\n\n\nRegulation (Economic Freedom of the World), Legal Systems and Property Rights\n\n\n(Economic Freedom of the World), and hours of power outages faced by the firm in a\n\n\ntypical month (ES). These interaction term controls are in addition the ones between\n\n\ncorruption and GDP per capita (logs, PPP adjusted and in constant 2011 international\n\n\ndollars) and between corruption and _Annual Sales_ (firm-size). Data source for GDP per\n\n\ncapita is WDI, World Bank. Regression results in table 5 and 6 confirm our prediction.\n\n\nThat is, while higher corruption increases the likelihood of a firm being financially\n\n\nconstrained, the increase is significantly smaller in countries that have better financial\n\n\ninstitutions.\n\n\n5.2 _Falsification test_\n\n\nNext, we provide results from the falsification test. That is, we check if like corruption,\n\n\nthe impact of more burdensome regulation and higher power outages on exacerbating\n\n\nfirm’s financial constraints is smaller (less positive or more negative) in countries with\n\n\nbetter quality of financial institutions as measured by Getting Credit sub-index and the\n\n\npresence of credit bureaus. A failure to find such an effect will raise our confidence\n\n\nagainst endogeneity concerns with our main results. For regulation and in separate\n\n\nregressions, we use two measures – a dummy variable equal to 1 if the firm was inspected\n\n\nby tax officials and 0 otherwise; and percentage of firm’s senior management’s time spent\n\n\n24\n\nMention: \"Gross Primary Education Enrollment rate\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"This education indicator enters the interaction terms as a country characteristic.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002217:25:0", "occurrence_id": "8e65e39933627dea6e20105c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: with high vs. low quality of financial institutions is a mere proxy for the differential effect\n\n\nof corruption across countries with varying levels of the quality of the broader business\n\n\nenvironment. The additional interaction term controls include interaction terms between\n\n\ncorruption (incidence and depth of petty corruption, separately) and the following: Rule\n\n\nof Law (Worldwide Governance Indicators), Gross Primary Education Enrollment rate\n\n\n(WDI), Gross Secondary Education Enrollment rate (WDI), Freedom from Business\n\n\nRegulation (Economic Freedom of the World), Legal Systems and Property Rights\n\n\n(Economic Freedom of the World), and hours of power outages faced by the firm in a\n\n\ntypical month (ES). These interaction term controls are in addition the ones between\n\n\ncorruption and GDP per capita (logs, PPP adjusted and in constant 2011 international\n\n\ndollars) and between corruption and _Annual Sales_ (firm-size). Data source for GDP per\n\n\ncapita is WDI, World Bank. Regression results in table 5 and 6 confirm our prediction.\n\n\nThat is, while higher corruption increases the likelihood of a firm being financially\n\n\nconstrained, the increase is significantly smaller in countries that have better financial\n\n\ninstitutions.\n\n\n5.2 _Falsification test_\n\n\nNext, we provide results from the falsification test. That is, we check if like corruption,\n\n\nthe impact of more burdensome regulation and higher power outages on exacerbating\n\n\nfirm’s financial constraints is smaller (less positive or more negative) in countries with\n\n\nbetter quality of financial institutions as measured by Getting Credit sub-index and the\n\n\npresence of credit bureaus. A failure to find such an effect will raise our confidence\n\n\nagainst endogeneity concerns with our main results. For regulation and in separate\n\n\nregressions, we use two measures – a dummy variable equal to 1 if the firm was inspected\n\n\nby tax officials and 0 otherwise; and percentage of firm’s senior management’s time spent\n\n\n24\n\nMention: \"Gross Secondary Education Enrollment rate\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Secondary enrolment is the second education control in the interactions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002217:25:0", "occurrence_id": "f2a2b5fce7ceabfcf32c3c74", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: with high vs. low quality of financial institutions is a mere proxy for the differential effect\n\n\nof corruption across countries with varying levels of the quality of the broader business\n\n\nenvironment. The additional interaction term controls include interaction terms between\n\n\ncorruption (incidence and depth of petty corruption, separately) and the following: Rule\n\n\nof Law (Worldwide Governance Indicators), Gross Primary Education Enrollment rate\n\n\n(WDI), Gross Secondary Education Enrollment rate (WDI), Freedom from Business\n\n\nRegulation (Economic Freedom of the World), Legal Systems and Property Rights\n\n\n(Economic Freedom of the World), and hours of power outages faced by the firm in a\n\n\ntypical month (ES). These interaction term controls are in addition the ones between\n\n\ncorruption and GDP per capita (logs, PPP adjusted and in constant 2011 international\n\n\ndollars) and between corruption and _Annual Sales_ (firm-size). Data source for GDP per\n\n\ncapita is WDI, World Bank. Regression results in table 5 and 6 confirm our prediction.\n\n\nThat is, while higher corruption increases the likelihood of a firm being financially\n\n\nconstrained, the increase is significantly smaller in countries that have better financial\n\n\ninstitutions.\n\n\n5.2 _Falsification test_\n\n\nNext, we provide results from the falsification test. That is, we check if like corruption,\n\n\nthe impact of more burdensome regulation and higher power outages on exacerbating\n\n\nfirm’s financial constraints is smaller (less positive or more negative) in countries with\n\n\nbetter quality of financial institutions as measured by Getting Credit sub-index and the\n\n\npresence of credit bureaus. A failure to find such an effect will raise our confidence\n\n\nagainst endogeneity concerns with our main results. For regulation and in separate\n\n\nregressions, we use two measures – a dummy variable equal to 1 if the firm was inspected\n\n\nby tax officials and 0 otherwise; and percentage of firm’s senior management’s time spent\n\n\n24\n\nMention: \"WDI\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"GDP per capita in constant international dollars is taken from this source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002217:25:0", "occurrence_id": "9b7b4bc9ec19ad7d2a87e0f2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: with high vs. low quality of financial institutions is a mere proxy for the differential effect\n\n\nof corruption across countries with varying levels of the quality of the broader business\n\n\nenvironment. The additional interaction term controls include interaction terms between\n\n\ncorruption (incidence and depth of petty corruption, separately) and the following: Rule\n\n\nof Law (Worldwide Governance Indicators), Gross Primary Education Enrollment rate\n\n\n(WDI), Gross Secondary Education Enrollment rate (WDI), Freedom from Business\n\n\nRegulation (Economic Freedom of the World), Legal Systems and Property Rights\n\n\n(Economic Freedom of the World), and hours of power outages faced by the firm in a\n\n\ntypical month (ES). These interaction term controls are in addition the ones between\n\n\ncorruption and GDP per capita (logs, PPP adjusted and in constant 2011 international\n\n\ndollars) and between corruption and _Annual Sales_ (firm-size). Data source for GDP per\n\n\ncapita is WDI, World Bank. Regression results in table 5 and 6 confirm our prediction.\n\n\nThat is, while higher corruption increases the likelihood of a firm being financially\n\n\nconstrained, the increase is significantly smaller in countries that have better financial\n\n\ninstitutions.\n\n\n5.2 _Falsification test_\n\n\nNext, we provide results from the falsification test. That is, we check if like corruption,\n\n\nthe impact of more burdensome regulation and higher power outages on exacerbating\n\n\nfirm’s financial constraints is smaller (less positive or more negative) in countries with\n\n\nbetter quality of financial institutions as measured by Getting Credit sub-index and the\n\n\npresence of credit bureaus. A failure to find such an effect will raise our confidence\n\n\nagainst endogeneity concerns with our main results. For regulation and in separate\n\n\nregressions, we use two measures – a dummy variable equal to 1 if the firm was inspected\n\n\nby tax officials and 0 otherwise; and percentage of firm’s senior management’s time spent\n\n\n24\n\nMention: \"Getting Credit sub-index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"It measures financial institution quality in the interaction with firms' financial constraints.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002218:3:0", "occurrence_id": "1901da0551ef3add51149100", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ###### **1 Introduction**\n\nThe COVID-19 (Coronavirus) pandemic and the associated containment measures are causing far-reaching\n\ndamage to firms around the world. Firms are suffering from reduced demand due to lockdowns and move\nment restrictions. Yet, the limited availability of firm censuses and survey data in many developing countries\n\nhampers policymakers’ ability to quantify the economic impact of the pandemic and the effect of policy\n\nmeasures to support firms. In this paper, we use a novel dataset of administrative corporate tax returns from\n\n10 low- and middle-income countries to provide plausible estimates of the direct effects of the lockdown\n\non firms’ profits, payrolls, and exit rates. We then derive the implications for tax revenues and government\n\nsupport policies. On average across countries, our lower-bound predictions (optimistic scenario) suggest\n\nthat only half of all firms remain profitable, tax revenue remitted by corporations falls by 1.5% of GDP, and\n\naggregate corporate losses increase by 2.9% of GDP.\n\n\nOur analysis relies on a few simple and transparent assumptions about the shock firms face and the\n\nadjustments they make. We simulate scenarios in which a demand shock induces a drop in firms’ sales over\n\neither three or five months – which can be thought of as the length of the lockdown with perfect compliance\n\nto governments’ instructions, or of reduced activity for a longer period with imperfect compliance. The\n\nseverity of the shock — the percentage drop in monthly sales — differs across economic sectors. We assume\n\nthat firms produce a unit of output with a Leontieff production function which requires capital, labor and\n\nmaterial inputs in fixed proportions, with the proportions estimated from each firm’s tax declaration. In our\n\nvery stylized world, firms can reduce their material costs proportionally to the drop in demand; they reduce\n\nlabor costs only when making losses because re-contracting workers is costly; and they cannot adjust their\n\nfixed costs. Implicitly, we assume that\n\nMention: \"dataset of administrative corporate tax returns\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Firm profits, payrolls and exit following lockdown are estimated from ten countries' returns.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002222:13:0", "occurrence_id": "a19d822fda5bb686b62d98f1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: We manually constructed the FDI regulations data by comparing the 1997 and 2002 Catalogue for the\n\nGuidance of Foreign Investment Industries. The 2002 revision was substantial and in strict accordance with\n\nthe commitments made by China’s central government in its negotiations with WTO members before 2001.\n\nThe changes are plausibly exogenous to China’s domestic situations, as the process took 15 years and\n\ninvolved numerous negotiations with many WTO member countries.\n\n\nThe Catalogue classifies sectors and products into four categories: (1) FDI was encouraged; (2) FDI\n\nwas permitted, these industries and products are not listed in the Catalogue; (3) FDI was restricted; and (4)\n\nFDI was prohibited. We first map each industry and product to the 4-digit industry in 2002 industry\n\nclassification, then assign value 1 to all industries that are encouraged, 2 to permitted industries, 3 to\n\nrestricted industries, and 4 to prohibited industries. We do these two steps for the 1997 and 2002 Catalogue\n\nrespectively, and then calculate the change in FDI policy for each industry from 1997 to 2002. Positive\n\nchanges indicate the industry became more restricted after 2002, while negative changes mean the industry\n\nbecame more open to FDI. About 300 industries experienced no changes in FDI regulations. Five industries\n\nhad mixed changes, where one or more products in the industry became more open while one or more\n\nproducts became more restricted. We consider the industries that became more open to FDI the treatment\n\ngroup, all other industries are the control group.\n\n\n**Prefecture level human capital stock data**\n\nTo test how local human capital stock affects FDI spillovers, we use China’s population census data in 2000\n\nand calculate the share of adult population aged between 18-55 with a bachelor’s degree or above at the\n\nprefecture level. We used the 2000 data to mitigate endogeneity, as FDI liberalization in 2002 could induce\n\nlabor migration, which affects local human capital stock afterwards.\n\n\n**Summary statistics**\n\nFrom pre-treatment period to post\n\nMention: \"Catalogue for the\n\nGuidance of Foreign Investment Industries\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"Comparing the two editions yields the manually constructed regulation dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002222:13:0", "occurrence_id": "78b349bdd5d18227b81871b7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: We manually constructed the FDI regulations data by comparing the 1997 and 2002 Catalogue for the\n\nGuidance of Foreign Investment Industries. The 2002 revision was substantial and in strict accordance with\n\nthe commitments made by China’s central government in its negotiations with WTO members before 2001.\n\nThe changes are plausibly exogenous to China’s domestic situations, as the process took 15 years and\n\ninvolved numerous negotiations with many WTO member countries.\n\n\nThe Catalogue classifies sectors and products into four categories: (1) FDI was encouraged; (2) FDI\n\nwas permitted, these industries and products are not listed in the Catalogue; (3) FDI was restricted; and (4)\n\nFDI was prohibited. We first map each industry and product to the 4-digit industry in 2002 industry\n\nclassification, then assign value 1 to all industries that are encouraged, 2 to permitted industries, 3 to\n\nrestricted industries, and 4 to prohibited industries. We do these two steps for the 1997 and 2002 Catalogue\n\nrespectively, and then calculate the change in FDI policy for each industry from 1997 to 2002. Positive\n\nchanges indicate the industry became more restricted after 2002, while negative changes mean the industry\n\nbecame more open to FDI. About 300 industries experienced no changes in FDI regulations. Five industries\n\nhad mixed changes, where one or more products in the industry became more open while one or more\n\nproducts became more restricted. We consider the industries that became more open to FDI the treatment\n\ngroup, all other industries are the control group.\n\n\n**Prefecture level human capital stock data**\n\nTo test how local human capital stock affects FDI spillovers, we use China’s population census data in 2000\n\nand calculate the share of adult population aged between 18-55 with a bachelor’s degree or above at the\n\nprefecture level. We used the 2000 data to mitigate endogeneity, as FDI liberalization in 2002 could induce\n\nlabor migration, which affects local human capital stock afterwards.\n\n\n**Summary statistics**\n\nFrom pre-treatment period to post\n\nMention: \"China’s population census data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The share of adults with tertiary education is computed at prefecture level.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002228:30:2", "occurrence_id": "5497b5bc4a8881602aa90ca1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: )\n0-1 quarters after -0.265 -0.126 -0.035 -0.022 -0.014 -0.024\n(0.195) (0.079) (0.025) (0.022) (0.022) (0.018)\n0-1 quarters after 0.092 0.075 0.048 0.035 0.015 0.042\nx Garments (0.338) (0.134) (0.033) (0.030) (0.025) (0.031)\n2+ quarters after 0.017 -0.047 -0.061*** -0.031* -0.031 -0.007\n(0.159) (0.058) (0.020) (0.019) (0.020) (0.016)\n2+ quarters after -0.567* 0.193 0.069** 0.069** 0.000 -0.012\nx Garments (0.292) (0.118) (0.028) (0.027) (0.026) (0.030)\n<u>Average</u> <u>35.810</u> <u>4.867</u> <u>12.740</u> <u>7.699</u> <u>5.039</u> <u>0.358</u>\n\n\n\nAge\n\n\n\nEduc\nation\n\n\n\n_Source_ : Author’s calculations from Cambodia SES 2007-2015.\n_Note_ : The table is identical to Table 2 except that it restricts the observations to those surveyed within 6\nquarters of the minimum wage hike. Standard errors, clustered by survey year _×_ psu shown in parentheses.\n***: p _<_ 0.01; *\n\nMention: \"Cambodia SES 2007-2015\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Observations surveyed within six quarters of the minimum wage hike are used for the sample.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002215:9:1", "occurrence_id": "d0e2ea6f08f6479fbca8009e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .6 where\napproximately 2.6 records are used for calling. Like other developing countries, multiple-SIM-card\nholding is common in The Gambia. We expect a certain overlap between the two-MNO subscribers,\nwhich might have resulted in over-representing the multiple-SIM-card holders. In this study the impact\nof the multi-SIM holding on the analysis result is considered to be limited since the two MNOs primarily\nmarket different socio-economic groups. One of them is a leading MNO in The Gambia and is popular\nin urban areas with high-speed internet services. The other MNO provides only voice and shortmessaging services with inexpensive plans, which are much popular in rural areas.\n\nThe preparation and analysis of CDR data under this project are based on a protocol to address concerns\nof privacy and confidentiality. Raw CDR data include several identifiers associated with each record,\nsuch as phone number, IMSI and IMEI. We employ a three-stage approach to anonymize these\nidentifiers in order to protect data privacy.\n\n  - First, identifiers are encrypted using a one-way function by the MNOs on their premises to\nprotect the data privacy.\n\n  - Second, lists of cell towers and the locations provided by the MNOs are pooled in the\nregulator’s premise and cell-tower locations are clustered using Ward’s hierarchical clustering,\nwith a maximum distance constraint of 1km from the centroid of the cluster. <sup>3</sup> [^3: The choices of a clustering methodology and a distance threshold are proposed by the World Bank COVID19\nMobility Task Force.] We then use the\ncentroid of the cluster to match and map the de-identified CDR data to their respective celltower locations. This process lowers the spatial granularity of cell-tower distributions.\n\n  - Third, the results of all indicators are aggregated at the administrative unit level. There are\ncertain concerns about reverse engineering for the re-identification of de-identified CDR data\n(Daniel Kondor, Hashemian, de Montjoye, & Ratti, 2018)\n\nMention: \"CDR data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"Preparation follows a protocol addressing privacy, since raw records contain personal identifiers.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002215:9:1", "occurrence_id": "4ac4b71f6cb54ce134bb8f79", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .6 where\napproximately 2.6 records are used for calling. Like other developing countries, multiple-SIM-card\nholding is common in The Gambia. We expect a certain overlap between the two-MNO subscribers,\nwhich might have resulted in over-representing the multiple-SIM-card holders. In this study the impact\nof the multi-SIM holding on the analysis result is considered to be limited since the two MNOs primarily\nmarket different socio-economic groups. One of them is a leading MNO in The Gambia and is popular\nin urban areas with high-speed internet services. The other MNO provides only voice and shortmessaging services with inexpensive plans, which are much popular in rural areas.\n\nThe preparation and analysis of CDR data under this project are based on a protocol to address concerns\nof privacy and confidentiality. Raw CDR data include several identifiers associated with each record,\nsuch as phone number, IMSI and IMEI. We employ a three-stage approach to anonymize these\nidentifiers in order to protect data privacy.\n\n  - First, identifiers are encrypted using a one-way function by the MNOs on their premises to\nprotect the data privacy.\n\n  - Second, lists of cell towers and the locations provided by the MNOs are pooled in the\nregulator’s premise and cell-tower locations are clustered using Ward’s hierarchical clustering,\nwith a maximum distance constraint of 1km from the centroid of the cluster. <sup>3</sup> [^3: The choices of a clustering methodology and a distance threshold are proposed by the World Bank COVID19\nMobility Task Force.] We then use the\ncentroid of the cluster to match and map the de-identified CDR data to their respective celltower locations. This process lowers the spatial granularity of cell-tower distributions.\n\n  - Third, the results of all indicators are aggregated at the administrative unit level. There are\ncertain concerns about reverse engineering for the re-identification of de-identified CDR data\n(Daniel Kondor, Hashemian, de Montjoye, & Ratti, 2018)\n\nMention: \"de-identified CDR data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Cluster centroids are matched to cell tower locations, reducing spatial granularity.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002235:29:1", "occurrence_id": "f6cdc661cd0160cfab3c19f4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: sup> 0.0521 0.0462\n\n\n(0.0290) (0.0283) (0.0275) (0.0275) (0.0277) (0.0415) (0.0422) (0.0293)\n\n\nObservations 1939 1939 1932 1932 1932 1932 1932 1932\n\n\nPanel C: Rural\n\n\nWomen 0.121 <sup>_∗∗∗_</sup> 0.126 <sup>_∗∗∗_</sup> 0.118 <sup>_∗∗∗_</sup> 0.118 <sup>_∗∗∗_</sup> 0.115 <sup>_∗∗∗_</sup> 0.0888 <sup>_∗∗_</sup> 0.0785 <sup>_∗∗_</sup> 0.111 <sup>_∗∗∗_</sup>\n\n\n(0.0250) (0.0241) (0.0240) (0.0240) (0.0242) (0.0381) (0.0388) (0.0250)\n\n\nObservations 1262 1262 1261 1261 1261 1261 1261 1261\n\n\nControls\n\n\nEducation, age and month X X X X X X X\n\n\nWFH with broadband access and essential sectors X X X X X X\n\n\nRural X X X X X\n\n\nInformality X X X X\n\n\nChildren X X\n\n\nFemale dominated sectors X X\n\n\n_Notes:_ Based on data from ENAHO 2019-2020 (second quarter) and Hatayama et al. (2020). Estimation\nusing OLS regressions. WFH is defined for workers in the top 25 percent of the distribution and considers\nindividuals with broadband access (i.e. households with\n\nMention: \"ENAHO 2019-2020\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Working-from-home incidence by sector, rurality and informality is estimated from this quarter.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002236:7:0", "occurrence_id": "174159a27592007971a37e0f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: #### **2 Data**\n\n###### **2.1 Lockdown data**\n\nLockdown is a binary variable where one indicates the month the following measures were\n\n\nimposed by trading partners of the Philippines: workplace closure, the closure of public\n\n\ntransport, stay at home requirements, restrictions on internal movement, and international\n\n\ntravel controls. This definition of lockdowns has been adopted from Majune (2020). Data for\n\n\nthese indicators is from the Oxford COVID-19 Government Response Tracker (Hale et al.,\n\n\n2020).\n\n\nTo explain the pattern of lockdown measures, we display the trend of the lockdown\n\n\nStringency Index for the Philippines and neighboring countries (Thailand, Malaysia, China,\n\n\nIndonesia, and Vietnam) between January 1, 2020, and December 31, 2020, in Figure 1.\n\n\nThe index is constructed by re-scaling the maximum ordinal values of nine indicators (one\n\n\nhealth measure and eight closures and containment policies) to range between 0 and 100.\n\n\nThe scores are then averaged to get a composite index for over 180 countries. A score of 0\n\n\nindicates no restrictions while 100 signals severe implementation of lockdown policies in a\n\n\ncountry (Hale et al., 2020).\n\n\nThe first lockdown measure in the Philippines was on January 31, 2020, when arrivals\n\n\nof people from some regions were banned. This was followed by the closure of the border,\n\n\nworkplaces, public transport, and imposition of stay-at-home requirements and restrictions\n\n\non internal movement on March 15, 2020, as reflected by a sudden jump in the index, from\n\n\n34 to 75. The index was 100 on March 22, 2020, indicating maximum implementation of\n\n\nall containment measures. However, it dropped significantly, on May 29, 2020, when work\n\nplace and public transport closure, stay at home requirements and restrictions on internal\n\n\nmovement were temporarily suspended. The easing of international travel controls started\n\n\non July 7, 2020. The Philippines had the joint least index\n\nMention: \"Oxford COVID-19 Government Response Tracker\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its stringency index and lockdown indicators describe the measures adopted.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002238:4:0", "occurrence_id": "a17f0ed36a90888e8c86e2d2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: during a financial crisis. Moreover, welfare gains from policy may become negative.\n\nEmpirically, I present evidence from Brazilian data that expectations of economic\n\nfundamentals respond to capital control announcements: first, the paper uses a high\nfrequency strategy to show that such announcements tend to affect daily forecasts\n\nof GDP collected by the Central Bank of Brazil; second, the paper constructs an\n\neconometric model to nowcast Brazilian GDP with real-time macroeconomic data,\n\nand shows that such announcements reduce the forecast errors of the model.\n\n\nIn the theoretical section, I incorporate learning from policy into a standard model\n\nresearchers use to study capital controls in emerging market economies. In this two\nsector small open economy model, borrowing by domestic agent is limited by tradable\n\nand nontradable income through a collateral constraint. This environment features\n\na pecuniary externality because individual agents do not internalize the effect of\n\ntheir borrowing decisions on the value of collaterals and hence on borrowing capacity.\n\nCapital controls are usually deployed by the government to mitigate the effect of over\nborrowing by agents arising from the pecuniary externality. I incorporate three novel\n\nfeatures in this setting to model learning from policy: first, the government knows\n\nmore about future economic fundamentals than the representative agent. Second,\n\nthe agent knows that the government is levying capital control taxes according to a\n\nsimple policy rule which reveals useful information previously unknown to her. Last\n\nbut not least, the agent learns from policy by completing a Bayesian updating of her\n\ninformation set when she observes policy changes.\n\n\nA version of my model is calibrated to the Brazilian economy and the quantitative\n\nmodel features unintended consequences of capital controls when people learn from\n\npolicy. I choose the Brazilian economy between 2006 and 2013 because the country\n\nexperienced large and volatile capital inflows during the period and actively used\n\ncapital controls to manage such inflows with some success. I solve the calibrated\nmodel for the decentralized equilibrium <sup>2</sup>, the equilibrium with policy interventions\n\nbut no learning from policy and most importantly, the equilibrium\n\nMention: \"real-time macroeconomic data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"It feeds the nowcasting model whose forecast errors fall after announcements.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002239:4:0", "occurrence_id": "748a1f408feff01d581e8135", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: to value that consumption, irrespective of source, at market prices. Diaries instead collect information\n\n\non food _acquired_ during the interview period regardless of when the food was consumed. There is little\n\n\nevidence on the best approach for managing the transition to recall interviews, with most countries\n\n\naccepting a break in the series on poverty, food security, and welfare aggregates, given that the survey\n\n\nmode affects indicators of poverty and inequality in the cross-section (Beegle et al., 2012) and over time\n\n\n(Battistin, 2003, Pudney, 2008, Attanasio and Pistaferri, 2016, and Coibion et al., 2020).\n\n\nThis paper studies how to obtain the most accurate measurements of population well-being in house\n\nhold expenditure surveys. Specifically, we define a household’s well-being as their _usual_ (or _average_ ) rate\n\n\nof consumption in a fixed period, which in our application is one week. Usual consumption is a longer-run\n\n\naverage of weekly intakes and the quantity grounding welfare comparisons on the theoretical concept\n\n\nof consumption smoothing. The empirical challenge then is to use survey measurements collected as a\n\n\nsnapshot, typically covering one week, to infer usual consumption. We develop a strategy to identify the\n\n\ndistribution of usual consumption, which is _unobserved_ because of the short-term nature of interviews,\n\n\nwithout applying any corrections to raw data or assuming _ex_ _ante_ which mode of collection - diaries or\n\n\nrecall questions - yields the most accurate measurement.\n\n\nWe frame the identification of consumer welfare in a general setting, encompassing “survey errors��  \n\ndefined here as differences between survey measurements and usual consumption - related to household\n\n\npreferences and misreporting behavior. Specifically, households with different consumption preferences in\n\n\nthe interview week but planning for the same dollar amount of consumption over a longer period (e.g.,\n\n\na month) will have the same well-being. Such _irregularity_ _of_ _\n\nMention: \"survey measurements\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Collected as a one-week snapshot, they must be used to infer usual consumption.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002243:9:0", "occurrence_id": "79923fbe0d5528a9c5435ff2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The preceding discussion is based on a static, cross-sectional view of poverty. As noted, Malaysia\n\ndoes not collect panel survey data on household income or consumption, so there is limited knowledge\n\nof poverty and income dynamics at the micro-level. However, some studies have investigated socio\neconomic intergenerational mobility based on retrospective data (Khalid 2018, Hamid et al. 2019). Khalid\n\n(2018) finds that the majority of younger-generation Malaysians (birth cohort 1975–1985) have higher\n\neducational attainment and better occupational skills than the previous generation (the cohort born\n\nbetween 1945–1960). Income mobility patterns across generations are mixed, but mobility was found to\n\nbe highest for children born to parents in the lowest income quintile. This is consistent with the image of\n\nreductions in poverty and inequality in official Malaysian statistics. For example, Khalid (2018) finds that\n\nchildren born in the poorest households will in general not be poor as adults, suggesting that long-run\n\nchronic poverty is uncommon. However, he also observes that upward intergenerational mobility was less\n\nlikely in rural areas than in urban ones. Despite positive findings about past mobility, Khalid argues that\n\nsuch patterns are unlikely to be repeated for children born in the 1990s and later, e.g., due to changes in\n\nMalaysia’s education policy. This connects with the criticisms discussed above that preferential policies\n\nfavor those who are already richer than the average Malaysian.\n\n\n7\n\nMention: \"panel survey data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"The country does not collect it, so little is known about micro-level income dynamics.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002243:9:0", "occurrence_id": "3b2837d5b97a4c5f2f607e55", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The preceding discussion is based on a static, cross-sectional view of poverty. As noted, Malaysia\n\ndoes not collect panel survey data on household income or consumption, so there is limited knowledge\n\nof poverty and income dynamics at the micro-level. However, some studies have investigated socio\neconomic intergenerational mobility based on retrospective data (Khalid 2018, Hamid et al. 2019). Khalid\n\n(2018) finds that the majority of younger-generation Malaysians (birth cohort 1975–1985) have higher\n\neducational attainment and better occupational skills than the previous generation (the cohort born\n\nbetween 1945–1960). Income mobility patterns across generations are mixed, but mobility was found to\n\nbe highest for children born to parents in the lowest income quintile. This is consistent with the image of\n\nreductions in poverty and inequality in official Malaysian statistics. For example, Khalid (2018) finds that\n\nchildren born in the poorest households will in general not be poor as adults, suggesting that long-run\n\nchronic poverty is uncommon. However, he also observes that upward intergenerational mobility was less\n\nlikely in rural areas than in urban ones. Despite positive findings about past mobility, Khalid argues that\n\nsuch patterns are unlikely to be repeated for children born in the 1990s and later, e.g., due to changes in\n\nMalaysia’s education policy. This connects with the criticisms discussed above that preferential policies\n\nfavor those who are already richer than the average Malaysian.\n\n\n7\n\nMention: \"retrospective data\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"Cited studies use it to study intergenerational socio-economic mobility.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002249:30:0", "occurrence_id": "89533d3fdaadeaff797dbca4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: mean value of the independent variable is 0.085 - that is, on average women are close to 0.085\n\n\nmines. Conditional on being close to a mine, the main independent variable is 1.33 - that is, a\n\n\nwoman sampled close to a mine is close to 1.33 mines on average. Panel C of table A3 shows\n\n\nthe effects on women’s labor market participation. We note that the estimates are similar in\n\n\ndirection as before, where mines are positively associated with service and sales jobs and with\n\n\ncash earnings, but negatively associated with agriculture and professional jobs.\n\n\nPanel D of table A3 shows the results if we drop the part of the sample that lives 20 km to 40\n\n\nkm away from a mine, and if we drop those that are sampled two years before mine opening.\n\n\nThe rationale for this is to have a cleaner control group, since those that live just outside our 20\n\n\nkm cutoff distance may also be “treated” by the mine, and the investment phase of the mine\n\n\nthat precedes initial production can generate substantial employment. Overall, the effects do\n\n\nnot change much except making the cash earnings coefficient larger and more significant. The\n\n\nincrease in cash earning opportunities is estimated at 7.5 to 7.8 percentage points compared\n\n\nwith 5.4 in the baseline estimation.\n\n\n**6.6 Employment and wages using the GLSS**\n\n\nThe DHS data do not provide detailed information regarding how much an individual earns for\n\n\nwork, or her wage rate, but the GLSS does collect such data. First, we try to replicate the results\n\n\nestimated with the DHS data. Panel A of table 11 indicates that agriculture becomes less\n\n\nimportant in mining communities for women (statistically insignificant), who mainly shift into\n\n\nservices and sales (statistically insignificant, except for strategy 2). Men are more likely to work\n\n\nas miners (statistically significant across all strategies).\n\n\n29\n\nMention: \"GLSS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"It collects individual earnings and wage rates that the other survey lacks.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002249:30:0", "occurrence_id": "d4a5dfa0180c73509e0516f5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: mean value of the independent variable is 0.085 - that is, on average women are close to 0.085\n\n\nmines. Conditional on being close to a mine, the main independent variable is 1.33 - that is, a\n\n\nwoman sampled close to a mine is close to 1.33 mines on average. Panel C of table A3 shows\n\n\nthe effects on women’s labor market participation. We note that the estimates are similar in\n\n\ndirection as before, where mines are positively associated with service and sales jobs and with\n\n\ncash earnings, but negatively associated with agriculture and professional jobs.\n\n\nPanel D of table A3 shows the results if we drop the part of the sample that lives 20 km to 40\n\n\nkm away from a mine, and if we drop those that are sampled two years before mine opening.\n\n\nThe rationale for this is to have a cleaner control group, since those that live just outside our 20\n\n\nkm cutoff distance may also be “treated” by the mine, and the investment phase of the mine\n\n\nthat precedes initial production can generate substantial employment. Overall, the effects do\n\n\nnot change much except making the cash earnings coefficient larger and more significant. The\n\n\nincrease in cash earning opportunities is estimated at 7.5 to 7.8 percentage points compared\n\n\nwith 5.4 in the baseline estimation.\n\n\n**6.6 Employment and wages using the GLSS**\n\n\nThe DHS data do not provide detailed information regarding how much an individual earns for\n\n\nwork, or her wage rate, but the GLSS does collect such data. First, we try to replicate the results\n\n\nestimated with the DHS data. Panel A of table 11 indicates that agriculture becomes less\n\n\nimportant in mining communities for women (statistically insignificant), who mainly shift into\n\n\nservices and sales (statistically insignificant, except for strategy 2). Men are more likely to work\n\n\nas miners (statistically significant across all strategies).\n\n\n29\n\nMention: \"DHS data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"It does not report how much individuals earn or their wage rate.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002250:7:1", "occurrence_id": "2f144da51a7f211c952930ac", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  2015) for recent media coverage of Vietnam’s\nperformance on the PISA. See also Glewwe et al. (2017) for a recent attempt to disentangle this education phenomenon.\n6 Furthermore, our study also contributes to an emerging literature that examines the negative effects of war on human\ncapital but mostly focuses on richer countries; see, for example, Shemyakina (2011), Kesternich _et al._ (2014), and\nSwee (2015). But in contrast with this literature’s focus on the _negative_ impacts of wars, we offer an investigation into\na _positive_ shock to education (supply) during wartime. The positive shock is caused by beneficial exposure to the\nDRV’s popular education program during the war. To some extent, a better understanding of positive shocks—rather\nthan negative shocks—is perhaps more relevant to policies that aim at future education expansions. Another major\ndifference is that while this literature (mostly) studies the generally war-related disruptive effects, we offer a specific\ninvestigation of the long-term impacts of schooling policies in the context of the Indochina War.\n\n\n6\n\nMention: \"PISA\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"Media coverage of the country's performance is cited as context for the phenomenon studied.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002251:51:1", "occurrence_id": "24d1b962d5c14e88fde9279f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _ <mark>71, no. 4 (2017): 607-628.</mark>\n\n<mark>Shemyakina, Olga. \"The effect of armed conflict on accumulation of schooling: Results from</mark>\n<mark>Tajikistan.\"</mark> _<mark>Journal of Development Economics</mark>_ <mark>95, no. 2 (2011): 186-200.</mark>\n\n“The Beginning of the Construction Process after Liberation: Tutoring, a Great Human Educational\nAchievement Achieved by the Kuwaiti Teachers Association in the Worst Conditions.” (in Arabic).\nThe Teacher Magazine, August (1991): 37-38.\n\n<mark>UNESCO-IBE.</mark> <mark>“World</mark> <mark>Data</mark> <mark>on</mark> <mark>Education:</mark> <mark>Sixth</mark> <mark>edition</mark> <mark>2006-07.”</mark> <mark>(2007).</mark> <mark>Web:</mark>\n<mark>http://www.ibe.unesco.org/en/document/world-data-education-sixth-edition-2006-07.</mark>\n<mark>Accessed: January-18-2021.</mark>\n\n<mark>UNESCO-IBE. “World Data on Education: Seventh edition 2010-11.” (2012). Web:</mark>\n<mark>http://www.ibe.unesco.org/en/document/world-data-education-seventh-edition-2010-11.</mark>\n<mark>Accessed: January-18-2021.</mark>\n\n<mark>Weldeegzie, Samuel G.\n\nMention: \"World Data on Education\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"A reference list entry giving the editions and access date of this education database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002237:1:0", "occurrence_id": "51244bc6f251302d853ed979", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Policy Research Working Paper 9270\n\n### **Abstract**\n\nHardly any literature exists on the relationship between\nequivalence scales and poverty dynamics for transitional\ncountries. This paper offers a new study on the impacts\nof equivalence scale adjustments on poverty dynamics\nin the Russian Federation, using equivalence scales constructed from subjective wealth and more than 20 waves of\nhousehold panel survey data from the Russia Longitudinal\nMonitoring Survey. The analysis suggests that the equivalence scale elasticity is sensitive to household demographic\ncomposition. The adjustments for the equivalence of scales\n\n\n\nresult in lower estimates of poverty lines. The study decomposes poverty into chronic and transient components and\nfinds that chronic poverty is positively related to the adult\n\nscale parameter. However, chronic poverty is less sensitive\nto the child scale factor compared with the adult scale\nfactor. Interestingly, the direction of income mobility might\nchange depending on the specific scale parameters that are\nemployed. The results are robust to different measures of\nchronic poverty, income expectations, reference groups,\nfunctional forms, and various other specifications.\n\n\n\nThis paper is a product of the Development Data Group, Development Economics and the Office of the Chief Economist,\n\nEurope and Central Asia Region. It is part of a larger effort by the World Bank to provide open access to its research and\nmake a contribution to development policy discussions around the world. Policy Research Working Papers are also posted\non the Web at http://www.worldbank.org/prwp. The authors may be contacted at hdang@worldbank.org.\n\n\n_The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development_\n_issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the_\n_names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those_\n_of the authors. They do not necessarily represent the views of the\n\nMention: \"Russia Longitudinal\nMonitoring Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Over twenty panel waves supply the data for constructing equivalence scales.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002253:50:0", "occurrence_id": "09fc8bcddc21320ae927aea2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Appendix S2. Tariffs, Data and Descriptive Statistics**\n\nThe main source is the reform dataset compiled by the Research Department of the IMF (Prati,\nOnorato, and Papageorgiou 2013; Giuliano, Mishra, and Spilimbergo 2013) which cover an\nunbalanced sample of 151 countries from 1964 to 2004. We extend the data to 2014 using tariff\ndata from the World Integrated Trade Solution (WITS) and World Development Indicators\n(WDI). Tariff rates at the country-level are computed by aggregating product-level tariff data\nusing weighted averages, with weights given by the import share of each product, measured as\nfractions of value. We do extensive checks of the data to ensure that they do not contain\nspurious jumps, by checking each jump is supported by country and policy reports.\n\nAdvanced economies comprise 28 percent of the sample, emerging markets around 44 percent\nand low-income economies around 28 percent.\n\n**<u>Table S2.1.</u>** <u>Descriptive Statistics for Tariffs</u>\n\n|Col1|Number of<br>observations|Mean|Standard<br>Deviation|Minimum|Maximum|\n|---|---|---|---|---|---|\n|**Tariffs (%)**<br>All<br>AM<br>EM<br>LIC<br>**Annual Tariff Changes (ppt)**<br>All<br>AM<br>EM<br>LIC|5042<br>1427<br>2161<br>1422<br>4702<br>1392<br>1960<br>1320|11.2<br>5.7<br>12.5<br>15.1<br>-\n\nMention: \"reform dataset\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its dates of financial and trade reforms across 151 countries are the main source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002253:50:0", "occurrence_id": "f307f07dd3ef33a55895e728", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Appendix S2. Tariffs, Data and Descriptive Statistics**\n\nThe main source is the reform dataset compiled by the Research Department of the IMF (Prati,\nOnorato, and Papageorgiou 2013; Giuliano, Mishra, and Spilimbergo 2013) which cover an\nunbalanced sample of 151 countries from 1964 to 2004. We extend the data to 2014 using tariff\ndata from the World Integrated Trade Solution (WITS) and World Development Indicators\n(WDI). Tariff rates at the country-level are computed by aggregating product-level tariff data\nusing weighted averages, with weights given by the import share of each product, measured as\nfractions of value. We do extensive checks of the data to ensure that they do not contain\nspurious jumps, by checking each jump is supported by country and policy reports.\n\nAdvanced economies comprise 28 percent of the sample, emerging markets around 44 percent\nand low-income economies around 28 percent.\n\n**<u>Table S2.1.</u>** <u>Descriptive Statistics for Tariffs</u>\n\n|Col1|Number of<br>observations|Mean|Standard<br>Deviation|Minimum|Maximum|\n|---|---|---|---|---|---|\n|**Tariffs (%)**<br>All<br>AM<br>EM<br>LIC<br>**Annual Tariff Changes (ppt)**<br>All<br>AM<br>EM<br>LIC|5042<br>1427<br>2161<br>1422<br>4702<br>1392<br>1960<br>1320|11.2<br>5.7<br>12.5<br>15.1<br>-\n\nMention: \"World Integrated Trade Solution\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its tariff data extend the reform series beyond the original end year.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002253:50:0", "occurrence_id": "127339eb337f8c74fc0e283d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Appendix S2. Tariffs, Data and Descriptive Statistics**\n\nThe main source is the reform dataset compiled by the Research Department of the IMF (Prati,\nOnorato, and Papageorgiou 2013; Giuliano, Mishra, and Spilimbergo 2013) which cover an\nunbalanced sample of 151 countries from 1964 to 2004. We extend the data to 2014 using tariff\ndata from the World Integrated Trade Solution (WITS) and World Development Indicators\n(WDI). Tariff rates at the country-level are computed by aggregating product-level tariff data\nusing weighted averages, with weights given by the import share of each product, measured as\nfractions of value. We do extensive checks of the data to ensure that they do not contain\nspurious jumps, by checking each jump is supported by country and policy reports.\n\nAdvanced economies comprise 28 percent of the sample, emerging markets around 44 percent\nand low-income economies around 28 percent.\n\n**<u>Table S2.1.</u>** <u>Descriptive Statistics for Tariffs</u>\n\n|Col1|Number of<br>observations|Mean|Standard<br>Deviation|Minimum|Maximum|\n|---|---|---|---|---|---|\n|**Tariffs (%)**<br>All<br>AM<br>EM<br>LIC<br>**Annual Tariff Changes (ppt)**<br>All<br>AM<br>EM<br>LIC|5042<br>1427<br>2161<br>1422<br>4702<br>1392<br>1960<br>1320|11.2<br>5.7<br>12.5<br>15.1<br>-\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Together with the trade solution it supplies the extended tariff series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002258:6:1", "occurrence_id": "20a0a986e946ae14abf3027c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |9.3<br>|7%|5.46<br>|0.70<br>|24.00<br>|\n|Cost Coverage (ratio)|48|1.46<br>|1.17<br>|‐20%|0.35<br>|0.40<br>|2.34<br>|\n|Average Revenue (US$/m3 of water sold)|46|0.09<br>|0.10<br>|12%|0.05<br>|0.02<br>|0.21<br>|\n|Unit Cost (US$/m3 of water produced)|47|0.05<br>|0.07<br>|63%|0.04<br>|0.01<br>|0.17<br>|\n|Collection rate (%)|48|68<br>|80<br>|19%|14.59<br>|37.21<br>|121.38<br>|\n|Metering rate (%)|48|28<br>|21<br>|‐25%|33.96<br>|‐<br>|100.00<br>|\n\n\n\nSource: Author calculations based on IBNET data\n\nPourashavas performed poorly when compared with global averages and with the WASAs and city\ncorporations, except on indicators of efficiency.\n\n\n_Efficiency:_ Pourashavas outperformed the WASAs and five city corporations on NRW. NRW in the\npourashavas was 21.9 percent in 2010, which further decreased to 18.4 percent in 2016 compared to an\nNRW of 23 percent in 2016 for\n\nMention: \"IBNET data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Water utility indicators such as metering rates are compared with global averages.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002262:21:2", "occurrence_id": "ab879294c95a306e00d14532", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: /u> <u>0.72</u> <u>0.71</u> <u>0.57</u> <u>0.60</u> <u>0.58</u> <u>0.61</u> <u>0.59</u>\n\nThe dependent variable is in inverse hyperbolic sine. The regressions include country and year\nfixed effects. Robust standard errors in parenthesis, and  - stands for statistical significance\nat the 10% level, ** at the 5% level and *** at the 1% percent level.\n\n\non the right-hand side. While the results using SIPRI data suggest that both oil and gas\n\n\ndiscoveries lead to higher arms imports, the results using COMTRADE data suggest only\n\n\ngas discoveries are associated with larger arms imports. These are not statistically significant\n\n\nresults however. This may be because, while both oil and gas discoveries probably can trigger\n\n\narms imports, identifying their average effects separately requires a larger time period.\n\n\nThe second panel suggests that it might only be offshore discoveries that trigger arms\n\n\nimports. While this result is in line with the Angola and Mozambique anecdotes, it does\n\n\nnot appear to be in line with the evidence from Nordvik (2018) who suggested that while\n\n\nonshore oil motivates military build-ups, offshore oil does not. It also goes against the idea\n\n\nthat arms imports are to be used to protect onshore fields.\n\n\nThe lower panel looks at how the discoveries’ distance to the border affects their impact\n\n\non trade. The idea here is based on Caselli et al. (2015) who suggested that bordering\n\n\ncountries are more likely to engage in conflict when one of the countries has resources near\n\n\n21\n\nMention: \"SIPRI data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Its arms import figures indicate both oil and gas discoveries raise imports.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002262:21:2", "occurrence_id": "0ab52fe3caffcd4c289e6f10", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: /u> <u>0.72</u> <u>0.71</u> <u>0.57</u> <u>0.60</u> <u>0.58</u> <u>0.61</u> <u>0.59</u>\n\nThe dependent variable is in inverse hyperbolic sine. The regressions include country and year\nfixed effects. Robust standard errors in parenthesis, and  - stands for statistical significance\nat the 10% level, ** at the 5% level and *** at the 1% percent level.\n\n\non the right-hand side. While the results using SIPRI data suggest that both oil and gas\n\n\ndiscoveries lead to higher arms imports, the results using COMTRADE data suggest only\n\n\ngas discoveries are associated with larger arms imports. These are not statistically significant\n\n\nresults however. This may be because, while both oil and gas discoveries probably can trigger\n\n\narms imports, identifying their average effects separately requires a larger time period.\n\n\nThe second panel suggests that it might only be offshore discoveries that trigger arms\n\n\nimports. While this result is in line with the Angola and Mozambique anecdotes, it does\n\n\nnot appear to be in line with the evidence from Nordvik (2018) who suggested that while\n\n\nonshore oil motivates military build-ups, offshore oil does not. It also goes against the idea\n\n\nthat arms imports are to be used to protect onshore fields.\n\n\nThe lower panel looks at how the discoveries’ distance to the border affects their impact\n\n\non trade. The idea here is based on Caselli et al. (2015) who suggested that bordering\n\n\ncountries are more likely to engage in conflict when one of the countries has resources near\n\n\n21\n\nMention: \"COMTRADE data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its trade-based measure gives weaker results, indicating only gas discoveries raise imports.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002263:28:1", "occurrence_id": "680d6b9d5a5c612858697e47", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .637)</u>\n_N_ 9870 9870 9870 9870 9870\n_<u>R</u>_ <sup>2</sup> <u>0.892</u> <u>0.869</u> <u>0.803</u> <u>0.778</u> <u>0.557</u>\n\n\nPanel C: The share of labor allocated to coffee\n\n\nLand share of coffee   - LogPrice 0.183*** 0.144* 0.084 0.112 0.083\n<u>(0.069)</u> <u>(0.079)</u> <u>(0.087)</u> <u>(0.083)</u> <u>(0.098)</u>\n_N_ 9870 9870 9870 9870 9870\n_<u>R</u>_ <sup>2</sup> <u>0.818</u> <u>0.803</u> <u>0.679</u> <u>0.657</u> <u>0.529</u>\n\n\nNotes: This table is based on three rounds (2011, 2013, and 2015) of ESS survey. Robust standard errors\nin parenthesis. All regressions include household and zone-year fixed effects, and log rainfall. All dependent\nvariables are measured as inverse hyperbolic sine (IHS) transformation of hours per year to deal with zero values.\nThe share of labor allocated to coffee is defined as the ratio of labor supply on coffee fields to one _plus_ labor\nsupply on all fields, to account for zero values. Male labor supply includes labor supply by all male members\nabove 13 years of\n\nMention: \"ESS survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Three rounds are pooled with household and zone-year fixed effects and rainfall.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002264:7:0", "occurrence_id": "a255b984336b88074195d04d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \"PEDIGREBe1 **of** 0EC **CONVERSION** FACTORS for PER **CAPITA** GNP\n\nCOMPUTATIONS for the **BANf'S** OPERATIONAL **GUIDELXNHS** and ATLAS\n\n\n**I.** Introduction\n\n```\n    A. GNP per Capita and Country Classifications\n```\n\nThe Bank uses a variety of country classifications **for** its\n```\n operational, analytical, and statistical purposes. In the\n```\n\noperational context (known as Operational Guidelines within the\n\nBank), the Bank has established thresholds of per capita **GNP**\n\ncategories, based on the Atlas methodology,' to classify member\n\ncountries into groups eligible for various Bank procurement\n\npractices, and for loans and credits on different terms (e.g.,\ncivil works preferences, IDA credits, IBRD loans, and graduation).\n\nLikewise, the classifications used for analytical and statistical\n\npurposes, which cover both member and non-member economies, are\n\nbest represented in the Bank's publications World Development\n\nReport, World Development Indicators, and The World Bank Atlas. Per\n\ncapita **GNP** estimates are used **by** the Bank as the main criterion for\n\ngrouping countries into low-income, lower-middle income, upper\n```\nmiddle income, and high-income classifications.\n\n```\n\n**1** **A** brief description of **the** **Atlaa** methodology for computing **GNP** per capita\n**is** found **in** the **technical** **notes** **to** **the** **World** **Development** **Indicators** **(Washington,**\n```\nDCs The World Bank); various issues\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Named as one of the publications where comparable national income statistics appear.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002264:7:0", "occurrence_id": "c56fb778268928ea5856d262", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \"PEDIGREBe1 **of** 0EC **CONVERSION** FACTORS for PER **CAPITA** GNP\n\nCOMPUTATIONS for the **BANf'S** OPERATIONAL **GUIDELXNHS** and ATLAS\n\n\n**I.** Introduction\n\n```\n    A. GNP per Capita and Country Classifications\n```\n\nThe Bank uses a variety of country classifications **for** its\n```\n operational, analytical, and statistical purposes. In the\n```\n\noperational context (known as Operational Guidelines within the\n\nBank), the Bank has established thresholds of per capita **GNP**\n\ncategories, based on the Atlas methodology,' to classify member\n\ncountries into groups eligible for various Bank procurement\n\npractices, and for loans and credits on different terms (e.g.,\ncivil works preferences, IDA credits, IBRD loans, and graduation).\n\nLikewise, the classifications used for analytical and statistical\n\npurposes, which cover both member and non-member economies, are\n\nbest represented in the Bank's publications World Development\n\nReport, World Development Indicators, and The World Bank Atlas. Per\n\ncapita **GNP** estimates are used **by** the Bank as the main criterion for\n\ngrouping countries into low-income, lower-middle income, upper\n```\nmiddle income, and high-income classifications.\n\n```\n\n**1** **A** brief description of **the** **Atlaa** methodology for computing **GNP** per capita\n**is** found **in** the **technical** **notes** **to** **the** **World** **Development** **Indicators** **(Washington,**\n```\nDCs The World Bank); various issues\n\nMention: \"World Bank Atlas\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"The third publication listed as carrying these statistics.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002271:8:2", "occurrence_id": "58d6647a1d0bde9e3284eec9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  For the purposes of scoring, providers are assessed on seven actions for\nneonatal asphyxia (calling for help, placing the baby in a neutral position, checking the baby’s heartrate, checking\n\n\n15 Rao & Sheffel (2018).\n16 Escamilla et al. (2018).\n17 Das & Leonard (2006).\n18 Peabody et al. (2000).\n19 Das et al. (2008).\n20 Diarrhea at 7.6% of DALYs, Malaria at 7.9%, Diabetes at 1.3%, lower respiratory infections at 8.7%, tuberculosis\nat 3.3% and anemia at 1.4%, based on IHME (2020).\n\n\n6\n\nMention: \"IHME\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Disease shares of disability-adjusted life years are taken from this institute.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002277:13:1", "occurrence_id": "0a43152164b8fd1e1c3098b2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: s q produces a superior proxy for investment opportunities. They\n\n\nalso show that their adjustments are not sensitive to specific assumptions on the depreciation of\n\n\nintellectual capital. Thus, while these measures are, by construction, approximations, they are\n\n\narguably the best available.\n\n\nHence, as an alternate definition of invested capital, we replace the _INTANit_ in equation (2),\n\n\nwith the new definition of intangible capital from Peters and Taylor [2017], _ICAPit_ .\n\n\n_Invested Capitalit_ = _PPENTit_ + _ACTit_ + _ICAPit −_ _LCTit −_ _GDWLit_\n\n\n_−_ _max_ ( _CHEit −_ 0 _._ 02 _× SALEit,_ 0) (3)\n\n\nwhere _ICAPit_, is defined as the sum of externally purchased intangible capital (Compustat item\n\n\n11\n\nMention: \"Compustat\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Externally purchased intangible capital is taken from a numbered item in this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002278:9:0", "occurrence_id": "fd41fd46347857119d18c8c9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: markets, destroyed infrastructure, and an absence of state-provided electricity and piped water\nuntil 2006 (International Crisis Group, 2003).\n\nThree years after the last peace agreement, only about half of those who had been forcibly\ndisplaced returned home. While many former combatants had entered reintegration programs,\ndemobilized army officers and former security personnel continued to stage protests, some\nviolent, throughout the years (Amnesty International, 2007). By the end of 2006, approximately\n9% of Liberians still resided in other countries such as Guinea, Sierra Leone, Côte d’Ivoire,\nGhana and Nigeria (Amnesty International, 2008).\n\nViolence against women, including rape and IPV, was pervasive in post-conflict Liberia, as\nreported in the DHS conducted five years after the war. In a 2007 survey, over one-third of\nwomen in Liberia reported ever having experienced physical violence by a partner, and 11%\nreported partner sexual violence (Liberia Institute of Statistics, 2008). Just under half of all\nwomen (49%) reported having experienced some kind of violence (physical, sexual or\nemotional) from a partner. Sixty percent of women believed that a husband is justified in beating\nhis wife for at least one reason. Almost half of women surveyed (45%) indicated that they\nexperienced some kind of physical violence from the age of 15, and one-third of women\nexperienced this abuse in the past 12 months. During this time, other incidents of violent crime\nsuch as armed robbery and murder increased as well (Human Rights Watch, 2008).\n\nThe Human Development Index (HDI) provides a rare insight into Liberia’s development pre-,\nduring, and post-conflict. The scale aggregates information about lifespan, education and gross\ndomestic product (GDP) to create a global ranking of countries (Anand & Sen, 1994). In 1970,\nLiberia scored in the lowest quartile of development. The country’s progress\n\nMention: \"DHS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its post-war round reports the prevalence of physical violence against women by a partner.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002278:9:0", "occurrence_id": "104871a0c5456ee3dcf05515", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: markets, destroyed infrastructure, and an absence of state-provided electricity and piped water\nuntil 2006 (International Crisis Group, 2003).\n\nThree years after the last peace agreement, only about half of those who had been forcibly\ndisplaced returned home. While many former combatants had entered reintegration programs,\ndemobilized army officers and former security personnel continued to stage protests, some\nviolent, throughout the years (Amnesty International, 2007). By the end of 2006, approximately\n9% of Liberians still resided in other countries such as Guinea, Sierra Leone, Côte d’Ivoire,\nGhana and Nigeria (Amnesty International, 2008).\n\nViolence against women, including rape and IPV, was pervasive in post-conflict Liberia, as\nreported in the DHS conducted five years after the war. In a 2007 survey, over one-third of\nwomen in Liberia reported ever having experienced physical violence by a partner, and 11%\nreported partner sexual violence (Liberia Institute of Statistics, 2008). Just under half of all\nwomen (49%) reported having experienced some kind of violence (physical, sexual or\nemotional) from a partner. Sixty percent of women believed that a husband is justified in beating\nhis wife for at least one reason. Almost half of women surveyed (45%) indicated that they\nexperienced some kind of physical violence from the age of 15, and one-third of women\nexperienced this abuse in the past 12 months. During this time, other incidents of violent crime\nsuch as armed robbery and murder increased as well (Human Rights Watch, 2008).\n\nThe Human Development Index (HDI) provides a rare insight into Liberia’s development pre-,\nduring, and post-conflict. The scale aggregates information about lifespan, education and gross\ndomestic product (GDP) to create a global ranking of countries (Anand & Sen, 1994). In 1970,\nLiberia scored in the lowest quartile of development. The country’s progress\n\nMention: \"Human Development Index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its aggregate of lifespan, education and living standards gives insight into development across the conflict.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002283:24:1", "occurrence_id": "6bbfa10c38e2d882a90ab9e4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  _κσ,i,s_ ∆ _σs_ ), but as discussed\n\nearlier, the elasticity approach only provides a first-order approximation for headcount\n\npoverty as changes in mean incomes and inequality are discrete.\n\n\nTo project headcount poverty, we need values for mean income and inequality out to\n\n2030. As our baseline, we generate predictions for _y_ ¯ using data on real GDP per-capita\nfrom 1990-2022 from the Macro Poverty Outlook (MPO). <sup>18</sup> In particular, we estimate\nthe following MA(2) regression for each country _i_ : <sup>19</sup>\n\n\nlog( _gdp_ _~~p~~_ _ci,t_ ) = _bi,_ 0 + _bi,_ 1 _yeart_ + _ei,t_ + _θi,_ 1 _ei,t−_ 1 + _θi,_ 2 _ei,t−_ 2 (14)\n\n\nwhere log( _gdp_ _~~p~~_ _ci,t_ ) denotes the log of GDP per-capita in country _i_ in year _t_ . Using\n\nthe coefficients estimated from these regressions, we then predict real GDP per-capita for\n\n2023-2030 for each country. Then, using this GDP per-capita series (the MPO series up\n\nuntil 2022 plus our predicted series from 2023-2030), we calculate country specific growth\n\nrates in GDP per-capita out to 2030. These growth rates, combined with the latest\n\nobservation of mean income in each country from PovcalNet, allow us to predict mean\n\n\n17. For population measures, we use predicted country populations for 2022 generated by\nthe Macro and Poverty Outlook (MPO). See `[https://www.worldbank.org/en/publication/](https:\n\nMention: \"data on real GDP per-capita\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Country MA(2) regressions on this series generate the income path for poverty projections.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002283:24:1", "occurrence_id": "c3fa8af238517b802ca562f0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  _κσ,i,s_ ∆ _σs_ ), but as discussed\n\nearlier, the elasticity approach only provides a first-order approximation for headcount\n\npoverty as changes in mean incomes and inequality are discrete.\n\n\nTo project headcount poverty, we need values for mean income and inequality out to\n\n2030. As our baseline, we generate predictions for _y_ ¯ using data on real GDP per-capita\nfrom 1990-2022 from the Macro Poverty Outlook (MPO). <sup>18</sup> In particular, we estimate\nthe following MA(2) regression for each country _i_ : <sup>19</sup>\n\n\nlog( _gdp_ _~~p~~_ _ci,t_ ) = _bi,_ 0 + _bi,_ 1 _yeart_ + _ei,t_ + _θi,_ 1 _ei,t−_ 1 + _θi,_ 2 _ei,t−_ 2 (14)\n\n\nwhere log( _gdp_ _~~p~~_ _ci,t_ ) denotes the log of GDP per-capita in country _i_ in year _t_ . Using\n\nthe coefficients estimated from these regressions, we then predict real GDP per-capita for\n\n2023-2030 for each country. Then, using this GDP per-capita series (the MPO series up\n\nuntil 2022 plus our predicted series from 2023-2030), we calculate country specific growth\n\nrates in GDP per-capita out to 2030. These growth rates, combined with the latest\n\nobservation of mean income in each country from PovcalNet, allow us to predict mean\n\n\n17. For population measures, we use predicted country populations for 2022 generated by\nthe Macro and Poverty Outlook (MPO). See `[https://www.worldbank.org/en/publication/](https:\n\nMention: \"Macro Poverty Outlook\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It supplies both the historical GDP series and the population projections used.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002283:24:1", "occurrence_id": "d9ebdefa099d1c7e52cd480e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  _κσ,i,s_ ∆ _σs_ ), but as discussed\n\nearlier, the elasticity approach only provides a first-order approximation for headcount\n\npoverty as changes in mean incomes and inequality are discrete.\n\n\nTo project headcount poverty, we need values for mean income and inequality out to\n\n2030. As our baseline, we generate predictions for _y_ ¯ using data on real GDP per-capita\nfrom 1990-2022 from the Macro Poverty Outlook (MPO). <sup>18</sup> In particular, we estimate\nthe following MA(2) regression for each country _i_ : <sup>19</sup>\n\n\nlog( _gdp_ _~~p~~_ _ci,t_ ) = _bi,_ 0 + _bi,_ 1 _yeart_ + _ei,t_ + _θi,_ 1 _ei,t−_ 1 + _θi,_ 2 _ei,t−_ 2 (14)\n\n\nwhere log( _gdp_ _~~p~~_ _ci,t_ ) denotes the log of GDP per-capita in country _i_ in year _t_ . Using\n\nthe coefficients estimated from these regressions, we then predict real GDP per-capita for\n\n2023-2030 for each country. Then, using this GDP per-capita series (the MPO series up\n\nuntil 2022 plus our predicted series from 2023-2030), we calculate country specific growth\n\nrates in GDP per-capita out to 2030. These growth rates, combined with the latest\n\nobservation of mean income in each country from PovcalNet, allow us to predict mean\n\n\n17. For population measures, we use predicted country populations for 2022 generated by\nthe Macro and Poverty Outlook (MPO). See `[https://www.worldbank.org/en/publication/](https:\n\nMention: \"PovcalNet\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The latest observed mean income per country anchors the projected growth rates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002285:12:2", "occurrence_id": "4d078c7a9c9760188095fbfe", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 8 percentage points more likely than men to lose\n\n\ntheir jobs during the pandemic. This result is in line with our previous discussion: the\n\n\n11Our preferred specification includes this adjusted index that captures at the individual level the\ncombined effect of both the characteristics of the occupation and the availability of technology at home\non the ability to do teleworking. We also estimate models including separately the unadjusted WFH\nindex and household internet access. All our main results hold.\n\n12Unfortunately, the HFPS does not have information on hours worked. However, from other sources\n(e.g., SEDLAC) we know that self-employment exhibits a greater dispersion of weekly working hours\nthan salaried work.\n\n13Tables A.1 and A.2 show the results of estimating other more parsimonious specifications that gradually add controls. Models in Table A.1 explore conditional gender differences in the probability of\nexperiencing job losses. The results indicate that the probability of job losses is 7.7 percentage points\nhigher for women than for men with similar characteristics—i.e., same age, education, country, etc.\nModels in Table A.2 add the WFH index as a regressor. Given everything else, workers in jobs fully\ncompatible with remote work (WFH = 1) are 8.5 percentage points less likely to have lost their jobs\nduring the pandemic than workers in jobs incompatible with remote work (WFH = 0).\n\n\n11\n\nMention: \"SEDLAC\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It supplies the evidence that self-employment shows wider dispersion of weekly hours.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002286:5:0", "occurrence_id": "5e603a2ca85f12d0858bef84", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ucation with financial investment in the education production function of children. This\n\n\ncomplementarity may reflect peer and role model effects in addition to more efficient in\n\nvestment by educated parents, and can give rise to a convex CEF. Unlike the standard\n\n\nlinear model, relative mobility in our model depends on parent’s schooling level, and the\n\n\nurban and rural relative mobility curves can cross. The approach developed in this paper\n\n\nis of wider interest as it can be applied to household survey data from other developing\n\n\ncountries.\n\n\nFor the empirical analysis, we take advantage of the rich household panel data from the\n\n\nIndonesia Family Life Survey (IFLS) to make progress on two major issues highlighted in the\n\n\nliterature: (i) truncation bias due to coresident sample, and (ii) omitted variables bias due\n\n\nto ability heterogeneity. An important data constraint on the analysis of intergenerational\n\n\nmobility in developing countries is truncation bias caused by coresidency restrictions in\n\n\ndefining household membership in surveys (see Emran et al. (2018)). The IFLS data do\n\n\nnot suffer from any significant truncation as the survey collected information on nonresident\n\nparents of any household member older than 15. <sup>9</sup> A second important advantage of the IFLS\n\n\n2014 data used in this paper is that good quality measures of cognitive ability are available\n\n\nwhich allows us to control for ability heterogeneity across children in the regressions. Our\n\n\nmain analysis of intergenerational mobility is based on the 18-40 years old children in the\n\n\n2014 wave of IFLS who went to school in the 1990s and 2000s. We use the earlier rounds\n\n\nof the panel data (1993, 1997, 2000, 2007) to understand the effects of parent’s education\n\n\non financial investment in children’s schooling. Another important part of our empirical\n\n\nanalysis is the estimates of returns to education at the household level in rural vs. urban\n\nareas. <sup>10</sup> The availability of 5 rounds of panel\n\nMention: \"Indonesia Family Life Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its rich household panel addresses truncation bias and omitted variables.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002286:5:0", "occurrence_id": "e01e08c3f8e946bfd87d484a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ucation with financial investment in the education production function of children. This\n\n\ncomplementarity may reflect peer and role model effects in addition to more efficient in\n\nvestment by educated parents, and can give rise to a convex CEF. Unlike the standard\n\n\nlinear model, relative mobility in our model depends on parent’s schooling level, and the\n\n\nurban and rural relative mobility curves can cross. The approach developed in this paper\n\n\nis of wider interest as it can be applied to household survey data from other developing\n\n\ncountries.\n\n\nFor the empirical analysis, we take advantage of the rich household panel data from the\n\n\nIndonesia Family Life Survey (IFLS) to make progress on two major issues highlighted in the\n\n\nliterature: (i) truncation bias due to coresident sample, and (ii) omitted variables bias due\n\n\nto ability heterogeneity. An important data constraint on the analysis of intergenerational\n\n\nmobility in developing countries is truncation bias caused by coresidency restrictions in\n\n\ndefining household membership in surveys (see Emran et al. (2018)). The IFLS data do\n\n\nnot suffer from any significant truncation as the survey collected information on nonresident\n\nparents of any household member older than 15. <sup>9</sup> A second important advantage of the IFLS\n\n\n2014 data used in this paper is that good quality measures of cognitive ability are available\n\n\nwhich allows us to control for ability heterogeneity across children in the regressions. Our\n\n\nmain analysis of intergenerational mobility is based on the 18-40 years old children in the\n\n\n2014 wave of IFLS who went to school in the 1990s and 2000s. We use the earlier rounds\n\n\nof the panel data (1993, 1997, 2000, 2007) to understand the effects of parent’s education\n\n\non financial investment in children’s schooling. Another important part of our empirical\n\n\nanalysis is the estimates of returns to education at the household level in rural vs. urban\n\nareas. <sup>10</sup> The availability of 5 rounds of panel\n\nMention: \"IFLS data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"It avoids truncation because non-resident parents of members over fifteen were interviewed.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002286:5:0", "occurrence_id": "2297927344c8f1141b9d1881", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ucation with financial investment in the education production function of children. This\n\n\ncomplementarity may reflect peer and role model effects in addition to more efficient in\n\nvestment by educated parents, and can give rise to a convex CEF. Unlike the standard\n\n\nlinear model, relative mobility in our model depends on parent’s schooling level, and the\n\n\nurban and rural relative mobility curves can cross. The approach developed in this paper\n\n\nis of wider interest as it can be applied to household survey data from other developing\n\n\ncountries.\n\n\nFor the empirical analysis, we take advantage of the rich household panel data from the\n\n\nIndonesia Family Life Survey (IFLS) to make progress on two major issues highlighted in the\n\n\nliterature: (i) truncation bias due to coresident sample, and (ii) omitted variables bias due\n\n\nto ability heterogeneity. An important data constraint on the analysis of intergenerational\n\n\nmobility in developing countries is truncation bias caused by coresidency restrictions in\n\n\ndefining household membership in surveys (see Emran et al. (2018)). The IFLS data do\n\n\nnot suffer from any significant truncation as the survey collected information on nonresident\n\nparents of any household member older than 15. <sup>9</sup> A second important advantage of the IFLS\n\n\n2014 data used in this paper is that good quality measures of cognitive ability are available\n\n\nwhich allows us to control for ability heterogeneity across children in the regressions. Our\n\n\nmain analysis of intergenerational mobility is based on the 18-40 years old children in the\n\n\n2014 wave of IFLS who went to school in the 1990s and 2000s. We use the earlier rounds\n\n\nof the panel data (1993, 1997, 2000, 2007) to understand the effects of parent’s education\n\n\non financial investment in children’s schooling. Another important part of our empirical\n\n\nanalysis is the estimates of returns to education at the household level in rural vs. urban\n\nareas. <sup>10</sup> The availability of 5 rounds of panel\n\nMention: \"IFLS\n\n\n2014 data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Cognitive ability measures in this round allow ability heterogeneity to be controlled.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002286:5:0", "occurrence_id": "0401adcb3e4aafa5c8b780c0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ucation with financial investment in the education production function of children. This\n\n\ncomplementarity may reflect peer and role model effects in addition to more efficient in\n\nvestment by educated parents, and can give rise to a convex CEF. Unlike the standard\n\n\nlinear model, relative mobility in our model depends on parent’s schooling level, and the\n\n\nurban and rural relative mobility curves can cross. The approach developed in this paper\n\n\nis of wider interest as it can be applied to household survey data from other developing\n\n\ncountries.\n\n\nFor the empirical analysis, we take advantage of the rich household panel data from the\n\n\nIndonesia Family Life Survey (IFLS) to make progress on two major issues highlighted in the\n\n\nliterature: (i) truncation bias due to coresident sample, and (ii) omitted variables bias due\n\n\nto ability heterogeneity. An important data constraint on the analysis of intergenerational\n\n\nmobility in developing countries is truncation bias caused by coresidency restrictions in\n\n\ndefining household membership in surveys (see Emran et al. (2018)). The IFLS data do\n\n\nnot suffer from any significant truncation as the survey collected information on nonresident\n\nparents of any household member older than 15. <sup>9</sup> A second important advantage of the IFLS\n\n\n2014 data used in this paper is that good quality measures of cognitive ability are available\n\n\nwhich allows us to control for ability heterogeneity across children in the regressions. Our\n\n\nmain analysis of intergenerational mobility is based on the 18-40 years old children in the\n\n\n2014 wave of IFLS who went to school in the 1990s and 2000s. We use the earlier rounds\n\n\nof the panel data (1993, 1997, 2000, 2007) to understand the effects of parent’s education\n\n\non financial investment in children’s schooling. Another important part of our empirical\n\n\nanalysis is the estimates of returns to education at the household level in rural vs. urban\n\nareas. <sup>10</sup> The availability of 5 rounds of panel\n\nMention: \"2014 wave of IFLS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Adult children who attended school in the 1990s and 2000s form the mobility sample.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002290:14:0", "occurrence_id": "883b7f2d76491fb3cbf8f0ed", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: moderate, <sup>33</sup> suggesting that each indicator adds distinct information to the index. The index\ntakes values between 0 and 1 such that a district with a larger index value is more exposed.\nFigures 1 - 3 depict the spatial dispersion in treatment intensity across districts in India. <sup>34</sup>\n\n\n\n_Indexd_ = <sup>_<u></u>_</sup>\n\n__\n\n\n\n__\n\n\n\n_j_ = __\n\n\n\n_Indicatorj_ (1)\n\n\n\nI compare this conceptualization of shock-exposure to measures of treatment intensity used\nin past studies. The growing body of work that examines the effect of the Arab Spring\nprotests uses different measures of protest intensity such as number of protesters (Acemoglu\net al., 2018), number of martyrs (El-Mallakh et al., 2018) and number of fatalities, injuries\nand arrests (Bargain et al., 2019). I do not use such measures in this study for three key\nreasons. First, unlike the Arab Spring, most protests in India were peaceful. <sup>35</sup> Second, while\nthe details of protests during the Arab Spring have been documented well (see Egyptian\nRevolution Database, Survey of Young People in Egypt, Project Wiki Thawra) there is no\nsystematic documentation or database available in the Indian context. Third, the shock in\nIndia was not only confined to street protests; it also entailed intense media coverage, digital\nactivism, community mobilization and civic engagement. In this light, I try to adopt a more\ncomprehensive measure of treatment intensity, encompassing three distinct pathways of\nexposure, as discussed previously.\n\nRelatedly, few recent studies have used networks on online social media platforms to\nmeasure social connectedness. For instance, Bailey et al. (2018) analyzes patterns of social\nconnectedness between counties in the United States using friendship\n\nMention: \"Egyptian\nRevolution Database\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"Named as one resource documenting Arab Spring protests in detail.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002290:14:0", "occurrence_id": "b904b5d26d4d3f9cb638322a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: moderate, <sup>33</sup> suggesting that each indicator adds distinct information to the index. The index\ntakes values between 0 and 1 such that a district with a larger index value is more exposed.\nFigures 1 - 3 depict the spatial dispersion in treatment intensity across districts in India. <sup>34</sup>\n\n\n\n_Indexd_ = <sup>_<u></u>_</sup>\n\n__\n\n\n\n__\n\n\n\n_j_ = __\n\n\n\n_Indicatorj_ (1)\n\n\n\nI compare this conceptualization of shock-exposure to measures of treatment intensity used\nin past studies. The growing body of work that examines the effect of the Arab Spring\nprotests uses different measures of protest intensity such as number of protesters (Acemoglu\net al., 2018), number of martyrs (El-Mallakh et al., 2018) and number of fatalities, injuries\nand arrests (Bargain et al., 2019). I do not use such measures in this study for three key\nreasons. First, unlike the Arab Spring, most protests in India were peaceful. <sup>35</sup> Second, while\nthe details of protests during the Arab Spring have been documented well (see Egyptian\nRevolution Database, Survey of Young People in Egypt, Project Wiki Thawra) there is no\nsystematic documentation or database available in the Indian context. Third, the shock in\nIndia was not only confined to street protests; it also entailed intense media coverage, digital\nactivism, community mobilization and civic engagement. In this light, I try to adopt a more\ncomprehensive measure of treatment intensity, encompassing three distinct pathways of\nexposure, as discussed previously.\n\nRelatedly, few recent studies have used networks on online social media platforms to\nmeasure social connectedness. For instance, Bailey et al. (2018) analyzes patterns of social\nconnectedness between counties in the United States using friendship\n\nMention: \"Survey of Young People in Egypt\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"Listed alongside other sources that record the Arab Spring.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002290:14:0", "occurrence_id": "d5ef01c48cb70a1449171b05", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: moderate, <sup>33</sup> suggesting that each indicator adds distinct information to the index. The index\ntakes values between 0 and 1 such that a district with a larger index value is more exposed.\nFigures 1 - 3 depict the spatial dispersion in treatment intensity across districts in India. <sup>34</sup>\n\n\n\n_Indexd_ = <sup>_<u></u>_</sup>\n\n__\n\n\n\n__\n\n\n\n_j_ = __\n\n\n\n_Indicatorj_ (1)\n\n\n\nI compare this conceptualization of shock-exposure to measures of treatment intensity used\nin past studies. The growing body of work that examines the effect of the Arab Spring\nprotests uses different measures of protest intensity such as number of protesters (Acemoglu\net al., 2018), number of martyrs (El-Mallakh et al., 2018) and number of fatalities, injuries\nand arrests (Bargain et al., 2019). I do not use such measures in this study for three key\nreasons. First, unlike the Arab Spring, most protests in India were peaceful. <sup>35</sup> Second, while\nthe details of protests during the Arab Spring have been documented well (see Egyptian\nRevolution Database, Survey of Young People in Egypt, Project Wiki Thawra) there is no\nsystematic documentation or database available in the Indian context. Third, the shock in\nIndia was not only confined to street protests; it also entailed intense media coverage, digital\nactivism, community mobilization and civic engagement. In this light, I try to adopt a more\ncomprehensive measure of treatment intensity, encompassing three distinct pathways of\nexposure, as discussed previously.\n\nRelatedly, few recent studies have used networks on online social media platforms to\nmeasure social connectedness. For instance, Bailey et al. (2018) analyzes patterns of social\nconnectedness between counties in the United States using friendship\n\nMention: \"Project Wiki Thawra\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"The third documented source on Arab Spring protests.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002292:9:0", "occurrence_id": "821fa97f6c0efca33b511e01", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: (2015) highlight numerous ways in which the insured took advantage of the NHIS, such as\n\nfrequent and unnecessary visits to health facilities, impersonation, or even feigning sickness\n\nto collect drugs for non-insured persons.\n\n#### **3 Data and Research Design**\n\n###### **3.1 Data**\n\n\nTo analyze the effects of NHIS implementation on risk coping mechanisms with a focus\n\non child labor impacts, we use household data from the fifth round of the Ghana Living\n\nStandards Survey (GLSS), collected from October 2005 to September 2006. The fifth round\n\nof the GLSS interviewed 37,128 individuals in 8,687 households. We define a child as aged 5\n\nto 17 years and limit the sample to those households composed by at least one adult and one\n\nchild (30,060 individuals in 7,982 households), i.e., households with at least one potential\n\nchild laborer.\n\nBecause the timing of implementation of the NHIS is non-random, <sup>7</sup> our empirical strategy\n\nrelies on within district variation as opposed to cross-district variation. Correctly identifying\n\ngeographic information is therefore central to our analysis. The finest within-district level\n\nreported in the survey is the enumeration area (EA), which roughly corresponds to villages\n\n(in rural areas) and neighborhoods (in urban areas). EAs within the same district were\n\nsurveyed at different times, with no systematic order. Importantly, the date of the survey of\n\neach EA is orthogonal to key defining characteristics, such as population, economy, infrastructure, provision of education, and health care facilities. <sup>8</sup> This orthogonality is key for the\n\nidentification of the treatment effect of insurance, as we will explain below.\n\n\nWe link the survey data to information on the exact implementation date of the NHIS\n\nat the district level. We collected these data from several sources,\n\nMention: \"Ghana Living\n\nStandards Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"37,128 individuals in 8,687 households are used to study risk coping and child labour.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002297:10:0", "occurrence_id": "9236b99f992279ae3560abe0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 9 of 21\n\n**Figure 4. Examples of existing medium-resolution remote sensing LCLU products for northern**\n**Nigeria**\n\n\nA\n\n\nB\n\n\nC\n\n\nD\n\n\nE\n\n\n_Note:_ A. False color composition (754) of Landsat 8 image; B. European Space Agency land-cover 20-meter\nmap of Africa; C. GFSAD30AFCE; D. Tsinghua University 30-meter resolution land-cover and land-use (LCLU)\nmap; E. northern Nigeria 30-meter resolution Landsat 8–based LCLU map. Green: natural vegetation; light\nyellow: cropland.\n\nMention: \"European Space Agency land-cover 20-meter\nmap of Africa\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"One of the existing products compared against the derived map.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002297:10:0", "occurrence_id": "51a4733daecdf4189925b436", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 9 of 21\n\n**Figure 4. Examples of existing medium-resolution remote sensing LCLU products for northern**\n**Nigeria**\n\n\nA\n\n\nB\n\n\nC\n\n\nD\n\n\nE\n\n\n_Note:_ A. False color composition (754) of Landsat 8 image; B. European Space Agency land-cover 20-meter\nmap of Africa; C. GFSAD30AFCE; D. Tsinghua University 30-meter resolution land-cover and land-use (LCLU)\nmap; E. northern Nigeria 30-meter resolution Landsat 8–based LCLU map. Green: natural vegetation; light\nyellow: cropland.\n\nMention: \"GFSAD30AFCE\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Another existing land cover product in the comparison panel.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002311:12:1", "occurrence_id": "e3292706d5e212271c1d3d1f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: >𝑋𝑀</mark> 𝑋 **𝑀** <mark>𝑇</mark> 𝑖𝑖,𝑡𝑡𝑇 <mark>= (1 + 𝑔</mark> 𝑋𝑔 <mark>)𝑋𝑀</mark> 𝑋 **𝑀** <mark>𝑇</mark> 𝑖𝑖,𝑡𝑡−1𝑇 <mark>and 𝑋𝑀</mark> 𝑋 **𝑀** <mark>𝑇</mark> 𝑖𝑖,2010𝑇 <mark>= 𝑋</mark> 2010𝑋\n\n\n\n<mark>(11)</mark>\n\n\n\n𝑋 **𝑀** 𝑇 𝑔 𝑋 **𝑀** 𝑇 𝑋 **𝑀** 𝑇 𝑋\n\n<mark>where g</mark> X <mark>is the growth rate of variable X,</mark> _<mark>ω</mark>_ _ij_ <mark>: i the weight of country</mark> _<mark>j</mark>_ <mark>’s share imports in the exports</mark>\n\n<mark>of the domestic economy (with the weights calculated using the average rates from COMTRADE</mark>\n\n<mark>data for 2009-11) and</mark> _<mark>M</mark>_ _tj_ <sup>_KD_</sup> <mark>: is the volume of country</mark> _<mark>j</mark>_ <mark>’s total imports in constant price dollars at</mark>\n\n<mark>time t. We assume that small open economies are price takers. This implies that the price of</mark>\n\n\n<mark>the decision to export or to sell domestically from the point of view of a local producer will depend</mark>\n\n<mark\n\nMention: \"COMTRADE\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Average rates from it provide the weights for import shares.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002316:16:0", "occurrence_id": "832ccf80d2eccdbc0cb31570", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 5: State-level change in electricity and fuel prices: Mexico\n\n\n(a) Growth in electricity prices (%)\n\n\n(b) Growth in fuel prices (%)\n\n\nNote: Electricity prices are weighted by the tariff that correspond to\neach user using data from the EAIM. Fuel prices are calculated as a\nweighted average of the prices of coal, petroleum coke, diesel and\nnatural gas using 3-digits NAICS consumption as weights.\n_Source:_ Authors’ calculations with data from SENER.\n\n\n14\n\nMention: \"data from the EAIM\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Tariff categories from it weight the electricity price growth by user type.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002323:34:0", "occurrence_id": "fa68c847d8c52b1893ad3e09", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: heterogeneity however emerges across countries. While the gravity results suggest that TAs\n\nhave partly raised cross-border government procurement, the CHB values remind us that\n\ngovernment procurement remains vastly home biased, even if the trends show a slow opening\n\nup to trade.\n\n#### **6 Concluding remarks**\n\n\nThis paper estimates trade barriers in government procurement. In doing so, it analyses\n\nthe role of trade agreements to see if and how much they have contributed to reduce those\n\nbarriers.\n\n\nUsing new Inter-Country Input-Output (ICIO) tables from the TiVA database for 62\n\ncountries between 1995 and 2015, we obtain estimates of government purchases across\n\nsectors and countries of origin and use those in a standard gravity model. The estimates\n\nsuggest that governments are significantly more local in their purchasing decisions than\n\nprivate firms. The border effect is large and significantly higher for public markets, especially\n\nin services. Yet, we find that trade agreements and the EU single market in particular have\n\ncontributed to the opening up of government procurement, with the effect being robust in\n\nservices. Non-discriminatory provisions specific to government procurement are driving the\n\ntrade-creating effect of TAs in services, whereas the effects are overall weak in goods.\n\n\nThe estimates from the gravity model are then used to estimate indexes of home bias\n\nin government procurement and private markets. These measures capture the overall\n\ntendency to trade locally rather than internationally, and hence include also factors other\n\nthan protectionist trade policies. Home bias is larger in government procurement than in\n\nprivate markets. Two mechanisms can account for this difference: (i) governments being\n\non average more home biased than firms in purchases of both services and goods; and\n\n(ii) governments spending relatively more on sectors (goods and services) that are more\n\nhome biased. We find that this difference is however shrinking over time: home bias in\n\ngovernment procurement is declining faster than home bias in private markets.\n\n\nOur results have implications for trade negotiations that touch upon government\n\nprocurement issues.\n\nMention: \"TiVA database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its input-output tables for 62 countries yield government purchases by sector and origin.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002328:3:0", "occurrence_id": "c622e25ec725d90af2bfe02d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ##### **1. Introduction**\n\nThe rapid growth of international trade and global value chains in the past decades has been\nassociated with significant reduction in trade costs. But while some developing countries have\nbeen highly successful in integrating into global markets, others have been much less so.\nBorder inefficiency and regulatory uncertainty are often viewed as key barriers for developing\ncountries to boost trade, even when tariff levels are low (WTO 2015). This has motivated\nrenewed energy to improve, simplify, and harmonize the procedures and controls governing\nthe movements of goods across borders –i.e. the trade facilitation agenda.\n\n\nModern Preferential Trade Agreements (PTAs) increasingly cover detailed trade facilitation\nprovisions (Kieck 2020). In this paper, we investigate the effect of these provisions on export\noutcomes using firm level data from Peru. In particular, we focus on the role of trade\nfacilitation measures in simplifying and reducing uncertainty in import processes, as typically\nimport procedures are more cumbersome and administratively heavy. Streamlined import\nprocedures can boost Peruvian exports through two channels. First, enhanced efficiency and\nclarity in import procedures that results from the PTA in the partner country can reduce export\ncosts and uncertainty for Peruvian exporters. Second, increased import efficiency on Peru’s\nside can allow firms involved in Global Value Chains (referred to as GVC-firms) to access\ncritical inputs more quickly and reliably, and therefore enhance their export competitiveness.\n\n\nThe analysis relies on new data on the content of trade agreements (Mattoo, Rocha, and Ruta\n2020) to identify trade facilitation provisions in PTAs and combine this information with firm\nlevel data from Peru from the Exporter Dynamic Database (Fernandes, Freund, and Pierola\n2016). In particular, we exploit the information on the import side of exporting firms (whether\nthey import and what, from where) to investigate if trade facilitation provisions have a\ndifferential impact on GVC-firms through the importing channel. Intuitively,\n\nMention: \"Exporter Dynamic Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Firm-level import behaviour of Peruvian exporters is matched to trade facilitation provisions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002330:4:0", "occurrence_id": "4b619b03f9b622b991dbcae8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: primary channel through which protection has been ratcheted up around the world. We refer to\n\nsuch accords as Intellectual-Property-Related PTAs, or IPAs.\n\n\nThe TRIPS Agreement has been studied empirically in broad terms, with papers asking\n\nwhether its adoption by members has affected such economic outcomes as trade and foreign\n\ndirect investment (FDI). The role of PTAs that feature strong IPRs standards has attracted much\n\nless attention, however. These agreements, which have steadily proliferated since the mid-1990s,\n\nare a potentially significant determinant of trade and investment patterns and innovation\n\nactivities. As such, they deserve systematic study, which we undertake in this paper. Specifically,\n\nwe build on an existing working paper (Maskus and Ridley, 2020), which studied the impact of\n\nmembership in IPAs, where one partner is the US, the EU, or EFTA, on the composition of\n\naggregate and bilateral trade, controlling for compliance with TRIPS standards. Here we delve\n\ndeeper by using specific aspects of such agreements, as measured by the World Bank’s database\n\non the depth of trade agreements (Hofmann, et al, 2017).\n\n\nThe relationship between intellectual property standards and trade is theoretically\n\nambiguous and depends on context. Most important, patents and copyrights are different from\n\nimport barriers. A cut in import tariffs is a reduction in trade costs, implying generally higher\n\ntrade. Lower trade barriers expose domestic firms to competition, typically reducing market\n\npower and markups. Intellectual property rights, however, create temporary monopolies in the\n\nuse, including trade, of technologies and goods. These exclusive rights permit their owners to\n\ndecide where, when, and how they will produce, export, license, and sell protected products and\n\ntechnologies.\n\n\nA small number of studies have estimated the trade effects of TRIPS, using broad\n\nmeasures. Thus, Ivus (2010) found that a group of developing countries (those that were not\n\nBritish or French colonies) that were obligated by membership\n\nMention: \"database\n\non the depth of trade agreements\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Specific aspects of agreements are measured from this compilation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002318:15:1", "occurrence_id": "41a98ca1d061d4c89b3ee705", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .S. output gap is positive, the\n\nglobal economy is expected to expand as well. Then, tradable goods prices rise due to the rising demand in global\n\nmarkets, hence resulting in a sustainable RER depreciation and misalignments in small developing economies, all\n\nother things being equal;\n\n\n- intra-annual volatility of OECD producer price index as a proxy for the volatility of the tradable goods’ prices\n\nin world markets. This proxy is obtained by computing the standard deviation of OECD producer price index\n\nover the 4 quarters of each year between 1980 and 2015. The underlying assumption is that volatility in\n\ninternational markets is an additional cost to small economies, as they impact terms of trade and translate in higher\n\nmisalignments.\n\n\n14\n\nMention: \"OECD producer price index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Its standard deviation proxies volatility of tradable goods prices in world markets.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002321:1:0", "occurrence_id": "290029c59e87789d6a2f5b59", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Policy Research Working Paper 8843\n\n### **Abstract**\n\nThe Sustainable Development Goals recently adopted by\n\nthe United Nations represent an important step to identify shared global goals for development over the next\ntwo decades. Yet, the stated goals are not as straightforward and easy to interpret as they appear on the surface.\nReview of the Sustainable Development Goals indicators\n\nsuggests that some further refinements to their wordings\nand clarifications to their underlying objectives would be\nuseful. This paper brings attention to potential pitfalls with\n\n\n\ninterpretation, where different evaluation methods can lead\nto different conclusions about country performance. The\nreview of the United Nations’ Sustainable Development\nGoals database highlights the overwhelming challenge with\nmissing data: data are available for just over 50 percent of\nall the indicators and for just 19 percent of what is needed\nfor comprehensively tracking progress across countries and\nover time. The paper offers further reflections and proposes\nsome simple but cost-effective solutions to these challenges.\n\n\n\nThis paper is a product of the Development Data Group, Development Economics. It is part of a larger effort by the\nWorld Bank to provide open access to its research and make a contribution to development policy discussions around the\n\nworld. Policy Research Working Papers are also posted on the Web at http://www.worldbank.org/prwp. The authors may\nbe contacted athdang@worldbank.org.\n\n\n_The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development_\n_issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the_\n_names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those_\n_of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and_\n_its affiliated organizations, or those of the Executive Directors of the World Bank or the\n\nMention: \"the United Nations’ Sustainable Development\nGoals database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Its review shows data exist for barely half the indicators and under a fifth of country-indicator pairs.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002331:13:1", "occurrence_id": "5870dda66e56a29b104e28dc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \nlower values of our concentration measures can be interpreted as greater extent of export diversification. The Herfindahl\n\n\nindex is computed as follows:\n\n\n\n\n- \n\n\n𝐻�� ~~�~~ <sup><u>𝑥</u></sup> <u>�</u> <sup><u>�</u></sup>\n\n\n\n\n\n\n\n<u>∑��</u> 𝑥�\n\n\n\n\n<sup><u>𝑥</u></sup> <u>��</u> <sup><u>�</u></sup> 𝑥��\n\n\n\nwhere _x_ represents the amount of exports, the subscript 𝑖 stands for a specific product (or market) and 𝑛 is the total\n\n\nnumber of products (or markets). When a single export product (market) produces all the revenues, _H_ =1, as opposed\n\n\nto _H_ =0 when a large variety of products compose the revenues (Haddad et al. 2010; Lederman and Maloney 2012;\n\n\nCadot et al. 2013). Higher (lower) values of the Herfindahl index, _H_, imply lower (higher) degree of export\n\n\ndiversification. We use COMTRADE data on exports by product at a 4-digit disaggregated level from the SITC Revision\n\n\n1.0 to compute Herfindahl indices of export product concentration. Bilateral data from COMTRADE are also used to\n\n\ncompute Herfindahl indices of export market concentration.\n\n\n_Table 1_ displays the evolution over time of the Herfindahl indices of export product and market concentration\n\n\nin the periods 1990-94 and 2010-14. In the period under analysis, the region’s Herfindahl Index for export product\n\n\nconcentration declined from 0.41 in 1990-94 to 0.3 in 2010-14 —thus, signaling an improvement in terms of trade\n\n\ndiversification\n\nMention: \"COMTRADE data on exports by product\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Four-digit product export values are used to compute concentration indices.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002331:13:1", "occurrence_id": "ea4aa6ba0629981f7f85277f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \nlower values of our concentration measures can be interpreted as greater extent of export diversification. The Herfindahl\n\n\nindex is computed as follows:\n\n\n\n\n- \n\n\n𝐻�� ~~�~~ <sup><u>𝑥</u></sup> <u>�</u> <sup><u>�</u></sup>\n\n\n\n\n\n\n\n<u>∑��</u> 𝑥�\n\n\n\n\n<sup><u>𝑥</u></sup> <u>��</u> <sup><u>�</u></sup> 𝑥��\n\n\n\nwhere _x_ represents the amount of exports, the subscript 𝑖 stands for a specific product (or market) and 𝑛 is the total\n\n\nnumber of products (or markets). When a single export product (market) produces all the revenues, _H_ =1, as opposed\n\n\nto _H_ =0 when a large variety of products compose the revenues (Haddad et al. 2010; Lederman and Maloney 2012;\n\n\nCadot et al. 2013). Higher (lower) values of the Herfindahl index, _H_, imply lower (higher) degree of export\n\n\ndiversification. We use COMTRADE data on exports by product at a 4-digit disaggregated level from the SITC Revision\n\n\n1.0 to compute Herfindahl indices of export product concentration. Bilateral data from COMTRADE are also used to\n\n\ncompute Herfindahl indices of export market concentration.\n\n\n_Table 1_ displays the evolution over time of the Herfindahl indices of export product and market concentration\n\n\nin the periods 1990-94 and 2010-14. In the period under analysis, the region’s Herfindahl Index for export product\n\n\nconcentration declined from 0.41 in 1990-94 to 0.3 in 2010-14 —thus, signaling an improvement in terms of trade\n\n\ndiversification\n\nMention: \"Bilateral data from COMTRADE\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Partner-country flows give the market concentration index.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002335:30:0", "occurrence_id": "e69abdfb5c5f9479b36672ac", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: including WorldPop population data and OSM travel speeds simultaneously. Table 3 shows the\nwelfare gains for each of these robustness checks across both scenarios in each country. In general,\nthe numbers are very similar across specifications. In Appendix Table A.1, we show correlations\nbetween changes in infrastructure in each grid cell in each country under the baseline case and under\neach of the three robustness checks. The correlations are very high, suggesting our results are not\nsensitive to these different sources of data.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n**4.2** **Rural Road Quality**\n\nOne concern with our approach to measuring infrastructure quality is that the data we use may\noverstate the quality of roads in rural areas. <sup>22</sup> [^22: In addition, we do not take into account that weather may render some unpaved rural roads unusable during certain\ntimes of the year; thus, their true average quality may be lower than presented in the data.] We therefore explore the robustness of our results to\nthis measure by reducing travel speeds in very isolated areas, where road quality may be lower than\nmeasured. We focus here on Brazil and Bolivia, which both include large, remote, difficult-to-access\nareas where this problem may be especially acute. We identify which grid cells are “rural” in each\ncountry based on whether population density is below 20 people per square kilometer. <sup>23</sup> [^23: Chomitz et al. (2005) designate this threshold as “ultra-low density” in the context of LAC.] In Brazil, we\nfind that about 16% of the population lives in areas that we identify as ultra-low density; in Bolivia,\nthat figure is 29%. Given this set of rural grid cells within each country, we use the OSM version of\nour results as described in Section 4.1 and reduce speeds by 20% along edges with a rural destination\n_and_ <u>a rural origin, and by 10% if the origin</\n\nMention: \"WorldPop population data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Alternative population grids are substituted to test the welfare results.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002335:30:0", "occurrence_id": "6972333e0dab4f7ea2f929c2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: including WorldPop population data and OSM travel speeds simultaneously. Table 3 shows the\nwelfare gains for each of these robustness checks across both scenarios in each country. In general,\nthe numbers are very similar across specifications. In Appendix Table A.1, we show correlations\nbetween changes in infrastructure in each grid cell in each country under the baseline case and under\neach of the three robustness checks. The correlations are very high, suggesting our results are not\nsensitive to these different sources of data.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n**4.2** **Rural Road Quality**\n\nOne concern with our approach to measuring infrastructure quality is that the data we use may\noverstate the quality of roads in rural areas. <sup>22</sup> [^22: In addition, we do not take into account that weather may render some unpaved rural roads unusable during certain\ntimes of the year; thus, their true average quality may be lower than presented in the data.] We therefore explore the robustness of our results to\nthis measure by reducing travel speeds in very isolated areas, where road quality may be lower than\nmeasured. We focus here on Brazil and Bolivia, which both include large, remote, difficult-to-access\nareas where this problem may be especially acute. We identify which grid cells are “rural” in each\ncountry based on whether population density is below 20 people per square kilometer. <sup>23</sup> [^23: Chomitz et al. (2005) designate this threshold as “ultra-low density” in the context of LAC.] In Brazil, we\nfind that about 16% of the population lives in areas that we identify as ultra-low density; in Bolivia,\nthat figure is 29%. Given this set of rural grid cells within each country, we use the OSM version of\nour results as described in Section 4.1 and reduce speeds by 20% along edges with a rural destination\n_and_ <u>a rural origin, and by 10% if the origin</\n\nMention: \"OSM travel speeds\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Crowd-sourced speeds provide an alternative friction surface.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002337:43:0", "occurrence_id": "6b84fcde266521b40212bb04", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 1: China’s Growing Role in the World and the U.S.\n\n\n\n(a) China’s Share of World Totals\n\n\n\n(b) China’s Share of U.S. Trade\n\n\n\n\n\n\n\n\n\n\n\n_Notes:_ Panel (a) shows China’s share of Non-U.S. World GDP (dashed) and World Exports (solid). Panel\n(b) shows China’s share in U.S. exports (dashed) and imports (solid). Data source is UN Comtrade.\n\n\nFigure 2: China’s Aggregate Import Share and Duties\n\n\n\n\n\n_Notes:_ Tariffs are defined at the SITC level and calculated as the log of 1 + total duties over the total FOB\nvalue of imports. The dashed lines are the median tariff across goods and the 25 <sup>th</sup> and 75 <sup>th</sup> percentiles. The\nsolid line is the log import share of China over the total of U.S. imports from all countries.\n\n\n41\n\nMention: \"UN Comtrade\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"World export shares and China's trade shares are computed from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002339:18:1", "occurrence_id": "993104aba46ab9a6019d68e2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: , the quality of trade and\ntransport infrastructure, the competitiveness and quality of logistics services (trucking, forwarding, and\ncustoms brokerage), and the ability to track and trace consignments, Chad performs relatively poorly, and\nall three countries perform well below regional benchmark peers (Kenya, Uganda and Tanzania). Further\ncompounding Chad’s problems, Cameroon does not much better in its LPI score (Figure 13).\n\nFigure 13: Mali, Niger and Chad’s logistics performance\n\n\n_Source:_ World Bank’s Logistics Performance Index 2017. Note:\nHighest score means best performers. _Note:_ TCD=Chad;\nNER=Niger; MLI=Mali.\n\nWhile Mali and Chad show a mixed performance in terms of trading costs, Chad poorly performs trading\nacross borders compared with peers and faces exorbitant transactions costs in its transit partners (Cameroon,\nNigeria) (Table 4). The cost of being landlocked may be reflected by marked price differences in coastal and\nlandlocked countries. Products in Ndjamena may be 30 percent more expensive than in neighboring\nCameroonian cities. The World Bank Aggregate Trade Cost indicator also shows Chad with the highest costs\namong landlocked countries; while Mali and Niger show a trend decline in trade costs between 2004 and\n2014. <sup>28</sup> However, the cost to export (measured in US$ per container) is higher in Mali than the average in\nWAEMU and SSA countries, implying that the cost of domestic freight transport in Mali is also higher in light\n\n\n15\n\nMention: \"Logistics Performance Index 2017\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its scores rank the landlocked countries' logistics performance.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002339:18:1", "occurrence_id": "be943ceccd90e194e4ccaf34", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: , the quality of trade and\ntransport infrastructure, the competitiveness and quality of logistics services (trucking, forwarding, and\ncustoms brokerage), and the ability to track and trace consignments, Chad performs relatively poorly, and\nall three countries perform well below regional benchmark peers (Kenya, Uganda and Tanzania). Further\ncompounding Chad’s problems, Cameroon does not much better in its LPI score (Figure 13).\n\nFigure 13: Mali, Niger and Chad’s logistics performance\n\n\n_Source:_ World Bank’s Logistics Performance Index 2017. Note:\nHighest score means best performers. _Note:_ TCD=Chad;\nNER=Niger; MLI=Mali.\n\nWhile Mali and Chad show a mixed performance in terms of trading costs, Chad poorly performs trading\nacross borders compared with peers and faces exorbitant transactions costs in its transit partners (Cameroon,\nNigeria) (Table 4). The cost of being landlocked may be reflected by marked price differences in coastal and\nlandlocked countries. Products in Ndjamena may be 30 percent more expensive than in neighboring\nCameroonian cities. The World Bank Aggregate Trade Cost indicator also shows Chad with the highest costs\namong landlocked countries; while Mali and Niger show a trend decline in trade costs between 2004 and\n2014. <sup>28</sup> However, the cost to export (measured in US$ per container) is higher in Mali than the average in\nWAEMU and SSA countries, implying that the cost of domestic freight transport in Mali is also higher in light\n\n\n15\n\nMention: \"World Bank Aggregate Trade Cost indicator\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It shows the country with the highest trade costs among landlocked economies.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002341:12:1", "occurrence_id": "0f6a7d2d0564eaf0739fc3e4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  data, we use and improve upon the templates and concordances developed by Nicita,\n\n\nOlarreaga and Porto (2014). In short, we first aggregate goods in the household surveys to\n\n\n2-digit and 4-digit categories. We then aggregate tariff and trade data from COMTRADE\n\n\nto those categories. See Appendix A for details.\n\n\n10\n\nMention: \"tariff and trade data from COMTRADE\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"They are aggregated to the same categories as the goods in household surveys.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002350:6:0", "occurrence_id": "24dc2774ff2f1b79461879a7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: relatively low in developing countries, and total gasoline consumption is still far behind that in\n\nthe United States (Figure 3), the upward trend in these countries is substantial.\n\n\n**Figure 2: New Passenger Car Registrations for Selected Countries**\n\n\n_Notes: Figure 2 shows the new passenger car registrations_ _and changes from 2005 to 2017 using_\n_[data from the International Organization of Motor Vehicle Manufacturers (www.oica.net). Axis](http://www.oica.net/)_\n_on the right corresponds to China’s new passenger car registrations._\n\n\n**Figure 3: Fuel Consumption by Country**\n\n\n**Panel (a) Gasoline Consumption by Country**\n\n\n5\n\nMention: \"the International Organization of Motor Vehicle Manufacturers\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"New passenger car registrations by country are taken from this association.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002351:43:0", "occurrence_id": "cac771dba6289bde34c0cb20", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure A.1. Drivers of females being out of the labor force (OLF) - Women over age 25**\n(Marginal Effects on the probability of female inactivity)\n\n\n<u>Source: Author’s calculations using HCP Enquête National sur l’emploi, 2001-2018. Controls include: 5 age categories,</u>\nnumber of kids under 10 years old in the household, a 4-categories education variable (omitted category is ”No\neducation”), education level of head of household, a dummy that equals one if women were never married, the number of\ninactive women in the household and its square. Marginal effects evaluated at the mean of the covariates in the estimation\nsample. Sample restricted to women between 15-65 years old.\n\n\n42\n\nMention: \"HCP Enquête National sur l’emploi\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Marginal effects on female inactivity are estimated with age, children and education controls.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002358:20:1", "occurrence_id": "9a46d04610f7221db14fb8f1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ). In a 2015 evaluation of the program, 90\npercent of respondents said it had helped build trust between citizens and authorities (Masud\n2015).\n\n\n**Interoperability**\n\nWhen data sources are interoperable the potential use and value of data expand. As one example,\ninteroperability can help governments prioritize scarce resources by cutting costs such as by\neliminating duplicate or ghost recipients of social transfers—beneficiaries, often of pension funds,\nwho are no longer alive. This was the case in Argentina where the government identified\nnoneligible beneficiaries across various social programs using the country’s system of unique\ntaxpayer ID numbers to link data sets, generating an estimated savings of $143 million over eight\nyears (World Bank 2020a and World Bank 2018b).\n\nInteroperability can also help improve targeting. In Jordan, the Profile Global Registration\nSystem (ProGres), the main database used by the UNHCR to register refugees, has traditionally\nbeen used to target cash assistance to refugees outside camps. This database contains key\nsocioeconomic characteristics on almost all refugees since most register with UNCHR at some\npoint in time. ProGres does not contain information on expenditure of households, though,\npreventing transfers to be targeted explicitly to poor households. Combining ProGres with the\nJordanian Home Visits database, which contains only a subset of the population of interest, but\n\n\n19\n\nMention: \"Profile Global Registration\nSystem\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"It targets cash assistance but lacks household expenditure, so poverty targeting is impossible.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002358:20:1", "occurrence_id": "217ce6a1e78a7d14fc307199", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ). In a 2015 evaluation of the program, 90\npercent of respondents said it had helped build trust between citizens and authorities (Masud\n2015).\n\n\n**Interoperability**\n\nWhen data sources are interoperable the potential use and value of data expand. As one example,\ninteroperability can help governments prioritize scarce resources by cutting costs such as by\neliminating duplicate or ghost recipients of social transfers—beneficiaries, often of pension funds,\nwho are no longer alive. This was the case in Argentina where the government identified\nnoneligible beneficiaries across various social programs using the country’s system of unique\ntaxpayer ID numbers to link data sets, generating an estimated savings of $143 million over eight\nyears (World Bank 2020a and World Bank 2018b).\n\nInteroperability can also help improve targeting. In Jordan, the Profile Global Registration\nSystem (ProGres), the main database used by the UNHCR to register refugees, has traditionally\nbeen used to target cash assistance to refugees outside camps. This database contains key\nsocioeconomic characteristics on almost all refugees since most register with UNCHR at some\npoint in time. ProGres does not contain information on expenditure of households, though,\npreventing transfers to be targeted explicitly to poor households. Combining ProGres with the\nJordanian Home Visits database, which contains only a subset of the population of interest, but\n\n\n19\n\nMention: \"Jordanian Home Visits database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Combining it with the registration system adds welfare information for a subset of the population.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002345:32:0", "occurrence_id": "d76221f52e0e04b978232ede", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Notes: EAP stands for East Asia and Pacific region. ECA stands for Europe and Central Asia region. MENA\nstands for Middle East and North Africa region. LAC stands for Latin America and the Caribbean region.\nSAR stands South Asia region. SSA stands for Sub-Saharan Africa region.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nSource: Global Financial Inclusion as of October 2018\nNotes: GCC countries are Bahrain, Kuwait, Oman, Qatar, Saudi Arabia and UAE. Developing Oil Exporters\ninclude Algeria, Iran, Iraq, Libya, Yemen, Rep. Developing Oil Importers include Djibouti, Egypt, Jordan,\nLebanon, Morocco, Tunisia, West Bank & Gaza. Non-European Mediterranean Countries include Syria,\nCyprus, Lebanon, Israel, West Bank and Gaza, Egypt, Libya, Tunisia, Algeria and Morocco.\n\n\n31\n\nMention: \"Global Financial Inclusion\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Regional account ownership figures are read from this release.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002348:4:0", "occurrence_id": "c34ba70b8531824563f8a958", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: forgoing earning opportunities at home, a combination of mechanisms that has received\n\n\nlittle attention in the literature on terrorism.\n\n\nThis paper exploits a unique data set of Daesh’s personnel records to study how eco\n\nnomic opportunities and migration costs interact to explain the spatial pattern of foreign\n\n\nparticipation in transnational terrorist organizations. The data set contains information\n\n\non 3,965 foreign recruits from 59 countries, including their age and education. Dodwell,\n\n\nMilton and Rassler (2016) estimate that these data account for approximately 30 percent\n\n\nof the total number of foreign recruits who entered Syria between early 2013 and late\n\n\n2014. Our main explanatory variable is the unemployment rate in the countries of origin\n\n\nof these foreign recruits, a first-order measure of economic opportunity costs.\n\n\nThe individual information contained in the Daesh personnel records allows us to\n\n\nmove beyond cross-country correlations and control for any observed and unobserved\n\n\ncountry characteristics that may affect both terrorism participation and labor market op\n\nportunities, such as institutions, government policies, and state capacity (Fearon and\n\n\nLaitin 2003, Sanchez de la Sierra 2019). Specifically, we link the number of Daesh re\n\ncruits from a particular country and education group to the unemployment rate faced by\n\n\nworkers in that same country and with the same education level. We run panel regres\n\nsions that include country- and education-level fixed effects so that identification relies\n\n\non within-country correlations between the schooling gradient of the unemployment rate\n\n\nand the relative number of recruits from each schooling group. Therefore, we contribute\n\n\nplausibly causal estimates of the impact of economic conditions on terrorism participa\n\ntion that are informed by a new data source and a different identification strategy than in\n\n\nthe previous literature. <sup>5</sup>\n\n\nTheoretically, unemployment has an ambiguous effect on foreign terrorist recruitment.\n\n\nOn the one hand, unemployment lowers the economic opportunity cost of participa\n\ntion in terrorist activities and exacerbates grievances against the government (Collier and\n\n\nHoeffler 2004\n\nMention: \"Daesh’s personnel records\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"These unique records explain the spatial pattern of foreign participation in terrorism.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002348:4:0", "occurrence_id": "3925ac205bffb6487659cc3f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: forgoing earning opportunities at home, a combination of mechanisms that has received\n\n\nlittle attention in the literature on terrorism.\n\n\nThis paper exploits a unique data set of Daesh’s personnel records to study how eco\n\nnomic opportunities and migration costs interact to explain the spatial pattern of foreign\n\n\nparticipation in transnational terrorist organizations. The data set contains information\n\n\non 3,965 foreign recruits from 59 countries, including their age and education. Dodwell,\n\n\nMilton and Rassler (2016) estimate that these data account for approximately 30 percent\n\n\nof the total number of foreign recruits who entered Syria between early 2013 and late\n\n\n2014. Our main explanatory variable is the unemployment rate in the countries of origin\n\n\nof these foreign recruits, a first-order measure of economic opportunity costs.\n\n\nThe individual information contained in the Daesh personnel records allows us to\n\n\nmove beyond cross-country correlations and control for any observed and unobserved\n\n\ncountry characteristics that may affect both terrorism participation and labor market op\n\nportunities, such as institutions, government policies, and state capacity (Fearon and\n\n\nLaitin 2003, Sanchez de la Sierra 2019). Specifically, we link the number of Daesh re\n\ncruits from a particular country and education group to the unemployment rate faced by\n\n\nworkers in that same country and with the same education level. We run panel regres\n\nsions that include country- and education-level fixed effects so that identification relies\n\n\non within-country correlations between the schooling gradient of the unemployment rate\n\n\nand the relative number of recruits from each schooling group. Therefore, we contribute\n\n\nplausibly causal estimates of the impact of economic conditions on terrorism participa\n\ntion that are informed by a new data source and a different identification strategy than in\n\n\nthe previous literature. <sup>5</sup>\n\n\nTheoretically, unemployment has an ambiguous effect on foreign terrorist recruitment.\n\n\nOn the one hand, unemployment lowers the economic opportunity cost of participa\n\ntion in terrorist activities and exacerbates grievances against the government (Collier and\n\n\nHoeffler 2004\n\nMention: \"Daesh personnel records\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Individual-level information allows country characteristics to be controlled.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002364:10:1", "occurrence_id": "0aafd26b2815aab306d734d7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  South Asia (0.5–1.1 percent).\nAdditional spending will be required to reach the remaining targets under SDG 6, such as\nwastewater treatment and environmental water quality (SDG 6.3).\n\n  - _Energy (SDG 7)._ The investment required to achieve three targets related to SDG 7\n(universal access to modern energy services, doubled energy efficiency, doubled share of\nrenewable energy) has been estimated to be at least $600 billion to $800 billion per year\nover and above existing levels (Angelou et al. 2013). This would imply a doubling or\ntripling of financing for the sector.\n### • Climate action (SDG 13). The cost to developing economies of adapting to global warming\n\nof 2 <sup>o</sup> C has been estimated at $70 billion to $100 billion per year (Narain, Margulis, and\nEssam 2011). In 2016, an analysis conducted by the International Finance Corporation\n(IFC) found $23 trillion in climate smart investment opportunities (or gaps) in 21 emerging\neconomies by 2030 (IFC 2016). <sup>11</sup> [^11: This analysis was followed by a series of reports examining climate smart investment opportunities in particular\nsectors (IFC 2017a), regions (e.g., South Asia, IFC 2017b), and subsectors (e.g., in cities, IFC 2018).]\n\n\n9 Granular cost estimates are critical for costing exercises. For example, the WBG’s Road Costs Knowledge System\n(ROCKS) records costs related to road-related projects by the WBG, the African Development Bank, and the Asian\nDevelopment Bank. The database provides unit costs for comparable road work activities for SDG9-related costing\nexercises (Collier, Kirchberger, and Söderbom 2016; Bosio et al. 2018).\n10 Existing studies are skewed toward evaluating the cost of infrastructure-related SDGs, for which costs are more\nquantifiable than SDGs that will be achieved chiefly through policy change, such\n\nMention: \"Road Costs Knowledge System\n(ROCKS)\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It records road project unit costs across three development banks for costing exercises.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002365:28:0", "occurrence_id": "06f05aabc565c568e2f67937", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ###### **6. Data availability and gaps for economic impact assessments**\n\nThis section includes a non-exhaustive list of some of the key data sets that are\navailable for use in the different models described in sections 4 and 5.\n\n  - UNCTAD Comrade – includes values and weight of international trade between\ncountries; not mode-specific.\n\n  - OECD Transport cost database – an estimate of the transport costs for flows of\ntrade between countries.\n\n  - EUROSTAT –value, volume, and mode choice for trade flows between European\ncountries and between Europe and the rest of the world.\n\n  - Customs data – both trade value and volume information highly disaggregated to\nindividual shipments. The data are not available for every country.\n\n  - ECLAC trade database– a trade data set that includes trade value, volume, and\nmode choice between Latin American and Caribbean countries and the rest of the\nworld.\n\n  - UNCTAD LSCI database – a database describing the Liner Shipping Connectivity\nIndex for each country worldwide from 2006 to 2017.\n\n  - UNCTAD Container port throughput data – a data set containing the amount of\ncontainer traffic handled by each country worldwide from 2004 to 2018.\nInformation is aggregated at country level; traffic at the port level is not available.\n\n  - MDS Trans modal database on the schedule of liner shipping companies – a data\nset that contains the sequence of port calls and their schedule for the Liner\nShipping Companies worldwide. The data are available for a fee.\n\n\nOne key issue is that many of these data sources are incomplete or not publicly\navailable. This might restrict the full potential that they offer both in terms of geographical\nlimits (e.g. some trade data for specific countries may be missing) and time limits (e.g.\nOECD transport cost data ends in 2007 and has not been updated since). In particular,\nthere is an absence in some of these data sources for information on SIDS and LDCs, who\nmay not have national statistics\n\nMention: \"UNCTAD Comrade\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"Listed as a trade value and weight source that is not mode-specific.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002365:28:0", "occurrence_id": "90d45d6c2bf08ebdd143d512", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ###### **6. Data availability and gaps for economic impact assessments**\n\nThis section includes a non-exhaustive list of some of the key data sets that are\navailable for use in the different models described in sections 4 and 5.\n\n  - UNCTAD Comrade – includes values and weight of international trade between\ncountries; not mode-specific.\n\n  - OECD Transport cost database – an estimate of the transport costs for flows of\ntrade between countries.\n\n  - EUROSTAT –value, volume, and mode choice for trade flows between European\ncountries and between Europe and the rest of the world.\n\n  - Customs data – both trade value and volume information highly disaggregated to\nindividual shipments. The data are not available for every country.\n\n  - ECLAC trade database– a trade data set that includes trade value, volume, and\nmode choice between Latin American and Caribbean countries and the rest of the\nworld.\n\n  - UNCTAD LSCI database – a database describing the Liner Shipping Connectivity\nIndex for each country worldwide from 2006 to 2017.\n\n  - UNCTAD Container port throughput data – a data set containing the amount of\ncontainer traffic handled by each country worldwide from 2004 to 2018.\nInformation is aggregated at country level; traffic at the port level is not available.\n\n  - MDS Trans modal database on the schedule of liner shipping companies – a data\nset that contains the sequence of port calls and their schedule for the Liner\nShipping Companies worldwide. The data are available for a fee.\n\n\nOne key issue is that many of these data sources are incomplete or not publicly\navailable. This might restrict the full potential that they offer both in terms of geographical\nlimits (e.g. some trade data for specific countries may be missing) and time limits (e.g.\nOECD transport cost data ends in 2007 and has not been updated since). In particular,\nthere is an absence in some of these data sources for information on SIDS and LDCs, who\nmay not have national statistics\n\nMention: \"OECD Transport cost database\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"Catalogued as estimated transport costs for bilateral trade flows.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002365:28:0", "occurrence_id": "1c0a8cfe4b4f4a32857c27e5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ###### **6. Data availability and gaps for economic impact assessments**\n\nThis section includes a non-exhaustive list of some of the key data sets that are\navailable for use in the different models described in sections 4 and 5.\n\n  - UNCTAD Comrade – includes values and weight of international trade between\ncountries; not mode-specific.\n\n  - OECD Transport cost database – an estimate of the transport costs for flows of\ntrade between countries.\n\n  - EUROSTAT –value, volume, and mode choice for trade flows between European\ncountries and between Europe and the rest of the world.\n\n  - Customs data – both trade value and volume information highly disaggregated to\nindividual shipments. The data are not available for every country.\n\n  - ECLAC trade database– a trade data set that includes trade value, volume, and\nmode choice between Latin American and Caribbean countries and the rest of the\nworld.\n\n  - UNCTAD LSCI database – a database describing the Liner Shipping Connectivity\nIndex for each country worldwide from 2006 to 2017.\n\n  - UNCTAD Container port throughput data – a data set containing the amount of\ncontainer traffic handled by each country worldwide from 2004 to 2018.\nInformation is aggregated at country level; traffic at the port level is not available.\n\n  - MDS Trans modal database on the schedule of liner shipping companies – a data\nset that contains the sequence of port calls and their schedule for the Liner\nShipping Companies worldwide. The data are available for a fee.\n\n\nOne key issue is that many of these data sources are incomplete or not publicly\navailable. This might restrict the full potential that they offer both in terms of geographical\nlimits (e.g. some trade data for specific countries may be missing) and time limits (e.g.\nOECD transport cost data ends in 2007 and has not been updated since). In particular,\nthere is an absence in some of these data sources for information on SIDS and LDCs, who\nmay not have national statistics\n\nMention: \"EUROSTAT\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"Catalogued as covering value, volume and mode choice for European trade.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002365:28:0", "occurrence_id": "9b65775b92125cfc957931e1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ###### **6. Data availability and gaps for economic impact assessments**\n\nThis section includes a non-exhaustive list of some of the key data sets that are\navailable for use in the different models described in sections 4 and 5.\n\n  - UNCTAD Comrade – includes values and weight of international trade between\ncountries; not mode-specific.\n\n  - OECD Transport cost database – an estimate of the transport costs for flows of\ntrade between countries.\n\n  - EUROSTAT –value, volume, and mode choice for trade flows between European\ncountries and between Europe and the rest of the world.\n\n  - Customs data – both trade value and volume information highly disaggregated to\nindividual shipments. The data are not available for every country.\n\n  - ECLAC trade database– a trade data set that includes trade value, volume, and\nmode choice between Latin American and Caribbean countries and the rest of the\nworld.\n\n  - UNCTAD LSCI database – a database describing the Liner Shipping Connectivity\nIndex for each country worldwide from 2006 to 2017.\n\n  - UNCTAD Container port throughput data – a data set containing the amount of\ncontainer traffic handled by each country worldwide from 2004 to 2018.\nInformation is aggregated at country level; traffic at the port level is not available.\n\n  - MDS Trans modal database on the schedule of liner shipping companies – a data\nset that contains the sequence of port calls and their schedule for the Liner\nShipping Companies worldwide. The data are available for a fee.\n\n\nOne key issue is that many of these data sources are incomplete or not publicly\navailable. This might restrict the full potential that they offer both in terms of geographical\nlimits (e.g. some trade data for specific countries may be missing) and time limits (e.g.\nOECD transport cost data ends in 2007 and has not been updated since). In particular,\nthere is an absence in some of these data sources for information on SIDS and LDCs, who\nmay not have national statistics\n\nMention: \"ECLAC trade database\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"Catalogued as covering value, volume and mode choice for the region's trade.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002365:28:0", "occurrence_id": "c6f7b0a8e5158db5eb004e59", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ###### **6. Data availability and gaps for economic impact assessments**\n\nThis section includes a non-exhaustive list of some of the key data sets that are\navailable for use in the different models described in sections 4 and 5.\n\n  - UNCTAD Comrade – includes values and weight of international trade between\ncountries; not mode-specific.\n\n  - OECD Transport cost database – an estimate of the transport costs for flows of\ntrade between countries.\n\n  - EUROSTAT –value, volume, and mode choice for trade flows between European\ncountries and between Europe and the rest of the world.\n\n  - Customs data – both trade value and volume information highly disaggregated to\nindividual shipments. The data are not available for every country.\n\n  - ECLAC trade database– a trade data set that includes trade value, volume, and\nmode choice between Latin American and Caribbean countries and the rest of the\nworld.\n\n  - UNCTAD LSCI database – a database describing the Liner Shipping Connectivity\nIndex for each country worldwide from 2006 to 2017.\n\n  - UNCTAD Container port throughput data – a data set containing the amount of\ncontainer traffic handled by each country worldwide from 2004 to 2018.\nInformation is aggregated at country level; traffic at the port level is not available.\n\n  - MDS Trans modal database on the schedule of liner shipping companies – a data\nset that contains the sequence of port calls and their schedule for the Liner\nShipping Companies worldwide. The data are available for a fee.\n\n\nOne key issue is that many of these data sources are incomplete or not publicly\navailable. This might restrict the full potential that they offer both in terms of geographical\nlimits (e.g. some trade data for specific countries may be missing) and time limits (e.g.\nOECD transport cost data ends in 2007 and has not been updated since). In particular,\nthere is an absence in some of these data sources for information on SIDS and LDCs, who\nmay not have national statistics\n\nMention: \"UNCTAD LSCI database\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"Catalogued as the source of liner shipping connectivity for every country.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002365:28:0", "occurrence_id": "06623c73db24bcb94d0e4435", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ###### **6. Data availability and gaps for economic impact assessments**\n\nThis section includes a non-exhaustive list of some of the key data sets that are\navailable for use in the different models described in sections 4 and 5.\n\n  - UNCTAD Comrade – includes values and weight of international trade between\ncountries; not mode-specific.\n\n  - OECD Transport cost database – an estimate of the transport costs for flows of\ntrade between countries.\n\n  - EUROSTAT –value, volume, and mode choice for trade flows between European\ncountries and between Europe and the rest of the world.\n\n  - Customs data – both trade value and volume information highly disaggregated to\nindividual shipments. The data are not available for every country.\n\n  - ECLAC trade database– a trade data set that includes trade value, volume, and\nmode choice between Latin American and Caribbean countries and the rest of the\nworld.\n\n  - UNCTAD LSCI database – a database describing the Liner Shipping Connectivity\nIndex for each country worldwide from 2006 to 2017.\n\n  - UNCTAD Container port throughput data – a data set containing the amount of\ncontainer traffic handled by each country worldwide from 2004 to 2018.\nInformation is aggregated at country level; traffic at the port level is not available.\n\n  - MDS Trans modal database on the schedule of liner shipping companies – a data\nset that contains the sequence of port calls and their schedule for the Liner\nShipping Companies worldwide. The data are available for a fee.\n\n\nOne key issue is that many of these data sources are incomplete or not publicly\navailable. This might restrict the full potential that they offer both in terms of geographical\nlimits (e.g. some trade data for specific countries may be missing) and time limits (e.g.\nOECD transport cost data ends in 2007 and has not been updated since). In particular,\nthere is an absence in some of these data sources for information on SIDS and LDCs, who\nmay not have national statistics\n\nMention: \"UNCTAD Container port throughput data\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"Catalogued as country-level container traffic, without port detail.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002365:28:0", "occurrence_id": "d7f024d7efdcfd8a81e4dea8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ###### **6. Data availability and gaps for economic impact assessments**\n\nThis section includes a non-exhaustive list of some of the key data sets that are\navailable for use in the different models described in sections 4 and 5.\n\n  - UNCTAD Comrade – includes values and weight of international trade between\ncountries; not mode-specific.\n\n  - OECD Transport cost database – an estimate of the transport costs for flows of\ntrade between countries.\n\n  - EUROSTAT –value, volume, and mode choice for trade flows between European\ncountries and between Europe and the rest of the world.\n\n  - Customs data – both trade value and volume information highly disaggregated to\nindividual shipments. The data are not available for every country.\n\n  - ECLAC trade database– a trade data set that includes trade value, volume, and\nmode choice between Latin American and Caribbean countries and the rest of the\nworld.\n\n  - UNCTAD LSCI database – a database describing the Liner Shipping Connectivity\nIndex for each country worldwide from 2006 to 2017.\n\n  - UNCTAD Container port throughput data – a data set containing the amount of\ncontainer traffic handled by each country worldwide from 2004 to 2018.\nInformation is aggregated at country level; traffic at the port level is not available.\n\n  - MDS Trans modal database on the schedule of liner shipping companies – a data\nset that contains the sequence of port calls and their schedule for the Liner\nShipping Companies worldwide. The data are available for a fee.\n\n\nOne key issue is that many of these data sources are incomplete or not publicly\navailable. This might restrict the full potential that they offer both in terms of geographical\nlimits (e.g. some trade data for specific countries may be missing) and time limits (e.g.\nOECD transport cost data ends in 2007 and has not been updated since). In particular,\nthere is an absence in some of these data sources for information on SIDS and LDCs, who\nmay not have national statistics\n\nMention: \"MDS Trans modal database\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"Catalogued as containing the sequence and schedule of liner port calls.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002367:7:0", "occurrence_id": "b73497526ac1bc1fe25588af", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 2.1. Country level evidence: Which countries experience the most disruptions?\n\nThe World Bank’s Enterprise Surveys enable a descriptive analysis of which countries experience the most\ndisruptions to electricity, water, and transport infrastructure. This provides estimates for up to 139 low\nand middle-income countries, and can be complemented with (partially overlapping) data from the WEF\nfor 96 middle and high-income countries.\n\n\nOverall, the data portraits a consistent picture: there are substantial differences in the quality of\ninfrastructure across countries. The disruption data from the World Bank’s Enterprise Surveys\ndemonstrate this, and the infrastructure quality index by the World Economic Forum (WEF) confirms it.\nThe quality of infrastructure is generally linked closely to the level of economic development, proxied in\nFigure 2 by GDP per capita. For transport, water, electricity, and infrastructure more generally, more\ndeveloped countries tend to have fewer disruptions and higher perceived infrastructure reliability.\nHowever, the data also shows that there exist significant differences in infrastructure quality levels for\ncountries at the same income level. For instance, Bhutan’s (GDP p.c. $2,500) reliability of electricity is\ncomparable to many developing countries, while Nigeria ($2,476) ranks among the lowest in the whole\nsample. Many factors can drive these patterns, and the remainder of this section will discuss and explore\nthese aspects.\n\n\n6\n\nMention: \"Enterprise Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Disruptions to electricity, water and transport are described for up to 139 countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002367:7:0", "occurrence_id": "4c15a3e534a704d6501fac19", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 2.1. Country level evidence: Which countries experience the most disruptions?\n\nThe World Bank’s Enterprise Surveys enable a descriptive analysis of which countries experience the most\ndisruptions to electricity, water, and transport infrastructure. This provides estimates for up to 139 low\nand middle-income countries, and can be complemented with (partially overlapping) data from the WEF\nfor 96 middle and high-income countries.\n\n\nOverall, the data portraits a consistent picture: there are substantial differences in the quality of\ninfrastructure across countries. The disruption data from the World Bank’s Enterprise Surveys\ndemonstrate this, and the infrastructure quality index by the World Economic Forum (WEF) confirms it.\nThe quality of infrastructure is generally linked closely to the level of economic development, proxied in\nFigure 2 by GDP per capita. For transport, water, electricity, and infrastructure more generally, more\ndeveloped countries tend to have fewer disruptions and higher perceived infrastructure reliability.\nHowever, the data also shows that there exist significant differences in infrastructure quality levels for\ncountries at the same income level. For instance, Bhutan’s (GDP p.c. $2,500) reliability of electricity is\ncomparable to many developing countries, while Nigeria ($2,476) ranks among the lowest in the whole\nsample. Many factors can drive these patterns, and the remainder of this section will discuss and explore\nthese aspects.\n\n\n6\n\nMention: \"data from the WEF\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its partially overlapping coverage extends the picture to 96 middle and high-income countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002368:22:0", "occurrence_id": "9002311d95780797580690f4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: and fertility are other major threats to the identi\u001ccation strategy. The con\u001dict intensity variable\n\n\nis de\u001cned at the level of an individual's village of residence at the time of the survey. Systematic\n\n\nsorting by economic and physical status between the stayers at high con\u001dict villages and movers\n\n\nfrom the high con\u001dict to low con\u001dict villages is of concern, which will lead to overestimation of the\n\n\nimpact of con\u001dict. Unfortunately, I do not observe women's village of birth. However, the 2016\n\n\nNDHS asked each individual how long she has been residing at the place where she was surveyed\n\n\nand if not her entire life, the name of the district from which she migrated. Twenty-six percent of\n\n\nthe women in the sample are living in a di\u001berent district. Although the survey did not collect the\n\n\ninformation, marriage is the most likely reason for women's migration, as most women in Nepal\n\n\nmove to their husband's home permanently from their maternal home, which is also likely to be\n\n\ntheir place of birth.\n\n\nI de\u001cne con\u001dict at the level of women's districts of birth (district they moved from) and estimate\n\n\nthe same speci\u001ccation as Equation (7) except with district of birth xed e\u001bects and region of birth\n\n\ntrends and present the results in Table 6. Columns 1 to 3 examine the presences of di\u001berences\n\n\nin migration patterns between the control and treated cohorts. The di\u001berence-in-di\u001berences across\n\n\nthe speci\u001ccations are zero, suggesting no selective migration. Columns 4 to 9 re-estimate the main\n\n\nresults from Table 4. The di\u001berences-in-di\u001berences estimates, although reduced in magnitude, are\n\n\nstatistically signi\u001ccant and in the same direction as in the main speci\u001ccation for age\n\nMention: \"the 2016\n\n\nNDHS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It records duration of residence and origin district, allowing in-migration to be identified.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002369:48:0", "occurrence_id": "4e916a191ed0a6f37d5f4ef2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure A6: Predicted mobility around transit stations in countries not covered by Google trends.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n47\n\nMention: \"Google trends\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Countries it does not cover must be identified separately, bounding the analysis's reach.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002374:19:3", "occurrence_id": "a466096d223b2d1d7ddc5de1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  result of the 2SLS model\n\n\nwithout Addis Ababa fixed e↵ects in column 2 shows that underemployment rates are lower in\n\n\nlarger towns. Estimating the same models separately for private wage workers (column 5) and\n\n\nself-employed workers (column 8) suggests that the impacts are larger among the latter group.\n\n\n**5.4** **Nominal** **Wages**\n\n\nWe then examine the agglomeration e↵ects on labor productivity by regressing the log of nominal\n\n\nwages on the log of city population (Table 8). Caution is required to interpret the results since\n\n\nlabor income is reported only for wage workers in the UEUS/LFS, and thus, this analysis does not\n\n\ninclude self-employed workers. To account for this, we control for workers’ education levels and\n\n\nother characteristics in the models. The baseline OLS models show that the wage elasticity with\n\n\nrespect to the town population is positive and statistically significant (7.0 percent without Addis\n\n\nAbaba fixed e↵ect in column 1 and 6.0 percent with Addis Ababa fixed e↵ect in column 2). The\n\n\n18\n\nMention: \"UEUS/LFS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Labour income is reported only for wage workers, so the self-employed are excluded.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002376:12:0", "occurrence_id": "02286b0cba26c8ab55dbe0bb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: of the social cohesion index (i.e. “willingness to interact with outgroups”), participants were asked to\n\nimagine that they need to become part of a committee to decide what type of amenities should be\n\nprovided in the park that they saw in the VR experience. They were then told that in this committee they\n\nwould be working with one other person, who was always shown to be a member of the opposite sex and\n\nof a randomly assigned ethnic group. The selection of the social cohesion indicators was informed by Naef\n\nand Schupp (2009), Wilson (2017), and Glaeser, et. al. (2000) for the measures of trust; Strangor (2012)\n\non stereotypes and prejudice; and UNICEF’s Compilation of Tools for Measuring Social Cohesion,\n\nResilience, and Peacebuilding on inclusion and tolerance. They were relevant for the study context, where\n\nother data sources had shown that perceptions and trust toward other groups were low—according to\n\nthe 2018 World Values Survey, nearly 70 percent of the Pakistanis mistrust people of another religion and\n\n74 percent believe that most people cannot be trusted.\n\n\nThe variable definitions are provided in Table 1. <sup>14</sup> Because the survey included a wide set of\n\nquestions, four indices were constructed to reduce the dimensionality of the data and the number of\n\n\n\nstatistical tests performed:\n\n\n\n𝐼𝐼𝑖𝑖 = �𝑧𝑧𝑖\n\n𝑚𝑚\n\n**𝑖**\n\n\n\n𝑚𝑚\n\n**𝑖**\n\n\n\n(2)\n\n\n\nwhere 𝐼𝐼𝑖𝑖 refers to the index score for each individual 𝑖𝑖, defined as the sum of the z-scores 𝑧𝑧𝑖 **𝑖** of the\n\nvariable 𝑚𝑚 included in the index. The control mean and standard deviation were used to construct the z\n\n\nwhere 𝐼𝐼𝑖𝑖 refers to the index score for each individual 𝑖𝑖, defined as the sum of the z-scores 𝑧𝑧𝑖 **𝑖** of the\n\nvariable 𝑚𝑚 included in the index. The control mean\n\nMention: \"2018 World Values Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its shares reporting mistrust of other religions and of people generally justify the variable design.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002378:25:0", "occurrence_id": "7799d9bf4007a6b9c6da9697", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: behavioral responses to taxation (e.g., Blundell and MaCurdy, 1999), we assume that\nincome effects are negligible (so that _ξz_ <sup>_u_</sup> <sup>=</sup> <sup>_ξ_</sup> _z_ <sup>_c_</sup> <sup>and</sup> <sup>_ξ_</sup> _h_ <sup>_u_</sup> <sup>=</sup> <sup>_ξ_</sup> _h_ <sup>_c_</sup> <sup>).27</sup>\n\nAfter briefly discussing what the baseline results imply about the drivers of income\n\ninequality, our main analysis concerns performing the inversion from labor supply deci\nsions to primitives under different assumptions on the relevant elasticities, highlighting\n\nhow deviations from the baseline estimates change our inference about the determinants\n\nof income inequality. We show that (1) larger differences between the income and hours\n\nelasticities with respect to the tax rate (i.e., larger effort elasticities) and (2) larger income\n\neffects will both lead us to infer that preference heterogeneity is increasingly important\n\nin driving income differences between rich and poor. We also discuss briefly at the end of\n\nthis section how our findings change if allow for heterogeneity in the elasticity schedules\n\n(roughly in line with the findings of Gruber and Saez, 2002) and how we can account for\n\nlabor market frictions using survey data on actual and preferred hours.\n\n###### **5.1 Data on Incomes and Hours Worked**\n\n\nWe will use data on incomes and hours worked from the American Time Use Survey\n\n(ATUS), which is a\n\nMention: \"American Time Use Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"It supplies incomes and actual and preferred hours worked.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002379:13:1", "occurrence_id": "7fd68059831aa05a4e3b2a3b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \ncontent or leaf area at much higher temporal resolution. This allows to take corrective action in near\n\nreal time <sup>13</sup> and to assess channels for observed effects _ex post_ . Second, by providing an independent\n\nsource of information, they can allay concerns about bias in statistical data that may result from mis\nreporting to statistical agencies (e.g., to avoid taxes); partial coverage such as the omission of non\nregistered or household farms in the case of Ukraine; or other types of measurement error. <sup>14</sup>\n\n\nTo address these issues within the confines of this paper, <sup>15</sup> we complement yield analysis with an\n\naggregate analysis where the means/sums of four key vegetation indices over relevant crops’ entire\n\nvegetation period is used as outcome variables in regression equation (1). These indices, widely used in\n\nthe literature and practical work are the enhanced vegetation index (EVI), the leaf area index (LAI),\n\nfraction of photosynthetically active radiation (FAPAR), and the Land Surface Water Index (LSWI).\n\n\nThe EVI is based on the normalized difference vegetation index (NDVI) that is widely used in the literature\n\nand agronomy but minimizes canopy-soil variations and improves sensitivity in dense vegetations. The\n\n\n13 While several firms provide such services to large farms in Ukraine, they generally use satellite imagery to identify areas at risk and\ncomplement it with drone imagery or manual scouting.\n14 In our case, inability to match all the village councils is a relevant concern in this respect.\n15 While detailed discussion of temporal variation in these indices together with detailed climatic data would be of great interest to\nprovide insights on the mechanisms through which crop rotation and other factors affect yields, it is well beyond the scope of this paper.\n\n\n12\n\nMention: \"Land Surface Water Index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Named among the satellite products whose practical relevance is considered.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002361:19:0", "occurrence_id": "f7bd276b60d1cbebbccf94f5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 7: Sectors with high GVC participation and high FDI inn 2011 and 2017: no obvious**\n\n\n**Note** : FDI unit of measurement is log FDI where FDI is US$ millions. GVC participation is as\nindex between 0 and 1. Agriculture and manufacturing are in darker tone. Services sectors,\nincluding utilities and construction sector generally supplied domestically, are in lighter tone.\n\n**Source** : Asian Development Bank estimates based on Wang, Wei, Yu and Zhu (2017) using Asian\nDevelopment Bank Multi-Reginal Input Output Tables (2011 and 2017) and Financial Times fDi\nMarkets Database\n\n\nFor every country-sector, we will need to derive a metric of productivity based on its position either\n\n\nupstream or downstream, as well as the level of GVC participation. The metrics used to determine the\n\n\nproductivity of sectors in this framework using the MRIOT are the following:\n\n\n**_Domestic value-added (DVA)_** is defined as the sum of the domestic value-added in\ndomestically used final products and in final exports.\n\n\n**_Foreign value-added (FVA)_** is defined as the domestic and foreign value-added in\nintermediate imports.\n\n\n**_Domestic value-added share (DVA_share)_** is the ratio of _DVA_ over total output.\n\nMention: \"Asian\nDevelopment Bank Multi-Reginal Input Output Tables\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its tables locate each country-sector upstream or downstream in global value chains.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002361:19:0", "occurrence_id": "d75f4fbeca3b986abd7aa5ed", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 7: Sectors with high GVC participation and high FDI inn 2011 and 2017: no obvious**\n\n\n**Note** : FDI unit of measurement is log FDI where FDI is US$ millions. GVC participation is as\nindex between 0 and 1. Agriculture and manufacturing are in darker tone. Services sectors,\nincluding utilities and construction sector generally supplied domestically, are in lighter tone.\n\n**Source** : Asian Development Bank estimates based on Wang, Wei, Yu and Zhu (2017) using Asian\nDevelopment Bank Multi-Reginal Input Output Tables (2011 and 2017) and Financial Times fDi\nMarkets Database\n\n\nFor every country-sector, we will need to derive a metric of productivity based on its position either\n\n\nupstream or downstream, as well as the level of GVC participation. The metrics used to determine the\n\n\nproductivity of sectors in this framework using the MRIOT are the following:\n\n\n**_Domestic value-added (DVA)_** is defined as the sum of the domestic value-added in\ndomestically used final products and in final exports.\n\n\n**_Foreign value-added (FVA)_** is defined as the domestic and foreign value-added in\nintermediate imports.\n\n\n**_Domestic value-added share (DVA_share)_** is the ratio of _DVA_ over total output.\n\nMention: \"Financial Times fDi\nMarkets Database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Greenfield investment announcements complement the input-output evidence.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002384:10:0", "occurrence_id": "bdbea644ca43dbdf82b359ce", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Step 3. Tabulate summary statistics of individuals and dummy variables by nationality and state.**\n\n\nThe share and number of foreign workers are computed by main nationality and state, and by month from\nMarch 2017 through February 2018. We also take an annual average distribution.\n\n\n**Step 4. Make adjustments to the remittance statistics and compare the results obtained from**\n**remittance data analysis with the official numbers reported by MOHA** .\n\n\nBased on the remittance analysis, we make additional adjustments to estimate the total number of foreign\nworkers in Malaysia. As earlier in World Bank (2020), we recognize that not all foreign workers use MSPs\nto remit their earnings back home, and thus assume that the remittance data represent 77 percent of the\nforeign worker population, based on the World Bank Greenback 2.0 surveys of foreign workers in Johor,\nwhich finds that about 77 percent of foreign workers use MSPs to transfer money (World Bank, 2017).\nThen, the final estimated foreign worker population is compared with MOHA’s figures on work permit\nholders, using annual averages, to find the estimated number of irregular workers between March 2017\nand February 2018.\n\n\n**Step 5. Robustness check to provide a lower-bound estimate of the number of foreign workers.**\n\n\nAs a robustness check of our initial estimations, we took the share of foreign workers by nationality in\neach state from the MOHA data, and then applied these shares to the total number of foreign workers at\nthe state level we obtained from the remittance data (the figure prior to adjusting using the 77 percent\nremitter proportion in Step 4). After recalibrating the monthly numbers of foreign worker population by\nnationality, we calculated the differences in the number of foreign workers in each state between our\nresults and the MOHA data. When the differences are negative (that is, the number of foreign workers in\nthe remittance data set is smaller than in the MOHA\n\nMention: \"World Bank Greenback 2.0 surveys\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Its finding that 77 percent of foreign workers remit through formal providers sets the scaling factor.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002384:10:0", "occurrence_id": "a0179eb113875a4e3c2dfe19", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Step 3. Tabulate summary statistics of individuals and dummy variables by nationality and state.**\n\n\nThe share and number of foreign workers are computed by main nationality and state, and by month from\nMarch 2017 through February 2018. We also take an annual average distribution.\n\n\n**Step 4. Make adjustments to the remittance statistics and compare the results obtained from**\n**remittance data analysis with the official numbers reported by MOHA** .\n\n\nBased on the remittance analysis, we make additional adjustments to estimate the total number of foreign\nworkers in Malaysia. As earlier in World Bank (2020), we recognize that not all foreign workers use MSPs\nto remit their earnings back home, and thus assume that the remittance data represent 77 percent of the\nforeign worker population, based on the World Bank Greenback 2.0 surveys of foreign workers in Johor,\nwhich finds that about 77 percent of foreign workers use MSPs to transfer money (World Bank, 2017).\nThen, the final estimated foreign worker population is compared with MOHA’s figures on work permit\nholders, using annual averages, to find the estimated number of irregular workers between March 2017\nand February 2018.\n\n\n**Step 5. Robustness check to provide a lower-bound estimate of the number of foreign workers.**\n\n\nAs a robustness check of our initial estimations, we took the share of foreign workers by nationality in\neach state from the MOHA data, and then applied these shares to the total number of foreign workers at\nthe state level we obtained from the remittance data (the figure prior to adjusting using the 77 percent\nremitter proportion in Step 4). After recalibrating the monthly numbers of foreign worker population by\nnationality, we calculated the differences in the number of foreign workers in each state between our\nresults and the MOHA data. When the differences are negative (that is, the number of foreign workers in\nthe remittance data set is smaller than in the MOHA\n\nMention: \"MOHA data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Shares of foreign workers by nationality per state are applied to the remittance totals.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002385:17:2", "occurrence_id": "5b188160eb3a22b343d87637", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \naxis) assessments. The dotted lines in the figure represent test scores that are one standard deviation\n\n\nabove and below the average scores. Each dot in the figure represents a school. The color of the\n\n\ndots represents the poverty rates of the parishes (the smallest administrative territorial\n\n\ndisaggregation) where schools are located. The blue dots represent the schools located in areas\n\n\nwith the highest poverty rates, and the black dots represent the schools located in areas with the\n\nlowest poverty rates. <sup>7</sup>\n\n\nFigure 4 illustrates that schools with the worst-performing teachers also have the worst\n\nperforming students (bottom left panel of the chart). While this relationship does not necessarily\n\n\nimply a causality, it contributes to identifying the worst-performing institutions, presumably those\n\n\n7 Poverty is measured by the (census-based) Unsatisfied Basic Needs Index produced in 2010 by Ecuador’s National\nOffice of Census and Statistics.\n\n\n16\n\nMention: \"Unsatisfied Basic Needs Index\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"This census-based index is the measure of poverty in the analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002386:1:0", "occurrence_id": "a73f670abfc78e7eafca9228", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Policy Research Working Paper 9575\n\n#### **Abstract**\n\nEvaluating service delivery needs in data-poor environments\npresents a particularly difficult problem for policymakers. The places where the need for social services are most\nacute are often the very same places where assessing policy\ninterventions is the most challenging. This paper uses Twitter data to gain insights into service delivery needs in a\ndata-poor environment. Specifically, it examines the development priorities of citizens in the north-western region of\nPakistan between 2007 and 2020, using natural language\nprocessing techniques (NLP) and sentiment analysis of 9.5\n\n\n\nmillion tweets generated by 20,000 unique Twitter users.\nThe analysis reveals that service delivery priorities in this\n\ncontext are centered on access to education, healthcare,\nfood, and clean water. The findings provide baseline data\nfor future on-the-ground research and development initiatives. In addition, the methodology used in this paper\ndemonstrates both current resources and areas in need of\nfuture work in the use of NLP techniques in analyzing social\nmedia data in other contexts.\n\n\n\nThis paper is a product of the Development Data Group, Development Economics and the Social Sustainability and Inclusion\n\nGlobal Practice. It is part of a larger effort by the World Bank to provide open access to its research and make a contribution\nto development policy discussions around the world. Policy Research Working Papers are also posted on the Web at http://\nwww.worldbank.org/prwp. The authors may be contacted at alia_braley@berkeley.edu, sfraiberger@worldbank.org, and\netas@worldbank.org.\n\n\n_The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development_\n_issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the_\n_names of the authors and should be cited accordingly. The findings, interpretations, and\n\nMention: \"Twitter data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Citizens' stated development priorities are inferred where conventional data are absent.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002392:12:0", "occurrence_id": "909681ad92768b0aa7e402d7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: businesses. Therefore, we define the job loss shock as the loss of employment of previously\nsalaried household members. Only 2.5 percent of households experience this kind of job loss.\n\nOn average, 43 percent of households report unusually high food prices. Approximately\n\n\n\nsalaried household members. Only 2.5 percent of households experience this kind of job loss.\n\nOn average, 43 percent of households report unusually high food prices. Approximately\n26 percent of households did so in IHS 2 and IHS 3 and 71 percent in IHS 4, most likely\nreflecting the adverse weather conditions observed in 2016. The correlation between\ndrought and self-reported inflation for food is 12.3 percent. Delays in passing the cost on to\nconsumers and potential changes in importations or transportation costs all matter to this\nweak correlation.\n\nAbout 18 percent of households suffer the death of a member. The deceased was a\n\n\n\nweak correlation.\n\nAbout 18 percent of households suffer the death of a member. The deceased was a\nbreadwinner in around 20 percent of cases. Parts 4 to 7 of Table 1 present household\ncharacteristics. The average family is headed by a male with low education and has half of its\nworking-age members employed in agriculture. Moreover, around 22 percent of households\nreceive food aid. We remove food aid from consumption per capita when constructing our\ndependent variable. However, we do not include it as a control in our regression as it is\nresponsive to weather conditions and would therefore bias our coefficients on weather\nshocks. <sup>6</sup> Instead, we control for receiving aid unrelated to weather conditions (Malawi Social\nAction Fund [MASAF] program and educational aid). Finally, note that 84.2 percent of the\nhouseholds reside in rural areas.\n\n**2.2** **Impacts of Weather and Idiosyncratic Shocks on Consumption**\n\nAs a first step in the analysis of vulnerability to poverty following a household’s exposure to\ncommon and idiosyncratic\n\nMention: \"IHS 2\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Twenty-six percent of households reported unusually high food prices in this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002392:12:0", "occurrence_id": "474240c69c411599b7e5a7ed", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: businesses. Therefore, we define the job loss shock as the loss of employment of previously\nsalaried household members. Only 2.5 percent of households experience this kind of job loss.\n\nOn average, 43 percent of households report unusually high food prices. Approximately\n\n\n\nsalaried household members. Only 2.5 percent of households experience this kind of job loss.\n\nOn average, 43 percent of households report unusually high food prices. Approximately\n26 percent of households did so in IHS 2 and IHS 3 and 71 percent in IHS 4, most likely\nreflecting the adverse weather conditions observed in 2016. The correlation between\ndrought and self-reported inflation for food is 12.3 percent. Delays in passing the cost on to\nconsumers and potential changes in importations or transportation costs all matter to this\nweak correlation.\n\nAbout 18 percent of households suffer the death of a member. The deceased was a\n\n\n\nweak correlation.\n\nAbout 18 percent of households suffer the death of a member. The deceased was a\nbreadwinner in around 20 percent of cases. Parts 4 to 7 of Table 1 present household\ncharacteristics. The average family is headed by a male with low education and has half of its\nworking-age members employed in agriculture. Moreover, around 22 percent of households\nreceive food aid. We remove food aid from consumption per capita when constructing our\ndependent variable. However, we do not include it as a control in our regression as it is\nresponsive to weather conditions and would therefore bias our coefficients on weather\nshocks. <sup>6</sup> Instead, we control for receiving aid unrelated to weather conditions (Malawi Social\nAction Fund [MASAF] program and educational aid). Finally, note that 84.2 percent of the\nhouseholds reside in rural areas.\n\n**2.2** **Impacts of Weather and Idiosyncratic Shocks on Consumption**\n\nAs a first step in the analysis of vulnerability to poverty following a household’s exposure to\ncommon and idiosyncratic\n\nMention: \"IHS 3\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Its reported share of households facing high food prices is the second baseline.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002392:12:0", "occurrence_id": "3d60d5d1bc96241d9aebed1a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: businesses. Therefore, we define the job loss shock as the loss of employment of previously\nsalaried household members. Only 2.5 percent of households experience this kind of job loss.\n\nOn average, 43 percent of households report unusually high food prices. Approximately\n\n\n\nsalaried household members. Only 2.5 percent of households experience this kind of job loss.\n\nOn average, 43 percent of households report unusually high food prices. Approximately\n26 percent of households did so in IHS 2 and IHS 3 and 71 percent in IHS 4, most likely\nreflecting the adverse weather conditions observed in 2016. The correlation between\ndrought and self-reported inflation for food is 12.3 percent. Delays in passing the cost on to\nconsumers and potential changes in importations or transportation costs all matter to this\nweak correlation.\n\nAbout 18 percent of households suffer the death of a member. The deceased was a\n\n\n\nweak correlation.\n\nAbout 18 percent of households suffer the death of a member. The deceased was a\nbreadwinner in around 20 percent of cases. Parts 4 to 7 of Table 1 present household\ncharacteristics. The average family is headed by a male with low education and has half of its\nworking-age members employed in agriculture. Moreover, around 22 percent of households\nreceive food aid. We remove food aid from consumption per capita when constructing our\ndependent variable. However, we do not include it as a control in our regression as it is\nresponsive to weather conditions and would therefore bias our coefficients on weather\nshocks. <sup>6</sup> Instead, we control for receiving aid unrelated to weather conditions (Malawi Social\nAction Fund [MASAF] program and educational aid). Finally, note that 84.2 percent of the\nhouseholds reside in rural areas.\n\n**2.2** **Impacts of Weather and Idiosyncratic Shocks on Consumption**\n\nAs a first step in the analysis of vulnerability to poverty following a household’s exposure to\ncommon and idiosyncratic\n\nMention: \"IHS 4\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Seventy-one percent report unusually high food prices, reflecting adverse weather.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002393:5:1", "occurrence_id": "af4d8d191df0559356ac4c4a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  10 households from each EA using satellite imagery. Later, we added\n28 EAs to the original sample as part of an additional round of data collection. These were selected using a\nsimilar methodology with the objective to include more households from risk prone areas.\n\nTo capture enough households that had experienced floods, flood risk stratums were designed using the\nRamani Huria community flood map. The community maps were created between 2015-2017 and used\ninformation from community members to identify flood prone areas. This generated a “flood prone tag,”\nwhich was used to categorize EAs into “no risk”, “low to medium risk” and “high risk” depending on how\nmuch of the area was covered by the tag. The city was divided into these three categories and this\ncategorization was used to oversample in areas with flood risk. As a result, about 42.5% of households in\nthe final sample are located in areas inside or adjacent to the flood prone tag (see Table 9 in Appendix 1 for\nsample selection by strata). The final sample includes a satisfactory number of households in flood prone\nareas, as well as low-risk areas to be able to draw statistically sound conclusions about the population.\nHowever, it is important to keep in mind that the uncertainty of being affected by a flood is higher in lowrisk areas, which represents a much larger section of the city.\n\nIn November and December 2017, data were collected from 1,058 selected households in Dar es Salaam,\nand in September 2018, 282 households were added to the original sample. Additional households were\nadded to increase the number of data points located in high risk areas in order to improve precision in those\nareas and cover more affected households in the sample. All the selected households were randomly drawn\nwithin each EA using satellite imagery. Surveyors used GPS tags to locate the households and interviewed\nthem using CAPI software on tablets. Participation in the survey was voluntary and not compensated.\n\n2.2\n\nMention: \"the\nRamani Huria community flood map\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Community-identified flood-prone areas define the risk strata for sampling.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002396:8:1", "occurrence_id": "128a08f730648ee4665adb11", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  young people,\ngenerally but not always male, often residing in urban areas. Therefore for the LSMS-ISA surveys, the\nsecond methodology is employed.\n\nTo obtain the attrition adjustment factor, the probability that a sample household was successfully reinterviewed in round 2 is modeled with the linear logistic model at the level of the individual. A binary\nresponse variable is created by coding the response disposition for eligible household members that are\nnot interviewed in round 2 as zero, and household members that are interviewed as one. The eligibility\nrequirements vary between surveys (for example, excluding children under the age of 15 in the Tanzania\nLSMS-ISA calculations). In all surveys, the dead are excluded from the calculations because they cannot\nbe tracked. Members who moved abroad are generally excluded from tracking but remain eligible, as they\nmay return in later rounds.\n\nThese calculations use a logistic response propensity model, including the household and individual\ncharacteristics measured in the previous round as covariates. The details of this model will vary, perhaps\ndramatically in other types of surveys or those implemented in different contexts. General model building\ntechniques should be applied to determine the most appropriate set of covariates for a given dataset. For\nthe purposes of illustration, the following round 1 covariates were among those included in the LSMSISA survey attrition calculations:\n\n\n`o` being a split-off member targeted for tracking\n\n`o` demographic characteristics (age, gender, marital status, etc.)\n\n`o` residence of one or both parents in household\n\n`o` years of education\n\n`o` current school attendance\n\n`o` labor force participation\n\n`o` household consumption\n\n`o` household assets\n\n`o` household size\n\n7\n\nMention: \"LSMS-ISA surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"The second tracking methodology is applied to them when computing attrition adjustments.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002396:8:1", "occurrence_id": "de53c2926cd2311b8f2bffc5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  young people,\ngenerally but not always male, often residing in urban areas. Therefore for the LSMS-ISA surveys, the\nsecond methodology is employed.\n\nTo obtain the attrition adjustment factor, the probability that a sample household was successfully reinterviewed in round 2 is modeled with the linear logistic model at the level of the individual. A binary\nresponse variable is created by coding the response disposition for eligible household members that are\nnot interviewed in round 2 as zero, and household members that are interviewed as one. The eligibility\nrequirements vary between surveys (for example, excluding children under the age of 15 in the Tanzania\nLSMS-ISA calculations). In all surveys, the dead are excluded from the calculations because they cannot\nbe tracked. Members who moved abroad are generally excluded from tracking but remain eligible, as they\nmay return in later rounds.\n\nThese calculations use a logistic response propensity model, including the household and individual\ncharacteristics measured in the previous round as covariates. The details of this model will vary, perhaps\ndramatically in other types of surveys or those implemented in different contexts. General model building\ntechniques should be applied to determine the most appropriate set of covariates for a given dataset. For\nthe purposes of illustration, the following round 1 covariates were among those included in the LSMSISA survey attrition calculations:\n\n\n`o` being a split-off member targeted for tracking\n\n`o` demographic characteristics (age, gender, marital status, etc.)\n\n`o` residence of one or both parents in household\n\n`o` years of education\n\n`o` current school attendance\n\n`o` labor force participation\n\n`o` household consumption\n\n`o` household assets\n\n`o` household size\n\n7\n\nMention: \"LSMSISA survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Round one covariates such as split-off status and demographics enter its attrition calculations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002400:9:1", "occurrence_id": "cef79c44e9d732de224b21b7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Mbps<br>5 Mbps<br>6 Mbps|Available plans<br>3 Mbps<br>4 Mbps<br>5 Mbps<br>6 Mbps|Available plans<br>3 Mbps<br>4 Mbps<br>5 Mbps<br>6 Mbps|Available plans<br>3 Mbps<br>4 Mbps<br>5 Mbps<br>6 Mbps|Available plans<br>3 Mbps<br>4 Mbps<br>5 Mbps<br>6 Mbps|\n|Monthly plan<br>9,000 DJF<br>14,000 DJF<br>18,000 DJF<br>25,000 DJF<br>Yearly plan (15% discount)<br>91,800 DJF<br>142,800 DJF<br>183,600 DJF<br>255,000 DJF<br>|9,000 DJF|14,000 DJF|18,000 DJF|25,000 DJF|\n\n\n\nSource: Djibouti Telecom, <u>http://www.djiboutitelecom.dj/particulier-internet-adsl-particulier.html</u> [accessed November 22, 2018]\nNotes: * Internet for _particuliers_\n\n|capita consumption (in|DJF) by quintile|\n|---|---|\n|**Quintile**|**Average per capita**<br>**consumption (annual)**|\n|1|56,182|\n|2|108,506|\n|3|156,757|\n|4|224,548|\n|5|496,690|\n|National|208,224|\n\n\n\nSource: Calculations based on EDAM4.\nNote: Quintiles based on the distribution of consumption per capita.\n\n\n8 To put this\n\nMention: \"EDAM4\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Per-capita consumption by quintile is computed from this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002402:34:1", "occurrence_id": "4ba8616508464a9fb09d01d1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .687*** -0.341*** -0.783*** -0.748***\n\n\n\n\n[0.029] [0.068] [0.081]\n\n\n\n-1.449*** -1.248***\n\n\n\n\n[0.009] [0.018] [0.012]\n\n\n\n-0.859*** -0.109***\n\n\n\n\n[0.001] [0.010] [0.002]\n\n\n\n-1.232*** -1.236***\n\n\n\n\n[0.004] [0.008] [0.002]\n\n\n\n-0.783*** -0.748***\n\n\n\n\n[0.002] [0.005] [0.012]\n\n\n\n\n[0.002] [0.002]\n\n\n\n<u>Observations</u> <u>21,239</u> <u>6,766</u> <u>4,785</u> <u>5,743</u> <u>3,945</u>\nstandard errors in brackets, 100 bootstrap replications\n\n*** p<0.01, ** p<0.05, * p<0.1\n\nSource: 2012 1-2-3 Survey, DRC\n\n\n33\n\nMention: \"2012 1-2-3 Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Bootstrap estimates with standard errors in brackets are computed from this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002406:16:0", "occurrence_id": "21e8d6ae52eb726682f3ed39", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: are more intricate to treat and need further elaboration and assumptions. All details of the country’s\n\ntaxes and transfers system, and how they are identified and assigned to households using the data\n\nsources described above, are presented in the methodological Annex A.\n\nOne methodological contribution we make to the approach is to treat VAT non-compliance\n\nin a distributionally-sensitive manner. Incidence studies typically treat VAT as if all households\n\npay the statutory rates in their purchases, or as if they contribute an effective VAT rate calculated\n\nfrom national accounts. In the former case, households are assumed to fully comply with VAT in\n\nall their purchases. In the latter, the full compliance assumption is lifted, but all households are\n\nassumed to pay the same rate, regardless of their socio-economic background. The majority of\n\nCEQ studies fit the latter case, treating VAT evasion as uniformly distributed.\n\nIn developing countries, however, where the informal economy is more prevalent, VAT\n\ncompliance is a function of consumer behavior and purchase channel, and this varies across the\n\nhousehold income distribution. Lower income households tend to make a larger share of their\n\npurchases in informal outlets without invoice. Using this observation, we depart from previous\n\nstudies and lift the assumption of uniformly distributed non-compliance.\n\nTo do that, we use data from the IPSOS consumption survey to estimate the share of\n\npurchases made in places that normally do not invoice (like open bazaars) by each socio-economic\n\nsegment of the population. We then apply these shares of no-VAT purchases to adjust the\n\ncalculation of VAT in the purchases declared by each socio-economic segment in the HBS. As\n\nexpected, the share of purchases in places like open bazaars is higher among lower-income\n\nhouseholds than higher-income households. This small tweak in the methodology matters from a\n\ndistributional viewpoint. <sup>10</sup>\n\nAfter calculating taxes and transfers for all households, we aggregate them and compare\n\nthe aggregates to the public administrative records\n\nMention: \"IPSOS consumption survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Purchase shares in places that do not invoice are estimated by socio-economic segment.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002408:34:0", "occurrence_id": "71ca5a7fa55350ba03687844", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Appendix 6: Draft pandemic preparedness and response survey module for SDI\nThis novel module has been developed by pandemic preparedness experts and an earlier version was piloted in Bhutan (in preparation for data\ncollection for a health SDI in the coming year). Going forward, this information will be collected as part of standard health SDI surveys. The SDI\n[Team welcomes feedback on this module (please send comments to Kathryn Andrews kandrews@worldbank.org).](mailto:kandrews@worldbank.org)\n###### **_Bhutan Health Service Delivery Indicators Survey 2020_**\n\n**<u>Emergency Preparedness Module</u>**\n\n**_EMERGENCY PREPAREDNESS AND RESPONSE SERVICES_**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n|Number|Question|Response|Skip|Notes|\n|---|---|---|---|---|\n|1. <br>|Does this facility have a ‘Health Emergency and Disaster<br>Contingency Plan’ and/or ‘Public Health Emergency<br>Contingency Plan’?<br>_(verify)_|Yes=1<br>No=2<br>Don’t know=−96||In Bhutan, if the facility has a plan, it<br>will automatically have all of the gov-<br>required elements, so no need to ask<br>about the components of the plan|\n|2. <br>|Has this facility allocated funds earmarked for emergency?|Yes=1<br>No=2<br>Don’t know=−96||Preparedness|\n|3. <br>|Has this facility identified a surveillance focal point?|Yes=1<br>No=2<br>Don’t know=−96||JEE- Real-time Surveillance\n\nMention: \"health SDI surveys\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"The module will be collected as standard in future rounds of these service delivery surveys.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002409:24:0", "occurrence_id": "a3112da0d35393a23049d056", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Source: ILO Stats\n\n\nFigure 11: Public employment and unemployment rate\n\n\n\n\n\n\n\n\n\nSource: ILO Stats\n\n\n\n23\n\nMention: \"ILO Stats\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Public employment and unemployment series are plotted from this portal.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002412:9:0", "occurrence_id": "9fc9972d624ead2192afc9e0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: What would we expect the data to show from a theoretical perspective? If the\n\n\nnascent rate represents early stage activity, we expect this to be higher than the young\n\n\nentrepreneurship rate. This is because many people that take “some steps” towards\n\n\nstarting a business do not actually succeed. We also expect the young entrepreneurship\n\n\nrate to be larger than the formal rate, since many firms first are initially established under\n\n\nsole proprietorship but incorporated at a later stage. In fact, for the United States, these\n\n\nrates are 8.12%, 4.98% and 2.55% respectively. This does not, however, hold across\n\n\ndeveloped and developing countries.\n\n\nIn fact, it appears that in many countries - developed and developing - the young\n\n\nentrepreneurship rate and the nascent entrepreneurship rate are _less_ than the formal\n\n\nentrepreneurship rate. This is case not only in Hong Kong, but also in Latvia, the\n\n\nNetherlands, Norway, Israel, Iceland, New Zealand, Denmark, Sweden, Belgium, etc. In\n\n\nmany developed countries, therefore, the formal entrepreneurship rate actually exceeds\n\n\nthe young entrepreneurship rate and even the nascent entrepreneurship rate.\n\n\nThere are several possible explanations. In developing countries, a lower\n\n\ncorporate rate might actually represent a shift towards increased formalization of the\n\n\neconomy. Newly registered companies may represent some aspect of formalization,\n\n\nwhere businesses that were not previously LLCs have newly converted their legal status.\n\n\nIt is also important to note the unit of analysis is different in the datasets: GEM measures\n\n\nthe number of individual entrepreneurs, possibly overlooking individuals that are\n\n\ninvolved in multiple new businesses. The WBGES dataset instead measures number of\n\n\nbusinesses, and can capture this dynamic. However, a possible complication also results\n\n\nfrom the WBGES measure: Formal entrepreneurship includes both actual businesses _and_\n\n\nLLC’s that are a legal vehicle for purposes other than starting a new business. For\n\n\ninstance, entrepreneurs might use registrations to achieve other business ends such as\n\nreducing taxes (e.g. shell companies)\n\nMention: \"The WBGES dataset\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"It counts businesses rather than entrepreneurs, capturing people involved in several ventures.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002416:25:1", "occurrence_id": "af81fba5410c2b5f27386a47", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Youth Unemployment in\n\nEthiopia. Public Policy and Administration Research, ISSN, 2224(5731), 2225-0972.\n\n\nFranklin, S. (2016). Location, Search Costs and Youth Unemployment: Experimental Evidence from\n\nTransport Subsidies. The Economic Journal, 128(614).\n\n\nMengistae, T. (1999). Wage rates and job queues: does the public sector overpay in Ethiopia? The World\n\nBank.\n\n\nWieser, C., Abebe, G., & Asfaw, A. (forthcoming). How have firms fared in times of COVID-19 in Addis\n\nAbaba? Evidence from eight rounds of high-frequency phone surveys. Washington, D.C.: World Bank\n\nGroup.\n\n\nWieser, C., Ambel, A.A., Bundervoet, T. & Tsegay, A.H. (2020). Monitoring COVID-19 Impacts on\n\nHouseholds in Ethiopia : Results from a High-Frequency Phone Survey of Households. Washington, D.C.\n\n: World Bank Group.\n\n\n24\n\nMention: \"high-frequency phone surveys\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"A reference entry for a study of firm performance across eight rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002416:25:1", "occurrence_id": "1bc3baa66c2f1664120d0df1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Youth Unemployment in\n\nEthiopia. Public Policy and Administration Research, ISSN, 2224(5731), 2225-0972.\n\n\nFranklin, S. (2016). Location, Search Costs and Youth Unemployment: Experimental Evidence from\n\nTransport Subsidies. The Economic Journal, 128(614).\n\n\nMengistae, T. (1999). Wage rates and job queues: does the public sector overpay in Ethiopia? The World\n\nBank.\n\n\nWieser, C., Abebe, G., & Asfaw, A. (forthcoming). How have firms fared in times of COVID-19 in Addis\n\nAbaba? Evidence from eight rounds of high-frequency phone surveys. Washington, D.C.: World Bank\n\nGroup.\n\n\nWieser, C., Ambel, A.A., Bundervoet, T. & Tsegay, A.H. (2020). Monitoring COVID-19 Impacts on\n\nHouseholds in Ethiopia : Results from a High-Frequency Phone Survey of Households. Washington, D.C.\n\n: World Bank Group.\n\n\n24\n\nMention: \"High-Frequency Phone Survey of Households\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"A reference entry for the report on household pandemic impacts.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002420:67:0", "occurrence_id": "4d3137da18a99ceff26fcbfc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Source_ : Own calculations using WBSCPS, administrative data, and households’ surveys. The administrative data sources are the following: Annual Reports of the Social\nSecurity Institution ( _Relaҁão Anual de Informaҁões Sociais,_ RAIS) for Brazil for 2017, Social Security and Taxes database for 2019-2020 for Ecuador, _Observatorio_\n_Laboral para la Educacion (_ OLE) for Colombia for 2016, and _Ponte en Carrera_ for Peru for 2018. The household surveys are the original (unharmonized) surveys from\nthese sources: Brazil: PNADC ( _Pesquisa Nacional por Amosta de Domicílios Contínua_ ) for 2018; Colombia: GEIH ( _Gran Encuesta Integrada de Hogares_ ) for 2018;\nEcuador: ENEMDU ( _Encuesta Nacional Empleo, Desempleo y Subempleo_ ) for 2018; Peru: ENAHO ( _Encuesta Nacional de Hogares sobre Condiciones de Vida y Pobreza_ )\nfor 2017.\n_Notes_ : This table presents a validation exercise for the outcomes reported in the survey data. Graduation, dropout, time to degree, and employment are expressed as\npercentages. * indicates that wages are expressed in annual USD PPP (2017) for Brazil when using administrative data (column 3). Column 1 presents outcome averages\nas reported by program directors in the WBSCPS; means are weighted by the WBSCPS sampling weights (see definitions of outcomes from survey data in Appendix 1).\nIn column (2), we impute a formal employment rate based on the following survey questions: \"Regarding the graduates of the program in recent years, how many were\nemployed by a firm in the formal sector?\" and \"Regarding the graduates of the program in recent years, how many were self-employed in the\n\nMention: \"WBSCPS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Its outcomes are validated against administrative records and household surveys.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002420:67:0", "occurrence_id": "a07a6670a97657eade3208c8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Source_ : Own calculations using WBSCPS, administrative data, and households’ surveys. The administrative data sources are the following: Annual Reports of the Social\nSecurity Institution ( _Relaҁão Anual de Informaҁões Sociais,_ RAIS) for Brazil for 2017, Social Security and Taxes database for 2019-2020 for Ecuador, _Observatorio_\n_Laboral para la Educacion (_ OLE) for Colombia for 2016, and _Ponte en Carrera_ for Peru for 2018. The household surveys are the original (unharmonized) surveys from\nthese sources: Brazil: PNADC ( _Pesquisa Nacional por Amosta de Domicílios Contínua_ ) for 2018; Colombia: GEIH ( _Gran Encuesta Integrada de Hogares_ ) for 2018;\nEcuador: ENEMDU ( _Encuesta Nacional Empleo, Desempleo y Subempleo_ ) for 2018; Peru: ENAHO ( _Encuesta Nacional de Hogares sobre Condiciones de Vida y Pobreza_ )\nfor 2017.\n_Notes_ : This table presents a validation exercise for the outcomes reported in the survey data. Graduation, dropout, time to degree, and employment are expressed as\npercentages. * indicates that wages are expressed in annual USD PPP (2017) for Brazil when using administrative data (column 3). Column 1 presents outcome averages\nas reported by program directors in the WBSCPS; means are weighted by the WBSCPS sampling weights (see definitions of outcomes from survey data in Appendix 1).\nIn column (2), we impute a formal employment rate based on the following survey questions: \"Regarding the graduates of the program in recent years, how many were\nemployed by a firm in the formal sector?\" and \"Regarding the graduates of the program in recent years, how many were self-employed in the\n\nMention: \"Annual Reports of the Social\nSecurity Institution\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its annual social information report is the Brazilian administrative benchmark.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002420:67:0", "occurrence_id": "dd14dfa661311752f90c23c8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Source_ : Own calculations using WBSCPS, administrative data, and households’ surveys. The administrative data sources are the following: Annual Reports of the Social\nSecurity Institution ( _Relaҁão Anual de Informaҁões Sociais,_ RAIS) for Brazil for 2017, Social Security and Taxes database for 2019-2020 for Ecuador, _Observatorio_\n_Laboral para la Educacion (_ OLE) for Colombia for 2016, and _Ponte en Carrera_ for Peru for 2018. The household surveys are the original (unharmonized) surveys from\nthese sources: Brazil: PNADC ( _Pesquisa Nacional por Amosta de Domicílios Contínua_ ) for 2018; Colombia: GEIH ( _Gran Encuesta Integrada de Hogares_ ) for 2018;\nEcuador: ENEMDU ( _Encuesta Nacional Empleo, Desempleo y Subempleo_ ) for 2018; Peru: ENAHO ( _Encuesta Nacional de Hogares sobre Condiciones de Vida y Pobreza_ )\nfor 2017.\n_Notes_ : This table presents a validation exercise for the outcomes reported in the survey data. Graduation, dropout, time to degree, and employment are expressed as\npercentages. * indicates that wages are expressed in annual USD PPP (2017) for Brazil when using administrative data (column 3). Column 1 presents outcome averages\nas reported by program directors in the WBSCPS; means are weighted by the WBSCPS sampling weights (see definitions of outcomes from survey data in Appendix 1).\nIn column (2), we impute a formal employment rate based on the following survey questions: \"Regarding the graduates of the program in recent years, how many were\nemployed by a firm in the formal sector?\" and \"Regarding the graduates of the program in recent years, how many were self-employed in the\n\nMention: \"Social Security and Taxes database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"It provides the Ecuadorian administrative comparison.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002420:67:0", "occurrence_id": "9f185d19545483ea7feb28ee", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Source_ : Own calculations using WBSCPS, administrative data, and households’ surveys. The administrative data sources are the following: Annual Reports of the Social\nSecurity Institution ( _Relaҁão Anual de Informaҁões Sociais,_ RAIS) for Brazil for 2017, Social Security and Taxes database for 2019-2020 for Ecuador, _Observatorio_\n_Laboral para la Educacion (_ OLE) for Colombia for 2016, and _Ponte en Carrera_ for Peru for 2018. The household surveys are the original (unharmonized) surveys from\nthese sources: Brazil: PNADC ( _Pesquisa Nacional por Amosta de Domicílios Contínua_ ) for 2018; Colombia: GEIH ( _Gran Encuesta Integrada de Hogares_ ) for 2018;\nEcuador: ENEMDU ( _Encuesta Nacional Empleo, Desempleo y Subempleo_ ) for 2018; Peru: ENAHO ( _Encuesta Nacional de Hogares sobre Condiciones de Vida y Pobreza_ )\nfor 2017.\n_Notes_ : This table presents a validation exercise for the outcomes reported in the survey data. Graduation, dropout, time to degree, and employment are expressed as\npercentages. * indicates that wages are expressed in annual USD PPP (2017) for Brazil when using administrative data (column 3). Column 1 presents outcome averages\nas reported by program directors in the WBSCPS; means are weighted by the WBSCPS sampling weights (see definitions of outcomes from survey data in Appendix 1).\nIn column (2), we impute a formal employment rate based on the following survey questions: \"Regarding the graduates of the program in recent years, how many were\nemployed by a firm in the formal sector?\" and \"Regarding the graduates of the program in recent years, how many were self-employed in the\n\nMention: \"ENEMDU\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"The national employment survey is one household source in the validation exercise.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002420:67:0", "occurrence_id": "94b3f4ffc2199b799b80f166", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Source_ : Own calculations using WBSCPS, administrative data, and households’ surveys. The administrative data sources are the following: Annual Reports of the Social\nSecurity Institution ( _Relaҁão Anual de Informaҁões Sociais,_ RAIS) for Brazil for 2017, Social Security and Taxes database for 2019-2020 for Ecuador, _Observatorio_\n_Laboral para la Educacion (_ OLE) for Colombia for 2016, and _Ponte en Carrera_ for Peru for 2018. The household surveys are the original (unharmonized) surveys from\nthese sources: Brazil: PNADC ( _Pesquisa Nacional por Amosta de Domicílios Contínua_ ) for 2018; Colombia: GEIH ( _Gran Encuesta Integrada de Hogares_ ) for 2018;\nEcuador: ENEMDU ( _Encuesta Nacional Empleo, Desempleo y Subempleo_ ) for 2018; Peru: ENAHO ( _Encuesta Nacional de Hogares sobre Condiciones de Vida y Pobreza_ )\nfor 2017.\n_Notes_ : This table presents a validation exercise for the outcomes reported in the survey data. Graduation, dropout, time to degree, and employment are expressed as\npercentages. * indicates that wages are expressed in annual USD PPP (2017) for Brazil when using administrative data (column 3). Column 1 presents outcome averages\nas reported by program directors in the WBSCPS; means are weighted by the WBSCPS sampling weights (see definitions of outcomes from survey data in Appendix 1).\nIn column (2), we impute a formal employment rate based on the following survey questions: \"Regarding the graduates of the program in recent years, how many were\nemployed by a firm in the formal sector?\" and \"Regarding the graduates of the program in recent years, how many were self-employed in the\n\nMention: \"ENAHO\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"The Peruvian household conditions survey supplies the comparison for that country.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002426:58:0", "occurrence_id": "be1ce2457ed19e6b5dafc16b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure A1: Monthly Income by Education, 2018 Household Survey\n\nMention: \"2018 Household Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Monthly income by education level is tabulated from this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002429:1:0", "occurrence_id": "104cb7327b04b35fb9eef69d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Policy Research Working Paper 9240\n\n### **Abstract**\n\nThere is a crisis of demand brewing around the globe\n\nas social distancing becomes the norm to counter the\nCOVID-19 outbreak. So, which parts of the economy\n\nare most in the line of fire? Looking at jobs that can be\ndone at home or that require a high degree of face-to-face\ninteractions with consumers can capture complementary\nbut distinct mechanisms to assess this vulnerability. This\npaper uses data on 900 job titles from the Occupational\nInformation Network (O*NET) database for the United\n\n\n\nStates to demonstrate that there is substantial heterogeneity in vulnerability across industries, income groups, and\ngender. First, industries vary in whether they emphasize\nface-to-face interactions and home-based work and the two\ndo not always go hand-in-hand. Second, occupations that\nare less amenable to home-based work are largely concentrated among the lower wage deciles. Third, a larger share\nof women’s employment is accounted for by occupations\nthat are intensive in face-to-face interactions.\n\n\n\nThis paper is a product of the Finance, Competitiveness and Innovation Global Practice. It is part of a larger effort by the\nWorld Bank to provide open access to its research and make a contribution to development policy discussions around the\n\nworld. Policy Research Working Papers are also posted on the Web at http://www.worldbank.org/prwp. The authors may\nbe contacted at bavdiu@worldbank.org and gnayyar@worldbank.org.\n\n\n_The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development_\n_issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the_\n_names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely\n\nMention: \"Occupational\nInformation Network\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Nine hundred job titles are used to measure task heterogeneity in vulnerability.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002443:20:0", "occurrence_id": "da9915e01ae20d4d2b91fba0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **V.** **How to mitigate the procyclicality bias: Policy options budget composition,**\n**decentralization, and fiscal rules**\n\n\n**_Top-down solution: Fiscal rules_**\n\n\nThis section explores the correlation between the procyclical bias and the existence of fiscal rules and\n\nsome of their features. <sup>13</sup> [^13: We account for numerical fiscal rules. For a literature review on procedural rules, see Alesina and Perotti (1999).] National fiscal rules are generally recognized to be more effective than\n\n\nsupranational fiscal rules (Tapsoba, 2012). Indeed, supranational rules generally suffer from a problem\n\n\nof insufficient enforcement and compliance. Fiscal rules (FRs) are often criticized to amplify the\n\nprocyclical bias, <sup>14</sup> [^14: On procyclical bias, a petition signed by 1,100 economists and 11 Nobel economists in the New York Times claimed that\nattempts to strictly keep the budget balance (in US states) would aggravate recession (see Levinson, 1998).] but Guerguil et al. (2016) suggest that the design of the rules matters. Based on the\n\n\nfiscal rules data set from the Fiscal Affairs Department, we consider several characteristics of the\n\n\nnational FRs, namely (i) if a fiscal council also exists; (ii) the presence of effective monitoring\n\n\nmechanisms; (iii) the presence of effective enforcement mechanisms; (iv) the coverage of general\n\n\ngovernment for the rules; (v) the existence of a written legal basis for the rules (statutory or\n\n\nconstitutional); (vi) the presence of escape clauses; and (vii) the presence of other kind of flexibility \n\ninvestment-friendly rules and cyclically-adjusted rules (see Guerguil et al., 2016). The comparison of\n\n\nthe procyclicality bias between countries with and without these features shows (Table C.1 and Figure\n\n\nC.3) that countries with national FRs have a lower procyclical bias, though still positive. The other\n\n\ncharacteristics that\n\nMention: \"fiscal rules data set\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its characteristics of national fiscal rules are coded for the analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002447:15:0", "occurrence_id": "a620ee4a53075e1f581c3639", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **3.** **Empirical Evaluation of Household WTP for Electricity Access**\n\n\nTo empirically test the model predictions, we use unique survey data from rural\n\n\nareas of three representative Sub‐Saharan African countries with deep pockets of\n\n\nextreme poverty: Burkina Faso, Rwanda, and Senegal. The data were collected\n\n\nbetween 2010 and 2013 for impact evaluations on electricity access interventions. All\n\n\nthree countries are among the world’s poorest countries, classified as “low‐income”\n\n\nby the World Bank, and had very low rates of electricity access, particularly in ultra‐\n\n\npoor rural areas at the time the survey. <sup>6</sup> [^6: According to the World Bank Sustainable Energy for All (SE4ALL) database, rural electricity access in 2013\nwas 0.8% in Burkina-Faso, 7.7% in Rwanda, and 27.5% in Senegal.] The surveyed areas correspond to subcase b)\n\n\nshown in Figure 1 as very few (if any) households in the sample use diesel\n\n\ngeneration. The analysis thus focuses on the determinants of WTP by rural\n\n\nhouseholds for solar generation and grid electricity access.\n\n\n_3.1 Data and Quantification of WTP_\n\n\nIdentical WTP questions were asked in all three countries. Table 2 describes the basic\n\n\ninformation on the collected data sample.\n\n\nIn Burkina Faso, the data come from an impact evaluation of a Solar Home System\n\n\nelectrification intervention in the rural province of Kénédougou supported by the\n\n\nNetherlands’ Ministry of Foreign Affairs (see Bensch et al. 2013). For this purpose, a\n\n\nrandom sample in villages, drawn in 2010, was offered a solar home system (SHS) on\n\n\na fee‐for‐service basis. The WTP survey was only elicited among households that did\n\n\nnot have any electricity source at the time of the survey. Most of the remaining\n\n\nhouseholds had already had a private solar panel bought on local markets. Grid\n\n\nelectricity was not available among the surveyed population.\n\n\nIn Rwanda, the 2011 data are a representative random sample\n\nMention: \"Sustainable Energy for All\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its rural electricity access rates contextualise the low coverage in the surveyed areas.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002448:35:2", "occurrence_id": "beae1f1a174d2d67cf42c661", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: br>01mar2020<br>01apr2020<br>01may2020<br>(g) Actual reduction in bookings vs reduction Google search index: U.S.A|-50<br>-40<br>-30<br>-20<br>-10<br>0<br>% Reduction in Google search index<br>-100<br>-80<br>-60<br>-40<br>-20<br>0<br>% Reduction in actual bookings<br>01mar2020<br>01apr2020<br>01may2020<br>(g) Actual reduction in bookings vs reduction Google search index: U.S.A|\n\n\n\nActual bookings Google serch index\n\n\nNote: The graphs in this figure compare actual reductions in restaurant bookings versus reductions in the Google\nsearch index for “restaurant” for the seven countries where these data are available. The bookings data come from\n[OpenTable, https://www.opentable.com/state-of-industry](https://www.opentable.com/state-of-industry) <u>.</u>\n\nFigure B.1: Comparison of actual reductions in restaurant bookings versus reductions in Google search\nindex for “restaurant”\n\n\n34\n\nMention: \"OpenTable\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Restaurant bookings are compared with search index declines in seven countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002437:10:1", "occurrence_id": "84034ff1d6f103af11696860", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  these names show considerable variation in their self-reported caste-status. For the\npurposes of rigor, in our data we flag a name as “caste-neutral” if at least 15% of respondents report a\ndifferent status than the rest of the group. Using this threshold, of the 22,390,585 households in the SECC\ndata, approximately 23% reported caste-neutral names.\n\n\nIn the courts, the practice is not confined to just petitioners, respondents and advocates alone. Interestingly,\nof the 84 judges who have served in the Patna High Court during our sample period, we find that 35% have\ncaste-neutral last names.\n\n\nAdoption of caste-neutral names is a growing trend in India (Jayaraman, 2006). The records of the British\nIndian army in the year 1912, for example, contain more than 100,000 names, but we find no records of\nany individuals with the name _Sinha_ and very few instances of individuals with the last name _Kumar_\n(N=65). We do however find many instances of the name _Singh_ (N=8,562). We acknowledge of course,\nthat certain names may be overrepresented in colonial administrations, particularly in light of policies that\npreferred specific castes for these positions. But the fact that we see no records of Kumar in the census\ndocuments of that time either is noteworthy. It is hard to argue that migration, and fertility across these\ngroups would be so different over the past 100 years to drive the current name distribution in the population.\n\n\nThe rise of caste-neutral last names is widely seen all over India, but in Bihar the practice is believed to\nhave become widespread in the aftermath of the Bihar movement of the 1970s (Shah, 1977; Shah, 1977a;\nShah, 1977b; Singh, 2008). Further research into the timing of this shift, and a comparison with other parts\nof India would\n\nMention: \"SECC\ndata\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Of over twenty-two million households, roughly twenty-three percent report caste-neutral names.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002437:10:1", "occurrence_id": "6f31a9a60f2e7dd09a69a290", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  these names show considerable variation in their self-reported caste-status. For the\npurposes of rigor, in our data we flag a name as “caste-neutral” if at least 15% of respondents report a\ndifferent status than the rest of the group. Using this threshold, of the 22,390,585 households in the SECC\ndata, approximately 23% reported caste-neutral names.\n\n\nIn the courts, the practice is not confined to just petitioners, respondents and advocates alone. Interestingly,\nof the 84 judges who have served in the Patna High Court during our sample period, we find that 35% have\ncaste-neutral last names.\n\n\nAdoption of caste-neutral names is a growing trend in India (Jayaraman, 2006). The records of the British\nIndian army in the year 1912, for example, contain more than 100,000 names, but we find no records of\nany individuals with the name _Sinha_ and very few instances of individuals with the last name _Kumar_\n(N=65). We do however find many instances of the name _Singh_ (N=8,562). We acknowledge of course,\nthat certain names may be overrepresented in colonial administrations, particularly in light of policies that\npreferred specific castes for these positions. But the fact that we see no records of Kumar in the census\ndocuments of that time either is noteworthy. It is hard to argue that migration, and fertility across these\ngroups would be so different over the past 100 years to drive the current name distribution in the population.\n\n\nThe rise of caste-neutral last names is widely seen all over India, but in Bihar the practice is believed to\nhave become widespread in the aftermath of the Bihar movement of the 1970s (Shah, 1977; Shah, 1977a;\nShah, 1977b; Singh, 2008). Further research into the timing of this shift, and a comparison with other parts\nof India would\n\nMention: \"records of the British\nIndian army\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Over one hundred thousand recorded names show no instances of a particular caste name.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002451:29:0", "occurrence_id": "1ee69988f25f6acf31edf304", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <u>Table 2. Within-country correlates of FLFP in 2018. Robustness checks. OLS estimates</u>\n\n\n\nLatinobarometro 2018\n\n\n\nUrban areas\n\n\n\nWomen 25-55\n\n\n\nWomen with\n\n\n\nkid 4-6\n\n\n\nkid 4-6 barometro 2018 only\n\n<u>(1)</u> <u>(2)</u> <u>(3)</u> <u>(4)</u> <u>(5)</u>\n_Individual & Household characteristics_\nAge 0.034 0.034 0.0362 0.0408 0.0495\n\n[0.000811]*** [0.000809]*** [0.00141]*** [0.00232]*** [0.00104]***\nAge squared -0.000433 -0.000434 -0.000476 -0.000534 -0.000654\n\n[1.20e-05]*** [1.21e-05]*** [2.10e-05]*** [3.68e-05]*** [1.38e-05]***\nPrimary level 0.0227 0.0226 0.0165 0.0207 0.0250\n\n[0.00187]*** [0.00190]*** [0.00206]*** [0.00299]*** [0.00352]***\nSecondary level 0.115 0.115 0.101 0.113 0.105\n\n[0.00703]*** [0.00708]*** [0.00504]*** [0.0137]*** [0.00818]***\n\nMention: \"Latinobarometro 2018\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Within-country correlates of female labour force participation are estimated on this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002451:29:0", "occurrence_id": "f70252f8ae19cd263e4de62f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <u>Table 2. Within-country correlates of FLFP in 2018. Robustness checks. OLS estimates</u>\n\n\n\nLatinobarometro 2018\n\n\n\nUrban areas\n\n\n\nWomen 25-55\n\n\n\nWomen with\n\n\n\nkid 4-6\n\n\n\nkid 4-6 barometro 2018 only\n\n<u>(1)</u> <u>(2)</u> <u>(3)</u> <u>(4)</u> <u>(5)</u>\n_Individual & Household characteristics_\nAge 0.034 0.034 0.0362 0.0408 0.0495\n\n[0.000811]*** [0.000809]*** [0.00141]*** [0.00232]*** [0.00104]***\nAge squared -0.000433 -0.000434 -0.000476 -0.000534 -0.000654\n\n[1.20e-05]*** [1.21e-05]*** [2.10e-05]*** [3.68e-05]*** [1.38e-05]***\nPrimary level 0.0227 0.0226 0.0165 0.0207 0.0250\n\n[0.00187]*** [0.00190]*** [0.00206]*** [0.00299]*** [0.00352]***\nSecondary level 0.115 0.115 0.101 0.113 0.105\n\n[0.00703]*** [0.00708]*** [0.00504]*** [0.0137]*** [0.00818]***\n\nMention: \"barometro 2018\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"A specification restricted to mothers of pre-school children uses this round only.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002454:13:0", "occurrence_id": "896af8e17a9fd06cb8da4cf8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **3.3** **Patterns within countries**\n\nHow are gender gaps shaped by urbanization? Though we lack panel data or repeated crosssections to explore trends over time, we can compare gender gaps between urban and rural areas.\nOwn land has a central role in rural areas, as the main place to live and produce, but is less relevant\nin towns and cities where most jobs are outside of agriculture and housing can be rented. In urban\nareas, housing and residential land play a larger role than agricultural land, but even though city\ndwellers tend to earn more than their rural compatriots, they are often less likely to own a home\n(Pendall et al 2016 for the United States; Eurostat 2015 for European Union countries; Sato et al\n2011 for China). This reflects partly that land is relatively scarce in urban areas, leading to highpriced real estate, but also better access to financial services, as an alternative means to store\nwealth. Because of this, some caution must be exercised in comparing ownership of land and\nhousing property across rural and urban areas.\n\nFigure 5 plots gender gaps in property ownership by country, separately for rural and urban\nareas. Gender gaps in sole ownership are almost always larger in rural than in urban areas. This\npattern holds for the most part, though less pronounced, also for gender gaps in sole and joint\nownership, though there are some exceptions. This, however, does _not_ mean that women are more\nlikely to own property if they live in urban areas. It rather reflects that urban men are much _less_\nlikely to own property than their rural counterparts.\n\nTo further investigate distributional patterns, we estimate gender gaps by quintile, using the\nDHS household-level wealth index. As shown in Figure 6, gender gaps in sole property ownership\nare in most countries larger for the poorest than for the richest quintile, though there are some\ncountries where the opposite pattern holds (Burundi, Jordan,\n\nMention: \"DHS household-level wealth index\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Its quintiles are used to examine how property ownership gaps vary by wealth.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002461:0:0", "occurrence_id": "d03aa00677fc7087ed3dc2d8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: # WPS 4140\n\n## **Corporate Governance and Regulation:** **Can There Be Too Much of a Good Thing?**\n\n∗\n\n**Valentina G. Bruno and Stijn Claessens**\n\n\n**Abstract**\n\nFor a large number of companies from different countries, we analyze how company\ncorporate governance practices and country regulatory regimes interact in terms of\ncompany valuation. We confirm that company corporate governance practices play a\ncrucial role in efficient company functioning and shareholder protection, and\nconsequently positively impact valuation. We find little valuation impact from corporate\ngovernance measures at the country level, and evidence of possible over-regulation.\nCorporate governance appears more valuable for large companies and those that rely\nmore heavily on external financing, consistent with the hypothesis that the main role of\ncorporate governance is to protect external financiers.\n\nKeywords: Corporate governance practices, Regulatory regimes, Company valuation\n\nJEL Classifications: G34\n\nWorld Bank Policy Research Working Paper 4140, March 2007\n\n_The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the_\n_exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if_\n_the presentations are less than fully polished. The papers carry the names of the authors and should be cited_\n_accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the_\n_authors. They do not necessarily represent the view of the World Bank, its Executive Directors, or the countries_\n_they represent. Policy Research Working Papers are available online at http://econ.worldbank.org._\n\n______________________________\n\n\n\n∗\n\n\n\nWe thank the Institutional Shareholder Services (ISS) for providing us with data on companies’\ncorporate governance practices, Vidhi Chhaochharia, Luc La\n\nMention: \"Institutional Shareholder Services (ISS)\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It supplied the corporate governance practice data the paper is built on.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002464:25:2", "occurrence_id": "a11ac10783cae205cdde0a1f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .7</u> <u>7.0</u> <u>4.4</u>\n\n**Notes** : 1/ this is comparable to underwriting expense and operational expense ratios for the Mexican pension annuity\nindustry.\n\n\n\nerwriting expense and operational expense ratios for the Mexican pension annuity\n\n\n\n**Source** : Holzeu Lechner (2006).\n\n\n\nHence, the final elem\n\n\n\nHence, the final element of a company performance is given by the rate of return that can\n\nbe generated from reserves. As previously mentioned, pension annuity companies invest\nreserves in government debt. With increased fiscal discipline, sovereign risk and\ninflation pressure decreased considerably. The next figure reports the evolution of\ninflation and sovereign risk for Mexico. This is measured by the JP Morgan EMBI+; i.e.,\nthe interest rate difference between Mexico and US Treasury Bills. As of November 22\n2005, this was at historical lows of 112 basis points.\n\n\n\ne\n\n\n\nent of a company performance is given by the rate of return that can\nserves. As previously mentioned, pension annuity companies invest\n\n\n\npressure decreased considerably. The next figure reports the evolution of\n\n\n\ninflation and sovereign risk for Mexico. This is measured by the JP Morgan EMBI+; i.e.,\n\ne interest rate difference between Mexico and US Treasury Bills. As of November 22\n005, this was at historical lows of 112 basis points.\n\n\n\n\n- 23\n\nMention: \"the JP Morgan EMBI+\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The interest rate difference against US Treasury bills measures sovereign risk.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002467:24:0", "occurrence_id": "a7024bd88d2c5929dd71f8b7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 5.2 Hazard -Survival function - unemployment by gender no CI.pdf\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\ndoes not support the hypothesis of habituation, when using the GSOEP and the ECHP <sup>17</sup>,\nbut the results are not significant using the BHPS. The same exercise is executed here.\n\n\nUsing pooled data, the interaction term between unemployment and its duration, it\nis found that unemployment duration has a positive and significant effect on well-being,\nwhen one does not control for the other life variables. However, if one controls for life\nvariables, then duration has no effect on well-being. This hints that duration is associated\nwith the life variables mentioned above, but not directly with well-being. The transition\nmatrix shown before provides evidence that staying in unemployment has no different effect\non well-being than staying in employment. As pooled data is not rigourous, panel data\nspecifications are used. Introducing individual fixed effects makes it possible to regress the\nchange in well-being to the change in duration. A fixed effect logit is used to estimate this\neffect of duration on well-being, for the unemployed. The results show that the coefficient\nof duration is not significant. Hence, it can be said that the effect of unemployment on\nwell-being is independent from its duration. These results are shown in the appendix.\nTable A.8 shows the pooled results, and Table A.9 the fixed effects results.\n\n\n17GSOEP is the German Socio-Economic Panel ; ECHP is the European Community Household Panel\n\n\n23\n\nMention: \"GSOEP\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"A cited test of habituation rejects the hypothesis when this panel is used.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002467:24:0", "occurrence_id": "f974398299c11d47055ef72f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 5.2 Hazard -Survival function - unemployment by gender no CI.pdf\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\ndoes not support the hypothesis of habituation, when using the GSOEP and the ECHP <sup>17</sup>,\nbut the results are not significant using the BHPS. The same exercise is executed here.\n\n\nUsing pooled data, the interaction term between unemployment and its duration, it\nis found that unemployment duration has a positive and significant effect on well-being,\nwhen one does not control for the other life variables. However, if one controls for life\nvariables, then duration has no effect on well-being. This hints that duration is associated\nwith the life variables mentioned above, but not directly with well-being. The transition\nmatrix shown before provides evidence that staying in unemployment has no different effect\non well-being than staying in employment. As pooled data is not rigourous, panel data\nspecifications are used. Introducing individual fixed effects makes it possible to regress the\nchange in well-being to the change in duration. A fixed effect logit is used to estimate this\neffect of duration on well-being, for the unemployed. The results show that the coefficient\nof duration is not significant. Hence, it can be said that the effect of unemployment on\nwell-being is independent from its duration. These results are shown in the appendix.\nTable A.8 shows the pooled results, and Table A.9 the fixed effects results.\n\n\n17GSOEP is the German Socio-Economic Panel ; ECHP is the European Community Household Panel\n\n\n23\n\nMention: \"ECHP\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"The second panel in the cited comparative test.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002467:24:0", "occurrence_id": "fc7e90e565d49046f686f163", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 5.2 Hazard -Survival function - unemployment by gender no CI.pdf\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\ndoes not support the hypothesis of habituation, when using the GSOEP and the ECHP <sup>17</sup>,\nbut the results are not significant using the BHPS. The same exercise is executed here.\n\n\nUsing pooled data, the interaction term between unemployment and its duration, it\nis found that unemployment duration has a positive and significant effect on well-being,\nwhen one does not control for the other life variables. However, if one controls for life\nvariables, then duration has no effect on well-being. This hints that duration is associated\nwith the life variables mentioned above, but not directly with well-being. The transition\nmatrix shown before provides evidence that staying in unemployment has no different effect\non well-being than staying in employment. As pooled data is not rigourous, panel data\nspecifications are used. Introducing individual fixed effects makes it possible to regress the\nchange in well-being to the change in duration. A fixed effect logit is used to estimate this\neffect of duration on well-being, for the unemployed. The results show that the coefficient\nof duration is not significant. Hence, it can be said that the effect of unemployment on\nwell-being is independent from its duration. These results are shown in the appendix.\nTable A.8 shows the pooled results, and Table A.9 the fixed effects results.\n\n\n17GSOEP is the German Socio-Economic Panel ; ECHP is the European Community Household Panel\n\n\n23\n\nMention: \"BHPS\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"Results are insignificant when this panel is used.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002467:24:0", "occurrence_id": "8e9bf99cbc8a809bfccafcee", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 5.2 Hazard -Survival function - unemployment by gender no CI.pdf\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\ndoes not support the hypothesis of habituation, when using the GSOEP and the ECHP <sup>17</sup>,\nbut the results are not significant using the BHPS. The same exercise is executed here.\n\n\nUsing pooled data, the interaction term between unemployment and its duration, it\nis found that unemployment duration has a positive and significant effect on well-being,\nwhen one does not control for the other life variables. However, if one controls for life\nvariables, then duration has no effect on well-being. This hints that duration is associated\nwith the life variables mentioned above, but not directly with well-being. The transition\nmatrix shown before provides evidence that staying in unemployment has no different effect\non well-being than staying in employment. As pooled data is not rigourous, panel data\nspecifications are used. Introducing individual fixed effects makes it possible to regress the\nchange in well-being to the change in duration. A fixed effect logit is used to estimate this\neffect of duration on well-being, for the unemployed. The results show that the coefficient\nof duration is not significant. Hence, it can be said that the effect of unemployment on\nwell-being is independent from its duration. These results are shown in the appendix.\nTable A.8 shows the pooled results, and Table A.9 the fixed effects results.\n\n\n17GSOEP is the German Socio-Economic Panel ; ECHP is the European Community Household Panel\n\n\n23\n\nMention: \"German Socio-Economic Panel\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"A footnote expands the acronym used in the text.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002467:24:0", "occurrence_id": "52de856a977bf66724da4380", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 5.2 Hazard -Survival function - unemployment by gender no CI.pdf\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\ndoes not support the hypothesis of habituation, when using the GSOEP and the ECHP <sup>17</sup>,\nbut the results are not significant using the BHPS. The same exercise is executed here.\n\n\nUsing pooled data, the interaction term between unemployment and its duration, it\nis found that unemployment duration has a positive and significant effect on well-being,\nwhen one does not control for the other life variables. However, if one controls for life\nvariables, then duration has no effect on well-being. This hints that duration is associated\nwith the life variables mentioned above, but not directly with well-being. The transition\nmatrix shown before provides evidence that staying in unemployment has no different effect\non well-being than staying in employment. As pooled data is not rigourous, panel data\nspecifications are used. Introducing individual fixed effects makes it possible to regress the\nchange in well-being to the change in duration. A fixed effect logit is used to estimate this\neffect of duration on well-being, for the unemployed. The results show that the coefficient\nof duration is not significant. Hence, it can be said that the effect of unemployment on\nwell-being is independent from its duration. These results are shown in the appendix.\nTable A.8 shows the pooled results, and Table A.9 the fixed effects results.\n\n\n17GSOEP is the German Socio-Economic Panel ; ECHP is the European Community Household Panel\n\n\n23\n\nMention: \"European Community Household Panel\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"A footnote expands the second acronym used in the text.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002478:20:0", "occurrence_id": "49933bfdb32d5bedc699c3d4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: movements Mexico, Brazil and Colombia. <sup>8,9</sup> Although cointegration is sometimes given\n\n\nthe economic interpretation of capturing “long run” relations, fundamentally it is a\n\n\nstatistical relationship existing among non-stationary series. In our case, both relative\n\n\nsector sizes and the real exchange rate can plausibly be posited as I(1) and they always\n\n\nappear to be so in the analysis. Changing aggregate skill levels postulated in section 2\n\n\nsuggest no reason to assume the wages ex ante to be I(0) and even if they were, a slow\n\n\nadjustment process could lead to apparent non-stationary behavior within the sample\n\nanalyzed. <sup>10</sup>\n\n\nWe then use recursive cointegration estimation techniques to explore the stability\n\n\nof the cointegration space over time and we apply LR-test to test hypotheses on the\n\n\ncoefficients of the cointegration vectors in order to provide empirical evidence to test the\n\n\nvarious predictions of the model and to extract some conclusions about labor market\n\n\nrigidities and the source of appreciations in the three countries.\n\n\n**3.1 Data**\n\n\nWe use quarterly data for Mexico, Brazil and Colombia on the log-transformed\n\n\nreal exchange rate, _p_, the wage ratio of formal over self-employed workers, _WT/WN_, and\n\n\nthe ratio of the absolute size of formal over the self-employed sector, _nT/nN_ . Real\n\nexchange rates where taken from IFS, all labor market data were derived from national\n\n\nlabor force surveys ( _Pesquisa Mensual do Empreg_ - for Brazil, _Encuesta Nacional de_\n\n\n_Empleo Urbano_ for Mexico and _Encuesta Nacional de Hogares_, Colombia). The work\n\n\nforce was defined as only the active male population. Because we are interested in\n\n\ninformal workers and not educated workers opening consulting firms, all workers with\n\n\nmore than 12 years of education were\n\nMention: \"IFS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Real exchange rates are taken from this financial statistics compilation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002480:9:1", "occurrence_id": "9bb2f66c7fb4e907601146fa", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  sector to include up to seven\nforms of private enterprises: four types of registered domestic private enterprises, as well\nas household firms, cooperatives and foreign invested enterprises. Official statistics also\nrefer to the “private economy” ( _kinh te tu nhan_ ), which includes only sole proprietorship\ncompanies, while limited liability and shareholding companies are presented separately\nunder the vaguely-titled “mixed economy” ( _kinh te hon hop_ ) category.9 Official statistics\non private companies are also very difficult to obtain and flawed by methodological\ninconsistencies.10\n\n\n11. Reports on the private sector, in turn, have not always distinguished between\nformal companies and household enterprises, sometimes referring to both as formal\n\n\n6 The legality of private enterprises was further strengthened by the Constitution in 1992, which recognized the right of\nfreedom of business in various economic sectors, including the right to equal treatment among all economic sectors.\n7 Household enterprises were previously covered in a separate piece of legislation (Decree 66). Note that even though\nthey are now included in the 1999 Enterprise Law, they are still subject to different regulations than formally registered\nenterprises. Household enterprises are now considered as “quasi-formal” private enterprises. They benefit from simple\nregistration procedures but are restricted in certain business operations, such as the right to trade, scope of operation\nand land use rights. These restrictions are put in place to encourage household enterprises to become formal business\nentities. Household enterprises, however, do continue to enjoy significantly lower taxes as they do not pay VAT taxes\nand are in a better position to negotiate lower taxes with local officials.\n8 Also included is the newest form: a variation on the limited liability company that allows for single ownership unlike\nthe original form that still is defined to include multiple owners.\n9 Webster and Taussig (1999), p. 13.\n10 The GSO’s methods for calculating private sector statistics have varied across recent years. Methodologies appear to\nvary across provinces, too\n\nMention: \"Official statistics\non private companies\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"They are hard to obtain and affected by methodological inconsistencies.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002480:9:1", "occurrence_id": "9b088c29c45790d6aeeae843", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  sector to include up to seven\nforms of private enterprises: four types of registered domestic private enterprises, as well\nas household firms, cooperatives and foreign invested enterprises. Official statistics also\nrefer to the “private economy” ( _kinh te tu nhan_ ), which includes only sole proprietorship\ncompanies, while limited liability and shareholding companies are presented separately\nunder the vaguely-titled “mixed economy” ( _kinh te hon hop_ ) category.9 Official statistics\non private companies are also very difficult to obtain and flawed by methodological\ninconsistencies.10\n\n\n11. Reports on the private sector, in turn, have not always distinguished between\nformal companies and household enterprises, sometimes referring to both as formal\n\n\n6 The legality of private enterprises was further strengthened by the Constitution in 1992, which recognized the right of\nfreedom of business in various economic sectors, including the right to equal treatment among all economic sectors.\n7 Household enterprises were previously covered in a separate piece of legislation (Decree 66). Note that even though\nthey are now included in the 1999 Enterprise Law, they are still subject to different regulations than formally registered\nenterprises. Household enterprises are now considered as “quasi-formal” private enterprises. They benefit from simple\nregistration procedures but are restricted in certain business operations, such as the right to trade, scope of operation\nand land use rights. These restrictions are put in place to encourage household enterprises to become formal business\nentities. Household enterprises, however, do continue to enjoy significantly lower taxes as they do not pay VAT taxes\nand are in a better position to negotiate lower taxes with local officials.\n8 Also included is the newest form: a variation on the limited liability company that allows for single ownership unlike\nthe original form that still is defined to include multiple owners.\n9 Webster and Taussig (1999), p. 13.\n10 The GSO’s methods for calculating private sector statistics have varied across recent years. Methodologies appear to\nvary across provinces, too\n\nMention: \"private sector statistics\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Calculation methods have varied across years and provinces.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002485:6:0", "occurrence_id": "70502c733eccdb7e3287be85", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: location of food aid programs as a result of high fixed start up costs, earlier needs\n\n\nassessments during the second half of the 1980s have been observed to be good\n\n\npredictors of future food aid in Ethiopia (Jayne et al.). The selection of the instrumental\n\n\nvariables is discussed in more detail in Section 5.\n\n\nTo examine the potential differential effect of free food aid distribution (FD) and\n\n\nfood for work (FFW) the predicted amounts received of each kind are also separately\n\nincluded in the child growth regression (β _F_ F _jt_ =β _FD_ F _FDjt_ +β _FFW_ F _FFWjt)_ . The presence of\n\n\nintra-household gender differentiation is explored through the inclusion of interaction\n\n\nterms of the sex of the child with the shock variables.\n\n\n**Child Growth, Food Aid and Crop Failure in Rural Ethiopia**\n\n\nOver the period 1995-96, a series of three nationally representative household\n\n\nsurveys were conducted in Ethiopia as part of the Rural Integrated Household Survey\n\n\nProgram. We integrate information from these surveys either at the household level or the\n\n\nenumeration area (EA)/community level. Anthropometric information on pre-school\n\n\nchildren is provided by the 1995/96 Welfare Monitoring Survey (WMS), which covered\n\n\n12 randomly selected households in each EA/community. In this survey, 2,414 rural\n\nchildren under 5 years old were measured twice with a 6 months interval. <sup>3</sup> After\n\n\nexcluding cases with more than 12-month age difference reported, negative height\n\n\ngrowth, more than 25 cm growth in six months, and a HAZ-score beyond the [–6, +6]\n\nrange, 2,089 children remained spread over 469 EAs out of a total of 531 EAs. <sup>4</sup>\n\n\n3 While 5,012 children under 54 months old were measured during the first round, only 2\n\nMention: \"Rural Integrated Household Survey\n\n\nProgram\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Three nationally representative surveys are integrated at household and community level.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002485:6:0", "occurrence_id": "c3ded99d804c31ded0664a1c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: location of food aid programs as a result of high fixed start up costs, earlier needs\n\n\nassessments during the second half of the 1980s have been observed to be good\n\n\npredictors of future food aid in Ethiopia (Jayne et al.). The selection of the instrumental\n\n\nvariables is discussed in more detail in Section 5.\n\n\nTo examine the potential differential effect of free food aid distribution (FD) and\n\n\nfood for work (FFW) the predicted amounts received of each kind are also separately\n\nincluded in the child growth regression (β _F_ F _jt_ =β _FD_ F _FDjt_ +β _FFW_ F _FFWjt)_ . The presence of\n\n\nintra-household gender differentiation is explored through the inclusion of interaction\n\n\nterms of the sex of the child with the shock variables.\n\n\n**Child Growth, Food Aid and Crop Failure in Rural Ethiopia**\n\n\nOver the period 1995-96, a series of three nationally representative household\n\n\nsurveys were conducted in Ethiopia as part of the Rural Integrated Household Survey\n\n\nProgram. We integrate information from these surveys either at the household level or the\n\n\nenumeration area (EA)/community level. Anthropometric information on pre-school\n\n\nchildren is provided by the 1995/96 Welfare Monitoring Survey (WMS), which covered\n\n\n12 randomly selected households in each EA/community. In this survey, 2,414 rural\n\nchildren under 5 years old were measured twice with a 6 months interval. <sup>3</sup> After\n\n\nexcluding cases with more than 12-month age difference reported, negative height\n\n\ngrowth, more than 25 cm growth in six months, and a HAZ-score beyond the [–6, +6]\n\nrange, 2,089 children remained spread over 469 EAs out of a total of 531 EAs. <sup>4</sup>\n\n\n3 While 5,012 children under 54 months old were measured during the first round, only 2\n\nMention: \"Welfare Monitoring Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Twelve households per community were sampled and 2,414 rural children measured twice.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002488:21:5", "occurrence_id": "e25faafaabdde3ffd885eee4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .0560 0.0642 0.0540\n(0.0010) (0.0038) (0.0009) (0.0012) (0.0009) (0.0005)\n\n\n\nGovernment 0.0703 0.0540 0.0672 0.0560 0.0642 0.0540\n(0.0010) (0.0038) (0.0009) (0.0012) (0.0009) (0.0005)\n\nNo Signed Labour Card 0.2396 0.2991 0.2074 0.2730 0.2198 0.2883\n<u>(0.0017)</u> <u>(0.0033)</u> <u>(0.0015)</u> <u>(0.0023)</u> <u>(0.0015)</u> <u>(0.0021)</u>\n\n\n\nNo Signed Labour Card 0.2396 0.2991 0.2074 0.2730 0.2198 0.2883\n<u>(0.0017)</u> <u>(0.0033)</u> <u>(0.0015)</u> <u>(0.0023)</u> <u>(0.0015)</u> <u>(0.0021)</u>\n\n\nSource: computed from PNAD 1988, 1992 and 1998. Standard deviations of the dummy\nvariables are standard deviations of a proportion.\n\n\n\n21\n\nMention: \"PNAD 1988, 1992 and 1998\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Means and standard deviations of the dummy variables are computed across these rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002495:37:0", "occurrence_id": "8bf5eb3f420a3a5270977901", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: interaction.\n\nOverall, the results reported in this section suggest that there are aspects of an\nindustry’s dependence on financial markets that are not completely captured in the\nmeasure of liquidity needs. Nevertheless, the coefficient of the liquidity needs interaction is robust to the consideration of these alternative channels.\n\n### **7.2 Financial development or institutions?**\n\n\nThis section explores the possibility that the measure of financial development captures the effect of either the general level of economic development or institutional\nquality. The reasons to consider this possibility are that (i) a country’s financial\ndevelopment is correlated with the quality of its institutions (see La Porta et al.\n(1998)), (ii) the measure of a country’s liquidity needs may be related to its “institutional dependence”, <sup>50</sup> and (iii) countries with weak institutions are more likely to\nexperience political crises in which the whole institutional framework is altered. By\nexplicitly considering this channel, the regressions presented in this section attempt\nto disentangle the effects of financial and institutional development.\n\nResults from this analysis are presented in Table 9. Column (1) reproduces the\nbase regression to facilitate comparison. In column (2), the interaction of initial\nGDP per capita, which is a proxy for institutional development, and the measure of\nliquidity needs was added to the main specification. The results show that financial\ndevelopment bears the main responsibility for the differences in sectorial volatility,\nnot economic development.\n\nThe specifications reported in columns (3) to (5) add to the base specification\nthe interaction of a different measure of institutional development, the extent of _con-_\n_straints_ _on_ _the_ _executive_ from the Polity IV dataset, <sup>51</sup> <sup>_,_</sup> <sup>52</sup> with industry characteristics\nthat proxy\n\nMention: \"Polity IV dataset\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Its constraints on the executive measure is interacted with industry characteristics.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002500:13:0", "occurrence_id": "89148977644afe5300dd22bd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 7.** Catastrophe bonds issued on financial markets (Source: Cummins 2003).\n\n\n\n\n\n\n\n**Figure 8.** Reinsurance Pricing Volatility\n\n(1984 = 1.00)\n\n\nSource: Congressional Budget Office based on data from Paragon Reinsurance Risk Management Services.\n\n\n**5.** **Conclusions**\n\nThis paper investigates optimal risk financing strategies for a firm using self-retention,\ninsurance and contingent debt under different assumptions about pricing of these instruments.\nWhile several previous studies point out the importance of such a combination, to the best of\nour knowledge, this paper is the first effort to draw an analytical framework to study the\nunderlying tradeoffs between the costs of different risk financing instruments and how they\naffect the choice of an optimal risk financing strategy. The flexibility of the model employed\nby this research allows us first to distinguish between two sets of technical parameters faced\nby a risk-averse firm: the attachment points for each additional layer of risk financing needed\nto cover the firm’s risk exposure which ultimately results in a four layered structure, and the\nrisk-financing arrangements within each risk layer. The first set of parameters, the attachment\npoints, cannot be explicitly characterized in our model. These model limitations thus justify\nthe use of stochastic simulation tools in further research. The focus of this paper is the second\nset of parameters, the optimal risk-financing rules within each layer, which are examined in\ndetail, with an output being the explicit contract structure. By employing a plausible base case\nscenario risk financing structure, we analyze how changes in the costs of insurance coverage,\nthe level of risk assumed by the firm and the borrowing constraint affect the optimal risk\n\n\n14\n\nMention: \"Paragon Reinsurance Risk Management Services\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its reinsurance pricing series is indexed to show volatility since 1984.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002507:13:0", "occurrence_id": "7b6cd381516494bd66dd7e4a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: However, aggregate statistics on the number of official public telephones do not really\ntell the full story about access in La Paz and El Alto <sup>5</sup> . On the one hand, public telephones\nare not always very accessible since in order to avoid vandalism telephones in outlying\nlocations are often located inside retail outlets, and are therefore not available outside of\nnormal retail hours.\n\nOn the other hand, the informal sector has effectively ‘privatized’ the public telephone\nservice. First, many of the small street vendors or ‘anaqueleros’ have their own telephone\nlines (often coming straight out of the payment) which they rent out to the public for local\ncalls at a fixed charge for a maximum three minute call. Indeed, the current\ntelecommunications law explicitly allows for such small-scale resale of services. Second,\nvendors of public telephone coupons <sup>6</sup> have been observed to padlock public telephone\nbooths. This allows them to improve their sales by forcing any prospective callers to\npurchase coupons from them in return for securing access to the public telephone booth.\n\n**Impact on coverage.** To what extent did the increase in new connections feed through\ninto significant improvements in service coverage? A number of methodological\nconsiderations complicate the drawing of inferences about coverage rates from data on\nthe number of household connections <sup>7</sup> . Hence, as a cross-check, the resulting coverage\nrates were compared with those reported in the Encuesta Continua de Hogares, 1999.\nAny major discrepancies are reported in the text. Notwithstanding, some uncertainty over\nthe absolute level of connections, the results reveal interesting differences in the pattern\nof coverage trends between La Paz (Figure 3.6) and El Alto (Figure 3.7).\n\nIn the city of La Paz, progress in coverage of electricity, water and sewerage services has\nbeen modest since the reform. However, this reflects the\n\nMention: \"data on\nthe number of household connections\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Methodological issues complicate inferring coverage rates from connection counts.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002507:13:0", "occurrence_id": "bfa0e91067e7be22b15c3530", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: However, aggregate statistics on the number of official public telephones do not really\ntell the full story about access in La Paz and El Alto <sup>5</sup> . On the one hand, public telephones\nare not always very accessible since in order to avoid vandalism telephones in outlying\nlocations are often located inside retail outlets, and are therefore not available outside of\nnormal retail hours.\n\nOn the other hand, the informal sector has effectively ‘privatized’ the public telephone\nservice. First, many of the small street vendors or ‘anaqueleros’ have their own telephone\nlines (often coming straight out of the payment) which they rent out to the public for local\ncalls at a fixed charge for a maximum three minute call. Indeed, the current\ntelecommunications law explicitly allows for such small-scale resale of services. Second,\nvendors of public telephone coupons <sup>6</sup> have been observed to padlock public telephone\nbooths. This allows them to improve their sales by forcing any prospective callers to\npurchase coupons from them in return for securing access to the public telephone booth.\n\n**Impact on coverage.** To what extent did the increase in new connections feed through\ninto significant improvements in service coverage? A number of methodological\nconsiderations complicate the drawing of inferences about coverage rates from data on\nthe number of household connections <sup>7</sup> . Hence, as a cross-check, the resulting coverage\nrates were compared with those reported in the Encuesta Continua de Hogares, 1999.\nAny major discrepancies are reported in the text. Notwithstanding, some uncertainty over\nthe absolute level of connections, the results reveal interesting differences in the pattern\nof coverage trends between La Paz (Figure 3.6) and El Alto (Figure 3.7).\n\nIn the city of La Paz, progress in coverage of electricity, water and sewerage services has\nbeen modest since the reform. However, this reflects the\n\nMention: \"Encuesta Continua de Hogares\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its reported coverage rates are compared with those implied by connection counts.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002508:23:1", "occurrence_id": "b4b19b0c571327818b7d0d5f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \n7 Low income and middle income countries are those which are classified as such by the World Bank in the\n<u>World Development Report, 2000/01 (Washington, DC), p. 334. Low income includes countries with 1999</u>\nGNP per capita $756 or less; middle-income includes countries with 1999 GNP per capita of $756 to\n$9,265.\n\n8 In line with other cross-national studies of poverty, 1980 was selected as a cutoff point because the\npoverty data prior to that year are far less comprehensive. See, for example, Martin Ravallion and Shaohua\nChen, “What Can New Survey Data Tell Us About Recent Changes in Distribution and Poverty,” World\n<u>Bank Economic Review 11:2 (May 1997): 357-382; and Richard Adams, Jr., Economic Growth, Inequality</u>\n<u>and Poverty: Findings from a New Data Set, World Bank Research Working Paper 2972, Washington, DC,</u>\nFebruary 2003.\n\n9 For the purposes of this study, OECD (Europe) includes 21 countries: Austria, Belgium, Czech Republic,\nDenmark, Finland, France, Germany, Greece, Hungary, Ireland, Italy, Luxemburg, Netherlands, Norway,\nPoland, Portugal, Slovak Republic, Spain, Sweden, Switzerland and United Kingdom.\n\n10 The International Monetary Fund (IMF) records annual flow in international remittances in its\npublication, Balance of Payments Statistics Yearbook (Washington, DC).\n\n11 A recent IMF study estimated that informal transfers of remittance monies currently amount to $10\nbillion per annum. See Mohammed el-Qorchi, Samuel Maimbo and John Wilson, “The Hawala Informal\nFunds Transfer System: An Economic and Regulatory Analysis,” unpublished paper, International\nMonetary Fund, Washington, DC, August 2002, p. 64.\n\n12 For a full list of these 157 developing countries, see World Bank, World Development Report\n\nMention: \"Balance of Payments Statistics Yearbook\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It records annual international remittance flows.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002511:24:0", "occurrence_id": "db9cf553403398422fce1949", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Instrumental Variables**\n\nVolatility of Inflation Measured by the standard deviation of\nthe rate of change in the consumer price\nindex: annual percentage change in the\ncost to the average consumer of\nacquiring a fixed basket of goods and\nservices.\n\n\nReal Exchange Rate Misalignment Absolute deviation of the real exchange\nrate overvaluation from the equilibrium\nreal exchange rate (set to 1).\n\nThe extent of Real Exchange Rate\ndisequilibrium is defined as the difference\nbetween actual real effective exchange\nrate and its equilibrium level, given by\ncross-country purchasing power parity\ncomparisons.\n\n\nSystemic Banking Crises Number of years in which a country\nunderwent a systemic banking crisis, as a\nfraction of the number of years in the\ncorresponding period.\n\n\nVolatility of Terms of Trade shocks Standard deviation of the log difference\nof the terms of trade.\n\n\n\nThe World Bank (2002).\n\n\nLoayza and Kubota (2003)\n\n\nAuthors’s calculations using data from\nCaprio and Klingebiel (1999), and\nKaminsky and Reinhart (1998).\n\n\nThe World Bank (2000) \"World\nDevelopment Indicators\".\n\n\n25\n\nMention: \"World\nDevelopment Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It is one input into the authors' calculations behind the table.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002513:25:1", "occurrence_id": "260a9f6d256d215d834e3ea0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  staff, or that prevailing salaries for highly qualified civil servants are\nhigh enough. The point is rather that donors should not unduly distort the market for\nskilled labor. The resource injection from high donor-paid salaries potentially has a\npositive net impact on development, despite the adverse impact on the functioning of\ngovernment. However, those same benefits could be obtained, without the negative\nconsequences, from using the funds to increase the salaries of underpaid civil servants, or\nthrough general budget support. Budget support targeted for increasing salaries of senior\nand technical staff—particularly where the use of merit systems increases the\n\n\n28 Bureaucratic quality has been linked to growth in several cross-country studies, including Evans and\nRauch (1999) and Keefer and Knack (1997), who measure bureaucratic quality using ICRG data.\n\n\n24\n\nMention: \"ICRG data\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"Cited cross-country studies measure bureaucratic quality with it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002521:9:0", "occurrence_id": "1459cad82a7324d488ed9d4e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _B. Sources_\n\n\nBanking data come from two major sources. BANKSCOPE, maintained by Bureau Van Dijk,\n\n\ncontains financial and ownership information for about 4,000 major banks. BANKERS ALMANAC,\n\n\npublished since 1847, also contains a wealth of data, including detailed ownership data. <sup>9</sup> [^9: Reed Business Information, which is part of Anglo-Dutch Reed Elsevier, maintains Bankers Almanac.]\n\n\nTo obtain ownership information of banks, we also use annual reports, 20-F filings for\n\n\ncompanies with American Depositary Receipts, proxy statements, and country-specific publications.\n\n\nAlso, many individual banks and national institutions (e.g., Central Banks, regulatory authorities)\n\n\nmaintain websites that we used to compile ownership data.\n\n\nSince (1) non-financial institutions own bank shares, (2) BANKSCOPE and the BANKERS\n\n\nALMANAC only have information on financial institutions, and (3) we seek to trace bank ownership\n\n\nthrough corporations back to individuals, we need information on the ownership of non-financial\n\n\ninstitutions. WORLDSCOPE contains ownership data of firms. Thus, we use WORLDSCOPE along\n\n\nwith 20-F filings, company reports, and filings with national stock exchanges and securities regulations\n\n\nto identify the ultimate owners of corporations that own shares in banks. The ownership data are from\n\n\n2001 except in a few cases, where we use 2000 data. Because ownership patterns are very stable, this\n\n\nshould neither induce problems, nor bias the results.\n\n\n_C. Control rights_\n\n\nWe classify a bank as having a controlling owner if the shareholder has direct and indirect\n\n\nvoting rights that sum to 10 percent or more. If no shareholder holds 10 percent of the voting rights, we\n\n\nclassify the bank as widely held. Since 10 percent voting rights is frequently sufficient to exert control,\n\n\nthis cut-off is used extensively (e.g., LLS, 1999; LLSV, 2002). When we use a 20 percent criterion,\n\n\nhowever, we obtain the same conclusions as those discussed below.\n\n\n9 Reed Business\n\nMention: \"BANKSCOPE\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Financial and ownership information for about four thousand major banks is drawn from it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002521:9:0", "occurrence_id": "2d77bc8407287356ddba9641", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _B. Sources_\n\n\nBanking data come from two major sources. BANKSCOPE, maintained by Bureau Van Dijk,\n\n\ncontains financial and ownership information for about 4,000 major banks. BANKERS ALMANAC,\n\n\npublished since 1847, also contains a wealth of data, including detailed ownership data. <sup>9</sup> [^9: Reed Business Information, which is part of Anglo-Dutch Reed Elsevier, maintains Bankers Almanac.]\n\n\nTo obtain ownership information of banks, we also use annual reports, 20-F filings for\n\n\ncompanies with American Depositary Receipts, proxy statements, and country-specific publications.\n\n\nAlso, many individual banks and national institutions (e.g., Central Banks, regulatory authorities)\n\n\nmaintain websites that we used to compile ownership data.\n\n\nSince (1) non-financial institutions own bank shares, (2) BANKSCOPE and the BANKERS\n\n\nALMANAC only have information on financial institutions, and (3) we seek to trace bank ownership\n\n\nthrough corporations back to individuals, we need information on the ownership of non-financial\n\n\ninstitutions. WORLDSCOPE contains ownership data of firms. Thus, we use WORLDSCOPE along\n\n\nwith 20-F filings, company reports, and filings with national stock exchanges and securities regulations\n\n\nto identify the ultimate owners of corporations that own shares in banks. The ownership data are from\n\n\n2001 except in a few cases, where we use 2000 data. Because ownership patterns are very stable, this\n\n\nshould neither induce problems, nor bias the results.\n\n\n_C. Control rights_\n\n\nWe classify a bank as having a controlling owner if the shareholder has direct and indirect\n\n\nvoting rights that sum to 10 percent or more. If no shareholder holds 10 percent of the voting rights, we\n\n\nclassify the bank as widely held. Since 10 percent voting rights is frequently sufficient to exert control,\n\n\nthis cut-off is used extensively (e.g., LLS, 1999; LLSV, 2002). When we use a 20 percent criterion,\n\n\nhowever, we obtain the same conclusions as those discussed below.\n\n\n9 Reed Business\n\nMention: \"BANKERS ALMANAC\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It supplies complementary data including detailed ownership information.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002521:9:0", "occurrence_id": "2111213b60f97e55c7063905", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _B. Sources_\n\n\nBanking data come from two major sources. BANKSCOPE, maintained by Bureau Van Dijk,\n\n\ncontains financial and ownership information for about 4,000 major banks. BANKERS ALMANAC,\n\n\npublished since 1847, also contains a wealth of data, including detailed ownership data. <sup>9</sup> [^9: Reed Business Information, which is part of Anglo-Dutch Reed Elsevier, maintains Bankers Almanac.]\n\n\nTo obtain ownership information of banks, we also use annual reports, 20-F filings for\n\n\ncompanies with American Depositary Receipts, proxy statements, and country-specific publications.\n\n\nAlso, many individual banks and national institutions (e.g., Central Banks, regulatory authorities)\n\n\nmaintain websites that we used to compile ownership data.\n\n\nSince (1) non-financial institutions own bank shares, (2) BANKSCOPE and the BANKERS\n\n\nALMANAC only have information on financial institutions, and (3) we seek to trace bank ownership\n\n\nthrough corporations back to individuals, we need information on the ownership of non-financial\n\n\ninstitutions. WORLDSCOPE contains ownership data of firms. Thus, we use WORLDSCOPE along\n\n\nwith 20-F filings, company reports, and filings with national stock exchanges and securities regulations\n\n\nto identify the ultimate owners of corporations that own shares in banks. The ownership data are from\n\n\n2001 except in a few cases, where we use 2000 data. Because ownership patterns are very stable, this\n\n\nshould neither induce problems, nor bias the results.\n\n\n_C. Control rights_\n\n\nWe classify a bank as having a controlling owner if the shareholder has direct and indirect\n\n\nvoting rights that sum to 10 percent or more. If no shareholder holds 10 percent of the voting rights, we\n\n\nclassify the bank as widely held. Since 10 percent voting rights is frequently sufficient to exert control,\n\n\nthis cut-off is used extensively (e.g., LLS, 1999; LLSV, 2002). When we use a 20 percent criterion,\n\n\nhowever, we obtain the same conclusions as those discussed below.\n\n\n9 Reed Business\n\nMention: \"WORLDSCOPE\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its firm ownership data allow bank ownership to be traced through non-financial corporations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002528:19:6", "occurrence_id": "0702c9a3bcc11dba95b91b16", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ,9<br>31,7<br>22,0<br>23,8<br>38,7|58,3<br>39,3<br>29,9<br>20,1<br>20,1<br>34,4|\n\n\n\n_Source:_ Author’s calculations based on PNAD 1981-2001.\n\n\n20\n\n\n\nnone or less\nthan 1\n\nMention: \"PNAD 1981-2001\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Distributional shares by education are computed from this series of household surveys.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002535:14:1", "occurrence_id": "22ab9fcdd7ac50be59be80ce", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  6.6 6.6\n\n\nAverage farm price 30.4 28.9 29.3 28.1 28.3 28\n\n\nCCC export price 20 20 20 20 20 20\n\n\nExport unit value 33 28.6 29.2 32.2 32.7 32.8\n\n\nRotterdam c.i.f. Price 37.1 29.2 33.6 31.6 36 29\n\n\nSource: Skinner (1999)\n\nThe domestic support as measured by the aggregate measure of support (AMS) for U.S.\ngroundnuts was estimated at around $330 million during that year or $31 million more than the\naverage AMS for 1996-2001. The average cost of aggregate support in 1996-01 stood at $206\nper metric ton of groundnut produced in the U.S. (Table 7).\n\n\n**Table 7. U.S. aggregate measurement of support (million dollars)**\n\n<u>1986-88</u> <u>1995-96</u> <u>1996-97</u> <u>1997-98</u> <u>1999-2000</u> <u>2000-01</u>\n\nAMS Million US$ 347.2 414.6 299 305.76 300 330\n\nAMS/ Metric Ton of US Production <u>203.3</u> <u>264.1</u> <u>180.0</u> <u>190.5</u> <u>172.7</u> <u>222.8</u>\n\n\nSource: Skinner (1999), Hart and Babcock (2002) for 2000-01 AMS and USDA database for production data\n\n\nWith the 2002 farm bill, groundnut production quotas were eliminated (with a quota\nbuyout) and the groundnut price support program converted to a system of direct and countercyclical payments, and a price floor _cum\n\nMention: \"USDA database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Production quantities are taken from this agricultural statistics source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002514:31:6", "occurrence_id": "b60833bab78c3d459c5086a8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Product Accounts<br>|b Industrial classification numbers used for the Slovenian National Income and Product Accounts<br>|b Industrial classification numbers used for the Slovenian National Income and Product Accounts<br>|\n|c Industry’s share of total manufacturing output in Slovenia.  These sectors represent approximately two-thirds of Slovenian manufacturing output over the period.|c Industry’s share of total manufacturing output in Slovenia.  These sectors represent approximately two-thirds of Slovenian manufacturing output over the period.|c Industry’s share of total manufacturing output in Slovenia.  These sectors represent approximately two-thirds of Slovenian manufacturing output over the period.|c Industry’s share of total manufacturing output in Slovenia.  These sectors represent approximately two-thirds of Slovenian manufacturing output over the period.|c Industry’s share of total manufacturing output in Slovenia.  These sectors represent approximately two-thirds of Slovenian manufacturing output over the period.|c Industry’s share of total manufacturing output in Slovenia.  These sectors represent approximately two-thirds of Slovenian manufacturing output over the period.|c Industry’s share of total manufacturing output in Slovenia.  These sectors represent approximately two-thirds of Slovenian manufacturing output over the period.|c Industry’s share of total manufacturing output in Slovenia.  These sectors represent approximately two-thirds of Slovenian manufacturing output over the period.|c Industry’s share of total manufacturing output in Slovenia.  These sectors represent approximately two-thirds of Slovenian manufacturing output over the period.|c Industry’s share of total manufacturing output in Slovenia.  These sectors represent approximately two-thirds of Slovenian manufacturing output over the period.|c Industry’s share of total manufacturing output in Slovenia.  These sectors represent approximately two-thirds of Slovenian manufacturing output over the period.|c Industry’s share of total manufacturing output in Slovenia.  These sectors represent approximately two-thirds of Slovenian manufacturing output over the period.|\n\nMention: \"the Slovenian National Income and Product Accounts\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Its industrial classification numbers are used to align sectors across sources.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002538:13:3", "occurrence_id": "08f25baa8f0e7aec85557bc8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  <mark>11.5</mark> <mark>14.6</mark>\nNet domestic assets of RBI 12.0 7.0 <mark>6.6</mark> <mark>6.0</mark> <mark>4.8</mark> <mark>4.2</mark> <mark>2.5</mark> <mark>-2.2</mark>\n<u>Net public debt</u> <u>63.8</u> <u>53.8</u> <u><mark>55.2</mark></u> <u><mark>56.5</mark></u> <u><mark>60.8</mark></u> <u><mark>65.6</mark></u> <u><mark>70.2</mark></u> <u><mark>77.6</mark></u>\nSource: Reserve Bank of India, Monthly Bulletins\n\n\n36 The definition of net public debt also includes some other terms on RBI’s balance sheet, as shown in Table 5, which\nwe ignore here for simplicity.\n37 Strictly speaking, this holds approximately because net foreign assets and net domestic assets are both expressed as\nshares of GDP.\n38 Another way of making this point is to say that genuine assets can be created only by saving, not borrowing. To\nquote Flood and Jeanne (2000) “…reserves are expensive window dressing…by accumulating them the government\nbehaves as an insolvent household that would try to impress creditors by financing large cash balances with credit card\ndebt”.\n39 That is, RBI was exchanging an appreciating asset (rupee government securities) for a depreciating one (dollars) with\na lower nominal\n\nMention: \"Monthly Bulletins\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its published series supply the debt and financial aggregates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002540:21:1", "occurrence_id": "f5369156b4e29a03f9272524", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  creation through young firms, the more destruction there also is. So ExitUS should\n\n\nserve as a proxy for “natural entry” and when we replace EntryUS with it in the regression, the\n\n\ninteraction has the appropriate negative sign and is significant.\n\n\nIn column (ii) we also use the Dunn and Bradstreet data to calculate SME, the ratio of\n\n\nSmall and Medium Enterprises (SMEs), which we define as businesses with less than 250\n\n\nemployees. Since new firms are generally also small, we expect greater entry into industries with\n\n\nlarger shares of smaller firms. Indeed, we find a significantly negative coefficient, suggesting\n\n\nthat higher entry costs discourage entry to industries with larger shares of SMEs.\n\n\n21\n\nMention: \"Dunn and Bradstreet data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"It supplies the share of businesses with fewer than 250 employees.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002542:11:0", "occurrence_id": "91020d09a486bc2eb7570e47", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: children, missed changes in labor supply owing to migration are not likely to be significantly\n\n\ndistorting our results for the population means.\n\n\nIn this paper we relate changes in the price of rice to changes in the labor supply of\n\n\nchildren, young adults, and adults by gender. Let us first overview the rice price changes that\n\n\noccurred between 1993 and 1998. The sixth row in table 1 reports the consumer price of a\n\n\nkilogram of ordinary rice in 1993 and 1998 from the community-level price survey. <sup>9</sup> We deflate\n\n\nthe price of rice with the monthly consumer price index so that all prices are in thousands of\n\n\n1998 (January) Dong. One U.S. dollar corresponds to approximately 14,000 Dong in 1998. <sup>10</sup>\n\n\nThe average domestic price of rice increased by 28 percent relative to the rise in the consumer\n\n\nprice index (Table 1). <sup>11</sup> Figure 1 plots the real rice price data for the 147 panel communes that\n\n\nare revisited in both rounds of the survey. The horizontal axis is the real price of rice in a\n\n\ncommune in 1993, and the vertical axis is the real price of rice in a commune in 1998. The 45\n\ndegree line is also pictured. 95 percent of the panel communes in the VLSS experienced an\n\n\nincrease in the real price of rice between 1993 and 1998. The variation in rice price changes\n\n\nacross communities is substantial. Our study focuses on the 115 rural, panel communes which\n\n\nare a subset of the panel communes pictured in Figure 1. As in figure 1, there is considerable\n\n\n9 6 communes do not report the price in at least one of the survey rounds. We impute the rice price in these\ncommunes. Based on the unit value of rice purchased by households reported in the VLSS, we calculate the mean\nunit value of a kg of rice for a commune in a\n\nMention: \"community-level price survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Consumer prices per kilogram of ordinary rice are taken from this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002542:11:0", "occurrence_id": "11d90b85f07cbdf83e7ee2cd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: children, missed changes in labor supply owing to migration are not likely to be significantly\n\n\ndistorting our results for the population means.\n\n\nIn this paper we relate changes in the price of rice to changes in the labor supply of\n\n\nchildren, young adults, and adults by gender. Let us first overview the rice price changes that\n\n\noccurred between 1993 and 1998. The sixth row in table 1 reports the consumer price of a\n\n\nkilogram of ordinary rice in 1993 and 1998 from the community-level price survey. <sup>9</sup> We deflate\n\n\nthe price of rice with the monthly consumer price index so that all prices are in thousands of\n\n\n1998 (January) Dong. One U.S. dollar corresponds to approximately 14,000 Dong in 1998. <sup>10</sup>\n\n\nThe average domestic price of rice increased by 28 percent relative to the rise in the consumer\n\n\nprice index (Table 1). <sup>11</sup> Figure 1 plots the real rice price data for the 147 panel communes that\n\n\nare revisited in both rounds of the survey. The horizontal axis is the real price of rice in a\n\n\ncommune in 1993, and the vertical axis is the real price of rice in a commune in 1998. The 45\n\ndegree line is also pictured. 95 percent of the panel communes in the VLSS experienced an\n\n\nincrease in the real price of rice between 1993 and 1998. The variation in rice price changes\n\n\nacross communities is substantial. Our study focuses on the 115 rural, panel communes which\n\n\nare a subset of the panel communes pictured in Figure 1. As in figure 1, there is considerable\n\n\n9 6 communes do not report the price in at least one of the survey rounds. We impute the rice price in these\ncommunes. Based on the unit value of rice purchased by households reported in the VLSS, we calculate the mean\nunit value of a kg of rice for a commune in a\n\nMention: \"monthly consumer price index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"It deflates the rice prices so all values are in constant Dong.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002543:22:0", "occurrence_id": "49312d2142fcd62dd2178cad", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Benjamin and Brandt 2004). At the same time, different regions might have implemented\n\n\ndifferential policy changes and infrastructure improvements during the sample. Or, labor\n\n\nmobility across regions might be limited, which could lead to variation in the changes in the\n\n\nreturns to schooling across regions. These unobserved region-specific time-varying shocks could\n\n\nresult in spurious correlation between rice prices and economic activities. These concerns\n\n\nhowever can be addressed by interacting each region indicator with a year indicator. <sup>1</sup> [^1: There are 4 to 35 sampled communes per region (the mean and the median are both 25 communes per region).] Third,\n\n\nVietnamese communes differ in their accessibility. If more accessible communes are better\n\n\nintegrated into international and national rice markets, these communes might undergo bigger\n\n\nrice price changes. Similarly, accessibility might be associated with a range of time-varying\n\n\nfactors affecting labor supply. The VLSS provides information on whether regular\n\n\ntransportation is provided to a commune and whether the road to commune is paved. Based on\n\n\nthis information, two indicator variables for a communes’ accessibility are constructed and\n\n\ninteracted with the year indicator. Inclusion of these two interactions in the empirical framework\n\n\nallows for a different change in economic activities in accessible communes. Fourth,\n\n\ncommunities are interviewed at different times of the year in each round of the survey. This\n\n\nmight generate a spurious correlation between changes in child labor and rice prices due to\n\n\nseasonality. For example, rice prices may be low in harvest season and demand for child labor\n\n\nhigh. Rice prices may be high in off-season and demand for child labor low. Thus, a regression\n\n\nof rice prices on child labor would merely capture the fact that different communities are\n\n\ninterviewed at different times. In the empirical work, this can be controlled for by the inclusion\n\n\nof season indicators, an indicator for whether the interview took place at a rice harvest, and an\n\n\nindicator for whether an interview took place at rice planting time control for seasonal variation.\n\n\n**4.2 Empirical Framework**\n\n1\n\nMention: \"The VLSS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It records whether communes have regular transport and paved roads.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002544:34:1", "occurrence_id": "af707a239f6fcf997aa31707", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ATO 20 483 613 51 1,167\nGUERRERO 59 105 643 447 76 1,330\nHIDALGO 44 143 488 489 91 1,255\nJALISCO 42 289 388 797 108 1,624\nMICHOACAN 69 399 384 558 95 1,505\nMORELOS 15 48 420 927 64 1,474\nNAYARIT 6 14 441 679 81 1,221\nNUEVO LEON 6 411 939 104 1,460\nOAXACA 34 448 709 516 64 1,771\nPUEBLA 20 401 432 473 96 1,422\nQUERETARO 18 52 504 500 138 1,212\nQUINTANA ROO 5 45 385 809 85 1,329\nSAN LUIS POTOSI 35 444 464 497 90 1,530\nSINALOA 20 16 415 643 103 1,197\nSONORA 2 412 345 773 477 2,009\nTABASCO 20 409 544 484 71 1,528\nTAMAULIPAS 12 394 787 73 1,266\nTLAXCALA 6 533 604 79 1,222\nVERACRUZ 45 800 1,867 2,083 66 4,861\nYUCATAN 10 400 409 830 74 1,723\nZACATECAS 11 484 517 98 1,110\n<u>Total</u> <u>638</u> <u>5123</u> <u>15742</u> <u>27277</u> <u>4429</u> <u>53,209</u>\n<u>Source: Primary Education Assessment Survey, second round SEP, 1997</u>\n\n\n35\n\nMention: \"Primary Education Assessment Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"State-level school and student counts are tabulated from this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002546:11:3", "occurrence_id": "9a831eb10b4e60be3ba81056", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: employed, on a contract**<br>**that has been tendered or**<br>**advertised, admitted for a**<br>**period not exceeding 6**<br>**months in any 12 month**<br>**period,  with no mention of**<br>**an  economic needs test.**<br>Commitments are subject to<br>the application of a<br>numerical ceiling, except<br>where otherwise indicated,<br>for a particular sub-sector.<br>Commitments apply to a<br>specific list of professions.|Independent professionals should be<br>admitted for short periods of stay up to  a<br>maximum of 1 year.<br>|\n\n\n\nSource: Compiled by authors from WTO documents and national websites.\n\n12\n\nMention: \"WTO documents and national websites\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"The authors compiled the commitments table from these official sources.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002548:3:0", "occurrence_id": "47c60056db5da7c76165a79e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: areas and plateaus. The primary sampling units (PSU) are villages. A total of 380,305\n\n\nrural households were interviewed during the month of June 1992. The survey\n\n\nquestionnaire design follows closely that of the DHS. There are four basic survey\n\n\ninstruments: (1) a household questionnaire, (2) a questionnaire for women, (3) a\n\n\nquestionnaire for children, and (4) a village questionnaire (only for the rural sample).\n\n\nAll women who had given birth since 1976 are interviewed. All births between\n\n\n1977-92 are recorded with the outcomes, dates, place, and type of birth assistance. For\n\n\nchildren who died before age 5 at the time of the survey, information on date, place and\n\n\ncause of death are collected. Studies of several DHS show evidence of downward bias in\n\n\nreporting child deaths; i.e., the longer the recall period, the more likely the respondents\n\n\nmisreport the case. In order to minimize the possible recall bias on child deaths, we\n\n\nconsider live births in the five years preceding the survey, which give a sample of\n\n160,899 births. <sup>4</sup> [^4: The number of child deaths in urban areas (555 over a 5-year period ) is too small to permit a separate\nsurvival analysis, hence we focus on rural areas.] The sample distribution by province is presented in Table 1 in the\n\n\nappendix.\n\n\nSince the 1992 CHS followed closely on the heels of the 1990 census, it is worth\n\n\ncomparing estimates of province-level child mortality rates across the two data sources.\n\n\nAs summarized in Table 1, this comparison is encouraging; it shows that for most\n\n\nprovinces (22 out of 29) the mortality estimates from the CHS are roughly in line with\n\nthose from the census. <sup>5</sup> [^5: The seven provinces which have a considerably different estimate of IMR include Hanan, Jilin, Jiansu and\nGuanxi, Zhejian, Hubei, Guanqi.]\n\n\nThe CHS collects information on\n\nMention: \"1992 CHS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Its province-level child mortality estimates are compared with the census.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002548:3:0", "occurrence_id": "d47ee5b41c9dc43c917382c8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: areas and plateaus. The primary sampling units (PSU) are villages. A total of 380,305\n\n\nrural households were interviewed during the month of June 1992. The survey\n\n\nquestionnaire design follows closely that of the DHS. There are four basic survey\n\n\ninstruments: (1) a household questionnaire, (2) a questionnaire for women, (3) a\n\n\nquestionnaire for children, and (4) a village questionnaire (only for the rural sample).\n\n\nAll women who had given birth since 1976 are interviewed. All births between\n\n\n1977-92 are recorded with the outcomes, dates, place, and type of birth assistance. For\n\n\nchildren who died before age 5 at the time of the survey, information on date, place and\n\n\ncause of death are collected. Studies of several DHS show evidence of downward bias in\n\n\nreporting child deaths; i.e., the longer the recall period, the more likely the respondents\n\n\nmisreport the case. In order to minimize the possible recall bias on child deaths, we\n\n\nconsider live births in the five years preceding the survey, which give a sample of\n\n160,899 births. <sup>4</sup> [^4: The number of child deaths in urban areas (555 over a 5-year period ) is too small to permit a separate\nsurvival analysis, hence we focus on rural areas.] The sample distribution by province is presented in Table 1 in the\n\n\nappendix.\n\n\nSince the 1992 CHS followed closely on the heels of the 1990 census, it is worth\n\n\ncomparing estimates of province-level child mortality rates across the two data sources.\n\n\nAs summarized in Table 1, this comparison is encouraging; it shows that for most\n\n\nprovinces (22 out of 29) the mortality estimates from the CHS are roughly in line with\n\nthose from the census. <sup>5</sup> [^5: The seven provinces which have a considerably different estimate of IMR include Hanan, Jilin, Jiansu and\nGuanxi, Zhejian, Hubei, Guanqi.]\n\n\nThe CHS collects information on\n\nMention: \"1990 census\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its mortality estimates are the benchmark for the survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002551:48:0", "occurrence_id": "be861a2c683993f05b0726f6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Table 6. Correlation of Size and Constraint Variables\n\nSize is measured as the weighted average of average firm size in each of the bins, as described in the text,\nand also as the percentage of employment in firms with fewer than 10 workers. We also examine the\npercentage of employment in tradable goods sectors. On the constraints side, judicial efficiency is based on\nsurvey data from ITAM/GMA at year-end 1998 and is measured as the average of seven individual\nindicators (each on a scale from 0-5): Perceived quality of judges; perceived impartiality of judges;\nadequacy of resources for materials; efficiency in the enforcement of resolutions; efficiency of public\nministry of justice; efficiency of public registry of real estate property; and adequacy of local legislation\nfor the enforcement of contracts. Judicial efficiency is constructed such that a higher score indicates more\nefficiency. The measures of regulation and general trust are taken from responses to the Survey of\nGovernance and Development of Enterprises in Mexico conducted by ITESM in 2001. Regulation is the\nreported percentage of sales expended on regulatory costs, averaged at the state level. generalized trust is a\nresponse on a scale of 1 (disagree) to 7 (agree) to the statement: “Most people can be trusted.” Financial\ndevelopment is the ratio of private sector to GDP in 2000 from Mexican Statistical Institute (INEGI).\nForeign ownership is the share of firms with foreign ownership. Log population are from the Mexican\npopulation census of 2000 and log GDP per capita is from the Mexican Statistical Institute (INEGI), also\nfor 2000. Schooling is represented by the percentage of adult population in the state with at least lower\nsecondary (9 years) education, measured by the 1990 census. * denotes significance at 10 percent; **\ndenotes significance at 5 percent; and *** denotes significance at 1 percent.\n\n\n\nWeighted\naverage firm\n\n<u>size</u>\n\n% of jobs in firms\n<10 workers ***-0\n\nMention: \"survey data from ITAM/GMA\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Seven indicators of perceived judicial quality are averaged into an efficiency measure.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002551:48:0", "occurrence_id": "b216028a31c9cf3b4cf6abdf", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Table 6. Correlation of Size and Constraint Variables\n\nSize is measured as the weighted average of average firm size in each of the bins, as described in the text,\nand also as the percentage of employment in firms with fewer than 10 workers. We also examine the\npercentage of employment in tradable goods sectors. On the constraints side, judicial efficiency is based on\nsurvey data from ITAM/GMA at year-end 1998 and is measured as the average of seven individual\nindicators (each on a scale from 0-5): Perceived quality of judges; perceived impartiality of judges;\nadequacy of resources for materials; efficiency in the enforcement of resolutions; efficiency of public\nministry of justice; efficiency of public registry of real estate property; and adequacy of local legislation\nfor the enforcement of contracts. Judicial efficiency is constructed such that a higher score indicates more\nefficiency. The measures of regulation and general trust are taken from responses to the Survey of\nGovernance and Development of Enterprises in Mexico conducted by ITESM in 2001. Regulation is the\nreported percentage of sales expended on regulatory costs, averaged at the state level. generalized trust is a\nresponse on a scale of 1 (disagree) to 7 (agree) to the statement: “Most people can be trusted.” Financial\ndevelopment is the ratio of private sector to GDP in 2000 from Mexican Statistical Institute (INEGI).\nForeign ownership is the share of firms with foreign ownership. Log population are from the Mexican\npopulation census of 2000 and log GDP per capita is from the Mexican Statistical Institute (INEGI), also\nfor 2000. Schooling is represented by the percentage of adult population in the state with at least lower\nsecondary (9 years) education, measured by the 1990 census. * denotes significance at 10 percent; **\ndenotes significance at 5 percent; and *** denotes significance at 1 percent.\n\n\n\nWeighted\naverage firm\n\n<u>size</u>\n\n% of jobs in firms\n<10 workers ***-0\n\nMention: \"Survey of\nGovernance and Development of Enterprises in Mexico\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its responses give regulatory cost shares and generalized trust by state.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002551:48:0", "occurrence_id": "6e0b896e1f07f18d837b79a7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Table 6. Correlation of Size and Constraint Variables\n\nSize is measured as the weighted average of average firm size in each of the bins, as described in the text,\nand also as the percentage of employment in firms with fewer than 10 workers. We also examine the\npercentage of employment in tradable goods sectors. On the constraints side, judicial efficiency is based on\nsurvey data from ITAM/GMA at year-end 1998 and is measured as the average of seven individual\nindicators (each on a scale from 0-5): Perceived quality of judges; perceived impartiality of judges;\nadequacy of resources for materials; efficiency in the enforcement of resolutions; efficiency of public\nministry of justice; efficiency of public registry of real estate property; and adequacy of local legislation\nfor the enforcement of contracts. Judicial efficiency is constructed such that a higher score indicates more\nefficiency. The measures of regulation and general trust are taken from responses to the Survey of\nGovernance and Development of Enterprises in Mexico conducted by ITESM in 2001. Regulation is the\nreported percentage of sales expended on regulatory costs, averaged at the state level. generalized trust is a\nresponse on a scale of 1 (disagree) to 7 (agree) to the statement: “Most people can be trusted.” Financial\ndevelopment is the ratio of private sector to GDP in 2000 from Mexican Statistical Institute (INEGI).\nForeign ownership is the share of firms with foreign ownership. Log population are from the Mexican\npopulation census of 2000 and log GDP per capita is from the Mexican Statistical Institute (INEGI), also\nfor 2000. Schooling is represented by the percentage of adult population in the state with at least lower\nsecondary (9 years) education, measured by the 1990 census. * denotes significance at 10 percent; **\ndenotes significance at 5 percent; and *** denotes significance at 1 percent.\n\n\n\nWeighted\naverage firm\n\n<u>size</u>\n\n% of jobs in firms\n<10 workers ***-0\n\nMention: \"Mexican Statistical Institute\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Private sector credit relative to GDP is taken from this institute.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002551:48:0", "occurrence_id": "549cb31f9c19b65a73f4e1ad", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Table 6. Correlation of Size and Constraint Variables\n\nSize is measured as the weighted average of average firm size in each of the bins, as described in the text,\nand also as the percentage of employment in firms with fewer than 10 workers. We also examine the\npercentage of employment in tradable goods sectors. On the constraints side, judicial efficiency is based on\nsurvey data from ITAM/GMA at year-end 1998 and is measured as the average of seven individual\nindicators (each on a scale from 0-5): Perceived quality of judges; perceived impartiality of judges;\nadequacy of resources for materials; efficiency in the enforcement of resolutions; efficiency of public\nministry of justice; efficiency of public registry of real estate property; and adequacy of local legislation\nfor the enforcement of contracts. Judicial efficiency is constructed such that a higher score indicates more\nefficiency. The measures of regulation and general trust are taken from responses to the Survey of\nGovernance and Development of Enterprises in Mexico conducted by ITESM in 2001. Regulation is the\nreported percentage of sales expended on regulatory costs, averaged at the state level. generalized trust is a\nresponse on a scale of 1 (disagree) to 7 (agree) to the statement: “Most people can be trusted.” Financial\ndevelopment is the ratio of private sector to GDP in 2000 from Mexican Statistical Institute (INEGI).\nForeign ownership is the share of firms with foreign ownership. Log population are from the Mexican\npopulation census of 2000 and log GDP per capita is from the Mexican Statistical Institute (INEGI), also\nfor 2000. Schooling is represented by the percentage of adult population in the state with at least lower\nsecondary (9 years) education, measured by the 1990 census. * denotes significance at 10 percent; **\ndenotes significance at 5 percent; and *** denotes significance at 1 percent.\n\n\n\nWeighted\naverage firm\n\n<u>size</u>\n\n% of jobs in firms\n<10 workers ***-0\n\nMention: \"Mexican\npopulation census\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"State population levels are taken from this enumeration.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002551:48:0", "occurrence_id": "5316fb3fcb69da08831f6c71", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Table 6. Correlation of Size and Constraint Variables\n\nSize is measured as the weighted average of average firm size in each of the bins, as described in the text,\nand also as the percentage of employment in firms with fewer than 10 workers. We also examine the\npercentage of employment in tradable goods sectors. On the constraints side, judicial efficiency is based on\nsurvey data from ITAM/GMA at year-end 1998 and is measured as the average of seven individual\nindicators (each on a scale from 0-5): Perceived quality of judges; perceived impartiality of judges;\nadequacy of resources for materials; efficiency in the enforcement of resolutions; efficiency of public\nministry of justice; efficiency of public registry of real estate property; and adequacy of local legislation\nfor the enforcement of contracts. Judicial efficiency is constructed such that a higher score indicates more\nefficiency. The measures of regulation and general trust are taken from responses to the Survey of\nGovernance and Development of Enterprises in Mexico conducted by ITESM in 2001. Regulation is the\nreported percentage of sales expended on regulatory costs, averaged at the state level. generalized trust is a\nresponse on a scale of 1 (disagree) to 7 (agree) to the statement: “Most people can be trusted.” Financial\ndevelopment is the ratio of private sector to GDP in 2000 from Mexican Statistical Institute (INEGI).\nForeign ownership is the share of firms with foreign ownership. Log population are from the Mexican\npopulation census of 2000 and log GDP per capita is from the Mexican Statistical Institute (INEGI), also\nfor 2000. Schooling is represented by the percentage of adult population in the state with at least lower\nsecondary (9 years) education, measured by the 1990 census. * denotes significance at 10 percent; **\ndenotes significance at 5 percent; and *** denotes significance at 1 percent.\n\n\n\nWeighted\naverage firm\n\n<u>size</u>\n\n% of jobs in firms\n<10 workers ***-0\n\nMention: \"1990 census\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The share of adults with lower secondary education is measured here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002552:10:0", "occurrence_id": "c11b70a5d45cdc93840b52a7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: We also construct a variable XCONST that is the average for the years 1990-99 of\n\n\nthe Polity IV sub-index on executive constraints. XCONST measures the extent of\n\n\ninstitutionalized constraints on the decision-making powers of chief executives.\n\n\nXCONST ranges from 0 to 10 with a higher score indicating less _de facto_ operational\n\n\nindependence (and more accountability) of the country’s chief executive. XCONST\n\n\ncaptures the extent to which the political system enables minorities or non-elites to\n\n\ninfluence political elites in regular ways.\n\n\nNext, we construct a variable COALITION that takes a value of one if the\n\n\ngovernment of the country is a coalition government, and zero otherwise. The variable\n\n\nCOALITION is averaged for the years 1975-97 and is based on data collected by the\n\n\nWorld Bank. For more details on this database, see Beck et al. (2001). Unfortunately, we\n\n\ndo not have cross-country data on political campaign contributions from interest groups,\n\n\nunlike Kroszner and Strahan (2001) in their study of state-level deposit insurance limits\n\n\nin the United States. Therefore, we cannot control for the effect that money contributions\n\n\nmay have on the voting patterns of politicians.\n\n\nSecond, we control for structural differences within the banking industry. Private\n\n\ninterest theory emphasizes how different interest groups can provide funds and votes to\n\n\npoliticians who, in turn, control the regulatory decision to raise the coverage level of\n\n\ndeposit insurance. Private interest theory thus predicts that support for high coverage\n\n\nlevels is more likely in countries with greater political influence of risky banks relative to\n\n\nthat of safe banks. Public interest theory, on the other hand, would imply that support for\n\n\nhigh deposit insurance coverage is less likely in countries where the share of risky banks\n\n\nis large because the social costs of deposit insurance are directly related to the size of the\n\n\n10\n\nMention: \"Polity IV\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Its executive constraints sub-index is averaged across the decade.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002553:20:2", "occurrence_id": "f57b8d49acddfe59cfe26896", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \nPlot left fallow 1.05% 1.02% 1.29% 1.29% 0.72% 1.51% 0.91%\nOutput per acre (US $) 175 180 352 79 77 90 201\n**Impact of conflict (plots w conflict)**\nNo impact mentioned 22.62% 22.22% 18.92% 27.59% 38.89%\nSocial problems 41.97% 48.15% 44.60% 31.03% 27.78%\nContemporary economic loss 10.82% 6.17% 13.51% 13.79% 0.00%\n<u>Reduced investment</u> <u>24.59%</u> <u>23.46%</u> <u>22.97%</u> <u>27.59%</u> <u>33.33%</u>\n\n\nSource: Own analysis based on the 2001 EPRC/World Bank survey.\n\n\n20\n\nMention: \"the 2001 EPRC/World Bank survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Reported investment reductions are tabulated from this enterprise survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002558:8:0", "occurrence_id": "ccc0b2c118551a23514e57e4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Table 2 provides a detailed description of the types of information contained in\n\n\nthe resulting October Inquiry files for 1983 to 1999, the period on which we focus. Here\n\n\nwe only report on countries that reported at least one gender breakdown in the period\n\n\n1983-99. In total there were 83 countries which did report at least one gender breakdown\n\n\nduring this period. Freeman and Oostendorp (2000, 2002) provide information on all\n\n\ncountries available in the October Inquiry files, also for the countries which did not report\n\n\nany male-female breakdown.\n\n\nPanel A gives information on the size of the sample. It shows the maximum\n\n\nconceivable observations that the Inquiry would contain if each country reported a female\n\n\nand male wage statistic for each occupation yearly: over 227,171 male-female wage\n\npairs. <sup>7</sup> [^7: The maximum is the multiplicand of the number of countries (83) times the number of occupations (161)\ntimes the number of years (17).] The actual number of observations is smaller, largely because in many years most\n\n\ncountries do not report statistics. On average, countries report male-female wages for 6.7\n\n\nyears out of 17 possible years. As a result, 80,500 of the potential observations are\n\n\nmissing, because various countries did not report data in particular years. Moreover, in\n\n\nthe years when countries did report, they did not report data for every occupation. The\n\n\nmain point is that there are 13,020 country-year-occupation cells with female-male wage\n\n\ndata in the 1983-99 file.\n\n\nThere is a further complication. Many countries report more than one female-male\n\n\nwage pair for a single occupation. Some give hourly wage rates _and_ average earnings.\n\n\nOther give female-male wages for different locations. Nearly half the observations (45\n\n\npercent) contain multiple wage figures. Including multiple wages, there are 18,931\n\n\nfemale-male wage pairs.\n\n\nPanel B shows the frequency distribution of countries by the number of\n\n\noccupations they report; and the frequency distribution of occupations by the number of\n\n\ncountries that report statistics on them\n\nMention: \"October Inquiry files\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Country reports of gender breakdowns are compiled from these inquiry returns.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002559:25:0", "occurrence_id": "661e70478f7e63ebe2fedc72", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: focused on whether the Indian industrial strategy, which involved extensive use of industrial\n\n\nlicensing and tariff and non-tariff barriers, led Indian manufacturing to perform poorly\n\n\nrelative to other countries in particular those in East Asia (Bhagwati and Desai 1970, and\n\n\nBhagwati and Srinivasan 1975). <sup>8</sup> [^8: Between 1960 and 1995, manufacturing as a share of GDP grew from 9 percent to 24\npercent of GDP in Indonesia, 8 percent to 26 percent in Malaysia and 12.5 percent to 28\npercent in Thailand. In contrast, manufacturing output in India constituted 13 percent of GDP\nin 1960 but grew to only 18 percent of GDP by 1995. Source: World Development.\nIndicators.] Nationwide policies, however, have limited scope to explain\n\n\nthe marked cross-state variation in manufacturing performance that we observe in Figure 4.\n\n\nTo understand these patterns we need to focus on state specific policies and\n\n\nconditions. We examine three examples of research on how 1 <sup>st</sup> advantage factors and the\n\n\ninterplay between 1 <sup>st</sup> advantage and 2 <sup>nd</sup> advantage factors determine the pattern of\n\n\nmanufacturing growth in India. These examples illustrate how the framework we have\n\n\ndeveloped can be used build up a microeconomic evidence base on what matters.\n\n\n_Example [1]: 1_ <sup>_st_</sup> _advantage: labor regulation and registered manufacturing_\n\n\nLabor regulations have been identified as an important element of the investment climate in\n\n\nIndia (Stern, 2001; Dollar, Iarossi and Mengitsae, 2001, Sachs et al, 1999). Besley and\n\n\nBurgess (2002) examine whether labor regulation is an important source of 1 <sup>st</sup> advantage\n\n\nwhich could help explain differences in manufacturing performance across Indian states. To\n\n\ndo this they exploit two key facts: (i) labor regulations only apply to firms in the registered\n\n\nmanufacturing sector, and (ii) the Indian constitution empowers state governments to amend\n\n\ncentral legislation. The\n\nMention: \"World Development.\nIndicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Manufacturing output as a share of GDP is drawn from this compilation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002561:5:1", "occurrence_id": "53721cad3489ea30d50cebb4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  EFI Database includes nearly four decades of data, 1962-99. Our analysis covers a\nshorter period, 1982-97, due to the more limited availability of some of the explanatory variables in\nthe econometric model. The database contains data on Romanian trade flows mainly from 1990\nforward. Changes in national boundaries during the 1990s affect the definitions of several\nRomanian trading partners. The database includes Germany as a single country since 1991 and as\ntwo countries, West Germany (“Germany FR”) and East Germany (“Germany DR”) in prior years.\nIt includes Czechoslovakia, Yugoslavia (“ex-Yugoslavia”), and the Soviet Union (“ex-USSR”) as\nsingle countries before 1993; from 1993 forward, it includes the Czech Republic, Slovakia,\nMacedonia, Bosnia and Herzegovina, Croatia, Slovenia, Yugoslavia (i.e, Serbia and Montenegro),\nand the individual republics of the former Soviet Union (Russian Federation, etc.) as separate\ncountries. It reports data separately for Belgium and Luxembourg only after 1998.\n\n**2.2** **A relative measure of discrepancies in trade statistics**\n\n\nWe developed a relative measure of discrepancies in trade statistics. Denote imports and\nexports by the following variables:\n\n\n\n_I_ <sup>_R_</sup>\n\n\n\n_I_ <sup>_R_</sup> _ij_ Reported imports by country _i_ from country _j_\n\n_E_ <sup>_R_</sup> _ij_ Reported exports by country _j_ to country _i_\n\n\n\n_E_ <sup>_R_</sup> _ij_ Reported exports by country _j_ to country _i_\n\n\nThen the relative discrepancy is defined as:\n\n\n\n6 They also made an effort to fill in missing data on physical quantities, by dividing data on the value of trade by\nunit values\n\nMention: \"EFI Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"It supplies nearly four decades of trade data, though the analysis uses 1982-97.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002561:5:1", "occurrence_id": "37082f3e18dd806782587134", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  EFI Database includes nearly four decades of data, 1962-99. Our analysis covers a\nshorter period, 1982-97, due to the more limited availability of some of the explanatory variables in\nthe econometric model. The database contains data on Romanian trade flows mainly from 1990\nforward. Changes in national boundaries during the 1990s affect the definitions of several\nRomanian trading partners. The database includes Germany as a single country since 1991 and as\ntwo countries, West Germany (“Germany FR”) and East Germany (“Germany DR”) in prior years.\nIt includes Czechoslovakia, Yugoslavia (“ex-Yugoslavia”), and the Soviet Union (“ex-USSR”) as\nsingle countries before 1993; from 1993 forward, it includes the Czech Republic, Slovakia,\nMacedonia, Bosnia and Herzegovina, Croatia, Slovenia, Yugoslavia (i.e, Serbia and Montenegro),\nand the individual republics of the former Soviet Union (Russian Federation, etc.) as separate\ncountries. It reports data separately for Belgium and Luxembourg only after 1998.\n\n**2.2** **A relative measure of discrepancies in trade statistics**\n\n\nWe developed a relative measure of discrepancies in trade statistics. Denote imports and\nexports by the following variables:\n\n\n\n_I_ <sup>_R_</sup>\n\n\n\n_I_ <sup>_R_</sup> _ij_ Reported imports by country _i_ from country _j_\n\n_E_ <sup>_R_</sup> _ij_ Reported exports by country _j_ to country _i_\n\n\n\n_E_ <sup>_R_</sup> _ij_ Reported exports by country _j_ to country _i_\n\n\nThen the relative discrepancy is defined as:\n\n\n\n6 They also made an effort to fill in missing data on physical quantities, by dividing data on the value of trade by\nunit values\n\nMention: \"data on Romanian trade flows\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Coverage mainly from 1990 and shifting boundaries limit how far back the panel runs.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002564:30:1", "occurrence_id": "e8a336368a828ee31824fb1e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  country\n\nlevel characteristics used in Table IX with one change. I substitute capital control regulations\n\n\nwith the level of foreign direct investment, FDI, taken from the IMF’s Balance of Payment\n\n\nStatistics and YEARS_LIBER which is the number of years since the official date of\n\n\n30\n\nMention: \"the IMF’s Balance of Payment\n\n\nStatistics\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Foreign direct investment levels substitute for capital control regulations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002568:5:0", "occurrence_id": "04c7108a60cb4e8659686f30", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: To construct a water-resource indicator, we draw on two sources of information.\n\n\nThe first is an estimated geographic distribution of excess demand for water resources\n\n\n(surface and sub-surface) in Vörösmarty, et al. (2000). We use GIS to compute the total\n\n\npopulation residing in excess-demand areas identified by this research. The second\n\n\ninformation source is a database of deaths and injuries from floods maintained by the\n\n\nCentre for Research on the Epidemiology of Disasters (CRED, Université Catholique de\n\n\nLouvain). For each of the Bank's partner countries, we calculate the sum of deaths and\n\n\ninjuries for all recorded floods since 1960. In constructing an indicator for flood damage,\n\n\nwe weight deaths to injuries in the ratio 50:1. Using equal weights, we combine our\n\n\nindicators for demand pressure and floods into a composite indicator of water-related\n\n\nproblems. <sup>4</sup>\n\n\nWe derive our indicator for environmental policy and institutional problems from\n\n\ntwo sources. The first is the World Bank's Country Policy and Institutional Assessment\n\n\n(CPIA) database, which rates environmental policies and institutions on a numerical scale\n\n\nof 1 (the lowest) to 6. For this exercise, we reverse the scaling (1 becomes the highest)\n\n\nand normalize the ratings so that countries with the greatest problems score 100. To\n\n\nproxy the scale of the problems confronted by environmental institutions, we compute the\n\n\nmean value of our five thematic indicators (global emissions, pollution, natural resource\n\n\ndegradation, biodiversity threats, water-related problems). <sup>5,6</sup> To assure equal weighting\n\n\n4 Our index of demand pressure also provides a useful proxy for economic damage from drought\nconditions.  We are indebted to our colleagues in the Bank’s Middle East / North Africa region for this\nobservation.\n5 While the CPIA ratings provide useful information for comparing institutional needs, they are not\nsufficient for judging investment priorities because they\n\nMention: \"Country Policy and Institutional Assessment\n\n\n(CPIA) database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Environmental policies and institutions are rated on a reversed one-to-six scale.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002569:10:1", "occurrence_id": "147d3c5e866bc636d6591e8f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  indicators of employment protection legislation (EPL)**\n\n\nThe indicators of employment protection legislation are available for two periods (late 1980s and 1998) and focus on both\nregular and temporary contracts (see Nicoletti _et al._ 1999). Regulations for regular contracts include: i) procedural\ninconvenience that employers face when trying to dismiss a worker; ii) advance notice of the dismissal and severance\npayments; and iii) prevailing standards of, and penalties for, “unfair” dismissals. Indicators of the stringency of EPL for\ntemporary contracts include: i) the “objective” reasons under which they can be offered; ii) the maximum number of\nsuccessive renewals; and iii) the maximum cumulated duration of the contract. The EPL indicator used in the econometric\nanalysis is time-varying, with the shift in regime from the late 1980s stance to that of 1990s being defined on the basis of\ninformation about the timing of major EPL reforms (concerning both temporary and regular workers) in OECD countries.\nIt should be stressed that the EPL indicators are constructed on the basis of differences in regulatory settings across\nOECD countries. The focus is on excessive regulation that could unnecessarily restrict market mechanisms, because it\nmakes reallocation of resources difficult, hindering the response of the economy to structural changes.\n\n\nThe indicator of EPL interacting with industry firm size uses the cross-country average size of firms in each industry to\nminimize possible endogeneity of the firm dimension to regulations. The cross-country average of firm size in each\nindustry is calculated using a sample of 9 OECD countries. For each country, data are from the business register or the\nsocial security administration and cover all firms with one or more employees. See Bartelsman _et al._ (2003) for details.\n\n\n<u>continued</u>\n\n\n12 . The coefficient on human capital remains significant with the same sign if the original variable is used instead\nof the predicted one in the productivity regression\n\nMention: \"business register\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"All firms with at least one employee are counted to give industry-average firm size.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002574:14:0", "occurrence_id": "c8f8e1e6103a9f5fe2be89d5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: This information is taken from Kirilenko and Summers (2003). They report that in\n2001 Brazil’s bank debit tax yield was 1.45% of GDP with a statutory tax rate of 0.37%.\nFigure 2 depicts this simple rootfinding problem. The equation has two roots, being\n\nn ≈ 5 the economically meaningful one. Again, sensitivity analysis will be conducted to\nassess how results are affected by more complex transaction environments.\n\n\nFrom the steady state version of the law of motion of the aggregate capital stock,\nequation (SYS.4), an expression for the rate of depreciation of the stock of physical\n###### capital is obtained: δ = 1− (1− ((1− COR) KOR))⋅ ηss, where η ss −1 is the steady state\n\ngrowth rate of per capita GDP, COR is the consumption-output ratio and KOR is the\ncapital-output ratio. The rate of growth of per capita GDP is evaluated at 2.22% which\ncorresponds to the average value calculated for the 1972-1993 sample using World Bank\ndata on GDP and population. From the same database and for the same time period,\n\nCOR is estimated at 73.6%. KOR is fixed at 2.68 which corresponds to the average\nratio for the 1972-1990 period using Dhareshwar and Nehru’s (1993) data. Thus, δ is set\nat 7.84%.\n\n\nUsing time series of the Selic rate (return on government bonds) for 1980-2000,\nKanczuk (2002) finds that the real interest rate is close to 8% (1.9% on a quarterly basis).\nThis estimate is taken as proxy for the economy’s real rate of return RRR . The real rate\nof return and the marginal product of capital are related simply as RRR = r ss - δ . This\nalong with (SYS.6) yields an expression for the\n\nMention: \"World Bank\ndata on GDP and population\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Per-capita growth of 2.22 percent and capital ratios are averaged over the sample period.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002584:10:0", "occurrence_id": "af6c68afa2b11ab346e93a21", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: city, which implies that it cannot accommodate the population _N_ . A similar argument rules out the\n\n\nother cases where the inequality in the Proposition is violated, establishing the result (see the\n\n\nappendix).\n\n\n_2.3. A numerical simulation_\n\n\nTo further illustrate differences between the restricted and unrestricted cities, it is helpful to\n\n\nrely on a numerical simulation. The numerical example is based on U.S. parameter values, which\n\n\ncan be chosen with some degree of confidence. However, the results apply qualitatively to any\n\n\ncity, including the Bangalore case considered below.\n\n\nThe simulation is based on Cobb-Douglas forms for the utility and production functions.\n\n\nThe number of households _N_ for the city is set at 800,000, implying an effective population of over\n\n\n2 million in the case where a household realistically contains 2.6 persons. Income per household is\n\n\nset at the 2000 U.S. census value of $42,151. About two-thirds of the circular area of the city is\n\n\navailable for housing, with θ in (6) and (9) set equal to 4 (as opposed to 2π) radians. Other\n\n\nparameter values are discussed in the appendix.\n\n\nThe unrestricted city has an <sup>_~~x~~_</sup> 0 <sup>value of 21.4 miles. In the city,</sup> <sup>_h_</sup> <sup>(</sup> <sup>_S_</sup> <sup>) ranges from a value of</sup>\n\n\naround 17.5 at the CBD to 0.4 at <sup>_~~x~~_</sup> 0 . To interpret these values, note that FAR, as represented by\n\n\n_h_ ( _S_ ), is roughly equal to the number of floors in\n\nMention: \"the 2000 U.S. census\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Household income of $42,151 is set to this enumeration's value.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002586:8:1", "occurrence_id": "9f9a703aa951a54b66ed5e71", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |---|---|\n||||**Farebox**|**Farebox**|**Farebox**|**Farebox**|**Farebox**|**Farebox**|**Farebox**|||||||||||||||||||||||||||\n|||||||||||||||||||||||||||||||||||||\n|||||||||||||||||||||||||||||||||||||\n|||||||||||||||||||||||||||||||||||||\n|||||||||||||||||||||||||||||||||||||\n|||||||||||||||||||||||||||||||||||||\n|||||||||||||||||||||||||||||||||||||\n|||||||||||||||||||||||||||||||||||||\n\n\n\n**_Source: Frontier Economics calculations based on SRA data_**\n\n\nIt is clear that the maximum incentive payments available represent a\n\n\nrelatively small proportion of\n\nMention: \"SRA data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Incentive payments are compared against these returns.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002572:5:0", "occurrence_id": "b3cfa98cca59b2bac816588d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: donors have in many countries (including Uganda) supported the establishment of institutions to deal\n\n\nspecifically with issues of corruption. However, little is known about the extent to which such institutions\n\n\nhave been effective or the extent to which increased levels of accountability would help improve their\n\n\neffectiveness, as would be suggested by a number of recent studies that have pointed to an important role\n\n\nof the mass media to put policy making on a more informed basis (Stromberg 2001, Besley and Burgess\n\n\n2002) and demands to bring in new actors such as NGOs and to foster transparency and open debate as a\n\n\nmeans to expose corrupt practices (Merhotra 2002, Goetz and Jenkins 2002, Ebrahim 2003).\n\n\nUganda provides an interesting case to study this issue. Despite high levels of economic growth since the\n\n\nearly 1990s, the country has ranked continuously in the bottom part of Transparency International’s\n\ncorruption perception index. <sup>1</sup> [^1: While the expansion in the number of countries in the corruption perception index compiled by Transparency International makes a consistent\nassessment more difficult, Uganda ranked as 43 <sup>rd</sup> out of 51 countries in 1996 (the first year when the country was included), 74 <sup>th</sup> of 85 in 1998,] Low levels of transparency and accountability have been identified as\n\n\npotential obstacles to continued progress in reducing poverty (Francis and James 2003). The goal of\n\n\nreducing corruption features prominently in the country’s development strategy and an Inspector General\n\n\nof Government (IGG) and a Ministry of Ethics and Integrity have been set up to help put this into\n\n\npractice. In addition, the Leadership Code Act of 2002 provides for minimum standards of behavior and\n\n\nconduct by leaders serving in public offices (including government departments and statutory bodies) and\n\n\nfor an enforcement mechanism through the IGG’s office. It requires leaders to declare there incomes,\n\n\nassets and liabilities at the time of assumption of office and thereafter every two years (Republic of\n\n\nUganda,\n\nMention: \"Transparency International’s\n\ncorruption perception index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The country's persistent low ranking motivates the case study.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002577:23:0", "occurrence_id": "92dcbd24c14de635e8c6a04b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Source: IMF Government Finance Statistics, various years and World Bank Fiscal Indicators\n\n<u>(http://www1.worldbank.org/publicsector/decentralization/fiscalindicators.htm)</u>\n\n\nThe above discussion implicitly assumes that assignment of taxes entails control\n\n\nover the tax base, tax rates and collection authority. This need not be the case and higher\n\n\nlevel governments may instead, in the interest of harmonization and minimizing\n\n\ncollection and compliance costs, allow lower level governments to either levy a\n\n\nsupplementary rate on own base (tax base sharing) or agree to share the proceeds from\n\n\nspecific taxes in a pre-determined way (shared taxes). Under such arrangements, tax base\n\n\ndetermination usually rests with the higher-level government with lower levels of\n\n\ngovernment levying supplementary rates on the same base. Tax collection is by one level\n\n\nof government, generally the central government in market economies and the local\n\n\ngovernment in transition economies with proceeds shared downward or upward\n\n\ndepending upon revenue yields. Only a handful of developing and transition economies\n\n\nhave adopted tax base sharing. A sub-national surcharge on the personal income tax is\n\n\npermitted in Brazil and Croatia. Russia allows a surcharge on the corporate income tax.\n\n\nProvincial governments in Pakistan allow local governments to have a supplementary\n\n\nrate on property transfer taxes. While the practice of tax base sharing is uncommon,\n\n\nsharing the proceeds of various taxes on a tax by tax basis is frequently practiced in\n\n\nDTEs. In transition economies, in 1999, 49.3% of sub-national government revenues\n\n\n21\n\nMention: \"IMF Government Finance Statistics\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its annual publications supply the public finance aggregates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002577:23:0", "occurrence_id": "454470345eaa0e105e21bb38", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Source: IMF Government Finance Statistics, various years and World Bank Fiscal Indicators\n\n<u>(http://www1.worldbank.org/publicsector/decentralization/fiscalindicators.htm)</u>\n\n\nThe above discussion implicitly assumes that assignment of taxes entails control\n\n\nover the tax base, tax rates and collection authority. This need not be the case and higher\n\n\nlevel governments may instead, in the interest of harmonization and minimizing\n\n\ncollection and compliance costs, allow lower level governments to either levy a\n\n\nsupplementary rate on own base (tax base sharing) or agree to share the proceeds from\n\n\nspecific taxes in a pre-determined way (shared taxes). Under such arrangements, tax base\n\n\ndetermination usually rests with the higher-level government with lower levels of\n\n\ngovernment levying supplementary rates on the same base. Tax collection is by one level\n\n\nof government, generally the central government in market economies and the local\n\n\ngovernment in transition economies with proceeds shared downward or upward\n\n\ndepending upon revenue yields. Only a handful of developing and transition economies\n\n\nhave adopted tax base sharing. A sub-national surcharge on the personal income tax is\n\n\npermitted in Brazil and Croatia. Russia allows a surcharge on the corporate income tax.\n\n\nProvincial governments in Pakistan allow local governments to have a supplementary\n\n\nrate on property transfer taxes. While the practice of tax base sharing is uncommon,\n\n\nsharing the proceeds of various taxes on a tax by tax basis is frequently practiced in\n\n\nDTEs. In transition economies, in 1999, 49.3% of sub-national government revenues\n\n\n21\n\nMention: \"World Bank Fiscal Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its decentralization series complements the international statistics.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002596:50:0", "occurrence_id": "620e6b2e0bdba0800222e789", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Coffee productivity has seemingly not increased substantially in the Dominican Republic despite\nyears of efforts to shift to higher yielding varieties, and a slow shakeout of smaller-scale\nproducers (figure III.1). Average yields in the Dominican Republic are below other coffeeproducing countries in the region and appear to be trending downward. <sup>52</sup> These trends do not\nportend well for the competitiveness of the Dominican coffee sector. Low productivity is\ncompounded by increasing costs of production, particularly labor costs and high financing costs.\n\n**Figure III.1 Coffee yields, Dominican Republic and various countries, 1970-2002**\n\n\n1800\n\n\n1600\n\n\n1400\n\n\n1200\n\n\n\n1000\n\n\n800\n\n\n600\n\n\n400\n\n\n200\n\n\n0\n\n\n\n\n\n\n\n**Year**\n\n\nSource: FAO Statistical Database\n\nSeveral representative budgets were examined in an effort to measure the costs of coffee\nproduction. Estimated costs range from US$45-85 per hundredweight <sup>53</sup>, depending on\ntechnology use and yields. As noted above, these estimates are subject to a number of caveats,\nbut compare favorably to estimates from other sources in other countries in the region (table\nIII.11). Low yields, high production costs and poor quality beans are the main constraints to the\ncompetitiveness of the Dominican coffee sector.\n\n\n52 A recent diagnostic carried out for FEDECARES indicates that average yields of its 3,500 members are\nabout 80lbs/tarea. (or 550kg/ha).\n53 A recent study carried out for COCOCAFE (see Mercedes, 2002), indicates higher costs of production\nthan returns from coffee sales for low, medium and high technified coffee producers, although high\ntechnified producers have a lower cost of production per qq and receive a higher price per qq.\n\n\n51\n\nMention: \"FAO Statistical Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Coffee producer prices over three decades are plotted from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002606:39:0", "occurrence_id": "0acd5f1bb7f2dc6a5a95754d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ### **Appendix A: Poverty Indices and Inequality Measures based on PNAD 1981-99**\n\n**Table A1**\n\n\n**<u>P0</u>** <u>1981</u> <u>1985</u> <u>1988</u> <u>1990</u> <u>1993</u> <u>1995</u> <u>1997</u> <u>1998</u> <u>1999</u>\nBrazil 24.4 26.0 25.1 30.3 29.7 21.8 21.9 21.6 22.4\nNE 49.8 51.9 51.9 58.7 56.7 44.6 46.1 43.5 44.3\nSão Paulo 6.6 8.3 6.5 8.8 12.5 7.1 7.4 7.8 8.6\nMaranhão 64.0 62.8 56.1 64.5 65.4 54.4 58.2 54.1 52.0\nPiauí 69.3 69.8 68.1 71.5 62.1 51.6 57.6 50.2 51.8\nCeará 59.1 58.6 58.0 65.8 57.5 47.2 47.6 45.2 46.7\nRio Grande do Norte 49.4 55.7 47.7 57.0 54.5 40.4 39.8 38.3 39.7\nParaíba 58.3 57.0 57.4 60.7 56.9 40.8 42.9 40.5 39.4\nPernambuco 40.8 44.5 45.4 51.8 53.9 39.5 42.5 39.7 41.6\nHalagaos 42.\n\nMention: \"PNAD 1981-99\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Poverty headcount and inequality measures are computed across these rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002608:11:1", "occurrence_id": "de8ef70936a4bcbc596de5a8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \n0if _Oth_ 1 = 1and _Oth_ 1 _Edu_ ≤\n\n_Oth_ 1 _Edu_ - _WomEdu_ otherwise\n\n\n\n\n\n\n\n_Oth_ 1 _Edu_ - _WomEdu_\n\n\n\n(The variables _Oth_ 2 _Edu_, _Oth_ 3 _Edu_ and the relevant interaction variable sare similarly defined)\n\n\nAs this specification cannot be represented as a restricted version of the\noriginal model, a test for non-nested models is performed.\n\n##### **4. Data and Variables**\n\nThe empirical analysis is based on the 1996/97 Mozambique National\nHousehold Survey on Living Conditions (IAF) _._ <sup>11</sup> The survey was designed and\nimplemented by the National Statistics Institute in Mozambique, and was conducted\nfrom February 1996 to April 1997. The sample was selected in three stages and is\ngeographically stratified to ensure representativeness at both the provincial level and\nfor urban/rural areas. The overall sample consists of 39,961 individuals in 7,920\nhouseholds. <sup>12</sup> For the analysis of maternity services, the relevant subsample consists\nof women between 12 and 49 years old who report having delivered a child in the 12\nmonths preceding the survey (1,584 individuals in 1,509 households). Conversely, the\nanalysis of childhood immunizations is based on the subsample of children between\n12 and 48 months old (3,647 individuals in 3,046 households). <sup>13</sup> However, for the\ndecisions concerning immunizations, the education variable of primary interest relates\nto the mother or guardian. As some women have more than one child in the relevant\nage group, the subsample of mothers or guardians is smaller than the relevant\nsubsample of children (3,113 individuals). <sup>14</sup>\n\nMention: \"Mozambique National\nHousehold Survey on Living Conditions (IAF)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Fieldwork from February 1996 supports the empirical analysis and non-nested model test.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002610:5:1", "occurrence_id": "5790e753da349b80a3a44a0a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: arbitrary control group. The ENEU sample is representative of\n\n\nmetropolitan areas while the CONALEP graduate tracer survey is representative nationally. The\n\n\ndifference in geographical coverage of the two groups makes comparison difficult. Second, the\n\n\ncontrol groups were constructed ad hoc. The control groups included individuals between the\n\n\nages of 17 and 30: (A) those who have completed lower-secondary education; (B) those who\n\n\nhave completed non-professional, elementary vocational training (CECATI), and (C) those who\n\n\nhave completed one to three years of general academic (non-vocational) high school. Some\n\n\ndoubts remain with respect to the second group, since the ENEU survey does not distinguish\n\n\nbetween formal and informal training/technical courses.\n\n\nLee (1998) compares the individuals from the Encuesta de Egresados 1994 (the treatment\n\n\ngroup) with two other groups. One group comprises all 1991 graduates from upper-secondary\n\n\n6\n\nMention: \"ENEU sample\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"It covers only metropolitan areas, unlike the national graduate follow-up.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002610:5:1", "occurrence_id": "2e79d61319df0602c9d3385b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: arbitrary control group. The ENEU sample is representative of\n\n\nmetropolitan areas while the CONALEP graduate tracer survey is representative nationally. The\n\n\ndifference in geographical coverage of the two groups makes comparison difficult. Second, the\n\n\ncontrol groups were constructed ad hoc. The control groups included individuals between the\n\n\nages of 17 and 30: (A) those who have completed lower-secondary education; (B) those who\n\n\nhave completed non-professional, elementary vocational training (CECATI), and (C) those who\n\n\nhave completed one to three years of general academic (non-vocational) high school. Some\n\n\ndoubts remain with respect to the second group, since the ENEU survey does not distinguish\n\n\nbetween formal and informal training/technical courses.\n\n\nLee (1998) compares the individuals from the Encuesta de Egresados 1994 (the treatment\n\n\ngroup) with two other groups. One group comprises all 1991 graduates from upper-secondary\n\n\n6\n\nMention: \"CONALEP graduate tracer survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Nationally representative follow-up of vocational graduates provides the treatment group.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002610:5:1", "occurrence_id": "f89e47eed4e65cd1c42a8f35", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: arbitrary control group. The ENEU sample is representative of\n\n\nmetropolitan areas while the CONALEP graduate tracer survey is representative nationally. The\n\n\ndifference in geographical coverage of the two groups makes comparison difficult. Second, the\n\n\ncontrol groups were constructed ad hoc. The control groups included individuals between the\n\n\nages of 17 and 30: (A) those who have completed lower-secondary education; (B) those who\n\n\nhave completed non-professional, elementary vocational training (CECATI), and (C) those who\n\n\nhave completed one to three years of general academic (non-vocational) high school. Some\n\n\ndoubts remain with respect to the second group, since the ENEU survey does not distinguish\n\n\nbetween formal and informal training/technical courses.\n\n\nLee (1998) compares the individuals from the Encuesta de Egresados 1994 (the treatment\n\n\ngroup) with two other groups. One group comprises all 1991 graduates from upper-secondary\n\n\n6\n\nMention: \"ENEU survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"It cannot separate formal from informal training courses, leaving doubt about the comparison group.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002610:5:1", "occurrence_id": "222b1aab2efe836817417f3b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: arbitrary control group. The ENEU sample is representative of\n\n\nmetropolitan areas while the CONALEP graduate tracer survey is representative nationally. The\n\n\ndifference in geographical coverage of the two groups makes comparison difficult. Second, the\n\n\ncontrol groups were constructed ad hoc. The control groups included individuals between the\n\n\nages of 17 and 30: (A) those who have completed lower-secondary education; (B) those who\n\n\nhave completed non-professional, elementary vocational training (CECATI), and (C) those who\n\n\nhave completed one to three years of general academic (non-vocational) high school. Some\n\n\ndoubts remain with respect to the second group, since the ENEU survey does not distinguish\n\n\nbetween formal and informal training/technical courses.\n\n\nLee (1998) compares the individuals from the Encuesta de Egresados 1994 (the treatment\n\n\ngroup) with two other groups. One group comprises all 1991 graduates from upper-secondary\n\n\n6\n\nMention: \"Encuesta de Egresados 1994\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"A cited study uses this graduate survey as its treatment group.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002616:15:0", "occurrence_id": "ca42cb9869c200417d83e67b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Student 2000 Bgrd Sixth 45,676 3,189\nStudent 2001 Bgrd Third 48,903 3,268\nStudent 2001 Bgrd & Test Fifth 46,718 3,221\nStudent 2002 Bgrd Fourth 48,451 3,205\nStudent 2002 Bgrd & Test Sixth 46,365 3,165\nStudent 1992-2002 Drop First-Sixth 583,876 82,520\nTeacher 1998 Bgrd 23,357 3,562\nTeacher 2001 Bgrd 22,917 3,437\n<u>Teacher</u> <u>2002</u> <u>Bgrd</u> <u>20,634</u> <u>3,373</u>\nSource: SEP data.\n\n**<u>Table 4. Secondary School Databases</u>**\n\n**<u>Respondent</u>** **<u>Year</u>** **<u>Type</u>** **<u>Grade</u>** **<u>Obs</u>** **<u>Schools</u>**\nPrincipal 2000 Bgrd 1,406 1,403\nPrincipal 2001 Bgrd 1,184 1,181\nPrincipal 2002 Bgrd 860 860\nStudent 2000 Test Seventh 39,562 1,189\nStudent 2001 Bgrd & Test Eighth 39,562 1,189\nStudent 2002 Bgrd Eighth 53,919 1,982\n<u>Student</u> <u>2002</u> <u>Bgrd & Test</u> <u>Ninth</u> <u>53,202</u> <u>1,980</u>\nSource: SEP data.\n\n\nInformation on student test scores came from _Estándares\n\nMention: \"SEP data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Teacher record counts by year and grade are drawn from this ministry source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002621:14:2", "occurrence_id": "fe1bb8c140dc0356c96cd0cb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 12 .0003**\n\n**Niger**\n1992 -.0077 .0033* .00002 .0002\n1998 -.0237 .0041** -.0010 .0003**\nPooled -.0123 .0034** -.0004 .0003\n<u>Pooled region fixed-effects (6 groups)</u> <u>-.0124</u> <u>.0034**</u> <u>-.0004</u> <u>.0003</u>\n_Note_ : Selected results from a model that includes child’s age, adults’ education, characteristics of the head of household, and a set of\ncluster characteristics. Cluster characteristics are selected to be consistent across the two rounds of data and therefore have minor\ndifferences with those in the models included in Table 1. *, ** indicate significance of the underlying probit coefficients at the 5, and 1\npercent, respectively.\n_Source:_ Author’s calculation from DHS data.\n\n\n13\n\nMention: \"DHS data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Probit coefficients and their significance are estimated from these survey rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002593:8:0", "occurrence_id": "e58c0d8a8f9951425338ef7c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |Table 4.2: Sectoral GDP, 1985-1998|Col2|Col3|Col4|\n|---|---|---|---|\n|**Sector**|**1985**<br>**(R$ million)**|**1998**<br>**(R$ million)**|**Annual**<br>**Growth**<br>**(Percent)**|\n|Industry|2047|2340|1.0|\n|Services|1629|4052|7.3|\n|Agriculture|375|448|1.4|\n|TOTAL|4051|6840|4.1|\n|_Source_: Contas Regionais, IBGE|_Source_: Contas Regionais, IBGE|_Source_: Contas Regionais, IBGE|_Source_: Contas Regionais, IBGE|\n\n\nThe service sector has been the most dynamic in the last two decades. It increased its\nshare of GDP from 40 percent in 1985 to 59 percent in 1998 while the shares of industry\nand agriculture declined from 50 to 34 percent and from 9 to 7 percent, respectively.\n\nEconomic growth experienced in the last decades in RN is far from homogeneous\nthroughout the state. The large regional differences within the state are associated with\nthe expansion of economic activities in a few of the areas away from the Sertão. Of the\neight zones in RN only two, Litoral Oriental and Mossoroense, experienced an aboveaverage expansion of output during 1970–96. In the past three decades these two veryfast-growing zones on the coast also experienced by far the largest population inflow of\nthe eight zones (Table 4.3). The largest real GDP growth, above 500 percent during\n1970–96, took place\n\nMention: \"Contas Regionais, IBGE\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Sectoral value added and growth rates are taken from these regional accounts.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002615:23:0", "occurrence_id": "71ba63debb72036c0e9d6b5a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: While it is true that there is no general trend toward higher inequality within countries\n\n\nwhen looking at all the countries of the world, the picture is not so favorable if one looks only at\n\n\nrich countries and only at the last decade. The Luxembourg Income Study (LIS) has produced\n\n\ncomparable, high-quality income distribution data for most of the rich countries. This work finds\n\n\nno obvious trends in inequality up through the mid-to-late 1980s. Over the past decade, on the\n\n\nother hand, there have been increases in inequality in most of the rich countries. Because low\n\nskilled workers in these countries are now competing more with workers in the developing\n\n\nworld, it is certainly plausible that global economic integration creates pressures for higher\n\n\ninequality in rich countries, while having effects in poor countries that often go the other way.\n\n\nThe good news from the LIS studies is that “Domestic policies and institutions still have large\n\n\neffects on the level and trend of inequality within rich and middle-income nations, even in a\n\n\nglobalizing world…. [G]lobalization does not force any single outcome on any country.”\n\n\n(Smeeding, 2002)  In other words, among rich countries some have managed to maintain stable\n\n\nincome distributions in this era of globalization through their social and economic policies (on\n\n\ntaxes, education, welfare).\n\n\n_Trend #5. Wage inequality is rising worldwide._ Much of the concern about globalization\n\n\nin rich countries relates to workers and what is happening to wages and other labor issues. The\n\n\nmost comprehensive examination of globalization and wages used International Labour\n\n\nOrganization data on very detailed occupational wages going back two decades (Freeman,\n\n\nOostendrop, and Rama, 2001). These data look across countries at what is happening to wages\n\n\nfor very specific occupations (bricklayer, primary school teacher, nurse, auto worker). What the\n\n\nstudy found is that wages have generally been rising faster in globalizing developing countries\n\n\nthan in rich ones, and faster in\n\nMention: \"Luxembourg Income Study\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"It supplies comparable income distribution data for most rich countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002615:23:0", "occurrence_id": "06d1c21acaadf0512a9e1393", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: While it is true that there is no general trend toward higher inequality within countries\n\n\nwhen looking at all the countries of the world, the picture is not so favorable if one looks only at\n\n\nrich countries and only at the last decade. The Luxembourg Income Study (LIS) has produced\n\n\ncomparable, high-quality income distribution data for most of the rich countries. This work finds\n\n\nno obvious trends in inequality up through the mid-to-late 1980s. Over the past decade, on the\n\n\nother hand, there have been increases in inequality in most of the rich countries. Because low\n\nskilled workers in these countries are now competing more with workers in the developing\n\n\nworld, it is certainly plausible that global economic integration creates pressures for higher\n\n\ninequality in rich countries, while having effects in poor countries that often go the other way.\n\n\nThe good news from the LIS studies is that “Domestic policies and institutions still have large\n\n\neffects on the level and trend of inequality within rich and middle-income nations, even in a\n\n\nglobalizing world…. [G]lobalization does not force any single outcome on any country.”\n\n\n(Smeeding, 2002)  In other words, among rich countries some have managed to maintain stable\n\n\nincome distributions in this era of globalization through their social and economic policies (on\n\n\ntaxes, education, welfare).\n\n\n_Trend #5. Wage inequality is rising worldwide._ Much of the concern about globalization\n\n\nin rich countries relates to workers and what is happening to wages and other labor issues. The\n\n\nmost comprehensive examination of globalization and wages used International Labour\n\n\nOrganization data on very detailed occupational wages going back two decades (Freeman,\n\n\nOostendrop, and Rama, 2001). These data look across countries at what is happening to wages\n\n\nfor very specific occupations (bricklayer, primary school teacher, nurse, auto worker). What the\n\n\nstudy found is that wages have generally been rising faster in globalizing developing countries\n\n\nthan in rich ones, and faster in\n\nMention: \"LIS studies\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Its findings show domestic policy still shapes inequality within rich and middle-income nations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002615:23:0", "occurrence_id": "cadc74c98f45ff27298535ac", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: While it is true that there is no general trend toward higher inequality within countries\n\n\nwhen looking at all the countries of the world, the picture is not so favorable if one looks only at\n\n\nrich countries and only at the last decade. The Luxembourg Income Study (LIS) has produced\n\n\ncomparable, high-quality income distribution data for most of the rich countries. This work finds\n\n\nno obvious trends in inequality up through the mid-to-late 1980s. Over the past decade, on the\n\n\nother hand, there have been increases in inequality in most of the rich countries. Because low\n\nskilled workers in these countries are now competing more with workers in the developing\n\n\nworld, it is certainly plausible that global economic integration creates pressures for higher\n\n\ninequality in rich countries, while having effects in poor countries that often go the other way.\n\n\nThe good news from the LIS studies is that “Domestic policies and institutions still have large\n\n\neffects on the level and trend of inequality within rich and middle-income nations, even in a\n\n\nglobalizing world…. [G]lobalization does not force any single outcome on any country.”\n\n\n(Smeeding, 2002)  In other words, among rich countries some have managed to maintain stable\n\n\nincome distributions in this era of globalization through their social and economic policies (on\n\n\ntaxes, education, welfare).\n\n\n_Trend #5. Wage inequality is rising worldwide._ Much of the concern about globalization\n\n\nin rich countries relates to workers and what is happening to wages and other labor issues. The\n\n\nmost comprehensive examination of globalization and wages used International Labour\n\n\nOrganization data on very detailed occupational wages going back two decades (Freeman,\n\n\nOostendrop, and Rama, 2001). These data look across countries at what is happening to wages\n\n\nfor very specific occupations (bricklayer, primary school teacher, nurse, auto worker). What the\n\n\nstudy found is that wages have generally been rising faster in globalizing developing countries\n\n\nthan in rich ones, and faster in\n\nMention: \"International Labour\n\n\nOrganization data\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"A cited study on globalisation and wages uses its detailed occupational wage series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002618:36:1", "occurrence_id": "9e4b08b1f54eaaa022261bd0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \n**Household Budget Survey (1997)** **Household Budget Survey (2000)**\n\n\n**Regions** **Gini** **E0** **Gini** **E0**\n\nRiga 35.23 21.80 37.29 23.89\nKurzeme 32.06 17.67 34.87 20.76\nVidzeme 30.91 16.12 35.88 25.12\nZemgale 32.78 18.54 34.62 20.38\nLatgale 31.25 17.01 34.49 20.45\nUrban 34.25 20.16 36.58 22.98\nRural 32.09 17.51 35.38 21.43\n\n\n**National** **33.8** **19.67** **37.3** **24.12**\n\n\n37\n\nMention: \"Household Budget Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Regional Gini and inequality elasticity coefficients are computed for two rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002636:18:0", "occurrence_id": "f8a187d674e2fe73f5cbf294", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: policy perspective, as this is the group that is likely to be induced to enroll by other\ninterventions such as tuition subsidies.\n\n\nThe results of the instrumental variable approach implemented in this section\nsuggest that at least some of the effect of college education is causal in reducing smoking\nprevalence. A similar result is obtained by Currie and Moretti (2003) who investigate\nsmoking during pregnancy using the dates of college openings as an instrument for\nmaternal education. Both instruments are likely to be effective at the college margin, which\nis the relevant margin for smoking decisions as documented in figure 2.\n\n\n**5.** **Conclusions**\n\n\nThis paper starts by analyzing smoking histories reconstructed from National\nHealth Interview Surveys in the United States and retracing the evolution of smoking\npatterns across education groups. The main conclusion is that, after the gradual arrival of\nthe information about the dangers of tobacco, the smoking prevalence among more\neducated people declined earlier and more dramatically.\n\n\nThe time pattern of the decline in smoking prevalence for individuals with college\ndegrees suggests that at least some of the differential by education can be ascribed to a\ncausal role of education in giving access to and processing the information. Figure 2 shows\nthat the smoking prevalence among college graduates started to decline earlier than for the\nthree other categories. The inflection point is in 1954, 10 years before the first Surgeon\nGeneral’s Report, four years after the consensus in the medical profession, and one year\nafter the first article in the general press on the dangers of smoking. Further, the results of\nmultivariate regressions show that the effect of college education on smoking prevalence is\nrobust even after controlling for income, indicating that the effect of education goes over\nand above the ability to generate a higher income.\n\n\nThis paper also argues that, because the decision to quit smoking is usually taken\nafter education is completed, this margin of the smoking decision is cleaner to analyze than\nsmoking initiation that often occurs before schooling levels are definitive. A detailed study\nof the incidence of smoking\n\nMention: \"National\nHealth Interview Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Smoking histories are reconstructed to trace patterns across education groups.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002637:25:0", "occurrence_id": "6f711d4e8f77fcf792d46756", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Presence of explicit deposit insurance**\n\nby predominant ownership type\n\n\n0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80\n\n\nMainly Govt\n\n\nMainly Fgn\n\n\nFgn+Govt\n\n\nEqually shared\n\n\nMainly local pvt\n\n\n**Accountability of bank management**\n\nby predominant ownership type\n\n\n2.50 2.55 2.60 2.65 2.70 2.75 2.80 2.85\n\n\nMainly Govt\n\n\nMainly Fgn\n\n\nFgn+Govt\n\n\nEqually shared\n\n\nMainly local pvt\n\n\n\n**Share of countries with all top banks rated**\n\nby predominant ownership type\n\n\n0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0\n\n\nMainly Govt\n\n\nMainly Fgn\n\n\nFgn+Govt\n\n\nEqually shared\n\n\nMainly local pvt\n\n\n**Share of assets in listed banks**\n\nby predominant ownership type\n\n\n0.00 0.10 0.20 0.30 0.40 0.50 0.60 0\n\n\nMainly Govt\n\n\nMainly Fgn\n\n\nFgn+Govt\n\n\nEqually shared\n\n\nMainly local pvt\n\n\n\nFigure 4: _Selected indicators of market discipline preconditions by predominant_\n_ownership of banks, group means_\n\n(Source: World Bank Regulation and Supervision Survey; for definition of predominant type of\nownership see note to Table 2)\n\n\n25\n\nMention: \"World Bank Regulation and Supervision Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Market discipline preconditions are averaged by predominant bank ownership type.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002642:59:0", "occurrence_id": "0106c8bb7dfa621b899131a4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |Links|ountries.cfm|Col3|Web links: http://www.worldbank.org/html/fpd/transpo rt/rail/rdb.htm or http://www.worldbank.org/html/fpd/transpo rt/publicat/td-rw6.htm|Col5|Web link: http://www.eia.doe.gov/emeu/cabs/|Col7|The author may be contacted at: ccaldero@bcentral.cl|\n|---|---|---|---|---|---|---|---|\n|**Description**|<br>|<br>|The World Bank's Railway Database provides information on scale, output<br>and performance for over 90 railways worldwide permitting comparisons of<br>performance and facilitating target setting by individual railways.<br>|<br>|The Energy Information Administration provides policy-independent data,<br>forecasts, and analyses to promote sound policy making, efficient markets,<br>and public understanding regarding energy and its interaction with the<br>economy and the environment. Most of their energy data actually are<br>collected by EIA staff. They design and send their statistical surveys to<br>energy producers, users, transporters, and certain other businesses.<br>Companies and households report directly to them. Besides they have data<br>from other sources, such as trade associations and other government<br>agencies. The information access would be by geography, fuel, sector or<br>price.<br> <br>|<br>|The database includes both physical indicators of quantity and quality of<br>infrastructure endowments and measures of public\n\nMention: \"Railway Database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Catalogued as covering scale, output and performance for over ninety railways.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002642:59:0", "occurrence_id": "98020acc98723fd86f8fc452", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |Links|ountries.cfm|Col3|Web links: http://www.worldbank.org/html/fpd/transpo rt/rail/rdb.htm or http://www.worldbank.org/html/fpd/transpo rt/publicat/td-rw6.htm|Col5|Web link: http://www.eia.doe.gov/emeu/cabs/|Col7|The author may be contacted at: ccaldero@bcentral.cl|\n|---|---|---|---|---|---|---|---|\n|**Description**|<br>|<br>|The World Bank's Railway Database provides information on scale, output<br>and performance for over 90 railways worldwide permitting comparisons of<br>performance and facilitating target setting by individual railways.<br>|<br>|The Energy Information Administration provides policy-independent data,<br>forecasts, and analyses to promote sound policy making, efficient markets,<br>and public understanding regarding energy and its interaction with the<br>economy and the environment. Most of their energy data actually are<br>collected by EIA staff. They design and send their statistical surveys to<br>energy producers, users, transporters, and certain other businesses.<br>Companies and households report directly to them. Besides they have data<br>from other sources, such as trade associations and other government<br>agencies. The information access would be by geography, fuel, sector or<br>price.<br> <br>|<br>|The database includes both physical indicators of quantity and quality of<br>infrastructure endowments and measures of public\n\nMention: \"Energy Information Administration\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Catalogued as a provider of independent energy data, forecasts and analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002643:30:0", "occurrence_id": "d8ae95718bb7835efac120ac", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Efforts to address e-security should consider two tradeoffs. The first is between convenience\nand risk: as a system becomes more convenient to use, it may become more vulnerable to risks. The\nsecond tradeoff is between costs and risks: e-security measures using the latest technology may be\nextremely costly, especially for emerging markets. Once a system’s level of risk has been determined,\nsustainable resources must be provided to cover it. Among the most crucial is sufficient expertise in esecurity software and hardware. E-security is advancing quickly—both in terms of how systems can\nbe attacked and how they can be protected from such attacks. External experts should be hired to\nassess a system’s vulnerabilities, as should in-house security officers at the same level as chief\naccountants and comptrollers.\n\n\n**How Does the Expanded Program Work?**\n\n\nTo purchase securities through the Expanded Small Investors Program, an investor must have a\nchecking account with a participating bank. (Because the Philippines does not have an automated\nclearing house system, these banks settle transactions under the program.) To open an account, an\ninvestor must visit a bank branch and physically and electronically sign the account opening\nstatement. For each account a securities account is created in the Registry of Scripless securities in the\ninvestor’s name.\n\n\nThe investor can then buy securities over the Internet. Purchases are simultaneously debited\nfrom the investor’s bank account and credited to the account in the Registry of Scripless Securities.\nInterest and principal payments are credited to the bank account when due. The system does not allow\ninvestors to sell securities before they mature.\n\n\nThis system has different implications for different parties. The Department of Finance, given\nits difficult fiscal position and heavy dependence on domestic borrowing, would be predisposed to\nfavor initiatives that electronically issue securities directly to investors. This approach could eliminate\nlayers in the workflow and so cut borrowing costs.\n\n\nFor investors, the program is at least as convenient as other online financial\n\nMention: \"Registry of Scripless securities\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"A securities account in the investor's name is opened here when the account is created.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002643:30:0", "occurrence_id": "363c69f7c7022910e6af15c0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Efforts to address e-security should consider two tradeoffs. The first is between convenience\nand risk: as a system becomes more convenient to use, it may become more vulnerable to risks. The\nsecond tradeoff is between costs and risks: e-security measures using the latest technology may be\nextremely costly, especially for emerging markets. Once a system’s level of risk has been determined,\nsustainable resources must be provided to cover it. Among the most crucial is sufficient expertise in esecurity software and hardware. E-security is advancing quickly—both in terms of how systems can\nbe attacked and how they can be protected from such attacks. External experts should be hired to\nassess a system’s vulnerabilities, as should in-house security officers at the same level as chief\naccountants and comptrollers.\n\n\n**How Does the Expanded Program Work?**\n\n\nTo purchase securities through the Expanded Small Investors Program, an investor must have a\nchecking account with a participating bank. (Because the Philippines does not have an automated\nclearing house system, these banks settle transactions under the program.) To open an account, an\ninvestor must visit a bank branch and physically and electronically sign the account opening\nstatement. For each account a securities account is created in the Registry of Scripless securities in the\ninvestor’s name.\n\n\nThe investor can then buy securities over the Internet. Purchases are simultaneously debited\nfrom the investor’s bank account and credited to the account in the Registry of Scripless Securities.\nInterest and principal payments are credited to the bank account when due. The system does not allow\ninvestors to sell securities before they mature.\n\n\nThis system has different implications for different parties. The Department of Finance, given\nits difficult fiscal position and heavy dependence on domestic borrowing, would be predisposed to\nfavor initiatives that electronically issue securities directly to investors. This approach could eliminate\nlayers in the workflow and so cut borrowing costs.\n\n\nFor investors, the program is at least as convenient as other online financial\n\nMention: \"Registry of Scripless Securities\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Internet purchases are credited to accounts held in this registry.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002645:15:0", "occurrence_id": "999d603c53382eaef6a4c6b3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: the adoption of less-efficiency-oriented regulatory regimes. Second, guarantees and various\ntypes of insurance can reduce the overall risks of projects, and hence reduce the risk of equity\nflight from the sector (Erhardt and Irwin, 2003). Third, the level of debt can be limited by\nother instruments, such as leasing agreements and allowing specific arrangements with\ninfrastructure vendors. In some business lines, creative business development will allow\nimprovements, but in most regulated industries facing low-income wage earners as their main\nclient, cost control of some type is likely to be the solution. The main question that remains\nto be solved is then the extent to which cost cutting and creative business development appear\nthe best ways to improve the operators’ capability to finance their investments and debts.\n\n\n**Figure 1: Average book-to-market ratio for the companies in privatised infrastructure sectors**\n\n\n2.0\n\n\n1.5\n\n\n1.0\n\n\n0.5\n\n\n0.0\n\n1993 1994 1995 1996 1997 1998 1999 2000 2001 2002\n\n\nTransport service Transport infrastructure Electricity Gas Water\n\n\nSource: Datastream, authors’ calculations.\n\n\nOverall, these suggestions do not eliminate the fundamental problem that these\nsectors still need to find the necessary levels of finance to deliver on their investment\ncommitments. Ineffectiveness in reducing the equity flight from the sector is likely to reduce\nthe speed at which countries will be able to rely as much as they had hoped for on the private\nsector to help them finance their investment needs. But there is also a macroeconomic reason\nfor concern. It is not unreasonable to expect that for some countries foreign participation is\nhighly concentrated in these sectors, and increased international debt levels resulting from a\nswitch from equity in public services may have balance of payment effects.\n\n\nThese concerns do not imply that regulators should regulate the financial structure of\nthe company, but it certainly implies that it may be important for regulators to better monitor\nthe leverage rates and their evolution to minimize the risks of unexpected shocks. It will also\nrequire a much\n\nMention: \"Datastream\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Sectoral investment series by infrastructure industry are plotted from this source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002650:28:0", "occurrence_id": "083befb3d51b61447f0e42bb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: On the contrary, the brain drain is strongly underestimated in countries such\nas the Gambia, Madagascar, Gabon, Brunei, Papua New Guinea. These countries\nsend a relatively small number of emigrants to non-US areas. They are usually not\nidenti…ed in the OECD migration statistics. Consequently, their emigration rates are\nlargely underestimated.\n\nThe last …gure (…g 3.C) represents the percentage of emigrants on the basis of\nwhich educational attainment is extrapolated (let us call this proportion a ”reliability rate”). On average, our method is based on 92.7 percent of observations. A method\nrelying on US educational attainment is based on 44.1 percent of observations. However, the reliability rate is heterogenous bewteen countries. Figure 3.C represents the\nnumber of reliable country estimates as a function of the reliability criterion required\nby any user of the database. The function is obviously decreasing: the number of\nreliable observations decreases with the degree of reliability imposed by the user. For\nexample, setting the reliability rate at 80 percent, our database provides 175 reliable\ncountry observations in 2000 (160 reliable observations in 2000). Transposing the\nUS immigration structure to the total stock of immigrants, we would only provide\n23 reliable observations in 2000 (22 countries in 1990) at the same reliability rate.\nBy extending the number of observations and their degree of reliability, our method\nprovides useful material to examine the changes in the international distribution of\nmigration rates, to test for the (push and pull) determinants of migration per skill\ngroup or to evaluate the macroeconomic consequences of migration on source and\ndestination countries.\n\n### 4 Key indicators\n\n\nLet us summarize the results by a few …gures.\n\nImmigration structure in OECD areas. The intensity of the brain drain\nclearly depends on the structure of immigration in OECD countries. Figure 4 describes that structure. In …gure 4.A, it clearly appears that the US is the\n\nMention: \"OECD migration statistics\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Countries it fails to identify have largely underestimated emigration rates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002652:30:0", "occurrence_id": "8f4bb5bba01081d9d3edc4ae", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The paper, in comparing the quantitative data with that gathered from a large-scale\n\n\nethnographic World Bank study on conflict, has also hinted at some of the weaknesses of the\n\n\nPODES data. In particular, the comparison would suggest that PODES significantly under\n\nreports levels of conflict (both violent and not), and that this is true across the archipelago. The\n\n\npaper sets out two reasons for this: unclear definitions of conflict, and lack of incentives for\n\n\nrespondents to be honest about conflict levels. We argued that the key question with regard to\n\n\nthe usefulness of the data is whether under-reporting is biased (i.e. levels of under-reporting\n\n\ndiffer across locations) or whether it is systemic, with levels of underreporting similar across\n\n\nsites. We suggested that the latter is more likely to be the case, but that more research (perhaps\n\n\nby comparing different data sources on more areas) would need to be conducted to reach firm\n\n\nconclusions.\n\n\nWe conducted a quantitative analysis to find out which locally-specific factors are\n\n\ncorrelated with the presence of conflict. We found that poverty by itself has very little\n\n\ncorrelation with conflict. Changes in economic conditions, on the other hand do.\n\n\nUnemployment is universally closely associated with higher conflict rates; high in- and outward\n\n\nmigration for work/school have similar positive associations with conflict. Our land variables\n\n\nindicate that the share of converted land in rural areas is associated with higher levels of conflict;\n\n\nthe fraction of titled land has no correlation with conflict. The presence of a natural disaster is\n\n\nalso clearly associated with higher conflict.\n\n\nMuch of the headline conflict in Indonesia is reported as clashes between ethnic groups.\n\n\nOnly a small proportion (3.3%) of conflict reported in PODES was described as being ‘ethnic’.\n\n\nThe 2000 Census for the first time collected self-reported ethnicity and we used this data to\n\n\ninvestigate the correlation between conflict and ethnicity. We find no evidence that increased\n\n\n31\n\nMention: \"PODES data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Compared with ethnographic material, it significantly under-reports conflict.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002652:30:0", "occurrence_id": "10067ea206f74ec38419d8dd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The paper, in comparing the quantitative data with that gathered from a large-scale\n\n\nethnographic World Bank study on conflict, has also hinted at some of the weaknesses of the\n\n\nPODES data. In particular, the comparison would suggest that PODES significantly under\n\nreports levels of conflict (both violent and not), and that this is true across the archipelago. The\n\n\npaper sets out two reasons for this: unclear definitions of conflict, and lack of incentives for\n\n\nrespondents to be honest about conflict levels. We argued that the key question with regard to\n\n\nthe usefulness of the data is whether under-reporting is biased (i.e. levels of under-reporting\n\n\ndiffer across locations) or whether it is systemic, with levels of underreporting similar across\n\n\nsites. We suggested that the latter is more likely to be the case, but that more research (perhaps\n\n\nby comparing different data sources on more areas) would need to be conducted to reach firm\n\n\nconclusions.\n\n\nWe conducted a quantitative analysis to find out which locally-specific factors are\n\n\ncorrelated with the presence of conflict. We found that poverty by itself has very little\n\n\ncorrelation with conflict. Changes in economic conditions, on the other hand do.\n\n\nUnemployment is universally closely associated with higher conflict rates; high in- and outward\n\n\nmigration for work/school have similar positive associations with conflict. Our land variables\n\n\nindicate that the share of converted land in rural areas is associated with higher levels of conflict;\n\n\nthe fraction of titled land has no correlation with conflict. The presence of a natural disaster is\n\n\nalso clearly associated with higher conflict.\n\n\nMuch of the headline conflict in Indonesia is reported as clashes between ethnic groups.\n\n\nOnly a small proportion (3.3%) of conflict reported in PODES was described as being ‘ethnic’.\n\n\nThe 2000 Census for the first time collected self-reported ethnicity and we used this data to\n\n\ninvestigate the correlation between conflict and ethnicity. We find no evidence that increased\n\n\n31\n\nMention: \"2000 Census\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"First self-reported ethnicity collection here is correlated with conflict.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002653:34:0", "occurrence_id": "c22d0407b408e176648e24ec", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 9: Steel Prices**\n**($U.S. per metric ton)**\n\n\n\n\n\n\n\nAverage 12 month moving average price for beams, plate, hot-rolled, cold-rolled and rebar.\nSource WSD Pricetrack.\n\n\n35\n\nMention: \"WSD Pricetrack\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Twelve-month moving averages for beams, plate, hot- and cold-rolled and rebar are drawn from it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002661:9:0", "occurrence_id": "f96e5b72e8f24313c1e75d34", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **F igure 1**\n**A djusted W eekly H ours W orked**\n\n50.0\n\n\n45.0\n\n\n40.0\n\n\n35.0\n\n\n30.0\n\n\n25.0\n\n\n20.0\n\n\n15.0\n\n\n10.0\n\n\n5.0\n\n\n0.0\n\n\nBased on ENIGH, the main and secondary occupations listed on Table 4 show hours-worked, adjusted\n\n\nhours-worked and the total number of hours-worked by region. Figure 2 shows that hours-worked vary across\n\n\nschool levels, sectors and regions. Public primary school teachers work an average of 32 hours per week,\n\n\nwhile those in private schools work on an average of 34 hours. Public and private primary school teachers\n\n\nwork significantly less than their counterparts. This pattern changes in lower and upper secondary school\n\n\nlevel, since public school teachers work more hours on average. On the other hand, _university teachers in_\n\n\n_public schools_ and _the mixed-skilled group_ work an average of 47 hours per week, while people employed in\n\n\n_the_ _agricultural group_ or _the low skilled group_ work on an average of 43 hours per week. There is not a\n\n\nsignificant difference in the weekly hours-worked between public school teachers in urban areas and those\n\n\nteachers in the rural areas. In sum, total hours-worked for _the Professionals with a B.A in Education not_\n\n\n_teaching, the mixed-skilled group_, _the agricultural group_ and _the_ _low-skilled group_ is higher than the total\n\n\nhours-worked for teachers of basic education. This result also holds true in urban and rural regions.\n\n\n10\n\nMention: \"ENIGH\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Main and secondary occupation hours are adjusted and totalled by region.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002663:9:1", "occurrence_id": "72989b3288fbe6102276011e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .9|16.6|\n|Interest income|0.0|15.4|13.4|13.5|17.2|17.7|18.9|\n|Commercial bank|2.9|16.2|12.8|12.6|10.8|12.6|10.7|\n\n\n\nSource: Compilied from CDF statistics (2003)\n\n\nMoreover, out of BRAC’s total $160 million expenditure on development programmes in 2002,\n\n\nmore than 80 percent was financed from its own resources, through the interest income on\n\n\nmicrocredit as well as surplus from its commercial enterprises. Two facets of these trends are\n\n\nworth highlighting.\n\n\nFirst, the decisions to subsidize these operations were not free from controversy. The advocates\n\n\nfor funding these loan funds had to argue their case with officials within their own agencies who\n\n\nbelieved that the capital base for loan operations ought to be enhanced only by savings\n\n\nmobilization or borrowing from commercial sources. In retrospect, these decisions to help\n\n\ncontribute to MFIs loan funds were by and large correct as almost all of the MFIs that received\n\n\nthis support have either reached financial self-sufficiency or are well on their way to doing so.\n\n\nDonors also invested in organizational systems and MFI staff training in order to strengthen the\n\n\ncapacity to administer these growing programs.\n\n\nSecond, large NGOs in particular have been reasonably successful in “managing donors”.\n\n\nInvesting in the capital base of expanding MFIs with a strategic sustainability plan required\n\n\n6 The Department for International Development (DFID) has been one of the largest of the bilateral donors\nhaving provided around $130 million over a twenty year period to MFIs in Bangladesh.\n\n\n9\n\nMention: \"CDF statistics\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Interest income and commercial bank financing shares are compiled from these statistics.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002665:12:1", "occurrence_id": "35210a5a497a728b0d49b7a1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 92 1.70 -0.30 -0.07 -3.25\n1991-1995 1.72 0.15 0.08 0.02 1.47\n<u>1996-2000</u> <u>3.30</u> <u>0.56</u> <u>-0.18</u> <u>1.08</u> <u>1.84</u>\nSource: Data from IMF, World bank databases, Gabon Ministry of Finance\nDefinitions from McCarthy, Neary, and Zanalda (1994), Chami (1999)\n1/ Positive values means an unfavorable terms of trade shock, while a negative value means a favorable shock.\n2/ Negative values indicate the imports are higher than expected.\n3/ Positive values show that imports would have been higher under the\nasumption that real GDP had grown at its estimated expected rate.\n4/ Positive value implies diversification\n5/ Residual term that measures external borrowing or drawing down on reserves.\n\n\nIn the third phase, from 1991 to 1995, adverse terms of trade shocks of close to 9 percent\nof GDP affected the Gabonese economy. The response to this shock was a slight increase in\nimport intensity of close to 1 percent of GDP, an increase in economic compression of half a\npercent of GDP, and an insignificant increase in export promotion. What was needed was a\ncombination of expenditure switching to alter relative prices and shift demand and resources\ntoward the production of tradable goods and expenditure-reducing policies to reduce domestic\nabsorption and increase domestic savings. Due to Gabon’s membership in the CEMAC Zone,\nexchange rate adjustment was not an option, although devaluation was finally implemented only\nin 1994. As in other countries in the subregion, the internal adjustment strategy was a failure. The\nsituation was difficult among the six countries in the CEMAC zone, whose regional average per\ncapita GDP’s fell by 3.\n\nMention: \"World bank databases\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"One of three sources behind the terms of trade series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002674:27:1", "occurrence_id": "33227634f76f5a3eec25f9fa", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  in 1985 (OECD\n1987:44). Low pensions were raised considerably, especially pensions to farmers, while\ncoverage of pension rights was also extended to returning immigrants with no record of\ncontributions to social security programs. Another source of increased government spending was\nsupport to loss-making public corporations. The combined deficit of these entities rose from 2 to\n5.5 percent of GNP between 1979 and 1985. In addition, many private sector firms were unable\nto meet the competitive challenge of European firms that followed the dismantling of tariffs\n(Giannitsis 1993). Many of these firms were declared “problematic” and relied on financial\nsupport from state-controlled banks for their continuing survival.\n\n\nThe effect of these policies was a massive increase in the public sector borrowing\nrequirement that reached 17.7 percent of GNP by 1985. The level of general government debt\nrose from 39 percent of GDP in 1980 to 73 percent in 1985. <sup>14</sup>\n\n\nExternal debt grew from 16 percent of total debt in 1980 to 31 percent in 1985. Most of\nthe external debt was of a long-term nature, mostly syndicated loans from international banks,\nwith little exposure to interest rate and refinancing risk. However, being denominated in foreign\ncurrency, it was exposed to foreign currency risk and suffered from the periodic devaluations of\n\n14 Data from the 1987 OECD survey show that on a net basis the general government debt rose from 27\npercent of GNP in 1979 to 63 percent in 1985. Adding the debt of public corporations brought the total to 85\npercent of GNP, which was one of the highest ratios in the OECD area. Particularly disquieting was the speed at\nwhich the total public debt had grown, having soared in relation to GNP by nearly 55 percentage points between\n1979 and 1985.\n\n\n25\n\nMention: \"the 1987 OECD survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It records general government net debt rising from 27 to 63 percent of GNP.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002675:21:0", "occurrence_id": "a6743a7235f009b70f022499", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: multilateral development agencies to make a difference. It is hopeful that this opportunity will not\nbe lost, provided that investment climate and governance continue to improve, and capacity for\nregulation and supervision continue to enhance.\n\n### **7. Conclusions**\n\n\nThe purpose of this paper is to investigate the relationship between the efficiency of\ndomestic banks and foreign bank entry, and also to study possible role of the order of financial\nliberalization in this process. We focus on the banking sector of 30 countries for the period of\n1995-2002. The BANKSCOPE database is our main data source.\n\n\nThe first issue that we investigate is how foreign bank openness differs from one country\nto another and whether foreign bank entry is related to the level of economic development. We\nalso try to examine whether changes in the foreign bank share are significant determinant of\ndomestic banks’ performance indicators such as the net interest margin, non-interest income,\nprofits, overhead costs, and loan loss reserves. Finally, we check how these results change when\ncountries are grouped according to the sequence of their financial liberalization. Throughout our\nanalyses, we control for other bank variables and macroeconomic indicators.\n\n\nThe descriptive statistics indicate that there is no specific relationship between the level\nof economic development and openness to foreign banks. When the foreign bank share is\ninvestigated at the country level, it can be seen that this share changes a lot among countries.\nWhen countries are grouped according to their sequence of financial liberalization such as\ndomestic financial markets first, stock market first, or capital account first, it is observed that\ndomestic banks’ net interest margin, non-interest income, overhead costs, and loan loss reserve\nget the lowest values in the countries liberalized their stock market first. This indicates that\ncompetition is well-established in these countries. The highest values of non-interest income,\noverhead costs, and loan loss reserves are in the Asian and Latin American countries which\nliberalized their domestic financial markets first. Despite differences in the performance\n\nMention: \"The BANKSCOPE database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Banking sector data for thirty countries over this period are drawn from it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002677:32:0", "occurrence_id": "9a65518be286c3278ceea32e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **<u>Table 1. Data Summary</u>**\n\n\n<u>Ecuador</u> <u>Madagascar</u> <u>Cambodia</u>\n\n\n**Household Survey**\n\n\nYear 1994 1993-4 1997\n\n\n\nSource Encuesta de\nCondiciones de Vida\n(ECV)\n\n\n\nSource Encuesta de Enquête Permanente Auprès Cambodia SocioCondiciones de Vida des Ménages (EPM) Economic Survey\n(ECV) (CSES)\n\nSample Size 4,500 Households 4,508 Households 6010\n\n\n\nEnquête Permanente Auprès\ndes Ménages (EPM)\n\n\n\nReferences Hentschel and\nLanjouw (1996); and\nHentschel, Lanjouw,\nLanjouw and Poggi\n(2000)\n\n\n**Population Census**\n\n\n\nMistiaen, Özler,\nRazafimanantena and\nRazafindravonona (2002)\n\n\n\nFujii (2003)\n\n\n\n1990 1993 1998\nYear\n\n\n\nCoverage About 10 million\nindividuals in 2 million\n<u>households</u>\n\n\n\nabout 11.9 million\nindividuals\n<u>in 2.4 million households</u>\n\n\n\nAbout 11.0 million\nindividuals in\n<u>2.15 million households</u>\n\n\n\n33\n\nMention: \"Encuesta de\nCondiciones de Vida\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Four thousand five hundred households form the Ecuadorian sample.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002677:32:0", "occurrence_id": "93c8b6b2a8232ea88d254d40", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **<u>Table 1. Data Summary</u>**\n\n\n<u>Ecuador</u> <u>Madagascar</u> <u>Cambodia</u>\n\n\n**Household Survey**\n\n\nYear 1994 1993-4 1997\n\n\n\nSource Encuesta de\nCondiciones de Vida\n(ECV)\n\n\n\nSource Encuesta de Enquête Permanente Auprès Cambodia SocioCondiciones de Vida des Ménages (EPM) Economic Survey\n(ECV) (CSES)\n\nSample Size 4,500 Households 4,508 Households 6010\n\n\n\nEnquête Permanente Auprès\ndes Ménages (EPM)\n\n\n\nReferences Hentschel and\nLanjouw (1996); and\nHentschel, Lanjouw,\nLanjouw and Poggi\n(2000)\n\n\n**Population Census**\n\n\n\nMistiaen, Özler,\nRazafimanantena and\nRazafindravonona (2002)\n\n\n\nFujii (2003)\n\n\n\n1990 1993 1998\nYear\n\n\n\nCoverage About 10 million\nindividuals in 2 million\n<u>households</u>\n\n\n\nabout 11.9 million\nindividuals\n<u>in 2.4 million households</u>\n\n\n\nAbout 11.0 million\nindividuals in\n<u>2.15 million households</u>\n\n\n\n33\n\nMention: \"Encuesta de Enquête Permanente Auprès Cambodia\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its 4,508 households give the Malagasy consumption estimates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002677:32:0", "occurrence_id": "a53918e793c8c1995a10b525", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **<u>Table 1. Data Summary</u>**\n\n\n<u>Ecuador</u> <u>Madagascar</u> <u>Cambodia</u>\n\n\n**Household Survey**\n\n\nYear 1994 1993-4 1997\n\n\n\nSource Encuesta de\nCondiciones de Vida\n(ECV)\n\n\n\nSource Encuesta de Enquête Permanente Auprès Cambodia SocioCondiciones de Vida des Ménages (EPM) Economic Survey\n(ECV) (CSES)\n\nSample Size 4,500 Households 4,508 Households 6010\n\n\n\nEnquête Permanente Auprès\ndes Ménages (EPM)\n\n\n\nReferences Hentschel and\nLanjouw (1996); and\nHentschel, Lanjouw,\nLanjouw and Poggi\n(2000)\n\n\n**Population Census**\n\n\n\nMistiaen, Özler,\nRazafimanantena and\nRazafindravonona (2002)\n\n\n\nFujii (2003)\n\n\n\n1990 1993 1998\nYear\n\n\n\nCoverage About 10 million\nindividuals in 2 million\n<u>households</u>\n\n\n\nabout 11.9 million\nindividuals\n<u>in 2.4 million households</u>\n\n\n\nAbout 11.0 million\nindividuals in\n<u>2.15 million households</u>\n\n\n\n33\n\nMention: \"Economic Survey\n(ECV)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Six thousand ten households supply the Cambodian welfare data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002677:32:0", "occurrence_id": "1752528ca51c03603b244d1b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **<u>Table 1. Data Summary</u>**\n\n\n<u>Ecuador</u> <u>Madagascar</u> <u>Cambodia</u>\n\n\n**Household Survey**\n\n\nYear 1994 1993-4 1997\n\n\n\nSource Encuesta de\nCondiciones de Vida\n(ECV)\n\n\n\nSource Encuesta de Enquête Permanente Auprès Cambodia SocioCondiciones de Vida des Ménages (EPM) Economic Survey\n(ECV) (CSES)\n\nSample Size 4,500 Households 4,508 Households 6010\n\n\n\nEnquête Permanente Auprès\ndes Ménages (EPM)\n\n\n\nReferences Hentschel and\nLanjouw (1996); and\nHentschel, Lanjouw,\nLanjouw and Poggi\n(2000)\n\n\n**Population Census**\n\n\n\nMistiaen, Özler,\nRazafimanantena and\nRazafindravonona (2002)\n\n\n\nFujii (2003)\n\n\n\n1990 1993 1998\nYear\n\n\n\nCoverage About 10 million\nindividuals in 2 million\n<u>households</u>\n\n\n\nabout 11.9 million\nindividuals\n<u>in 2.4 million households</u>\n\n\n\nAbout 11.0 million\nindividuals in\n<u>2.15 million households</u>\n\n\n\n33\n\nMention: \"Enquête Permanente Auprès\ndes Ménages\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Listed in the data summary with references describing its design.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002683:25:0", "occurrence_id": "b441919947b59296395d6021", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Table 5A: Impact of SOFES on jobs on the side, Bachelor students, SOFES data**\nN <RD >RD <u>PROBIT</u>\nDS=0.08 (1) (2) (3) (4) (5) (6)\n±0.005 253 0.237 0.226 -0.010 -0.011\n(0.056) (0.057)\n\n\n\n±0.005 253 0.237 0.226 -0.010 -0.011 -0.030\n(0.056) (0.057) (0.054)\n\n±0.010 733 0.238 0.252 0.015 0.047 0.046\n(0.032) (0.034) (0.034)\n\n\n\n-0.011\n(0.057)\n\n0.047\n(0.034)\n\n0.018\n(0.028)\n\n0.026\n(0.025)\n\n\n\n±0.010 733 0.238 0.252 0.015 0.047 0.046\n(0.032) (0.034) (0.034)\n\n±0.015 1,051 0.249 0.235 -0.014 0.018 0.013\n(0.026) (0.028) (0.028)\n\n\n\n±0.015 1,051 0.249 0.235 -0.014 0.018 0.013\n(0.026) (0.028) (0.028)\n\n±0.020 1,292 0.245 0.246 0.001 0.026 0.023\n(0.024) (0.025) (0.025)\n\n\n\n±0.020 1,292 0.245 0.246 0.001 0.026 0.023\n(0.024) (0.025) (0.025)\n\n\nDS=0\n\nMention: \"SOFES data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Probit estimates of the effect on side jobs for bachelor students are computed here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002673:33:0", "occurrence_id": "d8ad2360f5fee4db880578b2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: estate values could more than compensate for the loss of stamp duties due to a substantial\nreduction in rates. Only in those scenarios that are shaded would the black economy tax effect\nbe less than total stamp tax collections.\n\n\nWithin this broad range of scenarios, it is possible to assess which parameter values are\nthe more plausible by relating these computations to other estimates of the percentage of GDP in\nthe “black” economy. One can work backward through the basic equations to compute the total\nvalue of black GDP driven by underdeclaration in real estate. The share of stamp duties in GDP\nis roughly 0.5 percent. <sup>39</sup> [^39: See Indian Public Finance Statistics 2001-2002, Table 1.2] If the velocity of black money is 5, underdeclaration is 40 percent, and\nthe stamp tax rate is 10 percent <sup>40</sup> [^40: The weighted average stamp tax rate for the sixteen major states is 9.7 percent as of 2001.], then the implied black economy share of GDP from real estate\nis about 17 percent. Recent estimates show about 23 percent of GDP is in the black economy in\nIndia. <sup>41</sup> [^41: See Friedrich Schneider (2002), “Size and Measurement of the Informal Economy in 110 Countries Around the\nWorld”, paper presented at the Workshop of the Australian Tax Centre, ANU, Canberra, Australia, July 17, 2002.] Earlier work by Acharya et al. (1986) on the black economy suggests a figure of about\n18 to 21 percent for 1980-1981. <sup>42</sup> [^42: See Shankar Acharya et al. (1986), _Aspects of the Black Economy in India_, National Institute of Public Finance\nand Policy (New Delhi, India).] Together these figures strongly suggest that the parameter\nvalues we have chosen are not outside the bounds of plausibility.\n\n**E. Stamp Duty Rates\n\nMention: \"Indian Public Finance Statistics\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Table 1.2 supplies the stamp duty share of GDP.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002699:14:0", "occurrence_id": "5f9f99c0da3a6bc8b8f01e16", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: French company law recognizes two main types of companies – commercial and civil.\n\n\nCommercial companies are of three types: unlimited liability companies, those where the\n\n\nliability of the shareholders is limited to their contribution to the capital of the company\n\n\nand those which comprise mixed shareholders- that is those who are personally liable and\n\n\nthose shareholders whose liability is limited. A civil company may not engage in any\n\n\ncommercial activity.\n\n\nA simple comparison of the some of the corporate laws in different EU countries\n\n\nreveal differences in tax regulations. The corporate tax rate is 25% in Germany (though\n\n\nsome types of companies have a special status), in the United Kingdom (UK) there is a\n\n\ntwo-tier tax rate with corporations whose profits are less than GBP 1.5 million paying\n\n\n19% and the others 30%. The corporate tax rate is around 35% in the Netherlands\n\n\n(Richard, ed., (1992- 2004)).\n\n\nThe procedures required to start a business also vary. In France, minimum capital\n\n\nrequirements amount to 7,500 Euros for a private limited liability company. In Germany,\n\n\nthe minimum capital requirement for the same is 25,000 Euros and in the UK there is no\n\n\nminimum capital requirement. To register and clear the proposed name of the company\n\n\nrequires eight days and 38 Euros in France, and in Germany it requires one day and there\n\n\nis no fee. To register new firms, there are altogether 10 distinct procedural/legal\n\n\nrequirements in France, requiring over 53 days to complete at a total cost of 704 Euros.\n\n\nIn the UK, the comparable numbers (at time of writing) are 6, 18 and $US 264, and for\n\n\nGermany, 9 steps, 45 days and 1,425 euros. <sup>4</sup>\n\n\n4 Source: Doing Business Database, The World Bank 2004.\n\n\n15\n\nMention: \"Doing Business Database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Steps, days and cost to start a business are taken from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002704:16:0", "occurrence_id": "f14766285ccae1ebf3b4fe61", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: - While there is ample evidence of growing intra-industry trade <sup>14</sup> in the Czech\nRepublic, Slovenia and Hungary, Latvia’s performance is near the bottom of\nall NMS’s. This suggests that Latvia’s industries and enterprises are not well\nintegrated with industries and enterprises in other EU countries. In other\nwords, they are not part of EU value chains. Compared to the early period of\ntransition (and even more so with the pre-transition period), intra-industry\ntrade between the more advanced NMS’s (the Czech and Slovak Republics,\nHungary and Poland) and the EU has steadily increased whereas Latvia is\nclearly an outsider. Most of the countries that have relatively low and stable\nintra-industry manufacturing trade have the largest share of non-manufactured\ngoods in total exports. In Latvia’s case, the low share of intra-industry trade\nsuggests that a high proportion of country’s manufactured exports consist of\nrelatively simple transformations of raw materials, processes that are not\nsuited to a division of labor and activities across different countries. The low\nshare of intra-industry trade also suggests that Latvian enterprises do not have\nthe skills required to become suppliers to foreign companies and find a higher\nvalue added niche in global value chains.\n\n\n**Figure 8 Indicators of intra-industry trade with the EU (15)**\n\n\n\n\n\n\n\n\n\n\n\n_Sources_ : Eurostat COMEXT Database, WIIW Industrial Database, own calculations.\n\n\n14 The ‘new’ trade theory suggests that intra-industry trade, measured by the share of simultaneous export\nand import of similar products, is motivated by product differentiation and economies of scale. Intraindustry trade is measured by Grubel-Lloyd index (GL=1-�ABS(xij+mij), where xij and mij are country\ni's export and import of NACE 3 digit sector j, respectively).\n\n\n17\n\nMention: \"Eurostat COMEXT Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Bilateral product flows behind the intra-industry trade indicators come from it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002704:16:0", "occurrence_id": "aef76116b9dc17170103c5e1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: - While there is ample evidence of growing intra-industry trade <sup>14</sup> in the Czech\nRepublic, Slovenia and Hungary, Latvia’s performance is near the bottom of\nall NMS’s. This suggests that Latvia’s industries and enterprises are not well\nintegrated with industries and enterprises in other EU countries. In other\nwords, they are not part of EU value chains. Compared to the early period of\ntransition (and even more so with the pre-transition period), intra-industry\ntrade between the more advanced NMS’s (the Czech and Slovak Republics,\nHungary and Poland) and the EU has steadily increased whereas Latvia is\nclearly an outsider. Most of the countries that have relatively low and stable\nintra-industry manufacturing trade have the largest share of non-manufactured\ngoods in total exports. In Latvia’s case, the low share of intra-industry trade\nsuggests that a high proportion of country’s manufactured exports consist of\nrelatively simple transformations of raw materials, processes that are not\nsuited to a division of labor and activities across different countries. The low\nshare of intra-industry trade also suggests that Latvian enterprises do not have\nthe skills required to become suppliers to foreign companies and find a higher\nvalue added niche in global value chains.\n\n\n**Figure 8 Indicators of intra-industry trade with the EU (15)**\n\n\n\n\n\n\n\n\n\n\n\n_Sources_ : Eurostat COMEXT Database, WIIW Industrial Database, own calculations.\n\n\n14 The ‘new’ trade theory suggests that intra-industry trade, measured by the share of simultaneous export\nand import of similar products, is motivated by product differentiation and economies of scale. Intraindustry trade is measured by Grubel-Lloyd index (GL=1-�ABS(xij+mij), where xij and mij are country\ni's export and import of NACE 3 digit sector j, respectively).\n\n\n17\n\nMention: \"WIIW Industrial Database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It supplies the sectoral detail for the same figure.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002705:10:0", "occurrence_id": "ac8f79e880292677127a25b9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: e¤ects of a change in the level of the standard, respectively, on the number of …rms and\n\n\n…rms’output. Assessing the relative strengths of these two e¤ects helps us determine the\n\n\noverall impact on imports of any regional initiatives that deal with standards.\n\n\nWe are particularly interested in exploring the possible asymmetric e¤ects of regional\n\n\nagreements on standards in a heterogeneous world. We assume that the world consists\n\n\nof two types of countries: type K and type L, which di¤er in terms of their …rms’costs\n\n\nof complying with standards, i.e. ci and Fi. Two broad cases are possible: (i) cK < cL\n\nand FK - FL or (ii) cK < cL and FK 6 FL. Our analysis focuses on case (i) for several\n\n\nreasons. First of all, case (i) is more analytically challenging, and once the implications\n\n\nin this case are established, those in case (ii) can be worked out quite straightforwardly.\n\n\nMore importantly, case (i) may more accurately depict situations where …rms from both\n\n\nrich and poor countries produce the same product. It is likely that …rms from richer\n\n\ncapital-abundant countries (type K) tend to adopt a more capital-intensive technology\n\n\nthat requires a relatively large Fi, while …rms from poorer labor-abundant countries (type\n\n\nL) tend to adopt a more labor-intensive technology that requires a relatively large ci. In\n\n\nfact, the World Bank’s (2004) Technical Barriers to Trade Survey reveals that a …rm’s\n\n\npercentage of compliance cost spent on product re-design for each export market (cor\n\nresponding to Fi in our paper) is positively correlated with the GDP per capita of the\n\n\ncountry in which it is located, whereas that spent on hiring additional labor for production\n\n\nand testing (corresponding to ci in our\n\nMention: \"Technical Barriers to Trade Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Firms' compliance spending on product re-design per export market is reported here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002709:10:1", "occurrence_id": "e102a7d29d0e184e5b3ed05b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  one might have expected a positive\nrelation between the level of mandated labor protection and income across countries (i.e.\nlabor protection is a normal good), the relationship is in fact weakly negative across our\nsample of countries.\n\n\n\n**Table 1 The synthetic indicators of the stringency of employment protection for**\n**<u>regular and temporary employment</u>**\nCountry (by region) Overall Rank Regular Rank Temporary Rank\nemployment employment employment\n\n\n\n<u>index</u>\n\n\n\nRank Temporary\n\n\n\nRank\n\n\n\nemployment\n\n\n\n<u>index</u>\n\n\n\nRank Regular\nemployment\n<u>protection index</u>\n\n\n\n_Sub-Saharan Africa (SAS)_\nBotswana 0.20 24 0.40 44 0 1\nCameroon 0.67 75 0.67 80 0.67 60\nCote d'Ivoire 0.53 55 0.22 12 0.83 70\nEthiopia 0.57 63 0.40 44 0.75 63\n<u>Ghana</u> <u>0.18</u> <u>17</u> <u>0.35</u> <u>31</u> <u>0</u> <u>1</u>\n\n\n11 In order to check the quality of our index, we compare it to the new indicator derived by the\nOECD (2004). The comparison can be made for 15 industrial and transition countries: Canada, Czech\nRepublic, France, Germany, Hungary, Italy, Mexico, Poland, Portugal, Slovak Republic, Spain, Sweden,\nTurkey, United Kingdom, and United States. The correlation between our weighted index for overall\nemployment regulation with the relevant overall OECD index is statistically significant at 1 per cent level\n(correlation coefficient= 0.82).\n\n\n10\n\nMention: \"overall OECD index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"The weighted employment regulation index correlates 0.82 with it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002713:31:1", "occurrence_id": "972111822c5766cf0b56d074", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .5 percent of income is spent on water services. The overall budget share is\nrelatively constant across consumption quintiles, although richer households spend less on cooking fuels\nand more on telecommunications. Only a tiny fraction of the poorest households have access to\ntelephones, but those that do so spend as much as 5 percent of their income on the service. Such\nbreakdowns, however useful for designing tariff structures to ensure affordability, are seldom available.\n\nvi Brook and Smith (2001).\nvii See Jalan and Ravaillon, 2001; Jayasuriya and Wodon, 2003; Leipziger et al., 2003; World Bank, 2004;\n\nand references cited therein.\nviii Except where otherwise specified, the sources for the data in this section are as follows:\n\nUSAID Demographic and Health Surveys (1997–2001, depending on the country) available at\n\n<u>http://www.measuredhs.com/: access to electricity services, proxy of water service quality, proxy</u>\nof sanitation service quality;\nWorld Bank’s _World Development Indicators_ (1997-2001, depending on the country): technical\n\nquality of electricy services, access to water services, sanitation services, to telecom services, and\ntechnical quality of transport services;\nWorld Economic Forum (2003): perceived quality of water services, perceived quality of electricity\n\nservices, of sanitation services, of ICT services, and of transport services.\nInternational Telecommunication Union: technical quality of telecoms services, access to telecom\n\nservices;\nAngel (2000): quality of housing;\nRoberts (2003): access to transport services.\nix Urban housing alone is worth more than twice that, but problems of double counting make it difficult to\n\nestimate the value of infrastructure stock, including housing (Angel, 2000, p. 288).\n\n\n32\n\nMention: \"USAID Demographic and Health Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Access to electricity and water service proxies are read from these rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002714:41:1", "occurrence_id": "41771592423ff3337d3b63f4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **\n\nFrom Section:\n\n\n1: Size of government\n\n\nBased on: Price Waterhouse, Individual Taxes: A\nWorldwide Summary (various issues)\n\n\n\n**Scale** **Description** **Years** **Source**\n\n\n\n% Total trade union membership, in percent of the total labor 1945- **A Database of Labor-Market Indicators Across**\n\nforce. Includes workers of both sexes in the public and the 2000 **Countries**\n\n\n\nTotal trade union membership, in percent of the total labor\nforce. Includes workers of both sexes in the public and the\nprivate sectors. In some countries, the union membership\nmay include unemployed and retired workers who pay their\ndues. Based on the number of active contributors declared\nby the trade unions themselves and on labor force\nestimates. When declared membership is larger than the\nlabor force, a 100 percent membership rate is reported.\n\n\n\n19452000\n(5 year)\n\n\n\nAvailability of part-time and fixed-term contracts\n\n\nWorking time requirements, including mandatory minimum\ndaily rest, maximum number of hours in a normal workweek,\npremium for overtime work, restrictions on weekly holiday,\nmandatory payment for nonworking days, (which includes\ndays of annual leave with pay and paid time off for\nholidays), and minimum wage legislation. The constitutional\nprinciples dealing with the minimum conditions of\nemployment are also coded.\n\n\n\nSurvey\nconducted\nin 1997,\nupdated\nto 2003\n\n\n\nM. Rama and R. Artecona, The World Bank, 2002.\n\n\n**Doing Business, The World Bank**\nsee Botero, Djankov, La Porta, Lopez-de-Silanes, and\nShleifer, \"The Regulation of Labor\", Working Paper 9756,\nNational Bureau of Economic Research, June 2003\nBased on: NATLEX database (International Labour\nOrganization); Constitutions, available online on the U.S.\nLaw Library of Congress website; International\nEncyclopaedia for Labour Law and Industrial Relations,\nand Social Security Programs Throughout the World\n\nMention: \"Database of Labor-Market Indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Trade union membership as a share of the labour force is taken from this compilation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002720:21:0", "occurrence_id": "050a31a3c242ed0c2cf46430", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: construction process. A check against the well-known Penn World Tables data (see\n\n\n<u>www.nber.org) for a range of countries suggests that physical capital/output ratios were</u>\n\n\ngenerally rising quite rapidly in the 1970-1990 period for which the capital accumulation\n\n\ndata are available. The fact that the growth rates of K/Q and Edn/Q are generally lower\n\n\nthan their growth relative to the labor input does, however, give reason for caution about\n\n\ncommon assumptions, such as Hicks-Neutral technical change in all sectors used by\n\n\nHarrigan (1997) and Kee (2001).\n\n\nDespite the evidence from many different types of empirical studies on the\n\n\npotential role of Rybczynski effects, the coincidence of high rates of accumulation of\n\n\nphysical and human capital over the period, and the rapid shift of developing countries\n\n\ninto exports of manufactures and services, is clearly not definitive evidence of causation.\n\n\nHowever, it is strongly suggestive, and needs to be examined in conjunction with changes\n\n\nin trade policy and in technology.\n\n\n**Protection Policies**\n\n\nProtection policy is frequently advocated as a means to promote industrial development.\n\n\nIt can certainly do this for import competing activities, such as production of consumer\n\n\ngoods. However, this production pattern locks producers into small, and typically slow\n\n\ngrowing, markets for their output. Further, it introduces a major discontinuity. Under a\n\n\nprotectionist policy regime, an exporter must not only have sufficient comparative\n\n\nadvantage to be able to compete in world markets, but must have sufficient advantage to\n\n\nbe able to compete _despite_ the cost increases resulting from protection levied on its\n\n\nintermediate inputs, and the adverse effects of real exchange rate devaluation on its costs\n\n\nfor factors and nontraded goods.\n\n\nDeveloping countries have increasingly come to recognize the adverse impacts of\n\n\nprotection on their export performance. and begun to adjust their policies towards more\n\n\nopen trade regimes. The most profound and far-reaching manifestation of developing\n\n\ncountries’ interest in greater participation in trade is evident\n\nMention: \"Penn World Tables data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Capital-output ratios from the constructed series are checked against it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002723:3:0", "occurrence_id": "8703d23e23b989513edf6457", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Tables:**\n\n\nTable 1: Causes of Corruption: Full Specification………………………………………19\n\n\nTable 2: Corruption in Developing Countries…………………………………………...22\n\n\nTable 3: Decentralization and Corruption: Unitary vs. Federal States…………………..23\n\n\n**Appendices:**\n\n\nAppendix A: Sources for the 1998 Corruption Perceptions Index………………………29\n\n\nAppendix B: Sample Space……………………………………………………………...30\n\n\nAppendix C: Data Sources and Descriptions…………………………………………….31\n\n\nAppendix D: Composite Measures………………………………………………………34\n\nMention: \"1998 Corruption Perceptions Index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Listed in the contents as an appendix documenting the study's sources.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002728:45:5", "occurrence_id": "9b472bb50b9fc5f2be415110", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: up> <sup>elasticity</sup> <sup>faced</sup> <sup>by</sup> <sup>Large</sup> <sup>for</sup> <sup>good</sup> <sup>i.</sup>\n\nδ: Small’s discount factor of time.\nτ <sup>T</sup> : The threat level of the tariff used by Large if Small stops cooperating.\ne˜ <sup>B</sup> [e <sup>B</sup> ]: Equilibrium e <sup>S</sup> under PTAs without [with] a CET.\nT : Total amount of numeraire transfer from Large to Small.\n\n\n\nA.5 Data Sources and De fi nitions of Variables\n\n\n\n∆t: Change in the bound advalorem MFN tariff rates between the pre- Uruguay Round and post-UR\nperiods for the HS 8-digit product i. Source: WTO.\nI <sup>any0</sup> [I <sup>evy0</sup> ]: Indicator variable equal to 1, if good i is imported at a duty-free preferential rate by EU\nunder any [all] of its PTAs in 1994. It excludes PTAs involving a common external tariff. Sources:\nEurostat’s COMEXT (trade fl ows) and UNCTAD’s TRAINS (preferential tariff rates).\nI <sup>any</sup> [I <sup>pos</sup> ]: Indicator equal to 1, if good i is imported by the EU under any of its PTAs at either\na duty-free or positive preferential tariff rate [a positive preferential rate only] in 1994\n\nMention: \"COMEXT\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It provides the trade flows in the preferential tariff indicators.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002728:45:5", "occurrence_id": "5dcd68d0edd3377159fea7cc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: up> <sup>elasticity</sup> <sup>faced</sup> <sup>by</sup> <sup>Large</sup> <sup>for</sup> <sup>good</sup> <sup>i.</sup>\n\nδ: Small’s discount factor of time.\nτ <sup>T</sup> : The threat level of the tariff used by Large if Small stops cooperating.\ne˜ <sup>B</sup> [e <sup>B</sup> ]: Equilibrium e <sup>S</sup> under PTAs without [with] a CET.\nT : Total amount of numeraire transfer from Large to Small.\n\n\n\nA.5 Data Sources and De fi nitions of Variables\n\n\n\n∆t: Change in the bound advalorem MFN tariff rates between the pre- Uruguay Round and post-UR\nperiods for the HS 8-digit product i. Source: WTO.\nI <sup>any0</sup> [I <sup>evy0</sup> ]: Indicator variable equal to 1, if good i is imported at a duty-free preferential rate by EU\nunder any [all] of its PTAs in 1994. It excludes PTAs involving a common external tariff. Sources:\nEurostat’s COMEXT (trade fl ows) and UNCTAD’s TRAINS (preferential tariff rates).\nI <sup>any</sup> [I <sup>pos</sup> ]: Indicator equal to 1, if good i is imported by the EU under any of its PTAs at either\na duty-free or positive preferential tariff rate [a positive preferential rate only] in 1994\n\nMention: \"TRAINS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It provides the preferential tariff rates in the same indicators.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002706:7:1", "occurrence_id": "354b39019ff46ffd37c79b0f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: br>31.2<br>33.1<br>29.9<br>29.3|\n|Source: Encuesta de Hogares por Muestro|Source: Encuesta de Hogares por Muestro|Source: Encuesta de Hogares por Muestro|Source: Encuesta de Hogares por Muestro|Source: Encuesta de Hogares por Muestro|\n\n\n\nFigure 2: Shares of employment for Informal Sector, Employees and Self-employed: 1992-2002\n\n\n80\n\n\n70\n\n\n60\n\n\n\n50\n\n\n40\n\n\n30\n\n\n20\n\n\n10\n\n\n0\n\n\n\n\n\n\n\n1992 1995 1996 1997 1998 1999 2000 2001 2002\n\n\n7\n\nMention: \"Encuesta de Hogares por Muestro\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its household series underlie the tabulated shares across the sample years.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002733:28:0", "occurrence_id": "f3bd01352de54fb5959b29ba", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 20. Unemployed by Gender and Job Search Methods, 1996-2002**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n_Source:_ Calculation based on LFS.\n\n\n**E. RELATIVE WAGES**\n\n\nOne of the most important features of a labor market is its wage structure. Are human capital related\nworkers’ characteristics – education, general, and firm specific experience - rewarded in Latvia in similar\nways as in established market economies? Empirical evidence from other transition countries suggests\nthat returns to education have increased sharply, while returns to experience have decreased compared to\nthe socialist wage setting system. However, the patterns of change and the end result differ across\ncountries, reflecting differences in historical background, social norms, and the industrial structure of the\neconomy.\n\n\nIn the following section, the major factors contributing to wage differences between individuals in Latvia\nare assessed. Particular attention is focused on differences in wages earned by individuals of differing\neducation levels, gender and ethnic wage differentials, and regional differences in wages. Such an\nassessment is of great importance in designing policies which target individuals and regions which may\nhave been disadvantaged in the move to a market-based economy.\n\n\n29\n\nMention: \"LFS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Unemployment by gender and job search method is calculated from it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002738:20:1", "occurrence_id": "40bdb312d64d931674085c23", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: /US$)<br>(5)|Total exposure<br>in dollar terms<br>(US$ bi)<br>(6)|\n|---|---|---|---|---|---|---|\n|December 00|114|-|114|22|2.0|58|\n|December 01|179|-|179|29|2.3|77|\n|March 02|179|-|179|29|2.3|77|\n|May 2002|179|14|193|30|2.5|77|\n|September 02|191|77|268|41|3.9|69|\n|December 02|139|91|230|37|3.5|65|\n|November 03|82|92|174|24|3.0|59|\n\n\n(1), (2), (3), (4) Source; Relatorios da Divida Publica\n(5) Average exchange rate for the month\n(6)=(3)/(5)\n\n\n\n\n\n\n\n\n\n\n#### **III. FISCAL RULES: POLICY ELIMINATED AS SOURCE OF** **INSTABILITY**\n\n\n\n47. The Brazilian experience of 2001-2003 shows that the coordination between fiscal\nand monetary policies is necessary to stabilize the economy. When monetary policy is\nnot able to react to a given shock, then fiscal policy has to be more responsive. There\nmay be circumstances when fiscal events, or expectations of them, are determinants of\nprice changes or of interest rates, and these expectations might not be consistent with the\ncentral bank’s objectives. Functional independence of the central bank, meaning its\nability to guarantee that its actions will influence inflation expectations and keep them\nwithin certain bounds to meet an inflation target, can only exist if fiscal expectations\n\nMention: \"Relatorios da Divida Publica\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Monthly public debt composition figures are taken from these reports.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002739:13:0", "occurrence_id": "52419ccdb79cb184b5056383", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: first firms identified in a 1999 ITU report in a ranking of the main 20 providers according to\n\n\ntheir turnover. It turns out that 8 of these firms are present in both the fixed and the mobile\n\n\nlines business. For fixed lines, the 9 top firms were then: Bell South Corporation, BT Group,\n\n\nDeutsche Telekom, France Telecom, SBC, Telecom Italia, Telefonica de España, Telstra, and\n\n\nVerizon. For mobile, the firms covered are: Bell South Corporation, Deutsche Telekom,\n\n\nFrance Telecom, SBC, Telecom Italia, Telefonica de España, Telstra, Verizon and Vodafone.\n\n\nBT is dropped from the mobile list while Vodaphone is not present in the fixed line list. For\n\n\nfixed lines we consider 35 countries (18 OECD and 17 non OECD) while for mobile, we\n\n\ncover 65 countries (20 OECD and 45 non OECD).\n\n\nA few words of caution on this list may be appropriate. First, there is a fair amount of\n\n\nturnover in ownership in this sector. Some of the firms in the 1999 list have now changed\n\n\nowners (and names). Our choice of operators is thus arbitrarily frozen in 1999 even if the\n\n\ninformation is collected for these companies in their form as of 2003. Second, there is a clear\n\n\nregional specialization of firms. There are only 4 firms with a strong presence in LDCs:\n\n\nFrance Telecom, SBC, Telecom Italia, Telefonica de España. This last one is particularly\n\n\ndominating in Latin America, Telstra in Asia and France Telecoms in Eastern Europe. Third,\n\n\nthe list does not do justice to the growing presence of regional actors in the various regional\n\n\nmarkets with strong South African firms present in Africa and strong Mexican and Brazilian\n\n\noperators present in Latin America for instance. However, most of these firms and many of\n\n\nthose included in the ITU list do not publish the necessary information on their webpages and\n\n\nhence we cannot make a fair assessment of their effective role in diluting or reinforcing\n\n\nmarket power in this sector. However, many of these firms\n\nMention: \"1999 ITU report\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its turnover ranking of the twenty largest providers supplies the firm list.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002739:13:0", "occurrence_id": "2c4ec8cfc2f5e1e4074ec2f4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: first firms identified in a 1999 ITU report in a ranking of the main 20 providers according to\n\n\ntheir turnover. It turns out that 8 of these firms are present in both the fixed and the mobile\n\n\nlines business. For fixed lines, the 9 top firms were then: Bell South Corporation, BT Group,\n\n\nDeutsche Telekom, France Telecom, SBC, Telecom Italia, Telefonica de España, Telstra, and\n\n\nVerizon. For mobile, the firms covered are: Bell South Corporation, Deutsche Telekom,\n\n\nFrance Telecom, SBC, Telecom Italia, Telefonica de España, Telstra, Verizon and Vodafone.\n\n\nBT is dropped from the mobile list while Vodaphone is not present in the fixed line list. For\n\n\nfixed lines we consider 35 countries (18 OECD and 17 non OECD) while for mobile, we\n\n\ncover 65 countries (20 OECD and 45 non OECD).\n\n\nA few words of caution on this list may be appropriate. First, there is a fair amount of\n\n\nturnover in ownership in this sector. Some of the firms in the 1999 list have now changed\n\n\nowners (and names). Our choice of operators is thus arbitrarily frozen in 1999 even if the\n\n\ninformation is collected for these companies in their form as of 2003. Second, there is a clear\n\n\nregional specialization of firms. There are only 4 firms with a strong presence in LDCs:\n\n\nFrance Telecom, SBC, Telecom Italia, Telefonica de España. This last one is particularly\n\n\ndominating in Latin America, Telstra in Asia and France Telecoms in Eastern Europe. Third,\n\n\nthe list does not do justice to the growing presence of regional actors in the various regional\n\n\nmarkets with strong South African firms present in Africa and strong Mexican and Brazilian\n\n\noperators present in Latin America for instance. However, most of these firms and many of\n\n\nthose included in the ITU list do not publish the necessary information on their webpages and\n\n\nhence we cannot make a fair assessment of their effective role in diluting or reinforcing\n\n\nmarket power in this sector. However, many of these firms\n\nMention: \"ITU list\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Most listed firms publish too little for a fair assessment.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002747:7:0", "occurrence_id": "869cc6c037da88ba5222df90", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: capacity is spread over a large number of similar enterprises. Finally, one of the key\n\n\ncompeting hypotheses is that in countries with geographical constraints to development it\n\n\nis more difficult to have successful projects (implied in Dalgaard et al. 2004). We\n\n\nconsider two approaches to including geography: a variable capturing the share of a\n\n\ncountry’s territory in the tropics (Gallup et al. 1999) and/or regional dummies for each of\n\n\nthe major regions of the world.\n\n\nWe start with OLS regressions and with a simple specification in which the\n\n\nproject success rate for each country in the 1990s is a function of log per capita GDP, the\n\n\nRule of Law measure, the Freedom House Index, and the overall quantity of aid relative\n\n\nto GDP (all averaged for the 1990s).\n\n\nAverage project success rate in 1990si = b0 + b1 Log (per capita GDP in 1990i) +\nb2 Log (average ICRG rule of law in 1990si) + b3 Log (average Freedom House\nindex in 1990si) + b4 Log (average aid to GDP in 1990si),\n\nwhere i is the country in which the WB projects take place.\n\n\nThere is a very strong positive relationship between institutional quality and\n\n\nproject success (Table 1, column 1). Interestingly, both political freedom (5 percent\n\n\nlevel) and the property rights/rule of law measure (1 percent level) enter with high\n\n\nstatistical significance, and the joint test on the two is even more significant. Initial\n\n\nincome level and aid enter negatively, but neither is statistically significant.\n\n\nIn column 2 we introduce geography in the form of the percent of the country’s\n\n\nterritory situated in the tropics. The coefficients on the two institutional variable barely\n\n\nchange and remain significant at 5 percent level. Tropical location is associated with\n\n\nfailure with a coefficient of –9.6, but the coefficient is not statistically significant.\n\n\nWe can use partial scatters to illustrate this relationship between\n\nMention: \"the Freedom House Index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"It enters the project success regression alongside per-capita income and aid.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002752:22:0", "occurrence_id": "dcbb46e7d08662d704af9885", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **<u>Part II : Application to Morocco</u>**\n\nWe now apply the methodology developed in Part I to assess the performance of the housing\nsubsidy system in Morocco. We have used data from a study undertaken by the Ministry of\nHousing in Morocco in 1998-1999, through an assistance program of the World Bank\n(PLURAM-ALGOE-PROMOCONSULT, 1999). The purpose of the study was to examine all the\ndimensions of housing financing in Morocco. In particular, it looked at all subsidies for housing,\nboth from a qualitative and quantitative perspective. One product of the study is a booklet which\nsummarizes expenditures on all types of housing subsidies in Morocco, explicit as well as\nimplicit, for the year 1995. We have used this and other materials <sup>36</sup> [^36: In particular, another study undertaken in the same conditions as the first, the study by EDESHAURBAPLAN-IHEM (1998), concentrated on the bottlenecks in the housing construction process.] to rate the different\ncomponents of the housing subsidy system based on public finance criteria described in Part I.\n\n\n<u>1) Housing in Morocco</u>\n\n_Housing supply_\n\nMorocco has a current population of about 30 million people and 5 million primary residences. In\n2000, 62% of households were owners, 28 % were renters, and 10 % had another status. <sup>37</sup> [^37: The rental sector is entirely private, except for 38,000 public units managed by the companies CIFM and\nSOPHAL.]\nSimilar to its neighbors, Algeria and Tunisia, Morocco faces a high urban population growth,\nwith an annual growth rate ranging from 3 to 4 %. Demand for new housing units is estimated at\nabout 120,000 units per year and is expected to continue growing in future due to the\ndemographic composition of the population, whereas annual supply of formal housing units is\nestimated around 90,000 units.\n\nThe housing stock can be broken down into two categories: formal\n\nMention: \"the Ministry of\nHousing in Morocco\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its study of the housing subsidy system provides the evaluation data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002754:24:2", "occurrence_id": "0b63ef2a9c00cf7c26e84999", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: >Guatemalan<br>quetzals<br>|6,798.98<br>(14,021.5)<br>|6,523.24<br>(8,351.7)<br>|7,042.01<br>(13,307.4)<br>|0.45<br>|-0.12<br>|\n|<br>N<br>|5,665<br>|1,063<br>|593<br>|<br>|<br>|\n\n\ninternational remittances (from USA). All values are\nweighted; standard deviations are in parentheses.\n\n\n\nIn 2000, 1 Guatemalan quetzal = US$0.128.\n\nSource: Guatemala 2000 ENCOVI Survey, Instituto Nacional de Estadistica.\n\n\n*Significant at the 0.05 level.\n**Significant at the 0.01 level.\n\nMention: \"Guatemala 2000 ENCOVI Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Weighted remittance regressions with standard deviations are estimated on this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002762:54:0", "occurrence_id": "c98364678f84b9f29745d35a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: raw materials shares than that obtained using data for a single year. Although the input\n\noutput matrices used do not cover our sample period, 1981-1991, we believe that input\n\noutput relationships are relatively stable over these two decades. Matrix _A_ has 22 rows\n\n\nand 17 columns corresponding to the Colombian national accounts classification of\n\n\nindustries. The number of rows exceeds the number of columns because some raw\n\n\nmaterials used in manufacturing originate in the primary sector. Note that by\n\n\n\n22\n\n\n\n22\n## ∑ aij =\n\n_i_ = 1\n\n\n\nconstruction∑ _aij_ = 1\n\n\n\n_i_ = 1\n\n\n\n_aij_ = 1. Hence, our domestic raw materials price indexes are weighted\n\n\n\naverages of producer price indexes: for each manufacturing industry _j_ = 1, …, 17 and\n\n\n\n22\n\n\n\n22\n## time t, the domestic raw materials price index is defined as p RMjt = ∑\n\n\n## p RMjt = ∑ aij pit . To perform\n\n\n\n=\n## ∑\n\n_i_ = 1\n\n\n\n_i_ = 1\n\n\n\nthis calculation we aggregate 29 manufacturing producer price indexes at the 3-digit ISIC\n\n\nrevision 2 into 17 producer price indexes at the broader Colombian national accounts\n\n\nclassification. We use production weights for the period 1975-1989 to aggregate these\n\n\nprice indexes. For the primary sectors included in the calculations we use wholesale price\n\n\nindexes.\n\n\nThe construction of exports price indexes is more involved because the series\n\n\navailable from Banco de la República (Colombia’s central bank) starts only in 1990. For the\n\n\nperiod 1975-1990, we construct export price indexes using detailed trade data from the\n\n\nDirección de Impuestos y Aduanas Nacionales (DIAN). Export transactions in 1975-1990\n\n\nare recorded at an 8-digit Colombian trade classification (NABANDINA) based on the\n\n\nBrussels Tariff Nomenclature. For each NABANDINA and year, we compute export prices\n\n\nin pesos per unit of weight by dividing the value\n\nMention: \"Dirección de Impuestos y Aduanas Nacionales\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Eight-digit export transactions are used to build export price indexes.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002763:16:0", "occurrence_id": "ea34d9f3b5eb98a4d9ca8b7d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Together, Beijing and Shanghai hold a large share of innovative industries in the country\n\n\naccording to the figures in Table 9, whether in terms of firm numbers or employee numbers.\n\n\nHowever, the innovative sectors are still small in the overall urban economy in Beijing and\n\n\nShanghai. For example, employment in the high- and new-tech enterprises amounts to about 4%\n\n\nof the total employment in Beijing and less than 1% in Shanghai. <sup>7</sup> [^7: However, according to a more extensive definition of high-tech industries which not limited within high-tech\nindustrial parks, the gross output value of high-tech industries in both Beijing and Shanghai has hold a large share in\ntotal industrial output, over 20%, in both cities in 2002 (See Beijing and Shanghai Statistic Yearbook 2003).] The disparities between\n\n\nBeijing and Shanghai in knowledge intensive business services and creative content providers\n\n\nseem even larger. The employment in the selected categories together adds up to nearly 10% of\n\n\nthe total employment in Beijing, much higher than that of the nation as a whole (3.3%), while\n\n\nthis figure is less than 4% for Shanghai, only a little higher than the national level.\n\n\n**Innovative industries in Beijing**\n\n\nBeijing has unique advantages as the national capital in the development of innovative\n\n\nindustries.\n\n\n_National R&D center for high-tech development_\n\n\nBeijing has an abundance of academic think-tanks, universities, research institutes and\n\n\nresearch and development (R&D) centers. More than 50 universities and colleges and over 200\n\n\nresearch bodies are located in the capital, helping to spawn the largest number of high-tech\n\n\nenterprises in China, and some of the nation’s top high-tech companies such as Lenovo and\n\n\nFounder. Table 10 compares the R&D investment in Beijing and Shanghai in 2000. The large\n\n\nnumber of state-owned higher education and R&D institutes received the lion’s share of\n\n\ngovernment funded R&D investment.\n\n\n**Table 10: The R&D Investment in Beijing and Shanghai\n\nMention: \"Beijing and Shanghai Statistic Yearbook 2003\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It reports high-tech industries exceeding a fifth of industrial output.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002773:11:0", "occurrence_id": "c39a31c2e2c2910cf3126089", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 11\n\n\nthe small rise in the Gini index in these two countries could have two alternative explanations:\n\n\nthat everyone at a given income experienced similar proportionate income increases but those\n\n\nwho were better off initially experienced somewhat larger proportionate increases; or that people\n\n\nwho were initially _worse off_ experienced larger proportionate increases but this pro-poor vertical\n\n\nredistribution was more than offset by the inequality-increasing redistribution associated with\n\n\npeople with similar initial incomes experiencing different income increases (horizontal\n\n\nredistribution) and reranking. Section IV finds evidence in favor of the latter explanation.\n\n\nWithout horizontal redistribution and reranking, the Gini coefficient for income in China might\n\n\nhave fallen between 1989 and 1997 by as much as 21.6 points. In the event, the horizontal\n\n\nredistribution and reranking involved in the move from the 1989 to 1997 income distributions\n\n\nwas so large that the Gini _increased_ —by 2.4 points. A similar picture—albeit less dramatic—\n\n\nemerged from the Vietnam panel for 1993-98.\n\n\nDo these results matter? If vertical redistribution ( _V_ ), horizontal redistribution ( _H_ ) and\n\n\nreranking ( _R_ ) matter only insofar as they contribute to the change in the Gini ( _∆_ _G_ ), knowing the\n\n\nrelative magnitudes of _V_, _H_ and _R_ is immaterial. If the policymaker is simply interested in _∆_ _G_,\n\n\nshe would not care whether a small value of _∆_ _G_ derives from small values of _V_, _H_ and _R_, or from\n\n\na large negative value of _V_ being offset by a large positive value of _H_ + _R_ . Nonzero values of _H_\n\n\nand _R_ would be considered bad by an inequality-averse policymaker because— _and only_\n\n\n_because_ —they reduce _∆_ _G_ for a given value of _V_ .\n\n\nThere is, in\n\nMention: \"the Vietnam panel\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Its distributional changes show a smaller but similar rise in inequality.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002775:10:2", "occurrence_id": "aed1d7766cf3b60615fc154b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 1995 Male<br>1995 Female|1995 Male<br>1995 Female|1995 Male<br>1995 Female|1995 Male<br>1995 Female|1995 Male<br>1995 Female|1995 Male<br>1995 Female|1995 Male<br>1995 Female|1995 Male<br>1995 Female|\n|1995 Male<br>1995 Female||||||||||||||||\n||||2003 Male<br>2003 Female|2003 Male<br>2003 Female|2003 Male<br>2003 Female|2003 Male<br>2003 Female|2003 Male<br>2003 Female|2003 Male<br>2003 Female|2003 Male<br>2003 Female|2003 Male<br>2003 Female|2003 Male<br>2003 Female|2003 Male<br>2003 Female|2003 Male<br>2003 Female|2003 Male<br>2003 Female|2003 Male<br>2003 Female|\n|||||||||||||||||\n|||||||||||||||||\n|||||||||||||||||\n|||||||||||||||||\n|||||||||||||||||\n|||||||||||||||||\n|||||||||||||||||\n|||||||||||||||||\n\n\nNo Complete Eduation Primary Complete Lower Sec. Complete Upper Sec. Complete Higher Education Technical\n\n<u>Source: Calculations based on ENE surveys.</u>\n\n\n8\n\nMention: \"ENE surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Education attainment profiles are calculated from these employment surveys.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002776:9:0", "occurrence_id": "afdea4355ae684a6aca5cddf", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 2 identifies the products, which, between 1990 and 2002, underwent a variation above the\n\n\ngeneral average for import rates, measured in terms of the ratio of import to domestic production\n\n\nvalues. This group encompasses the major enterprises which had demanded trade defense\n\n\nmeasures — among others, manufacturers of artificial yarn and fabrics whose share of imports in\n\n\ndomestic production rose from 2.9% in 1990 to 29.6% in 2002; pharmaceutical and cosmetic\n\n\nproducts, from 7.1% to 27.9%; chemical products, from 6.7% to 16.7%; rubber, from 4.9% to\n\n\n13.1%; plastic products, from 2.2% to 10.4% and metals from 2.5% to 8.2%. Other sectors, such as\n\n\nelectronics, motorcars and vehicles and auto parts were given additional protection through an\n\n\nincrease in tariffs.\n\n\nFigure 2\n\n\nThe Ratio of Import to Domestic Production values for Certain Products: 1990 to 2002 (%)\n\n\n\n\n\n**100**\n**90**\n**80**\n**70**\n**60**\n**50**\n**40**\n**30**\n**20**\n**10**\n**0**\n\n\nSource: Brazil’s national accounts, IBGE, several years. Author's calculations.\n\n\n**3. The Brazilian Antidumping and Safeguard System**\n\n\n**3.1 Legal, Institutional and Operational Aspects**\n\n\nUntil 1988, Brazil had two instruments to fight dumping which enabled the fiscal value of imported\n\n\ngoods to be changed in order to estimate import duties. One of them, known as the <u>minimum</u>\n\n\n10\n\nMention: \"Brazil’s national accounts, IBGE\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Import shares by product are computed from these accounts.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002782:13:1", "occurrence_id": "99aeca3bf9b7d7038f0b5f02", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  17,996.2 143.6\nTotal manufacturing (c) 2001 138,323.0 345,141 31,922.8 48,129.6 92.5\n_Percentage_\n\nElectronics manufacturing 2001 45.0 28.4 36.4 33.1\nPharmaceuticals manufacturing 2001 3.7 0.7 8.8 4.3\n<u>(a+b)</u> <u>2001</u> 48.7 29.1 45.1 37.4\n_Note_ : The high-tech manufacturing sector comprises SSIC35220 medicinal & pharmaceutical products and SSIC 384 electronics for 1990. Comprises SSIC 242\npharmaceutical products and SSIC31 electronics for 2001\n_Sources_ :\nEconomic Development Board (EDB). (1990). _Report on the Census of Industrial Production 1990_ . Singapore: EDB\nEconomic Development Board (EDB). (2001). _Report on the Census of Manufacturing Activities 2001_ . Singapore: EDB\n\n\n12\n\nMention: \"Census of Industrial Production 1990_\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The earlier benchmark for industry shares by classification.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002782:13:1", "occurrence_id": "806cee839eed5155ad2b73ea", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  17,996.2 143.6\nTotal manufacturing (c) 2001 138,323.0 345,141 31,922.8 48,129.6 92.5\n_Percentage_\n\nElectronics manufacturing 2001 45.0 28.4 36.4 33.1\nPharmaceuticals manufacturing 2001 3.7 0.7 8.8 4.3\n<u>(a+b)</u> <u>2001</u> 48.7 29.1 45.1 37.4\n_Note_ : The high-tech manufacturing sector comprises SSIC35220 medicinal & pharmaceutical products and SSIC 384 electronics for 1990. Comprises SSIC 242\npharmaceutical products and SSIC31 electronics for 2001\n_Sources_ :\nEconomic Development Board (EDB). (1990). _Report on the Census of Industrial Production 1990_ . Singapore: EDB\nEconomic Development Board (EDB). (2001). _Report on the Census of Manufacturing Activities 2001_ . Singapore: EDB\n\n\n12\n\nMention: \"Census of Manufacturing Activities 2001_\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Pharmaceutical and electronics output for 2001 is taken from this census.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002790:31:13", "occurrence_id": "914054cfa7bf641cfb14f031", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <br>0.027<br>0.000<br> <br>0.234 ***<br>0.030<br>0.000<br> <br>-0.214 ***<br>0.028<br>0.000<br> <br>-0.131 ***<br>0.029<br>0.000<br> <br>-0.140 ***<br>0.019<br>0.000<br> <br>5.306 ***<br>0.092<br>0.000|<br>0.063 **<br>0.030<br>0.032<br> <br>0.028<br>0.034<br>0.407<br> <br>-0.288 ***<br>0.028<br>0.000<br> <br>-0.043<br>0.030<br>0.153<br> <br>-0.292 ***<br>0.027<br>0.000<br> <br>-0.238 ***<br>0.029<br>0.000<br> <br>-0.127 ***<br>0.019<br>0.000<br> <br>6.049 ***<br>0.093<br>0.000|\n|Source: Authors estimations based on ENIGH 1992, 1996, and 2002. Note:1Rural area defined as locality with less than 15,000 inhabitants.<br>*** sign. at 1%, ** sign. at 5%, * sign. at 10%,\n\nMention: \"ENIGH 1992, 1996, and 2002\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Coefficients with significance stars are estimated across these rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002796:8:0", "occurrence_id": "62dbd5978343c1dd9d075307", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 31.7 percent). The incidence is also signi fi cantly greater for more highly educated workers,\n\n\nand for professionals relative to technical and clerical staff.\n\n\n<u>Table</u> <u>1:</u> <u>Employer</u> <u>sponsored</u> <u>training</u> <u>in</u> <u>1998-99</u>\n\n|Immigrant category|% of recent immigrants in the<br>category receiving training|% of Canadian-born in the<br>category receiving training|\n|---|---|---|\n|All|47.7|56.3|\n|With no post-secondary education|34.5|50.6|\n|With post-secondary education|55.4|61.9|\n|Non-permanent workers|31.7|43.2|\n|Permanent workers|49.2|57.7|\n|Professionals|63.8|69.4|\n|Technical/trades|44.6|53.2|\n|Clerical/administrative|50.3|55.7|\n\n\n\nSource: Workplace and Employee Survey, 1998-99. Sample size: 23,540.\n\n\nFor simplicity, the costs and bene fi ts from training are assumed to have the following\n\n\nstructure. Let c - 0 denote the exogenously given and constant cost of training each\n\n\nimmigrant, which is incurred in stage 1. Training involves imparting fi rm-speci fi c knowledge\n\n\nto a worker while currently employed (on the job training). Without loss of generality we\n\n\nassume that the efficiency unit of each immigrant in stage 1 equals unity. Let γ - 1 denote\n\n\nthe efficiency unit of a trained worker in stage 2 to his stage 1 employer. <sup>3</sup> Let\n\nMention: \"Workplace and Employee Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Training incidence by occupational group is tabulated from 23,540 respondents.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002798:10:0", "occurrence_id": "9028a2cc466d9cefc4bdef33", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Table 7 provides information on the importance of intra-industry trade in Central America based\n\non the adjusted Grubel-Loyed intra-industry trade index. <sup>5</sup> [^5: _AIIT_ = _<u>i</u>_ _<u>i</u>_] This index can take values between 0\n\n\n(no intra-industry trade) to 1 (all trade is intra-industry). There appears to be some importance of\n\n\nintra-industry trade within Central America, however, with the exception of Costa Rica (0.3)\n\n\nthere is virtually no evidence of intra-industry trade with the United States. For El Salvador and\n\n\nGuatemala intra-industry trade appears to be quite high with Mexico and Brazil.\n\n\n**Table 7. Intra-industry Trade**\n\n\nSource: Own calculation based on trade data from UN COMTRADE for the year 2001. A 5 digit level of\ndisaggregation is used for this exercise.\n\n\n**3.3. Business Cycle Synchronization and Trade**\n\n\nEmpirical evidence on trade integration and business cycle synchronization is somewhat mixed.\n\n\nWhile Frankel and Rose (1998), Choe (2001), Calderon, Chong and Stein (2002) and Calderon\n\n\n(2003) find that a higher trade intensity tends to increase business cycle synchronization, Shin\n\n\nand Wang (2003) find that increasing trade itself does not necessarily lead to more synchronized\n\n\nbusiness cycles, evidence for East Asia suggests that only the expansion of intra-industry trade\n\n\nhad such an effect. However, Garnier (2003) find only weak or no relations between intra\n\nindustry trade and business cycle synchronization for 16 industrialized countries and conclude\n\n\nthat intra-industry trade at most only partially explains business cycle transmission; the low\n\n\n\n_n_\n\n\n\n( _X_ _i_ + _M_ _i_ )\n\n\n\n_n_ _n_\n## ∑ ( X i + M i ) − ∑\n\n\n## ∑ ( X i + M i ) − ∑ X i −\n\n= _<u>i</u>_ _<\n\nMention: \"UN COMTRADE\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Five-digit trade flows give the intra-industry trade indices.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002802:8:0", "occurrence_id": "36dfbd53bf4c4bfad3affe8d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: grants them permanent residence. Thus, graduate training of foreign students may have\n\n\nlong-lasting impacts on innovation capacities.\n\n\nOn a negative note, Borjas (2002) speculates that foreign students in the United\n\n\nStates benefit the economy to the tune of $1 billion a year, but this gain is more than\n\n\noffset by the costs of taxpayer-financed grants and subsidies at public universities. In\n\n\nanother paper Borjas (2004) finds a strong negative correlation between the enrollment of\n\n\nnative men in U.S. graduate programs and the enrollment of foreign students. Institutions\n\n\nwhich experienced the largest increase in foreign enrollment were also institutions that\n\n\nexperienced the steepest fall in the enrollment of native males.\n\n\nFor our purposes the interesting aspect of this claim is that Borjas seems to\n\n\nsuggest that domestic and foreign graduate students are highly substitutable and display\n\n\nsimilar characteristics. In fact, other information indicates that this assertion is\n\n\nquestionable. Although data on the quality of domestic graduate student applicants\n\n\ncompared to their international counterparts are not readily available, results from\n\n\nTIMMS and other international tests indicate that the native U.S. student pool for\n\n\nengineering and science programs is likely to be limited due to lower math and science\n\n\nachievement. <sup>13</sup> This suggests that student populations are not readily substitutable and\n\n\nthat university technical training programs may have increased their demand for foreign\n\n\nstudents.\n\n\nAs a result it is not surprising that a recent study indicates that there has been a\n\n\nsharp drop in the proportion of PhDs in science and engineering awarded to U.S.-born\n\n\nmales between the early 1970s and 2000 (Freeman et al., 2004). In 1966 these students\n\n\naccounted for 71 percent of science and engineering PhD graduates, while six percent\n\n\n7\n\nMention: \"TIMMS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its international test results indicate a limited domestic pool in science and engineering.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002810:7:0", "occurrence_id": "8ea2e5683956d47b7a2bb06d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: developments to start with, their consolidation is driven by organized groups of _land-_\n\n\n_grabbers_ . While squatter housing may be initially developed by households on open\n\n\nlands in close proximity to employment centers, these settlements get consolidated by\n\n\ngroups who construct dwellings for sale and rental purposes.\n\n\n**Table 1: Population by housing category, Bhopal 2003**\n\n\n**_Number of_**\n\n\n\n**_Housing category_**\n\n\n\n**_sample_**\n**_<u>households</u>_**\n\n\n\n**_Percent_**\n\n\n\n1 Non-notified squatter settlement 221 8.8\n2 Notified squatter settlement 519 20.7\n3 Resettlement area 80 3.2\n4 Unauthorized colony 184 7.3\n5 BDA/MPHB plots 122 4.9\n6 BDA/MPHB flats 132 5.3\n7 Cooperative housing 85 3.4\n8 Employer housing 357 14.2\n9 Private builder's / colonies 306 12.2\n10 Core/historical city area 502 20.0\n**<u>Total</u>** **<u>2508</u>** **<u>100.0</u>**\n\n\nSource: Bhopal Urban Household Survey 2003\n\n\nFigure 2 shows the distribution of welfare levels across households in six broad\n\n\nhousing categories, where the welfare measure is per capita annual household\n\n\nconsumption in Rupees. There is significant variance within each housing category with\n\n\nhouseholds in slums having the lowest consumption expenditures.\n\n\n\n7\n\nMention: \"Bhopal Urban Household Survey 2003\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Two thousand five hundred eight households are distributed across housing categories and welfare levels.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002811:10:0", "occurrence_id": "dd47e66690545332670b65e2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: series (land, buildings, etc.) we applied estimated depreciation rates from Harris et al.\n(1994). <sup>18</sup> [^18: The assumed annual depreciation rate for buildings is 3.3 percent, for machinery 10 percent, and for vehicles\nand other fixed assets 20 percent. For land, we assumed no depreciation. These rates are very similar to estimates\npresented in Goeltom (1995).]\n\nSince the data contain no information on physical quantities of inputs used or output\nproduced by plants, we are forced to start with nominal values instead. These are deflated\nusing a set of 192 wholesale price indices for manufactured commodities, published by the\nIndonesian Statistical Office. The commodity indices are mapped to the 5-digit ISIC\nclassification using a concordance table provided by the Statistical Office. These detailed 5digit ISIC level deflators are applied to plant output and material inputs. Figures on\ninvestment and capital are deflated as follows. For buildings, we use a wholesale price index\n(WPI) published in the Statistical Yearbook of Indonesia, and for machinery and vehicles the\naverage of the WPIs for 5-digit sectors producing machinery and vehicles, respectively. For\nother assets, we employ the economy-wide WPI. Unfortunately, the Indonesian Statistical\nOffice does not publish a wholesale price index for energy, so we were constrained to use a\nCPI specific to energy instead.\n\nThe production function is assumed to be Cobb-Douglas and is estimated using the\nsemiparametric procedure suggested by Levinsohn and Petrin (2003). <sup>19</sup> [^19: The estimation was implemented in Stata 8 using the program described in Levinsohn et al. (2003). In 29 out of] As a proxy for\nunobserved productivity shocks that may influence the input decision of the plant, we\nemploy the amount of electricity consumed by each plant. The data contain information on\nelectricity consumption net of own production and sales to other plants, expressed in\nphysical quantities (kWh) which rules out measurement errors related to deflation. Our\nproductivity\n\nMention: \"Statistical Yearbook of Indonesia\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Its wholesale price index deflates buildings and machinery investment.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002812:10:1", "occurrence_id": "8dfa8a3d8404760e0de11568", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: </u> <u>?</u>\n**<u>Crisis episode</u>** **<u>×</u>** **<u>×</u>** **<u>?</u>** **<u>?</u>** **<u>?</u>** **<u>?</u>** **<u>?</u>** **<u>×</u>** **<u>×</u>** **<u>×</u>** **<u>×</u>** **<u>?</u>** **<u>×</u>**\n_Note: (?_ _) denotes crisis periods; ?: crisis episode, ×: non-crisis episode_\n_Source: Author_\n\n\nIndicators are derived from the following sources.\n\n- Liquidity type banking crises (demand deposits to total assets ratio): IFS line 24 divided by the sum of\n\nIFS line 24-27;\n\n- Solvency type banking crises (capital account to total assets ratio): IFS line 27A divided by the sum of\n\nIFS line 24-27;\n\n- Balance of payments crises (year-on-year growth rates of gross international reserves): total reserves\n\nminus gold (IFS line 1L.d);\n\n- Currency crises (end of period nominal exchange rates against the US dollar): IFS line ae;\n\n- Debt crises (external debt outstanding to exports ratio): external debt outstanding data is obtained from\n\nCEIC data base, countries’ publications and GDF, and outstanding is divided by IFS line 70;\n\n- Growth rate crises (year-on-year real growth rates):\n\nMention: \"CEIC data base\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"External debt outstanding is taken from it to build the debt crisis indicator.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002815:2:0", "occurrence_id": "e196b6394bfd1be9f4f0895c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: maquila-- might reduce migration. <sup>2</sup> [^2: Using data gathered in 25 Mexican communities, Massey and Espinosa (1997) create histories of migrants\nto the US and find initial migration to the US to be negatively related to the wage rate and the proportion of\nwomen in manufacturing. Jones (2000) examines 17 emigrant municipalities and argues in favor of an\nassociation between proximity to maquiladoras and employment growth and, in turn, declining US\nmigration.] In a times series context, Hanson and Spillimbergo\n\n\n(1999) find that border apprehensions are responsive to US-Mexico wage differentials, a\n\n\nfinding confirmed by McKenzie and Rapoport (2003) using a retrospective survey.\n\n\nDavila and Saenz (1990) find a negative relationship between lagged maquila\n\n\nemployment on the border and monthly border apprehensions as a proxy for migration\n\n\npressures. Unfortunately, more direct trade data on FDI and trade are available neither at\n\n\nthe level of periodicity or span to offer enough degrees of freedom to permit inference\n\n\nwith any confidence. Further, the massive peso crisis beginning roughly concomitant\n\n\nwith the signing of NAFTA led to a sharp spike in unemployment and a 25% fall in\n\n\nwages. These likely induced flows to the US that complicate inference on the more\n\n\nmodest effects that NAFTA might have had in the opposite direction.\n\n\nFor these reasons, this paper focuses on evaluating the mechanisms through which\n\n\nNAFTA related variables might work using data on migration flows _within_ Mexico where\n\n\nthe 2000 census data permit careful analysis and the 32 by 32 permutations offer\n\n\nsubstantial degrees of freedom. Figure 1 shows the rates of net migration (net flows as a\n\n\nfraction of the population in the initial period) by state and figure 2 FDI by state. A casual\n\n\nlook suggests that, in fact, there is a relationship between the magnitude of FDI and\n\n\nmigration to a locale. In investigating this possibility for FDI and other trade variables\n\n\nmore rigorously, we make three contributions to the literature on migration more\n\n\ngenerally, and then specifically on the impact of trade and FDI flows\n\nMention: \"data on migration flows\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Internal flows across thirty-two states are used to test NAFTA-related mechanisms.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002815:2:0", "occurrence_id": "992c0eb85458a80cba79c2ab", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: maquila-- might reduce migration. <sup>2</sup> [^2: Using data gathered in 25 Mexican communities, Massey and Espinosa (1997) create histories of migrants\nto the US and find initial migration to the US to be negatively related to the wage rate and the proportion of\nwomen in manufacturing. Jones (2000) examines 17 emigrant municipalities and argues in favor of an\nassociation between proximity to maquiladoras and employment growth and, in turn, declining US\nmigration.] In a times series context, Hanson and Spillimbergo\n\n\n(1999) find that border apprehensions are responsive to US-Mexico wage differentials, a\n\n\nfinding confirmed by McKenzie and Rapoport (2003) using a retrospective survey.\n\n\nDavila and Saenz (1990) find a negative relationship between lagged maquila\n\n\nemployment on the border and monthly border apprehensions as a proxy for migration\n\n\npressures. Unfortunately, more direct trade data on FDI and trade are available neither at\n\n\nthe level of periodicity or span to offer enough degrees of freedom to permit inference\n\n\nwith any confidence. Further, the massive peso crisis beginning roughly concomitant\n\n\nwith the signing of NAFTA led to a sharp spike in unemployment and a 25% fall in\n\n\nwages. These likely induced flows to the US that complicate inference on the more\n\n\nmodest effects that NAFTA might have had in the opposite direction.\n\n\nFor these reasons, this paper focuses on evaluating the mechanisms through which\n\n\nNAFTA related variables might work using data on migration flows _within_ Mexico where\n\n\nthe 2000 census data permit careful analysis and the 32 by 32 permutations offer\n\n\nsubstantial degrees of freedom. Figure 1 shows the rates of net migration (net flows as a\n\n\nfraction of the population in the initial period) by state and figure 2 FDI by state. A casual\n\n\nlook suggests that, in fact, there is a relationship between the magnitude of FDI and\n\n\nmigration to a locale. In investigating this possibility for FDI and other trade variables\n\n\nmore rigorously, we make three contributions to the literature on migration more\n\n\ngenerally, and then specifically on the impact of trade and FDI flows\n\nMention: \"2000 census data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Net migration rates by state pair are computed from this enumeration.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002820:30:0", "occurrence_id": "e651206c5fb702801b31c141", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <u>Appendix Table:</u> _<u>Main finance questions in recent LSMS surveys</u>_\n\n\n_Country_\n\n\n_Date of the survey_\n\n\n_Sample size_\n\n\n**_Loan or credit-related questions_**\n\nDid you borrow in the past 12 months? ***** ***** ***** ***** ***** ***** ***** ***** ***** ***** ***** ***** ***** ***** ***** ***** ***** ***** ***** ***** *****\n\n\nSources of the credit? ***** ***** ***** ***** ***** ***** ***** ***** ***** ***** ***** ***** ***** ***** ***** ***** ***** ***** ***** ***** ***** *****\nMain reason for borrowing? ***** ***** ***** ***** ***** ***** ***** ***** ***** ***** ***** ***** ***** ***** ***** *****\n\nHow much did you borrow? ***** ***** **\n\nMention: \"LSMS surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"Their finance modules are compared to harmonise borrowing questions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002788:9:1", "occurrence_id": "f70fcd87b8c1676e5f4dbcf5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  nominal wages for differences in price levels\n\n\nacross urban and rural areas and states, and for temporal changes in price levels between\n\n\n1993 and 1999, using the indexes developed by Deaton and Tarozzi (2000). Thus, all\n\n\nwages in the paper are expressed in terms of the 1999 average of India’s rural prices.\n\n\nWorkers engaged in salaried occupations were asked about the type of enterprise\n\nwhere they carried out their main work activity. In the 50 <sup>th</sup> round, all salaried workers\n\n\nwere asked to choose from three categories: public, private, or semipublic, and we\n\n\n4 These cross-sectional surveys use a two-stage stratified sampling design, and are representative of the\nnational, state, and so-called “NSS region” level. The first-stage sampling units (FSUs) are census villages\nin the rural areas, and the NSS Urban Frame Survey (UFS) are blocks in the urban areas. The second stage\nunits are the households in both urban and rural sectors. For more information on the survey and sample\ndesign see, e.g., NSSO, Ministry of Planning (1994, 2000).\n5 In the 50th round, the sector of employment – public or private – is defined only for those engaged in\nregular salaried/wage work (in all industrial activities), while in the 55 <sup>th</sup> round, it is defined for those who\nare employed in non-agricultural activities. In both rounds, wages are defined for everyone who reported\nbeing employed. In order to maintain consistency and comparability across the two rounds, the\nclassification of the sector of employment was defined only for those with regular salaried/wage\nemployment in non-agricultural industries. We also excluded casual workers in public works as well as\nindividuals who were sick or did not work for other reasons.\n\n10\n\nMention: \"NSS Urban Frame Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Its blocks serve as first-stage sampling units in urban areas.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002792:39:2", "occurrence_id": "b4d816146b851b18d41a549d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: : Households with a member\nworking overseas in Jun 1996 (according to Oct 1997 Survey of Overseas Filipinos) and that also appear in 1998 Annual Poverty\nIndicators Survey, and excluding households with incomplete data (see Data Appendix for details).\n\n\nafe_table **s** =m9Sat._hh\n\nMention: \"Oct 1997 Survey of Overseas Filipinos\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It identifies households with a member working overseas mid-1996.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002792:39:2", "occurrence_id": "a1a530769d8b322983cdded4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: : Households with a member\nworking overseas in Jun 1996 (according to Oct 1997 Survey of Overseas Filipinos) and that also appear in 1998 Annual Poverty\nIndicators Survey, and excluding households with incomplete data (see Data Appendix for details).\n\n\nafe_table **s** =m9Sat._hh\n\nMention: \"1998 Annual Poverty\nIndicators Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Matching households into this round gives the linked analysis sample.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002828:32:0", "occurrence_id": "0310f7609f9050cba63288d1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **<u>Source:</u>** Authors’ elaboration based on INDEC.\n\n\n\n31\n\nMention: \"INDEC\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The tabulated statistics are elaborated from this statistical institute.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002831:6:0", "occurrence_id": "92a3a20e00d2522abbfdc9a1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: about 8 percent of Kenyan exporters export to their main industrial market (the United\n\n\nKingdom), since 58 percent of Kenyan enterprises export, this represents over 4 percent of\n\n\nKenyan enterprises. The poor performance of Ethiopian exporters in regional markets probably\n\n\nreflects regional difficulties that have prevented Ethiopian enterprises from developing export\n\n\npartnerships with firms in neighboring countries (e.g., in Eritrea, Somalia and Sudan).\n\n\n**II.2** **Barriers to Trade**\n\n\nMany different factors affect firms’ propensities to export. Since the cost of\n\n\ntransportation will vary between sectors, enterprises will be more likely to export in some sectors\n\n\nthan they will be in others. Similarly, enterprise size is likely to be important—there are fixed\n\n\ncosts associating with exporting that larger enterprises will find it easier to bear. Finally,\n\n\ntransportation costs will vary between countries. Although exporting will generally be more\n\n\ndifficult for enterprises in landlocked countries, it is important to note that most of the small and\n\n\nmedium enterprises included in the investment climate surveys export to neighboring countries\n\n\n(including neighboring landlocked countries) rather than to more distant markets.\n\n\nAnother factor that might affect export performance is the existence of policy-induced\n\n\nbarriers to trade. Export taxes are the most direct policy-induced barrier. In practice, however,\n\n\nexport taxes—which were mostly levied on primary production in any case—have become far\n\n\nless important in recent years. For example, Senegal eliminated export taxes in the mid-1980s\n\nand Uganda abolished export taxes in the mid-1990s. <sup>6</sup> [^6: Mbaye and Golub (2003) note that Senegal had few export taxes other than on groundnuts and gold and that these were\neliminated in 1985. Similarly, Grenier _et al_ . (1999) note that export duties had become uncommon by the mid-1990s in\nTanzania. Milner _et al_ . (2000) note a similar point for Uganda.]\n\n\nIn addition to export taxes, tariffs and non-tariff barriers on imports might also affect\n\n\nexports. By increasing the cost of imports,\n\nMention: \"investment climate surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"They show small and medium firms export mainly to neighbouring markets.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002848:1:0", "occurrence_id": "7aed3d2f65f558c9c1fecbe5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Rural-urban gradient in Latin America and the Caribbean_ _page_ 1\n\n## **Motivation and overview**\n\n\nHow many rural people live in Latin American and the Caribbean (LAC)? The question is\nsimple to pose but not well-defined. “Rurality” has many aspects and connotations. For\npolicy purposes we need to specify precisely which aspects of rurality interest us, and define\ncriteria for identifying these aspects. Moreover, we should not expect that any of this criteria\nexhibit marked break points between rural and urban areas or people. However defined,\nrurality is a gradient, not a discrete condition.\n\n\nHere we apply two criteria for defining the rurality: population density, and remoteness from\nlarge cities. We argue that these criteria constitute important gradients along which\neconomic behavior and appropriate development interventions might vary substantially.\nWhere population densities are low, markets of all kinds are thin, and unit costs of delivering\nmost social services and many types of infrastructure are high. Where large urban areas are\ndistant, farmgate (or factory-gate) prices of outputs will be low and prices of inputs will be\nhigh, and it will be difficult to recruit skilled personnel to public service or private\nenterprises. Thus remoteness and low population density together define a set of rural areas\nthat face special challenges in development.\n\n\nRemoteness and population density alone, however, cannot begin to capture the\nheterogeneity that exists within rural areas. Many aspects of physical geography shape\nopportunities and constraints for rural development, and environment-development\nlinkages. Here we focus on two summary measures of physical geography: agroclimatic\nsuitability for cropping, and forest cover.\n\n\nWith rurality criteria and descriptors defined, it is possible to tabulate rural populations using\none of the existing global gridded population coverages. The two main candidate datasets –\nLandscan, and the GPW – each have advantages and disadvantages. We chose to use the\nGPW 3, in part out of anticipation of a\n\nMention: \"global gridded population coverages\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Rural populations are tabulated using one of these existing coverages.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002848:1:0", "occurrence_id": "bcdf77e5b35789048f6e9b24", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Rural-urban gradient in Latin America and the Caribbean_ _page_ 1\n\n## **Motivation and overview**\n\n\nHow many rural people live in Latin American and the Caribbean (LAC)? The question is\nsimple to pose but not well-defined. “Rurality” has many aspects and connotations. For\npolicy purposes we need to specify precisely which aspects of rurality interest us, and define\ncriteria for identifying these aspects. Moreover, we should not expect that any of this criteria\nexhibit marked break points between rural and urban areas or people. However defined,\nrurality is a gradient, not a discrete condition.\n\n\nHere we apply two criteria for defining the rurality: population density, and remoteness from\nlarge cities. We argue that these criteria constitute important gradients along which\neconomic behavior and appropriate development interventions might vary substantially.\nWhere population densities are low, markets of all kinds are thin, and unit costs of delivering\nmost social services and many types of infrastructure are high. Where large urban areas are\ndistant, farmgate (or factory-gate) prices of outputs will be low and prices of inputs will be\nhigh, and it will be difficult to recruit skilled personnel to public service or private\nenterprises. Thus remoteness and low population density together define a set of rural areas\nthat face special challenges in development.\n\n\nRemoteness and population density alone, however, cannot begin to capture the\nheterogeneity that exists within rural areas. Many aspects of physical geography shape\nopportunities and constraints for rural development, and environment-development\nlinkages. Here we focus on two summary measures of physical geography: agroclimatic\nsuitability for cropping, and forest cover.\n\n\nWith rurality criteria and descriptors defined, it is possible to tabulate rural populations using\none of the existing global gridded population coverages. The two main candidate datasets –\nLandscan, and the GPW – each have advantages and disadvantages. We chose to use the\nGPW 3, in part out of anticipation of a\n\nMention: \"Landscan\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"One of two candidate grids, each with advantages and disadvantages.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002848:1:0", "occurrence_id": "0e20ebed37bb31f9eef04f6e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Rural-urban gradient in Latin America and the Caribbean_ _page_ 1\n\n## **Motivation and overview**\n\n\nHow many rural people live in Latin American and the Caribbean (LAC)? The question is\nsimple to pose but not well-defined. “Rurality” has many aspects and connotations. For\npolicy purposes we need to specify precisely which aspects of rurality interest us, and define\ncriteria for identifying these aspects. Moreover, we should not expect that any of this criteria\nexhibit marked break points between rural and urban areas or people. However defined,\nrurality is a gradient, not a discrete condition.\n\n\nHere we apply two criteria for defining the rurality: population density, and remoteness from\nlarge cities. We argue that these criteria constitute important gradients along which\neconomic behavior and appropriate development interventions might vary substantially.\nWhere population densities are low, markets of all kinds are thin, and unit costs of delivering\nmost social services and many types of infrastructure are high. Where large urban areas are\ndistant, farmgate (or factory-gate) prices of outputs will be low and prices of inputs will be\nhigh, and it will be difficult to recruit skilled personnel to public service or private\nenterprises. Thus remoteness and low population density together define a set of rural areas\nthat face special challenges in development.\n\n\nRemoteness and population density alone, however, cannot begin to capture the\nheterogeneity that exists within rural areas. Many aspects of physical geography shape\nopportunities and constraints for rural development, and environment-development\nlinkages. Here we focus on two summary measures of physical geography: agroclimatic\nsuitability for cropping, and forest cover.\n\n\nWith rurality criteria and descriptors defined, it is possible to tabulate rural populations using\none of the existing global gridded population coverages. The two main candidate datasets –\nLandscan, and the GPW – each have advantages and disadvantages. We chose to use the\nGPW 3, in part out of anticipation of a\n\nMention: \"GPW\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Named as the second candidate grid.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002848:1:0", "occurrence_id": "9ab3f08e97cc6fa3b7266411", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Rural-urban gradient in Latin America and the Caribbean_ _page_ 1\n\n## **Motivation and overview**\n\n\nHow many rural people live in Latin American and the Caribbean (LAC)? The question is\nsimple to pose but not well-defined. “Rurality” has many aspects and connotations. For\npolicy purposes we need to specify precisely which aspects of rurality interest us, and define\ncriteria for identifying these aspects. Moreover, we should not expect that any of this criteria\nexhibit marked break points between rural and urban areas or people. However defined,\nrurality is a gradient, not a discrete condition.\n\n\nHere we apply two criteria for defining the rurality: population density, and remoteness from\nlarge cities. We argue that these criteria constitute important gradients along which\neconomic behavior and appropriate development interventions might vary substantially.\nWhere population densities are low, markets of all kinds are thin, and unit costs of delivering\nmost social services and many types of infrastructure are high. Where large urban areas are\ndistant, farmgate (or factory-gate) prices of outputs will be low and prices of inputs will be\nhigh, and it will be difficult to recruit skilled personnel to public service or private\nenterprises. Thus remoteness and low population density together define a set of rural areas\nthat face special challenges in development.\n\n\nRemoteness and population density alone, however, cannot begin to capture the\nheterogeneity that exists within rural areas. Many aspects of physical geography shape\nopportunities and constraints for rural development, and environment-development\nlinkages. Here we focus on two summary measures of physical geography: agroclimatic\nsuitability for cropping, and forest cover.\n\n\nWith rurality criteria and descriptors defined, it is possible to tabulate rural populations using\none of the existing global gridded population coverages. The two main candidate datasets –\nLandscan, and the GPW – each have advantages and disadvantages. We chose to use the\nGPW 3, in part out of anticipation of a\n\nMention: \"GPW 3\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Chosen to construct the rural population estimates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002853:16:2", "occurrence_id": "3747299ca30f0baaaad0fbfb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  equity + book value of liabilities over total\nassets. Real Sales Growth is sales growth adjusted for inflation (obtained from WDI). Interest coverage is defined as earnings before interest and taxes over\ninterest payments. Gross PPE is property plan and equipment (i.e., fixed assets) at purchase cost, net PPE excludes accumulated depreciation. The t-test\nindicates significance for equality of the medians (or means) between years 1996 and 2000. ***,**,*, and <sup>a</sup> indicate significance at 1 percent,5 percent,\n10 percent, and 15 percent, respectively.\n\n\n\n1995\n\n\n\n1996\n\n\n\n1997\n\n\n\n1998\n\n\n\n1999\n\n\n\n2000\n\n\n\n17\n\nMention: \"WDI\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Its inflation figures deflate real sales growth.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002859:37:3", "occurrence_id": "89ae43a194d8e27ce95c7faa", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: (.00)|37.22<br>(.00)|25.33<br>(.00)|24.74<br>(.01)|\n\n\n\nNote:\n1. - 0.01 significance level, ** 0.05 significance level, *** 0.10 significance level, and insignificant otherwise\n2. EDU2-2: Inputs: orthogonalized public spending on education per capita, teachers per pupil\nOutputs: gross primary and secondary enrollments\nEDU2-2n: same inputs as EDU2-2, outputs: net primary and secondary enrollment\nEDU3-2: literacy of adult is added to EDU2-2 as input\nEDU3-2n: literacy of adult is added to EDU2-2n as input\nEDU3-3: literacy of youth is added to EDU3-2 as output\nEDU3-3bl: same inputs as in EDU3-2,\n\noutputs: average years of school, first level complete, and second level complete (Barro-Lee education indicators)\nHEA2-2: Inputs: orthogonalized public spending on health per capita, literacy of adult\nOutputs: life expectancy at birth, and immunization DPT\nHEA3-2: orthogonalized private spending on health per capita is added to HEA2-2 as input\nHEA3-2m: Immunization Measles is in place of DPT in HEA3-2 as output\nHEA3-3: Immunization Measles is added to HEA3-2 as output\n\n\n36\n\nMention: \"Barro-Lee education indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Average years of schooling and completion levels are taken as outputs.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002860:2:1", "occurrence_id": "14b2742ac373519633279a86", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  the poor. However, informal sector\nlenders remain a strong presence in rural India, delivering finance to the poor, the vast majority of\nwhom still do not have access to either formal or semi-formal (microfinance) sectors. Scaling-up\naccess to finance for India’s rural poor, to meet their diverse financial needs (savings, credit,\ninsurance against unexpected events, etc.) through flexible products at competitive prices,\npresents a formidable challenge in a country as vast and varied as India. But the opportunities,\ntoo, are plentiful, and government has an important role to play in creating space and a flexible\narchitecture for innovations.\n\nThis paper begins with a brief review in Section II of the current level and pattern of access to\nformal finance for India’s rural households, with a focus on the poor. The review draws on a\nrecent survey conducted jointly by the World Bank and the National Council of Applied\nEconomic Research, India (NCAER) - the _Rural Finance Access Survey, 2003_ (RFAS-2003). It\nhighlights inadequacies in rural access to _formal_ finance and the extortionary terms of _informal_\nfinance, which provide a strong need and ample space for innovative _microfinance_ approaches—\nthat combine the safety and reliability of _formal_ finance with the convenience and flexibility of\n_informal_ finance—to serve the financial needs of India’s rural poor. Section III examines some\nof the key microfinance initiatives in India over the past decade. Section IV takes a closer look at\nthe most dominant among the recent microfinance approaches, SHG Bank Linkage. It evaluates\nthe success with which this approach has been able to reach out to the poor, analyzes the reasons\n\n\n1 The focus on poverty and finance was articulated most famously in the 1954 Reserve Bank of India (RBI)\nreport on the All-India Rural Credit Survey of 1951-52 (RBI, 1954).\n2 For a good discussion on the link between finance\n\nMention: \"Rural Finance Access Survey, 2003\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"It documents inadequate formal rural finance and extortionate informal terms.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002861:9:0", "occurrence_id": "fb1fcbaf1b1b685e5cbbc09a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: strong son preference, puts pressure on reducing the number of girls in the family. <sup>21</sup> [^21: Das Gupta and Bhat 1997. On son preference more broadly, see Miller 1981, Dyson and Moore\n1983, and Das Gupta 1987. On gender differentials in tribal populations, see Maharatna 2000.]\nHigher socio-economic groups have greater access to sex-selective abortion, and are\ntherefore in a position to care more equally for the children that are born. Lower socioeconomic groups have limited access to prenatal sex selection technology, and this may\nlead to unequal treatment of children who are unwanted. There is, for example,\nconsiderable evidence that girls are less likely than boys to be taken for medical\ntreatment, and the quality of treatment sought also differs ⎯ which could impact on their\nanthropometric status.\n\n**5.** **Assessment of the ICDS’ Program Placement**\n\nCoverage of the ICDS program is high: a substantial proportion of India’s villages are\ncovered by the ICDS today, and this number rose sharply during the 1990s. Of the\nvillages sampled by the NFHS, a third had an ICDS program in place in 1992, and more\nthan a half of the surveyed villages had it in 1998 (Table 2). A few states, notably Kerala,\nhad virtually complete coverage already by 1992. Program coverage is especially high in\nthe southern region, the northeastern region, and the non-poor states of the northern\nregion. The apparent decline in coverage in Tamil Nadu during the 1990s is probably the\nresult of re-classification of the nutritional supplementation program to the TINP (Tamil\nNadu Integrated Nutrition Program), which is broadly similar in concept to the ICDS.\n\n(i) Program placement across states\n\nFor ease of discussion, we have grouped India’s states by region: the South, the\nNortheast, <sup>22</sup> [^22: Sikkim is not included where data are required for 1992, since these are not available.] and the\n\nMention: \"NFHS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its village sample records the share with an ICDS programme over time.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002865:47:2", "occurrence_id": "a39c1bca391f156833d9ce4b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |\n|---|---|---|---|---|---|\n|Korea|27|41|39|36|40|\n\n\nSource : IMD 2002; recited from Jang, Ha-Sung(2003)\n\n|< Part B. WEF Survey|>|Col3|\n|---|---|---|\n|Country|Ranking|Ranking|\n|Korea|Competitiveness|Corporate Boards|\n|Korea|28|61|\n\n\n\nSource : WEF 2002; recited from Jang, Ha-Sung(2003)\n\n\n\n\n\n\n\n\n```\n48\n\n```\n\nMention: \"WEF Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its competitiveness and corporate board rankings are reproduced.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002873:33:2", "occurrence_id": "1008ff85d2a6d5964f563776", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: reported health status, or in\nthe relationship of health status to income level (FTB Tables 7.3, 7.4). The statistically\nsignificant reductions in the proportion of children with stunting (FTB Table 7.5), limit the extent\nto which the effect of trend improvements in nutrition and healthcare on the condition of the poor\nwere immediately interrupted by the crisis. Nevertheless, the IFLS healthcare professionals\nreported a systematic deterioration in respondents’ overall health status with, apparently, the\nlargest deterioration in the lowest income categories (FTB p. 87 and Table 7.11).\n\n\n26 Declines in median hourly wage rates are larger than those reported for PCEs (, suggesting consumption\nsmoothing, though whether this employed savings or borrowing is not known (FTB Table 3.2).\nParticipation in paid employment did not change much, but there was a considerable increase in female\nparticipation in unpaid (family) work.\n27 For the shorter period 1997-8, measuring the immediate impact on the panel of households in IFLS2 and\n2+, Thomas and Frankenberg (2005) report that the top quartile showed the largest percentage declines in\nPCE; the bottom quartile also suffered large percentage declines, with the middle income groups relatively\nunaffected. As measured by the standard deviation of log PCE, inequality declined.\n28 The relationships here are not all statistically significant (FTB, Table 5.3, 5.4).\n\n\n33\n\nMention: \"IFLS2 and\n2+\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"A cited study uses these rounds to report consumption declines by quartile.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002874:31:3", "occurrence_id": "55307db657599748b51e2a10", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  World Development Indicators.\nGDP\n\nCredit via banks Average domestic credit provided by banking sector (% of GDP) in the pre-crisis period. Source: World\nto GDP Development Indicators.\n\n\n\nAverage domestic credit provided by banking sector (% of GDP) in the pre-crisis period. Source: World\nDevelopment Indicators.\n\n\n\nInflation Average inflation in the pre-crisis period. Source: World Development Indicators.\n\n<u>GDP per capita</u> <u>Average GDP per capita in 1995 US $ in the pre-crisis period. Source: World Development Indicators.</u>\n\n\n32\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Pre-crisis domestic credit provided by banks as a share of GDP is taken from it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002881:17:1", "occurrence_id": "7579a02acf22505f8d67223c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  annual number of Bunionectomies (toedeformity repairs). In some cases, we had to rely on outpatient volume estimates from\nhealth care research firms, e.g., OptiStock, a vision care research firm, provided estimates\non the annual number of cataract procedures. See appendix 3 for the detailed list of\nsources from which outpatient volume data was estimated.\n\n_Foreign prices_\n\nVanbreda International, a Belgium-based employee benefits consulting and\nadministration firm, whose clients include about 124 international organizations with\noperations in 185 countries, provided the data on international prices for the selected\nprocedures in 21 countries from their own claims (see appendix 4 for details). In order to\nmake a meaningful comparison between US and international prices from a trade\nstandpoint, we added the cost of round-trip airfare to these countries from the US to the\ninternational prices. Since the psychological costs of going abroad are hard to quantify,\nand our focus is on the monetary gains from trade, we do not include the former in our\ncost comparison. Further, since there can be a lot of intra-procedural variation in prices,\nwe used the average of the three lowest foreign country prices for each procedure as the\nbenchmark foreign price (Column 5, Table 4). This minimizes any price bias arising\nfrom the paucity of claims data in certain countries.\n\n\n24 The number of time units can be larger than 4 for a major surgery, while only one time unit may be\ncharged for a minor procedure.\n\n\n18\n\nMention: \"outpatient volume data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Volumes are estimated from listed sources to derive cataract procedure counts.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002885:6:0", "occurrence_id": "b2cb93dfb69f4a27f51551d7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: as a VAT payer (SARS,2003). Some services provided within the financial sector are also\nexempt. (International VAT Monitor,1995:376).\n\nIn this paper, we take the economic and tax structure of the SAM, including the VAT and its\nexceptions, as the basis for the CGE model and we examine possible tax reforms that will\nimprove welfare impact.\n\n**2.2** **Specification of a VAT with Rebates on Intermediate Inputs**\n\nRather than treat consumption tax payments as a retail sales tax, we specify a European-style\nvalue added tax with rebates for intermediate input and investment purchases. With this\ntreatment, there is no cascading effect on prices of taxes on intermediate goods.\n\nWe assume the VAT is administered using the “invoice method.” All transactions are taxed at a\nfixed proportional rate regardless of whether they are final or intermediate transactions. Firms\ncan deduct taxes paid on intermediate inputs and that tax amount is reported on the invoices for\nintermediates. Import sales are subject to a VAT, export sales are not.\n\nTo calibrate the model, we assume the observed VAT tax revenue in the SAM is the gross\npayment by sector. <sup>8</sup> [^8: We use the effective tax rates based on tax collection data. The rates differ by sector and differ from the reported\nstatutory rate of 14% due to sector aggregation, the treatment of the informal sector and/or measurement error.] Producers must pay the VAT on retail sales and can collect a rebate on taxes\npaid by intermediate input use. For each production activity, we compute the rebate on\nintermediate expenditure using data on input-output relationships. We assume a Leontief\nproduction structure with fixed input-output shares defining intermediate demand.\n\nThe rebate to activity, A, is based on total intermediate input use:\n\n\n\n_tvat_ ( _C_ ) ⋅ _PQ_ ( _C_ ) ⋅\n\n= ∑\n\n\n\n_REBATE_ ( _A_ ) _tvat_ ( _C_ ) ⋅ _PQ_ ( _C_ ) ⋅ _QINT_ ( _C_, _A_ )\n\n=\n\nMention: \"SAM\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its economic and tax structure, including VAT exceptions, calibrates the model.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002893:31:1", "occurrence_id": "331825a883dab576611f35f0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  value added for each commodity _i_ . Let _Ei_\ndenote exports of each commodity. The first row in the table, the labor-output ratio\n\n\n\n\n<u>�</u>\n\n\n\nfor exports is then\n\n\n\n_<u>i</u>_ <sup>(</sup> <sup>_Li×Ei_</sup> <sup>)</sup>\n\n<u>�</u>\n\n\n\n. Commodities are classified as _untaxed_ if the taxes\n_i_ <sup>_Ei_</sup>\n\n\n\npaid on the commodity constitute less than 5% of the value of domestic output\nof the commodity, and _domestic_ if more than 5% is taxed. Let _Uj_ be the valueadded of untaxed goods. The labor-output ratio for _untaxed_ is then calculated\n\n\n\n\n\n\n\nas\n\n\n\n_<u>j</u>_ <sup>(</sup> <sup>_Lj_</sup> <sup>_×Uj_</sup> <sup>)</sup>\n\n<u>�</u>\n\n\n\n. The same procedure is used to calculate the labor-output ratio for\n_j_ <sup>_Uj_</sup>\n\n\n\n_domestic_ .\n\nThe administrative costs of tax collection presented in Table 5 are derived\nfrom various sources. For the United States: IRS Data Book, FY2002, available at: www.irs.gov/pub/irs-soi/02db30cs.xls; For other OECD countries - authors’ calculations based on the following tax agency annual reports: Australia ATO Annual Report 2002, available at www.ato.gov.au; United Kingdom - Inland Revenue, Annual Report for the year\n\nMention: \"IRS Data Book\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its figures give the administrative costs of tax collection in the United States.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002893:31:1", "occurrence_id": "74a92114d9968c9cb57c75ef", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  value added for each commodity _i_ . Let _Ei_\ndenote exports of each commodity. The first row in the table, the labor-output ratio\n\n\n\n\n<u>�</u>\n\n\n\nfor exports is then\n\n\n\n_<u>i</u>_ <sup>(</sup> <sup>_Li×Ei_</sup> <sup>)</sup>\n\n<u>�</u>\n\n\n\n. Commodities are classified as _untaxed_ if the taxes\n_i_ <sup>_Ei_</sup>\n\n\n\npaid on the commodity constitute less than 5% of the value of domestic output\nof the commodity, and _domestic_ if more than 5% is taxed. Let _Uj_ be the valueadded of untaxed goods. The labor-output ratio for _untaxed_ is then calculated\n\n\n\n\n\n\n\nas\n\n\n\n_<u>j</u>_ <sup>(</sup> <sup>_Lj_</sup> <sup>_×Uj_</sup> <sup>)</sup>\n\n<u>�</u>\n\n\n\n. The same procedure is used to calculate the labor-output ratio for\n_j_ <sup>_Uj_</sup>\n\n\n\n_domestic_ .\n\nThe administrative costs of tax collection presented in Table 5 are derived\nfrom various sources. For the United States: IRS Data Book, FY2002, available at: www.irs.gov/pub/irs-soi/02db30cs.xls; For other OECD countries - authors’ calculations based on the following tax agency annual reports: Australia ATO Annual Report 2002, available at www.ato.gov.au; United Kingdom - Inland Revenue, Annual Report for the year\n\nMention: \"Australia ATO Annual Report 2002\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"One tax agency report behind the comparative administrative costs.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002894:38:0", "occurrence_id": "38c22413516226e354a6a78b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 1** : Rich-countries’ GDP, average DM index, terms of trade, and real interest\nrate\nRich countries’ GDP is the GDP per capita of high income OECD countries expressed in 1995 US$. The DM\nindex is the Deaton-Miller commodity based terms of trade index, terms of trade is the ratio of exports to import\nprices from Penn World Tables 6.1. The international real interest rate corresponds to the nominal six month\nLIBOR offer rate in US dollars deflated by the change in the US Producer’s Price Index. All growth rates\ncorrespond to the change in the log of the different variables. The three series have been demeaned and\nstandardized to highlight their co-movement.\n**Panel A**\n\n\n\n\n\n**Panel B**\n\n\n\n\n\n38\n\nMention: \"Penn World Tables\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Export and import price ratios give the terms of trade.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002894:38:0", "occurrence_id": "2b5eda1cf7b4d06bc942c0c1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 1** : Rich-countries’ GDP, average DM index, terms of trade, and real interest\nrate\nRich countries’ GDP is the GDP per capita of high income OECD countries expressed in 1995 US$. The DM\nindex is the Deaton-Miller commodity based terms of trade index, terms of trade is the ratio of exports to import\nprices from Penn World Tables 6.1. The international real interest rate corresponds to the nominal six month\nLIBOR offer rate in US dollars deflated by the change in the US Producer’s Price Index. All growth rates\ncorrespond to the change in the log of the different variables. The three series have been demeaned and\nstandardized to highlight their co-movement.\n**Panel A**\n\n\n\n\n\n**Panel B**\n\n\n\n\n\n38\n\nMention: \"US Producer’s Price Index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Its change deflates the LIBOR rate to a real interest rate.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002909:17:0", "occurrence_id": "0589f2c560e941b1a0c68f1c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ### 3 Empirical Results\n\nI exemplify the use of the method by estimating the responses of quantities consumed and\n\n\nagricultural wage income in rural Mexico. The case of Mexico is, in principle, a good case\n\n\nstudy because of the importance of the agricultural sector. Mexico is also close to the United\n\n\nStates, and thus WTO reforms are likely to reach rural producers.\n\n\nIn Mexico, a good source of data is the Household Income and Expenditure National\n\n\nSurveys, ENIGH (Encuesta Nacional de Ingresos y Gastos de los Hogares). During the\n\n\n1990s, the data were collected every other year; I use data from the 1994, 1996, 1998, and\n\n\n2000 rounds. The surveys are representative of the population at the National level, and\n\n\ncover rural and urban areas. Since I am focusing on the impacts of price reforms on wage\n\n\nagricultural income, I use the rural modules only. Table 1 reports some summary statistics.\n\n\nI begin with sample sizes and number of clusters c. The ENIGH is carried out at different\n\n\nweeks in different districts. To make the assumption of common prices in a given cluster\n\n\nc reasonable, I define them as district-week pairs. There are between 3306 and 5007 rural\n\n\nhouseholds in the samples. There are 1095 clusters in the pooled sample, approximately 275\n\n\nin each ENIGH round. The number of households interviewed in each “cluster” is typically\n\n\nlarger than in other surveys, with at least 20 households in each of them.\n\n\nOn the consumption side, expenditure data include food and non-food items. Since\n\n\nquantities are only reported for foodstuff, I estimate the model for the following six goods:\n\n\ncorn, wheat, dairy, oils & fats, meat, and fruits & vegetables. These goods are aggregates\n\n\nof differentiated goods within each category. For instance, corn includes own-consumption,\n\n\ncorn grain, and finely grinded corn; meat includes different types of beef, poultry and pork\n\nMention: \"Household Income and Expenditure National\n\n\nSurveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"It is the good source of data reaching rural producers.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002909:17:0", "occurrence_id": "bb05d07f9ea494481c884fba", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ### 3 Empirical Results\n\nI exemplify the use of the method by estimating the responses of quantities consumed and\n\n\nagricultural wage income in rural Mexico. The case of Mexico is, in principle, a good case\n\n\nstudy because of the importance of the agricultural sector. Mexico is also close to the United\n\n\nStates, and thus WTO reforms are likely to reach rural producers.\n\n\nIn Mexico, a good source of data is the Household Income and Expenditure National\n\n\nSurveys, ENIGH (Encuesta Nacional de Ingresos y Gastos de los Hogares). During the\n\n\n1990s, the data were collected every other year; I use data from the 1994, 1996, 1998, and\n\n\n2000 rounds. The surveys are representative of the population at the National level, and\n\n\ncover rural and urban areas. Since I am focusing on the impacts of price reforms on wage\n\n\nagricultural income, I use the rural modules only. Table 1 reports some summary statistics.\n\n\nI begin with sample sizes and number of clusters c. The ENIGH is carried out at different\n\n\nweeks in different districts. To make the assumption of common prices in a given cluster\n\n\nc reasonable, I define them as district-week pairs. There are between 3306 and 5007 rural\n\n\nhouseholds in the samples. There are 1095 clusters in the pooled sample, approximately 275\n\n\nin each ENIGH round. The number of households interviewed in each “cluster” is typically\n\n\nlarger than in other surveys, with at least 20 households in each of them.\n\n\nOn the consumption side, expenditure data include food and non-food items. Since\n\n\nquantities are only reported for foodstuff, I estimate the model for the following six goods:\n\n\ncorn, wheat, dairy, oils & fats, meat, and fruits & vegetables. These goods are aggregates\n\n\nof differentiated goods within each category. For instance, corn includes own-consumption,\n\n\ncorn grain, and finely grinded corn; meat includes different types of beef, poultry and pork\n\nMention: \"ENIGH\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Collection in different weeks across districts complicates the common price assumption.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002923:11:1", "occurrence_id": "273098515eb1a73520732336", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  from the firms. This prevents us from validating the financial information\n\n\nprovided by the firms from public records. However, the WBES has been widely used\n\n\nand most recently for its firm-level variation by Beck, Demirguc-Kunt, and Maksimovic\n\n\n(2005). It has also been used by Transparency International in computing its Corruption\n\n\n12\n\nMention: \"the WBES\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Its firm-level variation is widely used although public records cannot validate financial responses.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002936:17:0", "occurrence_id": "7007e75653648744107756f8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: merged with the relevant cycle’s CN files. The House and Senate files were separated for\n\n\neach cycle. Congressional committee and sub-committee assignments for each Congress\n\n\nwere obtained from Congressional Quarterly (1991-1999).\n\n\nMapping from PACs to SIC-based agricultural-related sector makes use of the\n\nconcordance by Beaulieu and Magee (2002). <sup>10</sup> [^10: I am grateful to Chris Magee for making this available.] In general the mapping is many-PACs-to\n\nmany-SIC-codes. For many-to-one maps, the contributions were simply aggregated for each\n\n\nSIC code. For one-to-many maps, political contributions from each PAC were fractionally\n\n\nassigned equally to each SIC code into which the PAC mapped. To check the consistency of\n\n\nthe PAC data, they were compared with the data on the opensecrets.com website. Our data\n\n\nare very comparable in the aggregate, as well as those sectors for which opensecrets.com does\n\n\nreport data.\n\n\n_B._ _PAC Spending Across Agriculture-Related Sectors_\n\n\nFigure 2 shows the breakdown of PAC contributions by 2-digit SIC agriculture-related sectors\n\n\nin the 1991-92 election cycle (first bar) and the 1999-2000 cycle (second bar). The striking\n\n\nfeature here is that the contributions have held fairly steady over the ten-year period. It\n\n\nappears that the same PACs have contributed fairly predictable amounts of money over this\n\n\nperiod. Rather than use the 2-digit descriptions, subsequent figures break down farm PACs\n\n\nby seven products—wheat, dairy, sugar, vegetables and fruits, cotton, ranch and other. Figure\n\n\n3 depicts the total contribution by agriculture-related sectors for each of the five election\n\n\ncycles between 1991 and 2000. In order to compare farm PAC contributions with other\n\n\nagricultural related sectors, Figure 3 also shows contributions by PACs in the following major\n\n\nsectors: Farm Equipment, Agricultural Services, Crop Processing Services, Distribution,\n\n\nForestry/Nursery, and Food\n\nMention: \"CN files\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"House and Senate records are separated per cycle and merged with committee assignments.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002936:17:0", "occurrence_id": "45a6a2265d9c170556218532", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: merged with the relevant cycle’s CN files. The House and Senate files were separated for\n\n\neach cycle. Congressional committee and sub-committee assignments for each Congress\n\n\nwere obtained from Congressional Quarterly (1991-1999).\n\n\nMapping from PACs to SIC-based agricultural-related sector makes use of the\n\nconcordance by Beaulieu and Magee (2002). <sup>10</sup> [^10: I am grateful to Chris Magee for making this available.] In general the mapping is many-PACs-to\n\nmany-SIC-codes. For many-to-one maps, the contributions were simply aggregated for each\n\n\nSIC code. For one-to-many maps, political contributions from each PAC were fractionally\n\n\nassigned equally to each SIC code into which the PAC mapped. To check the consistency of\n\n\nthe PAC data, they were compared with the data on the opensecrets.com website. Our data\n\n\nare very comparable in the aggregate, as well as those sectors for which opensecrets.com does\n\n\nreport data.\n\n\n_B._ _PAC Spending Across Agriculture-Related Sectors_\n\n\nFigure 2 shows the breakdown of PAC contributions by 2-digit SIC agriculture-related sectors\n\n\nin the 1991-92 election cycle (first bar) and the 1999-2000 cycle (second bar). The striking\n\n\nfeature here is that the contributions have held fairly steady over the ten-year period. It\n\n\nappears that the same PACs have contributed fairly predictable amounts of money over this\n\n\nperiod. Rather than use the 2-digit descriptions, subsequent figures break down farm PACs\n\n\nby seven products—wheat, dairy, sugar, vegetables and fruits, cotton, ranch and other. Figure\n\n\n3 depicts the total contribution by agriculture-related sectors for each of the five election\n\n\ncycles between 1991 and 2000. In order to compare farm PAC contributions with other\n\n\nagricultural related sectors, Figure 3 also shows contributions by PACs in the following major\n\n\nsectors: Farm Equipment, Agricultural Services, Crop Processing Services, Distribution,\n\n\nForestry/Nursery, and Food\n\nMention: \"PAC data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Consistency is checked against a public transparency website.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002948:30:0", "occurrence_id": "8a4a82fd161273acc56b4a8e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Table 1: Distribution of Sampled Plots by Document Type in Buriram\n\n\n<u>Before</u> <u>LTP</u> <u>After</u> <u>LTP</u>\nAgricultural Residential Agricultural Residential\n<u>Plots</u> <u>Plots</u> <u>Plots</u> <u>Plots</u>\n_Document_ _Type_\n\nNS-4 0.10 0.04 0.40 0.30\nNS-3 or NS-3K 0.30 0.20 0.16 0.24\nOther 0.60 0.76 0.44 0.46\n\n\n<u>N.</u> <u>Observations</u> <u>278</u> <u>171</u> <u>749</u> <u>1,282</u>\n<u>Note:</u> <u>An</u> <u>observation</u> <u>is</u> <u>a</u> <u>plot.</u> <u>Data</u> <u>from</u> <u>before</u> <u>the</u> <u>LTP</u> <u>come</u> <u>from</u> <u>Table</u> <u>1</u> <u>in</u>\nChalamwong et al.(1988) study. Data after the LTP come from the Townsend-Thai\ndata set.\n\n\nTable 2: Land Reform Areas and Beneficiaries: 1975-2003\n\n\n<u>Public</u> <u>Land</u> <u>Private</u>\n\nMention: \"the Townsend-Thai\ndata set\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Household outcomes after the land titling programme come from this panel.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002887:7:1", "occurrence_id": "7eaf5dbff5a0897efcc43e8f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  growth regressions, based on three- or\nfive-year averages; second in an event study where we use all the available time variation.\nOur control variables in the growth analysis are classic and inspired from the robust sets\nproposed in Levine and Renelt (1992). <sup>10</sup> [^10: Doppelhofer, Miller and Sala-i-Martin (2004) recently used Bayesian techniques to isolate an alternative,\nupdated set of control variables. The approach is however purely cross-sectional, and its results cannot be\nused in the present context where within-country growth determinants are of the essence. For instance, the\nmost robust correlates of economic growth include the extents of confucianistic or protestant religions, or\ngeographic binary variables, none of which lend themselves to a panel estimation.] They include initial income, openness to trade,\npopulation growth, secondary schooling and the growth rate of the terms of trade. We also\nexperimented with the fiscal balance, with no changes in conclusions. Construction of these\nvariables is standard; the sources used are the World Development Indicators, the World\nEconomic Outlook and the International Financial Statistics.\n\n\nIn the event study, we track the response over time of investment, macroeconomic policy,\nand the terms of borrowing. Of these, only investment is readily available and its measure\nrelatively uncontroversial. We choose to investigate the dynamic response of two policy\nrelated variables: inflation and government expenditures. If the conduct of economic policy\ndoes indeed deteriorate at high levels of debt, we conjecture that at least one of these\nmeasures will show a systematic response. We also include an index computed by the World\nBank, the Country Performance International Assessment, or CPIA, meant to summarize in\none number an assessment of the overall quality of policy stance. <sup>11</sup> [^11: These data are confidential.] Because we are interested\nin the terms under which debt is contracted, we finally bring in additional information from\nGlobal Development Finance publications, with a view to isolating changes in the rate of\ninterest\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"One of three standard sources for the macro variables.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002887:7:1", "occurrence_id": "0ce804ac08da90dde5f92ae7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  growth regressions, based on three- or\nfive-year averages; second in an event study where we use all the available time variation.\nOur control variables in the growth analysis are classic and inspired from the robust sets\nproposed in Levine and Renelt (1992). <sup>10</sup> [^10: Doppelhofer, Miller and Sala-i-Martin (2004) recently used Bayesian techniques to isolate an alternative,\nupdated set of control variables. The approach is however purely cross-sectional, and its results cannot be\nused in the present context where within-country growth determinants are of the essence. For instance, the\nmost robust correlates of economic growth include the extents of confucianistic or protestant religions, or\ngeographic binary variables, none of which lend themselves to a panel estimation.] They include initial income, openness to trade,\npopulation growth, secondary schooling and the growth rate of the terms of trade. We also\nexperimented with the fiscal balance, with no changes in conclusions. Construction of these\nvariables is standard; the sources used are the World Development Indicators, the World\nEconomic Outlook and the International Financial Statistics.\n\n\nIn the event study, we track the response over time of investment, macroeconomic policy,\nand the terms of borrowing. Of these, only investment is readily available and its measure\nrelatively uncontroversial. We choose to investigate the dynamic response of two policy\nrelated variables: inflation and government expenditures. If the conduct of economic policy\ndoes indeed deteriorate at high levels of debt, we conjecture that at least one of these\nmeasures will show a systematic response. We also include an index computed by the World\nBank, the Country Performance International Assessment, or CPIA, meant to summarize in\none number an assessment of the overall quality of policy stance. <sup>11</sup> [^11: These data are confidential.] Because we are interested\nin the terms under which debt is contracted, we finally bring in additional information from\nGlobal Development Finance publications, with a view to isolating changes in the rate of\ninterest\n\nMention: \"World\nEconomic Outlook\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The second standard source behind the macro variables.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002887:7:1", "occurrence_id": "497854990d85c916284fcedc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  growth regressions, based on three- or\nfive-year averages; second in an event study where we use all the available time variation.\nOur control variables in the growth analysis are classic and inspired from the robust sets\nproposed in Levine and Renelt (1992). <sup>10</sup> [^10: Doppelhofer, Miller and Sala-i-Martin (2004) recently used Bayesian techniques to isolate an alternative,\nupdated set of control variables. The approach is however purely cross-sectional, and its results cannot be\nused in the present context where within-country growth determinants are of the essence. For instance, the\nmost robust correlates of economic growth include the extents of confucianistic or protestant religions, or\ngeographic binary variables, none of which lend themselves to a panel estimation.] They include initial income, openness to trade,\npopulation growth, secondary schooling and the growth rate of the terms of trade. We also\nexperimented with the fiscal balance, with no changes in conclusions. Construction of these\nvariables is standard; the sources used are the World Development Indicators, the World\nEconomic Outlook and the International Financial Statistics.\n\n\nIn the event study, we track the response over time of investment, macroeconomic policy,\nand the terms of borrowing. Of these, only investment is readily available and its measure\nrelatively uncontroversial. We choose to investigate the dynamic response of two policy\nrelated variables: inflation and government expenditures. If the conduct of economic policy\ndoes indeed deteriorate at high levels of debt, we conjecture that at least one of these\nmeasures will show a systematic response. We also include an index computed by the World\nBank, the Country Performance International Assessment, or CPIA, meant to summarize in\none number an assessment of the overall quality of policy stance. <sup>11</sup> [^11: These data are confidential.] Because we are interested\nin the terms under which debt is contracted, we finally bring in additional information from\nGlobal Development Finance publications, with a view to isolating changes in the rate of\ninterest\n\nMention: \"International Financial Statistics\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The third standard source behind the macro variables.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002887:7:1", "occurrence_id": "972999b4cb6fd0b40614832e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  growth regressions, based on three- or\nfive-year averages; second in an event study where we use all the available time variation.\nOur control variables in the growth analysis are classic and inspired from the robust sets\nproposed in Levine and Renelt (1992). <sup>10</sup> [^10: Doppelhofer, Miller and Sala-i-Martin (2004) recently used Bayesian techniques to isolate an alternative,\nupdated set of control variables. The approach is however purely cross-sectional, and its results cannot be\nused in the present context where within-country growth determinants are of the essence. For instance, the\nmost robust correlates of economic growth include the extents of confucianistic or protestant religions, or\ngeographic binary variables, none of which lend themselves to a panel estimation.] They include initial income, openness to trade,\npopulation growth, secondary schooling and the growth rate of the terms of trade. We also\nexperimented with the fiscal balance, with no changes in conclusions. Construction of these\nvariables is standard; the sources used are the World Development Indicators, the World\nEconomic Outlook and the International Financial Statistics.\n\n\nIn the event study, we track the response over time of investment, macroeconomic policy,\nand the terms of borrowing. Of these, only investment is readily available and its measure\nrelatively uncontroversial. We choose to investigate the dynamic response of two policy\nrelated variables: inflation and government expenditures. If the conduct of economic policy\ndoes indeed deteriorate at high levels of debt, we conjecture that at least one of these\nmeasures will show a systematic response. We also include an index computed by the World\nBank, the Country Performance International Assessment, or CPIA, meant to summarize in\none number an assessment of the overall quality of policy stance. <sup>11</sup> [^11: These data are confidential.] Because we are interested\nin the terms under which debt is contracted, we finally bring in additional information from\nGlobal Development Finance publications, with a view to isolating changes in the rate of\ninterest\n\nMention: \"Country Performance International Assessment\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Its single index summarises the overall quality of the policy stance.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002892:41:1", "occurrence_id": "38cc49325285f91322f9fc72", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  d t + εt, where PETROLt = total petrol consumed in road vehicles\nin year t, DIESELt = total diesel fuel consumed in road vehicles in year t, and t is a linear time trend.\nPETROL and DIESEL (measured in thousands of metric tons) come primarily from the IRF World Road\nStatistics yearbooks. These series were cross-checked and supplemented by data from OECD Energy\n\n\n41\n\nMention: \"IRF World Road\nStatistics yearbooks\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Petrol and diesel consumption series come primarily from these yearbooks.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002892:41:1", "occurrence_id": "23f7b73e46e0671b6294a254", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  d t + εt, where PETROLt = total petrol consumed in road vehicles\nin year t, DIESELt = total diesel fuel consumed in road vehicles in year t, and t is a linear time trend.\nPETROL and DIESEL (measured in thousands of metric tons) come primarily from the IRF World Road\nStatistics yearbooks. These series were cross-checked and supplemented by data from OECD Energy\n\n\n41\n\nMention: \"data from OECD Energy\n\n\n41\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"It cross-checks and supplements the fuel series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002950:26:2", "occurrence_id": "cd689c7a048d568e8dd56ac2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **\nBangladesh 12.7 7.4 21.2 29.9 16.2\nIndia 49.9 25.7 75.6 26.5 24.2\nIndonesia 5.0 4.3 6.2 8.0 4.3\nVietnam 37.1 13.1 44.8 29.1 12.3\nRest of South Asia 21.1 14.2 32.0 6.6 14.3\nSelected SSAfrica <sup>a</sup> 11.8 10.2 13.0 12.5 7.5\nRest of SSAfrica 21.2 18.0 23.6 26.2 14.0\nRest of the World 11.8 1.9 18.7 5.6 8.9\n\n_a_ The Selected Sub-Saharan African countries (for which national modules are available\nin the LINKAGE Model) include Botswana, Madagascar, Malawi, Mozambique, Tanzania,\nUganda, Zambia, Zimbabwe.\n\n\n_b_ Numbers in parentheses are the averages at the start of 2005 following WTO accession\nby China and end of MFA.\n\n_Source:_ Authors’ projections from the GTAP database Version 6.05 using the World\nBank’s LINKAGE model\n\n\n25\n\nMention: \"GTAP database Version 6.05\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Projections of trade shifts after China's accession and the MFA end are built on this version.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002955:4:1", "occurrence_id": "d0c6e2aeefc2ef738b2bae30", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  fell, as did GIS coverage. By 1998 nearly half the urban\n\n\npopulation lacked insurance coverage. A variety of reforms have been introduced, including the\n\n\n4 Figures on coverage are from China’s National Health Survey (NHS), undertaken by the Ministry of Health (MOH) in 1993,\n1998 and 2003. The figures for 1993 and 1998 are published in Gao et al. (2002). The urban insurance figures for 2003 are taken\nfrom “Health Services Utilization and Urban Health Insurance Reform in China”, a presentation by Ling Xu of China’s MOH in\nDecember 2004 at an MOH seminar on the 2003 NHS. The rural figures for 2003 are from “Main Findings from the 3rd NHS\nSurvey”, available online at www.moh.gov.cn, accessed on April 21st, 2005.\n\nMention: \"the 3rd NHS\nSurvey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its main findings give the rural health figures for that year.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002960:33:7", "occurrence_id": "72bd6b7e3a36f02747508887", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <br> <br> <br>**100.0**<br>**12.0**<br> <br>|\n\n\n_<u>Source: Authors’ construction:</u>_\n\n- Based on the 2001 SAM; VA = Value added; X = Production.\n** Capital/Labor ratio.\n*** Export intensity = exports as a share of output.\n**** Import intensity = import as a share of domestic demand\n***** Armington elasticities calculated from GTAP for our commodity aggregations. We symmetrically use these values for CET elasticities.\n\n\n\n34\n\nMention: \"2001 SAM\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Value added, production and factor ratios are constructed from this matrix.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002960:33:7", "occurrence_id": "6b150cd7c1871282b1e51210", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <br> <br> <br>**100.0**<br>**12.0**<br> <br>|\n\n\n_<u>Source: Authors’ construction:</u>_\n\n- Based on the 2001 SAM; VA = Value added; X = Production.\n** Capital/Labor ratio.\n*** Export intensity = exports as a share of output.\n**** Import intensity = import as a share of domestic demand\n***** Armington elasticities calculated from GTAP for our commodity aggregations. We symmetrically use these values for CET elasticities.\n\n\n\n34\n\nMention: \"GTAP\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Armington elasticities for the commodity aggregations are calculated from it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002968:19:0", "occurrence_id": "8092590414fc99a1667a113c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: level surveys conducted in 1999 and 2000 for over 10,000 firms in 81 countries. <sup>20</sup> [^20: For a detailed discussion of the survey see Batra, Kaufmann, and Stone (2002).] This database\n\n\nhas several advantages over other firm-level databases. First, the survey includes a broad variety\n\n\nof firms of different ownership structures, sectors, legal forms, and – most importantly –\n\n\ndifferent sizes; 80% of the surveyed firms are small or medium-sized, with fewer than 500\n\n\nemployees. Second, firm managers were asked about the obstacles they face in their operation\n\n\nand growth, including several questions related to the financial system.\n\n\nManagers of the surveyed firms were asked to rate how problematic general financing\n\n\nobstacles are for the operation and growth of their firm. Responses varied between a rating of\n\n\none (no obstacle), two (minor obstacle), three (moderate obstacle) and four (major obstacle).\n\n\n36% of all firms rate financing as a major obstacle, 27% as moderate, 18% as minor and 19% as\n\n\nno obstacle. In addition to growth obstacles and firm size, the survey also provides general\n\n\ninformation on firms such as size, sector and ownership.\n\n\nSelf-reported financing obstacles might be subject to biases if slow-growing firms or\n\n\nfirms with low efficiency and productivity report higher obstacles. Using the WBES database,\n\n\nBeck, Demirguc-Kunt and Maksimovic (2005) show that firms reporting higher financing\n\n\nobstacles indeed grow more slowly, but that this relationship is not due to reverse causation.\n\n\nFurther, as reported in Beck, Demirguc-Kunt and Levine (2004), firm-reported financing\n\n\nobstacles are negatively and significantly correlated with the efficiency of investment, as\n\n\nmeasured by Wurgler (2000). <sup>21</sup> [^21: This is an investment elasticity that gauges the extent to which a country increases investment in growing\nindustries and decreases investment in declining ones.]\n\n\nWhile our outreach indicators are available for up to 99 countries and the WBES dataset\n\n\ncovers\n\nMention: \"WBES database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"A cited study uses it to link reported financing obstacles with growth.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002969:16:1", "occurrence_id": "d39f41384067808879353941", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  if the country has controls\n\n\non cross-border financial credits. Second, we control for restrictions on on-shore foreign\n\n\ncurrency lending by including a dummy for the countries where such restrictions are in place.\n\n\nData for both variables come from the IMF _Annual Report on Exchange Arrangements and_\n\n\n12 Ideally, we would have liked to control as well for the share of non-financial FDI from the home to the host\ncountry, however, this information was missing for many of the countries in our sample.\n\n\n16\n\nMention: \"IMF _Annual Report on Exchange Arrangements\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"It supplies both the restriction dummies and the arrangement variables.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002970:0:1", "occurrence_id": "e9c8960d341fc4b634727070", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Working Papers are available online at_\n_http://econ.worldbank.org._\n\n\n∗ Corresponding author: Rita Almeida (ralmeida@worldbank.org). We thank the Brazilian Ministry of\nLabor for providing us the data on enforcement of labor regulation that we use in our paper and important\ninformation about the process of enforcement, especially Edgar Brandao, Sandra Brandao and Marcelo\nCampos. We thank George Clarke, Saul Estrin, Ana Fernandes, Mary Hallward-Driemeier, John\nHaltiwanger, Jennifer Hunt, Sara Lemos, Adriana Kugler, Stefano Scarpetta, Luis Serven, Rodrigo Soares\nand participants in several seminars for valuable suggestions. Carneiro thanks the hospitality of the Poverty\nUnit of the World Bank Research Group.\n\nMention: \"the Brazilian Ministry of\nLabor\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It provided the enforcement records of labour regulation used in the paper.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002979:6:0", "occurrence_id": "bdcda8280deda7c691346b96", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 7\n\n\n**Figure 3: Top ten revenue generating countries**\n\n\n\n\n\n\n\n\n\n\n\n_Source_ : World Bank Privatization Database\n\n\n12. Data on foreign vs. domestic privatization revenues is available only for the earlier\nperiod, 1990-1999. Foreign revenues comprised 50 percent of total privatization proceeds in\ndeveloping countries; three-quarters of foreign revenues were raised through direct investment\nand the balance through portfolio flows (World Bank 2001). The Latin American region\nrepresented 56 percent of all foreign investment as a result of sales in telecommunications,\nbanking, and oil and gas, followed by Europe and Central Asia (23 percent) and East Asia (15\npercent). China was the main recipient of foreign investment in East Asia and Pacific followed\nby Thailand, Indonesia, and Malaysia. Despite the decline in total privatization proceeds after\ntheir peak 1997 level, foreign investment in privatization remained steady and increased in\n1999 due to large transactions in East Asia and Latin America. Data on the FDI share in\nprivatization is not available for the 2000 to 2003 time period. A comparison of general FDI\ntrends and privatization trends (Figure 4) shows, however, that while total FDI declined from\n2001 onwards, privatization proceeds began to pick up, albeit modestly--largely as a result of\nshare sales of a few large companies via domestic stock markets.\n\nMention: \"World Bank Privatization Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Privatization revenues by country are taken from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002979:6:0", "occurrence_id": "2758b7c6d8d85182cf157939", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 7\n\n\n**Figure 3: Top ten revenue generating countries**\n\n\n\n\n\n\n\n\n\n\n\n_Source_ : World Bank Privatization Database\n\n\n12. Data on foreign vs. domestic privatization revenues is available only for the earlier\nperiod, 1990-1999. Foreign revenues comprised 50 percent of total privatization proceeds in\ndeveloping countries; three-quarters of foreign revenues were raised through direct investment\nand the balance through portfolio flows (World Bank 2001). The Latin American region\nrepresented 56 percent of all foreign investment as a result of sales in telecommunications,\nbanking, and oil and gas, followed by Europe and Central Asia (23 percent) and East Asia (15\npercent). China was the main recipient of foreign investment in East Asia and Pacific followed\nby Thailand, Indonesia, and Malaysia. Despite the decline in total privatization proceeds after\ntheir peak 1997 level, foreign investment in privatization remained steady and increased in\n1999 due to large transactions in East Asia and Latin America. Data on the FDI share in\nprivatization is not available for the 2000 to 2003 time period. A comparison of general FDI\ntrends and privatization trends (Figure 4) shows, however, that while total FDI declined from\n2001 onwards, privatization proceeds began to pick up, albeit modestly--largely as a result of\nshare sales of a few large companies via domestic stock markets.\n\nMention: \"Data on foreign vs. domestic privatization revenues\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Coverage only for the earlier period limits the comparison.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002979:6:0", "occurrence_id": "a65ffa815e6219104095cfc0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 7\n\n\n**Figure 3: Top ten revenue generating countries**\n\n\n\n\n\n\n\n\n\n\n\n_Source_ : World Bank Privatization Database\n\n\n12. Data on foreign vs. domestic privatization revenues is available only for the earlier\nperiod, 1990-1999. Foreign revenues comprised 50 percent of total privatization proceeds in\ndeveloping countries; three-quarters of foreign revenues were raised through direct investment\nand the balance through portfolio flows (World Bank 2001). The Latin American region\nrepresented 56 percent of all foreign investment as a result of sales in telecommunications,\nbanking, and oil and gas, followed by Europe and Central Asia (23 percent) and East Asia (15\npercent). China was the main recipient of foreign investment in East Asia and Pacific followed\nby Thailand, Indonesia, and Malaysia. Despite the decline in total privatization proceeds after\ntheir peak 1997 level, foreign investment in privatization remained steady and increased in\n1999 due to large transactions in East Asia and Latin America. Data on the FDI share in\nprivatization is not available for the 2000 to 2003 time period. A comparison of general FDI\ntrends and privatization trends (Figure 4) shows, however, that while total FDI declined from\n2001 onwards, privatization proceeds began to pick up, albeit modestly--largely as a result of\nshare sales of a few large companies via domestic stock markets.\n\nMention: \"Data on the FDI share in\nprivatization\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its unavailability for the later period prevents a comparable series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002988:29:1", "occurrence_id": "7593ac3e81d410889a591c43", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |Unit<br>|U.S. Patent & Trademark Office<br>|\n|Patepo<br>|Total patents<br>granted by the<br>EPO by year<br>for each<br>country<br>|Unit<br>|European Patent Office<br>|\n|Royal<br>|Royalties and<br>license fees,<br>payments<br>|Current U.S.  dollars<br>|-World Development Indicators through<br>World Bank electronic databases<br>|\n|Rdexp|R&D<br>expenditure|Constant U.S. dollars|-UNESCO:printed and electronic<br>versions of Statistical Yearbook.|\n\n\n\n7 Red Iberoamericana de Indicadores de Ciencia y Tecnología\n\n\n\n29\n\nMention: \"World Bank electronic databases\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The indicators are accessed through these portals.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:002989:61:1", "occurrence_id": "21adb675045ed793b48f03db", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 0082 (0.0044, 0.0120) 0.0172 (0.0146, 0.0198)\n(secondary enrollment, in logs) 2.14 6.7\nFinancial Depth 0.0045 (0.0016, 0.0074) 0.0066 (0.0051, 0.0081)\n(private domestic credit/GDP, in logs) 1.56 4.28\nTrade Openness 0.0048 (0.0009, 0.0088) 0.0096 (0.0065, 0.0127)\n(structure-adjusted trade volume/GDP, in logs) 1.24 3.14\nGovernment Burden -0.0145 (-0.0196, -0.0094) -0.0154 (-0.0202, -0.0106)\n(government consumption/GDP, in logs) -2.82 -3.18\nPublic Infrastructure 0.0116 (0.0090, 0.0142) 0.0071 (0.0045, 0.0097)\n(Main telephone lines per capita, in logs) 4.46 2.71\nGovernance 0.0018 (0.0002, 0.0034) -0.0012 (-0.0030, 0.0006)\n(1st principal component of ICRG indicators) 1.15 -0.68\n\n\nStabilization Policies:\n\n\nLack of Price Stability -0.006 (-0.0088, -0.0032) -0.0048 (-0.0073, -0.0023)\n(inflation rate, in log [100+inf rate]) -2.11 -1.89\nCyclical Volatility: -0.2771 (-0.3508, -0.2034)\n(Std. Dev.\n\nMention: \"ICRG indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Their first principal component measures governance.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003004:3:1", "occurrence_id": "c812f88ee2b47641e3b6d352", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ||||\n|||||||\n|||||||\n|||~~Taiwan, Ch~~|<br>~~ ina~~|~~Japan~~<br>||\n|||<br>Korea|Hong Ko<br><br>|ng, China||\n||||~~Singapore~~|||\n|||||||\n\n\n0 10,000 20,000 30,000 40,000 50,000\n\nGDP per capita ($US)\n\n\n\n\n\n\n\n\n\n\n- Data refer to 2001 the latest year for which complete data are available. The predicted figures are from a regression of the\nnatural logarithm of total health expenditure (expressed as a share of GDP) on the natural logarithm of GDP per capita. Data are\nfrom the 2005 World Health Report annexes in the cases of Japan, Korea and Singapore. The per capita income data for Hong\nKong (China) and Taiwan (China) are from the World Bank’s World Development Indicators. Data on health spending are from\n<u>http://www.hwfb.gov.hk/statistics/download/dha/en/table1.pdf in the case of Hong Kong (China) and from</u>\n<u>http://www.doh.gov.tw/ufile/doc/200411_Statistic%20of%20Expenditure%20for%20Health,%201991-2003.pdf</u> in the case of\nTaiwan (China).\n\nMention: \"2005 World Health Report annexes\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Health expenditure as a share of output is taken from its annexes for three economies.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003004:3:1", "occurrence_id": "e45d982b98b0a7db080fe3b9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ||||\n|||||||\n|||||||\n|||~~Taiwan, Ch~~|<br>~~ ina~~|~~Japan~~<br>||\n|||<br>Korea|Hong Ko<br><br>|ng, China||\n||||~~Singapore~~|||\n|||||||\n\n\n0 10,000 20,000 30,000 40,000 50,000\n\nGDP per capita ($US)\n\n\n\n\n\n\n\n\n\n\n- Data refer to 2001 the latest year for which complete data are available. The predicted figures are from a regression of the\nnatural logarithm of total health expenditure (expressed as a share of GDP) on the natural logarithm of GDP per capita. Data are\nfrom the 2005 World Health Report annexes in the cases of Japan, Korea and Singapore. The per capita income data for Hong\nKong (China) and Taiwan (China) are from the World Bank’s World Development Indicators. Data on health spending are from\n<u>http://www.hwfb.gov.hk/statistics/download/dha/en/table1.pdf in the case of Hong Kong (China) and from</u>\n<u>http://www.doh.gov.tw/ufile/doc/200411_Statistic%20of%20Expenditure%20for%20Health,%201991-2003.pdf</u> in the case of\nTaiwan (China).\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It supplies per-capita income for the remaining two economies.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003007:2:0", "occurrence_id": "6c9449553cf35eb63f89bb40", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **CREATING A POVERTY MAP FOR AZERBAIJAN**\n\n## **Abstract**\n\n\n‘Poverty maps’, that is graphic representations of spatially disaggregated estimates of\n\n\nwelfare, are being increasingly used to geographically target scare resources. The development\n\n\nof detailed poverty maps in many low resource settings is, however, hampered due to data\n\n\nconstraints. Data on income or consumption are often unavailable and, where they are, direct\n\n\nsurvey estimates for small areas are likely to yield unacceptably large standard errors due to\n\n\nlimited sample sizes. Census data offer the required level of coverage but do not generally\n\n\ncontain the appropriate information. This has led to the development of a range of alternative\n\n\nmethods aimed either at combining survey data with unit record data from the Census to produce\n\n\nestimates of income or expenditure for small areas or at developing alternative welfare rankings,\n\n\nsuch as asset indices, using existing Census data. This report develops a set of poverty maps for\n\n\nAzerbaijan which can be used by different users. Two alternative approaches to the measurement\n\n\nand mapping of welfare are adopted. First a map is derived using imputed household\n\n\nconsumption. This involves combining information from the 2002 Household Budget Survey\n\n\n(HBS) with 1999 Census data. Secondly an alternative map is constructed using an asset index\n\n\nbased on data from the 1999 Census to produce estimates of welfare at the rayon level. This\n\n\nprovides a unique opportunity to compare the welfare rankings obtained at the regional level\n\n\nunder the two alternative approaches. In order to visually present the spatially disgaggregated\n\n\nestimates of welfare in Azerbaijan, this project has also involved the production of a digital\n\n\ncensus map of Azerbaijan. This involved matching the Census enumeration areas to a digital\n\n\nsettlement map of Azerbaijan. An important externality of the research is that is now possible for\n\n\nthe State Statistical Committee of Azerbaijan to display graphically the results of the 1999\n\n\nCensus of Azerbaijan along with other data.\n\n\n1\n\nMention: \"2002 Household Budget Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Household consumption is imputed from it to produce the welfare map.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003007:2:0", "occurrence_id": "d0a95e2d4bb4fcba6010f4cc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **CREATING A POVERTY MAP FOR AZERBAIJAN**\n\n## **Abstract**\n\n\n‘Poverty maps’, that is graphic representations of spatially disaggregated estimates of\n\n\nwelfare, are being increasingly used to geographically target scare resources. The development\n\n\nof detailed poverty maps in many low resource settings is, however, hampered due to data\n\n\nconstraints. Data on income or consumption are often unavailable and, where they are, direct\n\n\nsurvey estimates for small areas are likely to yield unacceptably large standard errors due to\n\n\nlimited sample sizes. Census data offer the required level of coverage but do not generally\n\n\ncontain the appropriate information. This has led to the development of a range of alternative\n\n\nmethods aimed either at combining survey data with unit record data from the Census to produce\n\n\nestimates of income or expenditure for small areas or at developing alternative welfare rankings,\n\n\nsuch as asset indices, using existing Census data. This report develops a set of poverty maps for\n\n\nAzerbaijan which can be used by different users. Two alternative approaches to the measurement\n\n\nand mapping of welfare are adopted. First a map is derived using imputed household\n\n\nconsumption. This involves combining information from the 2002 Household Budget Survey\n\n\n(HBS) with 1999 Census data. Secondly an alternative map is constructed using an asset index\n\n\nbased on data from the 1999 Census to produce estimates of welfare at the rayon level. This\n\n\nprovides a unique opportunity to compare the welfare rankings obtained at the regional level\n\n\nunder the two alternative approaches. In order to visually present the spatially disgaggregated\n\n\nestimates of welfare in Azerbaijan, this project has also involved the production of a digital\n\n\ncensus map of Azerbaijan. This involved matching the Census enumeration areas to a digital\n\n\nsettlement map of Azerbaijan. An important externality of the research is that is now possible for\n\n\nthe State Statistical Committee of Azerbaijan to display graphically the results of the 1999\n\n\nCensus of Azerbaijan along with other data.\n\n\n1\n\nMention: \"1999 Census data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Its data are combined with the survey to build the first map.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003007:2:0", "occurrence_id": "0a691a51a3bc231b95de3221", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **CREATING A POVERTY MAP FOR AZERBAIJAN**\n\n## **Abstract**\n\n\n‘Poverty maps’, that is graphic representations of spatially disaggregated estimates of\n\n\nwelfare, are being increasingly used to geographically target scare resources. The development\n\n\nof detailed poverty maps in many low resource settings is, however, hampered due to data\n\n\nconstraints. Data on income or consumption are often unavailable and, where they are, direct\n\n\nsurvey estimates for small areas are likely to yield unacceptably large standard errors due to\n\n\nlimited sample sizes. Census data offer the required level of coverage but do not generally\n\n\ncontain the appropriate information. This has led to the development of a range of alternative\n\n\nmethods aimed either at combining survey data with unit record data from the Census to produce\n\n\nestimates of income or expenditure for small areas or at developing alternative welfare rankings,\n\n\nsuch as asset indices, using existing Census data. This report develops a set of poverty maps for\n\n\nAzerbaijan which can be used by different users. Two alternative approaches to the measurement\n\n\nand mapping of welfare are adopted. First a map is derived using imputed household\n\n\nconsumption. This involves combining information from the 2002 Household Budget Survey\n\n\n(HBS) with 1999 Census data. Secondly an alternative map is constructed using an asset index\n\n\nbased on data from the 1999 Census to produce estimates of welfare at the rayon level. This\n\n\nprovides a unique opportunity to compare the welfare rankings obtained at the regional level\n\n\nunder the two alternative approaches. In order to visually present the spatially disgaggregated\n\n\nestimates of welfare in Azerbaijan, this project has also involved the production of a digital\n\n\ncensus map of Azerbaijan. This involved matching the Census enumeration areas to a digital\n\n\nsettlement map of Azerbaijan. An important externality of the research is that is now possible for\n\n\nthe State Statistical Committee of Azerbaijan to display graphically the results of the 1999\n\n\nCensus of Azerbaijan along with other data.\n\n\n1\n\nMention: \"1999 Census\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"An asset index built from it yields rayon-level welfare estimates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003007:2:0", "occurrence_id": "2240f01bfc64a28c31ea284e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **CREATING A POVERTY MAP FOR AZERBAIJAN**\n\n## **Abstract**\n\n\n‘Poverty maps’, that is graphic representations of spatially disaggregated estimates of\n\n\nwelfare, are being increasingly used to geographically target scare resources. The development\n\n\nof detailed poverty maps in many low resource settings is, however, hampered due to data\n\n\nconstraints. Data on income or consumption are often unavailable and, where they are, direct\n\n\nsurvey estimates for small areas are likely to yield unacceptably large standard errors due to\n\n\nlimited sample sizes. Census data offer the required level of coverage but do not generally\n\n\ncontain the appropriate information. This has led to the development of a range of alternative\n\n\nmethods aimed either at combining survey data with unit record data from the Census to produce\n\n\nestimates of income or expenditure for small areas or at developing alternative welfare rankings,\n\n\nsuch as asset indices, using existing Census data. This report develops a set of poverty maps for\n\n\nAzerbaijan which can be used by different users. Two alternative approaches to the measurement\n\n\nand mapping of welfare are adopted. First a map is derived using imputed household\n\n\nconsumption. This involves combining information from the 2002 Household Budget Survey\n\n\n(HBS) with 1999 Census data. Secondly an alternative map is constructed using an asset index\n\n\nbased on data from the 1999 Census to produce estimates of welfare at the rayon level. This\n\n\nprovides a unique opportunity to compare the welfare rankings obtained at the regional level\n\n\nunder the two alternative approaches. In order to visually present the spatially disgaggregated\n\n\nestimates of welfare in Azerbaijan, this project has also involved the production of a digital\n\n\ncensus map of Azerbaijan. This involved matching the Census enumeration areas to a digital\n\n\nsettlement map of Azerbaijan. An important externality of the research is that is now possible for\n\n\nthe State Statistical Committee of Azerbaijan to display graphically the results of the 1999\n\n\nCensus of Azerbaijan along with other data.\n\n\n1\n\nMention: \"digital\n\n\nsettlement map of Azerbaijan\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Enumeration areas are matched to it to produce the census map.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003007:2:0", "occurrence_id": "490b53b7041fd8940fc27594", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **CREATING A POVERTY MAP FOR AZERBAIJAN**\n\n## **Abstract**\n\n\n‘Poverty maps’, that is graphic representations of spatially disaggregated estimates of\n\n\nwelfare, are being increasingly used to geographically target scare resources. The development\n\n\nof detailed poverty maps in many low resource settings is, however, hampered due to data\n\n\nconstraints. Data on income or consumption are often unavailable and, where they are, direct\n\n\nsurvey estimates for small areas are likely to yield unacceptably large standard errors due to\n\n\nlimited sample sizes. Census data offer the required level of coverage but do not generally\n\n\ncontain the appropriate information. This has led to the development of a range of alternative\n\n\nmethods aimed either at combining survey data with unit record data from the Census to produce\n\n\nestimates of income or expenditure for small areas or at developing alternative welfare rankings,\n\n\nsuch as asset indices, using existing Census data. This report develops a set of poverty maps for\n\n\nAzerbaijan which can be used by different users. Two alternative approaches to the measurement\n\n\nand mapping of welfare are adopted. First a map is derived using imputed household\n\n\nconsumption. This involves combining information from the 2002 Household Budget Survey\n\n\n(HBS) with 1999 Census data. Secondly an alternative map is constructed using an asset index\n\n\nbased on data from the 1999 Census to produce estimates of welfare at the rayon level. This\n\n\nprovides a unique opportunity to compare the welfare rankings obtained at the regional level\n\n\nunder the two alternative approaches. In order to visually present the spatially disgaggregated\n\n\nestimates of welfare in Azerbaijan, this project has also involved the production of a digital\n\n\ncensus map of Azerbaijan. This involved matching the Census enumeration areas to a digital\n\n\nsettlement map of Azerbaijan. An important externality of the research is that is now possible for\n\n\nthe State Statistical Committee of Azerbaijan to display graphically the results of the 1999\n\n\nCensus of Azerbaijan along with other data.\n\n\n1\n\nMention: \"1999\n\n\nCensus of Azerbaijan\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Its results can now be displayed graphically by the statistical committee.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003008:2:1", "occurrence_id": "60ec217089a41d69314536f7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  on disability. There is relatively little data of this kind in\ndeveloping countries: the datasets, and therefore the countries, for this analysis were\nselected on the basis of data availability. The countries were not selected to be\nrepresentative of developing countries in general.\n\n\nThis is clearly a heterogeneous group of countries. Population living on less than a\ndollar a day ranges from 55 percent in Burundi to two percent in Jamaica and Romania;\nunder-5 mortality—an indicator of basic health status—ranges from 206 per thousand live\nbirths in Mozambique to 15 in Romania (Table 1). There are three countries from Africa,\nfour countries from Asia, one country from the Caribbean, and one country from Eastern\nEurope. While country variety is good since the results will reflect on a range of\nunderlying conditions, little draws these countries together besides having the data\navailable for this analysis.\n\n\n3 LSMS data are available online at <u>http://www.worldbank.org/lsms; national socio-economic surveys are</u>\navailable from the countries’ national statistics offices; DHS data are available online at\n<u>http://www.measuredhs.com; MICS2 data are available online at</u> <u>http://www.childinfo.org.</u>\n\n\n2\n\nMention: \"LSMS data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"A footnote records where the survey can be downloaded.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003008:2:1", "occurrence_id": "d741918ae1f5aed2b1019263", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  on disability. There is relatively little data of this kind in\ndeveloping countries: the datasets, and therefore the countries, for this analysis were\nselected on the basis of data availability. The countries were not selected to be\nrepresentative of developing countries in general.\n\n\nThis is clearly a heterogeneous group of countries. Population living on less than a\ndollar a day ranges from 55 percent in Burundi to two percent in Jamaica and Romania;\nunder-5 mortality—an indicator of basic health status—ranges from 206 per thousand live\nbirths in Mozambique to 15 in Romania (Table 1). There are three countries from Africa,\nfour countries from Asia, one country from the Caribbean, and one country from Eastern\nEurope. While country variety is good since the results will reflect on a range of\nunderlying conditions, little draws these countries together besides having the data\navailable for this analysis.\n\n\n3 LSMS data are available online at <u>http://www.worldbank.org/lsms; national socio-economic surveys are</u>\navailable from the countries’ national statistics offices; DHS data are available online at\n<u>http://www.measuredhs.com; MICS2 data are available online at</u> <u>http://www.childinfo.org.</u>\n\n\n2\n\nMention: \"DHS data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"The same footnote gives its download location.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003008:2:1", "occurrence_id": "77e6745aa460f394143bf1d2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  on disability. There is relatively little data of this kind in\ndeveloping countries: the datasets, and therefore the countries, for this analysis were\nselected on the basis of data availability. The countries were not selected to be\nrepresentative of developing countries in general.\n\n\nThis is clearly a heterogeneous group of countries. Population living on less than a\ndollar a day ranges from 55 percent in Burundi to two percent in Jamaica and Romania;\nunder-5 mortality—an indicator of basic health status—ranges from 206 per thousand live\nbirths in Mozambique to 15 in Romania (Table 1). There are three countries from Africa,\nfour countries from Asia, one country from the Caribbean, and one country from Eastern\nEurope. While country variety is good since the results will reflect on a range of\nunderlying conditions, little draws these countries together besides having the data\navailable for this analysis.\n\n\n3 LSMS data are available online at <u>http://www.worldbank.org/lsms; national socio-economic surveys are</u>\navailable from the countries’ national statistics offices; DHS data are available online at\n<u>http://www.measuredhs.com; MICS2 data are available online at</u> <u>http://www.childinfo.org.</u>\n\n\n2\n\nMention: \"MICS2 data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"The footnote lists its download location as well.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003023:1:0", "occurrence_id": "44272f4c27163a3aba03b835", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Abstract**\n\n\nStandards and technical regulations are an increasingly prominent part of\nthe international trade policy debate. In particular, there has been\nconsiderable discussion of whether standards and regulations affect trade\ncosts and export prospects for developing countries. In this paper, we\nexamine how meeting foreign standards affects firms’ export performance,\nreflected in export propensity and market diversification. The analysis\ndraws on the World Bank Technical Barriers to Trade Survey database of\n619 firms in 17 developing countries. Our results indicate that standards\nand technical regulations in developed countries do affect firms' propensity\nto export in developing countries. In particular, testing procedures and\nlengthy inspection procedures by importers reduce exports by 9% and 3%,\nrespectively. Furthermore, in our model, the difference in standards across\nforeign countries causes diseconomy of scale for firms and affects decisions\nabout whether to enter export markets. The empirical analysis presented\nhere implies that standards impede exporters' market entry, reducing the\nlikelihood of exporting to more than three markets by 7%. In addition, we\nfind that firms that outsource components are more challenged by\ncompliance with multiple standards than those that do not outsource..\n\n\n2\n\nMention: \"the World Bank Technical Barriers to Trade Survey database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Six hundred nineteen firms across seventeen developing countries are analysed for export effects.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003028:24:0", "occurrence_id": "f1b3ae4b28cf739afa1ee4d2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Table 9. Share of variance in poverty changes due to growth** <sup>**a/**</sup>\n\n**(Based on Dollar and Kraay (2002) database)**\n\n\n<u>Headcount</u>\n\n**Gini Coefficient**\n\n**Mean Income /**\n\n**Poverty Line** **0.30** **0.40** **0.50** **0.60**\n**6** 0.28 0.24 0.19 0.14\n**3** 0.48 0.41 0.32 0.23\n**2** 0.66 0.58 0.47 0.33\n**1.5** 0.82 0.73 0.60 0.44\n**1** 0.98 0.94 0.84 0.66\n\n\n<u>Poverty Gap</u>\n\n**Gini Coefficient**\n\n**Mean Income /**\n\n**Poverty Line** **0.30** **0.40** **0.50** **0.60**\n**6** 0.25 0.20 0.15 0.11\n**3** 0.41 0.32 0.24 0.16\n**2** 0.55 0.43 0.31 0.20\n**1.5** 0.67 0.53 0.38 0.24\n**1** 0.83 0.68 0.50 0.32\n\n\n<u>Squared Poverty Gap</u>\n\n**Gini Coefficient**\n\n**Mean Income /**\n\n**Poverty Line** **0.30** **0.\n\nMention: \"Dollar and Kraay (2002) database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Variance shares of poverty changes attributable to growth are computed from it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003030:5:1", "occurrence_id": "6d88d6bf305ae23b37ba5d4a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ---|---|---|---|---|\n||||||||||||||||||\n|Poorest|Poorest|Poorest|Poorest|Poorest|Poorest|Poorest|Poorest|Poorest|Poorest|Poorest|Poorest|Poorest|Poorest||||\n||||||||||||||||||\n||||||||||||||||||\n||||||||||||||||||\n\n\nWealthiest\n\n\n\n_Source:_ World Development Report 2006, from Demographic Health Survey (DHS) data.\n_Note:_ The continuous dark line represents the percentage of children without access to a basic\nimmunization package in each country, while the endpoints of the whiskers indicate the percentages for the\ntop and the bottom quintile of the asset ownership distribution.\n\n- Indicates that the poorest quintile have higher access to childhood immunization services than the\nwealthiest quintile.\n\n\nThese statistical associations do not establish causality, of course. While endogeneity\nshould not be a concern in the associations presented above – because access to\nimmunization rates or child mortality today can not cause parental education decades\nearlier – omitted variables clearly exist. The descriptions presented above are essentially\n\nMention: \"Demographic Health Survey (DHS)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Child access to immunisation by wealth group is drawn from these data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003033:15:1", "occurrence_id": "394c5e48c137a48119107144", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  the PAC spending data are put is to measure lobbying organization. <sup>35</sup> Whether a good i\n\n\nhas political representation (Ii=1) or not (Ii=0) is determined on the basis of PAC spending\n\n\nsummed over this period. Goods are ordered by PAC spending and sets of I are defined, each set\n\n\ncorresponding to a percentile cutoff value. Models (4)-(6) are estimated using all these\n\n32 Kee, Nicita and Olarreaga (2004a). See e.g. Kohli (1991) for the GDP function method. In this method\nimports are treated as inputs into domestic production, given exogenous world prices, productivity and\nendowments.\n\n33 Fuller (1986). The idea behind this correction is to limit the influence of estimates that are large and\nalso have large standard errors. Without the correction, these large estimates would grossly overstate the\ntrue elasticity. The correction mutes their effect.\n\n34 <u>www.fec.gov.</u>\n\n35 Mapping PACs into SIC-based agriculture-related sectors was done via a concordance constructed by\nBeaulieu and Magee (2002). In most cases many PACs mapped into one SIC code, so SIC level\ncontributions are simply the aggregate of the mapped PACs. For one-to-many maps, political\ncontributions from each PAC were fractionally assigned equally to each SIC code into which the PAC\nmapped. To check for consistency we compared our data with the data on the opensecrets.com website.\nFor the few sectors in which opensecrets.com reports such information our data closely matched theirs.\n\n\n15\n\nMention: \"PAC spending data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Its allocation across industry codes measures political representation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003033:15:1", "occurrence_id": "18d92f01a9bd4444e0d657a0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  the PAC spending data are put is to measure lobbying organization. <sup>35</sup> Whether a good i\n\n\nhas political representation (Ii=1) or not (Ii=0) is determined on the basis of PAC spending\n\n\nsummed over this period. Goods are ordered by PAC spending and sets of I are defined, each set\n\n\ncorresponding to a percentile cutoff value. Models (4)-(6) are estimated using all these\n\n32 Kee, Nicita and Olarreaga (2004a). See e.g. Kohli (1991) for the GDP function method. In this method\nimports are treated as inputs into domestic production, given exogenous world prices, productivity and\nendowments.\n\n33 Fuller (1986). The idea behind this correction is to limit the influence of estimates that are large and\nalso have large standard errors. Without the correction, these large estimates would grossly overstate the\ntrue elasticity. The correction mutes their effect.\n\n34 <u>www.fec.gov.</u>\n\n35 Mapping PACs into SIC-based agriculture-related sectors was done via a concordance constructed by\nBeaulieu and Magee (2002). In most cases many PACs mapped into one SIC code, so SIC level\ncontributions are simply the aggregate of the mapped PACs. For one-to-many maps, political\ncontributions from each PAC were fractionally assigned equally to each SIC code into which the PAC\nmapped. To check for consistency we compared our data with the data on the opensecrets.com website.\nFor the few sectors in which opensecrets.com reports such information our data closely matched theirs.\n\n\n15\n\nMention: \"data on the opensecrets.com website\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"A close match with this public source supports the coded data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003039:4:1", "occurrence_id": "ec3cedc21d0a1e4d3351bf71", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  risen to 53%, well above the benchmark average (World\nBank 2002).\n\n**_Access:_** Not surprisingly, Lesotho performs better than comparable groups when it comes to access to\nimproved **_water_** sources. Around 76% of the population has access compared to around 65% for low\nincome and SSA. Lesotho’s overall performance compares favorably to South Asia and East Asia &\nPacific regions. Moreover, the urban/rural divide is much narrower than in other regions, with 74% of\n**_rural_** population having access to improved water sources compared to 56% in low-income countries and\n54% in SSA. About 88% of **_urban_** population has access to improved water sources, which is higher than\naverages in low-income and Sub-Saharan African countries (83%).\n\nIn terms of access to improved **_sanitation_**, however, Lesotho has 37% access rate, as reported in the\ndatabase, which is similar to SSA but just under low-income countries; note that more recent data cited\nabove (World Bank 2002) indicate that coverage has climbed to 53%. Other regions, however, are far\nmore advanced in this area with Middle East and North Africa at 77%, East Asia Pacific 60% and Latin\nAmerica and Caribbean at 77% access rates. Both **_rural_** (32%) and **_urban_** (61%) access rates for Lesotho\nare better than in low-income countries and the Sub-Saharan Africa region as a whole (30% and 60%,\nrespectively. Most recent World Bank sector data indicate that in the capital Maseru, the coverage of 65%\n\n\n5\n\nMention: \"World Bank sector data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its latest figures give water coverage in the capital.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003040:13:1", "occurrence_id": "92eee3e988fdc9edd99e8828", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ’s house and the plot; we only know whether the plot is outside his\n\nvillage or not. This could be a concern, given that both the cost of manure application\n\non a given plot and tenancy status might vary by distance to the homestead. However,\n\nin our sample, 92% of all cultivated plots are located within the village (93% of owned\n\nplots compared to 90% of leased plots). Given the size of most villages in Pakistan, the\n\nvast majority of plots are therefore, at most, only a kilometer or two from the farmer’s\nhome. <sup>22</sup> [^22: We can be more precise by examining data from the 1991 IFPRI survey, which did collect information\non distance to plot (but not on whether the plot was outside the village). Virtually all of the households\nin this survey were included in the 2001 PRHS sample as part of a panel. Among the 914 plots recorded\nin the 1991 survey, mean distance between household and plot is only 600 meters (median 300 meters).] Within this radius, transport cost differentials are likely to be minimal. More\n\n19 We choose k so that it is at least an order of magnitude less than the minimum value of the dependent\nvariable. Maximum likelihood estimation of k would fail, since the log Jacobian of the transformation\nof y is - log(y + k). At y = 0, the optimization algorithm would always try to set k as close to zero as\npossible.\n\n20 For example, the t-value for the leasing coefficient in row (3) of Table 4 is 5.91. The corresponding\nt-value is 5.92 when k = 0.01 and 5.93 when k = 0.001, even though the associated coefficients vary widely\nin magnitude.\n\n21 Both owners and tenant cultivators are asked to report plot value in the survey. One might worry that\nthese values partly re fl ect the extent of past manuring, although this\n\nMention: \"the 1991 IFPRI survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It recorded distance to plot, allowing a more precise statement.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003044:16:1", "occurrence_id": "6ade10258a88b392f82c103b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  to be better in both rural and rural areas than in the upper middle-income countries.\nThis should not be taken automatically to imply better service, for which comparative indicators in rural\nareas are not available. In fact, as noted in the earlier sections, other aggregate indicators of telecom\nservice performance (especially in local services) suggest considerable scope for improvement.\n\n**Conclusion**\n\nIn sum, we have benchmarked South Africa’s infrastructure performance on all four major infrastructure\nsectors in terms of key indicators of access, affordability-pricing, and quality using the new World Bank\ndatabase of infrastructure sector outcome indicators from 207 countries. The broad results suggest the\nfollowing: access remains a major issue in electricity and, especially, water and sanitation (particularly\nin rural areas), and so does performance in local telecom services. Even transport performance appears\n\n\n17\n\nMention: \"World Bank\ndatabase of infrastructure sector outcome indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Access, affordability and quality indicators for two hundred seven countries come from it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003048:47:1", "occurrence_id": "d635a5950848751a6b7ebdd8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  to\nthe use of formal banking products. As educational levels increase, the\npercentage of unbanked falls and the percentage of banked persons rises.\nRegression analysis shows a high correlation between educational levels\nand likelihood of being banked or unbanked. At the same time, it is\nnoteworthy that approximately 15% and 25% of the group with less than\nelementary and elementary school only respectively do have bank\naccounts in formal sector institutions.\n\n\n**Figure 3. 5 Education levels of the banked and unbanked**\n\n\n100%\n\n\n\n90%\n\n80%\n\n\n70%\n\n60%\n\n50%\n\n40%\n\n30%\n\n\n20%\n\n10%\n\n0%\n\n\n\n85%\n\n|Col1|Col2|\n|---|---|\n|||\n\n\n\nIncomplete\n\nPrimary\nEducation and\n\nless\n\n\n\n|Col1|Col2|\n|---|---|\n|||\n\n\nComplete\nPrimary Incomplete\nSecondary\n\n\n\n79%\n75%\n\n\n\n50%\n\n\n\n50%\n\n\n|Col1|61%|\n|---|---|\n|||\n|||\n\n\n|Col1|Col2|\n|---|---|\n|||\n\n\n\nIncomplete\n\nSuperior\nEducation\n\n\n\nComplete\nSecondary\n\nEducation\n\n\n\nComplete\n\nSuperior\nEducation and\n\nMore\n\n\n\nBanked Unbanked\n\n\n_Source: DANE household study of Bogotá, 2003_\n\n\n\n48\n\nMention: \"DANE household study of Bogotá\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Banking status by education level is tabulated from this household study.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003049:45:0", "occurrence_id": "97a5a696ad2159e4b0424a3a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |Table 23. Credit status|Col2|\n|---|---|\n||**Total**|\n|**Totally paid**<br>**In payment process**<br>**Without using it**<br>**Total**|21.9%<br>65.0%<br>13.1%<br>**100%**|\n|**Source: ENDESFI**|**Source: ENDESFI**|\n\n\nD. What is the problem with being unbanked in Mexico?\n\n\nThere are no formal studies of the costs paid by Mexicans as a result of having such a\nlarge percentage of the population outside of the banking system, but there are good reasons to\nbelieve that these costs could be significant. They must, for example, pay extra fees to pay their\nutility bills through a bank. The 83% who are paid by check must pay, direct commissions or\nthrough favors and time to cash salary checks. Focus group interviews and cost surveys carried\nout as part of a World Bank survey of Mexico City’s unbanked suggest that households earning\nbelow median income (about U.S. $2,500) can spend up to $100 – or 0.5% per year on such extra\nfees, not counting the time lost – a heavy burden for persons dependent on daily wages. In addition, the lack of a deposit account can make it difficult for households to build assets. The\nhouseholds may lack a safe place to guard any financial savings that they may have. More importantly, one’s access to formal-sector credit is linked to the maintenance of a bank account.\nAnd credit can be used to acquire such non-financial assets as a home or an automobile, to\nsmooth consumer purchases over time, or to finance small business activities. The cost of credit\noutside the formal banking system is\n\nMention: \"World Bank survey of Mexico City’s unbanked\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Focus groups and cost surveys show low-income households lose up to a hundred dollars a year to cash checks.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003052:41:0", "occurrence_id": "70f50cc860a6fb40640e7e05", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 40\n\n\n<u>Notes</u>\n\n\n1 International remittances are defined here as “workers’ remittances,” as listed in the 2004 issue of the\nIMF, Balance of Payments Statistics Yearbook. International remittances in this paper do not include other\nitems listed in the IMF Yearbook – such as “compensation of employees” and “migrant transfers” -because it is not clear if these items are, in fact, remittances.\n\n2 The following studies have analyzed the impact of remittances on poverty and inequality in Latin America\nand East Asia: Adams on Guatemala (2004), Taylor, Mora and Adams on Mexico (2005) and Yang and\nMartinez on the Philippines (2005).\n\n3The two known studies on remittances and poverty in Sub-Saharan Africa are: Gustafsson and Makonnen\non Lesotho (1993), and Litchfield and Waddington on Ghana (2003). Like the present paper, Litchfield\nand Waddington (2003) use the 1998/99 Ghana Living Standards Survey (GLSS 4) to analyze the impact of\nmigration and remittances on poverty in Ghana; however, their study focuses more on migration than on\nremittances.\n\n4 According to the Ghana Statistical Service (2000: Table 2), in 1998/99 the poverty headcount index in\nGhana was 39.5 percent. This index measures the share of the population living below the poverty line.\nFor more details, see text.\n\n5 According to the 1998/99 Ghana Living Standards Survey (GLSS 4), while roughly four times as many\nhouseholds in Ghana receive internal as opposed to international remittances (2139 versus 488 households,\nrespectively), the average value received of internal remittances is about one-half that of international\nremittances: 224,248 versus 485,617 cedis per capita per year, respectively.\n\n6 For other attempts to treat remittances as a substitute for home earnings and to predict (estimate) the\nincomes of households with and without migration, see Barham and Boucher (1998) and Adams (1991\n\nMention: \"Balance of Payments Statistics Yearbook\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Workers' remittances are defined as listed in this edition.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003052:41:0", "occurrence_id": "bb427aa75d2fca1df306227f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 40\n\n\n<u>Notes</u>\n\n\n1 International remittances are defined here as “workers’ remittances,” as listed in the 2004 issue of the\nIMF, Balance of Payments Statistics Yearbook. International remittances in this paper do not include other\nitems listed in the IMF Yearbook – such as “compensation of employees” and “migrant transfers” -because it is not clear if these items are, in fact, remittances.\n\n2 The following studies have analyzed the impact of remittances on poverty and inequality in Latin America\nand East Asia: Adams on Guatemala (2004), Taylor, Mora and Adams on Mexico (2005) and Yang and\nMartinez on the Philippines (2005).\n\n3The two known studies on remittances and poverty in Sub-Saharan Africa are: Gustafsson and Makonnen\non Lesotho (1993), and Litchfield and Waddington on Ghana (2003). Like the present paper, Litchfield\nand Waddington (2003) use the 1998/99 Ghana Living Standards Survey (GLSS 4) to analyze the impact of\nmigration and remittances on poverty in Ghana; however, their study focuses more on migration than on\nremittances.\n\n4 According to the Ghana Statistical Service (2000: Table 2), in 1998/99 the poverty headcount index in\nGhana was 39.5 percent. This index measures the share of the population living below the poverty line.\nFor more details, see text.\n\n5 According to the 1998/99 Ghana Living Standards Survey (GLSS 4), while roughly four times as many\nhouseholds in Ghana receive internal as opposed to international remittances (2139 versus 488 households,\nrespectively), the average value received of internal remittances is about one-half that of international\nremittances: 224,248 versus 485,617 cedis per capita per year, respectively.\n\n6 For other attempts to treat remittances as a substitute for home earnings and to predict (estimate) the\nincomes of households with and without migration, see Barham and Boucher (1998) and Adams (1991\n\nMention: \"Ghana Living Standards Survey\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"A cited study uses this round to analyse migration and remittances.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003055:6:0", "occurrence_id": "411f624770384da9eaa3bfea", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: employment in the textile sector. <sup>9</sup> [^9: This number may be underestimated as the Industriel Census 1995 (RI95) and Annual Inquiry in Industry] Employment was reported at about 135,000 individuals\n\nin 1999, and by 2001 total employment in the textile and apparel industry had risen to\n\nalmost 200,000 individuals (or about 2 percent of the active workforce). The textile\n\nindustry was in 2001 already the fourth most important sector of employment after\n\nagriculture (74%), commerce (6%) and public administration (2.5%).\n\n\nAbout three-quarters of the labor force is women, and about half of it is skilled. The labor\n\nforce is fairly young (average age is in the early 30s) and fairly educated, as the average\n\nworker reported about 8 years of formal education. The industry is localized\n\npredominantly in the capital, with about 90% of its employees residing in the urban and\n\nrural areas in the province of Antananarivo. The industry relies on temporary workers for\n\nabout 20 percent of its labor force and pays average wages of about the equivalent of 50\n\nUSD per month. Skilled workers earn substantially more, about 76 USD per month\n\ncompared to about 33 USD per month for unskilled employees. About 42 percent of the\n\nemployees in the industry report expenditures below the poverty line.\n\n\nDuring the period from 1997 to 2001, the characteristics of textile and apparel workers\n\ndid not change substantially. The fact that the percentage of skilled workers has been\n\ndeclining (especially between 1999 and 2001) together with a reduction in the average\n\nage of employees, suggests a possible shortage in the supply of skilled labor, or an\n\nincrease of reliance upon “in firm” training which is not observed in the data. The\n\npercentage of temporary workers has been declining as well, which could suggest a more\n\nmature industry no longer subjected to a high oscillation in orders. With the increase in\n\nemployment, the average wages paid to employees also rose substantially in the late\n\n1990s\n\nMention: \"Annual Inquiry in Industry\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It reported textile employment that the census appears to understate.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003056:33:5", "occurrence_id": "e0cefe28eee3351b9c86d58b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .171|98.914<br>47.364|98.914<br>47.364|\n|Total Expenditures (pesos)|14716.180<br>18670.410<br>14640.740|14716.180<br>18670.410<br>14640.740|14716.180<br>18670.410<br>14640.740|26.870<br>-0.513|26.870<br>-0.513|\n\n\n**Source: Analysis of ENHRUM data**\n\n\n\n32\n\nMention: \"ENHRUM data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Descriptive statistics and estimates are computed from these household data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003061:19:0", "occurrence_id": "00f1fe698923651d59b758e9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Management and Natural Resource Management (in Table 4, **Honesty** also strengthens Waste\n\n\nPollution regulation). This may explain the insignificance of some of the **Honesty** results in\n\n\nTable 2. Environmental lobbying, as captured by **ENGOs**, affects environmental policies\n\n\naddressing Air Pollution, Waste Pollution, and Ecosystem Management in a positive and\n\n\nstrongly significant fashion (consistent with the Table 2 results). **%Urban** is (negative)\n\n\nsignificant only for Waste Pollution regulation in Table 3.\n\n\n**V.** **CONCLUSION**\n\n\nAs developing countries undergo or consider some form of decentralization of government\n\n\nresponsibilities, concerns arise regarding the environmental impacts of decentralized policy\n\n\nmaking. However, empirical research underpinning the associated difficult policy decisions in\n\n\ndeveloping countries is severely lacking in this policy area.\n\n\nThis paper seeks to evaluate the empirical effects of environmental federalism across\n\n\ndeveloping countries. We develop a simple theory of environmental policy formation in\n\n\ncentralized and decentralized systems, where worker-, capital- and environmental lobby groups\n\n\ncompete for the favors of a semi-benevolent government. The prediction that emerges is that a\n\n\ndecentralized government regime yields weaker environmental policy than a centralized\n\n\ninstitutional design. This is due to a greater aggregate incentive for worker- and capital-owner\n\n\nlobbying for less stringent environmental policy under a decentralized system. Using a novel and\n\n\nunique World Bank dataset, we find empirical support for this prediction. This holds especially\n\n\nfor air pollution policies. Moreover, we also find that the effect of federalism is conditional on\n\n\nthe degree of trade openness. In particular, the negative effect of environmental federalism falls\n\n\nas the degree of trade openness rises. We believe this is the first cross-country evidence of a\n\n\nnegative effect of federal institutional designs in developing countries.\n\n\n19\n\nMention: \"World Bank dataset\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"This novel compilation supports the prediction about lobbying under decentralisation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003064:2:0", "occurrence_id": "c94e22f9b5b1ed19272f4b03", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: consistency between the micro data and the CGE results; however, by combining the two,\n\n\nit maps aggregate results from the CGE to the detailed information available in the\n\n\nhousehold survey and provides a much more nuanced and useful analysis of poverty\n\n\nimpact. This approach also allows decomposing the total effect on poverty into an\n\n\naggregate income growth component and a redistribution component.\n\n\nA useful insight from this analysis is that even if the total impact on poverty is not\n\n\ntoo large, its dispersion across households – due to their heterogeneity of factor\n\n\nendowments, inputs use, commodity production, and consumption preferences – is\n\n\nsignificant and this should help in designing compensatory policies. Additionally, our\n\n\ngrowth and redistribution decomposition shows that, at least in the short to medium run,\n\n\nredistribution can be as important as growth.\n\n\nThe paper is structured as follows: section 2 presents the CGE model, the micro\n\n\nmodule and the relevant data. The first part of section 3 describes the general equilibrium\n\n\nresults of the trade policy shocks, and the second part the poverty implications. The final\n\n\nsection presents some conclusions.\n\n###### **2 Measuring the effects of trade reforms on poverty: linking a** **CGE model to household surveys**\n\n\nThis section describes the main features of the CGE model and household survey\n\n\nmicro-simulation module.\n\n\n**_<u>The Nicaragua general equilibrium model and its data</u>_**\n\n\nA 2000 Social Accounting Matrix (SAM) represents the initial benchmark\n\n\nequilibrium for the CGE model. This SAM, which includes 39 sectors, 39 commodities, 3\n\n\nfactors (skilled and unskilled labor and one composite capital), an aggregate household\n\n\naccount, and other accounts (government, savings and investment, and the Rest of the\n\n\nWorld), has been assembled from various sources incorporating data from the 2000\n\n\nInput-Output Table and the 2001 Living Standards Measurement Survey (LSMS). As\n\n\n3\n\nMention: \"2001 Living Standards Measurement Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Household welfare data are combined with the input-output table.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003069:14:0", "occurrence_id": "d31e410b3dff71345243a1ea", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **_Networks and SME Entry_**\n\n\nTable 5 presents the results of a regression analysis that examines the power of\n\n\nthe ethnic network in determining firm size at start up. Size at entry is important because\n\n\nit is a predictor of future prospects, such as survival and ability to grow in the SSA\n\n\nenvironment (Biggs and Shah 2002). The hypothesis is that members of a network are\n\n\nable to start larger firms, because the network provides connections, information, and\n\n\ngovernance capabilities, which facilitate access to credit and other inputs, access to\n\n\ntechnologies and markets, or access to equity investments.\n\n\nWe examine first the effect on size at start-up of belonging to a minority ethnic\n\n\nnetwork without controlling for other possible determinants of size at start. Model I\n\n\nlooks at whether networked entrepreneurs start firms in a different size class compared\n\n\nwith indigenous-African firms. We know from the RPED surveys that indigenous\n\nAfrican SMEs in manufacturing in these countries lack the strong business networks\n\n\nenjoyed by ethnic minority entrepreneurs (Biggs and Srivastava 1996; Fafchamps 2004).\n\n\nIn particular, they lack effective multilateral reputation mechanisms able to share\n\n\ninformation on payment histories and enforce contracts through group sanctions. The\n\n\nresults of model I confirm that the coefficients for Asian and European networked firms\n\n\nare highly significant and positive, indicating that firms belonging to these stable and\n\n\ncohesive business groups start at twice the size of indigenous-African firms.\n\n\nBut this result could be influenced by other factors, such as education of the\n\n\nentrepreneurs, their initial assets or their access to start-up finance. Model II controls for\n\n\nhuman capital. The results demonstrate that educational attainment matters.\n\n\nEntrepreneurs with university or technical degrees start bigger firms. The coefficient for\n\n\nuniversity education, for example, suggests that entrepreneurs with a university degree\n\n\nstart firms approximately 50 percent larger than entrepreneurs with only primary\n\n\neducation. Nevertheless the network coefficient remains positive and highly significant\n\n\n15\n\nMention: \"the RPED surveys\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"They show local firms lack the business networks of foreign-owned ones.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003078:9:1", "occurrence_id": "ddddf07961123d9085cd43e1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  definitions of survey income are very\n\n\ndifferent, and the more countries we include, the more different they become. A huge\n\n\neffort, conducted by Luxembourg Income Study, has gone into standardization of\n\n\nnational definitions. A similar project is underway at the World Bank using Living\n\n\nStandards Measurement Surveys (LSMS). Yet the standardization covers only a small\n\n\nportion of all countries and surveys. The main differences arise in the treatment of self\n\nemployed income (what are business expenditures for the self-employed?), valuation of\n\n16 Disposable household income retrieved from surveys in West European nations amounts to about 60\npercent of GDI. A bulk of that difference is explained by health and education consumption. Publiclyfinanced health and education as share of GDI is much less in poor countries.\n\n17 On the additional pitfalls caused by the use of averages from national accounts and distributions from\nhousehold surveys, see Ravallion (2000), Deaton and Dreze (2002), and Deaton (2003).\n\n\n10\n\nMention: \"Living\n\n\nStandards Measurement Surveys\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"A standardisation project uses them to harmonise national definitions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003080:8:1", "occurrence_id": "fb43296a78ebe90eeac88689", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .445 0.196 0.115 0.328 0.147 0.088\n1992 0.456 0.209 0.128 0.325 0.150 0.093\n1993 0.466 0.215 0.130 0.326 0.151 0.093\n1995 0.380 0.164 0.095 0.277 0.117 0.070\n1996 0.378 0.167 0.101 0.273 0.122 0.075\n1997 0.370 0.162 0.096 0.273 0.116 0.071\n1998 0.363 0.156 0.091 0.251 0.110 0.066\n1999 0.375 0.161 0.094 0.256 0.112 0.067\n2001 0.375 0.166 0.100 0.258 0.113 0.069\n2002 0.365 0.156 0.090 0.245 0.102 0.060\n2003 0.373 0.163 0.097 0.249 0.106 0.064\n2004 0.345 0.145 0.083 0.222 0.093 0.054\n\n\nNotes: 1 - Rocha (1993) regional poverty lines. See Appendix 2 for details.  2 - The \"administrative poverty\nline\" is set as R$100 per person per month, in September 2004 values.\n\nSource: Authors' calculations from the PNADs.\n\n\nBehind the overall decline, poverty dynamics in Brazil over the last two decades have been\n\nmarked by considerable volatility, which largely reflected macroeconomic instability.\n\nUnsurprisingly, therefore, the volatility was more pronounced in the unstable decade of the\n\n1980s, with a sharp increase during the 1981-83 recession, and a substantial decline during the\n\nrecovery that took place between 1984 and 1986. All three measures are at their minimum in\n\n\n16 See Foster, Greer and Thorbecke (1984).\n17 The Thai (Indonesian) poverty incidence is calculated with respect\n\nMention: \"the PNADs\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Poverty line dynamics over two decades are calculated from them.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003081:22:3", "occurrence_id": "c2da8b28ebae29120cf06bc3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .2 16.1 19.2 21.1 ***\nBanana on parcel (%) 32.3 64.6 *** 23.9 49.2 ***\nRoot crops on parcel (%) 49.5 75.6 *** 38.5 77.3 ***\nFruit crops on parcel (%) 9.2 32.3 *** 4.0 18.9 ***\n<u>Vegetables on parcel (%)</u> <u>2.4</u> <u>3.1</u> <u>2.2</u> <u>2.8</u>\n<u>Number of observations</u> <u>2185</u> <u>148</u> <u>1546</u> <u>489</u>\n<u>1 Land sales values are only for owners but not for renters.</u>\n\n- significant at 10%; ** significant at 5%; *** significant at 1% for the t-statistic to test the equality of group means using customary tenure as a\nreference group. Note that 26.4 % and 93.5 % of the parcels under freehold and _mailo_ tenure systems are respectively occupied by tenants.\n_Source:_ Own computation from 2004 WB/Makerere/FASID land tenure survey.\n\n22\n\nMention: \"2004 WB/Makerere/FASID land tenure survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Tenancy shares under freehold and mailo tenure are computed from it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003088:11:0", "occurrence_id": "279b515bbd99b3798d9c3ef2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ultimately translate into higher household expenditure (both food and non-food), as well as human capital\n\n\ninvestment (in children, for example). The results show that the returns to road investment for household\n\n\nper capita expenditure are about 11 percent in the RDP and RRMIMP villages, a substantial gain in terms\n\n\nof higher consumption and income for rural households. This means that rural households in villages\n\n\ntargeted by the road development project have on average an 11 percent higher consumption per capita\n\nper year. <sup>20</sup> [^20: The returns to consumption seem to be higher for rural electrification than rural road investment (as high as 23\npercent for electrification as compared to 11 percent for rural road development).]\n\n\nIn addition to household consumption, average school participation among boys is about 20\n\n\npercent higher in RDP villages, not controlling for proximity to the project road; road development also\n\n\nhas a positive but barely significant impact on girls’ schooling in the same sample. In the RRMIMP data,\n\n\nthe project effect is fairly similar for boys and girls, leading to a 14 percent increase in schooling. Indeed,\n\n\nin our results only the project indicator seems to have had any significant impact on schooling in the\n\n\nRRMIMP sample. Higher school attendance of boys and girls are therefore some of social returns due to\n\n\nroad investment; additional social returns to be assessed from road investment include returns to health\n\n\nand nutrition, which are not pursued in this paper.\n\n\nFinally, as a robustness check, we estimate the non-price outcomes (HH per capita expenditure,\n\n\nLaspeyres quantity index, boys’ and girls’ schooling, as well as male and female labor supply) on the\n\n\nsame set of explanatory variables as well as the price variables discussed above (men’s agricultural wage,\n\n\nfertilizer price, and the agricultural price index facing the household). The net returns are presented in\n\n\ntable 3, and reflect the returns if we held the structural model (taking prices as exogenous) to be correct.\n\n\nOverall, the estimated effects\n\nMention: \"the RRMIMP data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Project effects on schooling for boys and girls are estimated here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003089:7:1", "occurrence_id": "a8b6b229d3f10436d9c166d5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ,840 320,274\n\nMax 34,102,743 106,596 8,348,993\n\nAverage 22,295,590 63,303 5,593,792\n\nTotal Independent 66,886,770 189,908 16,781,376\n\n% of Market 8% 8% 8%\n\nSource: HFSA raw data.\n\n\n13. The MPFs sponsored by financial institutions typically outsource asset\nmanagement and administration to other specialized companies within the same financial\ngroup. The other MPFs usually outsource these services as well, sometimes to the same\ncompanies, although there are MPFs sponsored by employers that manage their assets\ninternally. When asset management is outsourced, the asset management company\ntypically charges much lower fees to MPFs outside the financial group. There is a marked\nsegmentation in terms of fees that has not translated into increased switching by members\nfrom high fee funds to low fee funds. As noted below, this is due in part to the lack of\nreadily comparable information on returns and fees until recently.\n\n14. Finally, arrangements for revenue collection have remained deficient. Individual\nMPFs deal with many employers, including thousands of small and medium enterprises.\nAs a result, the industry has been considering the creation of a single company specialized\nin revenue collection, and owned by all the MPFs.\n\n\n**III.  PERFORMANCE OF THE SECOND PILLAR:** **GROWTH**\n\n\n15. Membership in the second pillar grew rapidly in the first years of operation, as\nshown in Figure 1. The initial jump in membership reflected the high rate of voluntary\nswitching of younger workers which had already rights under the old system, especially\nworkers under 40 years of age. In the following years, membership was expected to\nincrease at lower but steady rates, in line with the flow of new entrants to the labor force.\n\nMention: \"HFSA raw data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Assets and market shares of independent sponsors are tabulated from it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003095:5:1", "occurrence_id": "ef5e8939be436f1f22bdba06", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  not use factual information to measure\n\n\ndifferences in institutions and regulations across countries.\n\n\nRecently, the World Bank started to publish results from their Doing Business project (World\n\n\nBank 2004). The Doing Business database provides objective measures of business\n\n\nregulations and their enforcement. The Doing Business indicators are comparable across\n\n\neconomies and indicate the regulatory costs of business. They allow us to obtain information\n\n\non regulatory outcomes, such as time and money spent on bureaucratic procedures, and thus\n\n\nto investigate the efficiency of the government institutions in place. By focusing on evidence\n\n\n6\n\nMention: \"Doing Business database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"It provides objective measures of business regulation and its enforcement.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003096:33:1", "occurrence_id": "3a7560e2e3065c07114918a1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: * **0.17**\n\n**Informal Salaried** **0.16** **0.17**\n\n\n\nSmall Firms 0.08 0.10\n\n\n\nBig Firms 0.08 0.08\n\n**Formal Sector** **0.47** **0.47**\n\n\n\nOwners of Firms 0.03 0.04\n\n\n\nEmployees 0.44 0.44\n**Unassigned** 0.03 0.01\n\n\n\n**<u>Table 2: Characteristics of employed workers 1q1987 and 4q2002</u>**\n\n\nAuthor’s calculations using the ENEU. The table shows the mean age, years of schooling (school) weekly\nhours of work (Hours) and the relative average wage with respect to the formal sector.\n\n\n33\n\nMention: \"the ENEU\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Means for age, schooling, hours and relative wages are calculated from it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003100:39:0", "occurrence_id": "00898427474ca001338b5f8f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Table 2: Forecast of Global IT Services and India’s Opportunities\n\nGlobal Market Global Market India’s Exports India’s Global\n\n\n\nShare (2001, %)\n\n\n\nGlobal Market\n\n\n\nIndia’s Global\n\n\n\nPotential for\n\n\nExports\n\n\n\n(2001)\n\nUS $ b %\nShare\n\n\n\n(2005)\n\nUS $ b %\nShare\n\n\n\nIndia’s Exports\n\n\n(2001, US $ b)\n\n\n\n142.9 32.5 238.7 34.1 5.3 3.7\nProfessional Services\n\n\n\nIT Consulting\n\n\nSystems Integration\n\n\nCustom Applications\n\n\nNetwork Consulting &\n\n\nIntegration\n\n\nProduct Services\n\n\nIT Training &\n\n\nEducation\n\n\nH/W Support &\n\n\nInstallation\n\n\nPackaged Software\n\n\nSupport Services\n\n\nOutsourcing Services\n\n\nProcessing Services\n\n\nIS Outsourcing\n\n\nApplication Outsourcing\n\n\nNetwork Infrastructure\n\n\nManagement\n\n\nTotal\n\n\n\n21.3 4.8 31.5 4.5 0.1 0.3 Low-Medium\n\n\n81.1 18.4 142.1 20.3 0.1 0.1 Low-Medium\n\n\n19.3 4.4 25.3 3.6 4.5 23.1 High\n\n\n21.2 4.8 39.8 5.7 0.7 3.3 Low\n\n\n117.9 26.8 176.9 25.3 0.4 0.3\n\n\n\n23.4 5.3 41.8 6.0 0.1 0.3 Medium-High\n\n\n440.0 100.0 700.4 100.0 5.7 1.3\n\n\n\n25.5 5.8 40.9 5.8 \n\n44.4 10.1 49.4 7.1 \n\n\nHigh\n\n\nLow\n\n\n\n48.0 10.9 86.6 12.4 0.4 0.7 Medium-High\n\n\n179.2 40.7 284.8 40.7 0.1 0.0\n\n\n\n78.4 17.8 103.8 14.8 \n\n64.0 14.5 100.2 14.3 \n\n13.4 3.0 39.0 5.6 \n\n\nHigh\n\n\nHigh\n\n\nLow-Medium\n\n\n\nSource: Calculated from NASSCOM 2002: 24, 46 in D’Costa\n\nMention: \"NASSCOM 2002\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Sectoral figures are calculated from this industry publication.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003104:17:1", "occurrence_id": "63a58178d6f2b62df692e516", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  composed of 304\nestablishments included for the first time in that year, and 254 from the old sample not all of\nwhich are to be followed in subsequent years.\n\n\n**7.1.** **Capital**\n\nTo construct the establishment capital stock series, we follow a methodology close to Black and\nLynch (1997). The 1988 Census reports information on the capital stock. We use machinery\ncapital. We avoid overestimation of the amount of depreciation by calculating an average\ndepreciation rate by industrial sector and year. The resulting depreciation rate is then used for all\nfirms within each sector yearly. We further exclude the value of assets sold in our measure of\ncapital, assuming assets have been totally depreciated at that point. Thus, the equation for\nestimating the capital stock for years later than 1988 is:\n\n###### K jit = K jit − 1 + I jit −δ it K jit − 1 (22)\n\nwith\n\n\n18\n\nMention: \"The 1988 Census\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Machinery capital stock reported here anchors the establishment capital series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003109:7:0", "occurrence_id": "08db1817260bc1d8147657c5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: - 8 –\n\n\n_Figure 1: Coefficient of variation in wages and registered unemployment rates in transition_\n_countries 1991/2-1998_\n\n\n\n0,21\n\n\n0,19\n\n\n0,17\n\n\n0,15\n\n\n0,13\n\n\n0,11\n\n\n0,09\n\n\n0,07\n\n\n0,05\n\n\n\n\n\nEstonia\n\n\nBulgaria\n\nHungary\n\n\nPoland\n\n\nSlovenia\n\n\n\n0,55\n\n\n0,50\n\n\n0,45\n\n\n0,40\n\n\n0,35\n\n\n0,30\n\n\n0,25\n\n\n0,20\n\n\n\n\n\nHungary\n\nPoland\n\n\nBulgaria\n\n\n\n1991 1992 1993 1994 1995 1996 1997 1998\n\n\n\n1991 1992 1993 1994 1995 1996 1997 1998\n\n\n\nWages (national currency) Unemployment rate (%)\n\n\nSource: Regspec/AcessLab data base.\n\n\n_Figure 2: Coefficient of variation in GDP per capita and unemployment rates in transition_\n_countries 1999-2001 (NUTS III)_\n\n\n\n0,50\n\n\n0,45\n\n\n0,40\n\n\n0,35\n\n\n0,30\n\n\n0,25\n\n\n0,20\n\n\n0,15\n\n\n\nEstonia\nSlovakia\n\n\nPoland\n\n\nBulgaria\n\nLatvia\n\n\nSlovenia\n\n\n\n0,60\n\n\n0,55\n\n\n0,50\n\n\n0,45\n\n\n0,40\n\n\n0,35\n\n\n0,30\n\n\n0,25\n\n\n0,20\n\n\n0,15\n\n\n0,10\n\n\n\n\n\nEstonia\n\n\nPoland\nLatvia\n\n\n(NUTS II)\n\n\n\nSlovakia\n\n\n\n1999 2000 2001\n\n\n\n1999 2000 2001 2002 2003\n\n\n\nGDP per Capita (Euro) Unemployment rate (%)\n\n\nSource: Eurostat, New Cronos and Goskomstat. – Romania at Nuts II level.\n\n\nIndeed regional disparities in unemployment rates, employment rates and GDP per capita levels are\n\n\ncomparable to those in many of the high unemployment countries in the EU. This is documented in\n\n\nTable 2 which uses the most recent data available from the New Cronos database. Differences between\n\n\nthe regions with the highest and the lowest unemployment rate exceeded a factor of 3 in the year 2003\n\n\nin all but the smallest transition countries (Slovenia, Latvia, Lithuania) and Romania. GDP per capital\n\n\nlevels ranged from 70 - 80% to up to over 200% of the national average in the majority of transition\n\n\neconomies. Finally, differences between the regions\n\nMention: \"Regspec/AcessLab data base\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Wages and unemployment rates for the transition years are plotted from it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003109:7:0", "occurrence_id": "20a7986699cb1f81b4e5994d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: - 8 –\n\n\n_Figure 1: Coefficient of variation in wages and registered unemployment rates in transition_\n_countries 1991/2-1998_\n\n\n\n0,21\n\n\n0,19\n\n\n0,17\n\n\n0,15\n\n\n0,13\n\n\n0,11\n\n\n0,09\n\n\n0,07\n\n\n0,05\n\n\n\n\n\nEstonia\n\n\nBulgaria\n\nHungary\n\n\nPoland\n\n\nSlovenia\n\n\n\n0,55\n\n\n0,50\n\n\n0,45\n\n\n0,40\n\n\n0,35\n\n\n0,30\n\n\n0,25\n\n\n0,20\n\n\n\n\n\nHungary\n\nPoland\n\n\nBulgaria\n\n\n\n1991 1992 1993 1994 1995 1996 1997 1998\n\n\n\n1991 1992 1993 1994 1995 1996 1997 1998\n\n\n\nWages (national currency) Unemployment rate (%)\n\n\nSource: Regspec/AcessLab data base.\n\n\n_Figure 2: Coefficient of variation in GDP per capita and unemployment rates in transition_\n_countries 1999-2001 (NUTS III)_\n\n\n\n0,50\n\n\n0,45\n\n\n0,40\n\n\n0,35\n\n\n0,30\n\n\n0,25\n\n\n0,20\n\n\n0,15\n\n\n\nEstonia\nSlovakia\n\n\nPoland\n\n\nBulgaria\n\nLatvia\n\n\nSlovenia\n\n\n\n0,60\n\n\n0,55\n\n\n0,50\n\n\n0,45\n\n\n0,40\n\n\n0,35\n\n\n0,30\n\n\n0,25\n\n\n0,20\n\n\n0,15\n\n\n0,10\n\n\n\n\n\nEstonia\n\n\nPoland\nLatvia\n\n\n(NUTS II)\n\n\n\nSlovakia\n\n\n\n1999 2000 2001\n\n\n\n1999 2000 2001 2002 2003\n\n\n\nGDP per Capita (Euro) Unemployment rate (%)\n\n\nSource: Eurostat, New Cronos and Goskomstat. – Romania at Nuts II level.\n\n\nIndeed regional disparities in unemployment rates, employment rates and GDP per capita levels are\n\n\ncomparable to those in many of the high unemployment countries in the EU. This is documented in\n\n\nTable 2 which uses the most recent data available from the New Cronos database. Differences between\n\n\nthe regions with the highest and the lowest unemployment rate exceeded a factor of 3 in the year 2003\n\n\nin all but the smallest transition countries (Slovenia, Latvia, Lithuania) and Romania. GDP per capital\n\n\nlevels ranged from 70 - 80% to up to over 200% of the national average in the majority of transition\n\n\neconomies. Finally, differences between the regions\n\nMention: \"New Cronos database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its most recent figures document regional unemployment differences across candidate members.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003111:9:1", "occurrence_id": "49b04bd7309a962c3c1bb030", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  put recent experience in perspective. As discussed above, the PRS can best help focus\npolicies on poverty if it is consonant with domestic political systems. Fundamental to facilitating\nparticipation are the state policy and practice on governance, rule of law, respect for human\nrights, freedom of expression and association, corruption, defense expenditure, and conflict.\n\n\nThe prevailing governance environment in most PRS countries suggests that implementation of\nthe approach presents both opportunities and challenges. Figure 1 presents a scatter diagram of\ntwo often-quoted indicators of governance, namely the “control of corruption” from the\nGovernance Matters Scores, and the index of political rights from the Freedom House. Quite\nvividly—and not surprisingly, perhaps—the chart shows that the universe of countries that have\nengaged in the PRS process is overwhelmingly characterized by high levels of corruption and\nrestricted freedoms for their citizens. There are of course exceptions (the two countries in the\n“blessed” quadrant being Cape Verde and Madagascar), but overall the participatory processes\ninduced by the PRS initiative have evolved in very difficult environments. Consequently,\nexpectations with regard to their potential results have to be kept in perspective.\n\n\n13 Ackerman. Idem. p.17.\n\n\n10\n\nMention: \"Governance Matters Scores\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its control of corruption measure is one of the two governance indicators plotted.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003111:9:1", "occurrence_id": "6ab42f24ec756ccb564c53a3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  put recent experience in perspective. As discussed above, the PRS can best help focus\npolicies on poverty if it is consonant with domestic political systems. Fundamental to facilitating\nparticipation are the state policy and practice on governance, rule of law, respect for human\nrights, freedom of expression and association, corruption, defense expenditure, and conflict.\n\n\nThe prevailing governance environment in most PRS countries suggests that implementation of\nthe approach presents both opportunities and challenges. Figure 1 presents a scatter diagram of\ntwo often-quoted indicators of governance, namely the “control of corruption” from the\nGovernance Matters Scores, and the index of political rights from the Freedom House. Quite\nvividly—and not surprisingly, perhaps—the chart shows that the universe of countries that have\nengaged in the PRS process is overwhelmingly characterized by high levels of corruption and\nrestricted freedoms for their citizens. There are of course exceptions (the two countries in the\n“blessed” quadrant being Cape Verde and Madagascar), but overall the participatory processes\ninduced by the PRS initiative have evolved in very difficult environments. Consequently,\nexpectations with regard to their potential results have to be kept in perspective.\n\n\n13 Ackerman. Idem. p.17.\n\n\n10\n\nMention: \"index of political rights\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The second governance indicator in the comparison.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003113:5:1", "occurrence_id": "269e32e48ad769ce15d34ce1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  by the OECD\n\n\ncompared with $90 billion in the GTAP database). This is not surprising because the\n\n\nOECD estimates are the source for that part of the GTAP protection database. To that\n\n3 A side issue is how the OECD’s PSE compares with the Aggregate Measure of Support (AMS) that\nmembers notify to the WTO as part of their commitments under the Uruguay Round Agreement on\nAgriculture. The AMS refers only to the domestic support pillar, and it excludes measures that are not\nsubject to reduction commitments (so-called blue box and green box measures). For a comparison of the\nPSE and AMS methodologies, see Diakosavvas (2002). In 1999 (the most recent year for which there has\nbeen full notification to the WTO) the AMS was $88 billion for high-income countries and $2 billion for\ndeveloping countries. The $88 billion comprises $52 billion from market price support and $36 billion from\ndirect domestic subsidies. By contrast, the PSE for just OECD countries was $273 billion in that year. Of\nthat latter amount, $182 billion was from market price support measures and only the residual ($91 billion)\nwas direct domestic subsidies. That residual is further reduced, to $43 billion, when measures that are\nsomewhat decoupled from production are excluded. Since those decoupled measures are not counted as\npart of the AMS, that $43 billion is comparable with the $36 billion notified to the WTO as that component\nof the AMS.\n4 For the purpose of this calculation, we assume in this section that domestic and imported farm products\nare homogeneous, so that a tariff equal to 20 percent of the value of imports will raise the price of domestic\nproduction by 20 percent.\n\nMention: \"the GTAP database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its protection aggregate differs from the OECD estimate, partly by construction.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003116:7:0", "occurrence_id": "d3867953ec5f053dc0d8985a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: to the same municipal areas (“villages”). The paper uses the 1998 wave because this is\n\n\nthe only year for which detailed information on household migration (previous migration\n\n\nspells and area of birth of actual household members) was asked in the EHPM. Such\n\n\ninformation will be useful in the construction of instrumental variables for the analysis\n\n\nthat follows. The 1998 wave contains a total of 11,953 households belonging to 111\n\n\n“villages” in the country, with all regions nearly equally represented (3,740 households\n\n\nfrom the Metropolitan region, 2,903 from the Western region, 2,780 from the Central\n\n\nregion, and 2,952 from the Eastern region).\n\n\nA panel structure would be ideal to help solve the potential selection problem by\n\n\nincluding households’ fixed effects and, in this way, exploit the variability of remittances\n\n\nwithin a household across time. Unfortunately, this is not possible as each survey does\n\n\nnot follow the same group of households. The data are also incomplete in that, for the\n\n\nsender, only the number of migrant family members is known (the question refers to the\n\n\n“number of members from this household living abroad”).\n\n\nConcerning remittances, the survey questions include whether the household\n\n\nreceives transfers from abroad, as well as their amount and the frequency. Information on\n\n\nthe items (food, housing, health, education, savings) the family decides to spend money\n\n\non, from the amount remitted, is also available. However, as pointed out by Cox-Edwards\n\n\nand Ureta (2003), it is easy to doubt the reliability of such information, as people tend to\n\n\npool remittances and earned income when expenditure decisions are made. Recall bias\n\n\nalso may be an important source of measurement error in such questions.\n\n\n8\n\nMention: \"the EHPM\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its migration history and place of birth items support the instrumental variables.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003117:28:1", "occurrence_id": "40e87688dc51af28baed35c3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  the Post Harvest Surveys 1997-2002.\n\n\nTable 2\nPercentage of Farmers Growing Cotton\n\n1997           - 2002\n\n\n<u>Province</u> <u>1997</u> <u>1998</u> <u>1999</u> <u>2000</u> <u>2001</u> <u>2002</u>\n\n\nCentral 24.6 22.6 16.6 10.3 14.7 20.2\nEastern 35.2 32.7 31.7 20.4 32.1 39.0\nSouthern 9.9 10.7 11.7 4.3 8.8 12.8\n<u>Lusaka</u> <u>5.4</u> <u>3.3</u> <u>4.7</u> <u>0.4</u> <u>5.1</u> <u>8.2</u>\n\n\nCopperbelt 0.8 0.6 0.3 0.0 0.0 0.0\nLuapula 0.0 0.0 0.0 0.0 0.0 0.0\nNorthern 0.0 0.1 0.0 0.0 0.0 0.0\nNWwestern 0.3 0.0 0.0 0.0 0.0 0.2\n<u>Western</u> <u>1.3</u> <u>0.6</u> <u>0.4</u> <u>0.1</u> <u>0.1</u> <u>0.0</u>\n\n\n<u>Total</u> <u>11.0</u> <u>10.4</u> <u>9.4<\n\nMention: \"the Post Harvest Surveys 1997-2002\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Provincial shares of farmers growing cotton are drawn from these rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003120:18:0", "occurrence_id": "3a436c0a0747f676441fa348", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: somewhat biased toward industrialized countries, especially in the earlier years of the panel; this\n\n\nis because many country pairs that consist of less-developed countries do not have bilateral tariff\n\n\ndata for either one. Therefore we perform all regressions with and without controlling for tariffs.\n\n\nWe use the sum of absolute value of latitude for a country-pair as an instrument for\n\n\ninstitutional quality variables. The latitude variables are taken from the CIA Factbook.\n\n\nIn Table 2 we report summary statistics for all our regressors. Sixty-five percent of the\n\n\ncountry-pairs in our sample do not share the same legal origin. Our institutional quality variables\n\n\nexhibit significant variation. It is noteworthy that the average tariff variable has some extreme\n\n\noutliers, usually due to countries that trade very few products and impose a high tariff vis-à-vis\n\n\neach other <sup>24</sup> . In Table 3 we report correlations that were of interest to us a priori <sup>25</sup> . Notably\n\n\namong these correlations is the small positive correlation between trade flows and different legal\n\n\norigins. The institutional quality variables are highly correlated among themselves. The\n\n\ninstitutional indicators also exhibit a relatively high correlation with trade flows, telephone lines\n\n\nand with roads. Our two infrastructure indicators are also correlated, which will lead us not to\n\n\nestimate specifications using both in order to avoid multi-colinearity problems. Not surprisingly,\n\n\ntariffs are negatively correlated with trade. They are also negatively correlated with institutional\n\n\nquality.\n\n\n24 It is also indicative that average tariffs between trade-pairs (even trade-weighted) mask huge variation in tariffs\namong different products for each trade-pair.\n25 All correlations are statistically significant at the 1% level except for correl(regional,comcol), which is not\nstatistically significant at conventional levels.\n\n\n18\n\nMention: \"CIA Factbook\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Latitude variables used as instruments are taken from it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003121:21:4", "occurrence_id": "6afe5f54365492fe817ace72", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 84<br>76<br>72<br>82<br>90<br>91<br>78<br>89<br>64<br>73<br>83<br>79<br>73<br>75<br>62<br>64<br>74<br>66<br>71<br>66<br>92<br>83<br>83<br>83<br>76<br>77<br>91<br>87<br>20<br>25<br>81<br>78<br>61<br>49<br>81<br>70<br>21<br>42<br>63<br>56|\n||All above products|8,097<br>19,626|242|48.2<br>55.5|<br>|69<br>71|\n|0 to 9|All goods|16,803<br>35,378|211|100<br>100|<br>|55<br>55|\n\n\nSource: Derived from the UN COMTRADE database as reported by the EU.\n\n\nThe composition of exports of top-twenty performers in terms of technology content and\nnatural resource intensity also points to significant realignment in Turkish export offer. The share\n\nMention: \"the UN COMTRADE database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Export composition for the top performers is derived from it as reported by the EU.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003124:18:1", "occurrence_id": "aca22e6580a0f8cbfaffb8bc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  compromises. Option five would allocate all the\n\n\n18 The rationale is that HDI measures the state of welfare of the people and GRDP per capita measures\neconomic potential and activity of one region (Warta Keuangan, 2 <sup>nd</sup> Edition 2005).\n19 Simulation two uses poverty indicators from DAU 2005 basic data from MoF.\n\n\n17\n\nMention: \"basic data from MoF\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Poverty indicators underlying the grant simulation come from these fiscal records.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003125:56:0", "occurrence_id": "83951a3e0908a1b1507570f7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Flat Returns_\n\n\nWhile skill supply increased, has is kept up with demand? If demand is outpacing supply,\none would expect to see the wage gap between workers with skills and without skills to\nincrease. Figure 24.A (all wage workers) and Figure 24.B (monthly wage workers only)\nsuggest that, first, more education is associated with higher wages at any point in time;\nand, second, real monthly wages trended upward until 1996, fell during the crisis, and\nremained roughly constant since, especially for secondary or higher education levels.\nWage data from the national Socio-Economic Surveys (SES) from 1986 to 2002 for wage\nworkers confirms this pattern, although the actual wage time series differ somewhat\n(Figure 24.C).\n\n\nCombining wages and education, it is possible to estimate returns to education over time\nusing Mincer regressions. In its most basic form, log hourly wages are regressed on\neducation, controlling for age, aged squared, gender, province and urbanization. <sup>21</sup> [^21: The calculation yields, under certain conditions, private returns. Social returns would require\nincorporating the public costs of providing education net of any external effects (through endogenous\ngrowth other channels) of education (Psacharopoulos 1995). The analysis makes no attempt at measuring\nthese effects, implicitly assuming that they have not changed differentially across skill levels over time.] Figure\n24.D shows the coefficients on education levels from a series of such cross-sectional\nregressions for wage workers. Education is divided into six attainment levels (less than\nprimary, primary, lower secondary, upper secondary, vocational and university) and the\nreturns are estimated relative to workers with less than primary education, the omitted\ncategory. The returns relative to less than primary education are flat over time and there\nis no trend towards higher wage differentials across skill levels. If anything, there is a\nsmall downward shift of returns to university post crisis.\n\n\nA robustness check is to restrict the sample to workers on a monthly\n\nMention: \"national Socio-Economic Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Wages of wage workers over this period confirm the education pattern.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003130:22:1", "occurrence_id": "fb0c32f2f15758ce5f4f09af", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  _18_ _0.2_ _7.4_ -1.0 -4.5 -0.8\n\nSoutheast Asia _1_ _1_ _-1.5_ _18.4_ -2.3 -8.4 -1.4\n\nSouth Asia _21_ _3_ _-1.0_ _24.5_ -1.0 -8.8 -0.6\n\nMiddle East and North Africa _8_ _7_ _0.3_ _-3.3_ -1.6 -7.8 -1.5\n\nSub-Saharan Africa (excl S. Africa) _5_ _13_ _1.1_ _-1.5_ -4.6 -7.5 -4.4\n\nLatin America and Carib. _6_ _4_ _-0.5_ _4.9_ -2.5 -8.9 -2.1\n\n**<u>World</u>** **_<u>100</u>_** **_<u>100</u>_** **_<u>0.0</u>_** **_<u>0.0</u>_** **<u>0.2</u>** **<u>-1.1</u>** **<u>-1.0</u>**\n\n\na Exports minus imports, both valued at f.o.b. prices as in the GTAP database 6.05\n\nSource: Authors’ GTAP model simulation results and (for columns 1 to 4) the GTAP database\n\nMention: \"the GTAP database 6.05\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Export and import values at f.o.b. prices are taken from this version.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003136:6:2", "occurrence_id": "2466f36786a14d624d9954c4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .4 3.9 2.5 5.4 4.8 0.9\n% with university education 3.2 4.6 0.7 5.4 5.2 0.7\n\n\n<u>No of. Uncensored observations</u> <u>2011</u> <u>1293</u> <u>718</u> <u>878</u> <u>270</u> <u>453</u>\n_<u>Note: standard deviations (STD) are not reported for dummy variables.</u>_\n_Source: Author's calculation based on EDAMS 1996._\n\n\nWages are the sum of cash earnings. Monthly wages <sup>6</sup> are obtained by dividing reported\n\n\nmonthly wages by imputed monthly hours. The average monthly earning is USD 288 with\n\nmales earning proportionally more on average (USD 354 vs. USD 169 for females). %.\n\n\nPublic sector monthly wages are on average higher by USD 26.9 relative to wages in the\n\nprivate formal sector.\n\n\nForty-three percent of workers are employed in the public sector, 14 percent in the private\n\n\nformal sector, and 22 percent in the informal sector. Males make up 64.3 percent of overall\n\nwage earners. More than 75 percent of male workers are married and more than 58 percent\n\n\nwork in the public sector. In contrast, 74 percent of female workers are unmarried and only\n\n\n6 Daily wages are converted in monthly figures by multiplying with 26 days. Similarly weekly wages are\nconverted to monthly wages by multiplying with 4.2 weeks.\n\nMention: \"EDAMS 1996\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Monthly wages are constructed from cash earnings reported here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003146:10:0", "occurrence_id": "9f31831182f0de9edc0e3adc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: sales rather than earnings to avoid the problems generated by the volatility and manipulability of\n\n\nearnings. To control for time-varying industry-level effects, we include each firm’s global\n\n\nindustry _q_, which is computed by averaging across all corporations within the firm’s industry.\n\n\nTo control for country factors, we include real GDP growth, which comes form the\n\n\nWorld Bank World Development Indicators. In robustness tests, we control for additional\n\n\ncountry traits that might affect not only a firm’s _q_ but also its willingness and ability to access\n\n\ninternational markets, including a country’s institutional quality, shareholder rights, legal origin,\n\n\ndomestic market capitalization, and an index of accounting standards.\n\n\nAfter removing financial firms (since highly leveraged and heavily regulated financial\n\n\ninstitutions could be valued differently from nonfinancial firms), firms with missing data, firms\n\n\nfrom the United States and the United Kingdom (since these are financial centers where most\n\n\ninternationalization is taking place), and firms with less than three observations, we are left with\n\n\na sample of 9,096 firms from 74 countries covering the period 1989 to 2000, totaling 66,963\n\n\nfirm-year observations. Appendix Table 1 lists the countries, the number of domestic and\n\n\ninternational firms per country, the coverage period for each country, and summary statistics on\n\n\n_q_ . Some countries do not have any international firms. We keep these in the sample as a control\n\n\ngroup, but emphasize that this paper’s results hold when we exclude countries with zero or only\n\n\none international firm. Also, Japanese firms represent about 30 percent of the total firms in our\n\n\nsample. We therefore re-did our analyses excluding Japanese firms and reached the same\n\n\nconclusions reported below.\n\n\n**3. Results: Before and After Internationalization**\n\n\nThis section tests whether there is a significant increase in _q_ after firms internationalize.\n\n\nWe compare the average valuation of firms in the years before they internationalize to average\n\n\n10\n\nMention: \"World Bank World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Real output growth enters as the country-level control.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003151:18:0", "occurrence_id": "b9cd1b712d4ce9a56791581a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: good match. All financial institutions are required by law to obtain information about a borrower\n\n\nfrom the CIB before granting or renewing a loan. This provision was introduced in the effort to\n\n\nreduce non-performing loans in the banking system.\n\n\nWhile the CIB has a long and positive track record, several shortcomings of the existing\n\n\nsystem should be noted. First, the system is still limited by the lack of computization. Second,\n\n\nbanks effectively only obtain current information about existing borrowers, but no historic\n\n\ninformation, i.e. information about past payment performance. One can imagine defaulting\n\n\nborrower paying up to one institution before taking out new loans with another institution – the\n\n\nCIB reports do not seem to be able to capture such behavior. Third, only financial institutions\n\n\nparticipate in the information sharing agreement, other non-financial institutions such as utility\n\n\ncompanies, which are a source of valuable credit information about customers in other countries\n\n\nare excluded.  Most importantly, however, the CIB has a stability focus – credit information\n\n\nsharing has the objective to prevent future loan losses, which is an important and valid aspect of\n\n\ncredit information sharing. As important, however, is the function of credit information sharing\n\n\nin enhancing competitiveness and access to credit by allowing borrowers to create “reputation\n\n\ncollateral”, i.e. allow them to use their track record as reference in future loan applications. The\n\n\nlack of historical information limits such a role of the CIB. More importantly, the threshold of\n\n\n50,000 BDT (800 USD) excludes the class of borrowers that around the world rely most on\n\n\nreputation collateral – small and opaque enterprises. While technical problems are certainly\n\n\nrecognized, a move towards including all borrowers irrespective of the size of their loans should\n\n\nbe envisioned in the medium-term. Another important step would be to link the existing CIB for\n\n\nbanks and financial institutions to the credit registry to be created for MFIs. Finally, there does\n\n\nnot seem any reason to prevent private credit information sharing, i.e. to\n\nMention: \"CIB reports\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Sequential defaulting across institutions is not captured by these reports.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003151:18:0", "occurrence_id": "d69e22bdb234e1ff041fd0af", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: good match. All financial institutions are required by law to obtain information about a borrower\n\n\nfrom the CIB before granting or renewing a loan. This provision was introduced in the effort to\n\n\nreduce non-performing loans in the banking system.\n\n\nWhile the CIB has a long and positive track record, several shortcomings of the existing\n\n\nsystem should be noted. First, the system is still limited by the lack of computization. Second,\n\n\nbanks effectively only obtain current information about existing borrowers, but no historic\n\n\ninformation, i.e. information about past payment performance. One can imagine defaulting\n\n\nborrower paying up to one institution before taking out new loans with another institution – the\n\n\nCIB reports do not seem to be able to capture such behavior. Third, only financial institutions\n\n\nparticipate in the information sharing agreement, other non-financial institutions such as utility\n\n\ncompanies, which are a source of valuable credit information about customers in other countries\n\n\nare excluded.  Most importantly, however, the CIB has a stability focus – credit information\n\n\nsharing has the objective to prevent future loan losses, which is an important and valid aspect of\n\n\ncredit information sharing. As important, however, is the function of credit information sharing\n\n\nin enhancing competitiveness and access to credit by allowing borrowers to create “reputation\n\n\ncollateral”, i.e. allow them to use their track record as reference in future loan applications. The\n\n\nlack of historical information limits such a role of the CIB. More importantly, the threshold of\n\n\n50,000 BDT (800 USD) excludes the class of borrowers that around the world rely most on\n\n\nreputation collateral – small and opaque enterprises. While technical problems are certainly\n\n\nrecognized, a move towards including all borrowers irrespective of the size of their loans should\n\n\nbe envisioned in the medium-term. Another important step would be to link the existing CIB for\n\n\nbanks and financial institutions to the credit registry to be created for MFIs. Finally, there does\n\n\nnot seem any reason to prevent private credit information sharing, i.e. to\n\nMention: \"credit registry\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"A planned registry for microfinance institutions would be linked to the existing system.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003153:13:0", "occurrence_id": "397f8422e267a518fd8da423", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: index of the degree to which the government protects private property and enforces laws\n\n\nthat protect private property. The data are for 2000 and were obtained from the Index of\n\n\nEconomic Freedom constructed by the Heritage Foundation. The index is available for a\n\n\nlarge number of countries and has been recently used in several papers including\n\n\nJohnson, Kaufmann, and Zoido-Lobaton (1998), LLSV (1999, 2002), Beck, Demirguc\n\nKunt and Levine (2003) and Claessens and Laeven (2003). Within our sample, property\n\n\nrights varies from a score of 5 for countries with good property rights protection like the\n\n\nUnited States to 1 for countries like the Congo Democratic Republic, Libya and Vietnam.\n\n\nThe countries in our sample belong to different legal traditions and the data on\n\n\nlegal families is taken from LLSV (1998). Since the literature has argued that common\n\n\nlaw countries have a significant advantage over civil and socialist law traditions and\n\n\nregressions typically only distinguish between common law countries and civil law\n\n\ncountries (see, for instance, LLSV(1998), Stulz and Williamson (2003), Beck, Demirguc\n\nKunt and Levine (2003)), we use the dummy variable **Common Law,** which takes the\n\n\nvalue 1 for English common law countries and 0 for countries of all other legal traditions.\n\n\nMoreover, it is not clear what the distinctions between the civil law (Scandinavian,\n\n\nGerman and French) countries really mean. For instance, Nenova (2003) shows that the\n\n\nbenefits from control are lower in countries with a Scandinavian civil law tradition than\n\n\nin common law countries while Coffee (2001) argues that social norms rather than legal\n\n\nregimes can explain these lower benefits of control. Small sample sizes of German and\n\n\nScandinavian civil law countries also prevent us from making finer distinctions between\n\n\nthe civil law countries.\n\n\n14\n\nMention: \"Index of\n\n\nEconomic Freedom\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its property rights measure is the variable used for that year.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003155:10:0", "occurrence_id": "6e1a22e4c77b572ad50b1d2b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: where Y is GDP, _dist_ is the bilateral distance between country _i_ and country _j_, _bord_ is a\ndummy for whether the two countries share a border, _isl_ is a dummy for being an island,\n_llock_ is a dummy for whether the country is landlocked, and _open_ measures the degree of\nopenness.\n\nWe estimate foreign market access here at the city level rather than at the country level,\nby replacing the distance variable _dist_ by the variable “international distance,”  What we\nrefer to here as national supplier access is the same thing as the supplier access variable\nin the Redding and Venables model. In addition, we measure supplier access at the local\nor city level as well, by modifying equation (13) to a) replace own country GDP, _Yii_, by\nown local GDP as defined above, and b) replace bilateral distance by our “international\ndistance”; and c) setting domestic distance, _distii_ to unity.\n\nTo obtain the three access variables listed in Table 1, we needed the list of trading\npartners of our countries of interest and estimates of the ρ -parameters of equations (12)\nand (13). Redding and Venables (2000) obtained the parameters from the estimation of a\ngravity equation on aggregate data on  a cross-section of 101 countries. We too have\nused their estimates here. <sup>13</sup> [^13: We thank Steve Redding for kindly providing the estimates. The equation estimated in Redding and Venables\n(2000) is] We identified trading partners of the countries of interest\nbased on the COMTRADE Database. <sup>14</sup> [^14: Because trade statistics were incomplete for Africa, we had to use the “mirrored” (reported by partners) data for\nAfrican countries. We limited this exercise to the top ten trading partners of each country since these accounted for\nmore than ninety percent of annual trade volumes in each case.] Country GDP figures were obtained from World\nBank’s World Development Indicators. Domestic\n\nMention: \"COMTRADE Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Trading partners are identified from it, using mirrored reports where statistics are incomplete.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003155:10:0", "occurrence_id": "c4de7e6a6ed9e4182dffcf85", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: where Y is GDP, _dist_ is the bilateral distance between country _i_ and country _j_, _bord_ is a\ndummy for whether the two countries share a border, _isl_ is a dummy for being an island,\n_llock_ is a dummy for whether the country is landlocked, and _open_ measures the degree of\nopenness.\n\nWe estimate foreign market access here at the city level rather than at the country level,\nby replacing the distance variable _dist_ by the variable “international distance,”  What we\nrefer to here as national supplier access is the same thing as the supplier access variable\nin the Redding and Venables model. In addition, we measure supplier access at the local\nor city level as well, by modifying equation (13) to a) replace own country GDP, _Yii_, by\nown local GDP as defined above, and b) replace bilateral distance by our “international\ndistance”; and c) setting domestic distance, _distii_ to unity.\n\nTo obtain the three access variables listed in Table 1, we needed the list of trading\npartners of our countries of interest and estimates of the ρ -parameters of equations (12)\nand (13). Redding and Venables (2000) obtained the parameters from the estimation of a\ngravity equation on aggregate data on  a cross-section of 101 countries. We too have\nused their estimates here. <sup>13</sup> [^13: We thank Steve Redding for kindly providing the estimates. The equation estimated in Redding and Venables\n(2000) is] We identified trading partners of the countries of interest\nbased on the COMTRADE Database. <sup>14</sup> [^14: Because trade statistics were incomplete for Africa, we had to use the “mirrored” (reported by partners) data for\nAfrican countries. We limited this exercise to the top ten trading partners of each country since these accounted for\nmore than ninety percent of annual trade volumes in each case.] Country GDP figures were obtained from World\nBank’s World Development Indicators. Domestic\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Country output figures are taken from this compilation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003156:18:0", "occurrence_id": "2b39b3e14ed8836b9d88e339", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 8a: Income share on subsistence consumption by LAC urban quintiles**\n**(current US$)**\n\n\n10%\n\n\n\n9%\n\n\n8%\n\n\n7%\n\n\n6%\n\n\n5%\n\n\n4%\n\n\n3%\n\n\n2%\n\n\n1%\n\n\n0%\n\n\n\n\n|Col1|Col2|\n|---|---|\n|||\n|||\n||Quintile 1<br>Quintile 2<br>Quintile 3<br>Quintile 4<br>Quintile 5|\n|||\n|||\n|||\n|||\n\n\n\n1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19\n\n\nCost of subsistence consumption (US$/m)\n\n\nSource: Own elaboration from LSMS household survey data\n\n**Figure 8b: Income share on subsistence consumption by LAC urban quintiles (PPP)**\n\n\n\n\n\n\n\n\n\nSource: Own elaboration from LSMS household survey data\n\n\n19\n\nMention: \"LSMS household survey data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Subsistence consumption costs and income shares by quintile are elaborated from them.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003164:25:0", "occurrence_id": "c2f42523f4ad07ec04133e71", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **<u>Annex Table 4. Openness Rankings of Country: GATS vs. Actual Practice</u>**\n\n\n\nRanked by average\nGATS commitment\n\n\n\nRanked by average\n\n\n\n<u>Rank</u> <u>Country</u> <u>index</u> <u>Country</u> <u>EBRD score</u>\n\n1 Moldova 78.4 Hungary 3.1\n2 Kyrgyz Rep. 59.3 Poland 2.9\n3 Latvia 59.1 Estonia 2.68\n4 Georgia 53.5 Czech Republic 2.6\n5 Croatia 49.3 Slovenia 2.47\n6 Lithuania 46.7 Latvia 2.3\n7 Hungary 44.4 Croatia 2.21\n8 Estonia 44.3 Lithuania 2.19\n9 Macedonia, FYR 42.1 Slovak Republic 2.1\n10 Armenia 41.9 Bulgaria 2.08\n11 Slovak Rep. 31.6 Romania 1.96\n12 Slovenia 31.1 Macedonia, FYR 1.82\n13 Czech Republic 30.8 Armenia 1.82\n14 Bulgaria 26.4 Moldova 1.8\n15 Romania 25 Georgia 1.62\n<u>16 Poland</u> <u>24.5</u> <u>Kyrgyz Rep.</u> <u>1.56</u>\n_Source_ : EBRD (2004) and author’s calculations based on GATS schedules.\n\n\n\n<u>Rank</u> <u>Country</u>\n\n\n\n<u>index</u> <u>Country</u>\n\n\n\n25\n\nMention: \"GATS schedules\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Commitment indices are calculated from these schedules.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003170:2:1", "occurrence_id": "c65c86510bca8f1dc332e683", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \n4 A recent survey of central banks in 40 countries reveals that most countries (90 percent of the sample to be exact)\ncollect remittance statistics from commercial banks, while less than 40 percent gather information from money\ntransfer companies and post offices (De Luna Martinez, 2005). Therefore, balance of payment statistics tend to\nbetter reflect the portion of remittances that is transferred through banks.\n\n\n2\n\nMention: \"survey of central banks\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It found ninety percent of forty countries compile remittance statistics.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003176:35:0", "occurrence_id": "b485327f51de5ed9742f6dcf", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 3**\n\n\n\n**Domestic Stock Market Development**\n\n\n\nThis figure shows the evolution of domestic stock market capitalization over GDP and value traded domestically over\nGDP. The series are averages across countries. The data for G-7 countries are averages for Canada, France, Germany,\nItaly, Japan, United Kingdom, and United States. The data for East Asian countries are averages for Hong Kong,\nIndonesia, Korea, Malaysia, Philippines, Taiwan, and Thailand. The data for Latin American countries are averages\nfor Argentina, Brazil, Chile, Colombia, Mexico, Peru, and Venezuela.\n\n\n\n**Market Capitalization / GDP**\n\n\n160%\n\n\n140%\n\n\n120%\n\n\n100%\n\n\n80%\n\n\n60%\n\n\n40%\n\n\n20%\n\n\n0%\n\n\n**Value Traded Domestically / GDP**\n\n\n120%\n\n\n100%\n\n\n80%\n\n\n60%\n\n\n40%\n\n\n20%\n\n\n0%\n\n\nLatin American Countries G-7 Countries East Asian Countries\n\n\nSource: S&P Stock Markets Factbook, World Bank\n\n\n35\n\nMention: \"S&P Stock Markets Factbook\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Market capitalisation shares by country group are taken from this factbook.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003181:14:0", "occurrence_id": "4897115cf6f3c17cb3de712f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Changes in methods, as well as in content, can reduce over-time comparability of\n\n\nindicators. Admirably, WEF has tried to increase the response rate of its Executive\n\n\nOpinion Survey, to enhance accuracy by making the sample more representative. <sup>12</sup>\n\n\nHowever, progress on this front can affect apparent trends. Suppose executives with the\n\n\nstrongest opinions are the most likely to respond, and that strong opinions tend to be\n\n\nunfavorable. An increase in the response rate from one year to the next would then\n\n\nreduce the negative bias, but the year-on-year _change_ would be biased toward showing\n\n\nimprovement.\n\n\n**3. Composite Corruption Indexes**\n\n\nThere are at least three possible justifications for constructing a single corruption\n\n\nindex from multiple, distinct sources of corruption indicators. The first motive\n\n\nemphasizes substantive content: individual indicators, or even several indicators from one\n\n\nsource such as the BEEPS, may be defined too narrowly for certain purposes. For\n\n\nexample, no matter how many corruption indicators one aggregates from the BEEPS, the\n\n\nresulting index still reflects only corrupt interactions between firms and public officials.\n\n\nThe second motive is to reduce measurement error. Given the obvious difficulties\n\n\nin measuring corruption, any one source may be highly inaccurate. However, if errors in\n\n\nmeasurement are largely independent across sources, the errors will tend to cancel out\n\n\nwhen data are aggregated from multiple sources.\n\n\nThe third motive is to cover a larger number of countries. No one source covers\n\n\nall countries. Some sources do not overlap at all in country coverage, for example the\n\n\nUNECA’s African Governance Indicators and Nations in Transit.\n\n\n12 The mean number of responses per country increased from 84 in 2004 to 94 in 2005.\n\n\n15\n\nMention: \"Executive\n\n\nOpinion Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Response rates rose and the sample was made more representative, improving accuracy.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003181:14:0", "occurrence_id": "e6cbed557fba46bba0c6177a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Changes in methods, as well as in content, can reduce over-time comparability of\n\n\nindicators. Admirably, WEF has tried to increase the response rate of its Executive\n\n\nOpinion Survey, to enhance accuracy by making the sample more representative. <sup>12</sup>\n\n\nHowever, progress on this front can affect apparent trends. Suppose executives with the\n\n\nstrongest opinions are the most likely to respond, and that strong opinions tend to be\n\n\nunfavorable. An increase in the response rate from one year to the next would then\n\n\nreduce the negative bias, but the year-on-year _change_ would be biased toward showing\n\n\nimprovement.\n\n\n**3. Composite Corruption Indexes**\n\n\nThere are at least three possible justifications for constructing a single corruption\n\n\nindex from multiple, distinct sources of corruption indicators. The first motive\n\n\nemphasizes substantive content: individual indicators, or even several indicators from one\n\n\nsource such as the BEEPS, may be defined too narrowly for certain purposes. For\n\n\nexample, no matter how many corruption indicators one aggregates from the BEEPS, the\n\n\nresulting index still reflects only corrupt interactions between firms and public officials.\n\n\nThe second motive is to reduce measurement error. Given the obvious difficulties\n\n\nin measuring corruption, any one source may be highly inaccurate. However, if errors in\n\n\nmeasurement are largely independent across sources, the errors will tend to cancel out\n\n\nwhen data are aggregated from multiple sources.\n\n\nThe third motive is to cover a larger number of countries. No one source covers\n\n\nall countries. Some sources do not overlap at all in country coverage, for example the\n\n\nUNECA’s African Governance Indicators and Nations in Transit.\n\n\n12 The mean number of responses per country increased from 84 in 2004 to 94 in 2005.\n\n\n15\n\nMention: \"BEEPS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its individual indicators may be defined too narrowly for some purposes.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003181:14:0", "occurrence_id": "1266b33e30075e1c104b9a31", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Changes in methods, as well as in content, can reduce over-time comparability of\n\n\nindicators. Admirably, WEF has tried to increase the response rate of its Executive\n\n\nOpinion Survey, to enhance accuracy by making the sample more representative. <sup>12</sup>\n\n\nHowever, progress on this front can affect apparent trends. Suppose executives with the\n\n\nstrongest opinions are the most likely to respond, and that strong opinions tend to be\n\n\nunfavorable. An increase in the response rate from one year to the next would then\n\n\nreduce the negative bias, but the year-on-year _change_ would be biased toward showing\n\n\nimprovement.\n\n\n**3. Composite Corruption Indexes**\n\n\nThere are at least three possible justifications for constructing a single corruption\n\n\nindex from multiple, distinct sources of corruption indicators. The first motive\n\n\nemphasizes substantive content: individual indicators, or even several indicators from one\n\n\nsource such as the BEEPS, may be defined too narrowly for certain purposes. For\n\n\nexample, no matter how many corruption indicators one aggregates from the BEEPS, the\n\n\nresulting index still reflects only corrupt interactions between firms and public officials.\n\n\nThe second motive is to reduce measurement error. Given the obvious difficulties\n\n\nin measuring corruption, any one source may be highly inaccurate. However, if errors in\n\n\nmeasurement are largely independent across sources, the errors will tend to cancel out\n\n\nwhen data are aggregated from multiple sources.\n\n\nThe third motive is to cover a larger number of countries. No one source covers\n\n\nall countries. Some sources do not overlap at all in country coverage, for example the\n\n\nUNECA’s African Governance Indicators and Nations in Transit.\n\n\n12 The mean number of responses per country increased from 84 in 2004 to 94 in 2005.\n\n\n15\n\nMention: \"African Governance Indicators and Nations in Transit\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Their country coverage does not overlap at all, limiting comparability.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003186:15:0", "occurrence_id": "1470f8870b8e707e962e7c00", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: and over) and for the group of all adults. In general the labor force participation of women of all\n\n\nages is quite low (for no age group do overall labor force participation rates exceed 18%). In\n\n\naddition, the majority of women who do work tend to participate in unsalaried activities: this is\n\n\nparticularly true of women over the age of 35. It is interesting to note the decreasing relative\n\n\nparticipation in salaried work versus unsalaried activities with age of women. Men, on the other\n\n\nhand, show a very high labor force participation rate, over 90% for men between the ages of 24\n\n\nand 55. The majority of men are salaried workers, although the percentage in salaried work\n\n\ntends to decrease with age. For male workers over the age of 55, almost half participate in non\n\nsalaried activities.\n\n\nIn particular, our sample includes 183,646 individual observations (89,207 observations\n\n\non adult males and 94,439 observations on adult females) for 4 rounds of the PROGRESA\n\n\nevaluation sample (Nov 97, Oct 98, Jun 99 and Nov 99) that are divided in treatment and control\n\n\nlocalities. The analysis is first carried on the sample of those who are classified as “poor” <sup>11</sup> [^11: This is the poor status after the _densifcacion_, the revision of the eligibility that raised the number of\nhouseholds eligible for the program from 52% to 78%. It has to be noticed that the fraction of households\nthat actually ended up receiving the PROGRESA cash transfers during the two-year interval covered by\nthe evaluation sample is just under 65%, due to administrative errors and delays in the final registration\nof beneficiary households.], and\n\n\nthus eligible to receive the program if they live in treatment localities (E=1). We then re-do the\n\n\nsame estimation on the sample of those who are classified as non-eligible (E=0).\n\n\nThe vector **_X_** in equation (1) consists of individual and demographic composition\n\n\nvariables\n\nMention: \"PROGRESA\n\n\nevaluation sample\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Nearly two hundred thousand adult observations across four rounds are analysed by treatment status.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003188:53:2", "occurrence_id": "872985bc6cac3f3ea2957775", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  3.40 N/a\nPakistan (E) 2.1 4 4 0.58 17.8 3.75 8.27\nS. Africa (E) 4.5 4 4 0.83 5.6 2.68 11.95\nArgentina (F) 2.8 4 1 0.50 N/a 5.44 10.31\nBrazil (F) 3.7 3 2 0.25 N/a 3.44 11.43\n<u>Mexico (F)</u> <u>3.5</u> <u>0</u> <u>0</u> <u>0.58</u> <u>N/a</u> <u>4.76</u> <u>10.79</u>\n\nFor definition of each of the variables, see Appendix A.1.\nData sources: Corruption Perception Index, Transparency International (2005); Anti-director rights (also see Table\n2-C), LLSV (1998) and DLLS 2005; Creditor rights (also see Table 2-B), LLSV (1998) and DMS (2005);\nDisclosure Requirement, LLS (2006); Earnings Management Score, Leuz, Nanda and Wysocki (2003); Legal\nFormalism Index, DLLS (2003); Legality Index, Berkowitz, Pistor and Richard (2003).\n\n\n53\n\nMention: \"Corruption Perception Index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Its score enters the regression as the corruption control.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003188:53:2", "occurrence_id": "8ff42acbd476f63203de6417", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  3.40 N/a\nPakistan (E) 2.1 4 4 0.58 17.8 3.75 8.27\nS. Africa (E) 4.5 4 4 0.83 5.6 2.68 11.95\nArgentina (F) 2.8 4 1 0.50 N/a 5.44 10.31\nBrazil (F) 3.7 3 2 0.25 N/a 3.44 11.43\n<u>Mexico (F)</u> <u>3.5</u> <u>0</u> <u>0</u> <u>0.58</u> <u>N/a</u> <u>4.76</u> <u>10.79</u>\n\nFor definition of each of the variables, see Appendix A.1.\nData sources: Corruption Perception Index, Transparency International (2005); Anti-director rights (also see Table\n2-C), LLSV (1998) and DLLS 2005; Creditor rights (also see Table 2-B), LLSV (1998) and DMS (2005);\nDisclosure Requirement, LLS (2006); Earnings Management Score, Leuz, Nanda and Wysocki (2003); Legal\nFormalism Index, DLLS (2003); Legality Index, Berkowitz, Pistor and Richard (2003).\n\n\n53\n\nMention: \"DLLS 2005\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Its updated anti-director rights measure is used alongside the earlier version.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003188:53:2", "occurrence_id": "d1050f4cc828cce4f5e1c0b7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  3.40 N/a\nPakistan (E) 2.1 4 4 0.58 17.8 3.75 8.27\nS. Africa (E) 4.5 4 4 0.83 5.6 2.68 11.95\nArgentina (F) 2.8 4 1 0.50 N/a 5.44 10.31\nBrazil (F) 3.7 3 2 0.25 N/a 3.44 11.43\n<u>Mexico (F)</u> <u>3.5</u> <u>0</u> <u>0</u> <u>0.58</u> <u>N/a</u> <u>4.76</u> <u>10.79</u>\n\nFor definition of each of the variables, see Appendix A.1.\nData sources: Corruption Perception Index, Transparency International (2005); Anti-director rights (also see Table\n2-C), LLSV (1998) and DLLS 2005; Creditor rights (also see Table 2-B), LLSV (1998) and DMS (2005);\nDisclosure Requirement, LLS (2006); Earnings Management Score, Leuz, Nanda and Wysocki (2003); Legal\nFormalism Index, DLLS (2003); Legality Index, Berkowitz, Pistor and Richard (2003).\n\n\n53\n\nMention: \"Legal\nFormalism Index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Its index of legal procedure formalism enters as a regressor.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003188:53:2", "occurrence_id": "90e3726e73412118b4014a07", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  3.40 N/a\nPakistan (E) 2.1 4 4 0.58 17.8 3.75 8.27\nS. Africa (E) 4.5 4 4 0.83 5.6 2.68 11.95\nArgentina (F) 2.8 4 1 0.50 N/a 5.44 10.31\nBrazil (F) 3.7 3 2 0.25 N/a 3.44 11.43\n<u>Mexico (F)</u> <u>3.5</u> <u>0</u> <u>0</u> <u>0.58</u> <u>N/a</u> <u>4.76</u> <u>10.79</u>\n\nFor definition of each of the variables, see Appendix A.1.\nData sources: Corruption Perception Index, Transparency International (2005); Anti-director rights (also see Table\n2-C), LLSV (1998) and DLLS 2005; Creditor rights (also see Table 2-B), LLSV (1998) and DMS (2005);\nDisclosure Requirement, LLS (2006); Earnings Management Score, Leuz, Nanda and Wysocki (2003); Legal\nFormalism Index, DLLS (2003); Legality Index, Berkowitz, Pistor and Richard (2003).\n\n\n53\n\nMention: \"Legality Index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Its measure of legal rule enforcement enters as a regressor.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003198:31:2", "occurrence_id": "189f3d651a928f873f3421de", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  or developed with equipment and machinery\nsuppliers.\nTechnology Acquired from \nInstitutions\nDummy variable equal to 1 if one of the most important ways in which the firm acquires technological\ninnovations is either from universities and public institutions, from business or industry associations, or\nfrom trade fairs or study groups.\nProduct Innovation\nDummy variable equal to 1 if the firm introduced new product line(s) or ugraded existing product line(s)\nin the three years prior to the survey.\nValue Added per Employee \n(log)\nFirm's nominal sales minus materials costs divided by total number of workers in the firm converted from\nlocal currency units to 2000 USD using the country's GDP deflator (Source: World Development\nIndicators) and the exchange rate of the local currency relative to the US dollar in 2000 (Source:\nInternational Financial Statistics, IMF). \nGDP per Capita (log)\nValues in 1995 in constant 2000 USD  (Source: World Development Indicators)\nNote: The source is the Investment Climate Surveys unless otherwise stated\n\nMention: \"World Development\nIndicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Its GDP deflator converts local wage figures into constant dollars.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003198:31:2", "occurrence_id": "5948275a9ae12d224d4f7f2c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  or developed with equipment and machinery\nsuppliers.\nTechnology Acquired from \nInstitutions\nDummy variable equal to 1 if one of the most important ways in which the firm acquires technological\ninnovations is either from universities and public institutions, from business or industry associations, or\nfrom trade fairs or study groups.\nProduct Innovation\nDummy variable equal to 1 if the firm introduced new product line(s) or ugraded existing product line(s)\nin the three years prior to the survey.\nValue Added per Employee \n(log)\nFirm's nominal sales minus materials costs divided by total number of workers in the firm converted from\nlocal currency units to 2000 USD using the country's GDP deflator (Source: World Development\nIndicators) and the exchange rate of the local currency relative to the US dollar in 2000 (Source:\nInternational Financial Statistics, IMF). \nGDP per Capita (log)\nValues in 1995 in constant 2000 USD  (Source: World Development Indicators)\nNote: The source is the Investment Climate Surveys unless otherwise stated\n\nMention: \"International Financial Statistics\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Its exchange rates convert local currency into dollars.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003198:31:2", "occurrence_id": "2424021f925ddbfa866618e4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  or developed with equipment and machinery\nsuppliers.\nTechnology Acquired from \nInstitutions\nDummy variable equal to 1 if one of the most important ways in which the firm acquires technological\ninnovations is either from universities and public institutions, from business or industry associations, or\nfrom trade fairs or study groups.\nProduct Innovation\nDummy variable equal to 1 if the firm introduced new product line(s) or ugraded existing product line(s)\nin the three years prior to the survey.\nValue Added per Employee \n(log)\nFirm's nominal sales minus materials costs divided by total number of workers in the firm converted from\nlocal currency units to 2000 USD using the country's GDP deflator (Source: World Development\nIndicators) and the exchange rate of the local currency relative to the US dollar in 2000 (Source:\nInternational Financial Statistics, IMF). \nGDP per Capita (log)\nValues in 1995 in constant 2000 USD  (Source: World Development Indicators)\nNote: The source is the Investment Climate Surveys unless otherwise stated\n\nMention: \"Investment Climate Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Unless otherwise stated, the firm variables come from these surveys.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003208:10:0", "occurrence_id": "37dca0fd56c1e85983a7d87b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: and a systematic sampling rule used to select the households to be interviewed. We\n\n\nadministered a screening questionnaire to determine whether the household contained a\n\n\nperson between the ages of 18 and 65 who was employed and commuted regularly to his\n\n\nor her place of work. We also required respondents to have at least an 8 <sup>th</sup> grade\n\n\neducation, in order to understand the risk information provided in the survey, and to have\n\n\nresided in Delhi for at least three months. The questionnaire was administered to 1,200\n\n\nrespondents during October – December of 2005, following two pretests involving 601\n\n\nhouseholds.\n\n\n**IV. Sample Characteristics and Responses**\n\n\n**A. Individual Characteristics of the Respondents**\n\n\nDescriptive statistics of the sample respondents are displayed in table 2. The top\n\n\npanel of table 2 reports information about demographic and socioeconomic\n\n\ncharacteristics of the sample. Briefly, we note that the average age of the respondent is 35\n\n\nand that 95% of the respondents are male. The high proportion of males reflects the fact\n\n\nthat only 15% of women in Delhi work outside the home (National Sample Survey,\n\n\n2005). About 48% of respondents have a high school diploma or vocational degree, and\n\n\n28% a bachelor’s degree or better. Mean household income is Rs. 135,000 a year, or\n\n\napproximately $15,350 in purchasing power parity terms. Mean earnings (personal\n\n\nincome) are approximately $10,250 in PPP terms.\n\n\nAs shown in the table, we place our respondents in three income groups,\n\n\ndepending on personal income. Individuals earning less than Rs. 8,000 a month constitute\n\n\nour low-income group, which accounts for 44% of the sample. Individuals with monthly\n\n\n8 Pretests of the questionnaire revealed that changes in the bids on the payment card did not have a\n\n\n                        - 10\n\nMention: \"National Sample Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It reports that only fifteen percent of women in the city work outside the home.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003210:27:0", "occurrence_id": "3575f6cb5dc841f2c6ade5e1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **6.** **Conclusions**\n\n\nThe recent literature on decentralization of anti-poverty programs in developing\n\ncountries has highlighted the need for quantitative empirical analysis of specific\n\nprograms, with particular attention to any evidence relating to whether programs might\n\nbe diverted from their original mandate, as a result of the influence of local elites. In this\n\npaper, we have analyzed a unique combination of data sets from the Ecuadorian Social\n\nInvestment Fund (FISE), during 1993-1996. The combination of a detailed, parish-level\n\npoverty map, full program administrative data on project implementation, and province\nlevel electoral results has allowed us to investigate the relationship between local income\n\ndistribution and project choice in an unprecedented way.\n\nTo help our understanding of the role played by elites in community-level\n\ndecision-making, we presented a simple model of project choice between local public\n\ngoods and excludable goods that are particularly demanded by the poor. The model was\n\ndesigned specifically for this case study, where a key distinction among a menu of\n\nprojects - all of which may appear to be pro-poor - was between projects that would\n\nbenefit the entire community (through increased provision of local public goods, such as\n\nschooling, health clinics or roads), and those that provided excludable private goods that\n\nthe poor were in great need of, but which brought no direct benefit to the elite. <sup>37</sup> [^37: We acknowledge that building latrines for the poor may have positive health externalities for the nonpoor. Whether the elite take this effect into consideration during the decision-making process is uncertain.] The\n\nmain examples of such private goods were latrines, and latrine construction projects\n\naccounted for 13% of all FISE projects.\n\nUnder the assumption that influence over local decision-making processes is a\n\nnon-decreasing function of income, the model generates two basic predictions: other\n\nthings equal, latrine projects would be more frequently found in poorer villages, and in\n\nless unequal villages. Inequality would reduce the probability that a community received\n\nlatrine projects, controlling for poverty and need, through an “elite\n\nMention: \"full program administrative data on project implementation\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Project-level implementation records identify where investments were made.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003210:27:0", "occurrence_id": "3b10735786887c701470a1d4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **6.** **Conclusions**\n\n\nThe recent literature on decentralization of anti-poverty programs in developing\n\ncountries has highlighted the need for quantitative empirical analysis of specific\n\nprograms, with particular attention to any evidence relating to whether programs might\n\nbe diverted from their original mandate, as a result of the influence of local elites. In this\n\npaper, we have analyzed a unique combination of data sets from the Ecuadorian Social\n\nInvestment Fund (FISE), during 1993-1996. The combination of a detailed, parish-level\n\npoverty map, full program administrative data on project implementation, and province\nlevel electoral results has allowed us to investigate the relationship between local income\n\ndistribution and project choice in an unprecedented way.\n\nTo help our understanding of the role played by elites in community-level\n\ndecision-making, we presented a simple model of project choice between local public\n\ngoods and excludable goods that are particularly demanded by the poor. The model was\n\ndesigned specifically for this case study, where a key distinction among a menu of\n\nprojects - all of which may appear to be pro-poor - was between projects that would\n\nbenefit the entire community (through increased provision of local public goods, such as\n\nschooling, health clinics or roads), and those that provided excludable private goods that\n\nthe poor were in great need of, but which brought no direct benefit to the elite. <sup>37</sup> [^37: We acknowledge that building latrines for the poor may have positive health externalities for the nonpoor. Whether the elite take this effect into consideration during the decision-making process is uncertain.] The\n\nmain examples of such private goods were latrines, and latrine construction projects\n\naccounted for 13% of all FISE projects.\n\nUnder the assumption that influence over local decision-making processes is a\n\nnon-decreasing function of income, the model generates two basic predictions: other\n\nthings equal, latrine projects would be more frequently found in poorer villages, and in\n\nless unequal villages. Inequality would reduce the probability that a community received\n\nlatrine projects, controlling for poverty and need, through an “elite\n\nMention: \"province\nlevel electoral results\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"They allow the link between local income distribution and project choice to be examined.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003214:32:2", "occurrence_id": "96c679fb693784e1145bb95a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <br>0.20<br>0.71|\n\n\n\nData:VLSS93 and VLSS98\nRobust standard errors in brackets, clustered at the year-of-birth level\n\n- significant at 10%; ** significant at 5%; *** significant at 1%\nHousehold controls include logarithm of per capita real consumption; family controls consist of family size, parents' education, height, and weight,\nparents' age at time of first birth\nCommune-level controls include commune dummies and logarithm of cohort size\nAn auspicious year is a year that is auspicious for either boys or girls\nA child is born lucky if he is a boy born in year auspicious for boys, or she is a girl born in a year auspicious for girls\n\n\n33\n\nMention: \"VLSS93\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Estimates with standard errors clustered by year of birth are computed on this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003217:14:0", "occurrence_id": "3bc5757e547c830dc8f99889", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 14\n\n\n\nFigure 12\nMonetary Impact of Treasury’s\n\nOperations 1999- 2003\n(ratio to the monetary base\n\n\n1999 2000 2001 2002 2003\n\n\n\n5.0\n\n\n4.5\n\n\n4.0\n\n\n3.5\n\n\n3.0\n\n\n2.5\n\n\n2.0\n\n\n\nFigure 11\nMoney Base as a Share of GDP\n\n(seasonally adjusted data\n\n\n95 96 97 98 99 00 01 02 03\n\n\n\n.4\n\n\n.3\n\n\n.2\n\n\n.1\n\n\n.0\n\n\n-.1\n\n\n-.2\n\n\n-.3\n\n\n\nSource: World Bank staff calculations based on\nBacen data\n\n\n\nSource: World Bank calculations based on Bacen\ndata\n\n\n\nMoreover, base money grew in lockstep with the faster depreciation of the\ncurrency. This pressured inflation, which accelerated between June and December,\nreaching a peak of 5.8 percent per month in November (Figure 13). Monetary growth\nand rising inflation increased the government’s revenue from money creation up to the\nequivalent of 2.0 percent of GDP <sup>5</sup> [^5: The figures reported in the text and in the graph are estimated by multiplying the base money as a share of\nGDP times the growth rate on base money. Eliana Cardoso (1998) estimates the average inflation tax\nrevenue in Brazil during the 50 years ending 1995 at 2 percent of GDP.] (Figure14). In these circumstances, the authorities’\ncredibility faltered. It is interesting to note that the seignorage peak occurred in the first\nquarter of 2003, a few months after the public debt ratio had stabilized and the exchange\nrate had appreciated. This implies that interest rates could not be lowered as quickly as\nmany would have desired.\n\n\n\nFigure 13\nMonthly Inflation Rate in Brazil 2000-2003\n\n(seasonally adjusted General Price Level\n\nIGP-DI)\n\n\n6\n\n\n5\n\n\n4\n\n\n3\n\n\n2\n\n\n1\n\n\n0\n\n\n\n2.5\n\n\n2.0\n\n\n1.5\n\n\n1.0\n\n\n0.5\n\n\n0.0\n\n\n-0.5\n\n\n\nFigure 14\nSeignorage from Money Creation\n\n2000-2003\n(% of GDP)\n\n\n2000 2001 2002 2003\n\n\n\n-1\n\n\n\n2001 2002 2003\n\n\n\nSource: FGV Source: World Bank calculations described in text.\n\nMention: \"Bacen data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Monetary aggregates are plotted from the central bank's series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003222:16:0", "occurrence_id": "52d06fe019f954d5fc1db75c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |Table 10. There is little|e evidence of exclusion of poor households|Col3|Col4|\n|---|---|---|---|\n||**Gini Coefficients**|**Gini Coefficients**|**Gini Coefficients**|\n||**2002**|**2003**|**2004**|\n|Azerbaijan|36.6|24.8|26.8|\n|Baku|36.3|27.9|27.5|\n|Non-Baku urban|38.0|25.5|28.4|\n|Rural|35.8|22.2|24.9|\n\n\nSource: Azerbaijan HIES data.\n\n**_Hypothesis 5_**\n\nUnder the assumptions for hypothesis 5, the Gini coefficient for Azerbaijan increase more\nappreciably than under any other hypotheses. Here we made the assumption that the extremely\nlow inequality in Azerbaijan HIES data arose from lack of and/or inadequate participation of rich\nhouseholds and significant amount of transfers among Azeri households. Table 11 presents the\nGini coefficients after removing transfers and replacing the consumption per capita for missing\nhouseholds following the Pareto distribution assumption under hypotheses 2 and 3. After\naccounting for upper truncation and transfers, the Gini coefficient more than double in most\ncases (see Table 12). After removing all transfer incomes and accounting for refusal of rich\nhouseholds, the Gini coefficient in 2002 increases to over 200 percent for Azerbaijan as the\nwhole. Inequalities in 2003 and 2004 similarly climb by about 145 and 194 percent respectively.\n\nThe overall findings here suggest that the main culprit for extremely low inequality in the\nAzerbaijan HIES data is not only refusal of rich households to participate in the survey, but also\nsignificant infusion of transfers to the Azeri household consumption. Transfers from both\ngovernment and private sources are important sources of income. Furthermore the\n\nMention: \"Azerbaijan HIES data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Household income and expenditure series feed the regional inequality projections.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003228:63:3", "occurrence_id": "de8a12e56e0c21e565068ef2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  0.002 0.035 0.003 0.005\n**_<mark>Sibiria</mark>_** 0.000 0.001 0.069 0.000 0.010 0.000 0.005 0.000 0.021 0.141 0.000 0.013 0.000 0.013\n**_<u><mark>Far East</mark></u>_** <u>0.000</u> <u>0.121</u> <u>0.077</u> <u>0.018</u> <u>0.006</u> <u>0.001</u> <u>0.008</u> <u>0.000</u> <u>0.096</u> <u>0.046</u> <u>0.004</u> <u>0.012</u> <u>0.000</u> <u>0.010</u>\n\n<mark>*</mark> <mark>COM (communication and transport) includes RLW TRK PIP MAR AIR TRO TMS PST</mark>\n<mark>**</mark> <mark>MIN(mining) includes FME NFM COA OLE</mark>\n<mark>***</mark> <mark>MNF (manufacturing) includes MWO TPP CNM CLO OTHT OIN</mark>\n\n\n**2.2** **Regions of Russia 2003**\n\n\nTable 3.5 of “Regions of Russia 2003�� reports the distribution of labor force by regions\n\n\nand types of ownership in 2002. The types of ownership include state, city, private, NGO, mixed\n\nRussian <sup\n\nMention: \"Regions of Russia 2003\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its table on labour force distribution by region and occupation supplies the inputs.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003233:13:0", "occurrence_id": "9ec6edf6aa2e4b1217e205f6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: spread during the earlier outbreaks in DRC and West Africa. Control measures cur‐\nrently in use aim at reducing the spread of the disease through burning infected coffee\ntrees on site, delaying replanting at least by one to two seasons, and restricting move‐\nment of infected material and soil.\n\nThe wilt disease, which was first confirmed in Uganda in 1993, has been moni‐\ntored through field reports and numerous scientific surveys. The disease was the sub‐\nject of a workshop (hosted by the Government of Uganda in July 1997), attended by 60\nparticipants from most East and Central Africa coffee producing countries (Sewaya\n1999, p. 37). The workshop’s resolutions were: (i) immediate action should be taken to\ncontain the disease; (ii) a project proposal should be presented to the ICO and EU for\nfunding; and (iii) Uganda and DRC should take the lead in the investigations. In terms\nof research, however, it appears that no immediate progress was made. A few years\nlater, for example, Lutakome (2001, p. 66) noted that “Research work will be intensified\nto develop a resistant coffee cultivar. Epidemiological study results on the disease will\nalso be made available to farmers.”\n\nThe latest UCDA survey confirmed that all robusta‐producing districts have\nbeen affected by the disease. In some cases entire coffee fields have been destroyed. Al‐\nthough precise estimates on the impact of the disease are not available, one can gauge\nthe effects of the disease based on the 2003 review study which estimated that of the to‐\ntal of 240,000 hectares in all 21 robusta‐growing (traditional) districts, 122,400 hectares\nhave been infected (equivalent to about 136 million robusta trees), representing a 51\npercent cumulative infection rate (UCDA, _2003/04 Annual Report_, p. 29). There is consid‐\nerable variation among districts\n\nMention: \"UCDA survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It confirmed the disease had reached all robusta-producing districts.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003235:59:1", "occurrence_id": "0ad177ff91602bffbf27b397", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ||\n\n\n\n1950 1970 1990 2010 2030 1950 1970 1990 2010 2030 1950 1970 1990 2010 2030\n\n|Kenya|Col2|\n|---|---|\n|||\n|||\n|||\n|||\n\n\n|Laos|Col2|\n|---|---|\n|||\n|||\n|||\n|||\n|||\n|||\n\n\n|Liberia|Col2|\n|---|---|\n|||\n|||\n|||\n|||\n\n\n\n1950 1970 1990 2010 2030 1950 1970 1990 2010 2030 1950 1970 1990 2010 2030\n\n|Madagascar|Col2|\n|---|---|\n|||\n|||\n|||\n|||\n|||\n|||\n\n\n|Malawi|Col2|\n|---|---|\n|||\n|||\n|||\n|||\n\n\n|Maldives|Col2|\n|---|---|\n|||\n|||\n|||\n|||\n|||\n|||\n\n\n\n1950 1970 1990 2010 2030 1950 1970 1990 2010 2030 1950 1970 1990 2010 2030\n\n|Mali|Col2|Col3|\n|---|---|---|\n||||\n||||\n||||\n||||\n||||\n||||\n\n\n|Mauritania|Col2|\n|---|---|\n|||\n|||\n|||\n|||\n|||\n\n\n|Mozambique|Col2|\n|---|---|\n|||\n|||\n|||\n|||\n|||\n|||\n\n\n\n1950 1970 1990 2010 2030 1950 1970 1990 2010 2030 1950 1970 1990 2010 2030\n#### Year\n###### United Nations Medium Variant Projections\n\n\n60\n\nMention: \"United Nations Medium Variant Projections\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The population paths shown are this projection variant.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003241:20:1", "occurrence_id": "14ab591033d1eca02a52a27a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: hat consumption as a share of total expenditure (%)|Col2|Col3|Col4|\n|---|---|---|---|\n|**Income Categories**<br>**All**<br>**Male Head**<br>**Female Head** <br>**Households**|**Income Categories**<br>**All**<br>**Male Head**<br>**Female Head** <br>**Households**|**Income Categories**<br>**All**<br>**Male Head**<br>**Female Head** <br>**Households**|**Income Categories**<br>**All**<br>**Male Head**<br>**Female Head** <br>**Households**|\n|All<br>poorest <br>2nd quartile <br>3rd quartile <br>richest|9.5<br>3.5<br>7.5<br>12.2<br>14.7|10.9<br>4.4<br>8.9<br>13.3<br>15.9|4.7 <br>1.1 <br>3.5 <br>8.2 <br>8.1|\n\n\n\n_Source: Authors' calculation based on EDAM 1996_\n\n\n**_5.5. Impact on income distribution_**\n\n**_Income inequality as measured by the Gini index rose slightly across all three scenarios which_**\n**_tentatively show that the devaluation exercise will not exacerbate inequality._** Again this appears\n\nMention: \"EDAM 1996\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Income distribution and Gini changes are calculated from this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003245:3:0", "occurrence_id": "16e760818b281f6664d1094a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: purchases across other sectors. <sup>2</sup> The basic insight here is that institutions lower the\n\n\ntransaction cost and hence benefit those sectors more which enter into a more complex\n\n\nweb of transactions with the rest of the economy. The measure suggested by Blanchard\n\n\nand Kremer is acknowledged by the authors to be a “noisy” one and has a number of\n\n\nlimitations. First, the measure does not include intra-sector transactions between firms\n\n\nlargely due to data limitations. Second, it does not take into account inputs supplied by a\n\n\nsector although there is no reason to believe that high transaction costs do not affect the\n\n\nsuppliers as well.\n\n\nNevertheless, an examination of the input-output matrices for a number of\n\n\ncountries reveals that in general input supplies and purchases were much more evenly\n\n\ndistributed for the services sectors relative to manufacturing. In the spirit of Blanchard\n\n\nand Kremer, it can be argued that the services sectors are more institutionally dependent\n\n\nthan the other sectors. As an example, institutional dependence of various sectors\n\n\ncalculated using India's input-output matrix is shown in Figure 1 where higher values\n\n\nindicate greater dependence. <sup>3</sup>\n\n\n2 Blanchard and Kremer (1997) look at the decline in output of manufacturing sectors in the former Soviet\nUnion countries between 1992 and 1994 and argue that the decline was more pronounced in the more\ninstitutionally dependent sectors.\n3 The graph uses GTAP data for India. To construct the graph we first computed the Gini coefficient for the\ninputs supplied to and bought from a sector to all other sectors. We then calculated 100 minus Gini for each\nsector to arrive at the figures shown in the graph. This is slightly different from the approach of Blanchard\nand Kremer who use the inverse of the Herfindahl index instead. Both these measures are widely used in\nthe literature.\n\n\n3\n\nMention: \"GTAP data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Sectoral input linkages for India are used to compute the concentration coefficient.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003249:37:0", "occurrence_id": "0d6fd5e13940974f9c0e2ba9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Table 4: Unconditional means of the three education outcomes examined**\n\n\n\nTreatment variable=T\nReceived PEC benefits in ALL 3 school\n\n<u>years</u>\n\n\n\nTreatment variable=T2\nReceived PEC benefits in ANY school year\n\n\n\nOutcome and Group 2000 2003 2003-2000 2000 2003 2003-2000\n\n\n**_Dropout_**\n\n\nPEC 4.64 4.23 -0.41 4.54 4.24 -0.31\n\n\nNon-PEC 4.04 3.82 -0.21 4.04 3.82 -0.21\n\n\nDifference: PEC-Non-PEC 0.60 0.40 -0.20 0.51 0.41 -0.09\n\n\n**_Failure_**\n\n\nPEC 5.18 4.31 -0.87 5.52 4.54 -0.98\n\n\nNon-PEC 7.17 6.12 -1.05 7.17 6.12 -1.05\n\n\nDifference: PEC-Non-PEC -1.99 -1.81 0.18 -1.65 -1.59 0.07\n\n\n**_Repetition_**\n\n\nPEC 4.99 3.93 -1.06 5.20 4.18 -1.02\n\n\nNon-PEC 6.67 5.68 -0.99 6.67 5.68 -0.99\n\n\nDifference: PEC-Non-PEC -1.68 -1.75 -0.08 -1.46 -1.50 -0.03\n\n\nSources: School Census 2000 and 2003.\nStandard errors in parentheses. Mexican primary schools,\nTreatment variable T: 1,767 PEC schools and 65,457 non-PEC schools\nTreatment variable T2: 9,244 PEC schools and 65,457 non-PEC schools\n\n\n37\n\nMention: \"School Census 2000 and 2003\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Primary school outcomes for treated and comparison schools are drawn from these enumerations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003251:11:1", "occurrence_id": "854e2896849abb2d03c8b2bb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \neffects of individual agreements, it is not appropriate for making generalizations about the _average_\n\n\neffect of all PTAs in existence. We are not aware of any empirical works that consider all PTAs\n\n\nin force in a gravity framework, and the fact that existing studies usually limit their attention to\n\n\nlarger and more well-known PTAs could easily bias any aggregation of these estimates. Therefore,\n\n\nthe first part of this section will discuss a new comprehensive database of PTAs, which improves\n\n\nupon the existing data sources by including a large number of agreements not notified to the WTO.\n\n\nRelying on this database will allow our model to track the average effects of PTAs on trade directly,\n\n\n14The sample was obtained by collecting individual agreement dummy coefficient estimates from each study. The\nrequirements for choosing a study were either a published article or an NBER working paper.\n\n\n9\n\nMention: \"comprehensive database of PTAs\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"This new compilation adds agreements never notified to the GATT or WTO.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003258:8:1", "occurrence_id": "fe9c80de72e746db1e21e72e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  the wide range of different countries featured in the conference,\ncertain similar patterns emerge. For all of these countries excepting Brazil, PPI for urban\ninfrastructure has not emerged as a major contributor to investment funding, certainly not\nsubstituting in a significant fashion, for government funds. Its contribution in Brazil, in\npercentage terms appears to be the most significant, where urban PPI accounts for about\n3% of total infrastructure investment spending and doubtless a larger share of urban\ninvestment spending. While in Brazil and the rest of Latin America, policymakers in the\n1990s hoped that increased PPI could fill the spending gap created by fiscal discipline,\nspending declined overall as a share in GDP. Indian cities both have very limited budgets\nand very pressing needs for infrastructure improvements, but urban PPI has been nearly\ninsignificant there. While Brazil and India faced harsh fiscal constraints, China’s\ngovernment spending was expanding rapidly, as was the economy. Apparently, however,\nthe contribution of PPI to urban infrastructure remains limited, less than 5%. Moreover, it\nappears that the PPI on urban toll roads was short-lived and disappeared when\ngovernment guarantees were withdrawn. Poland and South Africa’s local governments\nappear well resourced and many are quite well-managed, but PPI has also made\n\n\n9 Dollar amounts from the PPI database were converted at the annual average exchange rate in the\nyear of the transaction closure. Since transactions started in 1996, we counted an eight year period. The\nfigures from the Mohan report are in real terms and the data in the PPI database are in nominal terms. Since\nall the nominal figures are from years after the base year (1994-1995) for the investment requirements, this\nlow percentage share is biased somewhat upward.\n\n\n9\n\nMention: \"the PPI database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Transaction values are converted at annual average rates using this source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003260:23:1", "occurrence_id": "f5b18356822959656f7ce3ee", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: br>Proportion of population over 15 years with completed<br>high education with respect to population with less<br>than 4 years of schooling in 1987.<br>Barro and Lee (2000)<br>Production of coal<br>World Energy Council (2004)<br>Production of oil<br>World Energy Council (2004)<br>Average maximum lead content in gasoline<br>Octels world wide gasoline survey, prepaired base available on<br>http://www2.unine.ch/Jahia/site/irene/op/edit/pid/12149?matrix=101 <br>Capital to labor ratio<br>Sandeep Mahajan (PRMEP), World Bank 2001<br>Human development indicator for 1987<br>United Nations Database|\n\n\n```\n24\n\n```\n\nMention: \"Octels world wide gasoline survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Average maximum lead content in gasoline is taken from this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003260:23:1", "occurrence_id": "84a7cfda75019db28cbe190e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: br>Proportion of population over 15 years with completed<br>high education with respect to population with less<br>than 4 years of schooling in 1987.<br>Barro and Lee (2000)<br>Production of coal<br>World Energy Council (2004)<br>Production of oil<br>World Energy Council (2004)<br>Average maximum lead content in gasoline<br>Octels world wide gasoline survey, prepaired base available on<br>http://www2.unine.ch/Jahia/site/irene/op/edit/pid/12149?matrix=101 <br>Capital to labor ratio<br>Sandeep Mahajan (PRMEP), World Bank 2001<br>Human development indicator for 1987<br>United Nations Database|\n\n\n```\n24\n\n```\n\nMention: \"Human development indicator for 1987\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Its 1987 value enters the regression as a development control.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003260:23:1", "occurrence_id": "268b2466dd93bf155c0aa20f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: br>Proportion of population over 15 years with completed<br>high education with respect to population with less<br>than 4 years of schooling in 1987.<br>Barro and Lee (2000)<br>Production of coal<br>World Energy Council (2004)<br>Production of oil<br>World Energy Council (2004)<br>Average maximum lead content in gasoline<br>Octels world wide gasoline survey, prepaired base available on<br>http://www2.unine.ch/Jahia/site/irene/op/edit/pid/12149?matrix=101 <br>Capital to labor ratio<br>Sandeep Mahajan (PRMEP), World Bank 2001<br>Human development indicator for 1987<br>United Nations Database|\n\n\n```\n24\n\n```\n\nMention: \"United Nations Database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It is the source behind the development indicator.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003262:6:0", "occurrence_id": "7ff6fcfa27e82df6fa384cc3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: trade ban coupled with the binding export quota caused low prices of rice in the South\n\n\nand high prices in the North. The increase in export quota and the lifting of internal\n\n\nrestrictions has lead to increased rice prices (in real terms) everywhere but proportionally\n\n\nmore in the South (Benjamin & Brandt, 2004; Niimi et al, 2003; Seshan, 2005).\n\n\nThe unit prices in the household data also confirm this. The prices in Vietnam\n\n\nhave changed substantially between 1993 and 1998. The general inflation rate was 45.6%\n\n\n(which corresponds to a yearly inflation rate of 7.8%). The increase in nominal rice price\n\n\nhas been more than the inflation rate in all regions for the sales unit price and almost all\n\n\nregions for the production unit price (Table 1) pointing to an increase in real rice prices at\n\n\nthe household level.\n\n\n**Table 1. Change in Producer Rice Prices from 1993-1998, by Region.**\n\n\n**<u>Rice price</u>** **<u>Increase in Sales unit prices</u>** **<u>Increase in Production unit prices</u>**\nNorthern Uplands 57% 43%\nRed River Delta 62% 35%\nNorth Central Coast 72% 52%\nSouth Central Coast 98% 56%\nCentral Highlands 130% 86%\nSouth East 62% 54%\nMekong Delta 65% 69%\n<u>Total</u> <u>67%</u> <u>49%</u>\nSource: Author’s own calculations using VNLSS 93 and 98. The percentage change in the mean unit price\nof one kilogram of rice: from sales and production values and quantities respectively.\n\n### **_2.1 Data_**\n\n\nOur analysis is based on the panel data from Vietnam Living Standards Survey\n\n\n(VNLSS) for 1992/1993 and 1997/1998. This is a\n\nMention: \"VNLSS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Rice price changes are computed from unit values in these two rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003262:6:0", "occurrence_id": "0445d7189bfc08733f607147", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: trade ban coupled with the binding export quota caused low prices of rice in the South\n\n\nand high prices in the North. The increase in export quota and the lifting of internal\n\n\nrestrictions has lead to increased rice prices (in real terms) everywhere but proportionally\n\n\nmore in the South (Benjamin & Brandt, 2004; Niimi et al, 2003; Seshan, 2005).\n\n\nThe unit prices in the household data also confirm this. The prices in Vietnam\n\n\nhave changed substantially between 1993 and 1998. The general inflation rate was 45.6%\n\n\n(which corresponds to a yearly inflation rate of 7.8%). The increase in nominal rice price\n\n\nhas been more than the inflation rate in all regions for the sales unit price and almost all\n\n\nregions for the production unit price (Table 1) pointing to an increase in real rice prices at\n\n\nthe household level.\n\n\n**Table 1. Change in Producer Rice Prices from 1993-1998, by Region.**\n\n\n**<u>Rice price</u>** **<u>Increase in Sales unit prices</u>** **<u>Increase in Production unit prices</u>**\nNorthern Uplands 57% 43%\nRed River Delta 62% 35%\nNorth Central Coast 72% 52%\nSouth Central Coast 98% 56%\nCentral Highlands 130% 86%\nSouth East 62% 54%\nMekong Delta 65% 69%\n<u>Total</u> <u>67%</u> <u>49%</u>\nSource: Author’s own calculations using VNLSS 93 and 98. The percentage change in the mean unit price\nof one kilogram of rice: from sales and production values and quantities respectively.\n\n### **_2.1 Data_**\n\n\nOur analysis is based on the panel data from Vietnam Living Standards Survey\n\n\n(VNLSS) for 1992/1993 and 1997/1998. This is a\n\nMention: \"Vietnam Living Standards Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its two-round panel is the basis of the analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003264:34:0", "occurrence_id": "d418b5daeff72f6442edbfb1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Table 7: Urban Infrastructure Investment Ratios\n\n\n\n\n\n\n\n|Year|Ratio to Total Fixed<br>Asset Investment|Ratio to GDP|\n|---|---|---|\n|1980|1.2%|0.2%|\n|1985|2.5%|0.7%|\n|1990|2.7%|0.7%|\n|1995|4.0%|1.4%|\n|2000|5.8%|2.1%|\n\n\nSource: Construction Industry Publication of China as reported in Fu Tao (2004)\n\n\nGovernment experts have estimated that approximately 20 to 30 percent spent\n\n\non urban infrastructure construction during the 10 <sup>th</sup> Five Year Plan period, ending in\n\n\n2005, will be expended by local, provincial, and central governments directly from\n\n\nbudget (fiscal) resources. The remainder will be financed by a combination of\n\n\ndomestic loans, foreign and multilateral loans, revenue from asset earnings and asset\n\n\nsales, and various forms of market-based financing.\n\n### THE FISCAL FRAMEWORK AND INFRASTRUCTURE FINANCING OPTIONS\n\n\nPerhaps the best way to examine the linkage between the national fiscal\n\n\nframework and national fiscal policy, on the one hand, and urban infrastructure\n\n\n35\n\nMention: \"Construction Industry Publication of China\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Sectoral shares over two decades are reported in this publication.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003265:17:1", "occurrence_id": "d4aeecb254b584ffed7dfc43", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  <br>|\n|**_Pakistan_**|3 <br>|3 <br>|unspecified|3|3 <br>|\n|**_Philippines_**|3 <br>|3 <br>|unspecified|unspecified<br>|3 <br>|\n|**_Thailand_**|3 <br>|3 <br>|unspecified|3 <br>|3|\n|**_U.A.E._**|3|3|unspecified|3|unspecified|\n|∗  Dubai International Financial Centre<br>** Annex III details the legal bases and the provisions of these regulations<br>|∗  Dubai International Financial Centre<br>** Annex III details the legal bases and the provisions of these regulations<br>|∗  Dubai International Financial Centre<br>** Annex III details the legal bases and the provisions of these regulations<br>|∗  Dubai International Financial Centre<br>** Annex III details the legal bases and the provisions of these regulations<br>|∗  Dubai International Financial Centre<br>** Annex III details the legal bases and the provisions of these regulations<br>|∗  Dubai International Financial Centre<br>** Annex III details the legal bases and the provisions of these regulations<br>|\n|Source: Official country websites and central bank Annual Reports|Source: Official country websites and central bank Annual Reports|Source: Official country websites and central bank Annual Reports|Source: Official country websites and central bank Annual Reports|Source: Official country websites and central bank Annual Reports|Source: Official country websites and central bank Annual Reports|\n\n\nIn addition to _Shariah_ boards, most IIFS, particularly those complying with\n\n\nAAOIFI standards, have established another internal Shariah review structure, namely,\n\n_Shariah_ Review\n\nMention: \"Official country websites and central bank Annual Reports\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"Legal bases and provisions of the regulations were compiled from these official documents.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003274:11:0", "occurrence_id": "e8d363191ca47d6761c80bd5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **<u>Figure 3. Distribution of Bank Deposits in Zambia in March 2006</u>**\n\n\n\n\n\n\n\nSource: Data from commercial banks in Zambia\n\nSeveral studies indicate that the account holders are usually people living in urban areas\nand with a regular employment in the public sector or a large private firm. One additional\ncharacteristic is that the real value of small savings and deposit accounts declines over\ntime, as no interest is usually paid on small savings accounts denominated in Kwacha.\nOnly large firms receive positive interest rates on their deposits. In practice, for most\nhouseholds in Zambia bank savings and deposits are just a way to safeguard coins and\nnotes, not a practical instrument to save money over time. As illustrated in the following\nfigure, a 320,000 kwacha deposit (equivalent to $100 dollars) today would lose two thirds\nof its real value after 6 years if during this period the annual inflation rate remains at its\n2004 level (18%) and no nominal interest is paid, as is currently the case for small saving\ndeposits in Kwacha.\n\n\n**Figure 4. Real Value of a 320,000 Kwacha Deposit***\n\n\n\n\n\n*It assumes an annual inflation rate of 18% and no interest payments.\n\n\n\n12\n\nMention: \"Data from commercial banks in Zambia\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The distribution of bank deposits is compiled from these bank records.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003277:45:0", "occurrence_id": "e1b95f0cb5e95411d905453e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: infrastructure service _h_, the annual growth rate of real GDP per capita can be applied\nto the stock of public capital in _h_ at the base period. Elasticity values may vary,\ndepending on available estimates <sup>37</sup> . Actual stocks are used to estimate supply of\neach type of infrastructure services. Individual indicators of excess demand for each\ncomponent of infrastructure services are then calculated, by taking the ratio of supply\nto “predicted” demand. This ratio gives therefore an indicator of adequacy between\nsupply and demand; a fall in the ratio would indicate excessive pressure on existing\ninfrastructure and therefore a deterioration in quality. Second, a “composite” excess\ndemand indicator is calculated. To do so they use the same procedure used by\nHulten (1996) to calculate his quality index, that is, we standardized each of the three\nseries (by subtracting the mean and dividing by the standard error) and calculated\nthe unweighted, arithmetic average of the standardized series.\n\nMuch research has examined the issue of quality and congestion costs in\ninfrastructure, and their implications for private capital formation and the optimal\nallocation of public expenditure. But almost none has focused on congestion costs in\neducation. This is a particularly important factor in determining the quality of\nschooling in low-income countries, where (according to recent data from UNESCO\nand the World Bank) student-teacher ratios may dramatically exceed average ratios\nin industrial countries. For instance, at 44 to 1, the pupil-teacher ratio in sub-Saharan\nAfrica is on average three times higher than that of developed countries; moreover,\none in four countries in the region has ratios above 55 to 1 (see UNESCO (2005)).\n\nSimilarly, quality and congestion costs may be important in assessing the\neffect of health capital on growth. A recent press release by the World Health\nOrganization noted that hospitals in Sub-Saharan Africa are \"getting worse in terms\nof both the scope and quality of health care\n\nMention: \"UNESCO\nand the World Bank\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Their recent data show student-teacher ratios far above industrial-country averages.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003281:10:1", "occurrence_id": "ae5a2a0157d1f36ce91d650d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  in Malawi using 1997-98 household data\n\n\n(National Economic Council, National Statistical Office, Government of Malawi and\n\n\nInternational Food Policy Research Institute 2001). This model is also documented in Mukherjee\n\n\nand Benson (2003). This model is also similar to the model used in Chapter 2 of this volume to\n\n\nassess determinants of poverty. Our main extension is in the use of biomass as a determinant of\n\n\nwelfare.\n\n\nIn our reduced form model, the variable of interest on the right hand side is amount of biomass\n\n\nper hectare in the community. We postulate that household welfare and biomass stock have a\n\n\nnonlinear relationship. As biomass increases, welfare does not linearly increase but increases at a\n\n\n11\n\nMention: \"1997-98 household data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Poverty analysis in the country is conducted on these household observations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003283:29:0", "occurrence_id": "3d3dbf6751ddfcfae22ee890", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: primary source for these is the “Economic Freedom of the World (EFW)” database (see\nGwartney and Lawson [2004]). This database has been developed under the auspices of\nthe Fraser Institute in Canada with the aid of a worldwide network of economists and\nresearch institutes. In particular, we use indicators referring to hiring and firing practices,\nregulation of business activities and integrity of the legal system.\n\n\nDespite other indicators available in the literature for developing and emerging economies\n(e.g., the World Bank Doing Business database), the EFW tracks changes in regulations\nover time and is thus more suitable for our analysis of job flows that have indeed been\ninfluenced by policy changes over the period covered by our data (see Table 11 for details\non the regulatory variables).\n\n\n**Table** **11:** Institutional Variables, 1990s\n\n\n**<u>OVERALL</u>**\n<u>Variable</u> <u>Mean</u> <u>Std.</u> <u>Dev.</u> <u>Min</u> <u>Max</u>\nHiring and Firing Practices 5.261 1.515 2.878 7.700\nLaw&Order adj. Hiring and Firing Practices 4.113 2.019 0.000 7.209\nBusiness Regulations 3.490 1.389 1.100 5.900\nLaw&Order adj. Business Regulations 2.490 1.233 0.000 4.600\n<u>Law</u> <u>and</u> <u>Order</u> <u>2.280</u> <u>2.818</u> <u>0.000</u> <u>10.000</u>\n**<u>EU</u>** **<u\n\nMention: \"Economic Freedom of the World\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its regulatory measures are the primary source for the institutional variables.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003283:29:0", "occurrence_id": "eaaa8f152ce691c727dd42ae", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: primary source for these is the “Economic Freedom of the World (EFW)” database (see\nGwartney and Lawson [2004]). This database has been developed under the auspices of\nthe Fraser Institute in Canada with the aid of a worldwide network of economists and\nresearch institutes. In particular, we use indicators referring to hiring and firing practices,\nregulation of business activities and integrity of the legal system.\n\n\nDespite other indicators available in the literature for developing and emerging economies\n(e.g., the World Bank Doing Business database), the EFW tracks changes in regulations\nover time and is thus more suitable for our analysis of job flows that have indeed been\ninfluenced by policy changes over the period covered by our data (see Table 11 for details\non the regulatory variables).\n\n\n**Table** **11:** Institutional Variables, 1990s\n\n\n**<u>OVERALL</u>**\n<u>Variable</u> <u>Mean</u> <u>Std.</u> <u>Dev.</u> <u>Min</u> <u>Max</u>\nHiring and Firing Practices 5.261 1.515 2.878 7.700\nLaw&Order adj. Hiring and Firing Practices 4.113 2.019 0.000 7.209\nBusiness Regulations 3.490 1.389 1.100 5.900\nLaw&Order adj. Business Regulations 2.490 1.233 0.000 4.600\n<u>Law</u> <u>and</u> <u>Order</u> <u>2.280</u> <u>2.818</u> <u>0.000</u> <u>10.000</u>\n**<u>EU</u>** **<u\n\nMention: \"World Bank Doing Business database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"An alternative indicator set, less suitable here since it tracks regulation less over time.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003290:10:0", "occurrence_id": "721bd8ac408ee7144265d792", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: to unskilled workers. Under the assumption of wage equalization, the ratio of unskilled\n\n\nworkers in a sector to total unskilled workers in the country is calculated as the ratio of\n\n\npayments to unskilled workers in the sector to total payments to unskilled workers in the\n\n\neconomy. Regarding data for this calculation, only one observation per country or per\n\nsimilar countries is available from the original source (GTAP). <sup>6</sup> [^6: Given that, in most cases, the date of this observation differs significantly from the years of our poverty\nspell, we first use the GTAP data to compute the ratio of payments to unskilled workers to a sector’s value\nadded and assume it to be constant over time, for a given sector and country. We then use this ratio and the\nsector’s share in total value added during our spell (from World Bank, 2005) to compute the corresponding\nratio of unskilled workers in the sector to total unskilled workers in the country. Under wage equalization,\nthe ratio of unskilled workers in a sector to total unskilled workers can be written as\n## lk l =α k sk ∑α isi, where α k is the ratio of unskilled labor payments to sector’s k value added, and s k]\n\n\nThe resulting sample consists of 55 countries for 3-sector data and 51 countries\n\n\nfor 6-sector data. Appendix 1 provides the list of countries included in the sample, as\n\n\nwell as the initial and final years of their corresponding spell. Appendix 2 provides\n\n\ndefinitions and sources for all variables used in our empirical exercises, and Appendix 3\n\n\npresents basic summary statistics on the 51-country sample.\n\n\n**Poverty reduction and sectoral growth**\n\n\nWe are interested in estimating the effect of sectoral growth on poverty reduction.\n\n\nThe regression equation can then be written as,\n\n\n\n_I_\n\n\n\n_I_\n\nˆ\n## h j =δ 0 + ∑δ i ⋅ sij ⋅ y ˆ ij +ε j\n\n\n\nˆ\n## h j =δ 0 + ∑δ i ⋅ sij ⋅ y\n\nMention: \"GTAP\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The original single observation per country comes from this source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003293:11:0", "occurrence_id": "10830716c0c2f5499f8bdda6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 2000 US$, corrected for purchasing power (economic development), the log of initial\n\n\npoverty, the average GDP deflator (average inflation), the log of the average fraction of\n\n\nthe population in rural areas, the log of the average fraction of the population employed\n\n\nin the agricultural sector, and the log of the average value of trade (exports and imports)\n\n\nas a fraction of GDP.\n\n\nWe also take into account trade in food which may affect undernourishment and\n\n\nagricultural productivity. Specifically, in our panel regressions, we control for the yearly\n\n\ntotal food production per person in kilograms. We also control for the effects of\n\n\ninternational food trade, using the net food flow that leaves the country yearly, i.e., food\n\n\nexport minus food import, expressed in kilograms per person. For both total food\n\n\nproduction and trade data, we use the data as provided by FAOSTAT of the Food and\n\n\nAgriculture Organization of the United Nations, averaged for the following time periods:\n\n\n1979-81, 1990-92, 1993-95, 1995-97 and 2001-03. Finally, to investigate more general\n\n\neffects, we calculate a local producer price index based on yearly data from 1991-2001.\n\n\nThe price index is the weighted-average price of the following main food categories:\n\n\nwheat, rice, maize, oats, barley, sheep meat, chicken meat, and pig meat. The price of\n\n\neach category is weighted by its share in total production of all categories in the\n\n\nparticular country. The prices are in US dollars and corrected for “green” purchasing\n\n\npower parity. This PPP is calculated by FAO using a basket of agricultural products and\n\n\nrelated producer prices. To retain consistency, we make the periods for the price index as\n\n\nclose as possible to the periods defined by the FAO. Specifically, we average the data\n\n\nover the following periods: 1991-92, 1993-95, 1995-97, and 2001.\n\n\n<u>Correlations</u>\n\nMention: \"data as provided by FAOSTAT\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Per-capita food production and trade quantities are averaged from these series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003307:2:1", "occurrence_id": "1821badb1a31320384381c48", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  other countries. As well, some products and sectors may\n\nbe more sensitive to bilateral security measures (like social network goods or differentiated products\n\na la Rauch) than others for which transaction costs are less discriminatory across countries (standard\n\ngoods and primary products). An additional route to identify the counter security effect however, is\n\nto link terrorism acts to trade through a security measure that we can observe in the data.\n\n\nWe investigate empirically all these issues on US bilateral imports. We have chosen this country\n\nbecause it has been the main ”target” of transnational terrorism (i.e. one-third of the incidents over\n\nthe considered period target the U.S) and the one with the largest variation of ”origin” partners (more\n\nthan 95 countries have hit the U.S between 1968 and 2003). We combine, thereby, three datasets on\n\ntrade, terrorism incidents and business number of visas issued by the U.S.\n\n\nFirst for the trade data, we use bilateral imports of the United States at the product level (SITC4/5\n\ndigits) from the NBER World Trade Data complied by Feenstra and Lipsey (2005). As it provides\n\nonly values which exceed 100,000$ per year, we have completed it with the OECD-FLUBIL bilateral\n\ntrade dataset. Disaggregated data are needed here, in particular, in order to be able to capture the\n\ndifferentiated impact of counter-terrorism measures on trade across products, as mentioned above.\n\nBesides, as it will be shown next, as both bilateral trade and terrorism activities seem to be correlated\n\nwith the relative specialization of countries, one needs to condition out this effect in the regressions,\n\nwhich could not have been done by using highly aggregated data.\n\n\n1It might be argued that it is not the case anymore in the very recent years (say after 2000), where Al-Qaeda is\nbecoming more fragmented, acting as a Multinational with affiliates over the world.\n\nMention: \"NBER World Trade Data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Bilateral US imports at four- and five-digit product level are taken from it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003307:2:1", "occurrence_id": "fac5546459728994a2ca9484", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  other countries. As well, some products and sectors may\n\nbe more sensitive to bilateral security measures (like social network goods or differentiated products\n\na la Rauch) than others for which transaction costs are less discriminatory across countries (standard\n\ngoods and primary products). An additional route to identify the counter security effect however, is\n\nto link terrorism acts to trade through a security measure that we can observe in the data.\n\n\nWe investigate empirically all these issues on US bilateral imports. We have chosen this country\n\nbecause it has been the main ”target” of transnational terrorism (i.e. one-third of the incidents over\n\nthe considered period target the U.S) and the one with the largest variation of ”origin” partners (more\n\nthan 95 countries have hit the U.S between 1968 and 2003). We combine, thereby, three datasets on\n\ntrade, terrorism incidents and business number of visas issued by the U.S.\n\n\nFirst for the trade data, we use bilateral imports of the United States at the product level (SITC4/5\n\ndigits) from the NBER World Trade Data complied by Feenstra and Lipsey (2005). As it provides\n\nonly values which exceed 100,000$ per year, we have completed it with the OECD-FLUBIL bilateral\n\ntrade dataset. Disaggregated data are needed here, in particular, in order to be able to capture the\n\ndifferentiated impact of counter-terrorism measures on trade across products, as mentioned above.\n\nBesides, as it will be shown next, as both bilateral trade and terrorism activities seem to be correlated\n\nwith the relative specialization of countries, one needs to condition out this effect in the regressions,\n\nwhich could not have been done by using highly aggregated data.\n\n\n1It might be argued that it is not the case anymore in the very recent years (say after 2000), where Al-Qaeda is\nbecoming more fragmented, acting as a Multinational with affiliates over the world.\n\nMention: \"OECD-FLUBIL bilateral\n\ntrade dataset\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It fills in small trade values the primary source omits.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003308:24:1", "occurrence_id": "5f428b246f0d3da44d1b8703", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \nCotton (Seed) 3.3 1.3 -- 2.7 2.1 1.9 3.3 0.6 3.3\nCoffee (Green) 0.4 -- -- 0.9 0.8 -- -- 1.0 -Citrus 12.4 20.1 15.4 22.0 17.0 34.7 18.3 17.8 8.2\nPulses 0.8 1.1 1.6 0.7 0.8 1.9 2.7 0.6 1.4\nVegetables 16.5 17.2 25.6 17.9 14.9 27.1 26.7 12.9 19.2\n<u>Soybeans</u> <u>1.6</u> <u>2.5</u> <u>0.0</u> <u>2.6</u> <u>2.5</u> <u>2.6</u> <u>3.3</u> <u>0.8</u> <u>1.7</u>\n<u>Source: Caballero (2005)--calculations based on FAO’s AGROSTAT.</u>\n\n\nLabor productivity in Argentina’s agricultural sector is the highest in the region and\nway above the average for LAC countries. Labor productivity is compared across the LAC\nregion in Figure 7.1. Argentina and Uruguay, both land-abundant countries, have the\nhighest labor productivity, with a labor productivity of US$ 9,461 and US$ 7,807,\nrespectively, which is far more than double the LAC average of US$ 3,368. The high\nagricultural labor productivity in Argentina are mainly due to factors other than public\nexpenditure in the sector, which are only $136 per worker and\n\nMention: \"FAO’s AGROSTAT\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Agricultural labour productivity comparisons are calculated from it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003309:31:1", "occurrence_id": "91a41e8b0a61eea16df05eba", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: >\n\n\n\nNo education\n\nor\nprimary\nincomplete\n\n\n\nPrimary\ncomplete\n\nor\nsecondary\n<u>incomplete</u>\n\n\n\nTotal Argentina 17.9 48.9 24.5 8.7\nTotal Urban Argentina 15.6 49.0 26.1 9.4\nTotal Rural Argentina 38.6 49.2 9.2 3.1\nTotal Grouped Rural Argentina 29.9 52.9 12.8 4.4\n<u>Total Dispersed Rural Argentina</u> <u>42.8</u> <u>47.5</u> <u>7.3</u> <u>2.4</u>\n<u>Source: Own calculations based on INDEC National Population Census 2001.</u>\n\n\nSchool attendance by poor students still lags in rural Argentina. In dispersed rural\nareas, children from richer households have on average a higher school attendance, are\nless likely to repeat a school year, and have more completed years of schooling than\nchildren from poor households. Furthermore, data reveal a negative correlation between\npoverty and educational attainment in rural Argentina. The level of education of the\nextremely poor is the lowest, and it is also increasing more slowly than average.\n\n\n32\n\nMention: \"INDEC National Population Census 2001\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Rural school attendance and population shares are computed from this enumeration.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003310:15:1", "occurrence_id": "2688843943061d8639bd2286", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  missing values in the data force us to use a 4-year\naverage, even for the cross-section work. At best, we could estimate a pseudo-panel regression with an\nequivalent average for a previous period in the 1990’s. Even if we adopted this approach, we would\nconfront intractable problems with our road quality index. The cross-country index in Table 2 follows a\nsensible pattern but, as we have noted, World Development Indicator statistics on percent paved roads\nexhibit significant short-period fluctuations that seem very unlikely to reflect changes in actual road\nquality. In any case, we have no time-compatible measures for the CPIA index, which only exists for\nrecent years. The objective of this exercise is to predict the effect of a hypothetical change in road quality\nwhich, given the measurement issues, could only enter a pseudo-panel regression as a fixed effect. The\nsame is true for road distance. After considering all of these problems, we have decided against any\nattempt at panel estimation.\n20 We cannot test the effect of the South African Customs Union (SACU – Botswana, Lesotho, Namibia,\nSouth Africa, Swaziland)) because the relevant inter-country observations are missing in the data.\n21 Appendix Table A7 lists member countries of WAEMU, CEMAC, EAC, ECOWAS, COMESA and\nSADC.\n\n\n15\n\nMention: \"World Development Indicator statistics\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its paved-road shares fluctuate implausibly over short periods.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003310:15:1", "occurrence_id": "d8725fed3e614d4bc1f60f40", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  missing values in the data force us to use a 4-year\naverage, even for the cross-section work. At best, we could estimate a pseudo-panel regression with an\nequivalent average for a previous period in the 1990’s. Even if we adopted this approach, we would\nconfront intractable problems with our road quality index. The cross-country index in Table 2 follows a\nsensible pattern but, as we have noted, World Development Indicator statistics on percent paved roads\nexhibit significant short-period fluctuations that seem very unlikely to reflect changes in actual road\nquality. In any case, we have no time-compatible measures for the CPIA index, which only exists for\nrecent years. The objective of this exercise is to predict the effect of a hypothetical change in road quality\nwhich, given the measurement issues, could only enter a pseudo-panel regression as a fixed effect. The\nsame is true for road distance. After considering all of these problems, we have decided against any\nattempt at panel estimation.\n20 We cannot test the effect of the South African Customs Union (SACU – Botswana, Lesotho, Namibia,\nSouth Africa, Swaziland)) because the relevant inter-country observations are missing in the data.\n21 Appendix Table A7 lists member countries of WAEMU, CEMAC, EAC, ECOWAS, COMESA and\nSADC.\n\n\n15\n\nMention: \"CPIA index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"It exists only for recent years, so no time-compatible measure is available.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003311:20:2", "occurrence_id": "ecb5a8259aa825f257af9beb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  credit to the private sector, P_e is end-of period CPI, and P_a is period CPI (line 64M..ZF or, if not available, 64Q..ZF), and annual CPI (line 64..ZF) are from the\nis average annual CPI. electronic version of the IFS.\n\n\n\n20\n\nMention: \"IFS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Annual consumer price series are read from its electronic edition.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003318:3:1", "occurrence_id": "3cfaa93af1c9c32b35df5084", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  econometric methods for\n\n\ndetecting collusion. The fifth section presents an application of such methods to recent timber\n\n\nauction data from two Romanian forest directorates, Neamt and Suceava. Our results confirms\n\n\nthat data from Romanian timber auctions can be used to determine the likelihood of collusion,\n\n\nand it suggests that collusion reduced winning bids in Suceava in 2002 and perhaps also in\n\n\nNeamt, where less data were available. The sixth section concludes this paper.\n\n\n2 We review these studies in section 4.\n\n\n3 For example, tracts that sell for low prices might have few bidders not because of collusion but rather because they\n\n\nare of low quality. To control for this endogeneity between the quality of a tract and the number of bidders, Vincent\n\n\n(1996) used the predicted number of bidders rather than the actual number in a re-analysis of the data. He found less\n\n\nevidence of an impact of bidder number on winning prices.\n\n\n4\n\nMention: \"timber\n\n\nauction data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Auction records from two forest directorates support the collusion tests.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003318:3:1", "occurrence_id": "caec4f5ae9bf10ce0d316b63", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  econometric methods for\n\n\ndetecting collusion. The fifth section presents an application of such methods to recent timber\n\n\nauction data from two Romanian forest directorates, Neamt and Suceava. Our results confirms\n\n\nthat data from Romanian timber auctions can be used to determine the likelihood of collusion,\n\n\nand it suggests that collusion reduced winning bids in Suceava in 2002 and perhaps also in\n\n\nNeamt, where less data were available. The sixth section concludes this paper.\n\n\n2 We review these studies in section 4.\n\n\n3 For example, tracts that sell for low prices might have few bidders not because of collusion but rather because they\n\n\nare of low quality. To control for this endogeneity between the quality of a tract and the number of bidders, Vincent\n\n\n(1996) used the predicted number of bidders rather than the actual number in a re-analysis of the data. He found less\n\n\nevidence of an impact of bidder number on winning prices.\n\n\n4\n\nMention: \"data from Romanian timber auctions\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"They prove usable for detecting collusion and estimating its effect on bids.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003319:13:1", "occurrence_id": "73d71fbd92263e0d2e223e6b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  1994). In the\n\n\n1950s and 1960s, the auto complex (motor vehicles, steel, and especially roads) accounted for\n\n\nvirtually all of the World Bank’s loans to Japan not dedicated to electricity generation (World\n\n\nBank web page, projects database entries for Japan). While it is true that efforts by the\n\n\ngovernment in the 1950s and early 1960s to attain economies of scale by encouraging mergers\n\n\nof existing assemblers and discouraging new entry met stiff resistance from second-tier\n\n\n14\n\nMention: \"projects database entries for Japan\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Loan records identify the projects not dedicated to electricity generation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003326:20:0", "occurrence_id": "1585e83980620d33ce145a2d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **<mark>Appendix - Variable Definitions and Sources</mark>**\n\n\n\n**Variable** **Definition** **Source**\n_ForCont_ Ratio of the sum of assets of foreign banks in host country _i_ that are Bankscope\n\n\n\nRatio of the sum of assets of foreign banks in host country _i_ that are\nowned by banks from source country _j_, divided by the total amount\nof bank assets in host country _i_ .\n\n\n\nBankscope\n\n\n\n_Insthost_ The simple average of six indicators of quality of institutions in the Kaufmann, Kraay and Mastruzzi (2005)\n\nhost country (voice and accountability, politial instability and\nviolence, government effectiveness, regulatory quality, rule of law\nand control of corruption) measured in 2000.\n\n\n\n_DiffInst_ The absolute difference between quality of institutions of source and Kaufmann, Kraay and Mastruzzi (2005)\n\nhost countries measured in 2000, based on the simple average of the\nabsolute difference of each of the six indicators of quality of\ninstitutions.\n\n\n\n_CompAdv_\n\n\n\nInstitutional quality of the host country interacted with the difference\nbetween the institutional quality of the source country and that of the\ncompetitors. Institutional quality of competitors is measured by\ntaking the weighted average (based on GDP in current dollars) of the\ninstitutional quality of each of the possible source countries. All\nvariables are measured in 2000.\n\n\n\nKaufmann, Kraay and Mastruzzi (2005)\nand International Financial Statistics\n\n\n\n_Collinks_ Dummy which is 1 if host and source country have had colonial links CIA World Factbook (2005)\n\neither between colonizer and colony or between those countries\nhaving been colonized by the same colonizer after 1945, zero\notherwise.\n\n\n\n_Border_ Dummy which is 1 if the countries share a border, zero otherwise. CIA World Factbook (2005)\n\n_Comlang_ Dummy which is 1 if the countries share the same language, zero CIA World Factbook (2005\n\nMention: \"International Financial Statistics\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It supplies the country-level variables measured in that year.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003326:20:0", "occurrence_id": "10f82ab1eee8faf24bb16031", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **<mark>Appendix - Variable Definitions and Sources</mark>**\n\n\n\n**Variable** **Definition** **Source**\n_ForCont_ Ratio of the sum of assets of foreign banks in host country _i_ that are Bankscope\n\n\n\nRatio of the sum of assets of foreign banks in host country _i_ that are\nowned by banks from source country _j_, divided by the total amount\nof bank assets in host country _i_ .\n\n\n\nBankscope\n\n\n\n_Insthost_ The simple average of six indicators of quality of institutions in the Kaufmann, Kraay and Mastruzzi (2005)\n\nhost country (voice and accountability, politial instability and\nviolence, government effectiveness, regulatory quality, rule of law\nand control of corruption) measured in 2000.\n\n\n\n_DiffInst_ The absolute difference between quality of institutions of source and Kaufmann, Kraay and Mastruzzi (2005)\n\nhost countries measured in 2000, based on the simple average of the\nabsolute difference of each of the six indicators of quality of\ninstitutions.\n\n\n\n_CompAdv_\n\n\n\nInstitutional quality of the host country interacted with the difference\nbetween the institutional quality of the source country and that of the\ncompetitors. Institutional quality of competitors is measured by\ntaking the weighted average (based on GDP in current dollars) of the\ninstitutional quality of each of the possible source countries. All\nvariables are measured in 2000.\n\n\n\nKaufmann, Kraay and Mastruzzi (2005)\nand International Financial Statistics\n\n\n\n_Collinks_ Dummy which is 1 if host and source country have had colonial links CIA World Factbook (2005)\n\neither between colonizer and colony or between those countries\nhaving been colonized by the same colonizer after 1945, zero\notherwise.\n\n\n\n_Border_ Dummy which is 1 if the countries share a border, zero otherwise. CIA World Factbook (2005)\n\n_Comlang_ Dummy which is 1 if the countries share the same language, zero CIA World Factbook (2005\n\nMention: \"CIA World Factbook\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Colonial links between country pairs are coded from it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003329:2:0", "occurrence_id": "62115677aed896b268008a1d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 2\n\n\nand more on investment items -- like education and housing. Households receiving\n\n\nremittances also have a higher likelihood of investing in entrepreneurial activities.\n\n\nThis paper proceeds in five sections. Section 1 defines official international\n\n\nremittances and presents the new data set of 115 developing countries. Section 2 uses\n\n\nthis macro-level data set to estimate an econometric model to identity the determinants of\n\n\nremittances within the developing world. Moving to the level of the household, Section 3\n\n\nexamines how the receipt of international remittances affects the welfare and poverty\n\n\nstatus of households in various developing countries. Section 4 then analyzes how\n\n\nhouseholds consume and invest their remittance earnings, and the impact of these\n\n\nremittance expenditures on entrepreneurial activities and economic development. Section\n\n\n5 summarizes the findings and highlights areas for future work.\n\n\n1. Official International Remittances; Data Set of 115 Developing Countries\n\n\nIt should be stressed at the outset that a variety of data problems make it difficult\n\n\nto accurately examine the impact of international remittances on the developing world.\n\n\nThese data problems relate to the way in which information on international remittances\n\n\nis collected and compiled. While the International Monetary Fund (IMF) publishes\n\n\nannual records of the amount of “worker remittances” received by developing countries\n\n\nin its _Balance of Payments Statistics Yearbook_, these IMF data rely entirely on the\n\n\nreporting conventions of the central banks of some 185 developing countries. In other\n\n\nwords, while the central banks of many developing countries may receive remittances\n\n\nfrom their citizens working abroad, and correctly record these income flows as “worker\n\n\nremittances,” the central banks in other countries may report such remittance flows under\n\n\nanother line item in the balance of payments. In short, there is no reliable and consistent\n\n\nrule for enforcing the way that developing countries report “worker remittances” in their\n\n\nbalance of payments. This means that some countries which produce large numbers of\n\n\ninternational migrants – like the Philippines and the Russian Federation – report receiving\n\n\nlittle or no “worker remittances” because they (probably) record these remittance flows\n\nunder another category in their\n\nMention: \"data set of 115 developing countries\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The new compilation defines remittances and identifies their determinants.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003331:24:0", "occurrence_id": "48ad400a085a1f29218927b2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Table 1: Marital Status of Recent Mexican Immigrants in the U.S. Census**\n\n\n%\n\n<u>Males</u> <u>Females</u>\nMarried, with spouse present 14.4 48.3\nMarried, with spouse absent 26.9 8.7\nSeparated 1.8 3.7\nDivorced 1.5 2.4\nWidowed 0.2 0.7\n<u>Never married/Single</u> <u>55.2</u> <u>36.1</u>\n\n\nSource: US Census 5% public use sample (Ruggles et al, 2004)\nImmigrants born in Mexico who migrated to the U.S. in last two years\n\n\n**<u>Table 2: Summary Statistics of Key Variables</u>**\n\nNumber of All males 15-49 Migrants\n_<u>Individual level variables</u>_ <u>Observations</u> <u>Mean</u> <u>Std dev.</u> <u>Mean</u> <u>Std dev.</u>\n_ALL COMMUNITIES_\nProportion migrating in the last two years 55848 0.0099 1\nAge 55848 28.6 9.7 28.4 7.9\nYears of Education 55848 8.7 4.4 7.0 3.6\nProportion married 55848 0.55 0.75\nProportion who are household heads 55848 0.50 0.65\nYears of Education*Community migration prevalence 55848 1.26 1.3 1.90 1.4\n_COMMUNITIES WITH 100,000 OR LESS POPULATION_\nProportion migrating in the last two years 25531 0.014\nAge 25531 28.2 9.8 28.0 8.0\nYears of Education 25531 7.3 4.1 6.3 3\n\nMention: \"U.S. Census\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Marital status of recent Mexican immigrants is tabulated from it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003331:24:0", "occurrence_id": "ce29840cd42a402b607c1578", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Table 1: Marital Status of Recent Mexican Immigrants in the U.S. Census**\n\n\n%\n\n<u>Males</u> <u>Females</u>\nMarried, with spouse present 14.4 48.3\nMarried, with spouse absent 26.9 8.7\nSeparated 1.8 3.7\nDivorced 1.5 2.4\nWidowed 0.2 0.7\n<u>Never married/Single</u> <u>55.2</u> <u>36.1</u>\n\n\nSource: US Census 5% public use sample (Ruggles et al, 2004)\nImmigrants born in Mexico who migrated to the U.S. in last two years\n\n\n**<u>Table 2: Summary Statistics of Key Variables</u>**\n\nNumber of All males 15-49 Migrants\n_<u>Individual level variables</u>_ <u>Observations</u> <u>Mean</u> <u>Std dev.</u> <u>Mean</u> <u>Std dev.</u>\n_ALL COMMUNITIES_\nProportion migrating in the last two years 55848 0.0099 1\nAge 55848 28.6 9.7 28.4 7.9\nYears of Education 55848 8.7 4.4 7.0 3.6\nProportion married 55848 0.55 0.75\nProportion who are household heads 55848 0.50 0.65\nYears of Education*Community migration prevalence 55848 1.26 1.3 1.90 1.4\n_COMMUNITIES WITH 100,000 OR LESS POPULATION_\nProportion migrating in the last two years 25531 0.014\nAge 25531 28.2 9.8 28.0 8.0\nYears of Education 25531 7.3 4.1 6.3 3\n\nMention: \"US Census\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its five percent public use sample provides the migrant observations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003334:12:0", "occurrence_id": "5264b1cbd64647548b13ecea", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Educación Superior ICFES), which administers standardized tests in Colombia. The C600 and\n\n\nthe C100 surveys provide general data on an important array of school characteristics and have\n\n\nan identification code that allows researchers to link these datasets with the ICFES test scores.\n\n\nThe Ministry of Education collects the data directly via a questionnaire to all the schools in the\n\n\ncountry. Data are at the school and “schedule” level.\n\n\nThe school level data present information on the number of administrative personnel, the number\n\n\nand level of education of teachers, the number of teachers by subject areas, the number of\n\n\nphysicians in the school, the total number of students in the school by grade and by age, the\n\n\nnumber of students who failed a grade and the number of students who dropped out. These data\n\n\nare available for both 1999 and 2003.\n\n\nFor 1999, there is also information on the physical characteristics of schools. The main variables\n\n\nare: furniture in the school (chairs and desks), support materials such as computers, the number\n\n\nof computers exclusively for teaching purposes, the total number of laboratories, the number of\n\n\nlaboratories for specific subjects (physics, chemistry, biology, construction and farming), the\n\n\ntotal number of rooms, the number of classrooms, the number of libraries, the number of food\n\n\nfacilities and dorms, the number of other types of rooms and the number of sports facilities\n\n\n(soccer fields and basketball courts).\n\n\nThe surveys also breakdown the data at the schedule level, the time at which the school offers\n\n\nservices. There are four main schedules during which students can attend school: in the morning\n\n\n(usually from 7:00 am to 12:00 pm), in the afternoon (from 12:00 pm to 4:00 pm), in the evening\n\n\n(from 2:00 pm to 7:00 pm); or they can attend a school with a “complete” schedule (from 8:00\n\n\nam to 3:00 pm). The schedule mechanism was\n\nMention: \"C600\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its school characteristic items carry an identification code for matching.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003334:12:0", "occurrence_id": "c9744763ed2a887c26da90aa", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Educación Superior ICFES), which administers standardized tests in Colombia. The C600 and\n\n\nthe C100 surveys provide general data on an important array of school characteristics and have\n\n\nan identification code that allows researchers to link these datasets with the ICFES test scores.\n\n\nThe Ministry of Education collects the data directly via a questionnaire to all the schools in the\n\n\ncountry. Data are at the school and “schedule” level.\n\n\nThe school level data present information on the number of administrative personnel, the number\n\n\nand level of education of teachers, the number of teachers by subject areas, the number of\n\n\nphysicians in the school, the total number of students in the school by grade and by age, the\n\n\nnumber of students who failed a grade and the number of students who dropped out. These data\n\n\nare available for both 1999 and 2003.\n\n\nFor 1999, there is also information on the physical characteristics of schools. The main variables\n\n\nare: furniture in the school (chairs and desks), support materials such as computers, the number\n\n\nof computers exclusively for teaching purposes, the total number of laboratories, the number of\n\n\nlaboratories for specific subjects (physics, chemistry, biology, construction and farming), the\n\n\ntotal number of rooms, the number of classrooms, the number of libraries, the number of food\n\n\nfacilities and dorms, the number of other types of rooms and the number of sports facilities\n\n\n(soccer fields and basketball courts).\n\n\nThe surveys also breakdown the data at the schedule level, the time at which the school offers\n\n\nservices. There are four main schedules during which students can attend school: in the morning\n\n\n(usually from 7:00 am to 12:00 pm), in the afternoon (from 12:00 pm to 4:00 pm), in the evening\n\n\n(from 2:00 pm to 7:00 pm); or they can attend a school with a “complete” schedule (from 8:00\n\n\nam to 3:00 pm). The schedule mechanism was\n\nMention: \"C100 surveys\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The companion survey supplying the same school characteristics.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003336:8:1", "occurrence_id": "89f4e284d298fa1c473a49a6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \nln _Export_ _ _valueGermany_, _cpt_   - ln Im _port_ _ _valuecpt_ = _trade_ _ _gapcpt_ =α+ β _tariff_ _cpt_ +α _t_ +ε _cpt_\n\n\nwhere _p_ stands for a 6-digit product and _t_ for year. Our prior is that if the gap value is a good proxy for\n\n\ntariff evasion then the estimated coefficient of the tariff rate should be positive and significant.\n\n\nThe results, reported in Table 3, are consistent with the summary statistics presented earlier. The\n\n\nestimated coefficient on the tariff rate is positive and significant at the 1% level for all the countries but\n\n\nSlovenia and Poland. The higher the tariff rate, the lower the value of imports reported by the importing\n\n\ncountry relative to the reported exports (i.e, the higher the trade gap). A one-percentage-point increase in\n\n\nthe tariff level is associated with a 4.4% increase in the trade gap in Ukraine, 3.2% increase in the Russian\n\n\nFederation and 0.8% increase in Hungary. These results are in line with Fisman and Wei’s study which\n\nfinds a 3% increase. <sup>9</sup>\n\n\nIt is interesting to note that Ukraine, the country with the highest estimated elasticity, has the\n\n\nsecond highest prevalence of corruption in customs as reported in the BEEPS survey. Slovenia, a country\n\n\nfor which there is no statistically significant relationship, is ranked as the cleanest country in terms of\n\n\ncorruption in customs according to BEEPS. See Appendix II for more details. The insignificant\n\n\n9 Note that these calculations do not take into account the direct effect an increase in a tariff rate may have on the\nvolume of imports.\n\n\n9\n\nMention: \"the BEEPS survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It reports the prevalence of corruption in customs across countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003338:1:1", "occurrence_id": "5cfb6ced4fdd226cfb0ef7f6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  the shadow of mutual\n\nthreat across the marriages. As the watta-bride quoted above expresses so succinctly, a\n\nhusband who ‘mistreats’ his wife in this arrangement can expect his brother-in-law to\n\nretaliate in-kind against his sister.\n\n\nWe shall show that such reciprocal threats operating across marriages can be cred\nible and may, consequently, prevent inefficient marital outcomes. Husbands generally\n\nhave coercive power over their wives, through various forms of physical and emotional\n\nintimidation. In virilocal societies, this power is heightened by the wife’s residence in\n\n\n1 All of the quotations in this paper are from field interviews conducted as part of a qualitative survey\nundertaken in selected villages covered by the 2004-05 Pakistan Rural Household Survey.\n\n2 Similar marriage patterns prevail in south India, whereas in north India sibling exchange marriage is\ngenerally disapproved of (Mencher and Goldberg, 1967; Karve, 1993; Trautmann, 1993). Nevertheless,\nexchange marriage appears to have been common among Hindus in the Punjab, at least formerly (Rose,\n1908). Outside south Asia, exchange marriage has been documented in parts of China (Zhang, 2000).\nEarly anthropological accounts of exchange marriage in west Africa also mention the ‘retaliatory’ motive\nemphasized in this paper, but here it is typically in reaction to a wife’s ‘malfeasance’ (such as leaving her\nhusband) that her brother’s marriage is dissolved.(Bohannan, 1949; Meek, 1936).\n\n\n1\n\nMention: \"the 2004-05 Pakistan Rural Household Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Qualitative field interviews were conducted in villages this survey covers.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003340:8:0", "occurrence_id": "63d8f7257fe085f683602edf", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: environment matter, but there is less agreement as to whether absolute differences or differences\n\n\nvis-à-vis competitors are important. Galindo, Micco, and Serra (2003) argue that absolute legal,\n\n\nregulatory and institutional differences across countries can increase entry costs and reduce the\n\n\nparticipation of banks in foreign countries. Using bilateral foreign banking data for 176\n\n\ncountries, the authors find results that support their hypotheses. In other words, they find that\n\n\nforeign bank penetration is higher between countries that have similar legal origins, banking\n\n\nregulations, and institutional set ups. On the other hand, Claessens and Van Horen (2006) argue\n\n\nthat what is important in driving foreign bank participation is not the difference between the\n\n\ninstitutional environment in the host and source country, but rather the difference taking into\n\n\naccount the institutional quality of the competitors from other source countries. Using data for\n\n\nmost banks in all developing countries, the authors find that absolute levels of institutional\n\n\ndifferences between the host and source countries do not matter once they control for the relative\n\n\nadvantage the source country has compared to its competitors from other countries operating in\n\n\nthe same host.\n\n\nTo summarize, there is substantial evidence that factors such as market opportunities,\n\n\nregulatory barriers in the host countries along with economic, cultural and institutional\n\n\nsimilarities between home and host economies influence the decision of banks to operate\n\n\noverseas. Contrary to the wealth of evidence on these factors, the role that crises can play in\n\n\nbringing about foreign bank participation has only been discussed anecdotally. For example, in\n\n\ndiscussing the rise in foreign bank participation in Argentina, Brazil and Mexico, Peek and\n\n\nRosengren (2000) mention the Tequila Crisis in Mexico and Argentina and the 1999 Brazilian\n\n\ncrises as catalytic events. Moreno and Villar (2006) also point to the experience of Mexico in\n\n\nstating that foreign bank entry might be stimulated by the need for countries to recapitalize their\n\n\nbanking systems in the aftermath of crises. Similarly, Tschoegl (2003) discusses the importance\n\n\nof\n\nMention: \"bilateral foreign banking data for 176\n\n\ncountries\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"A cited study uses these bilateral data to test entry cost hypotheses.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003343:10:0", "occurrence_id": "a8f9fa6d270e9730d71e3ddf", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: take place between countries without bilateral aid flows. Moreover, we also expect to find a non\nzero correlation between the two error terms in the statistical model in a way that is consistent\n\n\nwith the truncation of potential outcomes. Alternatively, when investment decisions are not\n\nsubject to agency selection – that is, when firms optimize over Ω ( _st_ ) -- we expect no correlation\n\n\nbetween the error terms of the statistical model and any observed correspondence between\n\ninvestment flows and aid flows is attributed to coincidence.\n\n\n**5.** **Data Description**\n\n\n5.1. <u>AIJ Investments</u>\n\n\nOur dichotomous measure of AIJ investment is taken from data on AIJ projects reported to the\n\nUNFCCC and compiled by the SBSTA. As of February 2002, the UNFCCC reported on 156\n\n\nprojects. Among these, we include in our sample the 147 projects in which a letter of intent was\n\nsigned with the host country. The projects began between 1992 and 2001 and were distributed\n\namong 22 investor countries and 42 host countries (table 1). <sup>16</sup> [^16: The Bonn Decision, taken in 1995, gave rise to several new national programs. However, the Decision also\nbrought several established programs into the common AIJ framework. Consequently, some projects under the\nDutch, Norwegian, Swedish and US programs were negotiated prior to the Bonn decision.] For our initial analysis we focus on\n\nthe 65 unique investor-host partnerships that evolved over the study period; that is, for our\n\n\npreferred model we define an investment event as the joint agreement by an investor country and\n\na host country to participate in any number of AIJ projects. We do so for two reasons.\n\n\nFirst, this approach is more in keeping with descriptive studies that suggest agency\n\npreferences, sometimes tied to national policy objectives, were influential in screening outcomes.\n\n\nThe related point of interest then is whether the agency approval hurdle is cleared for any given\n\npotential partnership, rather than how frequently it is cleared. Second\n\nMention: \"the\n\nUNFCCC\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Project reports compiled here give the joint implementation investment count.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003346:4:0", "occurrence_id": "cac17d315f49c40c85e46ac6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 5\n\n\nrather than double differences (looking at differences in changes between ‘treated’ and\n\n\n‘untreated’ households before and after a program’s implementation). The reason for\n\n\nusing single differences is that there are no suitable baseline data that would allow\n\n\ndouble-differencing to be undertaken. Some of the households in the 2004 Vietnam\n\n\nHousehold and Living Standard Survey (VHLSS)—the survey used in what follows—\n\n\nwere interviewed two years before in the 2002 VHLSS. However, although the HCFP\n\n\nprogram did not exist at that date, its predecessors did, so already in 2002 there were poor\n\n\npeople who had been assigned a health card or who had been enrolled in the health\n\n\ninsurance program. If the aim is to get an estimate of the impact of being covered by one\n\n\nor other of the HCFP modalities versus not being covered (but eligible for the program),\n\n\nthe 2002 VHLSS cannot therefore serve as a baseline. Aside from this, only a minority of\n\n\nhouseholds in the 2004 survey can be found in the 2002 survey, and the health utilization\n\n\nquestions are different, making a double-difference somewhat heroic. The limitations of\n\n\nthe single-difference approach used in the paper are discussed further below.\n\n\nThe paper is organized as follows. Section II provides a brief description of the\n\n\nHCFP program. Section III outlines the methods used to estimate its impact. Section IV\n\n\npresents the data and evidence on targeting. Section V presents the estimation results of\n\n\nthe model used to estimate the propensity score for the analysis of impacts, and the\n\n\nresults of balancing tests. These estimation results also shed further light on the issue of\n\n\ntargeting. Section VI presents the estimates of the impacts of HCFP, and the final section\n\n\n(VII) contains a summary and discussion.\n\n\nII. **VIETNAM’S HEALTH CARE FUND FOR THE POOR**\n\n\nIn October 2002, through an edict known as Decision 139 <sup>5</sup>\n\nMention: \"2004 Vietnam\n\n\nHousehold and Living Standard Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Part of the sample was revisited, though a prior round limits double differencing.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003353:8:1", "occurrence_id": "c653507ed435916a9b8f9f5e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  in the sensitivity of results section that this parameter does\nnot impact the welfare results significantly. On the other hand, despite the fact that multinationals use\nKazakhstan labor less intensively than their Kazakhstan competitors, if multinationals use mostly\nKazakhstan labor, their expansion is likely to increase the demand for Kazakhstan labor in these sectors. <sup>11</sup> [^11: See Markusen, Rutherford and Tarr (2005) for a detailed explanation on why FDI may be a partial equilibrium\nsubstitute for domestic labor but a general equilibrium complement.]\nDue to lack of data in Kazakhstan, we used our estimates from our Russia model for this parameter; but\nwe perform sensitivity analysis with respect to this parameter.\n\n\n**Tariff and Export Tax Data.** We begin with tariff rate data at the four digit Harmonized System\nlevel based on government decree number 1389 of the Republic of Kazakhstan with updates and\namendments as of May 5, 2005. <sup>12</sup> [^12: The data were extracted from “Jurist” (a source of data on the laws of Kazakhstan) as of September 29, 2005. This\nis an on-line data source at <u>http://base.zakon.kz. Tariff rates for Kazakhstan may also be found at</u>\n<u>http://zakon.kz/,</u> which is an electronic reference database to the legislation of Kazakhstan.] We then aggregated the four digit data to two digit categories using a\nsimple average method of tariff rate aggregation. Subsequently, we mapped the two digit tariff lines to\nthe sectors of our model and again used a simple average of the tariff rates to obtain the tariff rates at the\nlevel of aggregation of our model. The export tax rate data were taken from the rates implied by the\ninput-output table.\n\n\nby which telecommunications prices are elevated due to the barriers and if we assume Kazakhstan domestic service\nproviders prices are unaffected, then we may solve for x from\n\nMention: \"Tariff and Export Tax Data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Four-digit tariff rates set by government decree are recorded here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003357:44:1", "occurrence_id": "ffd18a99d1808918ed2b1e83", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  of goods and services, expressed in current U.S. dollars. World Bank: World Development Indicators\n\n\nLaw and order Index Law and order index. International Country Risk Guide (ICRG)\n\nInvestor protection index Investor protection index. World Bank: Doing Business\n\n\n**Firm-Level**\nTotal assets Total assets at the end of the most recent fiscal year, expressed in U.S. dollars. Worldscope\n\nTwo-year average sales growth Geometric average annual sales growth over the last two years. Sales are expressed in U.S. dollars. Worldscope\n\n\n\nReturn on assets Sum of net income (before preferred dividends) and interest expenses on capitalized debt (after tax), over last year's total\nassets.\n\n\n\nWorldscope\n\n\n\nForeign sales to total sales Annual international sales over net sales or revenues. Worldscope\n\n\n44\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Output in current dollars is defined from this compilation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003358:4:1", "occurrence_id": "7dec6a263e34079c34a80c22", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Notes:</u>\n…Data not available or not applicable\n(a) Health Resources by WHO (1999)\n(b) Assuming that the number of hospital beds remains at the 1999 level.\n(c) Calculated from “World Population Prospects: The 2004 Revision Population Database”.\n(d) Approximate figure taken from STAT-USA Market Research Report at http://strategis.ic.gc.ca/epic/internet/inimr\nri.nsf/en/gr108259e.html.\n(e) Figure includes 1,241 beds at health centers.\n(f) The figure does not include Primary Health Care Units which are made up of trained village volunteers.\n(g) Comprising of general hospitals, district hospitals, special medical institutions which are all public and other private\n\nhospitals.\n(h) PHC and SPHC are Primary Health Centers and Supporting Primary Health Centers respectively.\n\nLooking at countries’ stock of medical personnel, the Philippines and Singapore stand out from\nthe rest of the group in terms of the number of physicians and nurses per 10,000 inhabitants\n(Table 3). <sup>5</sup> [^5: One must be careful when interpreting the numbers in Table 3 since the figures available are often exaggerated.\nMany of the registered health personnel are no longer working or work abroad (see Section III). Moreover, the\nfigures refer to different years. However, given the long training period for physicians, their numbers tend to be\nrelatively stable through time.] Thailand also enjoys a relatively high number of nurses per population, although\n\n4 According to Leebouapao (2004), more than 700 health centers were closed and the health system almost came to\na collapse. The Northern region was the worse hit. As of 2004, 29.1 percent of the villages were at least 16\nkilometers away from the nearest health center. To improve access to essential drugs to the poor, so-called Drug\nRevolving Funds (DRF) were established in the mid-1990s. The system is\n\nMention: \"Health Resources by WHO\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Hospital bed counts are taken from this compilation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003358:4:1", "occurrence_id": "c1d172b2b83bd8daba97ec18", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Notes:</u>\n…Data not available or not applicable\n(a) Health Resources by WHO (1999)\n(b) Assuming that the number of hospital beds remains at the 1999 level.\n(c) Calculated from “World Population Prospects: The 2004 Revision Population Database”.\n(d) Approximate figure taken from STAT-USA Market Research Report at http://strategis.ic.gc.ca/epic/internet/inimr\nri.nsf/en/gr108259e.html.\n(e) Figure includes 1,241 beds at health centers.\n(f) The figure does not include Primary Health Care Units which are made up of trained village volunteers.\n(g) Comprising of general hospitals, district hospitals, special medical institutions which are all public and other private\n\nhospitals.\n(h) PHC and SPHC are Primary Health Centers and Supporting Primary Health Centers respectively.\n\nLooking at countries’ stock of medical personnel, the Philippines and Singapore stand out from\nthe rest of the group in terms of the number of physicians and nurses per 10,000 inhabitants\n(Table 3). <sup>5</sup> [^5: One must be careful when interpreting the numbers in Table 3 since the figures available are often exaggerated.\nMany of the registered health personnel are no longer working or work abroad (see Section III). Moreover, the\nfigures refer to different years. However, given the long training period for physicians, their numbers tend to be\nrelatively stable through time.] Thailand also enjoys a relatively high number of nurses per population, although\n\n4 According to Leebouapao (2004), more than 700 health centers were closed and the health system almost came to\na collapse. The Northern region was the worse hit. As of 2004, 29.1 percent of the villages were at least 16\nkilometers away from the nearest health center. To improve access to essential drugs to the poor, so-called Drug\nRevolving Funds (DRF) were established in the mid-1990s. The system is\n\nMention: \"World Population Prospects\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Population denominators for the per-capita figures come from this revision.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003358:4:1", "occurrence_id": "3961940fb9cdd795ac72bff4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Notes:</u>\n…Data not available or not applicable\n(a) Health Resources by WHO (1999)\n(b) Assuming that the number of hospital beds remains at the 1999 level.\n(c) Calculated from “World Population Prospects: The 2004 Revision Population Database”.\n(d) Approximate figure taken from STAT-USA Market Research Report at http://strategis.ic.gc.ca/epic/internet/inimr\nri.nsf/en/gr108259e.html.\n(e) Figure includes 1,241 beds at health centers.\n(f) The figure does not include Primary Health Care Units which are made up of trained village volunteers.\n(g) Comprising of general hospitals, district hospitals, special medical institutions which are all public and other private\n\nhospitals.\n(h) PHC and SPHC are Primary Health Centers and Supporting Primary Health Centers respectively.\n\nLooking at countries’ stock of medical personnel, the Philippines and Singapore stand out from\nthe rest of the group in terms of the number of physicians and nurses per 10,000 inhabitants\n(Table 3). <sup>5</sup> [^5: One must be careful when interpreting the numbers in Table 3 since the figures available are often exaggerated.\nMany of the registered health personnel are no longer working or work abroad (see Section III). Moreover, the\nfigures refer to different years. However, given the long training period for physicians, their numbers tend to be\nrelatively stable through time.] Thailand also enjoys a relatively high number of nurses per population, although\n\n4 According to Leebouapao (2004), more than 700 health centers were closed and the health system almost came to\na collapse. The Northern region was the worse hit. As of 2004, 29.1 percent of the villages were at least 16\nkilometers away from the nearest health center. To improve access to essential drugs to the poor, so-called Drug\nRevolving Funds (DRF) were established in the mid-1990s. The system is\n\nMention: \"2004 Revision Population Database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The specific database version behind the population calculations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003358:4:1", "occurrence_id": "9f68fb98aa774aa61facc4de", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Notes:</u>\n…Data not available or not applicable\n(a) Health Resources by WHO (1999)\n(b) Assuming that the number of hospital beds remains at the 1999 level.\n(c) Calculated from “World Population Prospects: The 2004 Revision Population Database”.\n(d) Approximate figure taken from STAT-USA Market Research Report at http://strategis.ic.gc.ca/epic/internet/inimr\nri.nsf/en/gr108259e.html.\n(e) Figure includes 1,241 beds at health centers.\n(f) The figure does not include Primary Health Care Units which are made up of trained village volunteers.\n(g) Comprising of general hospitals, district hospitals, special medical institutions which are all public and other private\n\nhospitals.\n(h) PHC and SPHC are Primary Health Centers and Supporting Primary Health Centers respectively.\n\nLooking at countries’ stock of medical personnel, the Philippines and Singapore stand out from\nthe rest of the group in terms of the number of physicians and nurses per 10,000 inhabitants\n(Table 3). <sup>5</sup> [^5: One must be careful when interpreting the numbers in Table 3 since the figures available are often exaggerated.\nMany of the registered health personnel are no longer working or work abroad (see Section III). Moreover, the\nfigures refer to different years. However, given the long training period for physicians, their numbers tend to be\nrelatively stable through time.] Thailand also enjoys a relatively high number of nurses per population, although\n\n4 According to Leebouapao (2004), more than 700 health centers were closed and the health system almost came to\na collapse. The Northern region was the worse hit. As of 2004, 29.1 percent of the villages were at least 16\nkilometers away from the nearest health center. To improve access to essential drugs to the poor, so-called Drug\nRevolving Funds (DRF) were established in the mid-1990s. The system is\n\nMention: \"STAT-USA Market Research Report\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"An approximate industry figure is taken from this report.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003360:9:0", "occurrence_id": "90ce5740290f3cee34c07a04", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: - Fifth and finally, a simple practical point. For many purposes we do recognize\n\n\nthat it can be of interest for some users to make comparisons of governance\n\n\nbased on particular individual data sources. To facilitate this, with the 2005\n\n\nrelease of the aggregate governance indicators, we have made all but three of\n\nour underlying data sources available to users on our website. <sup>8</sup> [^8: These three sources are the Country Policy and Institutional Assessments produced by the\nWorld Bank, the African Development Bank, and the Asian Development Bank, which for the\nmost part are treated as confidential by these organizations. Only in the past few years has\nlimited, but growing disclosure of this data been made by these organizations, and full public\naccess to the detailed disaggregated and historical data on which we rely is still not permitted..]\n\n\n**_<u>Critique 3: Changes over time in some of the individual indicators underlying the</u>_**\n\n\n**_WGI aggregate indicators reflect corrections of past errors rather than actual_**\n\n\n**_changes._**\n\n\nThis critique of several of our underlying indicators is made by K in the context of\n\n\nhis discussion of trends in governance in countries in the ECA region between 2002 and\n\n\n2005 . We do not think that the evidence provided by K supports his claim. K argues\n\n\nthat in many cases individual data sources change their assessments of governance in a\n\n\ncountry simply to correct past errors in their assessments: countries that were rated \"too\n\n\nhigh\" in the past get lower scores, and vice versa. K then goes on to document that\n\n\namong ECA countries there is \"regression to the mean\", in the sense that changes in\n\n\ngovernance tend to be negatively correlated with initial scores, and interpret this as\n\n\nevidence that risk rating agencies change their scores to correct past \"errors\".\n\n\nWe believe that this is an overinterpretation of his results based on a regression\n\n\nthat has a much simpler explanation. To be specific,\n\nMention: \"aggregate governance indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"inform\", \"summary\": \"With this release nearly all underlying sources were made publicly available.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003363:20:0", "occurrence_id": "bcc354e04fa10b85b9281197", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: these institutional variables separately in the regressions. Since these variables are\n\n\nalso correlated with income of the country we follow the previously used methodology\n\n\nand regress the institutional variables on per-capita income and population and take\n\n\nthe residuals as representative of deviations from income-conditional expected values\n\n\nfor each of the four indexes.\n\n\nAlternative variables measuring institutional quality were obtained from two sources.\n\n\nA proxy for the level of corruption was obtained from the Transparency International\n\n\nCorruption Perceptions Index for the period 1996-2003. The Index ranks countries\n\n\nin terms of the degree to which corruption is perceived to exist among public officials\n\n\nand politicians and focuses on corruption in the public sector and defines corruption\n\n\nas the abuse of public office for private gain.\n\n\nSeveral other variables measuring the quality of institutions and governance were\n\n\ntaken from Kaufmann, Kraay and Mastruzzi (2005). The authors estimate six di\n\nmensions of governance covering 209 countries and territories for five time periods:\n\n\n1996, 1998, 2000, 2002 and 2004. Data for the year 1997, 1999, 2001 and 2003 were\n\n\ninterpolated. The variables were used to check the robustness of our previous re\n\nsults: government effectiveness (measuring the competence of the bureaucracy and\n\n\nthe quality of public service delivery), political stability (measuring the likelihood of\n\n\nviolent threats to, or changes in, government, including terrorism), regulatory quality\n\n\n(measuring the incidence of market-unfriendly policies), rule of law (measuring the\n\n\nquality of contract enforcement, the police, and the courts, as well as the likelihood of\n\n\ncrime and violence), voice and accountability (measuring political, civil and human\n\n\nrights). The six indicators are measured in units ranging from about -2.5 to 2.5, with\n\n\nhigher values corresponding to better governance outcomes. The results using the\n\n\n20\n\nMention: \"the Transparency International\n\n\nCorruption Perceptions Index\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its country rankings proxy the perceived level of corruption over this period.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003365:13:0", "occurrence_id": "5831c59952c80d219f51ae7c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <sup>13</sup>\n\n\nTo assess this question, the following system and its analog for long-run financial\n\n\nsector development are estimated:\n\n\n(3) _Security of property rights_ ( _1975_ ) = _β1_ + _β2ln(initial financial sector) i_ +\n\n\n_β3(_ **political/institutional variable** _) i_ + **Xi ′β4** + _εi_\n\n\n(4) _Growth of the financial sector i_ ( _1975-2000_ ) = _α1_ + _α2ln(initial financial sector) i_ +\n\n\n_α3(_ **predicted security of property rights)** + **Xi ′a4** + _μi_\n\n\nThese regressions simply ask whether the component of secure property rights that can be\n\n\nexplained by politics or political institutions is a significant determinant of financial sector\n\n\ndevelopment.\n\n\n**Data**\n\nThe regressions below require data on financial sector development and the political\n\n\ncharacteristics of countries. For the first, estimates below rely on _private credit_, total credit\n\n\nextended to the private sector by banks and other financial institutions. This variable is the\n\n\npreferred measure of financial sector development in Beck, et al. (2001) and is taken from\n\n\nthe Financial Structures Database of the World Bank (see Beck, Demirgüç-Kunt and Levine\n\n\n2000). Beck, et al. (2001) describe this as the preferred measure of financial sector\n\n\ndevelopment.\n\n\nFour different measures of political institutions or competition are evaluated here.\n\n\nAll are taken from the _Database of Political Institutions_ (Beck, et al. 2001), running from 1975 –\n\n\n2000. In contrast to the measures used in much of the literature, those here are all objective\n\n\nand easily replicable by others. The _checks_ indicator is an objective counterpart to a\n\n\nsubjective measure often used\n\nMention: \"Financial Structures Database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its bank size measure is the preferred indicator of financial development.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003368:55:0", "occurrence_id": "8070fe35399268880d41b236", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <u>Variable</u> <u>Definition</u> <u>Source</u>\n\n\nOther Percentage of new investment over the last year coming from other sources Investment Climate Survey\n\nShare of borrowing in foreign\ncurrency Percentage share of total borrowing (loans, accounts payable) that is denominated in foreign currency? Investment Climate Survey\n\n\n<u>Necessity of collateral</u> <u>Did the financing require collateral or a deposit? Yes (1) No (0)</u> <u>Investment Climate Survey</u>\n\n**_<u>Historical Instruments</u>_**\n\n\n\nLa Porta, Lopez-de-Silanes, Shleifer and\nVishny (1999)\n\n\nLa Porta, Lopez-de-Silanes, Shleifer and\nVishny (1999)\n\n\nLa Porta, Lopez-de-Silanes, Shleifer and\nVishny (1999)\n\n\nLa Porta, Lopez-de-Silanes, Shleifer and\nVishny (1999)\n\n\nAlesina et al. (2003)\n\n\nLa Porta, Lopez-de-Silanes, Shleifer and\nVishny (1999)\n\n\n\nEnglish Common Law\n\n\n\nEnglish Common Law is a dummy variable that takes the value 1 for Common Law Countries and 0 for\nCivil Law and Socialist Law Countries\n\n\n\nCatholic Dummy Variables that takes the value 1 if the dominant religious group in the country are Catholics\n\n\nMuslim Dummy Variables that takes the value 1 if the dominant religious group in the country are Muslims\n\n\nProtest Dummy Variables that takes the value 1 if the dominant religious group in the country are Protestants\n\n\n\nEthnic Fractionalization\n\n\n\nEthnic Fractionalization is the probability that two randomly selected individuals in a country do not\nbelong to the same ethnic group\n\n\n\nLatitude Latitude is the absolute value of the latitude of the country scaled between zero and one.\n\n\n\n56\n\nMention: \"Investment Climate Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"It defines the share of new investment from other sources and foreign-currency borrowing.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003377:2:0", "occurrence_id": "cb03224d44fda452d18b3a00", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **1. Introduction**\n\n\nFrom its inception, the Penn World Tables (PWT), building on the International\n\n\nComparisons Program (ICP) of the United Nations, has sought to compare the _standard of living_\n\n\nof individuals in different countries. That is, the term “real GDP per capita” as reported in the\n\n\nPWT is intended to represent the ability to purchase goods and services by a representative agent\n\n\nin the economy. The same is true of benchmark comparisons as published by the United Nations,\n\n\nEurostat or OECD. But this expenditure-side interpretation of real GDP is quite different from\n\n\nthe uses to which benchmark ICP and PWT data are frequently applied, such as in the “millions”\n\n\nof growth regressions (Sala-i-Martin, 1997) where “real GDP” is intended to reflect the\n\nproduction side of the economy. <sup>1</sup> [^1: There are many other applications where expenditure-side real GDP is used when output-side real GDP is really\nneeded. Recent examples are Bernard and Jones (1996), Jones and Hall (1999) and Acemoglu and Ventura (2002).]\n\n\nThe goal of this paper is to carefully distinguish the production or output-side measure of\n\nreal GDP, denoted _real GDP_ <sup>_o_</sup>, from the expenditure-side measure, denoted _real GDP_ <sup>_e_</sup> . Real\n\nGDP <sup>o</sup> reflects the production possibilities of an economy, whereas real GDP <sup>e</sup> measures real\n\n\nexpenditures which includes an adjustment for the trade balance. The reason these concepts were\n\n\nnot distinguished in the ICP and PWT is that these projects treat the net foreign balance in an\n\n\nunsatisfactory way. While there may have been some data justifications for that treatment in\n\n\nbenchmark studies of the 1970s, this is no longer the case. The treatment of exports and imports\n\nproposed in this paper will not only remove the ambiguity presently surrounding real GDP <sup>e</sup> and\n\nreal GDP <sup>\n\nMention: \"Penn World Tables\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its expenditure-side real output comparisons build on the UN comparison programme.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003377:2:0", "occurrence_id": "de4b2fdbff15cb22f947ce55", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **1. Introduction**\n\n\nFrom its inception, the Penn World Tables (PWT), building on the International\n\n\nComparisons Program (ICP) of the United Nations, has sought to compare the _standard of living_\n\n\nof individuals in different countries. That is, the term “real GDP per capita” as reported in the\n\n\nPWT is intended to represent the ability to purchase goods and services by a representative agent\n\n\nin the economy. The same is true of benchmark comparisons as published by the United Nations,\n\n\nEurostat or OECD. But this expenditure-side interpretation of real GDP is quite different from\n\n\nthe uses to which benchmark ICP and PWT data are frequently applied, such as in the “millions”\n\n\nof growth regressions (Sala-i-Martin, 1997) where “real GDP” is intended to reflect the\n\nproduction side of the economy. <sup>1</sup> [^1: There are many other applications where expenditure-side real GDP is used when output-side real GDP is really\nneeded. Recent examples are Bernard and Jones (1996), Jones and Hall (1999) and Acemoglu and Ventura (2002).]\n\n\nThe goal of this paper is to carefully distinguish the production or output-side measure of\n\nreal GDP, denoted _real GDP_ <sup>_o_</sup>, from the expenditure-side measure, denoted _real GDP_ <sup>_e_</sup> . Real\n\nGDP <sup>o</sup> reflects the production possibilities of an economy, whereas real GDP <sup>e</sup> measures real\n\n\nexpenditures which includes an adjustment for the trade balance. The reason these concepts were\n\n\nnot distinguished in the ICP and PWT is that these projects treat the net foreign balance in an\n\n\nunsatisfactory way. While there may have been some data justifications for that treatment in\n\n\nbenchmark studies of the 1970s, this is no longer the case. The treatment of exports and imports\n\nproposed in this paper will not only remove the ambiguity presently surrounding real GDP <sup>e</sup> and\n\nreal GDP <sup>\n\nMention: \"PWT data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Their intended interpretation differs from the use made in growth regressions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003385:31:0", "occurrence_id": "99606e75e2101f389e2c34ac", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **_Table 1_**\n\n\nNumber of <u>Growth Rates, 1990–2005</u>\nCountries Pop. (million) PPP GDP GDP per capita    Exports\n\nUSA 1 296 3.0 1.9 4.7\n\nJapan 1 128 1.3 1.0 3.8\n\nRest of high\nIncome 22 496 2.3 1.8 5.4\n\nChina 1 1304 9.6 8.7 15.0\n\nIndia 1 1095 5.7 4.0 8.9\n\nRest of dev\nAsia (RODA) 17 843 4.5 2.7 8.5\n\nLatin America 29 539 3.0 1.4 7.9\n\nSub-Saharan\nAfrica 41 711 2.7 0.2 3.9\n\nEastern Europe/ 14\nCentral Asia 324 0.2 0.1 8.6\n\nMiddle East/\nN Africa 11 275 3.9 2.0 5.9\n\nWorld 138 6012 3.4 2.0 6.1\n\n\n<u>Sources: World Development Indicators; World Bank WITS database.</u>\n\n\n31\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Regional trade aggregates are drawn from this compilation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003385:31:0", "occurrence_id": "88acae4ba8551e1c68823183", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **_Table 1_**\n\n\nNumber of <u>Growth Rates, 1990–2005</u>\nCountries Pop. (million) PPP GDP GDP per capita    Exports\n\nUSA 1 296 3.0 1.9 4.7\n\nJapan 1 128 1.3 1.0 3.8\n\nRest of high\nIncome 22 496 2.3 1.8 5.4\n\nChina 1 1304 9.6 8.7 15.0\n\nIndia 1 1095 5.7 4.0 8.9\n\nRest of dev\nAsia (RODA) 17 843 4.5 2.7 8.5\n\nLatin America 29 539 3.0 1.4 7.9\n\nSub-Saharan\nAfrica 41 711 2.7 0.2 3.9\n\nEastern Europe/ 14\nCentral Asia 324 0.2 0.1 8.6\n\nMiddle East/\nN Africa 11 275 3.9 2.0 5.9\n\nWorld 138 6012 3.4 2.0 6.1\n\n\n<u>Sources: World Development Indicators; World Bank WITS database.</u>\n\n\n31\n\nMention: \"World Bank WITS database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It supplies the companion trade statistics.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003387:23:1", "occurrence_id": "0d804f6605cead5e2b8c7974", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  the industrial\n\nproduction index and its trend, seasonal and irregular\n\ncomponents using a structural time series model.\n\n\n\n**Inflation Gap** Deviation of actual Inflation from Inflation target. Chile: The monthly inflation target\nseries provided by Gallego at al.\n(2002).\nColombia: Central Bank Website\n(www.banrep.gov.co).\n**Real Exchange** Chilean Central Bank Website\n\n\n\n**Real Exchange** Chilean Central Bank Website\n\n**Rate** (www.bcentral.cl) and Colombian\n\nCentral Bank Website\n(www.banrep.gov.co).\n**Real Exchange** Cyclical Component of Real Exchange Rate, obtained Author’s Calculations.\n\n\n\n**Rate**\n\n\n\nAuthor’s Calculations.\n\n\n\n**Rate Gap**\n\n\n**Country Risk**\n\n**Premium**\n\n\n**Terms of**\n\n**Trade**\n\n\n\nCyclical Component of Real Exchange Rate, obtained\nby fitting a structural time series model (Kalman Filter)\n\n\n\nto real exchange rate series.\n\nChile: premium on international bonds issued by\nChilean corporations. Colombia: Series constructed\n\nusing spread of Colombian bonds in international\nmarkets over US T-bills. Updated using EMBI+ until\n\n2002\n\n\n\nCentral Bank of Chile, Rincon &\n\nVillar (2000) & J.P. Morgan,\n\nobtained from Thomson’s\n\nDatastream.\n\n\nBennett & Valdes (2001) &\nColombian Central Bank Website\n\n\n\n(www.banrep.gov.co).\n**Junk Bond** Difference in Yields between US High-Yield Bonds Merrill Lynch **,** obtained from\n\n\n\nDifference in Yields between US High-Yield Bonds\n\n\n\nMerrill Lynch **,** obtained from\n\nThomson’s Datastream.\n\n\nIMF’s International Financial\n\n\n\n**Spread**\n\n\n\nand 10-Year US Treasury Bonds. A number of\ndifferent High-Yield indexes where used such as\n\nMaster II (H0A0) and High Yield 175 (X0A\n\nMention: \"monthly inflation target\nseries\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Published targets define the inflation gap for one country in the sample.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003387:23:1", "occurrence_id": "66993c93d554056fad266df5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  the industrial\n\nproduction index and its trend, seasonal and irregular\n\ncomponents using a structural time series model.\n\n\n\n**Inflation Gap** Deviation of actual Inflation from Inflation target. Chile: The monthly inflation target\nseries provided by Gallego at al.\n(2002).\nColombia: Central Bank Website\n(www.banrep.gov.co).\n**Real Exchange** Chilean Central Bank Website\n\n\n\n**Real Exchange** Chilean Central Bank Website\n\n**Rate** (www.bcentral.cl) and Colombian\n\nCentral Bank Website\n(www.banrep.gov.co).\n**Real Exchange** Cyclical Component of Real Exchange Rate, obtained Author’s Calculations.\n\n\n\n**Rate**\n\n\n\nAuthor’s Calculations.\n\n\n\n**Rate Gap**\n\n\n**Country Risk**\n\n**Premium**\n\n\n**Terms of**\n\n**Trade**\n\n\n\nCyclical Component of Real Exchange Rate, obtained\nby fitting a structural time series model (Kalman Filter)\n\n\n\nto real exchange rate series.\n\nChile: premium on international bonds issued by\nChilean corporations. Colombia: Series constructed\n\nusing spread of Colombian bonds in international\nmarkets over US T-bills. Updated using EMBI+ until\n\n2002\n\n\n\nCentral Bank of Chile, Rincon &\n\nVillar (2000) & J.P. Morgan,\n\nobtained from Thomson’s\n\nDatastream.\n\n\nBennett & Valdes (2001) &\nColombian Central Bank Website\n\n\n\n(www.banrep.gov.co).\n**Junk Bond** Difference in Yields between US High-Yield Bonds Merrill Lynch **,** obtained from\n\n\n\nDifference in Yields between US High-Yield Bonds\n\n\n\nMerrill Lynch **,** obtained from\n\nThomson’s Datastream.\n\n\nIMF’s International Financial\n\n\n\n**Spread**\n\n\n\nand 10-Year US Treasury Bonds. A number of\ndifferent High-Yield indexes where used such as\n\nMaster II (H0A0) and High Yield 175 (X0A\n\nMention: \"EMBI+\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its spreads update the sovereign risk measure.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003387:23:1", "occurrence_id": "827000bfe64bfaff97a06767", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  the industrial\n\nproduction index and its trend, seasonal and irregular\n\ncomponents using a structural time series model.\n\n\n\n**Inflation Gap** Deviation of actual Inflation from Inflation target. Chile: The monthly inflation target\nseries provided by Gallego at al.\n(2002).\nColombia: Central Bank Website\n(www.banrep.gov.co).\n**Real Exchange** Chilean Central Bank Website\n\n\n\n**Real Exchange** Chilean Central Bank Website\n\n**Rate** (www.bcentral.cl) and Colombian\n\nCentral Bank Website\n(www.banrep.gov.co).\n**Real Exchange** Cyclical Component of Real Exchange Rate, obtained Author’s Calculations.\n\n\n\n**Rate**\n\n\n\nAuthor’s Calculations.\n\n\n\n**Rate Gap**\n\n\n**Country Risk**\n\n**Premium**\n\n\n**Terms of**\n\n**Trade**\n\n\n\nCyclical Component of Real Exchange Rate, obtained\nby fitting a structural time series model (Kalman Filter)\n\n\n\nto real exchange rate series.\n\nChile: premium on international bonds issued by\nChilean corporations. Colombia: Series constructed\n\nusing spread of Colombian bonds in international\nmarkets over US T-bills. Updated using EMBI+ until\n\n2002\n\n\n\nCentral Bank of Chile, Rincon &\n\nVillar (2000) & J.P. Morgan,\n\nobtained from Thomson’s\n\nDatastream.\n\n\nBennett & Valdes (2001) &\nColombian Central Bank Website\n\n\n\n(www.banrep.gov.co).\n**Junk Bond** Difference in Yields between US High-Yield Bonds Merrill Lynch **,** obtained from\n\n\n\nDifference in Yields between US High-Yield Bonds\n\n\n\nMerrill Lynch **,** obtained from\n\nThomson’s Datastream.\n\n\nIMF’s International Financial\n\n\n\n**Spread**\n\n\n\nand 10-Year US Treasury Bonds. A number of\ndifferent High-Yield indexes where used such as\n\nMaster II (H0A0) and High Yield 175 (X0A\n\nMention: \"Thomson’s\n\nDatastream\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The underlying market series were obtained through this vendor.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003389:9:1", "occurrence_id": "42959d46cdf4d6964e308b31", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ||||||\n|||||||||\n|||||||||\n|||||||||\n|||||||||\n|||||||||\n\n\n\nage\n\n\n\ngirls\n\n\n\nFigure 2 - Distribution of children according to school attendance and current work status, by age\nand gender.\nSource: Own calculations based on World Bank, Fortaleza Survey 2003.\nNote: There is one other category—attending school only—which is not plotted.\n\n\nFigure 2 highlights that, after the age of 16, there is an increasing share of teenagers who, though\n\n\nnot having a job, drop out of school. By the age of 18, almost half of the teenagers is neither in\n\n\nschool nor at work. These figures cast some doubt on the relevance of child labor as the driving\n\n\nforce pulling teenagers out of school in urban Brazil.\n\n\n9\n\nMention: \"Fortaleza Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Attendance and work status by age and gender are tabulated from it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003394:7:0", "occurrence_id": "d8a644f34fd09a8e5a307e09", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Table 2. Descriptive Statistics: Bosnia and Herzegovina, 2001-2004\n\n(In units as indicated)\n\n\nIn percent In percent\n<u>of the labor force</u> <u>of the working age population</u>\n2001 2002 2004 2001 2002 2004\n\n\nMen 65 64 63 49 49 50\nAge Groups\n\nAge 15-24 14 15 15 22 23 19\nAge 25-44 52 51 50 41 40 41\nAge 45-54 26 25 26 21 21 23\nAge 55+ 8 9 9 17 17 17\nEducational Attainment\n\nUnfinished Elementary 7 7 6 13 13 12\nElementary 21 22 22 31 31 28\nVocational 42 40 38 32 31 31\nHigh School 20 21 24 18 19 23\nCollege (2+ years) 5 4 4 3 3 3\nUniversity 5 5 5 3 3 3\nUnemployment Rate 16 22 22\nLabor Force Participation 48 53 59\n\n\n_Source_ : LSMS and authors' calculations.\n\n\nTable 2 reports selected descriptive statistics of the sample used in our empirical\nanalysis. The distributions of age, gender and skills are presented for each year (2001,\n2002 and 2004). The labor force participation rate increased from 40 percent in 2001 to\n59 percent in 2004. Despite this increase, the participation rate in Bosnia and\nHerzegovina is low compared to other ECA countries, mostly due to the very low labor\nforce participation rate among women. Average unemployment rate is high, and\nincreased from 16 percent in 2001 to 22 percent in 2004.  The characteristics of the work\nforce (and the working population) are fairly stable, where youth represent around 22\npercent of the working population (the decline from 2001 to 2004 is due to the ageing for\nthe panel). The educational attainment is relatively high, with over 70 percent with\nvocational education, secondary education or more. Interestingly the level of secondary\neducation has increased significantly over the sample years, offset by a corresponding\ndecrease in vocational education.\n\n\nWhat do we know of youth labor market transitions in the ECA region? Appendix\nTable 1 reports summary data\n\nMention: \"LSMS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Education, unemployment and participation statistics are computed from it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003396:27:1", "occurrence_id": "ed3d3ed43f085cd884a4c428", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 0.09)_ <br> <br> <br>.407<br>(_0.04)_ <br>.597<br>(_0.20)_ <br>.680<br>(_0.32)_ <br>1.298<br>(_0.09)_ <br>1.224<br>(_0.15)_ <br>1.049<br>(_0.94)_ <br>.883<br>(_0.83)_ <br>.536<br>(_0.38)_ <br>.656<br>(_0.50)_ <br>2.399<br>(_0.41)_ <br>2.093<br>(_0.45)_|\n|_N (_country-years_) _<br>Log-likelihood<br>Pseudo R2|1553<br>1678<br>-202.658<br>-215.868<br>0.038<br>0.040|1019<br>1058<br>-149.379<br>-156.230<br>0.041<br>0.046|\n\n\nNote: Logistic estimation with clustered standard errors. All time-varying independent\nvariables lagged one year. Odds-ratios reported with _p_ -values in parentheses: _coefficient values_\n_below one indicate the odds of conflict increase with increases in the corresponding variable_ . _p_ -values for\nyears and party age calculated assuming the interaction term is zero. Political variables from\nthe Database of Political Institutions, conflict variables from Correlates of War, and control\nvariables from Fearon and Laitin (2003). All variables are\n\nMention: \"Database of Political Institutions\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its political variables enter the estimation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003396:27:1", "occurrence_id": "e1e666e7026002accaab40c6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 0.09)_ <br> <br> <br>.407<br>(_0.04)_ <br>.597<br>(_0.20)_ <br>.680<br>(_0.32)_ <br>1.298<br>(_0.09)_ <br>1.224<br>(_0.15)_ <br>1.049<br>(_0.94)_ <br>.883<br>(_0.83)_ <br>.536<br>(_0.38)_ <br>.656<br>(_0.50)_ <br>2.399<br>(_0.41)_ <br>2.093<br>(_0.45)_|\n|_N (_country-years_) _<br>Log-likelihood<br>Pseudo R2|1553<br>1678<br>-202.658<br>-215.868<br>0.038<br>0.040|1019<br>1058<br>-149.379<br>-156.230<br>0.041<br>0.046|\n\n\nNote: Logistic estimation with clustered standard errors. All time-varying independent\nvariables lagged one year. Odds-ratios reported with _p_ -values in parentheses: _coefficient values_\n_below one indicate the odds of conflict increase with increases in the corresponding variable_ . _p_ -values for\nyears and party age calculated assuming the interaction term is zero. Political variables from\nthe Database of Political Institutions, conflict variables from Correlates of War, and control\nvariables from Fearon and Laitin (2003). All variables are\n\nMention: \"Correlates of War\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its conflict coding supplies the war variables.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003402:11:0", "occurrence_id": "d17af6ff9dedc2c6c2bc20d6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: belligerents. <sup>10</sup> [^10: In the Binningsbø, Elster & Gates (2005) dataset there are 291 conflict periods, of which, 105 reoccur, indicating a\nrelatively high risk of peace failure.] To account for the obvious non-independence of peace events we calculate robust\n\n\nstandard errors. A number of wars and post-conflict peace periods were still on-going in 2003 (the\n\n\nlast year of our data); we censored these cases.\n\n##### **Data and operationalizations**\n\nThe Binningsbø, Elster & Gates (2005) data provide an overview of how post-conflict societies\n\n\ndeal/have dealt with the wrongdoings of their recent pasts. In order to define a ‘post-conflict\n\n\nsociety’ they rely on the Uppsala-PRIO armed conflict dataset (Gleditsch et al., 2002), which\n\n\ndefines a conflict as ‘a contested incompatibility that concerns government and/or territory where\n\n\nthe use of armed force between two parties, of which at least one is the government of a state,\n\nresults in at least 25 battle-related deaths’ (Strand, Wilhelmsen & Gleditsch, 2004: 3). <sup>11</sup> [^11: Version 3.0 (December 2004).] To allow\n\n\nfor time-varying covariates the dataset is disaggregated into ‘peace years’ for each individual\n\n\npost-conflict period. The first ‘peace year’ for each subject starts the first day of peace after\n\n\nconflict and ends 31st December the same year. Next ‘peace year’ starts 1st January the following\n\n\nyear and ends 31st December, or the date the conflict resumes (failure). The subject enters the\n\n\ndataset again the first day of peace after (if) the resumed conflict ends. Post-conflict peace periods\n\n\nthat did not fail (no resumed conflict) before the last day of observation in the dataset (31\n\n\nDecember 2003) are right-censored. Begin and end dates of the conflicts and peace transition\n\n\nperiods are based on Gates & Strand’s (2004) event dataset. Following Gates &\n\nMention: \"Uppsala-PRIO armed conflict dataset\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its conflict definition is used to identify post-conflict societies.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003404:14:0", "occurrence_id": "8fab93fabf4c3cceec2ef440", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: #### _4.3 Control Variables and Statistical Model_\n\nAchen’s (2002) ‘Rule of Three’, states that every analysis with more than three variables on the\nright-hand side will invariably be invalidated by serious problems of multicollinearity. I do not\nadhere to this rule, but believe that it is wise to keep the control variables at a minimum,\nespecially given the limited sample size under study.\n\nAs noted by Hegre & Sambanis (2005) three core variables are almost always included in\nmodels of civil war onset: the natural log of population; the natural log of per capita GDP; and\nthe length of peacetime until the outbreak of a war (i.e. the time since the last conflict). I include\nthese three controls, of which the first is particularly relevant due to potential problems of\nspuriousness, given that inequality may be related to the actual level of economic development in\na society. Data on population size stem from the WDI (World Bank, 2004). The variable is\ninterpolated and log-transformed. To proxy economic development I use log-transformed GDP\nper Capita measured in constant 1995 US$, also from the WDI (World Bank, 2004). The variable\nis lagged one year.\n\nAs suggested by Beck, Katz & Tucker (1998), I control for temporal dependence through\na variable measuring time since the last conflict and three cubic splines that approximate the\ndiscrete time hazard rates for conflict. Time since the last conflict counts the number of whole\nyears since the end of the last conflict (peace years). The statistical tests were conducted using\nSTATA, version 8.2 (StataCorp., 2003), and all models were estimated by logit regressions with\nrobust standard errors clustered by countries.\n\n### **5 Results**\n\n\nThe findings from the empirical tests of theoretical propositions are presented in Tables 1–4. All\nmodels\n\nMention: \"WDI\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Population size is taken from it, then interpolated and logged.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003413:2:0", "occurrence_id": "1862c1e6cf262593d2f394dc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: the arms supply in a particular country, i.e. the more easily individual combatants can obtain\n\n\nweapons through independent suppliers, the more difficult it will be to mount and maintain a\n\n\nunited and coordinated insurgency.\n\n\nThere are a number of reasons why small arms have been all but ignored in the quantitative\n\n\nanalysis of conflict. The historic state-centric bias of defense economics led to an almost\n\n\nexclusive focus on inter-state military strategy. In relation to military weapons, research has\n\n\nprincipally been concerned with the development and acquisition of large-scale military\n\n\ntechnology, such as nuclear weapons. Perhaps the most important reason for the dearth of\n\n\nattention given to the role of weapons in civil war is that usable data have been unavailable. The\n\n\npolicy research community, led by the Small Arms Survey (SAS), the UN’s Small Arms and\n\n\nDemobilisation Unit, the Bonn International Center for Conversion, and the Norwegian Initiative\n\n\non Small Arms Transfers (NISAT), has produced a great deal of survey and case-study work.\n\n\nHowever, no statistical analysis of the growing volume of survey information has yet taken\n\n\nplace.\n\n\n**DATA**\n\n\nExisting data on aspects of the small arms market are extremely limited. Since 2001, the Small\n\n\nArms Survey has gathered a range of information on small arms products, stockpiles, producers\n\n\nand trade. Despite occasional references to observed prices, the Survey has not regularly\n\n\ncollected price data which would be of most benefit for generating inferential statistics.\n\n\nCollecting price data for panel analysis requires an operational definition of the variable of\n\n\ninterest that will provide consistency across time and countries. In the case of small arms there is\n\n\nan obvious choice: the AK-47 assault rifle. Of the estimated 500 million firearms worldwide,\n\n\napproximately 100 million belong to the Kalashnikov family, three-quarters of which are AK\n\n47s (Small Arms Survey 2004).\n\n\nThe pervasiveness of this weapon may be explained in large part by its simplicity. The AK-47\n\n\nwas initially designed for\n\nMention: \"Small Arms Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Named as a lead source in the community compiling light weapons data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003417:18:0", "occurrence_id": "8dd8ae1ea9c5301a4b3dd81b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Expectedly, the following 16 countries are included in our comparison sample of\n\n\ndeveloping economies for the variables from the PolityIV database (i.e., when the institutional\n\n\nvariables are executive constraints or responsible government): Benin, Chile, Ghana, Hungary,\n\n\nIran, Jordan, Kenya, Mali, Mexico, Nicaragua, Pakistan, Rwanda, Senegal, Syria, Tunisia, and\n\n\nUganda. <sup>6</sup> [^6: In the case of executive constraints, Brazil is in, but Hungary is out.] We use 32 countries for the ICRG-data-based regressions (i.e., when the institutional\n\n\nvariables are the control of corruption <sup>7</sup> [^7: The list of countries when the institution is measured as the rule of law is similar but not identical, and is\navailable upon request.] ): Bangladesh, Bolivia, Cameroon, Colombia, El Salvador,\n\n\nGhana, Guatemala, Haiti, Honduras, Hong Kong, India, Indonesia, Kenya, Mali, Mexico, Niger,\n\n\nPakistan, Panama, Papua New Guinea, Paraguay, Philippines, Sierra Leone, Sudan, Syrian Arab\n\n\nRepublic, Thailand, Togo, Trinidad and Tobago, Turkey, Uganda, Venezuela, Zambia, Zimbabwe.\n\n\nTable 2 reports standard descriptive statistics for these variables across the relevant\n\n\nsamples of countries.\n\n\nTable 3 reports our cross country regression results. These regressions incorporate year\n\n\nfixed effects to absorb global macro factors. Because of expected serial correlations for\n\n\nobservations within the same country, the standard errors are estimated based on country\n\n\nclustering.\n\n\nThe institutional measures of good government that capture the strength of constraints on\n\n\nexecutive power are statistically significant positive determinants of FDI. In contrast, the two\n\n\nvariables related to general government quality (i.e., control of corruption and the rule of law) are\n\n\nnot associated with significantly more FDI. These results show that conditional on government\n\n\ntract record and general development levels, limits on executive power indeed attracts FDI, yet\n\n\nmany other aspects of general government\n\nMention: \"the PolityIV database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its executive constraints and responsible government variables define the comparison sample.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003420:21:1", "occurrence_id": "aac2e9d489d4aa5a5d0a2d6d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  under the BAU\ncases which already incorporate some expansion of non-fossil fuels sectors. The\nadditional improvements include (a) a 20 percent increase in hydroelectric\ncapacity relative to the base case, and (b) a 30 percent increase in the share of\nnuclear power in new investments for power generation, (c) the share of\nbiofuels is progressively increased to 10 percent of the total amount of fuels\nproduced by China and India. The shares of wind and solar energy increase\nsignificantly from a very low base but not enough to offset the reduction in the\nuse of traditional biomass. (d) Energy efficiency is also increased by 15 percent\nin the use of coal for industry and by 8 percent in the use of coal for electricity\ngeneration in the new capital stock installed after 2005.\n\n\n3. **Supply and demand side scenarios** (designated with an **S&D** ) that combine\nthe efficiency improvements and fuel-switching measures above, and are in line\nwith Chinese and Indian energy strategies. (Sarma, Margo, and Sachdeva, 1998;\nLiu, 2003).\n\n\nThe BAU and ALT scenarios are each simulated in two different contexts: (a) the\nbase case used for reference purposes (i.e., BAU and BAU-H) which assumes, perhaps\n\n\n44 The IEA suggests that end use efficiency improvements are the source of the greatest potential in managing energy\ndemand and mitigation of CO<->2<-> emissions. Over the 2002–2030 period, improvements in end-use efficiency could\ncontribute to more than 50 percent in the reduction in emissions for a group of 11 IEA countries (Australia, Denmark,\nFinland, France, Germany, Japan, Italy, Norway, Sweden, U.K. and U.S.) for which IEA has complete time series data\n(see Bradley, 2006).\n45 Note that fuel switching is often also accompanied by simultaneous improvements in energy efficiency.\n\n\n22\n\nMention: \"IEA\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its assessment identifies end-use efficiency as the largest potential.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003433:23:1", "occurrence_id": "15f2fdaada845ace88f7892e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  by design, focuses on medium and large firms with\nover 100 employees, and larger firms tend on average to train more than small ones. The\nBEEPS and ICS surveys, by contrast, tend to include a higher proportion of smaller\nfirms. To compare the incidence of training in Russia to that of OECD and selected\ncomparator developing countries, we adopt a common weighting scheme based\n(arbitrarily) on the size distribution of firms in the India ICS survey. <sup>15</sup> The weighted\nincidence of formal in-service training is calculated for the pooled LME and SE samples\nfrom Russia, and for 16 other comparator countries: 6 from the OECD, 3 from East Asia,\n2 from South Asia, 2 from Latin America, and 3 from Eastern Europe. <sup>16</sup>\n\nFigure 10 shows estimates of training incidence in OECD countries and in four regions,\nand separately for the 17 countries used to calculate regional averages. The selected\n\n14 Olga Lazareva, Irina Denisova, and Serguey Tsukhlo (2006), “Hiring or Retraining: Russian Firms’\nExperience”, HSE Working paper WP3/2006/11, 2006.\n15 The size distribution of micro (15 or fewer workers), small (16-100 workers), medium (101-250 workers)\nand large firms (over 250 workers) in India is 40, 44, 7 and 8 percent; the corresponding size distribution\nfor the pooled LME and SE surveys in Russia are 12, 16, 29 and 43 percent\n16 The 17 countries clustered by region include Germany, Greece, Ireland, Portugal, South Korea and Spain\n(OECD), China, Malaysia and Thailand (East Asia), India and Sri Lanka (South Asia), Brazil and Chile\n(Latin America), Bulgaria, Lithuania, Serbia and Russia (Eastern Europe).\n\n\n24\n\nMention: \"ICS surveys\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Their samples include more small firms than the alternative survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003433:23:1", "occurrence_id": "3dedd72252e4342ab06b27b2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  by design, focuses on medium and large firms with\nover 100 employees, and larger firms tend on average to train more than small ones. The\nBEEPS and ICS surveys, by contrast, tend to include a higher proportion of smaller\nfirms. To compare the incidence of training in Russia to that of OECD and selected\ncomparator developing countries, we adopt a common weighting scheme based\n(arbitrarily) on the size distribution of firms in the India ICS survey. <sup>15</sup> The weighted\nincidence of formal in-service training is calculated for the pooled LME and SE samples\nfrom Russia, and for 16 other comparator countries: 6 from the OECD, 3 from East Asia,\n2 from South Asia, 2 from Latin America, and 3 from Eastern Europe. <sup>16</sup>\n\nFigure 10 shows estimates of training incidence in OECD countries and in four regions,\nand separately for the 17 countries used to calculate regional averages. The selected\n\n14 Olga Lazareva, Irina Denisova, and Serguey Tsukhlo (2006), “Hiring or Retraining: Russian Firms’\nExperience”, HSE Working paper WP3/2006/11, 2006.\n15 The size distribution of micro (15 or fewer workers), small (16-100 workers), medium (101-250 workers)\nand large firms (over 250 workers) in India is 40, 44, 7 and 8 percent; the corresponding size distribution\nfor the pooled LME and SE surveys in Russia are 12, 16, 29 and 43 percent\n16 The 17 countries clustered by region include Germany, Greece, Ireland, Portugal, South Korea and Spain\n(OECD), China, Malaysia and Thailand (East Asia), India and Sri Lanka (South Asia), Brazil and Chile\n(Latin America), Bulgaria, Lithuania, Serbia and Russia (Eastern Europe).\n\n\n24\n\nMention: \"India ICS survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Its firm size distribution supplies the common weighting scheme.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003434:9:1", "occurrence_id": "e29fcecdfb7cd3fa241c78ec", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 11,076|200,870|57,420|43,623|15,846|\n|Scientific and Technical Journal<br>Articles / Mil. Pop.|16.5|10.7|704.0|451.6|529.8|109.5|\n\n\nSource: WBI KAM database.\n\n\n\n\n\nA.3.3. Strong business participation in R&D\n\n\nEnterprises are now becoming big players in both financing and conducting R&D\n\n\nactivities in China (tables 6 and 7). In 2003, the business sector accounted for 62.4% of\n\n\ntotal R&D funding. However, it should be noted that public enterprises undertake most of\n\n\nthe R&D in the business sector - roughly 50% of all R&D in China. This means that the\n\n\namount being done by non-public enterprises is just about 10% of the total R&D effort,\n\n\nand it appears that a large share of this is being done by multinational companies\n\n\n(MNCs), not by domestic Chinese firms. The total R&D input in 2004 by state-owned\n\n\nenterprises, supervised by the Central State Assets Supervision and Administration\n\n\nCommission (SASAC), came to 76.8 billion yuan, accounting for 1.5% of firms’ annual\n\n\n10\n\nMention: \"WBI KAM database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Research output indicators per capita are drawn from this knowledge database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003440:4:1", "occurrence_id": "ac4ced92a38cc46d9a46920b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  22 percent of Mexican microenterprises with less\n\n\nthan US$1000 of non-land capital investment keep records, according to the 1998\n\nENAMIN survey. <sup>3</sup> Studies during the 1980s found only 17 percent of small industry\n\n\nowners kept records in Sierra Leone, and less than 10 percent in Jamaica and rural\n\n\nBangladesh (Liedholm, 1991). It is therefore of general interest to determine what can be\n\n\ndone to elicit reasonable information on profits from small firm owners.\n\n\n2 Fieldwork was carried out by the Kandy Consulting Group.\n3 See McKenzie and Woodruff (2006) for more details on this survey.\n\n\n5\n\nMention: \"the 1998\n\nENAMIN survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"It records that few microenterprises with small capital keep records.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003441:11:0", "occurrence_id": "200909dd745f20e4459a7b50", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: the change in monthly profits from one quarter to the next was less than -90 percent or\n\ngreater than 962 percent, corresponding to the 1 <sup>st</sup> and 99 <sup>th</sup> percentiles of the distribution\n\n\nof changes in profits. The estimated nominal return on the treatment is 5.9 percent per\n\n\nmonth after this trimming.\n\n\nInflation in Sri Lanka was low over the period studied, with point-to-point\n\n\ninflation of 4.0 percent between March 2005 and March 2006. The second row of Table 2\n\n\nexamines the treatment effect on real profits, where the monthly Sri Lanka Consumers’\n\nPrice Index was used to deflate nominal profits. <sup>6</sup> [^6: Source: Sri Lanka Department of Census and Statistics,\n<u>http://www.statistics.gov.lk/price/slcpi/slcpi_monthly.htm</u> [accessed February 17, 2007].\n<u>7 Recall that none of the firms had paid non-family workers in the baseline, and less than 5 percent of firms</u>\nhad non-family workers at any time during the panel survey.] After trimming based on extreme\n\n\nchanges in profits, we see that the real monthly increase in profits was 5.7 percent of the\n\n\ntreatment amount.\n\n\n_Impact on Labor Hours_\n\n\nThis treatment effect represents the increase in real monthly profits from our\n\n\ntreatment. The treatment increased capital stock. However, labor hours may have also\n\n\nhave been affected if firm owners adjusted their own labor hours, or if the resulting rise\n\n\nin profits caused owners to substitute towards leisure. We therefore empirically\n\n\ninvestigate the impact of the treatment on labor hours by estimating equation (1), using\n\nown labor hours and family worker labor hours as the dependent variables. <sup>7</sup> The\n\n\ntreatment is found to have a positive and significant impact on own labor hours: a 10,000\n\n\nrupee treatment increases own weekly hours by 3.4 hours (t\n\nMention: \"Sri Lanka Consumers’\n\nPrice Index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Monthly movements in this index deflate nominal profits.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003444:38:1", "occurrence_id": "2afb2493dba64bf31958d4e9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 923.86 61019.07\n\n_Investment_ 1980-1999 0.50 26462.34\nBefore (1980-1991) 1265.90 11323.9\nAfter (1992-1999) -1830.70 39024.2\n\n_Average Tariffs_ 1980-1999 100.27 50.11\nBefore (1980-1991) 153.13 25.04\nAfter (1992-1999) 65.09 25.83\n\n_Average Non-Tariff Barriers_ 1980-1999 66..88 21.99\nBefore (1980-1991) 85.35 14.52\nAfter (1992-1999) 56.44 19.44\n\n_Output Share of Delicensed Industries_ 1980-1999 53.50 45.20\nBefore (1980-1991) 28.01 38.55\nAfter (1992-1999) 87.35 27.83\n\n_Share of Export>Import Industries_ 1991 0.55 0.48\n\nNotes:\n1) Output and Net Value Added are deflated by WPI obtained from Reserve Bank of India Database on Indian Economy at\nhttp://www.rbi.org.in/. Base year for WPI is 1981.\n2) Development expenditures are deflated by the GDP deflator.\n3) Number of workers includes direct and contracted employment.\n\n\n38\n\nMention: \"Reserve Bank of India Database on Indian Economy\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Its wholesale price series with that base year deflates output and value added.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003446:30:1", "occurrence_id": "622ca1bda6a8edfd0575b1e3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The index ranges from 0 to 6, with a score of 1 assigned for each of<br>the following 6 features of the credit information system:<br>. Both positive and negative credit information is distributed.<br>. Data on both firms and individuals are distributed.<br>. Data from retailers, trade creditors or utilities as well as financial<br>institutions are distributed.<br>. More than 2 years of historical data are distributed.<br>. Data on loans above 1% of income per capita are distributed.<br>. By law, borrowers have the right to access their data.<br>Accounting standard ratings in 1990 (0 worst-100 best). Source: LLSV<br>(1998)|\n|Legal<br>Institutions|D Common Law<br>Shareholder Rights<br>Creditor Rights<br>Corruption|Dummy =1 if legal origin is common law, and =0 if the origin is civil law<br>or socialist. Source: LLSV (1998)<br>Shareholder right index (i.e. new anti-director index) (0 least - 6 most<br>rights). Source: LLSV (1998)<br>Creditor right index (0 least - 4 most rights). Source: LLSV (1998) and<br>Djankov, McLiesh and Shleifer (2006)<br>Corruption perception index (reversed: 0 least - 10 most corrupted).<br>Source: Transparency International|\n|Financial<br>Development|Deposit /GDP<br>Private Bond /GDP<br>Bank Costs /Assets<br>D Financial Center<br>D Banking Crisis|Demand, time and saving deposits in deposit money banks and other<br>financial institutions as a share of GDP. Source:\n\nMention: \"Corruption perception index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Its reversed scale enters as the corruption control.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003451:53:2", "occurrence_id": "b815b5c41bbe4208842f063d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  household final consumption; investment = gross capital formation; trade balance = exports\n\nof goods and services - imports of goods and services. All variables except net exports and government debt are expressed in per capita terms and in\n\nlogaritms; all variables filtered with the Hodrick-Prescott filter. Persistence is defined as first order autocorrelation.\n\nTax data are from the IMF Government Finance Statistics database. Government debt figures are constructed as indicated in the text. The remaining\n\ndata are taken from the World Development Indicators database.\n\n\n54\n\nMention: \"IMF Government Finance Statistics database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Tax revenue series are taken from this compilation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003451:53:2", "occurrence_id": "97ff0a269e8ada6cb9e7e159", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  household final consumption; investment = gross capital formation; trade balance = exports\n\nof goods and services - imports of goods and services. All variables except net exports and government debt are expressed in per capita terms and in\n\nlogaritms; all variables filtered with the Hodrick-Prescott filter. Persistence is defined as first order autocorrelation.\n\nTax data are from the IMF Government Finance Statistics database. Government debt figures are constructed as indicated in the text. The remaining\n\ndata are taken from the World Development Indicators database.\n\n\n54\n\nMention: \"World Development Indicators database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The remaining macro variables come from this compilation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003461:6:0", "occurrence_id": "64bc8ea15a911ecc76c3b945", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The explanatory variables for the core specification of the probability of civil wars includes\n\n\nthe log of real GDP per capita in 1960 (lgdp60) and the log of the population in 1960 (lpop60). We\n\n\ncalculate the mean of the institutional variable of the period in which data is available. In the case\n\n\nof the ICRG, the period is 1984-1999 and for the protection against expropriation risk we have the\n\n\naverage for the 1985-1995 period. We first show some preliminary results using the logit\n\n\nspecification, and then we analyze the relationship between institutions and civil war addressing the\n\n\nendogeneity problem between institutions and conflict, using the IV-2SLS approach.\n\n\nWe first present the estimation of a logit model for the probability of civil wars, using a\n\n\nsample of 211 countries. The logit equation is the following:\n\n\nPr _ob_ ( _conflicti_ 60 - 05 = )1 = Λ ( α+ β 1 lg _dpi_ 60 + β 2 _lpopi_ 60 + β 3 _av_ _ _inst_ ),\n\n\nwhere _conflict_ is a dummy variable that has a value of 1 if the country had a civil war during the\n\n\nperiod 1960-2005 and zero otherwise, α is a constant _, lgdp_ is the log of real per capita income in\n\n\n1960, _lpop_ is the log of the population of the country in 1960, _av_inst_ is the average of the\n\n\ninstitutional measure between 1985 and 1995 for the protection against expropriation risk, between\n\n\n1984 and 1999 for the ICRG law and order measure. The ideal set-up would be to have information\n\n\non the quality of institutions before 1960, however data are not available for that period. We\n\n\naddress the endogeneity issue in the next section.\n\n\nIn column 1 of Table 3 we analyze the effect of the two core variables, population and per\n\n\ncapita income on the probability of civil war\n\nMention: \"ICRG\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Its values are averaged over the period for which data exist.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003463:18:1", "occurrence_id": "d23e13c24ea6deb6834ef90a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .22)\n\n\n\nOil Revenue\n\n\nConstant\n\n\n\n1.45\n\n\n(9.87)\n\n\n1.76\n\n\n(0.83)\n\n\n\nNo. of Obs. 36 36 36\n\n\nR-sq. overall 0.73 0.75 0.53\n\n\nDW 1.13 1.46 0.73\n\n\nNote: 1. Data source: IMF World Economic Outlook database. 2. t-statistics in parentheses.\n\n\nWe included a dummy term for the period up to 1984 and now also include asymmetric adjustment\n\n\ncoefficients: a dummy that equals zero in years of rising revenue and one in years of falling revenue is\n\n\nentered multiplicatively with the oil revenue term. One should expect a negative coefficient under the\n\n\nVoracity hypothesis, and a positive coefficient under the DO hypothesis. Table 2 lists the results.\n\n\n19\n\nMention: \"IMF World Economic Outlook database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"All regression inputs are drawn from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003465:72:0", "occurrence_id": "5a02f372f02d531f0f32ce15", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Source: Customs Adviser to the TTFSE Project and Authors’ Calculation. (2002_\n\n_survey one month period)_\n\n\n**3.** **USA Land Borders**\n\n|Border Post|State of Entry|Baseline|Average|95th Percentile|\n|---|---|---|---|---|\n|Ambassador bridge<br>Blaine Crossing<br>Blue Water<br>Peace Bridge<br>World Trade|Michigan<br>Washington<br>Michigan<br>New York<br>Texas|5.7<br>8.1<br>11.1<br>8.3<br>12.2|8.8<br>17.3<br>34.2<br>21.5<br>31.2|13.9<br>35.6.3<br>80.3<br>83.4<br>54.9|\n\n\n\n_Time in minutes_\n_Source: DOT (2002 survey). Office of Freight Management ._\n\n\n\n73\n\nMention: \"DOT (2002 survey)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Transport times by corridor are compiled from this freight survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003469:12:0", "occurrence_id": "fd586ba0ae87dc26e2f3ee16", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: continuously augment the stock of knowledge and ensure a sufficient flow of innovations\n\n\n(see Table 3).\n\n\n**Table 3 – R & D Expenditures and Sectoral Distribution in 2004**\n\n\n\n\n\n\n\n|Country|Share of R&D by Sector|Col3|Col4|GERD as percent of<br>GDP|\n|---|---|---|---|---|\n|Country|Industry|Higher Education|Government|Government|\n|Finland<br>Japan<br>Korea<br>United States<br>EU-25<br>China|70.1<br>75.2<br>76.7<br>70.1<br>63.3<br>66.8|19.8<br>13.4<br>9.9<br>13.6<br>22.1<br>10.2|9.5<br>9.5<br>12.1<br>12.2<br>13.4<br>23.0|3.51<br>3.13<br>2.85<br>2.68<br>1.81<br>1.23|\n\n\nSource: OECD Science, Technology and Industry Outlook 2006.\n\n\nGovernments have tended to shoulder more of the financing burden for basic\n\n\nresearch in universities and research institutes which adds to the pool of scientific\n\n\nknowledge <sup>29</sup> [^29: Some of the key research underlying innovation by Pharmaceutical industry in the U.S. was done in\npublic labs or with government fundings.] although they do support downstream and applied research by corporations\n\n\nas well. In the majority of cases, resources from the public sector (including for\n\n\nuniversity based research) account for between 20 and 40 percent of spending on R&D\n\n\n(Table 3). The balance comes from the private sector\n\nMention: \"OECD Science, Technology and Industry Outlook 2006\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Research financing shares by source are reported in this outlook.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003471:19:0", "occurrence_id": "92d45978ed9bcb1b4276d2e9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: An idiosyncratic factor that contributed to the overall decrease (as a proportion of\nGDP) in the banking system’s commercial credit portfolio has been the ‘de-leveraging’ of\nBanamex. Grupo Financiero Banamex was purchased by Citigroup in August 2001. At\naround the same time, Banamex began to significantly lower its commercial credit\nexposure, a process that lasted until mid-2004 (see figure 20). Banamex’s credit portfolio\nmovement contrasts sharply to the rest of the banking system which grew by almost 40\npercent in real terms since 2000, thereby halving its share of performing loans to 7\npercent by end-2005.\n\n\nFigure 20: Real Growth in Bank Commercial Credit for Banamex Versus Rest of\n\nBanking System (2000-2005)\n\n\n\n50%\n\n40%\n30%\n20%\n10%\n0%\n\n-10%\n-20%\n\n-30%\n-40%\n-50%\n\n\n\n\n\n<mark>Total Banks (ex Banamex)</mark> <mark>Banamex</mark>\nSource: BOM, CNBV and INEGI.\nNote: Some numbers might differ from other Figures because they are based on CNBV data (e.g.\nperforming loans to the private sector shown above exclude the FOBAPROA portfolio) and are deflated\nusing the retail price index as opposed to the GDP deflator.\n\n\nCorporate insolvency procedures have been significantly improved, but the extent of\ntheir contribution to greater bank lending remains unclear. A new law in 2000 (see Box\n1) radically changed the commercial insolvency (“ _concurso mercantil_ ”) procedural\nframework and eliminated the incidence of suspension of payments by insolvent debtors\nthat was prevalent and that delayed enforcement proceedings indefinitely <sup>17</sup> [^17: See Sánchez-Mejorada y Velasco C. (2004 and 2005), Méjan Carrer L.M. (2005), Gamboa-Cavazos M.\n(January 2005) and Martínez, Algaba, Estrella, de Haro y Galván-Duque (2005).] . The record\nto-date under the new law has been generally positive (around 40 percent of filed cases\nhave already been concluded and out-of-court negotiated\n\nMention: \"CNBV data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its supervisory figures exclude a large bad-debt portfolio and are used for the loan series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003473:28:1", "occurrence_id": "59b528f867db319f80bf84a9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  13.1 1.4 1.9 96.7\nLiberia 400 959 3.2 2.0 3.0 0.9 0.3 98.9\nMali 64 327 0.5 0.7 5.6 32.1 3.9 64.0\nNiger 53 222 0.4 0.5 4.9 33.4 0.3 66.3\nNigeria 9,219 33,209 73.0 70.7 4.4 1.8 0.2 98.1\nSenegal 391 1,269 3.1 2.7 4.0 10.3 7.2 82.5\nSierra Leone 146 185 1.2 0.4 0.8 2.0 0.0 98.0\nTogo 188 554 1.5 1.2 3.7 2.3 0.3 97.4\n\nMemorandum items:\n\nWAEMU countries 2/ 1,982 9,752 15.7 20.8 5.5 11.1 4.0 84.8\nNon-WAEMU countries 3/ 10,648 37,239 84.3 79.2 4.3 1.6 0.3 98.1\n\n\n9,113,\nWorld exports 746,464 972 … … 8.7 1.7 3.3 95.0\n\nSources: IMF Direction of Trade Statistics, World Economic Outlook database; and staff estimates.\n\n1/ For Cape Verde and Guinea 1976 and 1981 respectively.\n2/ Benin, Burkina Faso, Cote d'Ivoire, Guinea-Bissau, Mali, Niger, Senegal, and Togo\n3/ Cape Verde, The Gambia, Ghana, Guinea, Liberia, Nigeria, and Sierra Leone\n\n\n27\n\nMention: \"World Economic Outlook database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"World export volumes are taken from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003477:36:0", "occurrence_id": "b3bd0ce73205c7dfc2e802a7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **<u>Table A2: Descriptives for Self-Employment Activities (cont)</u>**\n\n<u>2004</u> <u>2005</u>\n<u>Panel F: Future Work Preferences</u>\n\nRemain in MEP Project 71% 90%\nFind wage employment 10% 8%\nStart another self-employment activity 19% 2%\n\n<u>Panel G: Project Sustainability</u>\n\nDefinitely sustainable 82% 80%\nProbably sustainable 16% 16%\nProbably not sustainable 0% 2%\nDefinitely not sustainable 0% 2%\nNo Knowledge 2% 0%\n\n<u>Panel H: Income Sustainability</u>\n\nDefinitely sustainable 77% 66%\nProbably sustainable 18% 25%\nProbably not sustainable 1% 8%\nDefinitely not sustainable 1% 2%\nNo Knowledge 3% 0%\n\n<u>Source: Author's calculations based on MEP sample (2004, 2005). Table reports descriptive</u>\nstatistics in the baseline period for different project characteristics. For example, the first number\nin Panel F means that 71% of the MEP beneficiaries in 2004 report that if they could choose in the\nfuture they would remain in MEP.\n\n\n**<u>Table A4: Descriptives for Project Rentability</u>**\n\n\n\nNegative Low/medium High\n<u>Rentability</u> <u>Rentability</u> <u>Rentability</u>\n\n<u>(1)</u> <u>(2)</u> <u>(3)</u>\nShare Females 0.8 0.5 0.7\nAv. Years Schooling 8.5 8.0 8.5\nAge 39.1 40.1 38.5\nShare of HH Heads 0.8 0.7 0.6\nNumber of Children in the HH 2.0 2.0 1.9\nSize\n\nMention: \"MEP sample (2004, 2005)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Baseline project characteristics and sustainability perceptions are tabulated from these rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003481:1:0", "occurrence_id": "90209eb0f13f4e662ac31623", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **1.** **Introduction**\n\n\nA new survey of 1,948 retail stores in India compiled by the World Bank’s Enterprise\n\n\nSurveys shows that 19% of the stores use computers for their business. Computer usage\n\n\nis higher among stores that find labor regulations to be more burdensome. Specifically,\n\n\n16% of the stores who report labor regulations as “no obstacle” for their business use\n\n\ncomputers. Corresponding figures for stores reporting labor regulations as “minor”,\n\n\n“moderate”, “major or very severe” obstacles are 24%, 30% and 34%, respectively. We\n\n\ntest if computer usage in India’s retail sector is driven by the severity of labor regulations.\n\n\nExisting empirical work has paid little attention to the technology-labor\n\n\nregulations nexus. <sup>1</sup> Few studies which do exist treat the choice of technology as\n\n\nexogenous to labor regulations. For example, Card et al (1999) take computer usage in\n\n\nthe Canada, France and the US as an exogenous technological shock. They find that the\n\n\neffect of this shock on wages and employment depends on labor regulations in the three\n\n\ncountries. In recent theoretical contributions, Alesina and Zeira (2006) and Blanchard and\n\n\nPhilippon (2006) have argued that the choice of technology should be treated as\n\n\nendogenous to labor regulations. Our findings confirm this argument.\n\n\n**2. Data and Main Variables**\n\n\nWe use store level data collected by the World Bank in 2006. <sup>2</sup> [^2: The survey and methodology for data collection are available at <u>www.enterprisesurveys.org.</u>] The data are a pure cross\n\n\nsection of 1,948 stores spread over 16 states and 41 cities of India. Our dependent\n\n\nvariable, _Computers_, is a dummy which equals 1 if a store uses a computer for running its\n\n\n1 See, for example, Botero, Djankov, La Porta, Silanes and Shleifer (2004).\n2 The\n\nMention: \"survey of 1,948 retail stores\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Computer use among retail stores is reported from this sample.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003481:1:0", "occurrence_id": "29877235cb21b0ae06944b34", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **1.** **Introduction**\n\n\nA new survey of 1,948 retail stores in India compiled by the World Bank’s Enterprise\n\n\nSurveys shows that 19% of the stores use computers for their business. Computer usage\n\n\nis higher among stores that find labor regulations to be more burdensome. Specifically,\n\n\n16% of the stores who report labor regulations as “no obstacle” for their business use\n\n\ncomputers. Corresponding figures for stores reporting labor regulations as “minor”,\n\n\n“moderate”, “major or very severe” obstacles are 24%, 30% and 34%, respectively. We\n\n\ntest if computer usage in India’s retail sector is driven by the severity of labor regulations.\n\n\nExisting empirical work has paid little attention to the technology-labor\n\n\nregulations nexus. <sup>1</sup> Few studies which do exist treat the choice of technology as\n\n\nexogenous to labor regulations. For example, Card et al (1999) take computer usage in\n\n\nthe Canada, France and the US as an exogenous technological shock. They find that the\n\n\neffect of this shock on wages and employment depends on labor regulations in the three\n\n\ncountries. In recent theoretical contributions, Alesina and Zeira (2006) and Blanchard and\n\n\nPhilippon (2006) have argued that the choice of technology should be treated as\n\n\nendogenous to labor regulations. Our findings confirm this argument.\n\n\n**2. Data and Main Variables**\n\n\nWe use store level data collected by the World Bank in 2006. <sup>2</sup> [^2: The survey and methodology for data collection are available at <u>www.enterprisesurveys.org.</u>] The data are a pure cross\n\n\nsection of 1,948 stores spread over 16 states and 41 cities of India. Our dependent\n\n\nvariable, _Computers_, is a dummy which equals 1 if a store uses a computer for running its\n\n\n1 See, for example, Botero, Djankov, La Porta, Silanes and Shleifer (2004).\n2 The\n\nMention: \"Enterprise\n\n\nSurveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The programme that compiled the retail store sample.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003493:17:1", "occurrence_id": "e66f3842f1ce6ccd166f1f49", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: </u> <u>-3.1</u> <u>0.2</u> <u>-2.4</u> <u>4.5</u>\nMedium conflict state and union 4.3 3.6 3.3 3.7\n<u>Low conflict state and union</u> <u>10.6 *</u> <u>9.1</u> <u>*</u> <u>7.8</u> <u>*</u> <u>6.5</u> <u>*</u>\n_Controls for:_\nStudent characteristics incl. incl. incl. incl.\nFamily background incl. incl. incl. incl.\nHome incentives and inputs incl. incl. incl. incl.\nPseudo R <sup>2</sup> 0.14 0.15 0.15 0.15\n<u>Observations=12,332</u> <u>-</u> <u>-</u> <u>-</u> <u>-</u>\nSource: Estimation with Quantile Regressions with Bootstrapped SE using PISA 2003; institutional variables;\nfull results available upon request\n\n - Denotes significance at the 99% level; **Denotes significance at 95 % level\n\n\nIt is not enough to have low levels of conflict with unions, although it helps. More\n\n\nimportantly, paying teachers more will not necessarily reduce conflicts, and there is no evidence\n\n\nthat it will lead to better learning outcomes (Figure 1). States with low levels of conflict and\n\n\nhigh teacher wages do very well. Even better are states that have complete and comprehensive\n\n\naccountability systems. The accountability system for Colima (World Bank 2005), the best\n\n\nperforming Mexican state, is characterized by\n\nMention: \"PISA 2003\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Quantile regressions with bootstrapped standard errors are estimated on this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003494:19:0", "occurrence_id": "d79aebe6f8b22467a87b9047", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: development is thus assumed to be associated with a better institutional framework\n\nincluding a debt management strategy and its public availability. We used also a\n\nquadratic of this variable in the model to capture possible non-linearities, however,\n\nit appeared to be insigni…cant and is not reported in the estimation results. The\n\nmeasure of GDP per capita is the PPP converted gross domestic product from the\n\nPenn World Tables (Heston et al, 2006).\n\n\nIndebtedness of government  - increases in this variable, de…ned as the ratio of\n\ntotal government debt to GDP, should result in an e¤ort to consolidate government\n\n…nances and adopt a debt management strategy. One can also expect that if this\n\nindicator reaches high levels the government may give up on debt management and\n\nfocus on debt renegotiations. Although debt renegotiations could be seen as a part of\n\nthe debt management strategy we do not include them in our indicator yi: Therefore,\n\ninclusion of a quadratic of government indebtedness into our PROBIT model can be\n\njusti…ed. As in the case of the GDP per capita the quadratic term appeared to be\n\ninsigni…cant and is not reported in the presented estimation results. The total debt\nto-GDP ratio was obtained from the GDF & WDI Central database of the World\n\nBank and the EIU database.\n\n\nGovernment share of GDP  - this variable is used to approximate the importance\n\nof public sector (government) in economic performance of a country. One may expect\n\nthat a larger share of government on real GDP would result in a greater e¤ort to\n\nstabilize government …nances in the sake of greater macroeconomic stability. Sim\nilarly, if government actions are important for an economy the public will require\n\nhigher transparency and accountability from the government. Existence of a public\n\ndebt management strategy is thus deemed to represent increased e¤orts of the gov\nernment to stabilize its …nances and meet\n\nMention: \"Penn World Tables\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Output per capita measured at purchasing power parity comes from this version.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003494:19:0", "occurrence_id": "d330056f22c4b457de907431", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: development is thus assumed to be associated with a better institutional framework\n\nincluding a debt management strategy and its public availability. We used also a\n\nquadratic of this variable in the model to capture possible non-linearities, however,\n\nit appeared to be insigni…cant and is not reported in the estimation results. The\n\nmeasure of GDP per capita is the PPP converted gross domestic product from the\n\nPenn World Tables (Heston et al, 2006).\n\n\nIndebtedness of government  - increases in this variable, de…ned as the ratio of\n\ntotal government debt to GDP, should result in an e¤ort to consolidate government\n\n…nances and adopt a debt management strategy. One can also expect that if this\n\nindicator reaches high levels the government may give up on debt management and\n\nfocus on debt renegotiations. Although debt renegotiations could be seen as a part of\n\nthe debt management strategy we do not include them in our indicator yi: Therefore,\n\ninclusion of a quadratic of government indebtedness into our PROBIT model can be\n\njusti…ed. As in the case of the GDP per capita the quadratic term appeared to be\n\ninsigni…cant and is not reported in the presented estimation results. The total debt\nto-GDP ratio was obtained from the GDF & WDI Central database of the World\n\nBank and the EIU database.\n\n\nGovernment share of GDP  - this variable is used to approximate the importance\n\nof public sector (government) in economic performance of a country. One may expect\n\nthat a larger share of government on real GDP would result in a greater e¤ort to\n\nstabilize government …nances in the sake of greater macroeconomic stability. Sim\nilarly, if government actions are important for an economy the public will require\n\nhigher transparency and accountability from the government. Existence of a public\n\ndebt management strategy is thus deemed to represent increased e¤orts of the gov\nernment to stabilize its …nances and meet\n\nMention: \"GDF & WDI Central database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The total debt to output ratio is obtained from this central database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003494:19:0", "occurrence_id": "6bb0938285293c0de9f55fb5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: development is thus assumed to be associated with a better institutional framework\n\nincluding a debt management strategy and its public availability. We used also a\n\nquadratic of this variable in the model to capture possible non-linearities, however,\n\nit appeared to be insigni…cant and is not reported in the estimation results. The\n\nmeasure of GDP per capita is the PPP converted gross domestic product from the\n\nPenn World Tables (Heston et al, 2006).\n\n\nIndebtedness of government  - increases in this variable, de…ned as the ratio of\n\ntotal government debt to GDP, should result in an e¤ort to consolidate government\n\n…nances and adopt a debt management strategy. One can also expect that if this\n\nindicator reaches high levels the government may give up on debt management and\n\nfocus on debt renegotiations. Although debt renegotiations could be seen as a part of\n\nthe debt management strategy we do not include them in our indicator yi: Therefore,\n\ninclusion of a quadratic of government indebtedness into our PROBIT model can be\n\njusti…ed. As in the case of the GDP per capita the quadratic term appeared to be\n\ninsigni…cant and is not reported in the presented estimation results. The total debt\nto-GDP ratio was obtained from the GDF & WDI Central database of the World\n\nBank and the EIU database.\n\n\nGovernment share of GDP  - this variable is used to approximate the importance\n\nof public sector (government) in economic performance of a country. One may expect\n\nthat a larger share of government on real GDP would result in a greater e¤ort to\n\nstabilize government …nances in the sake of greater macroeconomic stability. Sim\nilarly, if government actions are important for an economy the public will require\n\nhigher transparency and accountability from the government. Existence of a public\n\ndebt management strategy is thus deemed to represent increased e¤orts of the gov\nernment to stabilize its …nances and meet\n\nMention: \"EIU database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It supplies the companion country data behind the debt ratio.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003497:16:0", "occurrence_id": "9a75d4cfde8e0d57f4352a18", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: important aspect of the runoff regime to simulate correctly. A genetic mutation algorithm\n\n\n(Frontline Systems Inc. 2003) was used to optimize the parameters. This approach enables\n\n\nthe whole of the ‘solution space’ to be searched and generally allows solutions to be obtained\n\n\nmore rapidly than alternative optimization methods. The AWCmult parameters were\n\n\nconstrained to lie within a range 0.1m to 5.0m depending on the likely rooting depth of\n\n\nvegetation within the specified land use class (Table 3).\n\n\nUsing a dedicated computer, parameter calibration took approximately 15 seconds per grid\n\n\ncell. Thus the total time for calibrating parameters for all cells in Africa was three days. A\n\n\ncomparison between average annual runoff generated using WatBal and the UNH runoff\n\n\ndata, at the scale of ‘food production units’ (of which there are 97 in Africa) shows a\n\n\nreasonable fit (Figure 8).\n\n\n**_2.4_** **_Runoff simulation_**\n\n\nWhen the model was calibrated it was run to estimate runoff, on a monthly time step, for all\n\n\ngrid cells for the period 1961–1990. The cell data were then converted to runoff in each\n\n\ndistrict. The percentage of each district’s area in each associated grid cell was determined\n\n\nusing a blank global 0.5 x 0.5 degree grid and geoprocessing tools in ArcView. This enabled\n\n\nthe weighted average runoff produced by the grid cells contributing to runoff within any\n\n\ngiven district to be determined. For example, at any given time step, all districts in the same\n\n\ngrid cell will have the same runoff. However, if a district was divided across a number of grid\n\n\ncells (as illustrated in Figure 9), runoff for that district was calculated based on the percentage\n\n\narea of the district in each of the grid cells.\n\n\n**_2.5 Flow simulation_**\n\n\nThe flow routing model uses the STN-30p drainage network (Section 2.2.3\n\nMention: \"UNH runoff\n\n\ndata\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Modelled runoff is compared with these data at food production unit scale.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003502:16:1", "occurrence_id": "8ab70cca9d7fc667dece453e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  under the three scenarios outlined\n\n\nabove for the 22 countries with the highest burden of TB. Region-specific life\n\n\nexpectancies were derived from the Disease Control Priorities Project (www.dcp2.org).\n\n\nWe followed WHO methods in assuming that a person faces the same probability of\n\n\ndeath at each subsequent age as the existing population. This is equivalent to supposing\n\n\n13\n\nMention: \"the Disease Control Priorities Project\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Region-specific life expectancies are derived from this project.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003509:4:0", "occurrence_id": "4748a9089c4e0c4f59ebfcaa", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Recent literature has focused on the “discovery process” of exporting (Hausmann-Rodrik,\n\n\n2003). It contends that firms in developing economies tend under-invest in discovery\n\n\nbecause would-be first movers into export markets fear their initially high returns would be\n\n\neroded by subsequent new entry, resulting in an under-investment in searching for new\n\n\nexport activities. This hypothesis of “super-easy entry” is seen as the key market failure\n\n\nimpeding diversification. If this could be overcome, then presumably diversification\n\n\nwould increase and exports would grow more rapidly.\n\n\nA policy corollary is that governments can usefully deploy industrial policies to stimulate\n\n\ndiscovery and hence diversification. Klinger and Lederman (2004) find that overall export\n\n\ndiversification increases at low levels of development. They create a model to test the\n\n\nhypothesis that the threat of imitation inhibits the rate of “discovery”. They proxy barriers\n\n\nto entry by using the average time it takes to register a formal firm (from the _Doing_\n\n\n_Business_ surveys, World Bank), and find that indeed barriers to entry are associated with\n\n\nincreased discovery and diversification. From this they deduce that some type of subsidy\n\n\nto the diversification process is warranted.\n\n\nThis view of discovery may eventually be sustained by subsequent research. However,\n\n\nthere are intuitive reasons to continue investigation of this hypothesis. First, it is not clear\n\n\nthat aggregate data affecting all markets apply equally across industries and hence to the\n\n\nindividual industry failing to export. Second, anecdotal evidence often points to the\n\n\nopposite conclusion: that imitating entry broadens the market through agglomeration and\n\n\nindustrial level economies of scale affecting key inputs and lowers transportation costs for\n\n\nall firms in the industry. Finally, even if market failures constrain discovery, it is not clear\n\n\nthat policy makers should devote scarce resources to this problem – if discovery turns out\n\n\nto be a small part of the export growth problem.\n\n\nThis last concern is the point of departure for this paper. The discovery channel is only\n\n\none way of looking at the export diversification\n\nMention: \"_Doing_\n\n\n_Business_ surveys\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Average firm registration time from them proxies entry barriers.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003512:10:1", "occurrence_id": "ce53a9a6abecffef38de359d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  a language barrier, or they may be intentional\n\n\nincorrect answers. The final number of useable surveys is 9064 and their distribution by\n\n\ncountry is shown in Table 1.\n\n\nData on climate were gathered from two sources. We relied on temperature data from\n\n\nsatellites operated by the Department of Defense (Basist et al. 2001). The Defense\n\n\nDepartment uses a set of polar orbiting satellites that pass above each location on earth\n\n\nbetween 6am and 6pm every day. These satellites are equipped with sensors that measure\n\n\nsurface temperature by detecting microwaves that pass through clouds (Weng & Grody\n\n\n1998). The precipitation data come from the Africa Rainfall and Temperature Evaluation\n\n\nSystem (ARTES) (World Bank 2003). This dataset, created by the National Oceanic and\n\n\nAtmospheric Association’s Climate Prediction Center, is based on ground station\n\n\nmeasurements of precipitation. The temperatures for each country in the sample are shown in\n\n\n9\n\nMention: \"temperature data from\n\n\nsatellites\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Polar-orbiting satellite readings supply the temperature series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003512:10:1", "occurrence_id": "c5367e1eeea8e70f48009fbd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  a language barrier, or they may be intentional\n\n\nincorrect answers. The final number of useable surveys is 9064 and their distribution by\n\n\ncountry is shown in Table 1.\n\n\nData on climate were gathered from two sources. We relied on temperature data from\n\n\nsatellites operated by the Department of Defense (Basist et al. 2001). The Defense\n\n\nDepartment uses a set of polar orbiting satellites that pass above each location on earth\n\n\nbetween 6am and 6pm every day. These satellites are equipped with sensors that measure\n\n\nsurface temperature by detecting microwaves that pass through clouds (Weng & Grody\n\n\n1998). The precipitation data come from the Africa Rainfall and Temperature Evaluation\n\n\nSystem (ARTES) (World Bank 2003). This dataset, created by the National Oceanic and\n\n\nAtmospheric Association’s Climate Prediction Center, is based on ground station\n\n\nmeasurements of precipitation. The temperatures for each country in the sample are shown in\n\n\n9\n\nMention: \"Africa Rainfall and Temperature Evaluation\n\n\nSystem\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Precipitation estimates for the continent come from this system.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003514:10:0", "occurrence_id": "f0d53c3b304590c0fde11df1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: assumption in turn implies that the next three months would be spring, the three months after that\n\n\nwould be summer, and August, September and October would be fall (in the north). The choice\n\n\nof these particular seasonal definitions is motivated by the fact that they provided the best fit\n\n\nwith the data and that they reflected the mid-point for key rainy seasons in the sample. We\n\n\nadjusted for the fact that seasons in the southern hemisphere occur at exactly opposite months of\n\n\nthe year from northern hemisphere seasons.\n\n\nSoil data were obtained from FAO (2003). The FAO data provide information about the major\n\n\nand minor soils in each location. Data concerning the hydrology were obtained from the\n\n\nUniversity of Colorado (IWMI & University of Colorado 2003). Using a hydrological model for\n\n\nAfrica, the hydrology team calculated flow and runoff for each district in the surveyed countries.\n\n\n**4. Results**\n\n\nTable 2 summarizes the crop choice categories in the African analysis. The most popular crop\n\n\nchoice in the sample is maize. This crop is capable of growing across a range of climates because\n\n\nthere are a number of varieties of maize grown in Africa. The second most popular crop portfolio\n\n\nis maize-groundnut. We hyphenate crop combinations throughout the paper. Cowpea-sorghum,\n\n\nsorghum and millet-groundnut are also popular choices. While growing maize alone is the most\n\n\npopular choice in Zambia and Ghana, farmers in Cameroon prefer to grow maize-groundnut.\n\n\nDrought tolerant crops were the revealed preference of farmers in Burkina Faso, Niger and\n\n\nSenegal (sorghum, cowpea, or cowpea-sorghum). In Egypt and South Africa, as expected, the\n\n\npopular choices are maize and wheat. South African farmers also selected a variety of other crops\n\n\n(which did not belong to any of the other revealed categories). Farmers in Zambia and Zimbabwe\n\n\nindicated a preference for maize and maize-groundnut.\n\n\nIn Table 3, we examine the determinants of crop\n\nMention: \"FAO data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Major and minor soil types by location are taken from this source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003516:13:0", "occurrence_id": "984990d09d8ddc30381de387", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: exiting from product lines where returns are being squeezed is the strategy promising the\n\n\nbest returns (Hummels and Klenow 2005).\n\n\nOne measure of the development of technological capability in these countries\n\n\nand on the potential for diversification is provided by the statistics for patents registered\n\n\nin the United States between 2001 and 2005. Like articles in refereed scientific journals,\n\n\npatents are only a partial indicator of scientific prowess, because many patents are never\n\n\nused for any purpose (just as a high fraction of scientific papers are never cited or used to\n\n\nadvance technology). Patents are also registered by companies for purposes of cross\n\nlicensing and to defend against lawsuits. Although such factors limit the usefulness of\n\n\npatent data, the data nonetheless convey a sense of the scale of technology development\n\n\nand the areas in which it is most intense.\n\n\nBy the end of 2006, China ranked 24th in the world in the number of total patents\n\n\ngranted by the USPTO, with 3,178 patents. For the five years ending in 2006, it ranked\n\n\n20th in the world, with 2,053 patents. While these scores represent a significant increase\n\nover 2001, when China ranked 24 <sup>th</sup> in the world, patents from China are only now\n\n\nbeginning to make their presence felt on the United States Patent and Trademark Office\n\n\n(USPTO).\n\n\nWhat is most striking from the USPTO data is the narrow focus of patenting in\n\n\nthe sample of countries relative to the United States and Japan. For instance, the top 10\n\n\npatent classes account for 39 percent of US patents granted to China (Table 3) but just 21\n\n\npercent for the United States. The main fields of patent dominance (life sciences and\n\n\nsemiconductors for the United States, electronic components for Japan) account for much\n\n\nless than 10 percent of all patents in these countries (Tables 4 and 5). By comparison, for\n\n\nFinland, the Republic of Korea, and Taiwan (China), electronics or telecommunications\n\n\naccount for 30–40\n\nMention: \"USPTO data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Its records reveal how narrowly patenting is concentrated across the sample.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003521:1:0", "occurrence_id": "47014364ec4949f0edf28ee6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Policy Research Working Paper 4314\n\n### **Abstract**\n\n\nA new dataset of 1,948 retail stores in India compiled\nby the World Bank’s Enterprise Surveys shows that 27\npercent of the stores report labor regulations as a problem\nfor their business. Using these data we analyze the effect\nof labor regulation on employment at the store level. We\n\n\n\nfind that stricter labor regulation has a strong negative\neffect on employment. Our estimates show that labor\nreforms are likely to increase employment by 22 percent\nof the current level for an average store.\n\n\n\nThis paper—a product of the Financial and Private Sector Development Vice-Presidency, Enterprise Analysis Unit —is part\nof a larger effort in the department to to better understand the functioning of private sector. Policy Research Working Papers\nare also posted on the Web at http://econ.worldbank.org. The author may be contacted at mamin@worldbank.org.\n\n\n_The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development_\n_issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the_\n_names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those_\n_of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and_\n_its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent._\n\n\nProduced by the Research Support Team\n\nMention: \"Enterprise Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The retail store dataset on labour regulation complaints is compiled by this programme.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003523:32:1", "occurrence_id": "f9fc573a043127000ea542c8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 69 4.63\nAccess to formal credit 357 0.49 0.50 0 1\nCapacity utilization 357 70.53 22.98 4 100\nFirm exports 357 0.14 0.34 0 1\n<u>Firm sells to small customers</u> <u>357</u> <u>0.79</u> <u>0.41</u> <u>0</u> <u>1</u>\n\n\n**Source** : Authors’ estimates based on 2006 Enterprise Surveys.\n\nMention: \"2006 Enterprise Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Firm characteristics and outcomes are estimated from this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003529:29:2", "occurrence_id": "d9b7be2c6f6bc6bdc044fbe2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  domestic appliances,\ncleaning accessories, personal health products, recreation. Non Eligible: Alcoholic drinks and tobacco, cost of housing,\nelectricity and fuels, other services related to housing, health care, public transport, private transport, education.\n\n\nTable 7: Effect of SARE on Consumer Prices\n\n\n\n.\n\n\n\nThe two columns on the right compare changes across eligible vs non-eligible industries\nusing specification 3. <sup>20</sup> Banco de Mexico only reports prices for 46 cities, 29 of which had\n\nimplemented SARE during our sample period. We report results using both: only SARE\n\nmunicipalities and using all cities which includes non-SARE cities as controls. The results\n\nin the four specifications show that SARE had _no_ significant effect on prices.\n\n\n20We classified 8 industries as eligible for SARE and the remaining 8 as non-eligible, making them as\n\ncompatible as possible with our previous definition. Table 7 lists our classification.\n\n\n28\n\nMention: \"Banco de Mexico\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its price coverage is limited to forty-six cities, restricting the comparison.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003532:26:0", "occurrence_id": "7214101b662ccd4efa1211c0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **_6.1._** **_Statistics Used to Measure Livestock Mortality_**\n\n\nThe GoM has been conducting an annual census of animals in Mongolia for well over 50 years.\nThe procedures are well-established and have numerous laws that attempt to protect the integrity\nof the process. Nonetheless, there are potential problems with these data once an insurance\nproduct is developed to pay based on the mortality rates. Considerable effort was made to\nunderstand the data process during the development of the pilot. It was learned that\napproximately 90 percent of the losses occur while herders are in their winter camps. Even\nthough herders move their animals and can cross _soum_ boundaries, it was also learned that where\nthey have their winter camp is the designated _soum_ in the census. Herders typically have a pattern\nof keeping their animals in the same camps by season. Local government leaders know the\nfamilies and know their animals. These experts are used to assist in counting and reporting\nproblems. New systems to track the same families will be developed at the National Statistics\nOffice (NSO).\n\n\nTo assure a timely payment, a new animal census will be conducted during the pilot stage during\nthe month of May. Mortality rates of adult animals will be based on the Census at the end of\nDecember and the May Census. There are very few animals slaughtered during this period. Thus,\nthis statistic should reflect actual mortality better than numbers that would extend into the heavy\nslaughtering periods. During the pilot, the traditional methods of reporting mortality and a sample\nsurvey will be used to monitor any emerging problems. Additionally, a special technical\nassistance program is to be provided by the U.S. Department of Agriculture Statistics Service.\nThis assistance will be targeted at a lower cost and more accurate procedure that will use sample\nsurvey techniques. At the end of the pilot, an evaluation will be made about the sustainability and\nreliability of alternative methods for estimating mortality rates.\n\n\n**_\n\nMention: \"annual census of animals\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Over fifty years of these counts provide the livestock mortality base.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003532:26:0", "occurrence_id": "33bf97d7926af91499097365", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **_6.1._** **_Statistics Used to Measure Livestock Mortality_**\n\n\nThe GoM has been conducting an annual census of animals in Mongolia for well over 50 years.\nThe procedures are well-established and have numerous laws that attempt to protect the integrity\nof the process. Nonetheless, there are potential problems with these data once an insurance\nproduct is developed to pay based on the mortality rates. Considerable effort was made to\nunderstand the data process during the development of the pilot. It was learned that\napproximately 90 percent of the losses occur while herders are in their winter camps. Even\nthough herders move their animals and can cross _soum_ boundaries, it was also learned that where\nthey have their winter camp is the designated _soum_ in the census. Herders typically have a pattern\nof keeping their animals in the same camps by season. Local government leaders know the\nfamilies and know their animals. These experts are used to assist in counting and reporting\nproblems. New systems to track the same families will be developed at the National Statistics\nOffice (NSO).\n\n\nTo assure a timely payment, a new animal census will be conducted during the pilot stage during\nthe month of May. Mortality rates of adult animals will be based on the Census at the end of\nDecember and the May Census. There are very few animals slaughtered during this period. Thus,\nthis statistic should reflect actual mortality better than numbers that would extend into the heavy\nslaughtering periods. During the pilot, the traditional methods of reporting mortality and a sample\nsurvey will be used to monitor any emerging problems. Additionally, a special technical\nassistance program is to be provided by the U.S. Department of Agriculture Statistics Service.\nThis assistance will be targeted at a lower cost and more accurate procedure that will use sample\nsurvey techniques. At the end of the pilot, an evaluation will be made about the sustainability and\nreliability of alternative methods for estimating mortality rates.\n\n\n**_\n\nMention: \"animal census\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"A new census is planned during the pilot stage to assure timely payment.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003532:26:0", "occurrence_id": "8e7eabc5488d8310ed299d7d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **_6.1._** **_Statistics Used to Measure Livestock Mortality_**\n\n\nThe GoM has been conducting an annual census of animals in Mongolia for well over 50 years.\nThe procedures are well-established and have numerous laws that attempt to protect the integrity\nof the process. Nonetheless, there are potential problems with these data once an insurance\nproduct is developed to pay based on the mortality rates. Considerable effort was made to\nunderstand the data process during the development of the pilot. It was learned that\napproximately 90 percent of the losses occur while herders are in their winter camps. Even\nthough herders move their animals and can cross _soum_ boundaries, it was also learned that where\nthey have their winter camp is the designated _soum_ in the census. Herders typically have a pattern\nof keeping their animals in the same camps by season. Local government leaders know the\nfamilies and know their animals. These experts are used to assist in counting and reporting\nproblems. New systems to track the same families will be developed at the National Statistics\nOffice (NSO).\n\n\nTo assure a timely payment, a new animal census will be conducted during the pilot stage during\nthe month of May. Mortality rates of adult animals will be based on the Census at the end of\nDecember and the May Census. There are very few animals slaughtered during this period. Thus,\nthis statistic should reflect actual mortality better than numbers that would extend into the heavy\nslaughtering periods. During the pilot, the traditional methods of reporting mortality and a sample\nsurvey will be used to monitor any emerging problems. Additionally, a special technical\nassistance program is to be provided by the U.S. Department of Agriculture Statistics Service.\nThis assistance will be targeted at a lower cost and more accurate procedure that will use sample\nsurvey techniques. At the end of the pilot, an evaluation will be made about the sustainability and\nreliability of alternative methods for estimating mortality rates.\n\n\n**_\n\nMention: \"May Census\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Paired with the year-end count it yields adult mortality rates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003535:20:0", "occurrence_id": "284c357266462d4325cd92cd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The evidence thus far suggests that while there is a significantly positive partial\n\n\ncorrelation between disbursements on IDA lending and debt service paid to IDA, we do\n\n\nnot think that this correlation should be interpreted as evidence of defensive lending.\n\n\nThere is however a further possibility for defensive lending: it is possible that defensive\n\n\nlending is not done overtly, by increasing disbursements beyond what is prescribed by\n\n\nthe PBA formula to countries with incipient debt servicing difficulties. Rather it could be\n\ndone more covertly through manipulation of the variables driving PBA allocations. <sup>23</sup> In\n\n\nparticular, it is possible that defensive lending is accomplished by giving better-than\n\nwarranted CPIA scores to countries with incipient debt servicing difficulties. This would\n\n\nresult in higher-than-warranted commitments and eventual disbursements to the country,\n\n\nbut yet would not evidence itself as a significant partial correlation between\n\n\ndisbursements and debt service after controlling for policy performance.\n\n\nA simple way to test for this possibility is to examine whether measures of\n\n\ncountry indebtedness have predictive power for IDA country performance ratings after\n\n\ncontrolling for other measures of policy performance. We do this in Table 4, where we\n\n\nuse data from the Worldwide Governance Indicators project (Kaufmann, Kraay, and\n\n\nMastruzzi (2006)) as alternative measures of policy performance. We consider two of\n\n\nthe six measures from this dataset, Rule of Law, and Government Effectiveness, as\n\n\nones that correspond most closely to what the IDA country performance ratings are\n\nmeasuring. <sup>24</sup> Although these indicators have the advantage of covering all IDA\n\n\nborrowers, they have a more limited time dimension and are available only 1996, 1998,\n\n\n2000, and annually since 2002.  We then report results of pooled and fixed-effects\n\n\nregressions in this shorter panel, including in turn the stock of IDA debt and the flow of\n\n\nIDA debt service as proxies for incipient debt servicing difficulties that might be met with\n\n\nthis interaction to be statistically significant, possibly because we consider a\n\nMention: \"the Worldwide Governance Indicators project\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Its measures serve as alternative policy performance controls.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003536:6:4", "occurrence_id": "a5a16d9cf9df8a6c23799c1c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  data from Ministry of Finance).\n\nAlthough it is common practice to make international comparisons of selected expenditure\nindicators, such as the one presented above, these figures often are not ‘comparable’ when other\ndimensions are taken into consideration. A variety of economic and demographic factors should\nbe considered, however, when comparing education expenditures across countries. Factors that\nmay bias a one-dimensional comparison include the size of the public sector, the size of the\neconomy as a whole, the national income level, the level of government revenues, the publicprivate participation shares, the number of service beneficiaries, the prices of key inputs. Among\nmany other studies, Verbina and Chowdhury (2002), Fernandez and Regerson (2001), Arze et al\n(2005), discuss the determinants of public education spending. Results from an estimation that\ncontrols for several determinants of education expenditure allocations, including population,\npopulation density, GDP per capita, the level of fiscal decentralization, and budget balance;\nIndonesia’s level of education spending could be expected to be around 17 percent of its overall\n\n\n5\n\nMention: \"Ministry of Finance\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Expenditure figures used in the comparison come from this ministry.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003537:10:0", "occurrence_id": "8675131dea9f137db53e87e2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: deregulation process, followed by de novo branching. Table A1 presents the deregulation dates. <sup>1</sup> [^1: We have data for 50 states and the District of Columbia. Consistent with the literature on branch deregulation, we\ndrop Delaware and South Dakota because the structure of their banking systems were heavily affected by laws that\nmade them centers for the credit card industry, and because South Dakota lacks some financial and banking data.]\n\n\nTwelve states deregulated before the start of our sample period in 1976. Arkansas, Iowa and\n\n\nMinnesota were the last states to deregulate, only after the passage of the Riegle-Neal Act in 1994.\n\n\n**1.2. Income distribution data**\n\n\nInformation on the distribution of income is from the March Supplement of the Current\n\n\nPopulation Survey (CPS), which is a survey of about 60,000 households across the states of the\n\n\nU.S. The CPS provides information on total personal income, income from wages, proprietor\n\n\nincome, as well as detailed demographic information. The CPS is not a true panel; it is\n\n\ncross-sectional survey that is repeated each year. The CPS does not trace individuals through time,\n\n\nbut rather replaces 50% of the sample each year, so that each household is included for two\n\n\nconsecutive years. Each individual is assigned a sample weight corresponding to his or her\n\n\nrepresentativeness in the sample, which we use in our analyses.\n\n\nWe measure the distribution of income for each state and year over the period 1976-2005\n\n\nin four ways. First, the natural logarithm of the Gini coefficient of income distribution is derived\n\n\nfrom the Lorenz curve (Gini), where larger values imply greater income inequality. Our second\n\n\nmeasure of income distribution is the coefficient of variation of income, i.e. the variance of\n\n\nincome divided by the mean (Variance). Third, we use the difference between the natural\n\n\nlogarithm of incomes of those at the 90 <sup>th</sup> percentile and those at the 10 <sup>th\n\nMention: \"Current\n\n\nPopulation Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its March supplement of roughly sixty thousand households supplies the income distribution.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003538:20:0", "occurrence_id": "3fdde6f9f48d4ec70f580414", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: sector strategy and a road social contract in consultation with stakeholders that will guide\nexpenditures and provide an agreed framework for measuring successful delivery. In\nChile, the Ministry of Public Works is moving towards an in integrated planning process\nwhich will use a supply-chain approach to ensure investments are made based on multimodal analysis using improved economic and engineering assessment.\n\n**<u>Box 9: EITI</u>**\n\nDemand estimation will also improve the economic return to projects (World Bank,\n2006b). This involves not only improved economic analysis of projects, but also better\ncost forecasting. Predicted outcomes should be compared with predicted and actual\noutcomes of a relevant reference class of past projects to determine their likely accuracy,\nand forecasts should be published, publicly debated, and subject to independent review\n(Kenny, 2007).\n\nThere is a further need to improve expenditure tracking through levels of government. In\nsome cases, infrastructure services involve multiple levels of government simultaneously,\nand this can create particularly complex governance challenges. Rural roads is one such\ncase (see Box 10). Tanzania’s Pro Poor Expenditure Tracking Survey in 2001 covered\nfour sectors including water and rural roads. The Tanzania PETS found significant\nleakage. One district had recorded only 10 percent of receipts from central government\nfor road funds over 1999-2000, for example –this encouraged the Treasury to advertise\nall transfers from central to local government in the media (Sundet, 2004).\n\nThe response to expenditure tracking analysis can be improved if civil society is actively\nengaged. For example, in South Africa the Public Service Accountability Monitor based\nat Rhodes University provides a database of information on budget allocations and\nresources available to government departments as well as plans for their use, progress\nreports and recommendations made by oversight bodies (Sohail and Cavill, 2007).\n\n\n19\n\nMention: \"Pro Poor Expenditure Tracking Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Four sectors including water and rural roads were tracked, revealing leakage.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003538:20:0", "occurrence_id": "10c1e86532a67b4467db2d35", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: sector strategy and a road social contract in consultation with stakeholders that will guide\nexpenditures and provide an agreed framework for measuring successful delivery. In\nChile, the Ministry of Public Works is moving towards an in integrated planning process\nwhich will use a supply-chain approach to ensure investments are made based on multimodal analysis using improved economic and engineering assessment.\n\n**<u>Box 9: EITI</u>**\n\nDemand estimation will also improve the economic return to projects (World Bank,\n2006b). This involves not only improved economic analysis of projects, but also better\ncost forecasting. Predicted outcomes should be compared with predicted and actual\noutcomes of a relevant reference class of past projects to determine their likely accuracy,\nand forecasts should be published, publicly debated, and subject to independent review\n(Kenny, 2007).\n\nThere is a further need to improve expenditure tracking through levels of government. In\nsome cases, infrastructure services involve multiple levels of government simultaneously,\nand this can create particularly complex governance challenges. Rural roads is one such\ncase (see Box 10). Tanzania’s Pro Poor Expenditure Tracking Survey in 2001 covered\nfour sectors including water and rural roads. The Tanzania PETS found significant\nleakage. One district had recorded only 10 percent of receipts from central government\nfor road funds over 1999-2000, for example –this encouraged the Treasury to advertise\nall transfers from central to local government in the media (Sundet, 2004).\n\nThe response to expenditure tracking analysis can be improved if civil society is actively\nengaged. For example, in South Africa the Public Service Accountability Monitor based\nat Rhodes University provides a database of information on budget allocations and\nresources available to government departments as well as plans for their use, progress\nreports and recommendations made by oversight bodies (Sohail and Cavill, 2007).\n\n\n19\n\nMention: \"Tanzania PETS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Its findings show one district received only a tenth of central receipts.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003538:20:0", "occurrence_id": "7da7d552f1fb0f816b191170", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: sector strategy and a road social contract in consultation with stakeholders that will guide\nexpenditures and provide an agreed framework for measuring successful delivery. In\nChile, the Ministry of Public Works is moving towards an in integrated planning process\nwhich will use a supply-chain approach to ensure investments are made based on multimodal analysis using improved economic and engineering assessment.\n\n**<u>Box 9: EITI</u>**\n\nDemand estimation will also improve the economic return to projects (World Bank,\n2006b). This involves not only improved economic analysis of projects, but also better\ncost forecasting. Predicted outcomes should be compared with predicted and actual\noutcomes of a relevant reference class of past projects to determine their likely accuracy,\nand forecasts should be published, publicly debated, and subject to independent review\n(Kenny, 2007).\n\nThere is a further need to improve expenditure tracking through levels of government. In\nsome cases, infrastructure services involve multiple levels of government simultaneously,\nand this can create particularly complex governance challenges. Rural roads is one such\ncase (see Box 10). Tanzania’s Pro Poor Expenditure Tracking Survey in 2001 covered\nfour sectors including water and rural roads. The Tanzania PETS found significant\nleakage. One district had recorded only 10 percent of receipts from central government\nfor road funds over 1999-2000, for example –this encouraged the Treasury to advertise\nall transfers from central to local government in the media (Sundet, 2004).\n\nThe response to expenditure tracking analysis can be improved if civil society is actively\nengaged. For example, in South Africa the Public Service Accountability Monitor based\nat Rhodes University provides a database of information on budget allocations and\nresources available to government departments as well as plans for their use, progress\nreports and recommendations made by oversight bodies (Sohail and Cavill, 2007).\n\n\n19\n\nMention: \"Public Service Accountability Monitor\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It provides budget allocation and resource information for accountability analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003550:24:0", "occurrence_id": "50b900f2736726a4a695065f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 2) _IMPikt_ = λ 0 + λ 1 _GOVit_ + λ 2 _TPit_ + λ 3 _INFit_ + λ 4 _TotDemand_ _t_ +γ ' _X_ _it_ + _u_ 2\n\n\n**4.1 Description of variables and data sources**\n\n\nWe use US and EU imports of clothing products classified at the heading level (4 digit) of HS\n\nchapters 61-63. Trade data are obtained from the USITC website and the EU COMEXT\n\n\ndatabase. Data on the quality of governance are obtained from the Worldwide Governance\n\n\nIndicators (2006). As the six indicators contained in that dataset are highly correlated, including\n\n\nthem separately inevitably leads to problems of multicollinearity. Following Francois and\n\n\nManchin (2007) we use principal component analysis and retain one vector which accounts for\n\n\n80% of the total observed variance. All elements of the vector are positive, making\n\n\ninterpretation of the vector’s coefficient in the regression simple. An increase in any of the\n\n\nmeasured dimensions of government quality increases our aggregate measure and hence the\n\n\neffect on the propensity to export a given clothing product to a given market is qualitatively\n\n\nsimilar for all the contained indicators. In general, the quality of governance is better in\n\n\ncountries with higher per capita income. In order to isolate the relative quality of governance\n\n\n(per capita income is an explanatory variable in our gravity model), we regress this vector of\n\n\ngovernance on the log of per capita income and include the residuals as an independent variable\n\n\nin the regression.\n\n\nWe also use principal component analysis for indicators of infrastructure availability (main\n\n\ntelephone lines, internet users per 1000 inhabitants, road density (relative to both inhabitants\n\n\nand country size)) to deal with the collinearity of the individual variables. The retained vector is\n\n\nagain regressed on the log of per capita income and residuals are then used in the regression\n\n\nequation. We use the simple average of import\n\nMention: \"EU COMEXT\n\n\ndatabase\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Textile trade at four-digit heading level is obtained from it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003550:24:0", "occurrence_id": "ddb1536dd9f09c80cfad6e94", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 2) _IMPikt_ = λ 0 + λ 1 _GOVit_ + λ 2 _TPit_ + λ 3 _INFit_ + λ 4 _TotDemand_ _t_ +γ ' _X_ _it_ + _u_ 2\n\n\n**4.1 Description of variables and data sources**\n\n\nWe use US and EU imports of clothing products classified at the heading level (4 digit) of HS\n\nchapters 61-63. Trade data are obtained from the USITC website and the EU COMEXT\n\n\ndatabase. Data on the quality of governance are obtained from the Worldwide Governance\n\n\nIndicators (2006). As the six indicators contained in that dataset are highly correlated, including\n\n\nthem separately inevitably leads to problems of multicollinearity. Following Francois and\n\n\nManchin (2007) we use principal component analysis and retain one vector which accounts for\n\n\n80% of the total observed variance. All elements of the vector are positive, making\n\n\ninterpretation of the vector’s coefficient in the regression simple. An increase in any of the\n\n\nmeasured dimensions of government quality increases our aggregate measure and hence the\n\n\neffect on the propensity to export a given clothing product to a given market is qualitatively\n\n\nsimilar for all the contained indicators. In general, the quality of governance is better in\n\n\ncountries with higher per capita income. In order to isolate the relative quality of governance\n\n\n(per capita income is an explanatory variable in our gravity model), we regress this vector of\n\n\ngovernance on the log of per capita income and include the residuals as an independent variable\n\n\nin the regression.\n\n\nWe also use principal component analysis for indicators of infrastructure availability (main\n\n\ntelephone lines, internet users per 1000 inhabitants, road density (relative to both inhabitants\n\n\nand country size)) to deal with the collinearity of the individual variables. The retained vector is\n\n\nagain regressed on the log of per capita income and residuals are then used in the regression\n\n\nequation. We use the simple average of import\n\nMention: \"Worldwide Governance\n\n\nIndicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Its highly correlated composite indicators are used as governance controls.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003564:19:1", "occurrence_id": "3742056cc6c5332375bb3b36", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  covered in the WIIW database. Financial\n\n\nintermediation displays the highest average productivity across sub-sectors in Croatia,\n\n\nMacedonia, and Ukraine, and the second highest productivity in another 5 of the 10\n\ncountries. <sup>15</sup> Hotels and restaurants exhibit the lowest productivity across sub-sectors in 7\n\n\nof the 9 countries and the second lowest productivity in another country of the group of\n\nEU-5 countries and South Eastern European countries. <sup>16</sup>\n\n\n14 For brevity, we do not show the corresponding figures which are simply a different aggregation (by\ncountry) of the same statistics shown in Figures 8 and 9.\n15 Poland is an exception to the relatively high productivity of financial intermediation as this sub-sector\nexhibits the second lowest productivity in the country, only higher than that of hotels and restaurants.\n16 Ukraine is excluded for lack of data on hotels and restaurants.\n\n\n17\n\nMention: \"WIIW database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Sub-sector productivity comparisons are computed from it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003568:14:0", "occurrence_id": "8bd22b57148fcb0e9a1c0549", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: estimators show a positive partial correlation between FCS in China and India and those\n\n\nin the rest of the world.\n\n\nThe second panel of the table shows the results using the U.S. sector data. As with the\n\n\nOECD and UNCTAD data, there is an unconditional positive correlation between\n\n\nChinese FCS and FCS in LAC in the U.S. data. In contrast, here we cannot reject the\n\n\nhypothesis that the overall impact of India FCS in LAC is different from zero. These\n\n\nresults are robust to the use of the alternative estimators. The impact of China in the\n\n\nmanufacturing sector also appears to be either positive or not significant. The evidence of\n\n\nthe impact of India in the manufacturing sector is somewhat mixed. Indian FCS in the\n\n\nmanufacturing sector has a negative and significant coefficient in the Poisson estimation.\n\n\nHowever, this finding is not robust across estimators. We find no effect of this variable\n\n\nusing the OLS and an overall positive effect using the Negative Binomial estimator.\n\n\nTable 2 reports the results from the fully specified KCM, based on the OECD and\n\n\nUNCTAD data. Again, the bottom panel contains the point estimates of the China and\n\n\nIndia effects on LAC FCS. The first column shows the OLS results for our benchmark\n\n\nmodel. The results for most of the control variables are similar to those in the KCM\n\n\nliterature and consistent with the theory. GDPSUM and GDPDIF have the expected signs\n\n\nand are statistically significant. The coefficient in the skill difference is negative, which is\n\n\ncontrary to the theory but consistent with other empirical findings (e.g., Blonigen et al\n\n\n2003). We do not find a different effect for the negative skill difference. Both measures\n\n\nof investment cost are negative and statistically significant. The distance and the trade\n\n\ncosts variables also have a negative coefficient and are statistically different from zero.\n\n\n12\n\nMention: \"OECD and UNCTAD data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"The US results are compared with the correlations found in these sources.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003573:35:0", "occurrence_id": "32d4a971fa35f3528247568a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 1**\n\n\n\n\n\nSource: ALSMS05\n\n\n**Figure 2**\n\n\n\n\n\nSource: ALSMS05\n\n\n**Figure 3**\n\n\n\n\n\n\n\nSource: ALSMS05\n\n\n  - 34\n\nMention: \"ALSMS05\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"All three figures are drawn from this household survey round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003530:10:0", "occurrence_id": "a37f6b4d5b0ed85467336757", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: assets, and household characteristics that may affect crop output, and φ s are the parameters to be\n\n\nestimated. Assuming that the disturbance term is composed of two additive components where _vi_ is pure\n\nwhite noise, and _ui~ N_ <sup>_+_</sup> _(0,_ δ _u2)_ is a term that follows a truncated normal distribution with zero mean that\n\n\ncan be interpreted as producer’s level of technical inefficiency (Coelli 1995). A measure of ability or\n\n\ntechnical efficiency _TEi = exp(-ui)_ can then be predicted directly.\n\n\nTo compare effects of market transactions to those of non-market transactions (i.e., inheritance,\n\n\ngift, dowry, etc), we run an identical ordered probit model to identify key determinants for non-market\n\n\nland transfers with some modifications of the variables to be included in the ordered probit model. For\n\n\nexample, the entire argument of transaction costs associated with land sale and land purchase will not be\n\n\nrelevant to inheritance and gift exchange. Correspondingly, we treat the two thresholds in the ordered\n\n\nprobit model as constant rather than variable. As discussed in the estimation strategy section, we treated\n\n\nthe lower and upper bounds of the ordered probit model as constant because the transaction costs are\n\n\nunlikely to be relevant to non-market transactions. To allow direct comparability for robustness check, we\n\n\nalso estimate the ordered probit model with variable upper and lower bounds for the non-market land\n\n\ntransaction (see appendix table 2), and results are similar to those reported earlier.\n\n\n**3. Data sources and descriptive evidence**\n\n\nThe main source of data used in our analysis is two rounds of NCAER’s ARIS/REDS survey that were\n\n\nconducted in 1982 and 1999, respectively. This survey, the first rounds of which were implemented in\n\n\n1968-71 to evaluate the impact of an agricultural development program, covers all\n\nMention: \"NCAER’s ARIS/REDS survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Two rounds of this farmer survey are the main data for the analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003577:27:0", "occurrence_id": "f06a874a4d93014c00176729", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: arithmetic tells us that the standard deviation of measurement error is\n\n\nassessments. <sup>18</sup> [^18: For details on this calculation see Kaufmann, Kraay, and Mastruzzi (2004, 2006). Gelb, Ngo and Ye\n(2004) implement a similar calculation comparing the African Development Bank and World Bank CPIA\nscores. Their conclusion that the CPIA ratings are quite precise is largely driven by the fact that they focus\non the aggregate CPIA scores (across all macro, structural, social and public sector dimensions), which are\nvery highly correlated between the two institutions. In the case of the CPIA, here we focus on one of 16\nspecific questions, and at this level of disaggregation the correlation between the two sets of ratings is\nconsiderably lower.] We report this standard deviation in Table 3 for several pairs of\n\n\nindicators that we have discussed in the paper, and it ranges from 0.70 to 1.53 across\n\n\nour examples.  By way of comparison, the standard errors associated with aggregate\n\n\nindicators such as the WGI are much smaller, reflecting the benefits of aggregation in\n\n\nreducing noise in the individual indicators. For example the standard error for the\n\n\nestimate of Control of Corruption for a typical country in 2006 on the WGI is just 0.17, or\n\n\nless than a quarter of the standard error of the most precise pair of individual indicators\n\n\nin this example.\n\n\nTo appreciate the magnitude of this measurement error, it is useful to go one\n\n\nstep further and calculate the width of a 90% confidence interval for governance based\n\n\non any one of these individual indicators, based on the additional assumption that\n\n\ngovernance and the error term are jointly normally distributed. This width of the\n\n\ncolumn of Table 3. Since our assumptions imply that 95 percent of countries would have\n\n\ngovernance levels between -2 and 2, these figures imply that a 90 percent confidence\n\n\ninterval for governance for any individual country would span between half and two\n\nthirds of the entire most-likely range of governance outcomes!\n\n\n_Why Aggregate Indicators?_\n\nMention: \"WGI\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Aggregation gives it much smaller standard errors than individual indicators.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003578:16:0", "occurrence_id": "ca9c821438ebb216b6564a1f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 14\n\n##### **_Social health insurance status_**\n\n\nWe define our SHI dummy _SHIit_ as taking a value of one if in country _i_ at time _t_\n\n\nearmarked payroll taxes for health care were collected from formal-sector workers and there was\n\n\na SHI agency in place. The required information was obtained mainly from the _Health Systems in_\n\n\n_Transition_ ( _HiT)_ document series published by the European Observatory on Health Systems and\n\n\nPolicies, a partnership between the European Office of the World Health Organization (WHO)\n\n\nand governmental, national and international agencies. World Bank reports and consultations\n\n\nwith its staff working in the ECA region were also used in order to obtain data on countries for\n\n\nwhich HiTs have not yet been published, and to double-check the information assembled through\n\n\nthe HiTs. <sup>19</sup>\n\n\nOur derivation of the SHI status indicator is shown in Table 1. Our strict definition means\n\n\nthat we end up classifying as non-SHI some country-year combinations that are often—we\n\n\nbelieve, erroneously—classified as SHI (such as Latvia and Poland). Furthermore, we classify\n\n\nRomania as SHI only after 1998; despite the fact that payroll taxes were used somewhat before\n\n\nthen, it was not until 1998 that SHI was fully set up with a SHI agency and with payroll\n\n\ncontributions making up the majority of health care revenues. We explore the sensitivity of our\n\n\nresults to not classifying these as SHI countries, by re-running our models with the three of them\n\n\nclassified as SHI for the years indicated in Table 1.\n\n\n19 Our classification is consistent with the evidence presented in Langenbrunner et al. (in press) which came to our\nattention after the empirical work for the paper hade been completed.\n\nMention: \"HiTs\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Their information is double-checked against other sources for missing cases.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003599:4:0", "occurrence_id": "7ef240b37d1f2f5e93dcc44d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: and poverty effects of changes in macroeconomic policy and/or trends in global markets.\nIt complements a global CGE analysis with global microsimulations based on\nstandardized household surveys. The tool pools most of the currently available household\nsurveys covering 1.2 million households in 63 developing countries; household\ninformation from developed countries comes from the Luxemburg Income Study dataset.\nThese micro data are complemented with more aggregate information for countries where\nno surveys are available; the final dataset covers 91 percent of the world’s population\n(see Annex 1 for details).\n\nThe paper is organized as follows. The next section sketches the methodology,\nassumptions, and data behind the GIDD. Section 3 presents the macroeconomic results of\nthe baseline scenario, showing the importance of China and India for global growth and\ntrade. Section 4 assesses the importance of growth in China and India for the changes in\nthe global income distribution and the emergence of a global middle class. The final\nsection offers concluding remarks.\n\n#### 2 Methodology\n\nThe empirical analysis in this paper relies on two tools developed at the Development\nEconomic Prospects Group of the World Bank: the LINKAGE global computable general\nequilibrium (CGE) model and the GIDD, which combines a consistent set of price and\nvolume changes from the CGE model with expected changes in demographic structure to\ncreate a hypothetical distribution of income in 2030. We begin with a brief description of\nthe LINKAGE model and then proceed to introduce the GIDD framework and its ability\nto map macroeconomic outcomes to disaggregated household survey data.\n\n###### 2.1  LINKAGE: a global dynamic multi-sectoral model\n\nThe forward-looking scenarios in this paper have been produced with the World Bank’s\nLINKAGE model. At its core, LINKAGE is essentially a neo-classical growth model,\nwith aggregate growth predicated on assumptions regarding the growth of the labor force,\nsavings/investment decisions (and therefore capital accumulation) and productivity.\nUnlike more simple growth models\n\nMention: \"Luxemburg Income Study dataset\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Household information for developed countries comes from this dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003607:2:0", "occurrence_id": "305814d81efc5cece54a3f1d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ##### **Help or Hindrance?** **The Impact of Harmonized Standards on African Exports**\n\nWitold Czubala, Ben Shepherd and John S. Wilson <sup>*, †</sup>\n\nThe World Bank\n\n1818 H St. NW\nWashington D.C. 20433\n\n\n**Keywords:** International trade; Technical barriers to trade; Product standards; Economic\ndevelopment; Textiles and clothing.\n\n\n**JEL Codes:** F13; F15.\n\n\n- The first author is with the Bank of America, and second a Consultant and third author a Lead Economist in the\nDevelopment Research Group—Trade, The World Bank. This work is part of a broader project on trade facilitation\nand development supported through a Trust Fund of the U.K. Department for International Development.\nCompilation of the World Bank EU Standards Database (EUSDB) was supported by a Trust Fund of the U.S.\nAgency for International Development. Sincere thanks to the following for their superb assistance in assembling the\nWorld Bank EUSDB: Jérôme Beaufils, Teresa Bianchi, Françoise Blarez, Anne-Claire Hoyaux, Mark-André\nSaucier-Nadeau, and Mark Troger. We are also grateful to a number of World Bank colleagues for their comments\nand suggestions: Enrique Aldaz-Carroll, Silja Baller, Paul Brenton, and Bernard Hoekman, as well as participants at\na World Bank workshop on Trade Costs on April 30, 2007.  Comments to: <u>bshepherd@worldbank.org</u>\n(corresponding author), <u>jswilson@worldbank.org, or witold.czubala@gmail.com.</u>\n\n- The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not\nnecessarily represent\n\nMention: \"World Bank EU Standards Database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Its compilation was financed by a development trust fund.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003611:12:0", "occurrence_id": "05ab575ba304256f6e292328", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Madrasah teachers). The samples were drawn in two stages. 91 clusters <sup>1</sup> were selected at\n\n\nthe first stage as a subsample of the 361 clusters included in the Bangladesh\n\n\nDemographic and Health Survey (BDHS) of 2004. The second sampling stage selected\n\n\none adult from each household. Opinion leaders were selected from among those who\n\n\nwere resident in and around the cluster, having knowledge of and influences on the\n\n\npeople of the cluster. On average 49 adults and 5-6 opinion leaders were interviewed in\n\n\neach cluster. Out of the 49 adults interviewed in a cluster, roughly 16 were married\n\n\nwomen age 15-25, 16 married women age 45-59 and 17 married men age 25-50.\n\n\nInterviews were conducted in April-May 2006.\n\n\nWe have two estimation samples: older women (1431 initial observations) and\n\n\nyounger women (1543 initial observations). Explanatory variables are missing for some\n\n\nobservations, which cause a drop in sample sizes for the final/effective analyses samples.\n\n\nOur final samples thus are: older women (1408 observations) and younger women (1534\n\n\nobservations). Sample drops of these magnitudes do not seem to be cause for concern\n\n\nregarding the representativeness of the estimation samples. The means for the analyses\n\n\nsamples are reported in Table 2.\n\n\n[Table 2 about here]\n\n\nIn analyzing the difference in patterns between the two cohorts of women in the\n\n\nsample, we capture intergenerational change. Of course, it is entirely possible that the\n\n\ndifference could well be a function of age and life-cycle and not of cohort. That being so,\n\n\nwe believe that once we control for a number of demographic characteristics, we do\n\n\ncapture the effect of change over time.\n\n\n1 A cluster is a census defined village that corresponds roughly to a mouza village in rural areas and a\ncensus block (part of a mohollah) in an urban area.\n\n\n11\n\nMention: \"Bangladesh\n\n\nDemographic and Health Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its cluster frame supplies the first-stage sample for the follow-up survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003617:20:2", "occurrence_id": "68043a14994e31aa5f08ab55", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Source_: Briceño and Klytchnikova (2006)<br>_Note_: Data are the most recent available for 2000–04.|_Source_: Briceño and Klytchnikova (2006)<br>_Note_: Data are the most recent available for 2000–04.|_Source_: Briceño and Klytchnikova (2006)<br>_Note_: Data are the most recent available for 2000–04.|\n\n\n_How bad is the affordability problem?_\n\n\nThe access gap is only part of the problem. There is also an affordability issue.\nInfrastructure practitioners rely on rules of thumb to get a sense of the affordability\nproblem in any sector. One such rule of thumb (developed by the World Health\nOrganization) is that households should not need to spend more than 5% of their income\non their water and sanitations need—3.5 for water alone. In the case of electricity, there\nis no such formal ‘rule of thumb’ but many also assume that 4-6% is a maximum to\nspend on energy. Overall, the more general informal rule is the one that suggests that\npoor households should not have to spend more than 15% of their income on\ninfrastructure services.\n\n\nArmed with these rules of thumbs, it should be easy to get a sense of the extent of the\naffordability problem across the world. It is however impossible, because there is no\nsystematic formal monitoring of this crucial issue. There are however a number of recent\nbooks which have documented quite carefully the problem within Africa, Eastern Europe,\n\n\n22. In the Demographic and Health Survey data, the poorer and richer are defined based on an asset index used as a proxy\nof the welfare level. In the Living Standard Measurement Survey data, households are ranked by total per capita\nexpenditure.\n\n\n18\n\nMention: \"Demographic and Health Survey data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Wealth groups are defined from an asset index in these data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003617:20:2", "occurrence_id": "8fcff9841d2807f683fc2772", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Source_: Briceño and Klytchnikova (2006)<br>_Note_: Data are the most recent available for 2000–04.|_Source_: Briceño and Klytchnikova (2006)<br>_Note_: Data are the most recent available for 2000–04.|_Source_: Briceño and Klytchnikova (2006)<br>_Note_: Data are the most recent available for 2000–04.|\n\n\n_How bad is the affordability problem?_\n\n\nThe access gap is only part of the problem. There is also an affordability issue.\nInfrastructure practitioners rely on rules of thumb to get a sense of the affordability\nproblem in any sector. One such rule of thumb (developed by the World Health\nOrganization) is that households should not need to spend more than 5% of their income\non their water and sanitations need—3.5 for water alone. In the case of electricity, there\nis no such formal ‘rule of thumb’ but many also assume that 4-6% is a maximum to\nspend on energy. Overall, the more general informal rule is the one that suggests that\npoor households should not have to spend more than 15% of their income on\ninfrastructure services.\n\n\nArmed with these rules of thumbs, it should be easy to get a sense of the extent of the\naffordability problem across the world. It is however impossible, because there is no\nsystematic formal monitoring of this crucial issue. There are however a number of recent\nbooks which have documented quite carefully the problem within Africa, Eastern Europe,\n\n\n22. In the Demographic and Health Survey data, the poorer and richer are defined based on an asset index used as a proxy\nof the welfare level. In the Living Standard Measurement Survey data, households are ranked by total per capita\nexpenditure.\n\n\n18\n\nMention: \"Living Standard Measurement Survey data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Households here are ranked by per-capita total expenditure instead.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003618:27:23", "occurrence_id": "da30e98ad1118f1547385c31", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <br>(Millions of kilogram)<br>(7<br>|05<br>9)|(6|1)<br>|(58)|\n|||||||||||||||\n|||||||||||||||\n|||||||||||||||\n|||||||||||||||\n|||||||||||||||\n|||||||||||||||\n|||||||||||||||\n|||||||||||||||\n|||||||||||||||\n|||||||||||||||\n\n\n\nSource: _WITS COMTRADE_ database.\n\nMention: \"_WITS COMTRADE_ database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The tabulated trade flows are drawn from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003626:12:1", "occurrence_id": "21ca8e37ee6fceb3d54ad0d3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  population\n\n\napproaching 50 million people. The sampling process then consisted of three stages.\n\n\nFirst, a stratified random sample of 75 census tracts was selected. Second, interviewers\n\n\ncarried out a door-to-door listing within each census tract in order to determine which\n\n\nhouseholds had a Nikkei member. Third, the survey questionnaire was then administered\n\n\nto households identified as Nikkei. We now describe the details of each step.\n\n\n4 Population numbers from the 2000 Census are taken from\n<u>http://www.ibge.gov.br/english/estatistica/populacao/censo2000/</u> [accessed February 8, 2007].\n\n\n                               - - 11\n\nMention: \"the 2000 Census\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Population counts used to scale the estimates are taken from this enumeration.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003628:15:0", "occurrence_id": "545de49c0b8b73ad5a523108", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: marriage. People living in urban districts are generally more exposed to information\n\n\ncampaigns promoting the use of condoms and there is higher availability in urban\n\n\nareas. As expected, being currently and formally married is negatively and\n\n\nsignificantly associated with condom use in the last intercourse with the spouse.\n\n\nHowever, married women (both currently and formerly married) are less likely than\n\n\nsingle women to use a condom when they have extramarital sex. This is a potential\n\n\nsource of vulnerability which should be taken into consideration for prevention\n\n\nefforts. The same result was found in Burkina Faso, Cameroon and Kenya (De\n\n\nWalque 2006). Before the end of the apartheid regime in South Africa, about 60% of\n\n\nthe Lesotho male workforce was employed in the mining industry in South Africa.\n\n\nDuring this period anecdotal evidence suggests that some married women started to\n\n\nengage in sexual relationships with other partners. In the Demographic and Health\n\n\nSurvey, 11.6% of married women report having had extramarital sex in the last 12\n\n\nmonths. This is substantially higher than in other African countries (De Walque,\n\n2006) <sup>1</sup> [^1: It is not clear whether this is due to an actual difference in behavior or to a difference in reporting.] . The existence of long-term extra-marital relationships might increase the trust\n\n\nbetween partners and therefore the use of condoms might not appear necessary\n\n\nanymore.\n\n\nThe probability of using a condom inside and outside the marriage increases\n\n\nwith educational achievement. The relationship is robust, but statistically significant\n\n\nonly for individuals with secondary education or more. As expected, wealth positively\n\n\naffects condom use, both for sexual intercourse with the spouse and with other\n\n\npartners.\n\n\n**4.3 Fidelity in the last 12 months**\n\n\nThis section shows the results of the likelihood of having had extra-marital sex in the\n\n\nlast year.\n\n\n[Insert table 6]\n\n\nThe dependent variable we used in table 6 is equal to one if an individual had extra\n\nmarital sex during the last 12 months. We include only currently married or\n\n\n1\n\nMention: \"the Demographic and Health\n\n\nSurvey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"It reports the share of married women with extramarital partners.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003630:18:1", "occurrence_id": "a5f00fca99cbdcce6a05c2f6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  7: Internet Hosts per 10,000 Inhabitants in 2003\n\n\n300\n\n\n\n250\n\n\n200\n\n\n150\n\n\n100\n\n\n50\n\n\n0\n\n\n\n\n\n\n\n\n\n\n\n\n|276.3|Col2|Col3|Col4|\n|---|---|---|---|\n|276.3||||\n|||||\n|||||\n|||90.0|90.0|\n|||||\n|||||\n|0.82<br>13.91<br>18.55<br>0.39<br>1.010.98<br>5.9||||\n\n\n\nSource: Authors’ calculation based on International Telecommunication Union (2004) data\n\n\nImprovement in information technology services is as equally important as development\nof physical infrastructure. Biggs et al. (2000) examines the case in Nepal where service is\npoor and hinders business activities. For example, phone lines usually do not work well\noutside of the cities. Services are not available all day, and when available, voice quality\nis often poor (Biggs et al., 2000). Continued efforts to deregulate telecommunications,\nopen markets to competition and investment, and expand financing for infrastructure\n\n\n17\n\nMention: \"International Telecommunication Union (2004)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Indicators of information technology services are calculated from this source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003638:37:5", "occurrence_id": "b2a9a41e8276d51aba08fdac", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  0.032\n\n\n\nj= β i ∀ j=P,S,T 88.89 52 0.001 72.39 52 0\n\n\n\n**Pooling restrictions**\n\nH0: ρ i= ρ 24.06 25 0.516 29.0\n\n\n\nH0: ρ i= ρ 24.06 25 0.516 29.06 25\n\n\n\n.06 25 0.516 29.06 25 0.261\n\n\n\nj = β j ∀\n\n\n\n**ns**\n24.06 25 0.516 29.06 25 0.261\n52.64 50 0.372 61.37 50 0.130\n\n\n\nH0: β ij = β j ∀ j=P,T\n\n<u>H0:</u> <u>β</u> <u>ij =</u> <u>β</u> j <u>∀</u>\n\n\n\n64 50 0.372 61.37 50 0.\n\n\n\n\n<u>-</u> - - <u>96.16</u> 76 0.059\n\n\n\n<u>H0:</u> <u>β</u> <u>ij =</u> <u>β</u> j <u>∀</u> <u>j=S,T &</u> <u>π</u> <u>i=</u> <u>π</u> <u>=0</u> <u>-</u> - - <u>96.16</u> 76 0\n\n\n\nSource: PNAD 1985-2004; Author's calculation.\n\nNote: ** significant at 5%;* significant at 10%.\n\n\n\nS\n\n\n\n36\n\nMention: \"PNAD 1985-2004\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Coefficients with significance stars are computed across these rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003642:21:0", "occurrence_id": "0d3d2392d385cab1d44fbb7a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |Legal Status|Legal Status is a binary variable that measures whether an enterprise is registered as<br>a sole proprietorship or as a company. 0 indicates sole proprietorship, while 1<br>indicates the enterprise is registered as some form of company.<br>Source: World Bank Enterprise Surveys (www.enterprisesurveys.org).|\n|---|---|\n|Binary Bribe Index|Binary Bribe index is a variable that measures whether and enterprise has been<br>asked for an informal gift or payment in the past year for one or more of the<br>following: Connection/access to public utilities, obtaining business permits,<br>obtaining government contracts, to deal with safety, fire, or environmental<br>inspections, for tax purposes, for customs, courts, or to influence legislation. )<br>indicates no bribes were solicited, 1 indicates that at least one bribe was solicited.<br>On average, 68% of enterprises report at least one bribe.<br>Source: World Bank Enterprise Surveys (www.enterprisesurveys.org).|\n\n\n20\n\nMention: \"World Bank Enterprise Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Ownership form and bribe incidence are defined from this survey programme.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003646:25:1", "occurrence_id": "a560db92ea984e5b253eee57", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ’s policies and institutions by<br>analyzing 16 dimensions. Governance-related<br>dimensions are: Property rights and rules-based<br>governance; Quality of budgetary management; Quality<br>of public administration; Transparency, accountability &<br>corruption in the public sector<br>(Also serves as one of the sources for the KKM and<br>CPI)|2005|On a scale from 1 (worst) to 6<br>(best), Mauritania scored:<br>3.0 for property rights and rules-<br>based governance;<br>2.0 for quality of budgetary<br>management;<br>3.0 for quality of public<br>administration;<br>2.5 for transparency,<br>accountability & corruption<br>|\n|**Business Environment**<br>**Indicators**|Doing Business<br>(DB)<br>|Evaluates ease of doing business by monitoring ten<br>categories:<br>1. Starting a business; 2. Dealing with licenses; 3.<br>Employing workers; 4. Registering property 5. Getting<br>credit; 6. Protecting investors; 7.Paying taxes; 8.<br>Trading across borders; 9. Enforcing contracts; 10.<br>Closing a business.|2006|Overall rank: 148 out of 175<br>In top half for Registering<br>property and Enforcing<br>contracts<br>Among lowest performers in<br>Starting a business and Paying<br>taxes|\n|**Business Environment**<br>**Indicators**|Investment Climate<br>Survey (ICS)|Survey of private sector firms and employees to<br>evaluate the overall business environment\n\nMention: \"KKM\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"One underlying source feeds this composite governance measure.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003646:25:1", "occurrence_id": "317ea033725331cb8bf1eb48", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ’s policies and institutions by<br>analyzing 16 dimensions. Governance-related<br>dimensions are: Property rights and rules-based<br>governance; Quality of budgetary management; Quality<br>of public administration; Transparency, accountability &<br>corruption in the public sector<br>(Also serves as one of the sources for the KKM and<br>CPI)|2005|On a scale from 1 (worst) to 6<br>(best), Mauritania scored:<br>3.0 for property rights and rules-<br>based governance;<br>2.0 for quality of budgetary<br>management;<br>3.0 for quality of public<br>administration;<br>2.5 for transparency,<br>accountability & corruption<br>|\n|**Business Environment**<br>**Indicators**|Doing Business<br>(DB)<br>|Evaluates ease of doing business by monitoring ten<br>categories:<br>1. Starting a business; 2. Dealing with licenses; 3.<br>Employing workers; 4. Registering property 5. Getting<br>credit; 6. Protecting investors; 7.Paying taxes; 8.<br>Trading across borders; 9. Enforcing contracts; 10.<br>Closing a business.|2006|Overall rank: 148 out of 175<br>In top half for Registering<br>property and Enforcing<br>contracts<br>Among lowest performers in<br>Starting a business and Paying<br>taxes|\n|**Business Environment**<br>**Indicators**|Investment Climate<br>Survey (ICS)|Survey of private sector firms and employees to<br>evaluate the overall business environment\n\nMention: \"CPI\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The same source also feeds this corruption index.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003649:4:0", "occurrence_id": "77f699d9517e301e48f002dd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: of exogenous changes to which land markets need to respond increases disproportionately. The\n\n\ntransaction costs associated with renting out land may well affect land owners’ decisions on whether to,\n\n\ne.g., adopt new technology or take up off-farm employment, thus having implications for nonfarm\n\n\ngrowth.\n\n\nGiven its limited land endowment that is under increasing threat from degradation, making the most\n\n\nproductive use of available land resources is important for Ethiopia to escape the threat of starvation and\n\n\ndependence on food aid. Similarly, the population’s continued high level of dependence on agriculture\n\n\nimplies that well-functioning land rental markets that would allow those with little comparative advantage\n\n\nin agriculture to take up nonagricultural employment will be critical to lay the basis for overall\n\n\ndevelopment and a more diversified rural economy. If the experience of other countries that followed\n\n\nsimilar strategies is anything to go by, land rental markets will have to assume a major role in allowing\n\n\nmovement of labor out of agriculture, and transferring land to more productive producers. However, past\n\n\nstudies of the extent efficiency of rental markets in Ethiopia yield contradictory results; while some are\n\n\nunable to reject the hypothesis of frictionless adjustment through rental markets (Pender and Fafchamps\n\n\n2006), others provide strong evidence for land markets allowing at best partial adjustment (Teklu and\n\n\nLemi 2004, Ghebru and Holden 2006, Holden and Ghebru 2006).\n\n\nIn this paper, we use a large panel data set from Ethiopia that contains a substantial share of\n\n\ncultivating households owned and sharecropped land simultaneously, in addition to being more detailed\n\n\nand larger than what had been available in other studies. To assess the efficiency of sharecropping as a\n\n\nkey contractual arrangement in land markets, we estimate a production function and input demand\n\n\nfunctions using household fixed effects. To explore the extent to which land markets help farmers\n\n\ncapitalize on existing opportunities, we use a friction model (Rosett 1959, Skoufias 1995) to identify\n\n\nwhether rental markets allow producers to attain their desired level of\n\nMention: \"large panel data set from Ethiopia\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Households holding owned and sharecropped land simultaneously are observed in it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003659:19:5", "occurrence_id": "316d7757423fa00e0cef5935", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  (table A.4, row 1).\nb/ Average annual growth rate over the period in parenthesis\nc/ HI stands for “High-Income” countries as defined by World Bank (July 2007);\n“HI” sample = 32 fiscal events, “non HI” sample = 63 events.\n\n_<u>Source: Authors’ computation from GFS and PWT 6.2 data, see Annex A.2.</u>_\n\n\n18\n\nMention: \"GFS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Fiscal event variables are computed from these finance statistics.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003659:19:5", "occurrence_id": "5afa732ac437e4bd01be7bd8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  (table A.4, row 1).\nb/ Average annual growth rate over the period in parenthesis\nc/ HI stands for “High-Income” countries as defined by World Bank (July 2007);\n“HI” sample = 32 fiscal events, “non HI” sample = 63 events.\n\n_<u>Source: Authors’ computation from GFS and PWT 6.2 data, see Annex A.2.</u>_\n\n\n18\n\nMention: \"PWT 6.2 data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its version supplies the output and price controls.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003663:41:2", "occurrence_id": "a36793d734d99c27b2da258c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 3 1.7 **0.9**\nBrazil **1.7** 0.3 **0.0** 0.4 0.3 **0.2**\nFinland **1.5** 0.9 **0.2** 1.8 0.7 **0.5**\nSpain **1.4** 0.6 **0.5** 0.4 0.6 **1.4**\nAustralia **1.3** 1.3 **2.0** 0.6 0.4 **2.6**\nNorway **1.2** 0.7 **0.4** 0.7 0.7 **0.2**\nSingapore **1.2** 1.9 **1.5** 1.7 2.5 **0.6**\nKorea, Rep. **1.2** 0.4 **0.0** 0.0 0.8 **0.9**\n<u>Other</u> **<u>0.0</u>** <u>21.9</u> **<u>24.0</u>** <u>21.5</u> <u>30.1</u> **<u>34.8</u>**\n\nSource: IMF Direction of Trade Statistics\nNote: The twenty-three IMF trading partners are shown, while the remaining twenty-seven trading\npartners captured in the flexible weighting scheme are grouped in the other category.\n\nDespite the differences in the two weighting schemes, using a flexible weighting\nscheme appears to have very little impact on measures of Tanzania’s REER. Figure 1\nshows the\n\nMention: \"IMF Direction of Trade Statistics\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Bilateral trading partner shares are drawn from this compilation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003664:7:1", "occurrence_id": "d9149ac50787cc1ff58a51a8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: >65|20<br> <br>69<br>17<br>3 <br>25<br>24<br> <br>42<br>5 <br>7 <br>30<br> <br>131|33<br> <br>105<br>20<br>4 <br>30<br>51<br> <br>58<br>7 <br>8 <br>43<br> <br>196|\n\n\n\nNotes: Food is defined as raw food in SITC Revision 2, excluding all cash crops,\n\nprocessed food and seafood (see detailed classifications in Appendix Table 1).\n\nThe classification of country groups is provided in Appendix Tables 2 and 3.\n\n\nSource: Based on partner data from UN COMTRADE Statistics.\n\n\nAs expected, most of the countries in the world are net food importers, i.e. 131 out of\n196. This ratio is consistent with any other product group where the exporters tend to be\nmore specialized than the importers. 69 middle-income countries out of 105 and 42 out of 58\nlow-income countries are net food importers.\n\n\nIn terms of percentages, 39 percent of industrial countries, 34 percent of middle-income\ncountries, and 28 percent of low-income countries are net food exporters. If we exclude oil\nexporters, countries in conflict, and island states, these ratios increase to almost 51 percent\n\n\n11 There are some islands and territories that belong to industrial countries such as US or British Virgin Islands\netc. We have excluded these countries.\n\n\n6\n\nMention: \"UN COMTRADE Statistics\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Partner-level flows identify net food importers across nearly two hundred countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003666:7:0", "occurrence_id": "19f3b0fb0acc7165184e742c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: - 6 \n\nIn the 3G mobile telecommunications experiences in European countries, for instance, the\n\n\nnumber of bidding firms per license ranges from 0.75 to 2.6. In the United Kingdom, five\n\n\nlicenses were auctioned among 13 bidders. On the other hand, there was no competition in\n\n\nthe Austrian and Swiss cases, meaning that the number of bidders was eventually the same as\n\n\nthe number of licenses to be issued (Klemperer, 2002; van Damme, 2002). NOA (2007)\n\n\nindicates that half of the past U.K. PFI projects attracted at most two or three bids. Similarly,\n\n\nin the railway concession cases in Argentina and Brazil during the 1990s, two to four firms\n\n\nparticipated in the bidding process. Particularly, the freight train contracts, which may be less\n\n\nprofitable from the private sector point of view, attracted only one or two bidders.\n\n\nHow many bidders are required for sufficient competition? Gupta (2002) shows that the\n\n\nhighway construction market (in Florida) becomes competitive with about six to eight\n\nbidders. <sup>3</sup> [^3: It means that the marginal impact of one more bidder on the equilibrium bid is not statistically significant\nwhen the number of participants exceeds eight.] As per Iimi (2006), large-scale official development projects—mostly in the\n\n\ninfrastructure sector—require eight firms to minimize the contract prices. In the U.S.\n\n\noffshore oil lease market, a similar level of bidder participation seems necessary for\n\n\ncompetition purposes (Brannman _et al_ ., 1987).\n\n\nDeveloping countries may have an even more limited competition in PPP infrastructure\n\n\nauctions. The PPI Database shows that about 60 percent of PPP infrastructure transactions in\n\ndeveloping countries involved one to three bidders (Figure 1). <sup>4</sup> [^4: The database covers only transactions of which total investment commitments exceed US$ 1 million.] About 15 percent of contracts\n\n\nattracted only one bidder even under the competitive bidding circumstances. The degree of\n\n\ncompetition varies\n\nMention: \"PPI Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"It shows most infrastructure transactions involved only a few bidders.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003661:19:2", "occurrence_id": "f00c2adfba4dc25c4504b918", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: >20.92%</u> <u>51.95%</u> <u>22.30%</u> <u>4.83%</u> <u>100.00%</u>\n_Source:_ Own calculation based on NBS/World Bank survey (landlords’ and tenants’ assessments for 1106 contracts).\n\n\n18\n\nMention: \"NBS/World Bank survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Landlord and tenant assessments across eleven hundred contracts are tabulated from it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003668:9:0", "occurrence_id": "7ceda4788ca405e6d4627d70", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: move to the location of the consumer (or vice versa). As the conventional definition of\n\n\ntrade – where a product crosses the frontier – would miss out on the latter type of\n\n\ninternational transactions, the WTO has defined trade to span four modes of supply:\n\n\n  - _Mode 1 – Cross-border_ : services supplied from the territory of one country into\nthe territory of another.\n\n  - _Mode 2 – Consumption abroad_ : services supplied in the territory of a country to\nthe consumers of another.\n\n  - _Mode 3 – Commercial presence_ : services supplied through any type of business\nor professional establishment of one country in the territory of another (i.e., FDI).\n\n  - _Mode 4 – Presence of natural persons_ : services supplied by nationals of a\ncountry in the territory of another.\n\n  \n\nData on services trade are notoriously weak. The primary source is the balance of\n\n\npayments, where measured flows correspond mainly to modes 1 and 2. IMF data reveal\n\nthat service imports roughly tripled over the decade from 1994 to 2004. <sup>11</sup> For the 178\n\n\ncountries for which the IMF reports data the median value of imports of services in 1994\n\n\nwas some US$550 million. As of 2004, this had increased to US$1.5 billion. For the\n\n\nworld’s largest services trader, the US, total reported imports rose from almost US$200\n\n\nbillion to US$340 billion.\n\n\nNotwithstanding this growth in the absolute value of services trade, world service\n\n\nimports as a percentage of GDP rose slightly from 11 percent in 1994 to 12 percent of\n\n\nGDP in 2004 (Table 1). This masks declines in the services share in developing\n\n\ncountries. The rising importance of service imports in the aggregate stems mainly from\n\n\nincreases within the most developed countries (from 8 percent to 12.7 percent), driven by\n\n\nstrong growth in producer and business services imports (whose ratio to GDP rose from\n\n\n3.3 percent to 6.7 percent\n\nMention: \"IMF data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its balance of payments records show service imports roughly tripling over the decade.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003670:6:1", "occurrence_id": "4225355955bdf91be3b4f1aa", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  these differences—the high inequality (in various dimensions) and low\n\n\npopulation density—have probably influenced another evident difference to China: African\n\n\ncountries tend to have weaker state institutions, with an adverse feedback effect on ( _inter alia_ )\n\nthe quality and quantity of key social services and infrastructure. <sup>16</sup>\n\n\nThese are all important differences between China around 1980 and the typical African\n\n\ncountry today. However, even if there were no such differences, it is also relevant that SSA is 48\n\n\ncountries not one. There is a quite fundamental difference in the degree of internal (economic,\n\n\n10 For recent evidence on this point using country-level panel data see Ravallion (2007).\n11 Tabulations from the 1982 Census indicate that 36% of the population of China were in the 0-15 age group.\n12 This process of demographic transition probably began around 1970. Note that 1980 was also the year in\nwhich the controversial “one-child policy” was introduced, though with the widespread exceptions allowed locally\n(particularly in rural areas); the target level of fertility is closer to 1.5 children per couple; see Baochang et al.\n(2007). While fertility rates had started to fall well before the one-child policy, there can be little doubt that this\npolicy helped in bringing the birth rate below replacement levels from the early 1990s.\n13 See Cleland et al. (2006).\n14 For a critical perspective on Herbst’s theory see Robinson (2002). Also see the discussion in Clapham\n(2001) on the role played by colonialism and post-colonial international relations.\n15 For evidence on this point at a global level see Klasen and Nestmann (2006).\n16 The strength of the Chinese state (at all levels) is evident; on the weaknesses of most African states see\nHerbst (2000), Clapham (2001) and van de Walle (2001).\n\n\n5\n\nMention: \"1982 Census\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its tabulations show over a third of the population aged under sixteen.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003679:9:1", "occurrence_id": "9c504495fbbc336a2cde85d0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  2007\n\n\n\n**Private investment fell despite rising growth.** The private sector found serious\n\n\nimpediments to investing in the poor transportation network, the declining quality of\n\neducation and the high cost of inputs, particularly electricity. <sup>9</sup> Between 2000 and 2004,\n\n\nprivate investment - deterred by political instability and systemic uncertainties - did not\n\n\nrespond to higher growth (Figure 4 and Figure 5), and private sector credit fell by 40 percent\n\n\nas a share of GDP.\n\n\n7 In recent years, foreign investments have grown from a small base (from 2005 to 2006, net FDI inflows\nhave reached $2 billion, growing to 14 percent of the overall private investment), but are still too small to\nmatter, and many investment approvals have not translated into actual flows.\n8 There is evidence that busts in the real estate tend to be longer-lived than in other asset markets, and tend\nto be associated with more severe downturns in overall economic activity. For example, in industrial\ncountries, the evidence for housing market busts suggests they have an average duration of 4-5 years\ncompared to 2-3 years for equity market busts (World Bank, 2006a).\n9 Based on the competitiveness ranking of the World Economic Forum on basic infrastructure, the\nPhilippines placed 89th out of 102 countries surveyed.\n\n\n10\n\nMention: \"the World Economic Forum\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its infrastructure competitiveness ranking places the country eighty-ninth of one hundred two.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003680:4:0", "occurrence_id": "d72c4eb6a9dbf734c02e9ea6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: #### **2. STYLIZED FACTS**\n\nThis section presents stylized facts on the evolution of export diversification which\nhelp us to disentangle the relative importance of the intensive and extensive margins\nof trade. We begin with a description of the dataset used for both, the stylized facts,\nand the econometric analysis.\n\n\n2.1. **_The dataset_**\n\n\nThe ideal dataset to deal with export diversification issues from a microeconomic\npoint of view would consist of firm-level export data for each country pair; however,\nsuch data are not available to researchers. We use instead the highest level of\ninternationally comparable disaggregated country-level trade data that are currently\navailable, namely the 6-digit level of the Harmonised System (HS) from Comtrade\ndatabase. <sup>3</sup> [^3: For completeness we use mirror data (imports) from Comtrade.] Our time frame is the period from 1990 to 2005, and the sample includes a\nselection of 24 developed and developing economies. The set of countries\nencompasses Argentina (ARG), Australia (AUS), Bangladesh (BGD), Brazil (BRA),\nCanada (CAN), Chile (CHL), China (CHN), EU (EEC15), Ghana (GHA), Indonesia\n(IDN), India (IND), Japan (JPN), Kenya (KEN), Morocco (MAR), Mexico (MEX),\nMauritius (MUS), Malaysia (MYS), Peru (PER), Thailand (THA), Tunisia (TUN),\nTurkey (TUR), Uganda (UGA), USA and South Africa (ZAF).\n\nThe group of countries that has been selected accounts for a very significant share of\nworld trade. In year 1995, the selected group of countries accounted for two-thirds of\nworld trade. When looking at trade among developing countries alone, the developing\ncountries included in the sample also account for two-thirds of the total amount of\ntrade\n\nMention: \"Comtrade\ndatabase\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Six-digit Harmonised System country-level flows are the most disaggregated available here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003682:28:0", "occurrence_id": "b162b271ecb75e7ec9989fb7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Monitoring and evaluation**\n\n\n69. Financing is likely to come from multiple sources. Timely and consistent\ntracking of budgets/commitments and execution is a critical ingredient for assessing\nreconstruction progress. Financial information is especially sensitive, particularly if\ndonors and private contributions are high. An on-going assessment of commitments and\ndisbursements across all sources (public/private, domestic/international) would highlight\nthe extent of any reconstruction delays and potential financing gaps. Good PFM systems\nwould then measure whether the financial resources are being translated into outcomes.\nWhen execution is decentralized across various types of domestic or international\nagencies or local governments, data systems need to pay particular attention to proper\naccounting. For instance, commitments need to be separated from disbursements,\nemergency projects from reconstruction projects, and financing institutions from\nimplementing partners (Agustina, 2007).\n\n\n70. In recent years, several “tracking systems” have been developed to monitor\nfinancial information and improve aid management of the recovery process. The\nexistence of credible and integrated financial tracking systems has become more critical\nin recent years because of the unprecedented levels of funding after the tsunami, together\nwith other large-scale post-conflict reconstrucion efforts (e.g. Afghanistan and Sudan)\nwith large funding.\n\n\n71. The most prominent aid management and reconstruction tracking instrument is\nthe Development Assistance Database (DAD). The DAD has been applied in a number of\ncountries including Afghanistan, Sri Lanka, the Maldives and Indonesia. However,\ncoverage has been limited, often focusing on UN and other core donor activities.\nAlthough the DAD includes sophisticated technical specifications, including visual\nbreakdowns to the village level, it has faced many implementation challenges,\nparticularly in Aceh and Nias.\n\n\n72. While information technology has sometimes been an impediment to the\neffective implementation of information systems, it has typically not been the main\nobstacle for effective monitoring. Instead, the main challenges are related to data\ncollection and analysis:\n\n  - In Aceh and Nias, a labor\n\nMention: \"Development Assistance Database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It is the leading instrument for tracking aid and reconstruction across several countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003683:8:0", "occurrence_id": "75f8a07f090adfda63e5d37b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: privately owned companies, since publicly owned companies are likely to face structurally\n\n\ndifferent credit/informality trade offs.\n\n\nOne of the critiques often raised to the IC surveys is that they are not fully representative\n\n\nof the country firms’ size distribution. In particular, since in most countries firms’ distribution is\n\n\nheavily skewed towards micro and small firms, large firms were over-sampled to allow\n\n\nmeaningful analysis for all firm categories. This is a potential concern for our analysis: if smaller\n\n\nfirms are more informal than large firms, (i.e. they report lower share of sales for tax purposes)\n\n\nwe are likely to underestimate aggregate tax compliance. As most surveys do not report weights,\n\n\nwe address this concern by weighing tax compliance in each firm category (micro, small,\n\n\nmedium, large, and very large) using data from the International Financial Corporation (IFC)\n\n\nSME database as well as other individual census data on the actual distribution of firms by size in\n\n\neach country (see discussion further on).\n\n\nTo proxy for informality, we use the percentage of sales that firms report to declare for tax\n\n\npurpose (FORMALITY). The distinct advantage of using this variable is that, unlike most\n\n\nindirect estimates of informality, FORMALITY captures a well defined concept: tax compliance.\n\n\nClearly, the complementary drawback is that we measure only a portion of the informal economy.\n\n\nAlthough it is likely that FORMALITY suffers from measurement error, the lack of alternative\n\n\naudit-based measures of tax compliance that can be compared homogenously across a large\n\n\nnumber of developing countries makes this variable a particularly useful starting point to address\n\n\nthese issues.\n\n\nDescriptive statistics reported in tables 1-3 indicate that the degree of tax compliance\n\n\nincreases linearly with firm size, ranging from an average of 72% of sales declared in micro firms\n\n\nto an average of 86% in very large firms. Tax compliance is higher in high-tech technology\n\n\nindustries, particularly in “metals and machinery” and “chemical\n\nMention: \"IC surveys\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"They are criticised as unrepresentative of national firm size distributions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003683:8:0", "occurrence_id": "bd35566029abf1100e31a65c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: privately owned companies, since publicly owned companies are likely to face structurally\n\n\ndifferent credit/informality trade offs.\n\n\nOne of the critiques often raised to the IC surveys is that they are not fully representative\n\n\nof the country firms’ size distribution. In particular, since in most countries firms’ distribution is\n\n\nheavily skewed towards micro and small firms, large firms were over-sampled to allow\n\n\nmeaningful analysis for all firm categories. This is a potential concern for our analysis: if smaller\n\n\nfirms are more informal than large firms, (i.e. they report lower share of sales for tax purposes)\n\n\nwe are likely to underestimate aggregate tax compliance. As most surveys do not report weights,\n\n\nwe address this concern by weighing tax compliance in each firm category (micro, small,\n\n\nmedium, large, and very large) using data from the International Financial Corporation (IFC)\n\n\nSME database as well as other individual census data on the actual distribution of firms by size in\n\n\neach country (see discussion further on).\n\n\nTo proxy for informality, we use the percentage of sales that firms report to declare for tax\n\n\npurpose (FORMALITY). The distinct advantage of using this variable is that, unlike most\n\n\nindirect estimates of informality, FORMALITY captures a well defined concept: tax compliance.\n\n\nClearly, the complementary drawback is that we measure only a portion of the informal economy.\n\n\nAlthough it is likely that FORMALITY suffers from measurement error, the lack of alternative\n\n\naudit-based measures of tax compliance that can be compared homogenously across a large\n\n\nnumber of developing countries makes this variable a particularly useful starting point to address\n\n\nthese issues.\n\n\nDescriptive statistics reported in tables 1-3 indicate that the degree of tax compliance\n\n\nincreases linearly with firm size, ranging from an average of 72% of sales declared in micro firms\n\n\nto an average of 86% in very large firms. Tax compliance is higher in high-tech technology\n\n\nindustries, particularly in “metals and machinery” and “chemical\n\nMention: \"SME database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its size distribution data support the stratification into five firm classes.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003692:12:0", "occurrence_id": "b2962a96d2ae9c35bc2eb8fe", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: males in this state. This is likely to be explained by females having household chores and\ntaking care of (their) children. Some 15 percent of youth have dropped out from school—\nrather than graduated. Dropping out of school is a serious risk factor, because it is\nconnected with other risk factors and negative outcomes. For example, Eckstein and\nWolpin (1999) argue that dropouts have different unobserved traits than those completing\nan education, such as lower motivation, ability or expectations of return, higher value of\nleisure, and lower consumption value of attending school. Lack of education and having\nthese traits likely affect the chance of a positive outcome negatively in a number of ways,\nthrough the relation to jobs and income, but also through its impact on safe sex or\ndomestic violence where education may help reduce risk, because of precautionary\nmeasures.\n\n\n\n\n\n\n\n\n\n|Figure 3.3: Activity for Female Youth in<br>Argentina, 2005|Figure 3.4: Activity for Male Youth in Argentina,<br>2005|\n|---|---|\n|<br>In School<br>66%<br>Work<br>16%<br>Idle /<br>Unem-<br>ployed<br>18%<br>_Source_: Own calculations on ESCVJA (four urban areas)|<br>Idle / Unem-<br>ployed<br>8%<br>Work<br>26%<br>In School<br>66%<br>_Source_: Own calculations on ESCVJA (four urban areas)|\n\n\nChild and teenage work are often considered significant risk factors, because they\ncan have devastating effects on human capital accumulation over time (see, e.g., Fares\nand Raju 2007). In the sample analyzed here, 18 percent of female youth and 27 percent\nof male youth had their first job before\n\nMention: \"ESCVJA\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"School attendance, work and idle shares are calculated for four urban areas.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003696:17:0", "occurrence_id": "8428f0c3701bb84515fe096e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: and I omit any analysis of farm employment and rural labor markets. Moreover, and\n\n\npartly for this reason, I restrict the analysis to the impact of changes in the prices of\n\n\nagro-manufactures exports, leaving aside the impacts of changes in the export prices of\n\n\nagricultural commodities. <sup>9</sup>\n\n\nTable 1 reports the main trends in three variables of interest: the employment status of\n\n\nthe individual, monthly wages, and the level of the price of agro-industrial exports. In the\n\n\nTable, I report data on all workers in all sectors. <sup>10</sup> The sample is restricted to include all\n\n\n“employed” workers, leaving aside the self-employed. Sample sizes of employed, unemployed\n\n\nand inactive individuals of working age are reported as well.\n\n\nThe average wage of unskilled workers (primary education) increased until 1995 and\n\n\ndeclined afterwards to reach, in 1999, a level comparable to that in 1992. The wages of\n\n\nsemi-skilled workers (secondary education) increased until 1996, and then stabilized. In\n\n\n1999, average wages declined, but were still higher than in 1992. The wages of skilled\n\n\nworkers (college education) fluctuated more: they increased until 1995, declined in 1996,\n\n\nand then recovered again until 1998. In 1999, wages are higher than in 1992, indicating\n\n\npositive growth during the period.\n\n\nThe Argentine Statistical Institute provides data on the international prices of Argentine\n\n\nexports of agro-manufactured goods, which include dairy product, beef, oils and fats, and\n\n\nfood-processing goods. These export prices show an increasing trend (with fluctuations)\n\n\nfrom 1992 to 1997, and a decline in more recent years. <sup>11</sup>\n\n\nIn the EPH survey, there are three employment categories: employed, unemployed and\n\n\ninactive. The employment categories are defined on a weekly basis—this type of reporting is\n\n\n15\n\nMention: \"EPH survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its three employment categories define the measured activity statuses.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003700:21:0", "occurrence_id": "9543ada059a2dbdfc8a49880", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: In conclusion, between 60 and more than 80 percent of variability in the global\n\n\nincome position can be attributed to the circumstances beyond individual control. Sixty\n\n\npercent represents the lower bound where only the mean income of the country and\n\n\ncountry’s inequality are allowed to play a role. Eighty percent or more is obtained when\n\n\nwe include person’s parental income as part of circumstance, and use either base-case or\n\n\npessimistic assumptions regarding income mobility in various parts of the world. In any\n\n\ncase, the part which remains for effort and “episodic luck” (to use John Roemer’s\n\n\nfelicitous phrase) remains relatively small.\n\n\nFinally, we compare the actual role of location to a hypothetical case where all\n\ncountries’ mean incomes are equal. <sup>24</sup> [^24: This is the situation referred by Roemer (2007) as Equality of opportunity of degree 1. Incidentally, if\nall mean incomes were equalized the global Gini would be only 37.4 vs. the actual Gini of 64.2 (based on\nthe World Income Distribution 2002 dataset).] We still “allocate” people to different countries and\n\n\nincome classes in our Rawlsian lottery, but now location implies only a difference in the\n\n\nincome distributions between the countries (different Gini coefficients), not the\n\n\ndifference in the average income. The results are shown in column (9) Table 3. The\n\n\ncoefficient on income class more than doubles compared to regression (3), and when we\n\n\ndecompose the two effects, income class is found to explain more than 90 percent of\n\n\nvariability in the global income position, while location (through its specific inequality)\n\naccounts for less than 5 percent. <sup>25</sup> [^25: Historically, something similar might have obtained in the early 19th century when, according to the\nBourguignon and Morrisson (2002) study of the long-run global inequality, class (within-national\ninequality) explained about 90 percent of world inequality between individuals.] This counterfactual also allows us\n\nMention: \"the World Income Distribution 2002 dataset\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Its figures give the actual and equalised global Gini coefficients.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003704:27:4", "occurrence_id": "7a939e05d431a1cc09e9acd9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 359 6,279 1.8 1.0 0.6 44.8 35.4 25.1 1.8 -2.9 -17.2 2.5 1.7 1.8\nSouth America 9,177 16,908 18,004 2.5 1.6 1.9 87.0 86.2 83.5 3.4 2.6 -0.1 4.6 2.6 1.8\nChina 13,424 97,361 146,259 3.6 9.5 15.1 94.2 99.1 99.5 18.6 17.0 12.0 8.2 4.6 3.6\n<u>Rest of the World</u> <u>318,274</u> <u>773,822</u> <u>674,749</u> <u>85.6</u> <u>75.1</u> <u>69.4</u> <u>88.7</u> <u>90.5</u> <u>87.2</u> <u>6.3</u> <u>3.2</u> <u>-6.5</u> <u>5.2</u> <u>2.7</u> <u>2.5</u>\n_<u>Notes</u>_ <u>: Annual real growth rates calculated using US CPI as deflator. Average tariffs are calculated duties divided by the value of imports.</u>\n\n_Source_ : Authors' calculations based on US Customs Data.\n\n\n26\n\nMention: \"US Customs Data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Effective tariffs are computed as duties collected over import values from these records.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003707:27:0", "occurrence_id": "a108b85a60de8825bf8d60b0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 3: Height for Age Z-Scores, By Months of War Exposure\n\n\n\n\n\nNotes: Kernel-weighted local polynomial regression (using Epanechnikov kernel) of height for age zscores on months of civil war exposure. Data source: World Bank and Burundi Statistics Institute 1998\nPriority Survey.\n\n\n25\n\nMention: \"1998\nPriority Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Child nutrition z-scores linked to civil war exposure come from this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003714:24:0", "occurrence_id": "5b87d49f1d9452156831bb16", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 1: Estimated male and female hazards of ever—migration, Albania 2005 LSMS**\n\n\n\n22\n\nMention: \"LSMS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Male and female migration hazards are estimated from this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003719:8:1", "occurrence_id": "9567fa88f432b55da1e426ad", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  with at least one member\nparticipating in productive and non-productive associations. Column 1 indicates that\nnationally about 6.4% of all households have at least one member participating in a\nproductive association, the majority of which reside in rural areas (11% in rural areas vs.\n6% in urban areas). Columns 2, 3, and 4 display respectively the share of households\nparticipating in local committees, credit unions, and professional associations by regions\nand socio-economic group. Similarly to what occurs nationally, participation in local\ncommittees and in professional associations is higher in rural areas and among the nonpoor. Household participation in credit unions is larger in urban areas. Column 5 presents\nthe share of households having at least one member participating in non-productive\nassociations, mainly churches and clubs.\n\nTable 2 provides descriptive statistics on access to social programs and credit.\nThe 2005 EMNV includes a module that gathers information on whether or not\nhouseholds benefited from social programs. Out of a list of 23 programs, we identified 2\nsocial assistance programs that provide free or subsidized assets to promote investment\namong the poor: agricultural programs and housing programs. Agricultural programs\ninclude information on three ongoing programs in Nicaragua: _libra por libra_, _huertos_\n_caseros_, and the _forestation_ programs. Data do not permit to analyze each program\nindependently. These programs provide households working in agriculture with some\ntype of investment asset; such as improved seeds, plants, and trees. Housing programs\nprovide households with land, labor, and/or materials to build or remodel their dwelling.\nColumns 1 and 2 in Table 2 display the share of households benefiting from the\naforementioned programs. Descriptive statistics indicate that only about 3 (1) percent of\nall households claim to have benefited from agricultural (housing) programs; most of\n\nMention: \"2005 EMNV\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its module records household benefit from social programmes and credit access.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003726:5:3", "occurrence_id": "c88566cfc4a3ac28688b2a34", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: eychelles<br>0.1<br>2.3<br>Côte d'Ivoire<br>-0.3<br>2.0<br>Zimbabwe<br>-5.5<br>-2.6<br> <br> <br> <br>   Average<br>1.8<br>2.3<br> <br> <br> <br> <br> <br> <br> <br> <br> <br> <br> <br> <br> <br> <br> <br> <br> <br>|\n|Source:World Bank World Development Indicators database.<br>1The cut-off point of 4 percent is based primarily on the performance during 2000-06. Somalia is not included for lack of data. At the World<br>Bank, Djibouti is classified as part of Middle East and not part of Sub-Saharan Africa.<br>*Data for Liberia was available only for 1999-2006.|Source:World Bank World Development Indicators database.<br>1The cut-off point of 4 percent is based primarily on the performance during 2000-06. Somalia is not included for lack of data. At the World<br>Bank, Djibouti is classified as part of Middle East and not part of Sub-Saharan Africa.<br>*Data for Liberia was available only for 1999-2006.|\n\n\n**_B._** **_Performance by different types of countries_**\n\nNonetheless, the range of growth among individual countries can vary widely, anywhere\nfrom −5.5 percent to a high of 23.1 percent. Alternative taxonomies of countries are often\nnecessary to characterize the growth story better. Table 2 presents the growth\nperformance for frequently used country groups for the 1980s, the 1990s\n\nMention: \"World Bank World Development Indicators database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Growth comparisons over this window are drawn from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003728:11:1", "occurrence_id": "6a9370c2cbc74b85df8de1c7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  are estimated with the survey data set. The regressors\n\n\ncontain only the variables in the geographic data set or the variables that also appear in\n\n\nthe census data set. The left-handside variable is then imputed to each census record\n\n\nand aggregated to obtain poverty and inequality measures of interest. Using a Monte\n\n2 According to the World Development Indicators, the Purchasing Power Parity conversion factor (for\n\n\n1998) was USD 1=VND 2,673.\n\n\n10\n\nMention: \"survey data set\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Regressions restricted to geographic variables are estimated on it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003728:11:1", "occurrence_id": "bd7452b7751e2cb60e196abd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  are estimated with the survey data set. The regressors\n\n\ncontain only the variables in the geographic data set or the variables that also appear in\n\n\nthe census data set. The left-handside variable is then imputed to each census record\n\n\nand aggregated to obtain poverty and inequality measures of interest. Using a Monte\n\n2 According to the World Development Indicators, the Purchasing Power Parity conversion factor (for\n\n\n1998) was USD 1=VND 2,673.\n\n\n10\n\nMention: \"census data set\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Imputed outcomes are aggregated over its records to obtain poverty measures.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003728:11:1", "occurrence_id": "5ce8bac2ff3905e49d2f2653", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  are estimated with the survey data set. The regressors\n\n\ncontain only the variables in the geographic data set or the variables that also appear in\n\n\nthe census data set. The left-handside variable is then imputed to each census record\n\n\nand aggregated to obtain poverty and inequality measures of interest. Using a Monte\n\n2 According to the World Development Indicators, the Purchasing Power Parity conversion factor (for\n\n\n1998) was USD 1=VND 2,673.\n\n\n10\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its purchasing power conversion factor translates the currency.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003732:26:1", "occurrence_id": "46ff5ab22cd518f83177c23b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 14). Nevertheless, I check for this possibility by adding to the baseline specification\n\nthe interaction of the generic credit-chain distance and the (log) average real GDP per capita (from\n\nPWT) during 1980-2000. The results, reported in column (1) of Table 12, show that consistently\n\nwith the measurement hypothesis linkages are more important in developed countries, but this does\n\nnot significantly affect the findings regarding credit chains and sectoral comovement. Column (2)\n\nshows similar results when also controlling for a country’s overall degree of volatility (as measured\n\nby the standard deviation of the growth rate of real value added during 1980-2003).\n\nThe results obtained using the ratio of short term debt to payables (see Table 8) indicate that\n\nthe actual use of formal financing from financial intermediaries vis-a-vis suppliers credit partly\n\nalleviates the transmission of shocks through credit chains. Building on these results, column\n\n(3) checks whether the availability of formal financing affects the importance of the credit-chain\n\nmechanism by allowing the parameter _α_ to depend on a country’s degree of financial development,\n\nwhich amounts to add a triple interaction term. Surprisingly, the results indicate that the credit\n\nchain mechanism is stronger in more financially developed countries. <sup>19</sup> A possible interpretation of\n\nthis finding is that financial development also increases the supply of trade credit, at least in this\n\nsample, which would be consistent with Demirguc-Kunt and Maksimovic (2001) who maintain that\n\nthese sources of financing are complements. If this is the case, increasing the availability of formal\n\n\n19This finding is robust to controlling for the simple interaction between financial development and GDP per capita\nand generic credit chain distance.\n\n\n24\n\nMention: \"PWT\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Average real income per head over this window enters the interaction term.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003733:66:0", "occurrence_id": "612eba80221bdbae3fe16979", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Table A.2**\n**Evidence on Educational Attainment of Migrants**\n\n**from the China Urban Labor Survey (2001)**\n\n\nSource Community\n<u>All Provinces</u> <u>4 RCRE Provinces</u>\n\nEducation\nElementary or Less 0.247 0.220\n\nSome Middle School 0.086 0.096\n\nMiddle School 0.485 0.501\n\nSome High School 0.039 0.045\n\nHigh School 0.120 0.120\n\nSome Post Secondary 0.009 0.011\n\nCollege 0.010 0.012\n\nNumber of Observations 2,463 481\n\n<u>*Source: China Urban Labor Survey (see discussion on note of Table 1).</u>\n\n\n65\n\nMention: \"China Urban Labor Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Educational attainment of migrants by source community is tabulated from it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003735:13:1", "occurrence_id": "4e36248d30412b266a1a42b5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  on poverty might claim that their\n\n\nfindings can be extended to the national level. Free public access to RLMS and NOBUS\n\n\ndata also ensures that studies based on such surveys could be replicated. Neither HBS,\n\n\nnor regional or specialized surveys are accessible by Russian researchers. No paper in our\n\n\nsample presents a clear and detailed description of the sample design methodology,\n\n\nconstruction of the sample frame, or survey instruments – all of which form a standard set\n\n\nfor any empirical publication.\n\n\n5 NOBUS is a literal translation of the Russian abbreviation “НОБУС” that stands for “Национальное\nобследование благосостояния населения и его участия в социальных программах”. A detailed\ndescription of the survey and the actual data could be found at: http://nobus.worldbank.org.ru.\n\n\n12\n\nMention: \"RLMS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Free public access means results based on it can be replicated.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003735:13:1", "occurrence_id": "1e40d3c6a04176f37bcb9d5c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  on poverty might claim that their\n\n\nfindings can be extended to the national level. Free public access to RLMS and NOBUS\n\n\ndata also ensures that studies based on such surveys could be replicated. Neither HBS,\n\n\nnor regional or specialized surveys are accessible by Russian researchers. No paper in our\n\n\nsample presents a clear and detailed description of the sample design methodology,\n\n\nconstruction of the sample frame, or survey instruments – all of which form a standard set\n\n\nfor any empirical publication.\n\n\n5 NOBUS is a literal translation of the Russian abbreviation “НОБУС” that stands for “Национальное\nобследование благосостояния населения и его участия в социальных программах”. A detailed\ndescription of the survey and the actual data could be found at: http://nobus.worldbank.org.ru.\n\n\n12\n\nMention: \"NOBUS\n\n\ndata\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Its open access likewise supports replication of the findings.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003736:9:0", "occurrence_id": "58df4905327cd725fbaea246", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 2. Preference for Equality and Income Inequality\n\n\n\n9.0\n\n\n8.0\n\n\n7.0\n\n\n6.0\n\n\n5.0\n\n\n4.0\n\n\n3.0\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n20 25 30 35 40 45 50 55 60\n\n\nGini Coefficient\n\nSource: World Bank (2005) WDI database; WVS 1999-2001; and authors' calculations. Lower values of preference for equality indicate\ngreater preference for income equality.\n\n\nHowever, underlying levels of income and inequality are not the whole story. Table\n2 presents mean scores by quartiles and suggests that in addition to income and inequality\nthere are distinct geographic patterns in preferences. The lowest quartile representing the\ngreatest preference for equality is dominated by countries from Central and Eastern Europe.\nPreferences in this group are similar to more advanced economies such as Austria (4.6),\nFinland (4.6), and France (4.8). However, Lithuania, part of the former Soviet Union, also\nfalls in this group. The top-most quartile reflecting the lowest preferences for equality is\ndominated by countries from the former Soviet Union (FSU) such as Estonia, Moldova,\nUkraine and Russia. However, Bulgaria also falls in this group. The preferences of these\ncountries are roughly comparable to those of low-income and middle-income developing\ncountries such as Bangladesh (7.6), Peru (7.5), and Uganda (7.2). The exceptions\nnotwithstanding, it would appear that if countries display a strong preference for equality\nthey are likely to be from Eastern Europe, while if they display a low preference for equality\nthey are likely to be from the former Soviet Union.\n\n\nThere are several reasons why countries in Central and Eastern Europe may be more\ninclined towards greater equality than countries that descended from the Soviet Union. First,\nthis group of countries endured the economic disruption that ensued after the fall of the\nBerlin wall in 1989 and the end of communism relatively well. The decline in output was\nsmaller for countries of Central and\n\nMention: \"WDI database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Inequality coefficients plotted against equality preferences come from it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003736:9:0", "occurrence_id": "c08489e1a0d40482374e15e4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 2. Preference for Equality and Income Inequality\n\n\n\n9.0\n\n\n8.0\n\n\n7.0\n\n\n6.0\n\n\n5.0\n\n\n4.0\n\n\n3.0\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n20 25 30 35 40 45 50 55 60\n\n\nGini Coefficient\n\nSource: World Bank (2005) WDI database; WVS 1999-2001; and authors' calculations. Lower values of preference for equality indicate\ngreater preference for income equality.\n\n\nHowever, underlying levels of income and inequality are not the whole story. Table\n2 presents mean scores by quartiles and suggests that in addition to income and inequality\nthere are distinct geographic patterns in preferences. The lowest quartile representing the\ngreatest preference for equality is dominated by countries from Central and Eastern Europe.\nPreferences in this group are similar to more advanced economies such as Austria (4.6),\nFinland (4.6), and France (4.8). However, Lithuania, part of the former Soviet Union, also\nfalls in this group. The top-most quartile reflecting the lowest preferences for equality is\ndominated by countries from the former Soviet Union (FSU) such as Estonia, Moldova,\nUkraine and Russia. However, Bulgaria also falls in this group. The preferences of these\ncountries are roughly comparable to those of low-income and middle-income developing\ncountries such as Bangladesh (7.6), Peru (7.5), and Uganda (7.2). The exceptions\nnotwithstanding, it would appear that if countries display a strong preference for equality\nthey are likely to be from Eastern Europe, while if they display a low preference for equality\nthey are likely to be from the former Soviet Union.\n\n\nThere are several reasons why countries in Central and Eastern Europe may be more\ninclined towards greater equality than countries that descended from the Soviet Union. First,\nthis group of countries endured the economic disruption that ensued after the fall of the\nBerlin wall in 1989 and the end of communism relatively well. The decline in output was\nsmaller for countries of Central and\n\nMention: \"WVS 1999-2001\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its values measure preference for income equality.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003740:7:0", "occurrence_id": "a2a8c02a6a19b5dedcdf30ed", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: neighborhood relations in the spatial weights are only proxies for the true neighbors. The\n\n\nunderlying assumption is that the spatial variation among sampled “neighbors” is representative\n\n\nof that among the true neighbors, an assumption commonly taken in spatial hedonic models. We\n\n\nare comfortable that the sample design employed in these surveys supports the conclusions we\n\n\ndraw from this analysis. However, ideally it would be desirable to employ sampling designs that\n\n\nconsider neighborhood structure explicitly, for instance by sampling households randomly and\n\n\nthen including a number of direct neighbors for each households. Very few surveys have\n\n\nemployed such as design, the US American Housing Survey being an exception (Ioannides 2002).\n\n\n**IV.** **Data and Variables**\n\n\nDetailed household data for developing country cities are scarce. Most such data are collected at\n\n\nthe national level through Living Standards Measurement Surveys or Demographic and Health\n\n\nSurveys. These tend to distinguish urban and rural areas, but the number of observations in each\n\n\nurban area is too small for city-specific analysis. In this paper we use comprehensive and\n\n\ngeographically referenced information from urban household surveys for 2905 households in\n\n\nBangalore and 2508 households in Bhopal (for further details on this data see Deichmann et al.\n\n\n2003, Lall et al. 2004, Lall and Deichmann 2006b). These surveys were conducted in 2001 and\n\n\n2003 respectively. The available variables differ slightly across the two cities since the original\n\n\nsurveys were not totally identical. The key indicator in hedonic housing market analysis is the\n\n\nprice or rent of the dwelling unit. Due to high official transaction costs, most recorded sales in\n\n\nIndian real estate markets do not reflect actual transaction amounts. Rents are also often\n\n\nartificially low due to rent control and therefore do not always match actual market rents. The\n\n\nsurveys therefore asked households to report what they believe would be the market rent for a\n\n\nsimilar house in their neighborhood. Therefore this variable represents an estimated value\n\n\nreported by the surveyed individual. The survey\n\nMention: \"US American Housing Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"A rare design that samples neighbours alongside households.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003740:7:0", "occurrence_id": "cb2402fba65be680c6cca5bb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: neighborhood relations in the spatial weights are only proxies for the true neighbors. The\n\n\nunderlying assumption is that the spatial variation among sampled “neighbors” is representative\n\n\nof that among the true neighbors, an assumption commonly taken in spatial hedonic models. We\n\n\nare comfortable that the sample design employed in these surveys supports the conclusions we\n\n\ndraw from this analysis. However, ideally it would be desirable to employ sampling designs that\n\n\nconsider neighborhood structure explicitly, for instance by sampling households randomly and\n\n\nthen including a number of direct neighbors for each households. Very few surveys have\n\n\nemployed such as design, the US American Housing Survey being an exception (Ioannides 2002).\n\n\n**IV.** **Data and Variables**\n\n\nDetailed household data for developing country cities are scarce. Most such data are collected at\n\n\nthe national level through Living Standards Measurement Surveys or Demographic and Health\n\n\nSurveys. These tend to distinguish urban and rural areas, but the number of observations in each\n\n\nurban area is too small for city-specific analysis. In this paper we use comprehensive and\n\n\ngeographically referenced information from urban household surveys for 2905 households in\n\n\nBangalore and 2508 households in Bhopal (for further details on this data see Deichmann et al.\n\n\n2003, Lall et al. 2004, Lall and Deichmann 2006b). These surveys were conducted in 2001 and\n\n\n2003 respectively. The available variables differ slightly across the two cities since the original\n\n\nsurveys were not totally identical. The key indicator in hedonic housing market analysis is the\n\n\nprice or rent of the dwelling unit. Due to high official transaction costs, most recorded sales in\n\n\nIndian real estate markets do not reflect actual transaction amounts. Rents are also often\n\n\nartificially low due to rent control and therefore do not always match actual market rents. The\n\n\nsurveys therefore asked households to report what they believe would be the market rent for a\n\n\nsimilar house in their neighborhood. Therefore this variable represents an estimated value\n\n\nreported by the surveyed individual. The survey\n\nMention: \"Living Standards Measurement Surveys\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Their national scope leaves too few observations within each city.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003740:7:0", "occurrence_id": "1b8d9c4f08676ee72e5cd787", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: neighborhood relations in the spatial weights are only proxies for the true neighbors. The\n\n\nunderlying assumption is that the spatial variation among sampled “neighbors” is representative\n\n\nof that among the true neighbors, an assumption commonly taken in spatial hedonic models. We\n\n\nare comfortable that the sample design employed in these surveys supports the conclusions we\n\n\ndraw from this analysis. However, ideally it would be desirable to employ sampling designs that\n\n\nconsider neighborhood structure explicitly, for instance by sampling households randomly and\n\n\nthen including a number of direct neighbors for each households. Very few surveys have\n\n\nemployed such as design, the US American Housing Survey being an exception (Ioannides 2002).\n\n\n**IV.** **Data and Variables**\n\n\nDetailed household data for developing country cities are scarce. Most such data are collected at\n\n\nthe national level through Living Standards Measurement Surveys or Demographic and Health\n\n\nSurveys. These tend to distinguish urban and rural areas, but the number of observations in each\n\n\nurban area is too small for city-specific analysis. In this paper we use comprehensive and\n\n\ngeographically referenced information from urban household surveys for 2905 households in\n\n\nBangalore and 2508 households in Bhopal (for further details on this data see Deichmann et al.\n\n\n2003, Lall et al. 2004, Lall and Deichmann 2006b). These surveys were conducted in 2001 and\n\n\n2003 respectively. The available variables differ slightly across the two cities since the original\n\n\nsurveys were not totally identical. The key indicator in hedonic housing market analysis is the\n\n\nprice or rent of the dwelling unit. Due to high official transaction costs, most recorded sales in\n\n\nIndian real estate markets do not reflect actual transaction amounts. Rents are also often\n\n\nartificially low due to rent control and therefore do not always match actual market rents. The\n\n\nsurveys therefore asked households to report what they believe would be the market rent for a\n\n\nsimilar house in their neighborhood. Therefore this variable represents an estimated value\n\n\nreported by the surveyed individual. The survey\n\nMention: \"Demographic and Health\n\n\nSurveys\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Like the other programme, they split urban from rural but not across cities.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003741:5:0", "occurrence_id": "01700edc55343c5081a9faeb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: non-linear. <sup>3</sup> [^3: For an interesting theoretical analysis that highlights the tension between the increasing returns and risk\nin determining technological change and production structure in agriculture, see Dixit, 1993.]\n\n\nWe use data from the Nepal Living Standard Survey (NLSS) of 1995/96 to uncover the\n\n\nrole played by the extent of the market in agricultural specialization and commercialization\n\n\nat the village level. Crop agriculture in Nepal, like many other developing economies, is\n\n\ncharacterized by low degree of commercialization and specialization on average. There are,\n\n\nhowever, striking di ff erences among villages in terms of the level of agricultural development\n\n\ncovering the entire range from completely specialized production of non-staple crops to nearly\n\n\ncomplete subsistence agriculture. The stark geographical di ff erences in Nepal also resulted\n\n\nin large variation in sizes of the urban centers (from population of 10,000 in smaller towns\n\n\nto 421,000 in the capital city Kathmandu in 1991). These large variations in the level of\n\n\nagricultural development along with that in access to and size of urban markets enable us to\n\n\nempirically characterize the relationship between agricultural specialization and the extent of\n\n\nthe market. The results based on the Nepalese data are, however, of more general interest as\n\n\nthey are likely to be applicable to many other developing countries which are characterized by\n\n\nrelative isolation of rural areas due to poor infrastructure as well as low level of agricultural\n\n\ndevelopment. <sup>4</sup> [^4: These would include regions like Northern India and Pakistan, and many other developing countries in\nAfrica and Asia.]\n\n\nWe analyze two dimensions of structural change: the pattern of product diversi fi cation\n\n\n(crop specialization), and the degree of market production as opposed to home production\n\n\n(i.e., commercialization of agriculture). The pattern of crop specialization is measured by\n\n\nHer fi ndahl index of concentration of cropland use, and by the share of land devoted to non\n\ncereal crops. Sales of non-rice crops and all crops as a percentage\n\nMention: \"Nepal Living Standard Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its data uncover how market extent shapes agricultural specialisation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003744:35:0", "occurrence_id": "a6e09640fc38d8f7abbf36d0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: measures of investment climate into a real model. In addition, the current model structure does not\ninclude features that allow FDI to have certain “special” effects on the domestic economy, such as\nincreasing returns to scale and endogenous mark-ups in a Dixit-Stiglitz framework, product differentiation by firm ownership and location of production, or the use of business services (provided by\nboth domestic firms and foreign multinationals) as essential inputs into the production process (see\nLejour and Rojas-Romagosa, 2006, for a review of relevant literature). Instead, this section considers some basic links between preferential trade liberalization and FDI inflows through reduced-form\nparameter estimates obtained from econometric studies. Although clearly simplistic, the approach\nis very transparent while still allowing us to evaluate the different channels of transmission of FDI\nshocks.\n\nIn order to implement the FDI scenarios in this section, we make several modifications to the\ndataset and the dynamic updating equations. To accurately model the increase in foreign capital\nstock following the potential acceleration in FDI, we need to distinguish capital by ownership type\n(domestic or foreign) and, for the agricultural sectors, to distinguish capital from land. The latter\nis important because, as shown in section 4.4, capital intensities in agriculture are much higher\nthan capital intensities in manufacturing and services, which is a result of attributing all land and\nnatural resource income to aggregate capital, as well as potential misclassification of self-employed\nincome. This issue was not particularly relevant for the trade simulations but is very important\nfor the scenarios in this section, since the base year distribution of capital stocks by sector largely\ndetermines the allocation of new investment. Unfortunately, data on capital and land value added is\nnot readily available for Honduras, and therefore we rely on the Central America SAM from GTAP\nto separate the two. For each farm sector in this SAM, capital accounts for 47 percent of the total\ncapital-land value added bundle, and we use this information to split payments\n\nMention: \"Central America SAM\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It separates capital from land value added where national data are missing.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003750:5:1", "occurrence_id": "0a418341acbdfb9aed4ee627", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \nincarcerated drug offenders at approximately $40 billion (Executive Office of the President\n2004).\n\nWhile the costs of prohibition to consuming countries appear to be quite high, one\nunintended effect of prohibition is to impose even higher costs (at least relatively) on some\ndeveloping countries involved in the international “War on Drugs”.  The Uribe government in\nColombia committed the country to increasing defense expenditures from 3.6 percent of GDP\nin 2003 to 6 percent by 2006, increasing security forces from 250,000 (150,000 military plus\n100,000 police) to 850,000 over four years (Colombian Government, 2003). These expenditures\nare mostly intended to combat insurgency in the country, but the main source of insurgent\nfinancing is the revenues from participation in narcotics production and trafficking, made\nparticularly lucrative as a result of their illegality. In contrast, public expenditures on health in\nColombia were around 5 percent of the GDP in 2000 (World Development Indicators). The\nopportunity cost of these resources for developing countries, be it in terms of investments in\nhealth, education, or infrastructure, is almost surely larger than in richer countries and represents\na substantial cost of the prohibition strategy that is generally neglected.\n\nThese are the direct, fiscal costs of prohibition. They are particularly large in countries,\nsuch as Colombia, that have chosen to strongly confront drug production and trafficking. But\neven in developing countries that invest less in outright interdiction and prosecution, three other\ncategories of unintended costs of prohibition can be identified. These are not as clearly taken\ninto account in weighing the costs and benefits of prohibition. Prohibition damages public\nhealth, as it may worsen the social costs of given drug consumption; prohibition exposes farmers\nto the risk of catastrophic loss, while creating large rents for organized crime involved in drug\ntrafficking; and prohibition undermines public safety, security, and institutions in developing\ncountries.\n\n\n3\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It reports public health spending near five percent of output.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003753:28:2", "occurrence_id": "0ee22dc532381b50b587788e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 8.0<br>16.0<br>28.0<br>37.3<br>21.3<br>52.0<br>36.0<br>37.3<br>34.7<br>36.0<br>25.3<br>36.0<br>41.3<br>41.3<br>40.0<br>44.0<br>21.3<br>33.3<br>41.3<br>29.3<br>..<br>4.0<br>25.3<br>42.9<br>34.9<br>29.7<br>31.2<br>35.5<br>25.9<br>37.1<br>35.2<br>28.1<br>40.4<br>37.9<br>29.2<br> <br> <br> <br> <br> <br> <br>76.0<br>52.0<br>53.3<br>37.3<br>38.7<br>33.3<br>45.3<br>36.0<br>50.7<br>66.7<br>41.3<br>38.7|\n\n\n\nNotes: Percentages based on the 75 products of parts and components from textiles & clothing and\nmachinery & equipment at SITC 4/5-digit level in Revision 2.\nDue to the missing data, the 1995 data for Vietnam is replaced by 1997.\n\nSource: Computations based on UN COMTRADE Statistics.\n\n\nFrom Table 7, it is clear that East Asian revealed comparative advantage in processed\n\n\nexports of components and parts has been gaining over the years\n\nMention: \"UN COMTRADE Statistics\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Comparative advantage in component exports is computed from these statistics.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003756:16:0", "occurrence_id": "32e4207030291665b0392ab6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <u>Enforcement. Spectrum users must comply with the terms of authorization and the applicable</u>\nrules and regulations. Spectrum managers are particularly concerned with interference problems affecting public safety and security services such as emergency services (ambulance,\nfire, police) and navigational services (air, maritime), as well as commercial bands where\nspectrum holders have often invested and incurred large sunk costs and expect protection for\nthese investments. The spectrum authority needs a framework and process for responding and\ndealing with complaints from the public and users and for settling disputes. The powers,\nduties, and obligations of the spectrum manager and protection of rights for the public under\ncircumstances where inspection of property is necessary must be established by law. Penalties, remedies, and alternative dispute mechanisms help ensure rapid resolution.\n\n\n**2.6** **Implementation**\n\n\nImplementing the building blocks of spectrum administration requires support in terms of data,\nanalysis, administration, monitoring, and international affairs. Typically about one-third of all\npersonnel employed in regulating telecommunications and broadcasting is occupied in government administration of the spectrum. <sup>37</sup>\n\n\nDetailed knowledge of current use across the entire spectrum is required for effective planning. In addition to the basic data on frequency, users, and location, a national register contains\ninformation on the functions performed by the equipment, detailed technical characteristics,\nand costs of system implementation. A single national frequency register database should be\ncreated if one does not already exist.\n\n\nSpectrum allocation and assignment under traditional administration regimes is based on\ndetailed spectrum engineering analyses undertaken by the spectrum authority. This includes\nusing computer models of radio propagation patterns to ensure compatibility among uses\nand equipment, examine the likelihood of interference among users, and assess potential\nhealth and safety hazards. Comprehensive national records of individual spectrum assignments and use are maintained as needed for these engineering analyses as well as for administrative purposes.\n\n\nSpectrum use is monitored to obtain technical and operational information on current use\n\nMention: \"national frequency register database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"A recommendation that such a register be created where none exists.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003758:12:1", "occurrence_id": "d8cc054a97ce9bcd5306ebf6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 624<br>4,953<br>4,840<br>899<br>366<br>502<br>674<br>835<br>257<br>1,652<br>5,185<br>5,480<br>881<br>756<br>894<br>750<br>761<br>298<br>1,721<br>6,060<br>5,950<br>959<br>816<br>722<br>675<br>650<br>246<br>1,729<br>5,795<br>6,180<br>815<br>698<br>892<br>746<br>639<br>327<br>2,002<br>6,118<br>6,330<br>749<br>663<br>860<br>805<br>520<br>386<br>2,163<br>6,146<br>5,660<br>924<br>848<br>931<br>866<br>579<br>533<br>2,455<br>7,136<br>6,140<br>743<br>922<br>863<br>864<br>660<br>599<br>2,764<br>7,413<br>6,496|\n\n\nSource: COMTRADE database.\nNote: Excludes intra-EU trade. Arabian Peninsula include: Bahrain, Kuwait, Oman, Qatar, Saudi Arabia. United Arab Emirates and\nYemen.\n\n\n\nThe impact of avian flu on international trade flows is more evident by examining\n\n\nthe evolution of the exports of major exporters. Table 3 shows the rapid shifting in\n\n\noverall exports of unprepared poultry from HPAI affected countries to\n\nMention: \"COMTRADE database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Regional trade values excluding intra-community flows are drawn from it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003759:17:0", "occurrence_id": "dc1b0342f567dbc0d4528b77", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **5.  Productivity Performance and the Effects of Economic Reform**\n\n\n_Estimates of total factor productivity_\n\n\nThe estimate of capital's share obtained above, α = 0.43 was used to find a direct\n\n\nmeasure of change in total factor productivity for the periods 1961-2003, 1980-1989, and\n\n\n1995-2003. Table 7 reports the decomposition of output growth into the separate\n\n\ncontributions of capital, labor, and total factor productivity (TFP).\n\n\nTable 7\nDecomposition of Real GDP Growth in Rwanda, α = 0.43\n\n\n<u>Contribution of:</u>\n\n\nReal GDP Physical Labor TFP\n\n<u>Growth</u> <u>Capital</u>\n196103 3.6 1.66 1.57 0.37\n\n\n198089 3.22 3.73 1.86 -2.37\n\n\n1995<u>03</u> <u>11.5</u> <u>0.45</u> <u>3.27</u> <u>7.78</u>\n\n\nSource: World Bank database and computed by authors.\n\n\nThe findings show some rather stark contrasts evident between periods. In\n\n\nparticular, the residual calculation of TFP points to the possibility of a major shift in\n\n\neconomic structure beginning in 1995. The year 1995 was selected because of the\n\n\ncataclysmic events of 1994, which reflected the culmination of several years of economic\n\n\nand political instability in Rwanda. The contribution of TFP may appear to have been\n\n\nslightly positive for the entire period, but that was likely due to the extraordinary leap in\n\n\nTFP that took place since 1995. In the post-1995 period TFP accounted for about two\n\n16\n\nMention: \"World Bank database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Period growth comparisons are taken from it and extended by the authors.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003762:13:2", "occurrence_id": "9e47455daf2d2eeef4c6d647", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  64.0 [58.1] 28.9 [43.4] 32.4 [46.3] 8.1 [29.1] 34.1 [44.9]\n**RRS** 31.6 [67.0] 22.1 [41.8] 18.0 [52.9] 4.2 [27.8] 20.8 [47.8]\n**GBAO** 3.8 [58.7] 1.7 [43.8] 0.0 [43.4] 0.0 [33.0] 1.5 [44.4]\n\n**Men** 64.4 [60.9] 28.0 [43.8] 24.7 [50.2] 122.5 [53.5] 48.2 [54.8]\n**Women** 27.0 [61.7] 21.6 [42.4] 25.4 [46.7] 6.0 [29.2] 19.4 [42.7]\n**<u>Tajikistan</u>** <u>51.8 [61.20</u> <u>22.7 [42.7]</u> <u>25.1 [48.4]</u> <u>8.6 [29.7]</u> <u>28.8 [46.6]</u>\n**Source** : Authors’ calculations from Tajikistan Health PETS 2006 data\n**Note** : Gorno-Badakhshan Autonomous Oblast (GBAO) and Rayon under Republican Subordination (RRS) are regions\n\n\nOne would naturally expect the frequency of contact with the patients and the level of\n\n\nresponsibilities in the health facility to be positively correlated with both the prevalence\n\nMention: \"Tajikistan Health PETS 2006 data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Regional expenditure flows in the health sector are traced from these data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003764:7:0", "occurrence_id": "d426284cb9f4b2d01ca08cc7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: regulatory measures. We control for these factors using private credit to GDP ratio (from\n\n\nDjankov et al 2007) and number of procedures required to start a business, number of\n\n\nprocedures required to enforce a contract and an index of the difficulty of hiring labor\n\n\n(from World Bank’s Doing Business project, 2004).\n\n\nRegression results with these controls are reported in columns 1 and 2 of Table 2.\n\n\nEnglish common law continues to show a positive and significant effect on the dependent\n\n\nvariable. Private credit to GDP and all the Doing Business indicators show a significant\n\n\nrelationship with the dependent variable in the expected direction.\n\n\nNext, we control for the index of creditor rights from Djankov et al 2007 and\n\n\nmeasures of institutional quality as perceived by the firms (from WBES) which include:\n\n\nthe level of confidence in the legal system for upholding contract and property rights in a\n\n\ndispute ( _Property rights_ ), efficiency of courts in enforcing their decisions in business\n\n\ndisputes ( _Enforcement_ ), quality of power supply ( _Power_ ), and the severity of business\n\n\nregulations, high interest rates and labor regulations as obstacles for the firm’s operations.\n\n\nEach of the obstacles is ranked by firms on a 1-4 scale with a higher value indicating a\n\n\nmore severe obstacle. _Property rights_, _Enforcement_ and _Power_ take values 1-6 with a\n\n\nhigher value implying better quality. The motivation for the firm-level measures is to\n\n\ncapture possible heterogeneity in their enforcement and how they affect firms.\n\n\nRegression results in columns 3 and 4 of Table 2 show that controlling for the\n\n\nvariables discussed above does not change our main results much. Most of the firm-level\n\n\nmeasures show significant effects on government helpfulness in the expected direction:\n\n\nbetter quality of institutions and less severe obstacle implies a more helpful government.\n\n\n5\n\nMention: \"World Bank’s Doing Business project\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its procedure counts and hiring difficulty index enter as controls.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003765:23:1", "occurrence_id": "6babd5d62cfbdbcee9dd939d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  deviation worsening in infrastructure increases\ninventories by 27 percent to 47 percent relative to U.S. levels. The proxy for the lack of a free\nmarket is transfers and subsidies to private and public enterprises expressed as a fraction of GDP\nA one-standard deviation restriction on the free market increases raw materials inventories by 19\nto 30 percent.\n\n\n9 However, it is worth pointing out that the initials results of the research, using the aggregate inventory levels\ncomputed from the National Accounts data, were not inconsistent with the stylized observation that developing\ncountries hold more inventory than developed countries.\n\n\n21\n\nMention: \"the National Accounts data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Aggregate inventory levels computed from them are consistent with the observed pattern.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003767:28:1", "occurrence_id": "f7ab19cbcd9e00d8f748257b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  provide an appropriate be nchmark because the Chinese banking sector\ndose not operate on a commercial basis and is su bject to significant distortions, primarily\narising out of the continued dominant role of th e government in the sector. (1) The\nGOC’s ownership: the GOC continues to c ontrol the vast m ajority of financial\nintermediation in the banking sec tor; (2) SOCBs: they continue to be plagued by\nfunctional and operational problem s, and fund s continue to be a llocated in a m anner\nconsistent with the government policy; (3) Foreign banks: foreign investment in banking\nsector is tightly constrained, and foreign banks are subject to the same restrictions as the\nSOCBs. Therefore, foreign bank lending does not provide a suitable benchmark.\n\n**2. External Benchmark.** The DOC uses a cross-country average lending rate as the\nexternal benchmark in this proceeding. Th e countries are select ed based on sim ilar\nincome levels to China: (1) China is a lo wer-middle income country according to the\ncountry classification of the World Bank, 37 countries have been selected; (2) The DOC\ncomputed a sim ple average of these 37 lend ing rates from _International Financial_\n_Statistics_ in 2005, as the external benchm ark for short-term loans; (3) The DOC\ndeveloped a ratio of short-te rm and long-term lending for 2005, by using the short-ter m\nlending drawn from London Interbank Offered Rate (LIBOR) data and long-term interest\nrate from the interest rate swap market reported by the Federal Rese rve, to transfer the\nshort-term lending rate into long-term one.\n\n\n27\n\nMention: \"London Interbank Offered Rate (LIBOR) data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Short-term lending rates are drawn from these quotes.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003773:23:0", "occurrence_id": "0ddfc5ab8be4d625faea8e06", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: This paper uses data from the 2004 Quantitative Baseline Survey of the Tea sector\n\n\nto assess the potential impact of privatization of tea estates. The analysis is framed\n\n\nwithin the logic of causal inference. This entails a normalized comparison of outcomes\n\n\nto account for household heterogeneity in terms of observable and non-observable\n\n\ndeterminants of the outcomes of interest. These outcomes relate to productivity. Three\n\n\nmain findings emerge from this comparison. Productivity outcomes such as yield, time\n\n\ntaken to carry leaves to the collection point, and fertilizer use are generally better in the\n\n\nprivate sector than in the public sector. Also, male-headed households out perform\n\n\nfemale-headed households along all dimensions consider here. Finally, in a welfare\n\n\ncomparison between the tea and non-tea sectors, the former tend to be better off than the\n\n\nlatter.\n\n\n**References**\n\n\nBrambilla, Irene and Porto, Guido G. 2005. Farm Productivity and Market Structure:\n\n\nEvidence from Cotton Reforms in Zambia. Discussion Paper No.919. New\n\n\nHaven: Economic Growth Center (Yale University).\n\n\nDabalen, Andrew, Paternostro, Stefano and Gaëlle, Pierre. 2004. The Returns to\n\n\nParticipation in the Nonfarm Sector in Rural Rwanda. World Bank Policy\n\n\nResearch Working Paper No. 3462.\n\n\nDiop, Ndiame, Brenton, Paul and Asarkaya, Yakup. 2005. Trade Costs, Export\n\n\nDevelopment and Poverty in Rwanda. World Bank Policy Research Paper No.\n\n\n3784.\n\n\nEssama-Nssah, B. 2006. Propensity Score Matching and Policy Impact Analysis: A\n\n\nDemonstration in EViews. World Bank Policy Research Working Paper No.\n\n\n3877.\n\n\nHolland, Paul W. 1986. Statistics and Causal Inference. Journal of the American\n\n\nStatistical Association, Vol. 81, No. 396: 945-960.\n\n\n21\n\nMention: \"the 2004 Quantitative Baseline Survey of the Tea\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"It assesses the potential impact of privatising tea estates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003778:2:0", "occurrence_id": "3b7d69d1bea55542b589ee98", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ### **Making Poor Haitians Count**\n\n**Poverty in Rural and Urban Haiti**\n**Based on the First Household Survey for Haiti**\n\n##### By **Dorte Verner** World Bank 1\n\n\n1 I am very grateful to Nadim Khouri for inviting me to join the Haiti Rural Development Team, Michael\nJustesen for excellent and invaluable research assistance, and the Haiti Rural Development Team for\nanswering many questions. Moreover I am especially grateful to Willy Egset for data information and other\nsupport with the household data set. Without the help of all these people this paper would not have existed.\nThe findings, interpretations, and conclusions expressed in this paper are entirely those of the author.\n\nMention: \"Household Survey for Haiti\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The first such survey measures rural and urban poverty in the country.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003781:20:0", "occurrence_id": "33802826d2c690b76a3c8629", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Moreover, the intensive use of land has caused a large soil erosion problem for\nfarmers. Sixty-six percent of extreme poor farmers and 46 percent of the 20 percent\nrichest farmers reported that soil erosion is a severe problem on their land in 2001. Poor\nfarmers have little access to productivity enhancing plant nutrients and therefore see soil\nerosion as a more severe problem than rich farmers. Moreover, lack of titles or land\nownership is another problem faced by poor farmers. Of the 20 percent poorest farmers\n81 percent have a title or own their land compared to 92 percent of the 20 percent richest\nfarmers (Table 7). Data reveal little difference in travel time to plots between rich and\npoor farmers (Table 8).\n\n\n<u>Table 6: Reporting on Soil Erosion in Haiti, 2001 (percent)</u>\n\nQuintile\n<u>1 (poorest)</u> <u>2</u> <u>3</u> <u>4</u> <u>5 (richest)</u>\nYes, important 65.5 65.82 63.92 62.9 45.63\nYes, minor 7.82 7.47 8.73 6.47 7.32\n<u>No problem</u> <u>26.67</u> <u>26.71</u> <u>27.34</u> <u>31.63</u> <u>47.05</u>\n\n_Source:_ Own calculations based on HLCS 2001.\n\n\n<u>Table 7: Legal Title or Deed to Home and Land, 2001 (percent)</u>\nQuintile\n<u>1 (poorest)</u> <u>2</u> <u>3</u> <u>4</u> <u>5 (richest)\n\nMention: \"HLCS 2001\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Perceptions of problems and land title ownership are tabulated by quintile.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003782:12:0", "occurrence_id": "0b186d4e46cfaec8fb5fcc9c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: their θ s are higher than those of firms with higher variable trade costs, even though both types of\n\n\nfirms serve the same market. Thus, once we allow private characteristics of individual firms to\n\n\ngenerate variation in variable and fixed trade costs, lowering trade costs also increases the export\n\n\nintensity of individual firms.\n\n\n**3. Data**\n\n\nThe empirical part of this study uses the firm-level World Bank Investment Climate Survey (ICS)\n\n\ndata from the manufacturing sectors of seven Sub-Saharan African countries: Benin, Ethiopia, Kenya,\n\n\nMadagascar, Senegal, Tanzania, and Uganda. <sup>9</sup> [^9: The World Bank ICS has been renamed as the World Bank Enterprise Survey. Visit <u>[www.enterprisesurvey.org](http://www.enterprisesurvey.org/)</u> for more\ninformation.] These seven countries were chosen on the basis of\n\n\navailability of detailed information on firm-level export destinations in ICS. The ICS data are\n\n\ncollected to prepare Investment Climate Assessments (ICAs), diagnostic reports intended to serve as\n\n\nthe basis for policy reforms that will help improve these countries’ business environments with the\n\n\nsupport of the World Bank. <sup>10</sup> [^10: ICA reports for Sub-Saharan African countries are available from the World Bank Africa Regional Program on\n[Enterprise Development (RPED) website http://www.worldbank.org/rped).](http://www.worldbank.org/rped)] The data are collected through firm surveys that include a common set\n\n\nof questions for all countries surveyed, supplemented by country-specific questions to help each\n\n\ncountry assess its investment climate. The sample is selected by a stratified random sampling method\n\n\ncontrolling for size, sub-sector, and geographic distribution, based on the company registration\n\n\nrecords or manufacturing census information available from the government. More than 80 countries\n\n\nhave been surveyed since the program started in 2002. The sample size varies, ranging from about\n\n\n100 firms for some\n\nMention: \"World Bank Investment Climate Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Manufacturing firms in seven countries supply the export intensity measures.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003783:31:1", "occurrence_id": "01506f337fe0c2bb548781dc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  matrix of India from\nGlobal Trade Analysis Project (GTAP), Version 6.\nSkilled labor (column 1) denotes remunerations of skilled labor in the sector\nas a percentage of total labor remunerations. Intensity ratios for power\n(column 2) and roads (column 3) equal total expenditure incurred by firms in\nthe industry on the two respective inputs and expressed as a percentage of\n<u>total value added in the sector..</u>\n\n\n29\n\nMention: \"Global Trade Analysis Project (GTAP)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its version six matrix for India gives factor remunerations by sector.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003788:14:0", "occurrence_id": "5c50289aa6c7a06498fd964b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: - 13 \n\nThe extent to which the service quality of infrastructure has improved varies from country to\n\n\ncountry. Figures 13 to 15 show the difference between the 2002 and 2005 BEEPS results for\n\n\neach country. Obviously, more advanced countries tend to have experienced relatively\n\n\n\n20\n\n\n10\n\n\n0\n\n\n-10\n\n\n-20\n\n\n-30\n\n\n-40\n\n\n-50\n\n\n\n\n\n\n\n20\n\n\n10\n\n\n0\n\n\n-10\n\n\n-20\n\n\n-30\n\n\n-40\n\n\n-50\n\n\n\n\n|because they already achieved t<br>her improvement. Armenia, Geo<br>the best performing countries in<br>ange in electricity. Interestingly<br>consistent with each other with<br>development strategy and priori<br>3. Quality Improvements in Electricity in E<br>(16|Col2|Col3|Col4|Col5|Col6|Col7|Col8|h<br>r<br>,<br>i<br>t<br>0)|\n|---|---|---|---|---|---|---|---|---|\n||||||||||\n||||||||||\n||||||||||\n||||||||||\n||||||||||\n||||||||||\n||||||||||\n|Azerbaijan<br>Uzbekistan<br>Kyrgyz<br>Moldova<br>Georgia<br>Ukraine<br>Kazakhstan<br>Romania<br>Macedonia<br>Armenia<br>Slovakia<br>Russia<br>Estonia<br>Czech<br>Poland<br>Croatia<br>Slovenia<br>Hungary<br>Lithuania<\n\nMention: \"BEEPS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Differences between these two rounds show how perceptions changed by country.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003792:45:0", "occurrence_id": "88320991b1d4257443e257e7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Table 1**\n**Local Networks of Rural-Urban Migrants at Time of Migration**\n\n**Five-City CULS Migrant Survey***\n\n\nSource Community\n\nLocation\n\n\n\nAll\n<u>Provinces</u>\n\n\n\nRCRE\n<u>Provinces</u>\n\n\n\nShare of Migrants with:\n\n\nJob Arranged Before _First_ Migration Experience 0.52 0.57\n\n\nJob Arranged Before _Current_ Migration Experience 0.53 0.56\n\n\nSome Acquaintance from Home Village in City Before Migrating 0.91 0.94\n\n\n**Close Family Member in City Before Migration 0.35 0.35\n\n\n**Extended Family Member in City Before Migration 0.52 0.58\n\n\n**Hometown Acquaintances 0.65 0.67\n\n\nFive or Fewer Hometown Acquaintances 0.39 0.44\n\n\nMore than Five Hometown Acquaintances 0.27 0.24\n\n\nAt Least One Local Acquaintance 0.09 0.08\n\n\nNumber of Migrants 2,463 481\n\n\n*Respondents are holders of rural registration ( _hukou_ ). The survey was conducted in Fuzhou, Shanghai,\nShenyang, Wuhan and Xian during late 2001. Sample frames were assembled using information on\ndistribution of migrants within cities from the 2000 Population Census. After selecting neighborhoods through\na proportional population sampling procedure, sample frames were assembled using residents’ committee\nrecords of migrant households and registers of migrants living on construction sites and held by local by police\nstations. Very short-term migrants, who lack a residence that falls under the jurisdiction of either of these\nauthorities, are unlikely to have made it into the sample frame.\n**A _close family member_ is adult sibling or member of nuclear family (e.g., spouse, child, parent). An\n_extended family member_ refers to cousins or other relatives. _Hometown acquaintances_ are unrelated, but\nknown by the respondent. Note that migrants may have acquaintances in several\n\nMention: \"Five-City CULS Migrant Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Local networks at the time of migration are recorded in this migrant survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003792:45:0", "occurrence_id": "f3d0dfbf7eea85bc0c6feb9f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Table 1**\n**Local Networks of Rural-Urban Migrants at Time of Migration**\n\n**Five-City CULS Migrant Survey***\n\n\nSource Community\n\nLocation\n\n\n\nAll\n<u>Provinces</u>\n\n\n\nRCRE\n<u>Provinces</u>\n\n\n\nShare of Migrants with:\n\n\nJob Arranged Before _First_ Migration Experience 0.52 0.57\n\n\nJob Arranged Before _Current_ Migration Experience 0.53 0.56\n\n\nSome Acquaintance from Home Village in City Before Migrating 0.91 0.94\n\n\n**Close Family Member in City Before Migration 0.35 0.35\n\n\n**Extended Family Member in City Before Migration 0.52 0.58\n\n\n**Hometown Acquaintances 0.65 0.67\n\n\nFive or Fewer Hometown Acquaintances 0.39 0.44\n\n\nMore than Five Hometown Acquaintances 0.27 0.24\n\n\nAt Least One Local Acquaintance 0.09 0.08\n\n\nNumber of Migrants 2,463 481\n\n\n*Respondents are holders of rural registration ( _hukou_ ). The survey was conducted in Fuzhou, Shanghai,\nShenyang, Wuhan and Xian during late 2001. Sample frames were assembled using information on\ndistribution of migrants within cities from the 2000 Population Census. After selecting neighborhoods through\na proportional population sampling procedure, sample frames were assembled using residents’ committee\nrecords of migrant households and registers of migrants living on construction sites and held by local by police\nstations. Very short-term migrants, who lack a residence that falls under the jurisdiction of either of these\nauthorities, are unlikely to have made it into the sample frame.\n**A _close family member_ is adult sibling or member of nuclear family (e.g., spouse, child, parent). An\n_extended family member_ refers to cousins or other relatives. _Hometown acquaintances_ are unrelated, but\nknown by the respondent. Note that migrants may have acquaintances in several\n\nMention: \"2000 Population Census\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Migrant distribution across cities here defines the sample frames.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003792:45:0", "occurrence_id": "a7825e7fcd8a36b3da47e6b7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Table 1**\n**Local Networks of Rural-Urban Migrants at Time of Migration**\n\n**Five-City CULS Migrant Survey***\n\n\nSource Community\n\nLocation\n\n\n\nAll\n<u>Provinces</u>\n\n\n\nRCRE\n<u>Provinces</u>\n\n\n\nShare of Migrants with:\n\n\nJob Arranged Before _First_ Migration Experience 0.52 0.57\n\n\nJob Arranged Before _Current_ Migration Experience 0.53 0.56\n\n\nSome Acquaintance from Home Village in City Before Migrating 0.91 0.94\n\n\n**Close Family Member in City Before Migration 0.35 0.35\n\n\n**Extended Family Member in City Before Migration 0.52 0.58\n\n\n**Hometown Acquaintances 0.65 0.67\n\n\nFive or Fewer Hometown Acquaintances 0.39 0.44\n\n\nMore than Five Hometown Acquaintances 0.27 0.24\n\n\nAt Least One Local Acquaintance 0.09 0.08\n\n\nNumber of Migrants 2,463 481\n\n\n*Respondents are holders of rural registration ( _hukou_ ). The survey was conducted in Fuzhou, Shanghai,\nShenyang, Wuhan and Xian during late 2001. Sample frames were assembled using information on\ndistribution of migrants within cities from the 2000 Population Census. After selecting neighborhoods through\na proportional population sampling procedure, sample frames were assembled using residents’ committee\nrecords of migrant households and registers of migrants living on construction sites and held by local by police\nstations. Very short-term migrants, who lack a residence that falls under the jurisdiction of either of these\nauthorities, are unlikely to have made it into the sample frame.\n**A _close family member_ is adult sibling or member of nuclear family (e.g., spouse, child, parent). An\n_extended family member_ refers to cousins or other relatives. _Hometown acquaintances_ are unrelated, but\nknown by the respondent. Note that migrants may have acquaintances in several\n\nMention: \"residents’ committee\nrecords of migrant households\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"These local records complete the list of migrant households.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003792:45:0", "occurrence_id": "0bda2807eec6e6a6214ab08d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Table 1**\n**Local Networks of Rural-Urban Migrants at Time of Migration**\n\n**Five-City CULS Migrant Survey***\n\n\nSource Community\n\nLocation\n\n\n\nAll\n<u>Provinces</u>\n\n\n\nRCRE\n<u>Provinces</u>\n\n\n\nShare of Migrants with:\n\n\nJob Arranged Before _First_ Migration Experience 0.52 0.57\n\n\nJob Arranged Before _Current_ Migration Experience 0.53 0.56\n\n\nSome Acquaintance from Home Village in City Before Migrating 0.91 0.94\n\n\n**Close Family Member in City Before Migration 0.35 0.35\n\n\n**Extended Family Member in City Before Migration 0.52 0.58\n\n\n**Hometown Acquaintances 0.65 0.67\n\n\nFive or Fewer Hometown Acquaintances 0.39 0.44\n\n\nMore than Five Hometown Acquaintances 0.27 0.24\n\n\nAt Least One Local Acquaintance 0.09 0.08\n\n\nNumber of Migrants 2,463 481\n\n\n*Respondents are holders of rural registration ( _hukou_ ). The survey was conducted in Fuzhou, Shanghai,\nShenyang, Wuhan and Xian during late 2001. Sample frames were assembled using information on\ndistribution of migrants within cities from the 2000 Population Census. After selecting neighborhoods through\na proportional population sampling procedure, sample frames were assembled using residents’ committee\nrecords of migrant households and registers of migrants living on construction sites and held by local by police\nstations. Very short-term migrants, who lack a residence that falls under the jurisdiction of either of these\nauthorities, are unlikely to have made it into the sample frame.\n**A _close family member_ is adult sibling or member of nuclear family (e.g., spouse, child, parent). An\n_extended family member_ refers to cousins or other relatives. _Hometown acquaintances_ are unrelated, but\nknown by the respondent. Note that migrants may have acquaintances in several\n\nMention: \"registers of migrants living on construction sites\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Police-held registers cover short-term migrants without fixed residence.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003793:36:0", "occurrence_id": "1cad26f298b364f9d91bb729", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Endnotes**\n\n1 It is estimated that there were about 20,000 new establishments formed per annum after\nthe introduction of the Enterprise Law in 2000 (see World Bank, 2005).\n\n2 These are based on the authors’ calculations using the Vietnam Living Standards Survey\n(VLSS) 1992/93 and the Vietnam Household Living Standards Survey (VHLSS) 2004.\n\n3 This does not include short-term, unregistered movement or movement in the six months\npreceding the census date.\n\n4 See endnote 1.\n\n5 It should be noted that in this paper we define a remitter as a migrant who sent any\nmoney/goods home to their relatives and/or gave any money/goods to the relatives during\ntheir visits. The data do not allow us to make a distinction between the two types of activity.\nHence the value of remittances is the total value of the money/goods that the migrant\nsent/gave to his or her relatives during the 12 months prior to the interview.\n\n6 We investigated the magnitude of the correlations between the explanatory variables as a\nprelude to our regression analysis. In general, the correlations were modest in nature with\nthe average correlation coefficient about 0.09 in magnitude. The overwhelming majority of\nestimated correlation coefficients were found to be less than 0.3. Thus, we do not believe\nthat multicollinearity represents a serious issue for the econometric analysis undertaken\nhere. In addition, the role of this particular phenomenon is likely to be attenuated by the\nlarge sample size used in estimation.\n\n7 Unfortunately, the data do not allow us to distinguish whether the migrant works for the\ngovernment or in a manufacturing state-owned enterprise.\n\n8 Among those migrants who faced some difficulties, the main ones are reported in rank\norder to be housing problems, having no income source, and, related to this, not being able\nto find a job.\n\n9 The estimation of the migrant remittance function using cross-sectional data,\n\nMention: \"Vietnam Living Standards Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The earlier round behind the calculated comparisons.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003793:36:0", "occurrence_id": "47a41d3b2a3c5875b81109de", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Endnotes**\n\n1 It is estimated that there were about 20,000 new establishments formed per annum after\nthe introduction of the Enterprise Law in 2000 (see World Bank, 2005).\n\n2 These are based on the authors’ calculations using the Vietnam Living Standards Survey\n(VLSS) 1992/93 and the Vietnam Household Living Standards Survey (VHLSS) 2004.\n\n3 This does not include short-term, unregistered movement or movement in the six months\npreceding the census date.\n\n4 See endnote 1.\n\n5 It should be noted that in this paper we define a remitter as a migrant who sent any\nmoney/goods home to their relatives and/or gave any money/goods to the relatives during\ntheir visits. The data do not allow us to make a distinction between the two types of activity.\nHence the value of remittances is the total value of the money/goods that the migrant\nsent/gave to his or her relatives during the 12 months prior to the interview.\n\n6 We investigated the magnitude of the correlations between the explanatory variables as a\nprelude to our regression analysis. In general, the correlations were modest in nature with\nthe average correlation coefficient about 0.09 in magnitude. The overwhelming majority of\nestimated correlation coefficients were found to be less than 0.3. Thus, we do not believe\nthat multicollinearity represents a serious issue for the econometric analysis undertaken\nhere. In addition, the role of this particular phenomenon is likely to be attenuated by the\nlarge sample size used in estimation.\n\n7 Unfortunately, the data do not allow us to distinguish whether the migrant works for the\ngovernment or in a manufacturing state-owned enterprise.\n\n8 Among those migrants who faced some difficulties, the main ones are reported in rank\norder to be housing problems, having no income source, and, related to this, not being able\nto find a job.\n\n9 The estimation of the migrant remittance function using cross-sectional data,\n\nMention: \"Vietnam Household Living Standards Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The later round behind the calculated comparisons.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003804:22:0", "occurrence_id": "8c2f948c39cbdf3f7583a4e6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Private participation in infrastructure_\n\n\nPPI can expand access and quality of infrastructure services, with attendant increases in labor and\n\n\nservice productivity (Ramamurti, 1996; Agénor and Moreno-Dodson, 2006; ILO, 2002), support\n\n\nefficiency improvements and deployment of new technology in utilities, introduce alternative\n\n\nmechanisms to finance infrastructure and free scare resources to channel towards growth and\n\n\nother development policy priorities (Besant-Jones, 2006; ADB, JBIC, World Bank, 2005; Harris,\n\n\n2003) – all of which may positively impact growth. We therefore expect to find evidence of\n\n\ndifferential impact of infrastructure investments on a country’s growth depending upon\n\n\nprevalence of PPI. In this paper, the dummy variable “private” summarizes the information on the\n\n\ntiming of private participation in specific infrastructure subsectors in a country. The variable\n\n\ntakes the value 0 before the first year of private entry and the value 1 from that year on. Overall,\n\n\nprivate investors have invested approximately US$ 224 billion since 1990 (World Bank PPI\n\nDatabase) in East Asia’s infrastructure subsectors across 18 countries <sup>20</sup> [^20: Electricity, roads, water supply and sanitation, and telecom accounted, respectively, for 34%, 18%,\n16% and 7% of the number of PPI projects in the region. However, the distribution of projects has been\nskewed across East Asian countries, with over 77% of water and sanitation, 68% of roads, and 42% of\nelectricity projects located in China alone.] .\n\n\n**Table 6: Augmented Solow model with infrastructure variables and the effect of privatization**\n\n<u>Telecoms</u> <u>Electricity</u> <u>Roads</u> <u>Sanitation</u> <u>Water</u>\n<u>Bench</u> <u>Infrast</u> <u>Bench</u> <u>Infrast</u> <u>Bench</u\n\nMention: \"World Bank PPI\n\nDatabase\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Cumulative private investment in regional infrastructure is measured here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003768:39:0", "occurrence_id": "92b6ecbd3ab024798c93f009", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: #### **Appendix 1: The Surveys**\n\nThis Appendix describes the two main sources of information used in this paper: the\nIntegrated Household Survey and the Tobacco Clubs Survey. The household survey is the\nmain source of data for our analysis. The Tobacco Clubs Survey has a more modest role as\ncomplementary data to support some of the claims in the empirical analysis. It is becoming\nincreasingly common, especially in research on institutions in Africa, to supplement a major\ndata set with pieces of evidence from small-scale surveys or interviews with farmers. See\nUdry (2003) for a discussion of different research strategies and data collection efforts in\nAfrica.\n\n###### **The Malawi Integrated Household Survey (IHS)**\n\n\nThe sample design of the IHS is a two-stage stratified sampling procedure from a sample\nframe of Enumerator Areas (EA) from the 1998 Population Census. The survey covers the\nthree geographical regions (north, center and south), dividing the country into rural and\nurban areas. The urban stratum includes the four major urban areas (Lilongwe, Blantyre,\nMzuzu, and the Municipality of Zomba); all other areas are considered as rural. The total\nsample is 20 households per each of the 564 EA selected, resulting in 11,280 households\nsurveyed.\n\nThe Integrated Household Survey (IHS) consists of two questionnaires: a household\nquestionnaire, and a community questionnaire. The community questionnaires include\ntypical questions on socio-economic conditions at the village level. The household\nquestionnaire includes several modules: A) Household identification; B) Household roster\n(size, demographics; marital status); C) Education (schooling attainments); D) Health\n(recent illness or injury, chronic disease, treatments); E) Time use and labor (household\nwork, employment); F) Security and safety (crime, crime protection); G) Housing (dwelling\ncharacteristics); H) Consumption (over the last three\n\nMention: \"Integrated Household Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The main source of information for the analysis, sampled in two stratified stages.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003768:39:0", "occurrence_id": "26b27279e644901b097e9877", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: #### **Appendix 1: The Surveys**\n\nThis Appendix describes the two main sources of information used in this paper: the\nIntegrated Household Survey and the Tobacco Clubs Survey. The household survey is the\nmain source of data for our analysis. The Tobacco Clubs Survey has a more modest role as\ncomplementary data to support some of the claims in the empirical analysis. It is becoming\nincreasingly common, especially in research on institutions in Africa, to supplement a major\ndata set with pieces of evidence from small-scale surveys or interviews with farmers. See\nUdry (2003) for a discussion of different research strategies and data collection efforts in\nAfrica.\n\n###### **The Malawi Integrated Household Survey (IHS)**\n\n\nThe sample design of the IHS is a two-stage stratified sampling procedure from a sample\nframe of Enumerator Areas (EA) from the 1998 Population Census. The survey covers the\nthree geographical regions (north, center and south), dividing the country into rural and\nurban areas. The urban stratum includes the four major urban areas (Lilongwe, Blantyre,\nMzuzu, and the Municipality of Zomba); all other areas are considered as rural. The total\nsample is 20 households per each of the 564 EA selected, resulting in 11,280 households\nsurveyed.\n\nThe Integrated Household Survey (IHS) consists of two questionnaires: a household\nquestionnaire, and a community questionnaire. The community questionnaires include\ntypical questions on socio-economic conditions at the village level. The household\nquestionnaire includes several modules: A) Household identification; B) Household roster\n(size, demographics; marital status); C) Education (schooling attainments); D) Health\n(recent illness or injury, chronic disease, treatments); E) Time use and labor (household\nwork, employment); F) Security and safety (crime, crime protection); G) Housing (dwelling\ncharacteristics); H) Consumption (over the last three\n\nMention: \"Tobacco Clubs Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"A secondary source with a more modest role in the analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003768:39:0", "occurrence_id": "99c7921a08e7cc6f048b7820", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: #### **Appendix 1: The Surveys**\n\nThis Appendix describes the two main sources of information used in this paper: the\nIntegrated Household Survey and the Tobacco Clubs Survey. The household survey is the\nmain source of data for our analysis. The Tobacco Clubs Survey has a more modest role as\ncomplementary data to support some of the claims in the empirical analysis. It is becoming\nincreasingly common, especially in research on institutions in Africa, to supplement a major\ndata set with pieces of evidence from small-scale surveys or interviews with farmers. See\nUdry (2003) for a discussion of different research strategies and data collection efforts in\nAfrica.\n\n###### **The Malawi Integrated Household Survey (IHS)**\n\n\nThe sample design of the IHS is a two-stage stratified sampling procedure from a sample\nframe of Enumerator Areas (EA) from the 1998 Population Census. The survey covers the\nthree geographical regions (north, center and south), dividing the country into rural and\nurban areas. The urban stratum includes the four major urban areas (Lilongwe, Blantyre,\nMzuzu, and the Municipality of Zomba); all other areas are considered as rural. The total\nsample is 20 households per each of the 564 EA selected, resulting in 11,280 households\nsurveyed.\n\nThe Integrated Household Survey (IHS) consists of two questionnaires: a household\nquestionnaire, and a community questionnaire. The community questionnaires include\ntypical questions on socio-economic conditions at the village level. The household\nquestionnaire includes several modules: A) Household identification; B) Household roster\n(size, demographics; marital status); C) Education (schooling attainments); D) Health\n(recent illness or injury, chronic disease, treatments); E) Time use and labor (household\nwork, employment); F) Security and safety (crime, crime protection); G) Housing (dwelling\ncharacteristics); H) Consumption (over the last three\n\nMention: \"1998 Population Census\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Its enumerator areas form the sample frame for the household survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003810:8:0", "occurrence_id": "42bf98f5305943cecaad1e86", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: through each district in the surveyed countries. Data on elevation at the centroid of each\n\n\ndistrict were obtained from the United States Geological Survey (USGS 2004). The\n\n\nUSGS data are derived from a global digital elevation model with a horizontal grid\n\n\nspacing of 30 arc seconds (approximately one kilometer).\n\n###### **4. Empirical Results**\n\n\nAfrica contains a wide range of agricultural ecological zones across the continent. Due to\n\n\nits diverse ecological zones, farmers are expected to rely on different farm types, crops,\n\n\nlivestock, inputs, and outputs depending upon the characteristics of the farm location.\n\n\nTable 1 describes how farmers chose different farm types in the sample. Livestock was\n\n\nchosen in about 65 percent of the total farms. Irrigated farms account for about 25 percent\n\n\nof farmers. Across the sample, the proportion of each of the five distinct farm types is:\n\n\ncrop-only rainfed farms (28 percent), crop-only irrigated farms (10 percent), mixed\n\n\nrainfed farms (42 percent), mixed irrigated farms (14 percent), and livestock-only farms\n\n\n(6 percent).\n\n\nWe examine whether the choice of these farm types is sensitive to climate. We begin by\n\n\nexamining a cross section of farmers who face different climate conditions. We\n\n\nhypothesize that farm type choice is influenced by climate. Table 2 shows the results\n\n\nfrom a multinomial logit regression of the five farm types against a set of independent\n\n\nvariables which include climate variables in quadratic form, soil variables, household\n\n\ncharacteristics, water flow, pasture, and output prices. The choice of livestock-only is\n\n\nomitted as the base case. Several control variables are significant. When the farm has\n\n\nelectricity, farmers favor livestock-only farms over other choices. This may be because\n\n\nelectricity is needed for milk production and for storage of livestock products or it may\n\n\nbe because electricity is correlated with other missing variables that favor livestock.\n\n\nWhen Lithosol soils are dominant in a district, farmers tend\n\nMention: \"global digital elevation model\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Terrain restrictions derive from this one-kilometre grid.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003813:36:4", "occurrence_id": "04f60325cd38ab48d0adf3be", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: * significant at 1\n\nis as indicated at the top of the column and the independent variables\nrence-in-differences coefficient presented): young cohort * DHS 2000, hi\n\n\n\ncant at 10%\n\n\n\nsignificant\n\n\n\n%, *** sig\n\n\n\nNotes: Robust standard errors in brackets, clustered at the enumeration level. * significant at 10%, ** significant at 5%, *** significant at 1%.\nEach cell represents a separate regression where the dependent variable is as indicated at the top of the column and the independent variables\n\ninclude the following variables (in addition to the difference-in-difference-in-differences coefficient presented): young cohort * DHS 2000, high\n\nwar intensity *young cohort, high war intensity * DHS 2000, DHS 2000, female, non-poor, child age fixed effects, province fixed effects, and\n\nhousehold level controls as in previous tables. The difference-in-difference-in-differences variable is the interaction of a variable measuring war\nintensity, a dummy variable for being in the young cohort aged 6 to 15, and a dummy variable for being in the 2000 DHS. High war intensity\n\nmeasure A is defined as the proportion of days during the genocide during which killings occurred. High war intensity measure B is a dummy\n\nvariable indicating the 3 provinces that had the most number of genocide victims High war intensity measu C is defined as th. re e number of mass\n\ngraves sites and memorials per province. Regressions are restricted to individuals aged 6 to 35. Non-poor is a dummy variable indicating the\n\nindividual has more assets than the population mean. Assets used to calculate this wealth measure include piped running water, refrigerator, radio,\n\nfinished floor, bicycle, motorcycle, and car. Household level controls include age of household head, highest education level of any household\n\n\n\nic\n\n\n\nated at the to\n\nrences coeffi\n\n\n\np\n\n\n\np of the column a\n\ncient presented):\n\n\n\na\n\n\n\nhe indepe\nun\n\nMention: \"DHS 2000\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Its round supplies the difference-in-differences comparison across cohorts.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003813:36:4", "occurrence_id": "58acda919d9a02949d4fffbf", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: * significant at 1\n\nis as indicated at the top of the column and the independent variables\nrence-in-differences coefficient presented): young cohort * DHS 2000, hi\n\n\n\ncant at 10%\n\n\n\nsignificant\n\n\n\n%, *** sig\n\n\n\nNotes: Robust standard errors in brackets, clustered at the enumeration level. * significant at 10%, ** significant at 5%, *** significant at 1%.\nEach cell represents a separate regression where the dependent variable is as indicated at the top of the column and the independent variables\n\ninclude the following variables (in addition to the difference-in-difference-in-differences coefficient presented): young cohort * DHS 2000, high\n\nwar intensity *young cohort, high war intensity * DHS 2000, DHS 2000, female, non-poor, child age fixed effects, province fixed effects, and\n\nhousehold level controls as in previous tables. The difference-in-difference-in-differences variable is the interaction of a variable measuring war\nintensity, a dummy variable for being in the young cohort aged 6 to 15, and a dummy variable for being in the 2000 DHS. High war intensity\n\nmeasure A is defined as the proportion of days during the genocide during which killings occurred. High war intensity measure B is a dummy\n\nvariable indicating the 3 provinces that had the most number of genocide victims High war intensity measu C is defined as th. re e number of mass\n\ngraves sites and memorials per province. Regressions are restricted to individuals aged 6 to 35. Non-poor is a dummy variable indicating the\n\nindividual has more assets than the population mean. Assets used to calculate this wealth measure include piped running water, refrigerator, radio,\n\nfinished floor, bicycle, motorcycle, and car. Household level controls include age of household head, highest education level of any household\n\n\n\nic\n\n\n\nated at the to\n\nrences coeffi\n\n\n\np\n\n\n\np of the column a\n\ncient presented):\n\n\n\na\n\n\n\nhe indepe\nun\n\nMention: \"2000 DHS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"A dummy for this round marks the later observations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003818:23:0", "occurrence_id": "3a701ab468b85288bf235449", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: pole has the correct negative sign but it is not statistically signi fi cant. The estimated marginal\n\n\ne ff ects imply that the probability of being employed as a factory worker and as a service workers\n\n\ndeclines with an increase in distance from growth poles. The estimates in Table 5 also suggest\n\n\nthat services are somewhat concentrated around smaller towns implying importance of local\n\n\ndemand in the case of non-traded activities. This fi nding is similar to that reported in Foster\n\n\nand Rosenzweig (2004a). The results in Table 5 also con fi rm our earlier fi nding that the negative\n\n\ne ff ect of isolation is magni fi ed in regions with higher crop production potential.\n\n##### 5 Conclusions\n\n\nRural areas, home to most of the world’s poor, have experienced considerable employment\n\n\ndiversi fi cation into nonfarm activities during the last decade. Recognizing its potential for\n\n\nstimulating growth and reducing poverty, there has been a resurgence of interest among re\n\nsearchers and policy makers in identifying factors that can spark growth in the rural nonfarm\n\n\neconomy. While the roles of farm and urban linkages in promoting nonfarm activities are dis\n\ncussed frequently by development practitioners, empirical studies examined them in isolation of\n\n\neach other. Embedding the two linkages in a single speci fi cation, this paper provides empirical\n\n\nevidence on the relative role of farm and urban linkages in determining the probability of being\n\n\nemployed in di ff erent types of non-farm activities.\n\n\nThe econometric analysis, using individual level employment data from the Household In\n\ncome and Expenditure Survey, 2000 of Bangladesh, yields three main results. First, access to\n\n\nmajor urban centers matters greatly for high-return nonfarm activities. Second, agricultural\n\n\npotential of a village also matters, but through its interaction with access to urban centers.\n\n\nThe results speci fi cally suggest that the likelihood of being employed in high return jobs and\n\n\nin self-employment increases with a decrease in distance to growth pole and the negative e\n\nMention: \"Household In\n\ncome and Expenditure Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Individual employment data on non-farm activity come from this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003822:24:0", "occurrence_id": "8e7a1d6a281d7012ea47413b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: beneficial to concentrate initially on improving transport links with regional partners. In saying\n\n\nthis, we are conscious that infrastructure upgrading can involve significant costs. Whereas we\n\n\nemphasize the relative benefits of different options in this paper, informed policymaking would\n\n\nof course need to be based on a rigorous cost-benefit assessment in each case.\n\n\nSecond, e-business and connectivity would also seem to be very important for the region (see\n\n\n_Scenario 4_ ). ASEAN member countries may therefore also benefit from giving priority to\n\n\nimprovement of information technology infrastructure and, more generally, the competitiveness\n\n\nof backbone services sectors. Such an agenda could include both domestic regulatory reforms,\n\n\nand increased openness to international services trade in these sectors under the GATS.\n\n\nFinally, there is still progress to be made in reducing “traditional” trade barriers (such as tariffs),\n\n\nand in improving governance and transparency in such a way as to limit the role of unofficial\n\n\npayments in import/export transactions. While significant economic benefits could flow from\n\n\nthese steps, the data indicate at this stage that it may be preferable from a trade flow point of\n\n\nview to concentrate most heavily on the other two areas in the short-term. These conclusions are\n\n\nbroadly consistent with the data presented earlier indicating that tariff protection in Southeast\n\n\nAsia is generally at a low to moderate level. However, those same data—based on the Kee et al.\n\n\n(2006) OTRI—suggest that non-tariff barriers may play a significantly more important role. Due\n\n\nto lack of information at this point, we have not explicitly considered the role of non-tariff\n\n\nbarriers in our gravity model. This would be an important point for future research to elaborate\n\n\non, so as to give reform of these measures an appropriate place in the policy ranking we are\n\n\nsuggesting.\n\n\nBefore concluding this Section, it is important to stress that our results, like all simulation results,\n\n\nare subject to a number of caveats. First, our results in relation to tariffs, maritime port\n\nMention: \"Kee et al.\n\n\n(2006)\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its overall trade restrictiveness index underlies the barrier estimates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003827:7:1", "occurrence_id": "14197495a23653bc401814aa", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ) to an absolute poverty measure.\n\n\nWe begin by presenting our new cross-country data set on poverty lines in developing countries.\n\n###### **3. National poverty lines across developing countries**\n\n\nThe set of national poverty lines collected by RDV covered 33 countries (both developed\n\n\nand developing) and drew on specialized, country-specific, mostly academic, studies of poverty\n\nspanning 1980-90. Clearly, this data set is now rather old. <sup>13</sup>\n\n\nSince RDV, there has been considerable expansion in research and analysis on poverty in\n\n\ndeveloping countries through the World Bank’s country-level _Poverty Assessments_ (PA’s),\n\n\nwhich have now been done for many developing countries (though there were very few PA’s\n\n\navailable in 1990). These are core reports within the Bank’s program of analytic work at country\n\n\nlevel; each report describes the extent of poverty and its causes in that country. The Bank’s PA\n\n\n12 Ravallion and Lokshin (2007) provide survey-based evidence for one of the poorest countries,\nMalawi, indicating that relative deprivation is not a welfare concern to the bulk of the population, though\nthey do find that it matters to better off groups in urban areas. For a theoretical analysis in a model\ncombining risk sharing with positional goods see Ravallion (2008a).\n13 The only prior update was done by Ravallion (1998), who tested the effect of including a number\nof new observations for Africa (which was clearly under-represented in the Ravallion et al., 1991, data\nset). The main result of the earlier study by RDV on the relationship between poverty lines and mean\nconsumption was found to be quite robust to including these extra observations.\n\n\n6\n\nMention: \"Ravallion et al., 1991, data\nset\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"New African observations address its under-representation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003832:19:0", "occurrence_id": "f5b8f9dd00b21d11e6e85e78", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Table 2: School construction programme summary statistics_\n\n\n\n**<u>Panel A: All districts</u>** <sup>**<u>(a)</u>**</sup> **<u>Panel B: IFLS districts</u>** <sup>**<u>(b)</u>**</sup>\n**Mean** **Standard** **Mean** **Standard**\n**deviation** **deviation**\n\n\n\n**Mean** **Standard** **Mean** **Standard**\n**deviation** **deviation**\n\nTotal number of districts in 293 222\nsample\n\n\n\n293 222\n\n\n\nINPRES schools constructed\n(1973/74 - 1978/79)\n\nINPRES schools constructed\nper 1000 children\n(1973/74 - 1978/79)\n\nFraction of population that\nwere children in 1971\n\nEnrolment rate in the\npopulation in 1971\n\nAllocation of water and\nsanitation programmes in\n<u>1973/74</u>\n\n\n\n222 173 252 182\n\n\n2.3387 1.2621 2.2105 1.0706\n\n\n0.2868 0.0772 0.2714 0.0176\n\n\n0.1777 0.0969 0.1689 0.0606\n\n\n0.7064 0.5680 0.5525 0.2959\n\n\n\nNotes:\n(a) All districts for which school construction data were available\n(b) Districts with school construction data represented in IFLS survey (districts of birth)\nSources:\nData on school construction programme collected by Esther Duflo from INPRES instructions, Census\n(1971) and Ministry of Education and Culture.\n\n###### **6. Analytical framework**\n\n\n**Economic demand model**\n\n\nEmpirical analysis of health care choices is\n\nMention: \"IFLS survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Districts of birth covered by the school construction programme are identified here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003832:19:0", "occurrence_id": "5fec09e5ce0a1a44e5179cdd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Table 2: School construction programme summary statistics_\n\n\n\n**<u>Panel A: All districts</u>** <sup>**<u>(a)</u>**</sup> **<u>Panel B: IFLS districts</u>** <sup>**<u>(b)</u>**</sup>\n**Mean** **Standard** **Mean** **Standard**\n**deviation** **deviation**\n\n\n\n**Mean** **Standard** **Mean** **Standard**\n**deviation** **deviation**\n\nTotal number of districts in 293 222\nsample\n\n\n\n293 222\n\n\n\nINPRES schools constructed\n(1973/74 - 1978/79)\n\nINPRES schools constructed\nper 1000 children\n(1973/74 - 1978/79)\n\nFraction of population that\nwere children in 1971\n\nEnrolment rate in the\npopulation in 1971\n\nAllocation of water and\nsanitation programmes in\n<u>1973/74</u>\n\n\n\n222 173 252 182\n\n\n2.3387 1.2621 2.2105 1.0706\n\n\n0.2868 0.0772 0.2714 0.0176\n\n\n0.1777 0.0969 0.1689 0.0606\n\n\n0.7064 0.5680 0.5525 0.2959\n\n\n\nNotes:\n(a) All districts for which school construction data were available\n(b) Districts with school construction data represented in IFLS survey (districts of birth)\nSources:\nData on school construction programme collected by Esther Duflo from INPRES instructions, Census\n(1971) and Ministry of Education and Culture.\n\n###### **6. Analytical framework**\n\n\n**Economic demand model**\n\n\nEmpirical analysis of health care choices is\n\nMention: \"Census\n(1971)\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its population counts help locate the programme's districts.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003836:1:0", "occurrence_id": "e9fe21ced81e04bf1c4f0b6d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Policy Research Working Paper 4629\n\n### **Abstract**\n\n\nThe Addis Ababa Integrated Housing Development\nProgram aims to tackle the housing shortage and\nunemployment that prevail in Addis Ababa by deploying\nand supporting small and medium scale enterprises to\nconstruct low-cost housing using technologies novel for\nEthiopia. The motivation for such support is predicated\non the view that small firms create more jobs per unit\nof investment by virtue of being more labor intensive\nand that the jobs so created are concentrated among the\nlow-skilled and hence the poor. To assess whether the\nprogram has succeeded in biasing technology adoption\nin favor of labor and thereby contributed to poverty\nreduction, the impact of the program on technology\n\n\n\nusage, labor intensity, and earnings is investigated using\na unique matched workers-firms dataset, the Addis\nAbaba Construction Enterprise Survey. The data are\nrepresentative of all registered construction firms in\nAddis and were collected specifically for the purpose of\nanalyzing the impact of the program. The authors find\nthat program firms do not adopt different technologies\nand are not more labor intensive than non-program\nfirms. There is an earnings premium for program\nparticipants, who tend to be relatively well-educated,\nwhich is heterogeneous and highest for those at the\nbottom of the earnings distribution.\n\n\n\nThis paper—a product of the Poverty Reduction and Economic Management Department, Africa Region—is part of a\nlarger effort in the department to support evidence based policy making. Policy Research Working Papers are also posted\non the Web at http://econ.worldbank.org. The author may be contacted at cruggeriladerchi@worldbank.org.\n\n\n_The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development_\n_issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the_\n_names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed\n\nMention: \"Addis\nAbaba Construction Enterprise Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"A matched worker-firm dataset representative of registered construction firms.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003840:5:0", "occurrence_id": "d5b5ba8be1d20b7bcbfd9793", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 4\n\n\n**Figure 1: Fiscal Policy Volatility and GDP Volatility**\n\n\nSource: Calculations are based on WDI data. Volatility is defined as the standard deviation of growth rates.\n\n\n**Figure 2: Fiscal Policy Volatility and Household Consumption Volatility**\n\n\nSource: Calculations are based on WDI data. Volatility is defined as the standard deviation of growth rates\n\nThe relationship between fiscal policy volatility and output volatility (Figure 1) has been\nstudied by Fatas and Mihov in a series of papers, but the relationship with consumption\n\nMention: \"WDI data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Volatility of fiscal policy and output is computed as the standard deviation of growth from these data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003842:8:0", "occurrence_id": "adbeabbda0eba1b5df7357a4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: should expect a broad-based skill set, and the desire to juggle many tasks simultaneously,\n\n\nto be more important in a developing country context.\n\n\nAll of these factors collectively suggest that (self) selection of entrepreneurs from\n\n\nthe population may differ in developing and developed countries. Unfortunately, the\n\n\nmanner in which the entrepreneurial skill set has been assessed across countries is rather\n\n\nad hoc. This makes comparison of the factors associated with selection into\n\n\nentrepreneurship difficult to compare internationally.\n\n\nThe combined theoretical and empirical literature therefore suggests a rich set of\n\n\nfactors differentiating entrepreneurs from wage workers in both unconstrained and\n\n\nconstrained environments. We draw on this literature to identify the relevant factors for\n\n\nplacing the own account workers on the wage worker – entrepreneur scale. Our interest in\n\n\nundertaking this exercise is practical: From a policy perspective, should we focus on the\n\n\nown account workers as a potential source of job growth? Or should we instead focus on\n\n\nincreasing incomes of the self employed where possible, in the interest of poverty\n\n\nalleviation, while looking elsewhere for the generation of jobs which these own account\n\n\nworkers will fill?\n\n\n**Data and empirical approach**\n\n\nWe use data from a series of surveys carried out in Sri Lanka between 2005 and\n\n\n2007. The first is a survey of 618 own account workers, selected from three districts in\n\n\nsouth and southwestern Sri Lanka, which we refer to as the Sri Lankan Microenterprise\n\n\nSurvey (SLMS). The own account workers were surveyed quarterly between April 2005\n\n\nand April 2007, and then again in October 2007. Each survey included questions on the\n\n\noperating performance of the firms during the month preceding the survey, and questions\n\n\non additional topics which varied from quarter to quarter. For example, in the July 2005\n\n\nsurvey, we played a lottery game from which we measure attitudes toward risk. In the\n\n\nJanuary 2006 survey, we recorded data on the labor history of respondents; and in the\n\n\nOctober 2007, we gave the respondents a Raven progressive non-verbal test of reasoning.\n\n\nThe survey questions relevant to the\n\nMention: \"Sri Lankan Microenterprise\n\n\nSurvey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Own-account workers in three districts were followed quarterly.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003846:9:1", "occurrence_id": "94678cfdbe378461c2e64b4d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  3.1 3.7 2.4 2.7\nNicaragua -0.1 2.1 2.8 5.4 1.4 2.6\n<u>Peru</u> <u>1.1</u> <u>2.2</u> <u>2.2</u> <u>3.0</u> <u>1.7</u> <u>2.1</u>\n<u>Average</u> <u>1.4</u> <u>2.1</u> <u>2.4</u> <u>4.1</u> <u>1.9</u> <u>2.8</u>\n\n- Excludes the 1980s to compute the average growth rate.\n** Data from 1980-1993 are missing.\nFCT. Projections of growth under a continuous trend of reform scenario.\nFDC. Projections of growth under a dramatic change reform scenario.\nFCT1. Projections of growth resulting from merging FCT1 and average growth over 1961-2006.\nFDC. Projections of growth resulting from merging FDC and average growth over 1961-2006.\nSources: Own calculation using data from WDI and Loayza et al. (2005).\n\nA second reason of why past growth rates in per capita consumption are a poor proxy for\nexpected growth rates in the Latin American context is because of the potential role\nplayed by the reforms implement by the region in 1990s. Comparing the situation in the\nearly 2000s and early 1990s, Latin American countries have better macroeconomic\nframeworks, are more open to trade, have more developed financial sectors, and in\nvirtually all countries public spending in education has increased. All these elements are\nusually considered in the growth literature as growth enhancers. Moreover, per capita\nconsumption growth in\n\nMention: \"data from WDI\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Growth projections merge these series with an alternative estimator.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003851:38:0", "occurrence_id": "2967d25abff5445cd90575e7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: FIGURE 6. BOTTOM 5TH PERCENTILE OF MATHEMATICS ACHIEVEMENT, INCLUDING\n\nORISSA AND RAJASTHAN\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nSource: TIMSS Mathematics 2001, Grade 8, Exhibit 1.1; TIMSS Mathematics 2003, Grade 8, Exhibit D.1;\nauthors’ calculations from Rajasthan and Orissa Secondary School Survey, Grade 9, 2005. Most recent year\nis used when both are available.\n\n\nNotes: The Indian percentiles are calculated using the plausible values methodology described in the text.\n\nMention: \"Rajasthan and Orissa Secondary School Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Grade nine achievement percentiles are calculated from this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003853:11:0", "occurrence_id": "b9687bdee2161aa4952eb9a3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **IV.** **Bank Regulation and Supervision Around the World** : **What the Data Mean**\n\n\nIV.A. How reforms affect banking systems: Overview\n\n\nHow have reforms to bank regulations and supervisory practices affected national banking\n\n\nsystems? In countries that changed their regulatory policies, have these reforms reduced banking\n\n\nsystem fragility and boosted banking system development? Have these policy changes enhanced\n\n\nthe efficiency of intermediation and moderated corruption in the lending process? Answers to\n\n\nthese questions will help some countries adjust their reforms and help other countries avoid\n\n\nmistakes and select more appropriate reform strategies.\n\n\nGiven the available data, we conduct some illustrative simulations to assess the\n\n\npotentially impact of bank-regulatory reforms over the last decade on national banking systems.\n\n\nIn the first step, we estimate the relationships between bank regulations and banking-system\n\n\nfragility, development, efficiency, and corruption in lending. These estimates are based on\n\n\nSurvey I and analyses in BCL (2006). In the second step, we use the estimate coefficient from\n\n\nthe first step to compute the impact of regulatory reforms between Survey I and Survey III on\n\n\nbanking-system fragility, development, efficiency, and corruption. We make these computations\n\nfor each country and describe this simulation strategy in greater detail below. <sup>5</sup>\n\n\nIV.B Baseline regressions\n\n\nTable 1 presents estimates of the relationships between various bank regulations and banking\n\nsystem fragility, development, efficiency, and corruption. Since BCL (2006) explain these\n\nestimation processes in great detail, we provide a very brief synopsis of that description. <sup>6</sup>\n\n\n5 One difference between the estimates reported in this paper and our earlier work is that here we now use indexes\nbased on the summation of the individual questions, rather than computing the principal component of the individual\nquestions underlying the indexes. We do this because it makes it much more transparent to see how changes in an\nindividual question changed the index, and hence the estimated\n\nMention: \"Survey I\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Its estimates give banking fragility, efficiency and lending corruption.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003853:11:0", "occurrence_id": "6d88aea806e82f100b5f7581", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **IV.** **Bank Regulation and Supervision Around the World** : **What the Data Mean**\n\n\nIV.A. How reforms affect banking systems: Overview\n\n\nHow have reforms to bank regulations and supervisory practices affected national banking\n\n\nsystems? In countries that changed their regulatory policies, have these reforms reduced banking\n\n\nsystem fragility and boosted banking system development? Have these policy changes enhanced\n\n\nthe efficiency of intermediation and moderated corruption in the lending process? Answers to\n\n\nthese questions will help some countries adjust their reforms and help other countries avoid\n\n\nmistakes and select more appropriate reform strategies.\n\n\nGiven the available data, we conduct some illustrative simulations to assess the\n\n\npotentially impact of bank-regulatory reforms over the last decade on national banking systems.\n\n\nIn the first step, we estimate the relationships between bank regulations and banking-system\n\n\nfragility, development, efficiency, and corruption in lending. These estimates are based on\n\n\nSurvey I and analyses in BCL (2006). In the second step, we use the estimate coefficient from\n\n\nthe first step to compute the impact of regulatory reforms between Survey I and Survey III on\n\n\nbanking-system fragility, development, efficiency, and corruption. We make these computations\n\nfor each country and describe this simulation strategy in greater detail below. <sup>5</sup>\n\n\nIV.B Baseline regressions\n\n\nTable 1 presents estimates of the relationships between various bank regulations and banking\n\nsystem fragility, development, efficiency, and corruption. Since BCL (2006) explain these\n\nestimation processes in great detail, we provide a very brief synopsis of that description. <sup>6</sup>\n\n\n5 One difference between the estimates reported in this paper and our earlier work is that here we now use indexes\nbased on the summation of the individual questions, rather than computing the principal component of the individual\nquestions underlying the indexes. We do this because it makes it much more transparent to see how changes in an\nindividual question changed the index, and hence the estimated\n\nMention: \"Survey III\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Comparing it with the first round measures the impact of regulatory reform.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003855:44:1", "occurrence_id": "cd7bf51880ac7d1340890878", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  procedure by Geweke et al. (1986) outlined in Section III. Computations are based on\n10.000 Monte Carlo replications. The series have been smoothed using a 4 quarter moving average to remove high frequency fluctuations. O=Out of the Labour\nForce, U=Unemployment, E=Employment, S=Informal Self-Employed, I=Informal Salaried, and F=Formal Sector, all as proportions of working age population.\nData for Mexico (left panels) is drawn from the quarterly National Urban Labor Survey (ENEU) from 1987:Q1 to 2004:Q4. Data for Brazil (right panels) is drawn\nfrom the Monthly Employment Survey (PME), quarterly averaged from 1983:Q1 to 2001:Q2.\n\n\n43\n\nMention: \"quarterly National Urban Labor Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Formal and informal sector shares over this period come from this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003855:44:1", "occurrence_id": "51ecac804172bb61700b8c8e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  procedure by Geweke et al. (1986) outlined in Section III. Computations are based on\n10.000 Monte Carlo replications. The series have been smoothed using a 4 quarter moving average to remove high frequency fluctuations. O=Out of the Labour\nForce, U=Unemployment, E=Employment, S=Informal Self-Employed, I=Informal Salaried, and F=Formal Sector, all as proportions of working age population.\nData for Mexico (left panels) is drawn from the quarterly National Urban Labor Survey (ENEU) from 1987:Q1 to 2004:Q4. Data for Brazil (right panels) is drawn\nfrom the Monthly Employment Survey (PME), quarterly averaged from 1983:Q1 to 2001:Q2.\n\n\n43\n\nMention: \"Monthly Employment Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Quarterly averages of its monthly series give the comparison country.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003860:14:2", "occurrence_id": "4af1e44de6febfa0bfc31af8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  61 <sup>st</sup> round are broadly in keeping with our\n\n\nanalyses based on previous rounds of the NSS. Muslim status has no significant association with labor\n\n\nforce participation for men (but it has huge effects for women) in either urban or rural areas. Dalit status\n\n\nis on the other hand associated with higher labor force participation in rural areas but has no effect in\n\n\nurban areas. Higher levels of education are positively associated with men’s labor force participation\n\n\n(though not women’s as we have discussed elsewhere) and these effects are expectedly stronger in cities\n\n\nand towns than in villages since labor market opportunities are greater in the former. We have argued\n\n\nelsewhere that land is not merely an economic asset but also a marker of status and influences a number\n\n\nof outcomes. Here too we find that owning land is associated not only with greater likelihood of labor\n\n\nforce participation but also the likelihood of being in non-farm ventures.\n\n\nThe complexity of this analysis comes from the multinomial models, which estimate the probability of\n\n\nbeing in different employment categories compared to salaried work. Where the question of self\n\nemployment is concerned, the results from urban areas are more relevant than for rural areas. In the\n\n\n12\n\nMention: \"the NSS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Earlier rounds give the same lack of association between minority status and participation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003866:3:1", "occurrence_id": "a9ae3dcf3d26d978462edc81", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  groups. <sup>2</sup>\n\n\nThe distinction between inequality of opportunity and the more standard concept\n\n\nof inequality of outcomes is of interest to economists for at least three sets of reasons.\n\n\nFirst, if inequality of opportunity does affect attitudes to outcome inequality, then it may\n\n\naffect attitudes to redistribution and beliefs about social fairness. These beliefs may in\n\n\nturn affect the extent of redistribution actually implemented, and the level of investment\n\n\nand output generated. Alesina and Angeletos (2005) and Bénabou and Tirole (2006) are\n\n1 The World Value Survey is conducted by the Inter-university Consortium for Political and Social\nResearch, based at the University of Michigan, and contains responses from representative samples in 69\ncountries.\n2 Roemer (1998) was not, of course, the first economist or philosopher to argue that the space of\nopportunities was ethically the right one to focus on. Arneson (1989), Cohen (1989) and, to some extent,\nSen (1985) had made broadly similar points. By providing a simple, yet powerful, formalization of the\ndefinition of equal opportunities, however, Roemer contributed to an increase in interest in the concept\nfrom applied economists.\n\n\n2\n\nMention: \"The World Value Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"A footnote describes who conducts this attitudinal survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003868:30:0", "occurrence_id": "6fb8121a497137699336a0a8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Appendix Table A2. Variables - definitions and sources**\n\n**<u>Variable</u>** **<u>Definition</u>** **<u>Source</u>**\nBank Credit to GDP Total outstanding claims of deposit Beck, Demirguc-Kunt and Levine\nmoney banks on private sector as (2000)\nratio to GDP\n\n\n\nEnterprise Credit to GDP Total outstanding claims of deposit\nmoney banks on enterprise sector as\nratio to GDP\n\nHousehold Credit to GDP Total outstanding claims of deposit\nmoney banks on households as ratio\nto GDP\n\n\n\nSee Appendix Table A1\n\n\nSee Appendix Table A1\n\n\n\nGDP per capita World Development Indicators\n(WDI)\nSecondary enrolment Share of the respective age cohort WDI\nenrolled in secondary schools\n\n\n\nGovernment consumption Total govt. expenditures relative to\nGDP\n\n\n\nWDI\n\n\n\nTrade Ratio of exports and imports to GDP WDI\nInflation Average log difference in the WDI\nConsumer Price Index over the\nsample period\n\n\n\nLegal origin dummies Origin country of each country’s\nlegal system\n\n\n\nLegal origin dummies Origin country of each country’s La Porta, Lopez-de-Silanes, Shleifer\nlegal system and Vishny (1999)\n\nCatholic, Protestant and Muslim Share of population with the La Porta, Lopez-de-Silanes, Shleifer\npopulation shares respective religious belief in total and Vishny (1999)\n\n\n\nShare of population with the\nrespective religious belief in total\npopulation\n\n\n\nLa Porta, Lopez-de-Silanes, Shleifer\nand Vishny (1999)\n\n\n\nFinancial Structure Log ratio of stock market value\ntraded to bank credit to private\nsector\n\nUrbanization Share of population living in urban\nareas\n\nInformal economy Ratio of informal output relative to\nGDP\n\nManufacturing Share of manufacturing value added\nin GDP\n\nInstitutional Development Average of six principal component\nindicators measuring voice and\naccountability, political stability,\nabsence of corruption, rule of law,\nregulatory quality and government\neffectiveness.\n\nCreditor rights Rights of secured creditors inside\nand outside corporate bankruptcy\n\nCredit information sharing\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Output per head and secondary enrolment are defined from this compilation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003868:30:0", "occurrence_id": "3dcadc63a8d66f59db6d0054", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Appendix Table A2. Variables - definitions and sources**\n\n**<u>Variable</u>** **<u>Definition</u>** **<u>Source</u>**\nBank Credit to GDP Total outstanding claims of deposit Beck, Demirguc-Kunt and Levine\nmoney banks on private sector as (2000)\nratio to GDP\n\n\n\nEnterprise Credit to GDP Total outstanding claims of deposit\nmoney banks on enterprise sector as\nratio to GDP\n\nHousehold Credit to GDP Total outstanding claims of deposit\nmoney banks on households as ratio\nto GDP\n\n\n\nSee Appendix Table A1\n\n\nSee Appendix Table A1\n\n\n\nGDP per capita World Development Indicators\n(WDI)\nSecondary enrolment Share of the respective age cohort WDI\nenrolled in secondary schools\n\n\n\nGovernment consumption Total govt. expenditures relative to\nGDP\n\n\n\nWDI\n\n\n\nTrade Ratio of exports and imports to GDP WDI\nInflation Average log difference in the WDI\nConsumer Price Index over the\nsample period\n\n\n\nLegal origin dummies Origin country of each country’s\nlegal system\n\n\n\nLegal origin dummies Origin country of each country’s La Porta, Lopez-de-Silanes, Shleifer\nlegal system and Vishny (1999)\n\nCatholic, Protestant and Muslim Share of population with the La Porta, Lopez-de-Silanes, Shleifer\npopulation shares respective religious belief in total and Vishny (1999)\n\n\n\nShare of population with the\nrespective religious belief in total\npopulation\n\n\n\nLa Porta, Lopez-de-Silanes, Shleifer\nand Vishny (1999)\n\n\n\nFinancial Structure Log ratio of stock market value\ntraded to bank credit to private\nsector\n\nUrbanization Share of population living in urban\nareas\n\nInformal economy Ratio of informal output relative to\nGDP\n\nManufacturing Share of manufacturing value added\nin GDP\n\nInstitutional Development Average of six principal component\nindicators measuring voice and\naccountability, political stability,\nabsence of corruption, rule of law,\nregulatory quality and government\neffectiveness.\n\nCreditor rights Rights of secured creditors inside\nand outside corporate bankruptcy\n\nCredit information sharing\n\nMention: \"WDI\nConsumer Price Index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Its price series give average inflation over the sample.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003870:8:1", "occurrence_id": "c11d4e4bd5cdeca233ddcbb1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  or revenue side of the equation, when testing for the impact of public\nspending on growth.\n\nFourth, since not all categories of public spending are expected to have the same growth\nimpact, classifying them according to different criteria may be essential to differentiate\nthose items that are truly “productive” from those whose growth impact is negligible.\nHowever, data availability does not always allow for such classification <sup>25</sup> .\n\n\n22 Quantitative techniques such as Public Expenditure Tracking Surveys (PETS) that are being currently\nconducted in many developing countries show that the percentage of spending actually reaching the\npopulation is much lower than what was recorded in the budget as being spent, due to leakages and\ninadequate budget implementation.\n23 This issue was raised by Pritchet in 1996 who reckoned that only about 50% of public investments flows\nin developing countries contribute to creating new capital stock (see Pritchet, 2000). The idea has been\nfurther developed and tested by Hurlin and Arestoff, 2006.\n24 In Barro’s (1990) model, growth increases with expenditures and taxation at low levels, and then\ndecreases as the distortionary effects of taxation exceed the beneficial effects of public spending. In\naddition, at higher levels of public spending inefficiency may increase. That is the main conclusion of\nAfonso, Schuknecht, and Tanzi in reference to new EU members (forthcoming 2008).\n25 While aggregated expenditures are easier to find on an annual basis for a relatively long period of time,\nnot all the sectors report data every year and therefore functional classifications are not always possible. In\naddition, some countries have introduced recent changes in the budget nomenclature, which make sectoral\ncomparisons less meaningful.\n\n\n7\n\nMention: \"Public Expenditure Tracking Surveys\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"These quantitative surveys show how little spending reaches the front line.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003877:9:0", "occurrence_id": "d4d9de8a6b172ce64abfc95f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ##### Ro = ( 1 +δ n )( 1 +δ s ) Re = ( 1 +δ n )( 1 +δ s )( 1 +δ p ), (3)\n\nthat is, a combination of natural barriers, selection, and policy-induced wage\ndiscrimination. We will return to the subject of selection in detail below. It turns out that\nthe effect of this assumption on the estimates is closely related to the assumption of the\n##### independence of φ fh ( x ′ i ) and φ ff ( x ′ i ) .\n\n\n2.3 _Data_\n\nWe estimate equation (2) using unusually rich and standardized collection of individual\nlevel data sets on wage-earners compiled by the World Bank <sup>7</sup> [^7: The sources for all data are given in the appendix. The basic database is also described in Montenegro and Hirn\n(2008).] plus the US Census Public\nUse Microdata Sample (PUMS) five percent file.\n\nA series of steps brings us from the raw collection of data sets to the estimation sample.\nFirst, we remove all self-employed people and unpaid family workers from the data,\nleaving only wage-earners. This has the advantage of increasing the comparability and\naccuracy of the earnings measures, but has the disadvantage of eliminating a large portion\n(though not all) of the informal sector from the sample—especially many agriculturists in\nthe poorest countries. Second, we remove all people aged 14 or less and all people aged\n66 or greater, as well as all people reporting zero wage earnings. Third, we remove data\nfrom twelve transition countries because many of these countries were undergoing\nextraordinary instability of prices, wages, and currencies at the time the survey was\nadministered. Rather than picking and choosing among them we just eliminated the\ntransition countries as a block. <sup>8</sup> Fourth, we randomly delete US-born US-residents from\nthe PUMS to reduce the size of that group from about 6.\n\nMention: \"US Census Public\nUse Microdata Sample\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its five percent file is added to the basic database for estimation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003878:16:0", "occurrence_id": "59e1e17b99a5b1486f4e909b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **_V. Aid for Trade_** <sup>**_4_**</sup> [^4: This Section is based on Newfarmer (2008).]\n\n\nSince the Hong Kong ministerial in December 2005, Doha negotiators have\nendeavored to increase commitments on aid for trade. The purpose was to help\ndeveloping countries: (a) overcome supply side constraints that would otherwise impede\nmany from taking advantage of new Doha-associated market access, (b) implement any\nDoha commitments by providing technical assistance and finance; and (c) provide\nassistance for domestic adjustment. Ministers expressed strong support for the Least\nDeveloped Countries, highlighting the prominent role of the Enhanced Integrated\nFramework to provide technical assistance and catalyze further external support.\n\n\nThough measurement of aid for trade is complex, the WTO/OECD monitoring\nprogram indicates that these efforts are paying some dividends: aid for trade has\nincreased in volume to about $24 billion in 2006. Perhaps more important, the reported\nlevel of trade-related technical assistance has increased. At the same time, the Enhanced\nIntegrated Framework for LDCs is slowly taking shape, and donors have earmarked some\n$240 million in 2007 for its programs of technical assistance.\n\n\nSecond, the Aid for Trade initiative has raised the mutual awareness of the trade\nand development communities of the problems of the other. WTO negotiators have come\nto grasp that they cannot negotiate increases in aid for trade in Geneva because aggregate\nnumeric results are the sum of country-level strategies and of decisions to deploy scarce\ndevelopment assistance resources among competing needs. At the same time, Ministers\nof Finance and development practitioners have demonstrated increasing awareness of the\nimportance of competitiveness to economic growth, and that increased market access or\ntrade liberalization alone is insufficient to deal with binding supply-side constraints that\nimpinge on growth.\n\n\nThe pipeline of activities in the multilateral development banks and donors\nindicate that trade-related assistance may well increase. However, new demands are\nemerging in developing countries associated with the food crisis, and urgent needs for\nsocial programs may crowd out some aid for\n\nMention: \"WTO/OECD monitoring\nprogram\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It tracks the volume of trade-related aid, showing growth to twenty-four billion.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003880:5:0", "occurrence_id": "c56c87e28797e08f1db509c9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 1\n\n\n**Introduction**\n\n\n**Mauritania is rich in natural resources, but its human development is low.** Natural\nresources constitute the backbone of the Mauritanian economy. For decades, iron ore and\nfisheries have contributed to the public coffers. In 2006, the country joined the rank of\noil-exporters. However, both exports and growth have been unable to bring about\nsignificant human development over the years. The 2007 Human Development Index\n(HDI) ranked Mauritania 137 <sup>th</sup> out of a total of 177 countries (UNDP 2007).\n\n**To improve the living standards of the population and sustain broad-based growth,**\n**the authorities need to achieve the Millennium Development Goals (MDGs).** The\ngovernment has laid out an ambitious strategy to propel the country into the middleincome group in the medium-term (GIRM 2007a). <sup>2</sup> [^2: In 2006, the country’s Gross Domestic Product (GDP) was US$2.7 billion, which conferred a per\ncapita Gross National Income (GNI) of US$938 on its 2.9 million people.] To achieve this, Mauritania needs to\ndifferentiate growth beyond natural resources, and ensure that all segments of the\npopulation – 47 percent of which still lives in poverty (GIRM 2006) -- benefits from\ninclusive broad-based growth.\n\n**However, with current policies and resources (both national and external),**\n**Mauritania is unlikely to reach most of the MDGs.** As of today, only gender equality\nin primary and secondary education (MDG3) can be reached, while the poverty reduction\ngoal (MDG1) is at risk. The probability of reaching the other MDGs including the\nprimary school retention rate and all the health-related goals is minimal. The country may\nreach the goal for HIV/AIDS, given its low prevalence, and – given recent progress – the\ngoal for maternal mortality (GIRM 2007).\n\n**\n\nMention: \"2007 Human Development Index\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its ranking of one hundred thirty-seven of one hundred seventy-seven frames the development record.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003885:14:0", "occurrence_id": "506393156e0a5ccc22598240", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: corruption index and developing countries dummy variable are multicolinear, and\n\n\nbecause the institutional quality indices are highly and significant correlated, we do not\n\n\ninclude the developing country dummy in the following regressions.\n\n\nSpecification (5) includes as explanatory variables: HIV prevalence, the rule of\n\n\nlaw index constructed by Kaufmann et al. (2006) and data on ethnic fractionalization\n\n\nfrom Alesina et al. (2003). As expected, both the estimated coefficients for HIV\n\n\nprevalence and ethnic fractionalization present negative signs, although the latter is not\n\n\nstatistically significant, whereas the rule of law index presents a positive and statistically\n\n\nsignificant coefficient. Under this specification, a 1 percentage point increase in HIV\n\n\nprevalence would result in a 0.76 percentage point decrease in trust. Alternatively, a one\n\n\nstandard deviation increase in HIV prevalence will lead to just over 2.5 percentage point\n\n\ndecline in trust. The RESET test rejects the null hypothesis (model is linear) at the 10%\n\n\nlevel but not at 5%, which is acceptable given the small sample size. Including linguistic\n\n\ninstead of ethnic fractionalization does not qualitatively change the results. We also\n\n\ntested the regressions including a quadratic term for ethnic or linguistic fractionalization\n\n\nbecause according to Zak and Knack (2001) in settings with a large number of small\n\n\ngroups, no single group represents much of a threat to others; therefore, the effective\n\n\nsocial distance is greatest at an intermediate range of the fractionalization measure. The\n\nresults were qualitatively similar <sup>11</sup> [^11: Not reported in table1 for ease of exposition, but available upon request], the fractionalization variables were statistically\n\n\ninsignificant but HIV prevalence and the rule of law variable are statistically significant\n\n\nat conventional levels.\n\n\nSpecifications (7) through (11) regress trust on HIV prevalence, rule of law\n\n\nindex, the Gini coefficient for income inequality as a proxy for social distance, the log of\n\n\ninitial income, population density and average age of the population. Rule of law has\n\n\nalways a positive and significant effect on trust except in\n\nMention: \"Alesina et al. (2003)\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its ethnic fractionalization measure enters the regression.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003886:47:0", "occurrence_id": "6594327c3cd1ed778a33338e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **<u>Table 3 Outcomes of Peru’s Antidumping Investigations 1993–June 2004</u>**\n\n|Col1|1993–2000|2001–June 2004|\n|---|---|---|\n|Number of investigations completed|32|25|\n|Number of restrictive measures|11|18|\n|Number closed without restrictive measures|21|4|\n|Restrictive measures as percentage of<br>investigations completed|34|82|\n|Average number of definitive measures per<br>year|1.4|4.5|\n\n\n\n**Source: WPS3658**\n\n\n46\n\nMention: \"WPS3658\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Anti-corruption investigation statistics are reported in this working paper.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003894:16:0", "occurrence_id": "edfed2dd66fed7a8dc76ad11", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: In line with the distinction, labor costs, which are considered another important\n\n\ndeterminant of investment flows, may exert different impacts as well. It is argued that\n\n\nvertical FDI, by taking advantage of comparatively low production costs in the host\n\n\ncountry, shall be negatively related to it. In Bedi and Cieslik (2002), evidence shows a\n\n\nstrong negative correlation between FDI and wage levels in Poland. In case of horizontal\n\n\nFDI, generally a negative sign would be expected as well because cost considerations,\n\n\nalthough they may no longer be the top priority in investors’ decision making, still play a\n\n\nrole. However, this does not exclude the possibility that a positive relationship would\n\n\nshow up in some instances. One would expect a positive sign on the wage rate to reflect\n\n\nthe purchasing power and labor quality of the local market. To control for this factor, we\n\n\nuse the wage rate in manufacturing as a proxy. The International Labor Organization’s\n\n\nLABORSTA provides the data for individual country manufacturing wages in local\n\n\ncurrency. We convert the wage rates to US dollars using official average exchange rates\n\n\ndrawn from World Development Indicators. As for labor quality, we include the illiteracy\n\n\nrate among the host population as the proxy. The illiteracy rate data are from United\n\n\nNations Education, Scientific and Cultural Organization (UNESCO)’s Institute for\n\n\nStatistics and are calculated as the percentage of people aged 15 and above who are\n\n\nilliterate.\n\n\nIn terms of institutional variables, it is widely accepted that the host country’s\n\n\npolitical as well as financial risks are critical to its overall economic well-being. This is\n\n\nmore so in our case because almost all CEECs have experienced political instability and\n\n\neconomic slump at the beginning of the transition. As pointed out in Navaretti and\n\n\nVenables (2004), “political risk and instability seems to be an important deterrent to\n\n\n15\n\nMention: \"LABORSTA\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Manufacturing wages in local currency are taken from this ILO series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003894:16:0", "occurrence_id": "c21c0400cd077ef318800a2c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: In line with the distinction, labor costs, which are considered another important\n\n\ndeterminant of investment flows, may exert different impacts as well. It is argued that\n\n\nvertical FDI, by taking advantage of comparatively low production costs in the host\n\n\ncountry, shall be negatively related to it. In Bedi and Cieslik (2002), evidence shows a\n\n\nstrong negative correlation between FDI and wage levels in Poland. In case of horizontal\n\n\nFDI, generally a negative sign would be expected as well because cost considerations,\n\n\nalthough they may no longer be the top priority in investors’ decision making, still play a\n\n\nrole. However, this does not exclude the possibility that a positive relationship would\n\n\nshow up in some instances. One would expect a positive sign on the wage rate to reflect\n\n\nthe purchasing power and labor quality of the local market. To control for this factor, we\n\n\nuse the wage rate in manufacturing as a proxy. The International Labor Organization’s\n\n\nLABORSTA provides the data for individual country manufacturing wages in local\n\n\ncurrency. We convert the wage rates to US dollars using official average exchange rates\n\n\ndrawn from World Development Indicators. As for labor quality, we include the illiteracy\n\n\nrate among the host population as the proxy. The illiteracy rate data are from United\n\n\nNations Education, Scientific and Cultural Organization (UNESCO)’s Institute for\n\n\nStatistics and are calculated as the percentage of people aged 15 and above who are\n\n\nilliterate.\n\n\nIn terms of institutional variables, it is widely accepted that the host country’s\n\n\npolitical as well as financial risks are critical to its overall economic well-being. This is\n\n\nmore so in our case because almost all CEECs have experienced political instability and\n\n\neconomic slump at the beginning of the transition. As pointed out in Navaretti and\n\n\nVenables (2004), “political risk and instability seems to be an important deterrent to\n\n\n15\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Official average exchange rates here convert wages into dollars.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003898:29:0", "occurrence_id": "dc158f4b85de4ef2c9db61f9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |Col1|2000|Col3|2004|Col5|\n|---|---|---|---|---|\n||**RMB/capita**|**percent**|** RMB/capita**|** Percent**|\n|**County-to-Township Transfers**|**58**||**84**||\n|**Disposable Fiscal Resources**|**119**|**119**|**119**|**119**|\n|Total Extra-budgetary Income|55||53||\n|Total Self-raised Income (incl._tiliu_)|37|37|10|10|\n|_Tongchou_ (township levy)|20|20|0|0|\n|_Tiliu_(village accumulation)|14|14|0|0|\n|Other|4|4|10|10|\n|**Total Fiscal Revenues**|**211**||**182**||\n\n\nSource: fieldwork survey.\n\n\n**3.3. The Village**\n\n\nVillages figure prominently in the NSC as the basic unit of the xiaokang society in the\n\n\ncountryside. In 2005, China had a total of 640,139 administrative villages (China Statistical\n\n\nYearbook 2007). <sup>30</sup> [^30: The number of village committees has declined by about 11.5 percent in the five-year period from 1999 to] A typical “administrative village” <sup>31</sup> [^31: The traditional term “administrative village” is used here, since the formal term “village committee” has a] ( _cun_ ) generally has a population of\n\n\n1,000 to several thousand people, but can be as small as only a few hundred or as large as\n\n\n10,000 people. The average village in our sample survey of 100 had a population of\n\nMention: \"China Statistical\n\n\nYearbook 2007\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"It counts the administrative villages at that date.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003900:35:2", "occurrence_id": "1ef815c40579193f7a193dad", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Denmark**~~|<br>~~**Denmark**~~|\n||||||||\n\n\n15000 17000 19000 21000 23000 25000 27000\n\n\nGDP per capita\n\n\nFigure A2: Plotting vehicle stock fuel use (per 100 km) against income per capita (IEA\n\n\nand OECD data).\n\n\n\n15\n\n\n13\n\n\n11\n\n\n9\n\n\n7\n\n\n5\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n10000 15000 20000 25000 30000\n\n\n**Disposable income (USDppp per capita)**\n\n\nFigure A3: Fuel use of new car sales plotted against income per capita.\n\n\n\n34\n\nMention: \"IEA\n\n\nand OECD data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Vehicle stock, fuel use and income are plotted from these sources.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003907:9:0", "occurrence_id": "44e59a8a3cd5d97dd30f7b9d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: certain countries, generating either missing values or prices drawn from unusual outlets; for\n\nexample, Deaton and Heston (2009) point out that rice is hard to find in Ethiopia and teff (the\n\nstaple in Ethiopia) is hard to find in Thailand, say. One could avoid this problem by choosing\n\nmore representative country-specific bundles, but this would re-introduce the quality bias\n\ndiscussed above, which has plagued past ICP rounds. There is also a problem of “urban bias” in\n\nthe ICP price surveys for some counties; the next section describes our methods of addressing\n\nthis problem. As was argued in Ravallion et al. (1991), a further concern is that the weights\n\nattached to different commodities in the conventional PPP rate may not be appropriate for the\n\npoor; section 6 examines the sensitivity of our results to the use of alternative “PPPs for the\n\npoor” available for a subset of countries from Deaton and Dupriez (2009). Another limitation is\n\nthat the PPP is a national average. Just as the cost of living tends to be lower in poorer countries,\n\none expects it to be lower in poorer regions within one country, especially in rural areas.\n\nRavallion et al. (2007) have allowed for urban-rural cost of living differences facing the poor,\n\nand provided an urban-rural breakdown of our prior global poverty measures using the 1993\n\nPPP. We plan to update these estimates in future work.\n\nGiven that the bulk of the PPPs have risen for developing countries, the poverty count\n\nwill tend to rise at any given poverty line in $PPPs. However, the same changes in the PPPs also\n\nalter the (endogenous) international poverty line, given that it is anchored to the national poverty\n\nlines in the poorest countries. Next we turn to the poverty lines.\n\n###### **3. National and international poverty lines**\n\n\nIn setting an international poverty line using the 2005 ICP we\n\nMention: \"ICP price surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Some country rounds suffer an urban bias that the methods address.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003907:9:0", "occurrence_id": "7880b8291bc40c47c8b3802a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: certain countries, generating either missing values or prices drawn from unusual outlets; for\n\nexample, Deaton and Heston (2009) point out that rice is hard to find in Ethiopia and teff (the\n\nstaple in Ethiopia) is hard to find in Thailand, say. One could avoid this problem by choosing\n\nmore representative country-specific bundles, but this would re-introduce the quality bias\n\ndiscussed above, which has plagued past ICP rounds. There is also a problem of “urban bias” in\n\nthe ICP price surveys for some counties; the next section describes our methods of addressing\n\nthis problem. As was argued in Ravallion et al. (1991), a further concern is that the weights\n\nattached to different commodities in the conventional PPP rate may not be appropriate for the\n\npoor; section 6 examines the sensitivity of our results to the use of alternative “PPPs for the\n\npoor” available for a subset of countries from Deaton and Dupriez (2009). Another limitation is\n\nthat the PPP is a national average. Just as the cost of living tends to be lower in poorer countries,\n\none expects it to be lower in poorer regions within one country, especially in rural areas.\n\nRavallion et al. (2007) have allowed for urban-rural cost of living differences facing the poor,\n\nand provided an urban-rural breakdown of our prior global poverty measures using the 1993\n\nPPP. We plan to update these estimates in future work.\n\nGiven that the bulk of the PPPs have risen for developing countries, the poverty count\n\nwill tend to rise at any given poverty line in $PPPs. However, the same changes in the PPPs also\n\nalter the (endogenous) international poverty line, given that it is anchored to the national poverty\n\nlines in the poorest countries. Next we turn to the poverty lines.\n\n###### **3. National and international poverty lines**\n\n\nIn setting an international poverty line using the 2005 ICP we\n\nMention: \"2005 ICP\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"Its conversion factors set the international poverty line.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003911:4:0", "occurrence_id": "7e2b77e7348fde5b888672c4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: that the most important determinants of the speed at which a country adopts a broad range of\n\n\ntechnologies are its human capital endowment, government type, degree of openness to trade, and its\n\n\nadoption of predecessor technologies. More recently, Parente and Prescott (2005) argue that countries\n\n\nhave widely divergent living standards not because they have access to different stocks of knowledge,\n\n\nbut rather because they differ in the severity of the constraints imposed on the technology choices of\n\n\ntheir citizenry. Drawing on this literature, a recent study by the World Bank (2008) concludes that the\n\n\ndiffusion of international knowledge and technology to a developing country depends on two main\n\n\nfactors: (i) the extent of its access to international knowledge and technology (via trade, FDI, contacts\n\n\nwith its diaspora, and other communications channels) and (ii) its ability to absorb and adapt\n\n\ninternational knowledge and technology. This absorptive capacity is determined by the technological\n\n\nliteracy and skills of the country’s labor force, the availability of financing for innovative firms, sound\n\n\ngovernance, and a good business environment. Moreover, given the various market imperfections\n\n\nassociated with technology and knowledge diffusion, including increasing returns to scale and\n\n\ntechnological spillovers, that study also finds scope for carefully targeted proactive policies aimed\n\n\nspecifically and enhancing technology transfer.\n\n\nEmpirical evidence on the relative impact of various determinants of technology and knowledge\n\n\ndiffusion in developing countries at the micro level is still scarce, in part due to the limited availability\n\n\nof data. This paper aims to contribute to filling this gap by investigating how the “investment climate”,\n\n\nnamely the incentive regime faced by a firm (related to competition, ownership, and governance),\n\n\naccess to complementary inputs (skills, credit, and infrastructure), and access to international\n\n\nknowledge affect firms’ technology choices. Our analysis exploits a unique dataset - the Business\n\n\nEnterprise and Economic Performance Surveys (BEEPS) of firms in the Eastern Europe and Central\n\n\nAsia (ECA) region in 2002 and 2005 - which combines information on the\n\nMention: \"Business\n\n\nEnterprise and Economic Performance Surveys (BEEPS)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Firms across the region in these two rounds supply the technology and knowledge data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003912:5:0", "occurrence_id": "5e78e2b8ffe35180f2dbb0bb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: even bigger increases in worker hires. Nevertheless, one may interpret these findings as saying that more\n\n\nheavily regulated labor markets have more people who would be hurt by the adoption of pro-market\n\n\nreforms.\n\n\nA concrete example may make the above point clearer. Using an index of labor market rigidity that\n\n\ncomes from the Economic Freedom of the World report issued by the Fraser Institute, Chile has the most\n\n\nflexible labor market among the 14 countries studied while Argentina has the least flexible labor market.\n\n\nFirms in Chile report that, if labor regulations were made more flexible, the net gain to employment\n\n\nwould be 0.76% while Argentinian firms report a net gain of 2.82% of total employment. This finding\n\n\nshould not be surprising given the extensive literature that shows that making labor regulations more\n\n\nflexible generates gains to employment.\n\n\nThe focus of the current paper, however, is how these gains in net employment are accomplished.\n\n\nChilean firms report that the additional dismissals that would result from making labor regulations more\n\n\nflexible would be 0.94% of total employment, while Argentinian firms report an additional dismissal figure\n\n\nof 2.34%. We therefore see that, despite the fact that Argentina would gain more in terms of aggregate\n\n\nemployment by reforming its more rigid labor regulations, Argentina would also see a higher percentage\n\n\nof its workers losing their jobs. The larger net employment gains in Argentina arise because their higher\n\n\ndismissal figures are more than compensated for by even higher figures for additional hires. <sup>2</sup> [^2: Specifically, Chilean firms report that making labor regulations more flexible would result in additional hires as a\npercentage of employment of 1.70% while Argentinian firms report a figure of 5.15%. Details on these calculations are\npresented in the data and methodology section.]\n\n\nThe outline of the rest of the paper is as follows. Section 2 briefly reviews the literature on worker\n\n\nand job reallocations and places the current paper into this broader literature. Section 3 describes the\n\n\nessential features of the\n\nMention: \"Economic Freedom of the World report\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its labour market rigidity index ranks flexibility across the studied countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003918:19:1", "occurrence_id": "bbe011f39d5df1fad36d69bc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ,\n\n\nas does GDP and links to former colonies.\n\n\n10 Auxiliary analyses considered other country-level variables, including income growth rates, income\ninequality, foreign-born, exports, unemployment, tax revenue, and estimates of private charitable giving.\nIn some cases the variable was not significant in either survey, e.g., exports measured as a share of GDP.\nOther variables were significant in one survey and not another. For example, the percent foreign born\n(including refugees) was negative and significant in GI and non-significant in the WVS. In some cases,\nresults for these variables were significant but with differing signs in the two surveys. For example, the\naverage growth rate for the preceding 5 years has a positive effect in the GI sample, but negative in the\nWVS sample. Similarly, income inequality (measured by the Gini coefficient) was positive for the GI\ncountries, but negative for the WVS sample. Chong and Gradstein (2006) obtained a similar negative\neffect of inequality using the WVS, but it is contradicted by findings from the larger GI sample.\n\n\n17\n\nMention: \"WVS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Results for the foreign-born share differ in significance from the other survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003920:26:0", "occurrence_id": "d335d0b0b01ae6028ea89c16", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Table 1.** DHS Surveys and Mother/Birth Numbers\n\n|Country|Year(s) of Study|Total Mothers Observed|Total Births Observed|\n|---|---|---|---|\n|Armenia<br>Bangladesh<br>Benin<br>Bolivia<br>Brazil<br>Burkina Faso<br>Burundi<br>Cambodia<br>Cameroon<br>Central African Republic <br>Chad<br>Colombia<br>Comoros<br>Congo (Brazzaville)<br>Dominican Republic<br>Ecuador<br>Egypt<br>Ethiopia<br>Gabon<br>Ghana<br>Guatemala<br>Guinea<br>Haiti<br>Honduras<br>India<br>Indonesia<br>Ivory Coast<br>Kazakhstan<br>Kenya<br>Kyrgyzstan<br>Lesotho<br>Liberia<br>Madagascar<br>Malawi<br>Mali<br>Mexico<br>Moldova<br>Morocco<br>Mozambique<br>Namibia<br>Nepal<br>Nicaragua<br>Nigeria<br>Niger<br>Pakistan<br>Paraguay<br>Peru<br>Philippines<br>Rwanda<br>Senegal<br>South Africa<br>Sri Lanka<br>Sudan<br>Tanzania<br>Thailand<br>Togo<br>Trindad & Tobago<br>Tunisia<br>Turkey<br>Uganda<br>Uzbekistan<br>Vietnam<br>Yemen<br>\n\nMention: \"DHS Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Mothers and births observed by country and study year are listed from these surveys.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003924:31:2", "occurrence_id": "e88e16d11131cac64474b7e8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: </mark> <mark>64.5</mark>\nBonds 32.3 15.3 6.6\n<mark>Units of investment funds</mark> <mark>2.0</mark> <mark>2.6</mark> <mark>0.6</mark>\nOthers 0.4 2.4 0.4\n**<mark>Claims on the Foreign Sector</mark>** <mark>4.8</mark> <mark>9.0</mark> <mark>4.1</mark>\nUnits of mutual funds and shares 3.8 6.6 3.1\n<mark>Debt</mark> <mark>0.9</mark> <mark>2.1</mark> <mark>0.5</mark>\nTime deposits 0.1 0.2 0.5\n**<u><mark>Others</mark></u>** <u><mark>5.9</mark></u> <u><mark>7.0</mark></u> <u><mark>4.3</mark></u>\n**<u>Total assets (LCU million)</u>** <u>2,693.6</u> <u>41,044.8</u> <u>2,311.8</u>\nSource: SBS.\n\n\n50 See, for example, Shah (1997), Vittas (1997, 1998), Queisser (1999), Srinivas and Yermo (1999),\nSrinivas, Whitehouse and Yermo (2000), and Olivares (2005).\n51 In Peru, the legal framework refers to a minimum rate guarantee that is\n\nMention: \"SBS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Bank asset and concentration figures are drawn from this banking survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003925:10:2", "occurrence_id": "36a3017d85a646b799b68bcf", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  section of countries around\nthe world.\n17 The DB score shows the ordinal ranking of the country relative to all other countries where DB data are collected\nand analyzed. As they are ordinal scores, a country could undertake a major reform, yet not see much change it its\nDB score because other countries with similar scores are also reforming. This may be happening in Africa.\n18 For methodological details, see <u>[www.doingbusiness.org.](http://www.doingbusiness.org/)</u>\n19 Note that some dimensions of _Doing Business_, such as poor contract enforcement, show up to the entrepreneur as\nan increased risk, and hence the lender may raise the cost of finance.\n20 This is also found in South Africa by Habyarimana (2008).\n\n\n9\n\nMention: \"DB data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Its ordinal scores rank countries, so reforms may not move the position.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003936:35:2", "occurrence_id": "624a19695c834dcde8b7996e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 62|21.20|12.18|20.48|28.42|\n|Top 10|2.84|0.86|35.62|8.81|14.76|22.93|\n|Total||100.00|||100.00||\n\n\nNotes: Households ranked into decile groups on the basis of predicted per capita household\n\nexpenditure (excluding remittances). Predicted household expenditure is based on\nequation (10) in text. However, data on remittance reception and the amount of\nremittances received is based on observations. Columns (1) and (4) show the\nproportion of households receiving remittances in the given deciles. Columns (2)\nand (5) show the distribution of remittances by deciles. Columns (3) and (6) show\nremittances as a fraction of household expenditure (including remittances).\n\nSource: 2005/06 Ghana GLSS 5 Survey (sub-sample).\n\nMention: \"Ghana GLSS 5 Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"A subsample gives remittances as a share of household expenditure.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003939:23:0", "occurrence_id": "d5c328af9004b8e7aabf3149", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Chart 1: Global Tariffs and Trade, 1985-2006**\n\n\n\n%\n\n\n\n70.0\n\n\n60.0\n\n\n50.0\n\n\n40.0\n\n\n30.0\n\n\n20.0\n\n\n10.0\n\n\n0.0\n\n\n\n\n\n\n\n_1985_ _1986_ _1987_ _1988_ _1989_ _1990_ _1991_ _1992_ _1993_ _1994_ _1995_ _1996_ _1997_ _1998_ _1999_ _2000_ _2001_ _2002_ _2003_ _2004_ _2005_ _2006_\n\n\nSource: World Bank’s World Development Indicators, and WITS database\n\n\n\n80.0\n\n\n75.0\n\n\n70.0\n\n\n65.0\n\n\n60.0\n\n\n55.0\n\n\n50.0\n\n\n22\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The long annual series plotted here come from this compilation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003939:23:0", "occurrence_id": "ef5ed8fd4552f140091b64d2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Chart 1: Global Tariffs and Trade, 1985-2006**\n\n\n\n%\n\n\n\n70.0\n\n\n60.0\n\n\n50.0\n\n\n40.0\n\n\n30.0\n\n\n20.0\n\n\n10.0\n\n\n0.0\n\n\n\n\n\n\n\n_1985_ _1986_ _1987_ _1988_ _1989_ _1990_ _1991_ _1992_ _1993_ _1994_ _1995_ _1996_ _1997_ _1998_ _1999_ _2000_ _2001_ _2002_ _2003_ _2004_ _2005_ _2006_\n\n\nSource: World Bank’s World Development Indicators, and WITS database\n\n\n\n80.0\n\n\n75.0\n\n\n70.0\n\n\n65.0\n\n\n60.0\n\n\n55.0\n\n\n50.0\n\n\n22\n\nMention: \"WITS database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It supplies the companion trade series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003941:22:0", "occurrence_id": "bb27c96a6cc8d13aea1171c5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ### **Figure 1: Labor force participation by race and** **immigration status. Males.**\n\n\n\n0\n\n\n-0.01\n\n\n-0.02\n\n\n-0.03\n\n\n-0.04\n\n\n-0.05\n\n\n-0.06\n\n\n-0.07\n\n\n-0.08\n\n\n-0.09\n\n\n0.06\n\n\n0.04\n\n\n0.02\n\n\n0\n\n\n-0.02\n\n\n-0.04\n\n\n-0.06\n\n\n-0.08\n\n\n-0.1\n\n\n-0.12\n\n\n\nWhites\n\nout\n\n\n\nBlacks in Blacks\n\nout\n\n\n\nAsians in Asians\n\nout\n\n\n\nHispanics\n\nin\n\n\n\nHispanics\n\nout\n\n\n### **Figure 2: Labor force participation by race and** **immigration status. Females.**\n\n\n\nWhites\n\nout\n\n\n\nBlacks in Blacks\n\nout\n\n\n\nAsians in Asians\n\nout\n\n\n\nHispanics\n\nin\n\n\n\nHispanics\n\nout\n\n\n\nData from the 2000 Census IPUMS data (5% sample), population aged 25-65. The figures display the\nmarginal coefficients and their 95% confidence intervals on interaction terms for racial groups and foreign\nborn status (“in” for natives, “out” for foreign born) in probit regressions with labor force participation as\nthe dependent variable. “White natives” is the omitted dummy. The regression results are further detailed in\ntable 2, columns 1 (males) and 5 (females).\n\n\n21\n\nMention: \"IPUMS data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Marginal coefficients on migration flows are estimated on this five percent sample.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003950:26:1", "occurrence_id": "d044bb7960f85c7aa1281ba9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  experiment apply to other settings. In this section we provide suggestive (non\n\nexperimental) evidence that the pattern of returns by gender is not limited to Sri Lanka.\n\n\nWe use microenterprise data from detailed microenterprise surveys in Brazil and\n\n\nMexico. The Brazilian data come from the Economia Informal Urbano survey of 2003\n\n\n(see Fajnzylber et al. 2006) and Mexican data from pooling the 1992, 1994, 1996 and\n\n\n1998 National Survey of Microenterprises (ENAMIN) surveys (see McKenzie and\n\n\nWoodruff, 2006). For each survey we restrict our analysis to firm owners aged 18 to 65,\n\n\nworking more than 30 hours per week, with less than $US500 of capital stock, exclusive\n\n\n                               - - 25\n\nMention: \"Economia Informal Urbano survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The Brazilian microenterprise observations come from this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003950:26:1", "occurrence_id": "c0ae08d5b4f0153677914c10", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  experiment apply to other settings. In this section we provide suggestive (non\n\nexperimental) evidence that the pattern of returns by gender is not limited to Sri Lanka.\n\n\nWe use microenterprise data from detailed microenterprise surveys in Brazil and\n\n\nMexico. The Brazilian data come from the Economia Informal Urbano survey of 2003\n\n\n(see Fajnzylber et al. 2006) and Mexican data from pooling the 1992, 1994, 1996 and\n\n\n1998 National Survey of Microenterprises (ENAMIN) surveys (see McKenzie and\n\n\nWoodruff, 2006). For each survey we restrict our analysis to firm owners aged 18 to 65,\n\n\nworking more than 30 hours per week, with less than $US500 of capital stock, exclusive\n\n\n                               - - 25\n\nMention: \"National Survey of Microenterprises\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Four pooled rounds supply the comparison country's firms.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003951:32:0", "occurrence_id": "ee474d04ddaf79ca1f01b6c0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: influence of the duration of communist rule is insignificant when alternation in included as a\n\n\npredictor in the regression. A tentative conclusion from our results is that the mechanism through\n\n\nwhich years of communism matters is through its effect on political alternation. When we exclude\n\n\nyears of communism from the second stage regression (results not shown here), the estimated\n\n\ncoefficient for alternation increases.\n\n\nThe coefficient for democracy is near zero and is statistically insignificant in two\n\n\nformulations, which is consistent with Treisman’s work using different measures of democracy.\n\n\nHowever democracy significantly reduces the quality of governance in the IV regression. Thus,\n\n\npolitical alternation seems to crowd out the putatively positive effect of democracy.  It is more\n\n\ndifficult to explain why greater democracy, for a given number of alternations in power, would\n\n\nmake governance _worse_ .\n\n\nCumulative duration of war hurts governance:  the variable has a negative sign in all\n\n\nregressions. In the IV regression, each additional year of war reduces the rule of law indicator by\n\n\n12 percent, or one-sixth of a standard deviation of rule of law scores in our sample. <sup>22</sup> [^22: War has a more pronounced effect when the regressions are run up to 2002 or 2003. As wars on the\nterritories of transition countries have ceased, the impact of the variable on governance has somewhat faded.]\n\n\nAs predicted by our model, as well as the literature on the natural resource curse discussed\n\n\nabove, a higher share of fuel exports in GDP worsens governance.  The effect is particularly\n\n\nobvious in IV formulation where each 1 percent increase in fuel exports’ share in GDP reduces the\n\n\npredicted rule of law score by 0.14 points.\n\n\nAs a robustness check, we present in columns 4-6 of Table 6 regressions with another\n\n\nmeasure of governance from the KKM database, “control of corruption.”  This measure is defined\n\n\nas “the exercise of public power for private gain.” This measure is also scaled so that the average\n\n\nand median for the world are zero and the standard deviation is\n\nMention: \"the KKM database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Its control of corruption measure serves as the alternative governance specification.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003953:41:2", "occurrence_id": "7c20c12baeaa3e3d074ca47d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .0</u> <u>100.0</u> <u>100.0</u>\n**Source: APRA Life Office Market Reports**\n\n\nOne factor, which has a bearing on the popularity of lifetime annuities, is the value that\nthey provide to their purchaser as measured by the money’s worth ratios. Knox (2000)\nfound that the majority of money’s worth ratios for Australian annuities are in the range\nof 85-95 percent of the purchase price. He also observed that the results in Australia,\neven though the market is poorly developed, are consistent with major international\nmarkets and that the MWRs compare favorably with recent research for the UK and US\nmarkets. However, in more recent research Ganegoda (2007) reports that when compared\nwith previous calculations for Australian annuities, it appears that the money’s worth of\nAustralian annuities have dropped significantly reflecting high loadings, such as\n\n\n33 Only four companies are currently providing regular quotes on life annuities compared with 11 in 1998\n(Ganegoda 2007)\n\n\n40\n\nMention: \"APRA Life Office Market Reports\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Annuitant shares are compiled in these regulatory reports.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003956:49:0", "occurrence_id": "9894d9f1e03b9d370ec84e78", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |5|Figure 4a: IFLS Expenditure Data – Sample Size 30|Col3|Col4|Col5|\n|---|---|---|---|---|\n|2.5<br>0<br>.5<br>1<br>1.5<br>2<br>Baseline|||||\n|2.5<br>0<br>.5<br>1<br>1.5<br>2<br>Baseline|||||\n|2.5<br>0<br>.5<br>1<br>1.5<br>2<br>Baseline|||||\n|2.5<br>0<br>.5<br>1<br>1.5<br>2<br>Baseline|||||\n|2.5<br>0<br>.5<br>1<br>1.5<br>2<br>Baseline|||||\n|2.5<br>0<br>.5<br>1<br>1.5<br>2<br>Baseline|||||\n|2.5<br>0<br>.5<br>1<br>1.5<br>2<br>Baseline|||||\n|2.5<br>0<br>.5<br>1<br>1.5<br>2<br>Baseline|||||\n|0<br>.5<br>1<br>1.5<br>3 years later<br>-2<br>-1<br>0<br>1<\n\nMention: \"IFLS Expenditure Data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"A thirty-cluster subsample is used for the expenditure distribution plot.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003957:12:0", "occurrence_id": "79fe100eab98289fd0f6bb3b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: - utilizing alternative samples, such as a balanced panel to avoid attrition (see Appendix 2), or\n\n\nan unbalanced panel that includes observations from years 2000 and 2001.\n\n###### **V. CONCLUSIONS**\n\n\nUtilizing Croatia’s corporate tax reform in 2001 as a natural experiment, this study suggests a\n\n\nsignificant effect of corporate taxation on capital structure by demonstrating that lower taxes in\n\n\nCroatia resulted in lower firm leverage. Overall, the corporate tax reform in Croatia provides\n\n\nstrong evidence for a tax effect on firms’ financing policy. These findings are consistent with the\n\n\ntrade-off theory of capital structure, which suggests that lower taxes decrease the incentive to\n\n\nhold debt due to decreasing interest tax deductibility. Notably, the current study focuses on the\n\n\nsign and the significance of the effect from tax reform without making any further claims on the\n\n\nmagnitude of the change in debt/equity ratios, because the firm- and time-varying adjustment\n\n\nspeed of capital structures complicates the exact identification and economic interpretation of a\n\n\nmagnitude (Nivorozhkin, 2005).\n\n\nFrom a policy perspective, it could be argued that the reduction in corporate debt after the\n\n\ncorporate tax reform could have impacted firms’ access to finance. Just like other transition\n\n\neconomies, Croatia has a bank-based financial system where low levels of debt financing are\n\n\ntypically a supply-side phenomenon given the preferential bank financing towards state-owned\n\n\nenterprises and large firms. By reducing the optimal debt level, the tax reform may have \n\nindirectly- enhanced SMEs’ access to finance. This was important since access to finance was\n\n\none of the major obstacles for Croatian firms. More specifically, based on surveys from the\n\n\nEuropean Bank of Reconstruction and Development (EBRD), the portion of private SMEs in\n\n\nCroatia that reported obtaining financing from banks as a major obstacle in their operations was\n\nreduced from 48% in 1999 to 28% in 2005. <sup>16</sup> Besides access to finance, the observed post\n\nMention: \"European Bank of Reconstruction and Development (EBRD)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its survey records how many small firms cite finance as a major obstacle.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003966:17:0", "occurrence_id": "5169c9f5f9460532feed478f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: and political data from the World Bank’s _Database of Political Institutions_ (DPI). The fiscal variables\n\n\nmeasure total spending (excluding interest payments), total tax revenues, and fiscal deficits of central\n\ngovernments. <sup>20</sup>\n\n\nWe measure coalition governments in three different ways using data from the DPI, given the\n\n\nfinding that coefficient estimates are sensitive to the way coalitions are defined (Perotti and\n\n\nKontoupoulos, 2002). The simplest measure is that of an indicator variable which equals 1 if the party of\n\n\nthe executive leader in a country does not control a majority in all lawmaking houses, which we label\n\n“DIVIDED”. <sup>21</sup> The second measure is the probability (which lies between 0 and 1) that two legislators\n\n\nchosen at random from among parties that form the government will belong to two different parties,\n\n\nwhich is the variable labeled “GOVFRAC” in the DPI. The third measure is the number of political\n\n\nparties forming the government, similar to the measure used by Perotti and Kontoupoulos (2002), which\n\n\nwe calculate by simply counting the number of parties listed as government parties in the DPI—\n\n\n“NUMBER OF PARTIES”.\n\nWe compile available data for more than 70 countries over the period 1970 – 2006. <sup>22</sup> We specify\n\n\nthe time interval over which the variables are measured, _t_, as a decade. We calculate decadal averages for\n\n\neach country in our sample, for the 1970s (1970-1979), 1980s (1980-1989), 1990s (1990-1999), and\n\n\n2000s (2000-2006), using non-missing values within a decade. We analyze the correlation of political\n\n\nparty composition of government with fiscal policy over this country-decade sample. However, there is a\n\n\nlot of missing fiscal data, with observations on\n\nMention: \"data from the DPI\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Coalition governments are defined in three ways from this dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003968:23:0", "occurrence_id": "466c41e66db9a2404b5c4033", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: #### Annex 2. Lane Financial Long Term Index of Catastrophe Reinsurance prices\n\n**Lane Financial Long Term Index of Catastrophe Reinsurance**\n\n**Prices**\n\n\n**Ten Year Index** **Ten Year Index**\n12/31/1996 1.00 3/31/2003 1.05\n3/31/1997 0.97 6/30/2003 1.07\n6/30/1997 1.01 9/30/2003 1.03\n9/30/1997 0.93 <u>12/31/2003</u> 0.99\n<u>12/31/1997</u> 0.87 3/31/2004 0.92\n3/31/1998 0.85 6/30/2004 0.96\n6/30/1998 0.89 9/30/2004 0.92\n9/30/1998 0.84 <u>12/31/2004</u> 0.91\n<u>12/31/1998</u> 0.79 3/31/2005 0.87\n3/31/1999 0.80 6/30/2005 0.90\n6/30/1999 0.80 9/30/2005 0.86\n9/30/1999 0.79 <u>12/31/2005</u> 1.02\n<u>12/31/1999</u> 0.77 3/31/2006 1.08\n3/31/2000 0.76 6/30/2006 1.65\n6/30/2000 0.77 9/30/2006 1.57\n9/30/2000 0.76 <u>12/31/2006</u> 1.46\n<u>12/31/2000</u> 0.76 3/31/2007 1.30\n3/31/2001 0.76\n\nMention: \"Lane Financial Long Term Index of Catastrophe Reinsurance prices\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The ten-year price series tabulated in the annex is this index.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003968:23:0", "occurrence_id": "65efad66744059a14eaf0a3b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: #### Annex 2. Lane Financial Long Term Index of Catastrophe Reinsurance prices\n\n**Lane Financial Long Term Index of Catastrophe Reinsurance**\n\n**Prices**\n\n\n**Ten Year Index** **Ten Year Index**\n12/31/1996 1.00 3/31/2003 1.05\n3/31/1997 0.97 6/30/2003 1.07\n6/30/1997 1.01 9/30/2003 1.03\n9/30/1997 0.93 <u>12/31/2003</u> 0.99\n<u>12/31/1997</u> 0.87 3/31/2004 0.92\n3/31/1998 0.85 6/30/2004 0.96\n6/30/1998 0.89 9/30/2004 0.92\n9/30/1998 0.84 <u>12/31/2004</u> 0.91\n<u>12/31/1998</u> 0.79 3/31/2005 0.87\n3/31/1999 0.80 6/30/2005 0.90\n6/30/1999 0.80 9/30/2005 0.86\n9/30/1999 0.79 <u>12/31/2005</u> 1.02\n<u>12/31/1999</u> 0.77 3/31/2006 1.08\n3/31/2000 0.76 6/30/2006 1.65\n6/30/2000 0.77 9/30/2006 1.57\n9/30/2000 0.76 <u>12/31/2006</u> 1.46\n<u>12/31/2000</u> 0.76 3/31/2007 1.30\n3/31/2001 0.76\n\nMention: \"Lane Financial Long Term Index of Catastrophe Reinsurance\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Quarterly index values begin at unity in that year.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003969:16:0", "occurrence_id": "4e86e0a1a8b4dcacd8533648", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The main explanatory variable of interest, as mentioned, is the natural logarithm of\nthe absolute value of net exports of natural resources per worker. The trade data come from\nUNCOMTRADE, as downloaded from the World Bank’s data server. The commodity\ngroups of products classified as being intensive in the use of natural resources are Leamer’s\n(1984, 1995) non-manufacturing commodities – see Table A1.\n\nRegarding the conditioning variables, we follow the strategy in Lederman and\nMaloney (2007) by examining how changes in the specification of the model affect the\ncoefficient on the natural-resource variables. The initial multi-variate regression model\nspecification includes two variables in addition to the natural-resource proxies: (a) the share\nof years during 1980-1999 when the dummy variable for policy reform was observed, as\nper the data in Wacziarg and Welch (2002) and Sachs and Warner (1995); <sup>14</sup> [^14: The policy reform index is one if import tariffs are below 40%, if there are not significant non-tariff\nbarriers, if governments have privatized a significant share of public enterprises, and if the foreign-exchange\nblack market premium is below a certain threshold – see the Data Appendix for details.] (b) and the\naverage annual growth of the terms of trade during 1980-2005, which comes from the\nWorld Bank’s World Development Indicators data, as downloaded in early 2008.\n\nThe augmented model includes the executive constraints variable, which is the\ninstitutional variable used in Acemoglu and Robinson (2002, 2005) and Glaeser et al.\n(2004). The data come from the Polity IV database, and has a value that ranges between 1\nand 7, with higher values representing less executive-branch discretion. The econometric\nmodels include the natural logarithm of this variable, which makes it easier to interpret\ncoefficient estimates as elasticities. The literature argues that constraints on the Executive\nshould be negatively partially correlated, in the first stage of 2SLS estimations, with\nhistorical population density. <sup>15</sup> Due\n\nMention: \"UNCOMTRADE\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Net resource exports per worker are computed from these trade flows.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003969:16:0", "occurrence_id": "6c37a53b8abd536dc0c7a585", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The main explanatory variable of interest, as mentioned, is the natural logarithm of\nthe absolute value of net exports of natural resources per worker. The trade data come from\nUNCOMTRADE, as downloaded from the World Bank’s data server. The commodity\ngroups of products classified as being intensive in the use of natural resources are Leamer’s\n(1984, 1995) non-manufacturing commodities – see Table A1.\n\nRegarding the conditioning variables, we follow the strategy in Lederman and\nMaloney (2007) by examining how changes in the specification of the model affect the\ncoefficient on the natural-resource variables. The initial multi-variate regression model\nspecification includes two variables in addition to the natural-resource proxies: (a) the share\nof years during 1980-1999 when the dummy variable for policy reform was observed, as\nper the data in Wacziarg and Welch (2002) and Sachs and Warner (1995); <sup>14</sup> [^14: The policy reform index is one if import tariffs are below 40%, if there are not significant non-tariff\nbarriers, if governments have privatized a significant share of public enterprises, and if the foreign-exchange\nblack market premium is below a certain threshold – see the Data Appendix for details.] (b) and the\naverage annual growth of the terms of trade during 1980-2005, which comes from the\nWorld Bank’s World Development Indicators data, as downloaded in early 2008.\n\nThe augmented model includes the executive constraints variable, which is the\ninstitutional variable used in Acemoglu and Robinson (2002, 2005) and Glaeser et al.\n(2004). The data come from the Polity IV database, and has a value that ranges between 1\nand 7, with higher values representing less executive-branch discretion. The econometric\nmodels include the natural logarithm of this variable, which makes it easier to interpret\ncoefficient estimates as elasticities. The literature argues that constraints on the Executive\nshould be negatively partially correlated, in the first stage of 2SLS estimations, with\nhistorical population density. <sup>15</sup> Due\n\nMention: \"World Development Indicators data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Terms of trade growth over this window comes from this source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003969:16:0", "occurrence_id": "d1c835b11bb55c8d46c1136b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The main explanatory variable of interest, as mentioned, is the natural logarithm of\nthe absolute value of net exports of natural resources per worker. The trade data come from\nUNCOMTRADE, as downloaded from the World Bank’s data server. The commodity\ngroups of products classified as being intensive in the use of natural resources are Leamer’s\n(1984, 1995) non-manufacturing commodities – see Table A1.\n\nRegarding the conditioning variables, we follow the strategy in Lederman and\nMaloney (2007) by examining how changes in the specification of the model affect the\ncoefficient on the natural-resource variables. The initial multi-variate regression model\nspecification includes two variables in addition to the natural-resource proxies: (a) the share\nof years during 1980-1999 when the dummy variable for policy reform was observed, as\nper the data in Wacziarg and Welch (2002) and Sachs and Warner (1995); <sup>14</sup> [^14: The policy reform index is one if import tariffs are below 40%, if there are not significant non-tariff\nbarriers, if governments have privatized a significant share of public enterprises, and if the foreign-exchange\nblack market premium is below a certain threshold – see the Data Appendix for details.] (b) and the\naverage annual growth of the terms of trade during 1980-2005, which comes from the\nWorld Bank’s World Development Indicators data, as downloaded in early 2008.\n\nThe augmented model includes the executive constraints variable, which is the\ninstitutional variable used in Acemoglu and Robinson (2002, 2005) and Glaeser et al.\n(2004). The data come from the Polity IV database, and has a value that ranges between 1\nand 7, with higher values representing less executive-branch discretion. The econometric\nmodels include the natural logarithm of this variable, which makes it easier to interpret\ncoefficient estimates as elasticities. The literature argues that constraints on the Executive\nshould be negatively partially correlated, in the first stage of 2SLS estimations, with\nhistorical population density. <sup>15</sup> Due\n\nMention: \"Polity IV database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its executive constraints scale runs from one to seven.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003977:28:0", "occurrence_id": "083cbdef203eb5b8892f180b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Annex 2. Migration and Demographic Patterns**\n\n\nBeyond its direct effects on the size of the population, migration has long lasting\n\n\nindirect effects on the demographic dynamics, through changes in the age and gender\n\n\nstructure that impact birth and mortality rates.\n\n\nTheir magnitude depends both on the demographic structure of migrants (in and\n\n\nout migration, sex and age composition but also types of migration opposing permanent\n\n\nto temporary) and on the size of net migration flows in relation to the intensity of the\n\n\ndomestic demographic dynamics.\n\n\nA simple demographic model was thus developed to account for these effects. In a\n\n\nlong term perspective, the effects of migration add to (and interfere with) the\n\n\ndemographic transition. Two main parameters have been retained to reflect such an\n\n\ninteraction: the position of the “demographic cursor” with respect to the demographic\n\n\ntransition and the intensity of migration.\n\n\nThese two variables are, to a large extent, independent: the demographic cursor\n\n\ncommands fertility and mortality rates for the concerned age and gender categories, while\n\n\nmigration affects the percentage of individuals leaving (on a net basis) the country within\n\n\neach age and gender category. The age and gender characteristics retained for migrants\n\n\nfocused on young, predominantly male migrants and the parameter “migration” was\n\n\ntaken as a simple percentage of the basic pattern. The pace of the demographic transition\n\n\nwas aligned with the median scenario of the UN population statistics database and values\n\n\nrange from 0 to 1. The initial and final age structures for the population used in the model\n\n\nare presented below:\n\n\n**Demographic Patterns in Different Simulations**\n\n\n\n_Baseline Simulations_\n\n\n\n_Lower Dependency_\n\n\n\n_<u>Ratios</u>_\nEmigration rate for the workforce 1.00% 1.00%\nDemographic transition cursor (0% to 100%) 10% 75%\nTotal emigration rate 0.49% 0.56%\n\nNatural rate of population growth 2.00% 0.26%\nPost migration rate of population growth 1.51% -0\n\nMention: \"the UN population statistics database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its median scenario aligns the modelled demographic transition.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003982:28:1", "occurrence_id": "9a935b2de2da34d9a4a21143", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Undersecretariat of Foreign Trade (UFT Website); TurkSTAT 2002 Household consumption\n\n\nexpenditures and income survey statistics (TurkSTAT Website), ARIP Quantitative Household\n\n\nSurvey (QHS) commissioned by the Treasury and implemented by the G.G. Consulting et al.\n\n\n(conducted in 2002 and 2004), and the Turkish SAM for 2001 developed by GTAP. This\n\n\nregionalized SAM includes information on the regional level employment of labor and capital as\n\n\nwell as intermediate input use by crop, along with water charge and irrigated versus rainfed land\n\n\nrents by crop. Employment of capital, labor and data on intermediate input use are also entered\n\n\nfor activities at the national level in the SAM. Detailed income, consumption and saving\n\n\ninformation for the five-types of households (urban, and four rural) are included in the\n\n\ninstitutions block of the SAM. Also included in the institutions block are the data on government\n\n\nconsumption, net tax receipts, and public savings, along with the WUAs’ accounts. Finally,\n\n\nimport, export and tariff data concerning the EU-25 and the rest of the world constitute the\n\n\n27\n\nMention: \"TurkSTAT 2002 Household consumption\n\n\nexpenditures and income survey statistics\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Consumption and income aggregates are published from this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003983:4:1", "occurrence_id": "04f3a49cc030976c77c54485", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  on “trade” in services would ideally focus on foreign professionals working temporarily\nin the United States – reflecting the fact that multilateral and regional trade agreements treat traderelated labor mobility as distinct from immigration. There is, unfortunately, no data on the number of\nforeign professionals in the United States on temporary stay visas, but from the US census data it is\npossible to calculate the number of “foreign born” professionals. <sup>3</sup> [^3: Data is available on the number of individuals entering the United States on specialty occupation (H1B) visas,\nbut this data records the number of entries so that an individual may be counted more than once depending on\nthe number of times he or she leaves and returns to the United States.] We use this latter data to illustrate\nthe significant presence of foreign professionals.\n\n\nForeign professionals in the five services considered here, namely, accountants and auditors,\narchitects, engineers, physicians and surgeons, and lawyers, made up 17 percent of the total\nprofessionals of the US economy—which is smaller than the share of foreign goods and foreign\ncapital in the US economy, but higher than the share of foreign professionals in most other countries.\nIn absolute numbers, these five professions accounted for nearly 5.1 million jobs in 2000, of which,\nnearly 4.24 million were held by US-born professionals and the remaining 0.86 million by foreignborn professionals (many of whom have subsequently become US citizens).\n\n\nThe share of foreign professionals varies considerably across different professions, with\nprofessions that are less regulated and more intensive in science and technology-subjects tending to\nhave a larger foreign presence (Figure 1a). Foreign presence is the highest in the field of computer\nsoftware and medicine, with foreign computer software engineers and physicians and surgeons\naccounting for 29 and 27 percent of the total workforce in their respective fields. At the other extreme\nis the legal profession, where foreign-born lawyers account for only 6 percent of the total workforce.\n\n\n2 One exception is the excellent\n\nMention: \"US census data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Counts of foreign-born professionals are derived from it since visa data are absent.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:003989:4:0", "occurrence_id": "8fc93f058ca25c98aa98c65b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **_2.1 Trade Policy and the Booming_** **Maquila** **_Sector_**\n\n\nIn 1990, Honduras began the implementation of pro-trade reforms by unilaterally reducing tariffs\nand in 1994 it joined the General Agreement on Tariffs and Trade (GATT). The multilateral\nagreement became the base for Honduras’ trade policy granting, at least, Most Favored Nation\n(MFN) treatment to all its trading partners. Honduras is an active member of the Central\nAmerican Common Market. It has signed about a dozen bilateral investment treaties and free\ntrade agreements with countries including Canada, Colombia, Chile, the Dominican Republic,\nMexico, Panama, Switzerland, Taiwan (China), and the United States. Trade-oriented policies\ncontinue to be at the center of the development agenda in Honduras, which is participating in the\nnegotiation of a trade agreement between the European Union and Central America.\n\n\n**Figure 1 Trade Openness in Honduras, 1990–2006**\n\n\n_Source_ : Authors, based on data from the Banco Central de Honduras and Secretaría de Finanzas.\n\n\nAs a result of trade policy, the Honduran economy is developing into a more open and\nliberalized economy. Its tariff structure is low and more uniform than it used to be, and the\napplication of nontariff measures is very limited (figure 1). The simple average of implicit tariff\nrates decreased from more than 16 percent in 1991–92 to about 3.3 percent in 2005–06. <sup>3</sup> [^3: The average implicit tariff was calculated as import tariff revenue/imports fob, excluding oil.] In 2002,\nafter the damaging effects of Hurricane Mitch, international trade supported a rapid recovery led\nby the exports of _maquila_ manufacturing and agroprocessing industries.\n\n\nBetween 1999 and 2006, exports and domestic markets became more diversified and\nemployment and investment (domestic and foreign) grew, particularly in some nontraditional\nexport activities. The United States remains the country’s principal trading partner,\n\nMention: \"the Banco Central de Honduras\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Trade openness over this period is plotted from its records.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004003:30:0", "occurrence_id": "409b81fd00378c0d11ac105b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Second, asset price bubbles are not easy to spot in real time – at least at their early\n\n\nstages, when policy action might be most useful to stop them. Identifying a misalignment\n\n\nof asset prices requires a reliable assessment of their fundamentals. But available models\n\n\nof asset price fundamentals are inherently imperfect, and in addition some fundamentals\n\n\n- e.g., investors’ perceptions of risk and future asset productivity --- are not readily\n\n\nobservable. A simpler alternative would be to use of rules of thumb, such as a threshold\n\n\nabove which asset price increases would be deemed to involve bubbles. However, aside\n\n\nfrom their arbitrary nature, <sup>29</sup> [^29: How fast should asset prices rise to safely conclude that a bubble is at work? Too low a threshold would\nlead to the (incorrect) detection of too many bubbles. Even modest changes in key fundamentals – such as\ninvestors discount rates or their anticipations of the rate of growth of future dividends – can lead to large\nchanges in the prices of long-lived assets over short periods. In turn, too high a threshold (say 20% or 40%\nper year) could fail to detect many bubbles altogether. For example, during the recent U.S. housing bubble\nreal estate price increases remained consistenly below 15% per annum, according to the Case-Shiller U.S.\nnational home price index.] simple rules of thumb would likely lead to frequent\n\n\nidentification errors, especially in narrow and illiquid asset markets (such as those of\n\n\nmost emerging countries) characterized by high asset price volatility. And reacting to a\n\n\nmisidentified bubble may be very costly – e.g., tightening in response to an asset price\n\n\nrise that in reality is driven by improving fundamentals will both hamper growth and\n\n\ninterfere with the role of asset prices in allocating resources. Of course, one possible\n\n\nsolution to this problem is to react only to bubbles once they become self-evident, but by\n\n\nthen it may be too late for monetary policy to mitigate\n\nMention: \"the Case-Shiller U.S.\nnational home price index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its record shows annual price rises stayed below fifteen percent.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004008:39:0", "occurrence_id": "7873458c18b658bea5108895", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |Boys<br>-1.3<br>-1.4<br>z-score<br>-1.6<br>-1.8 Height-by-age<br>-2<br>-2.2<br>-2.5<br>J F M A M J J A S O N D|Girls<br>-1.3<br>-1.4<br>95% confidence interval<br>-1.6<br>-1.8<br>-2<br>-2.2 1992<br>1998<br>-2.5 2005<br>J F M A M J J A S O N D|\n|---|---|\n|-2.3<br>-2<br>-1.8<br>-1.6<br>-1.5<br>Weight-by-age z-score<br>J<br>F<br>M<br>A<br>M<br>J<br>J<br>A<br>S<br>O<br>N<br>D<br>Month of birth<br>Boys|-2.3<br>-2<br>-1.8<br>-1.6<br>-1.5<br>J<br>F<br>M<br>A<br>M<br>J<br>J<br>A<br>S<br>O<br>N<br>D<br>Month of birth<br>Girls|\n\n\n\n\n\n\n\n\n\n\n\nFigure 4: Health outcomes (WAZ and HAZ) by month of birth and gender.\n\nIndia NFHS 1992, 1998 and 2005\n\n\n\n\n\n\n|95% confidence interval\n\nMention: \"India NFHS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Child nutrition indices by birth month and gender are plotted across these rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004009:16:0", "occurrence_id": "e031dc63ad1dd3c79fdc024c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: When examining the female sample and male sample separately, it is found that the earnings\n\n\npremium for SMK graduates is only significant for the male sample. However, after using\n\nHeckman’s 2-step procedure to correct the sample selection bias, this male SMK wage\n\n\npremium becomes insignificant (Table 5).\n\n\n**Table 5: Regression of EARNINGS (in IDR) in 2000 (IFLS3, n=252, 515, 128, 258,**\n\n**124 & 257), with and without censoring correction**\n\n\nAll Female Male\n\n\n\nHeckman\n\n2-step\n\nestimates\n\n\n\nOLS\n\ncoefficient\n\nestimates\n\n\n\nHeckman\n\n2-step\n\nestimates\n\n\n\nSelection\n\nequation\n\nestimates\n\n\n\nHeckman\n\n2-step\n\nestimates\n\n\n\nIndependent\n\nvariable\n\n\n\nOLS\n\ncoefficient\n\nestimates\n\n\n\nRURAL -549,218* -558,768* -614,066* -635,735* -479,310 -370,741\n\n(285,538) (321,134) (355,545) (376,951) (458,486) (555,599)\n\n\nFEMALE 144,960 143,284\n\n(271,750) (269,855)\n\nAGE 378,989.*** 375,070*** 552,753*** 541,741*** 204,897 238,110\n\n(105,758) (121,991) (132,641) (149,049) (166,637) (191,663)\n\n\nSMK 576,895** 564,774* 371,828 353,463. 772,357* 931,816\n\n(273,775) (333,286) (341,856) (356,563) (439,088) (643,329)\n\nEBTANAS1 52,526** 53,763* 77,296** 80,018** 20,809 8,156\n\n(25,214) (31,871) (29,836) (34,\n\nMention: \"IFLS3\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Earnings regressions with censoring correction are estimated on this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004013:18:0", "occurrence_id": "7eb63e0ba721031d7956b292", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: literacy. Life expectancy is measured in years and literacy is an index value ranging\n\n\nfrom 0 to 100. We choose education and health as both are direct determinants of\n\n\ncapabilities and influence the freedom to choose the kind of life one likes. Education\n\n\nas well as health raises productivity and the ability to convert income and resources\n\n\ninto the favored way of life (Sen 2003: 55). The third dimension of human develop\n\nment, namely income, is not of interest for this paper, since there is much work on the\n\n\nrelation between democracy and economic development already available. Our data\n\n\non political institutions, especially democracy comes from the Polity IV Project of the\n\n\nCenter for International Development and Conflict Management at the University of\n\n\nMaryland. This dataset includes the Polity2 score ranging from 10 (highly democratic)\n\n\nto -10 (highly autocratic), while a zero score indicates a state between autocracy and\n\ndemocracy which we consider as not democratic. <sup>13</sup> Following Besley and Kudamatsu\n\n\n(2006) we calculate the fraction of democratic years over the past five years as our mea\n\nsure for democracy. The consideration of a period of five years captures the effect of\n\n\ndemocratic experience and reduces the uncertainty concerning the length of the delay\n\n\nuntil a change in the political system affects human development.\n\n\nOther variables that we expect to have an impact on human development or that\n\n\ndescribe possible conditions under which democracy affects human development are the\n\nfollowing: GDP per capita PPP in constant prices <sup>14</sup> from the Penn World Tables 6.2;\n\nGini coefficients <sup>15</sup> from the WIDER dataset with improvements in terms of compara\n\n13 According to the Polity2 measure, a system can be classified as democratic if three interdependent\nelements exists: 1) competitiveness of participation, institutions and procedures allow citizens to\nexpress their political preferences; 2) openness and competitiveness of executive recruitment and\nconstraints on the chief executive, so\n\nMention: \"Polity IV Project\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its democracy measure is the main political institution variable.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004013:18:0", "occurrence_id": "d9c0390210c70b24185576bd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: literacy. Life expectancy is measured in years and literacy is an index value ranging\n\n\nfrom 0 to 100. We choose education and health as both are direct determinants of\n\n\ncapabilities and influence the freedom to choose the kind of life one likes. Education\n\n\nas well as health raises productivity and the ability to convert income and resources\n\n\ninto the favored way of life (Sen 2003: 55). The third dimension of human develop\n\nment, namely income, is not of interest for this paper, since there is much work on the\n\n\nrelation between democracy and economic development already available. Our data\n\n\non political institutions, especially democracy comes from the Polity IV Project of the\n\n\nCenter for International Development and Conflict Management at the University of\n\n\nMaryland. This dataset includes the Polity2 score ranging from 10 (highly democratic)\n\n\nto -10 (highly autocratic), while a zero score indicates a state between autocracy and\n\ndemocracy which we consider as not democratic. <sup>13</sup> Following Besley and Kudamatsu\n\n\n(2006) we calculate the fraction of democratic years over the past five years as our mea\n\nsure for democracy. The consideration of a period of five years captures the effect of\n\n\ndemocratic experience and reduces the uncertainty concerning the length of the delay\n\n\nuntil a change in the political system affects human development.\n\n\nOther variables that we expect to have an impact on human development or that\n\n\ndescribe possible conditions under which democracy affects human development are the\n\nfollowing: GDP per capita PPP in constant prices <sup>14</sup> from the Penn World Tables 6.2;\n\nGini coefficients <sup>15</sup> from the WIDER dataset with improvements in terms of compara\n\n13 According to the Polity2 measure, a system can be classified as democratic if three interdependent\nelements exists: 1) competitiveness of participation, institutions and procedures allow citizens to\nexpress their political preferences; 2) openness and competitiveness of executive recruitment and\nconstraints on the chief executive, so\n\nMention: \"Penn World Tables\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Output per head in constant prices comes from this version.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004013:18:0", "occurrence_id": "dc300307ac96770cfaa0bbb9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: literacy. Life expectancy is measured in years and literacy is an index value ranging\n\n\nfrom 0 to 100. We choose education and health as both are direct determinants of\n\n\ncapabilities and influence the freedom to choose the kind of life one likes. Education\n\n\nas well as health raises productivity and the ability to convert income and resources\n\n\ninto the favored way of life (Sen 2003: 55). The third dimension of human develop\n\nment, namely income, is not of interest for this paper, since there is much work on the\n\n\nrelation between democracy and economic development already available. Our data\n\n\non political institutions, especially democracy comes from the Polity IV Project of the\n\n\nCenter for International Development and Conflict Management at the University of\n\n\nMaryland. This dataset includes the Polity2 score ranging from 10 (highly democratic)\n\n\nto -10 (highly autocratic), while a zero score indicates a state between autocracy and\n\ndemocracy which we consider as not democratic. <sup>13</sup> Following Besley and Kudamatsu\n\n\n(2006) we calculate the fraction of democratic years over the past five years as our mea\n\nsure for democracy. The consideration of a period of five years captures the effect of\n\n\ndemocratic experience and reduces the uncertainty concerning the length of the delay\n\n\nuntil a change in the political system affects human development.\n\n\nOther variables that we expect to have an impact on human development or that\n\n\ndescribe possible conditions under which democracy affects human development are the\n\nfollowing: GDP per capita PPP in constant prices <sup>14</sup> from the Penn World Tables 6.2;\n\nGini coefficients <sup>15</sup> from the WIDER dataset with improvements in terms of compara\n\n13 According to the Polity2 measure, a system can be classified as democratic if three interdependent\nelements exists: 1) competitiveness of participation, institutions and procedures allow citizens to\nexpress their political preferences; 2) openness and competitiveness of executive recruitment and\nconstraints on the chief executive, so\n\nMention: \"WIDER dataset\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its inequality coefficients are used with comparability improvements.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004016:32:0", "occurrence_id": "8456d0e619ef54fdd71e987a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 31\n\n\nTable 1: Health financing, provider payment methods, and existence of gatekeeper—late 2000s\n\n\n\nSHI/tax<u>finance</u>\n\n\n\nPrimary care providers Hospitals Gatekeeper\n\n\n\nAustria SHI Capitation + FFS DRGs + FFS (outpatient) No\nBelgium SHI FFS Budgets + FFS No\nCzech Republic SHI Capitation + FFS (previously FFS) Budgets (previously FFS) No\nFrance SHI Capitation + FFS Global budgets No\nGermany SHI Capitation DRGs No\nGreece SHI FFS Per diems No\nHungary SHI Capitation DRGs No\nJapan SHI FFS FFS No\nKorea SHI FFS FFS No\nLuxembourg SHI FFS Global budgets No\nMexico SHI Budgets Budgets No\nNetherlands SHI Capitation DRGs Yes\nPoland SHI Capitation DRGs No\nSlovak Republic SHI Capitation + FFS (previously FFS) Budgets (previously FFS) Partial\nSwitzerland SHI FFS Global budgets + per diems No\n<u>Turkey</u> <u>SHI</u> <u>Salary</u> <u>Global budgets + variations</u> <u>No</u>\nAustralia Tax FFS Global budgets Yes\nCanada Tax FFS Global budgets Yes\nDenmark Tax Capitation + FFS Global budgets Yes\nFinland Tax Salary, capitation + FFS FFS No\nIceland Tax Salary Global budgets No\nIreland Tax Capitation Global budgets Yes\nItaly Tax Capitation + FFS DRGs Yes\nNew Zealand Tax Capitation + FFS Global budgets Yes\nNorway Tax Salary, capitation + FFS DRGs Yes\nPortugal Tax Budgets Global budgets Yes\nSpain Tax Salary DRGs Yes\nSweden Tax Salary Global budgets + variations Yes\n<u>United Kingdom</u> <u>Tax</u> <u>Capitation + allowances + FFS</u> <u>DRGs (previously budgets)</u> <u>Yes</u>\nSources: WHO Health in Transition series (HiTS) <u>http://www.\n\nMention: \"WHO Health in Transition series (HiTS)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Payment arrangements by country are tabulated from this series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004020:19:1", "occurrence_id": "547aadc62b04103d77a24fa4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  account of the reported cases from the innetwork physicians and hospitals to the NHIF. Out-of-network actors do not have to\nreport their cases, the overall number therefore being unknown.\n\n**Table 8: Treatment Outside the Framework of EC Regulation 1408/71, 2005**\n\n\n\n\n\n\n\n|Treatment|Number of<br>cases|Value in Euro (not actual price,<br>according to NHIF provisions)|Euro/case|\n|---|---|---|---|\n|In-patient|705|249,240|353.53|\n|Out-patient|2,805|13,130.9|4.68|\n|Total|3,510|262,370|74.74|\n\n\nSource: Data provided by courtesy of the Hungarian NHIF, assembled by the author. The\nhospital/physician determines the price in advance. Only in-network hospitals/physicians have the duty to\nreport to the NHIF.\n\n**5 POTENTIAL FINANCIAL BENEFITS FROM CROSS-BORDER**\n**PURCHASING FOR SELECTED HEALTH CARE SERVICES**\nThe above presented data show that cross-border purchasing of health care is at the\nmoment a minor phenomenon regarding health services covered by the statutory Austrian\nhealth insurance system. This section will look at the potential benefits for Austrian\ninsurance institutions and individual patients regarding health services which are not\nalways fully covered: cataract surgeries, dental treatment, and cosmetic surgeries.\n\nIn principle, the statutory health insurance in Austria covers all treatments that are\nmedically indicated; only few areas are not covered such as specific cosmetic surgeries or\nfixed dentures. Each regional or occupational HIF has its own catalogue of benefits that\nare covered by it. The HIFs fund the public hospitals with lump sums without giving a\nlimit regarding the treatments covered. Each hospital then decides independently whether\n\n\n18\n\nMention: \"the Hungarian NHIF\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Hospital case numbers and prices were supplied by this insurer.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004022:15:1", "occurrence_id": "4c069dbb9cb9fd10e360b01d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  outperforms Barro and Lee's set in growth\n\n\nregressions. However, Cohen and Soto's data for Africa clearly underestimate official\n\n\nstatistics. According to the South African 1996 census, the share of educated individuals\n\n\namounts to 7.2 percent. Cohen and Soto report 3 percent (Barro and Lee report 6.9 percent).\n\n\nThe Kenyan 1999 census gives 2 percent while Cohen and Soto report 0.9 percent (1.2 for\n\n\nBarro and Lee).\n\n\nGenerally speaking, the Cohen and Soto data set predicts extremely low levels of human\n\ncapital for African countries <sup>21</sup> [^21: For this reason, Cohen and Soto (2001) exclude African countries from their growth regressions.] (the share of tertiary educated is lower than 1 percent in a large\n\nnumber of African countries) and a couple of other non-OECD countries <sup>22</sup> [^22: In Cyprus, the 2001 census gives 22%, compared with 4.6% in Cohen and Soto (17.1% in Barro and Lee).] . The Barro and\n\n\n17 For some countries, immigrants often travel back and forth between their new and old countries (e.g. Mexico).\nThey are likely to be counted as still being residents in their home country. For that reason, Carrington and\nDetragiache (1998) provide an upper bound (m=M/N) and a lower bound (m=M/(N+M)). Since the upper bound\nis not interpretable for a large number of countries (higher than one), we only report the lower bound.\n18 See <u>[http://esa.un.org/unpp.](http://esa.un.org/unpp)</u>\n19 See <u>[http://www.cia.gov/cia/publications/factbook.](http://www.cia.\n\nMention: \"South African 1996 census\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its educated share of 7.2 percent exceeds the alternative estimate.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004022:15:1", "occurrence_id": "69183fe63d8a73d5657f352e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  outperforms Barro and Lee's set in growth\n\n\nregressions. However, Cohen and Soto's data for Africa clearly underestimate official\n\n\nstatistics. According to the South African 1996 census, the share of educated individuals\n\n\namounts to 7.2 percent. Cohen and Soto report 3 percent (Barro and Lee report 6.9 percent).\n\n\nThe Kenyan 1999 census gives 2 percent while Cohen and Soto report 0.9 percent (1.2 for\n\n\nBarro and Lee).\n\n\nGenerally speaking, the Cohen and Soto data set predicts extremely low levels of human\n\ncapital for African countries <sup>21</sup> [^21: For this reason, Cohen and Soto (2001) exclude African countries from their growth regressions.] (the share of tertiary educated is lower than 1 percent in a large\n\nnumber of African countries) and a couple of other non-OECD countries <sup>22</sup> [^22: In Cyprus, the 2001 census gives 22%, compared with 4.6% in Cohen and Soto (17.1% in Barro and Lee).] . The Barro and\n\n\n17 For some countries, immigrants often travel back and forth between their new and old countries (e.g. Mexico).\nThey are likely to be counted as still being residents in their home country. For that reason, Carrington and\nDetragiache (1998) provide an upper bound (m=M/N) and a lower bound (m=M/(N+M)). Since the upper bound\nis not interpretable for a large number of countries (higher than one), we only report the lower bound.\n18 See <u>[http://esa.un.org/unpp.](http://esa.un.org/unpp)</u>\n19 See <u>[http://www.cia.gov/cia/publications/factbook.](http://www.cia.\n\nMention: \"Kenyan 1999 census\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its figure likewise exceeds the comparable estimates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004025:21:1", "occurrence_id": "ac713b833afc4d88b946408d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  from 1 (highest) to 7\nCivil Liberties Freedom House Index of civil liberties from 1 (highest) to 7\nMalaria Gallup et. al. Fraction of population in areas with malaria\nNavigable Water Gallup et. al. Fraction of land area near navigable water\nLandlocked Country Gallup et. al. Dummy for landlocked countries\nAir Distance Gallup et. al. Logarithm of minimal distance in km from\nNew York, Rotterdam, or Tokio\n<u>Tropical</u> <u>Area</u> <u>Gallup</u> <u>et.</u> <u>al.</u> <u>Fraction</u> <u>of</u> <u>land</u> <u>area</u> <u>in</u> <u>geographical</u> <u>tropics</u>\n\n\nNotes:\n1. PWT 6.2 refers to Penn World Table 6.2\n2. WDI 2005 refers to World Development Indicators 2005 from The World Bank\n\n\n20\n\nMention: \"Freedom House\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its civil liberties index runs from one to seven.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004025:21:1", "occurrence_id": "4ef4d307ef110dbbfb11c98a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  from 1 (highest) to 7\nCivil Liberties Freedom House Index of civil liberties from 1 (highest) to 7\nMalaria Gallup et. al. Fraction of population in areas with malaria\nNavigable Water Gallup et. al. Fraction of land area near navigable water\nLandlocked Country Gallup et. al. Dummy for landlocked countries\nAir Distance Gallup et. al. Logarithm of minimal distance in km from\nNew York, Rotterdam, or Tokio\n<u>Tropical</u> <u>Area</u> <u>Gallup</u> <u>et.</u> <u>al.</u> <u>Fraction</u> <u>of</u> <u>land</u> <u>area</u> <u>in</u> <u>geographical</u> <u>tropics</u>\n\n\nNotes:\n1. PWT 6.2 refers to Penn World Table 6.2\n2. WDI 2005 refers to World Development Indicators 2005 from The World Bank\n\n\n20\n\nMention: \"Penn World Table 6.2\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The abbreviation in the notes refers to this version.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004025:21:1", "occurrence_id": "fd5da31937d219ff1d49a889", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  from 1 (highest) to 7\nCivil Liberties Freedom House Index of civil liberties from 1 (highest) to 7\nMalaria Gallup et. al. Fraction of population in areas with malaria\nNavigable Water Gallup et. al. Fraction of land area near navigable water\nLandlocked Country Gallup et. al. Dummy for landlocked countries\nAir Distance Gallup et. al. Logarithm of minimal distance in km from\nNew York, Rotterdam, or Tokio\n<u>Tropical</u> <u>Area</u> <u>Gallup</u> <u>et.</u> <u>al.</u> <u>Fraction</u> <u>of</u> <u>land</u> <u>area</u> <u>in</u> <u>geographical</u> <u>tropics</u>\n\n\nNotes:\n1. PWT 6.2 refers to Penn World Table 6.2\n2. WDI 2005 refers to World Development Indicators 2005 from The World Bank\n\n\n20\n\nMention: \"WDI 2005\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The notes define this abbreviation for the indicator compilation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004026:7:0", "occurrence_id": "27f041aa7431673a04416a81", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **3.** **Data**\n\n\nThis paper uses Pakistan Integrated Household Survey (PIHS) 1998/1999,\n\n\nPIHS2001/2002, and PIHS2005/20006 <sup>3</sup> for data analysis. The Pakistan PIHS data include\n\n\ndifferent modules such as household demographic information, household assets module,\n\n\nemployment and schooling. The dataset also includes a detailed consumption module, which has\n\n\nbeen used to examine the relationship between wealth and child activity decisions.\n\n\nThis paper defines child labor as a child who is aged between 10 and 14, and employed.\n\n\nAlthough different organizations have used different age limits to define child labor, in general,\n\n\nthe minimum age of child employment is considered to be 15 years. This is largely based on the\n\n\nminimum age of completion of compulsory schooling. The lower bound of 10 years is used\n\n\nbecause surveys only ask household members aged 10 years old or above about their\n\n\nemployment activities. The definition of “work” is based on the question: “Did (the person) do\n\n\nany work for pay, profit or family gain during last month at least for one hour on any day?” If the\n\n\nhousehold member answers “yes”, he/she is considered to be employed.\n\n\nCommunity is defined as the primary sampling units (PSU) provided in the data. Since\n\n\nhouseholds are randomly selected within each PSU from lower income, middle income and high\n\n\nincome groups, the average at the community level is considered to be a good representative of\n\n\nprevailing conditions in the community.\n\n\n**4. Empirical Method**\n\n\nThe empirical estimation is based on the reduced form of equation 1. The empirical\n\n\nmethods applied have three notable features.\n\n\n3 A similar survey was also conducted in 2004/2005 but some critical questions in the labor module in 2004/2005\nare different from the surveys used in the paper. Therefore, the study can not use PIHS2004/2005.\n\n\n6\n\nMention: \"Pakistan Integrated Household Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its multiple rounds supply the household modules used in the analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004026:7:0", "occurrence_id": "552eac7c125acf655ccd3e59", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **3.** **Data**\n\n\nThis paper uses Pakistan Integrated Household Survey (PIHS) 1998/1999,\n\n\nPIHS2001/2002, and PIHS2005/20006 <sup>3</sup> for data analysis. The Pakistan PIHS data include\n\n\ndifferent modules such as household demographic information, household assets module,\n\n\nemployment and schooling. The dataset also includes a detailed consumption module, which has\n\n\nbeen used to examine the relationship between wealth and child activity decisions.\n\n\nThis paper defines child labor as a child who is aged between 10 and 14, and employed.\n\n\nAlthough different organizations have used different age limits to define child labor, in general,\n\n\nthe minimum age of child employment is considered to be 15 years. This is largely based on the\n\n\nminimum age of completion of compulsory schooling. The lower bound of 10 years is used\n\n\nbecause surveys only ask household members aged 10 years old or above about their\n\n\nemployment activities. The definition of “work” is based on the question: “Did (the person) do\n\n\nany work for pay, profit or family gain during last month at least for one hour on any day?” If the\n\n\nhousehold member answers “yes”, he/she is considered to be employed.\n\n\nCommunity is defined as the primary sampling units (PSU) provided in the data. Since\n\n\nhouseholds are randomly selected within each PSU from lower income, middle income and high\n\n\nincome groups, the average at the community level is considered to be a good representative of\n\n\nprevailing conditions in the community.\n\n\n**4. Empirical Method**\n\n\nThe empirical estimation is based on the reduced form of equation 1. The empirical\n\n\nmethods applied have three notable features.\n\n\n3 A similar survey was also conducted in 2004/2005 but some critical questions in the labor module in 2004/2005\nare different from the surveys used in the paper. Therefore, the study can not use PIHS2004/2005.\n\n\n6\n\nMention: \"PIHS2005/20006\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The latest round included in the analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004026:7:0", "occurrence_id": "52ff39ce6af3d8edd61ce0dc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **3.** **Data**\n\n\nThis paper uses Pakistan Integrated Household Survey (PIHS) 1998/1999,\n\n\nPIHS2001/2002, and PIHS2005/20006 <sup>3</sup> for data analysis. The Pakistan PIHS data include\n\n\ndifferent modules such as household demographic information, household assets module,\n\n\nemployment and schooling. The dataset also includes a detailed consumption module, which has\n\n\nbeen used to examine the relationship between wealth and child activity decisions.\n\n\nThis paper defines child labor as a child who is aged between 10 and 14, and employed.\n\n\nAlthough different organizations have used different age limits to define child labor, in general,\n\n\nthe minimum age of child employment is considered to be 15 years. This is largely based on the\n\n\nminimum age of completion of compulsory schooling. The lower bound of 10 years is used\n\n\nbecause surveys only ask household members aged 10 years old or above about their\n\n\nemployment activities. The definition of “work” is based on the question: “Did (the person) do\n\n\nany work for pay, profit or family gain during last month at least for one hour on any day?” If the\n\n\nhousehold member answers “yes”, he/she is considered to be employed.\n\n\nCommunity is defined as the primary sampling units (PSU) provided in the data. Since\n\n\nhouseholds are randomly selected within each PSU from lower income, middle income and high\n\n\nincome groups, the average at the community level is considered to be a good representative of\n\n\nprevailing conditions in the community.\n\n\n**4. Empirical Method**\n\n\nThe empirical estimation is based on the reduced form of equation 1. The empirical\n\n\nmethods applied have three notable features.\n\n\n3 A similar survey was also conducted in 2004/2005 but some critical questions in the labor module in 2004/2005\nare different from the surveys used in the paper. Therefore, the study can not use PIHS2004/2005.\n\n\n6\n\nMention: \"PIHS2004/2005\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its labour module differs, so this round is excluded from the study.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004030:18:0", "occurrence_id": "d0a214123fef6ae939e6c2e2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: #### **Figure 1**\n\nThe figure is a partial scatter plot showing the relationship between _Reform_ and _Democracy_ on average over the sample period, and controlling for GDP per\ncapita and regional fixed effects. The Y axis plots the residuals from the linear regression of _Reform_ on the control variables while the X axis contains residuals\nfrom a similar regression using _Democracy_ . Both these regressions are run over average values of all the variables taken over the sample period (2003-2008).\nAbbreviations of country names follow World Development Indicators, World Bank.\n\n\n17\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Country name abbreviations follow this compilation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004036:8:0", "occurrence_id": "6ab3b9f58cf24eb1a07a130a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **_2.2. AMADEUS Database_**\n\n\nFor the estimation of firm productivity, the analysis uses data from the May 2006 edition\n\n\nof the AMADEUS database, a comprehensive, pan-European commercial database compiled by\n\n\nBureau van Dijk. For each firm, AMADEUS provides accounting data in standardized financial\n\n\nformat for 24 balance sheet items, 25 profit and loss account items, 26 financial ratios, and\n\n\nadditional information including trade description and activity codes. The database includes up\n\n\nto ten years of information per firm through 2004, although coverage varies by country.\n\n\nAMADEUS is created by collecting standardized data received from 50 vendors across Europe,\n\nwhere the local source for these data is generally the office of the Registrar of Companies. <sup>5</sup> [^5: Further details about the AMADEUS database can be found on the product page of Bureau van Dijk’s website:\nhttp://www.bvdep.com/en/AMADEUS.html.]\n\n\nThe accounts for each firm are transformed into a universal format to allow for\n\n\ncomparison across countries. All accounting data is converted into U.S. dollars using period\n\n\naverage exchange rates, based on monthly series from the International Monetary Fund, nearest\n\n\nto the end date of each respective financial account. Nominal values are deflated using country\n\nlevel GDP deflators to express values in 2001 US dollars.  In addition, all firms are categorized\n\n\nby industry according to NACE Rev.1., and for the analysis, industry dummy variables are coded\n\n\nbased on the 4-digit activity code following NACE Rev.1 that AMADEUS assigns to each firm.\n\n\n**_2.3. Sample Selection_**\n\n\nThe econometric analysis of firm-level TFP growth and changes in the business\n\n\nenvironment uses a first-differenced equation in firm characteristics with two-year changes and\n\n\nrequires a panel of manufacturing firms from the AMADEUS database with complete\n\n\ninformation on production function variables for the years 2001 through 2004, the period that\n\n\ncorrespond to the 2002 and 2005 rounds of the BEEPS. Specifically, output, labor, material\n\n\ninputs, and capital\n\nMention: \"AMADEUS database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Firm-level productivity is estimated from this pan-European edition.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004039:9:1", "occurrence_id": "d1860422121ad7eb4da750d2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: avallion et al.,\n\n1991). AB then argued that the value of _k_ could also be based on national poverty lines above\n\nthose found in the poorest countries, by studying how those lines vary with man consumption in\n\nthe original data base of poverty lines used by Ravallion et al. (1991) to set the “$1-a-day” line.\n\nBy visual inspection of the Ravallion et al. (1991) data set on national poverty lines at 1985 PPP,\n\nAtkinson and Bourguignon set _k_ =0.37. Subsequently, Chen and Ravallion (2001) found that\n\n_k_ =1/3 gave a better fit with the Ravallion et al. (1991) data set at 1993 PPP.\n\nHowever, the AB line fails the WRA in that it has an elasticity of unity for all countries\n\nwith _M i_ - _Z_ - / _k_ . This is surely implausible. The idea that distribution-neutral growth has no\n\nimpact on the extent of poverty in new middle-income countries such as China would surely be\n\nvery hard to accept (not least, we would conjecture, in China). The violation of the WEA stems\n\nfrom the seemingly implausible assumption that the cost of social inclusion is directly\n\n\n8\n\nMention: \"Ravallion et al. (1991) data set\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its national poverty lines at that purchasing power basis informed the daily line.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004045:14:0", "occurrence_id": "32970b80108a2f193ffeb284", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: events. Educated (or richer) Hutu were also killed, both in the genocide as well as in the\n\n\nwar with the Rwandese Patriotic Front (RPF). The literature offers dramatic examples of\n\n\nthis. Mamdami (2001) reports, for example, that, in order to escape killing, Tutsi girls\n\n\n‘ _had to dress and behave like peasants_ ‘implying that the victims were also targeted\n\n\nbased on class (rich vs. poor or urban vs. rural). Since our dependent variable captures all\n\n\ndeaths in 1994, not just in the three months of the genocide, it includes all deaths in the\n\n\ngenocide and the war (Tutsi as well as Hutu) as well as deaths by disease (e.g. in the\n\n\nrefugee camps in Democratic Republic of Congo).\n\n\nOur econometric strategy is as follows: the 1992 DHS contains the ethnicity of\n\n\nthe respondents. We are thus able to analyse differences between Hutu and Tutsi in 1992.\n\n\nPanel A of table 2 analyzes the correlation between being Tutsi, and education and urban\n\n\nresidence in the female sample of the 1992 DHS. The results show that even when\n\n\ncontrolling for region and age, there is a strong positive correlation between education,\n\n\nurban status and being Tutsi. This confirms our fear that the relationship between\n\n\nmortality in 1994 and education and urban residence is at least partially driven by the fact\n\n\nthat Tutsi were more likely to die during the genocide.\n\n\nOur next step is then to find a proxy for being Tutsi, a variable that would be\n\n\ncorrelated with Tutsi but that would _not_ be correlated, by itself, i.e. independently from\n\n\nbeing Tutsi, with the probability of dying in 1994. We believe that the fact that Tutsi\n\n\nmarried at a later age satisfies these conditions. Panel B of table 2 indicates that even\n\n\nafter controlling for education, urban residence, region and age, Tutsi women married\n\n\nlater. Thus, age at marriage is positively correlated with being Tutsi. This fact had\n\n\n13\n\nMention: \"the 1992 DHS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Respondent ethnicity here allows a split between the two groups.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004048:3:0", "occurrence_id": "415fba0570bb94eb8614570b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: - 2 \n\n**I.** **INTRODUCTION**\n\n\nInfrastructure procurement is still a challenging task for governments, not only because\n\n\npublic works are expensive, but also because the nature of the works is often highly\n\n\ncomplicated and customized. Competitive bidding became prevalent among public\n\n\nprocurement systems throughout the world, but how to design the auction mechanism\n\n\nremains difficult in practice, especially when bidder participation is endogenously\n\n\ndetermined. Typical are large-scale electricity projects. Governments or public entities,\n\n\nnamely auctioneers, may be able to increase or decrease the number of bidders at will, but at\n\n\ncertain expense.\n\n\nThe endogeneity of bidder participation is a complex issue in auction theory as well as its\n\n\napplications. As is well known, competition is the single most important factor toward\n\n\nachieving economic efficiency and anti-corrupt allocation of an object through competitive\n\n\nbidding. Even in the traditional fixed- _n_ approach where the number of bidders is assumed\n\n\nfixed and commonly known by bidders, the equilibrium bid is expected to decrease with\n\ncompetition. <sup>1</sup> In the infrastructure procurement, it is of particular concern that the realized\n\n\ndegree of competition is often limited, regardless of whether it is contracted out through a\n\n\ntraditional method or public-private partnership (PPP) arrangement (e.g., Foster, 2005; NOA,\n\n2007; Estache and Iimi, 2008a). <sup>2</sup> [^2: According to the Private Participation in Infrastructure (PPI) database, the top 10 percent largest firms—\ndefined by the number of transactions that each company obtained—were awarded about half of total\ninfrastructure PPP contracts.]\n\n\nEstache and Iimi (2008a) confirm the importance of intensifying competition to contain\n\n\ninfrastructure procurement costs, but in a static sense. The previous analysis presumes that\n\n\n1 In the following sections of this paper, a government procurement auction is considered unless otherwise\nmentioned. Thus, bidder’s cost of procuring the contracted-out work is a piece of unknown private information.\nThe best contractor is a firm with the lowest possible\n\nMention: \"Private Participation in Infrastructure (PPI) database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It shows the ten percent largest firms winning most transactions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004055:9:0", "occurrence_id": "89e2f5de89d00ca6022809f7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: measure poverty status per se, our approach is to be more inclusive rather than less so.\nUsing the Planning Commission’s estimates, we assume 28% and 16% poverty rates for\nthe rural and urban sectors respectively in 1993 and 14% and 15% in 2004. We adopt this\nrelatively high cutoff to better capture households receiving BPL benefits. In our\nrobustness checks, we test this assumption using an alternative, less inclusive scenario\nbased on Deaton’s (2003) criteria. This assumes poverty rates of 17% and 11% for the\nrural and urban sectors in 1993 and 6% and 5% in 1999.\n\n**3.2.2. Which Households Are Eligible Under the SC/OBC Rule?**\n\nUnder the program rules, all households from a disadvantaged caste background are\neligible unless they are extremely wealthy, i.e. all SC and OBC households, excluding\nthose in which the household heads are gazetted government employees or pay income\ntaxes. The major challenge in identifying SC and OBC households in this data is that SC\nand OBC status are reported as separate mutually exclusive categories in NFHS-2 and\nNFHS-3, but in NFHS-1 only SC is reported.\n\nIn order to match SC and OBC classifications across all surveys, we first identify the\ncaste of the household head in NFHS-1. We then categorize that household as OBC if\nhouseholds of that caste are categorized as OBC in NFHS-2. This results in the recategorization of some castes that fall into SC in NFHS-1 as now OBC. In general, to\navoid confusion, for analytical purposes we conflate SC and OBC together into a single\ncategory.\n\nWhile we have no means of identifying top ranking government employees, this is likely\nto be an insignificant fraction of this sample. We do however address the income-tax\nqualification. In India, income tax-paying households are estimated to be the top 1-4% of\nthe\n\nMention: \"Planning Commission\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its rural and urban rates calibrate the poverty assumptions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004055:9:0", "occurrence_id": "e5fb1bea5d7e5a35a30420c9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: measure poverty status per se, our approach is to be more inclusive rather than less so.\nUsing the Planning Commission’s estimates, we assume 28% and 16% poverty rates for\nthe rural and urban sectors respectively in 1993 and 14% and 15% in 2004. We adopt this\nrelatively high cutoff to better capture households receiving BPL benefits. In our\nrobustness checks, we test this assumption using an alternative, less inclusive scenario\nbased on Deaton’s (2003) criteria. This assumes poverty rates of 17% and 11% for the\nrural and urban sectors in 1993 and 6% and 5% in 1999.\n\n**3.2.2. Which Households Are Eligible Under the SC/OBC Rule?**\n\nUnder the program rules, all households from a disadvantaged caste background are\neligible unless they are extremely wealthy, i.e. all SC and OBC households, excluding\nthose in which the household heads are gazetted government employees or pay income\ntaxes. The major challenge in identifying SC and OBC households in this data is that SC\nand OBC status are reported as separate mutually exclusive categories in NFHS-2 and\nNFHS-3, but in NFHS-1 only SC is reported.\n\nIn order to match SC and OBC classifications across all surveys, we first identify the\ncaste of the household head in NFHS-1. We then categorize that household as OBC if\nhouseholds of that caste are categorized as OBC in NFHS-2. This results in the recategorization of some castes that fall into SC in NFHS-1 as now OBC. In general, to\navoid confusion, for analytical purposes we conflate SC and OBC together into a single\ncategory.\n\nWhile we have no means of identifying top ranking government employees, this is likely\nto be an insignificant fraction of this sample. We do however address the income-tax\nqualification. In India, income tax-paying households are estimated to be the top 1-4% of\nthe\n\nMention: \"NFHS-2\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"It reports the two castes as separate exclusive categories.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004055:9:0", "occurrence_id": "3392ff52d1ed4916a8fce049", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: measure poverty status per se, our approach is to be more inclusive rather than less so.\nUsing the Planning Commission’s estimates, we assume 28% and 16% poverty rates for\nthe rural and urban sectors respectively in 1993 and 14% and 15% in 2004. We adopt this\nrelatively high cutoff to better capture households receiving BPL benefits. In our\nrobustness checks, we test this assumption using an alternative, less inclusive scenario\nbased on Deaton’s (2003) criteria. This assumes poverty rates of 17% and 11% for the\nrural and urban sectors in 1993 and 6% and 5% in 1999.\n\n**3.2.2. Which Households Are Eligible Under the SC/OBC Rule?**\n\nUnder the program rules, all households from a disadvantaged caste background are\neligible unless they are extremely wealthy, i.e. all SC and OBC households, excluding\nthose in which the household heads are gazetted government employees or pay income\ntaxes. The major challenge in identifying SC and OBC households in this data is that SC\nand OBC status are reported as separate mutually exclusive categories in NFHS-2 and\nNFHS-3, but in NFHS-1 only SC is reported.\n\nIn order to match SC and OBC classifications across all surveys, we first identify the\ncaste of the household head in NFHS-1. We then categorize that household as OBC if\nhouseholds of that caste are categorized as OBC in NFHS-2. This results in the recategorization of some castes that fall into SC in NFHS-1 as now OBC. In general, to\navoid confusion, for analytical purposes we conflate SC and OBC together into a single\ncategory.\n\nWhile we have no means of identifying top ranking government employees, this is likely\nto be an insignificant fraction of this sample. We do however address the income-tax\nqualification. In India, income tax-paying households are estimated to be the top 1-4% of\nthe\n\nMention: \"NFHS-3\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"It likewise reports the two castes separately.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004055:9:0", "occurrence_id": "41b6d25b4f3d975bab3e8b1b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: measure poverty status per se, our approach is to be more inclusive rather than less so.\nUsing the Planning Commission’s estimates, we assume 28% and 16% poverty rates for\nthe rural and urban sectors respectively in 1993 and 14% and 15% in 2004. We adopt this\nrelatively high cutoff to better capture households receiving BPL benefits. In our\nrobustness checks, we test this assumption using an alternative, less inclusive scenario\nbased on Deaton’s (2003) criteria. This assumes poverty rates of 17% and 11% for the\nrural and urban sectors in 1993 and 6% and 5% in 1999.\n\n**3.2.2. Which Households Are Eligible Under the SC/OBC Rule?**\n\nUnder the program rules, all households from a disadvantaged caste background are\neligible unless they are extremely wealthy, i.e. all SC and OBC households, excluding\nthose in which the household heads are gazetted government employees or pay income\ntaxes. The major challenge in identifying SC and OBC households in this data is that SC\nand OBC status are reported as separate mutually exclusive categories in NFHS-2 and\nNFHS-3, but in NFHS-1 only SC is reported.\n\nIn order to match SC and OBC classifications across all surveys, we first identify the\ncaste of the household head in NFHS-1. We then categorize that household as OBC if\nhouseholds of that caste are categorized as OBC in NFHS-2. This results in the recategorization of some castes that fall into SC in NFHS-1 as now OBC. In general, to\navoid confusion, for analytical purposes we conflate SC and OBC together into a single\ncategory.\n\nWhile we have no means of identifying top ranking government employees, this is likely\nto be an insignificant fraction of this sample. We do however address the income-tax\nqualification. In India, income tax-paying households are estimated to be the top 1-4% of\nthe\n\nMention: \"NFHS-1\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Only one caste category is reported here, requiring harmonisation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004058:15:0", "occurrence_id": "16898507675193cd5eecbdb9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Table 7 Street Lighting in UAs**\n\n\nCentral City\n\n\n\nNon Central City\n\n\nAverage\n\n\n\nUrban Agglomeration\n(Maximum, Minimum)\n\n\n\nNumber of Street\n\nLights\n\n\n\nHouseholds per\n\nStreet Light\n\n\n\nNumber of Street\n\nLights\n\n\n\nHouseholds per Street\n\nLight\n\n\n11\n\n\n\nHyderabad Urban\n\n\n\nHyderabad Urban 6,368\n\n18,138 36\nAgglomeration\n\n\n\n<u>(10,300, 1,465)</u> <u>(44, 2)</u>\n\nChennai Urban 5,922 199\n79,303 12\nAgglomeration\n\n<u>(1,7606, 25)</u> <u>(1,350, 4)</u>\n\nKolkata Urban 3,408 20\n78,354 12\nAgglomeration\n\n\n\n<u>(10,300, 1,465)</u>\n\n\n\n199\n\n\n\n<u>(1,7606, 25)</u>\n\n\n\n20\n\n\n\n(12,000, 55)\n\n\n\n(181, 3)\n\n\n\nDelhi Urban\nNA NA NA NA\nAgglomeration\n\n\n\nPune Urban\nAgglomeration\n\n\n\n2,323\n59,001 9\n\n\n\n<u>(4,000, 739)</u>\n\n\n\n82\n\n\n\n<u>(4,000, 739)</u> <u>(313, 4)</u>\n\n**Source: Census of India, 2001, Authors’ Computations**\n\n\n\n**3. Finances of Urban Local Bodies**\n\n\nWe analyze the finances of the ULBs by considering expenditures and revenues\n\n\nseparately. The main sources of information on expenditures and revenues of ULBs are the\n\n\nAnnual Administrative Reports, Budgets of the ULBs, the City Development Reports, and\n\n\nEnvironmental Status Reports, depending on availability at the city level. The smaller ULBs\n\n\nin India do not maintain systematic records of their finances. The data for this study were\n\n\ncollected through extensive field visits to all the ULBs by circulating questionnaires\n\n\n14\n\nMention: \"Census of India\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Agglomeration counts and population figures are computed from this enumeration.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004059:20:0", "occurrence_id": "e61f471c6429f17a010d5e32", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 19\n\n\n**The World Bank’s Agricultural Distortions database**\n\n\nThe database generated by the World Bank’s Agricultural Distortions project\n\n\n(Anderson and Valenzuela 2008), using a methodology summarized in Anderson et al.\n\n\n(2008), provides a timely opportunity to estimate global commodity trade and welfare\n\n\nreduction indexes for individual commodity markets. The database contains consistent\n\n\nestimates of annual nominal rates of assistance (NRAs) and consumer tax equivalents\n\n\n(CTEs) at the commodity level, for a set of agricultural products (called covered\n\n\nproducts). These products account for around 70 percent of total agricultural production\n\n\nin 75 countries (called focus countries), which in turn account for 92 percent of global\n\n\nagricultural GDP. The data cover a time period between 1955 and 2007 for the majority\n\n\nof countries but the country coverage is most complete for the years 1960 to 2004, so\n\n\nonly those are used here. Global NRAs and CTEs for various commodities are\n\n\nestimated, using as weights the values of production and consumption, respectively, at\n\n\nundistorted prices. <sup>4</sup>\n\n\nThe range of measures included in the Agricultural Distortions database NRA\n\n\nestimates is wide. By calculating domestic-to-border price ratios the estimates include\n\n\nassistance provided by all tariff and non-tariff trade measures, plus any domestic price\n\n\nsupport measures (positive or negative), plus an adjustment for the output-price\n\n\nequivalent of direct interventions on inputs. Where multiple exchange rates operate, an\n\n\n44 Appendix Tables 1 and 2 report those estimates for 28 major products, whose NRAs and CTEs cover 77\npercent of their global production on average.\n\nMention: \"World Bank’s Agricultural Distortions database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Nominal rate assistance estimates for products and countries are compiled here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004060:4:0", "occurrence_id": "a90f5b1b0cbcef9783ff9ba4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 2\n\n\nThere is a need for better indicators over time of the trade- and welfare-reducing effects\n\n\nof price-distorting policies than the existing weighted average NRA/CTE (or PSE/CSE)\n\n\nestimates for the farm sector, because the averaging process hides the fact that distortions vary\n\n\nacross industries within the sector. This is especially problematic in cases where NRAs have\n\n\nopposite signs, as when trade taxes apply to both imports and exports, or when dual exchange\n\n\nrates operate. In those cases the NRA may be close to zero even though the trade- and welfare\n\nreducing effects of those interventions could be substantial.\n\n\nRecent theoretical literature provides partial equilibrium indicators of the trade- and\n\n\nwelfare-reducing effects of import policies. The purpose of this paper is to draw on that literature\n\n\nin order to develop indexes to capture global distortions to agricultural incentives that are based\n\n\ndirectly on estimates of NRAs and CTEs. We then exploit the Agricultural Distortion database\n\n\nrecently compiled by the World Bank to generate the first set of estimates of consistent indexes\n\n\nfor the agricultural sector for both developing and high-income countries over the past half\n\n\ncentury.\n\n\nThe new World Bank global panel dataset contains comparable estimates of annual\n\n\nNRAs and CTEs for a wide range of agricultural products for around 75 countries that together\n\n\naccount for all but one-tenth of the world’s population, GDP and agricultural production\n\n\n(Anderson and Valenzuela 2008). Applying our indexes to these data takes us much closer to\n\n\nunderstanding the true trade and welfare effects of policies without needing a formal model of\n\n\nglobal markets or even price elasticity estimates.\n\n\nThe remainder of the paper is structured as follows. After a brief literature review the\n\n\nnext section presents the theory for estimating welfare- and trade-reduction index numbers in the\n\n\nimport-competing sub-sector. This is extended to cover the exportables sub-sector in the\n\n\nfollowing section. The World Bank’s Agricultural Distortions database is then discussed,\n\nMention: \"Agricultural Distortion database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"It generates the first consistent index estimates across countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004060:4:0", "occurrence_id": "eb87b4b2a425923f38325b77", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 2\n\n\nThere is a need for better indicators over time of the trade- and welfare-reducing effects\n\n\nof price-distorting policies than the existing weighted average NRA/CTE (or PSE/CSE)\n\n\nestimates for the farm sector, because the averaging process hides the fact that distortions vary\n\n\nacross industries within the sector. This is especially problematic in cases where NRAs have\n\n\nopposite signs, as when trade taxes apply to both imports and exports, or when dual exchange\n\n\nrates operate. In those cases the NRA may be close to zero even though the trade- and welfare\n\nreducing effects of those interventions could be substantial.\n\n\nRecent theoretical literature provides partial equilibrium indicators of the trade- and\n\n\nwelfare-reducing effects of import policies. The purpose of this paper is to draw on that literature\n\n\nin order to develop indexes to capture global distortions to agricultural incentives that are based\n\n\ndirectly on estimates of NRAs and CTEs. We then exploit the Agricultural Distortion database\n\n\nrecently compiled by the World Bank to generate the first set of estimates of consistent indexes\n\n\nfor the agricultural sector for both developing and high-income countries over the past half\n\n\ncentury.\n\n\nThe new World Bank global panel dataset contains comparable estimates of annual\n\n\nNRAs and CTEs for a wide range of agricultural products for around 75 countries that together\n\n\naccount for all but one-tenth of the world’s population, GDP and agricultural production\n\n\n(Anderson and Valenzuela 2008). Applying our indexes to these data takes us much closer to\n\n\nunderstanding the true trade and welfare effects of policies without needing a formal model of\n\n\nglobal markets or even price elasticity estimates.\n\n\nThe remainder of the paper is structured as follows. After a brief literature review the\n\n\nnext section presents the theory for estimating welfare- and trade-reduction index numbers in the\n\n\nimport-competing sub-sector. This is extended to cover the exportables sub-sector in the\n\n\nfollowing section. The World Bank’s Agricultural Distortions database is then discussed,\n\nMention: \"World Bank global panel dataset\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Annual estimates for around ninety products over half a century sit in this panel.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004060:4:0", "occurrence_id": "0838b62aa21c4f2de283cbcf", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 2\n\n\nThere is a need for better indicators over time of the trade- and welfare-reducing effects\n\n\nof price-distorting policies than the existing weighted average NRA/CTE (or PSE/CSE)\n\n\nestimates for the farm sector, because the averaging process hides the fact that distortions vary\n\n\nacross industries within the sector. This is especially problematic in cases where NRAs have\n\n\nopposite signs, as when trade taxes apply to both imports and exports, or when dual exchange\n\n\nrates operate. In those cases the NRA may be close to zero even though the trade- and welfare\n\nreducing effects of those interventions could be substantial.\n\n\nRecent theoretical literature provides partial equilibrium indicators of the trade- and\n\n\nwelfare-reducing effects of import policies. The purpose of this paper is to draw on that literature\n\n\nin order to develop indexes to capture global distortions to agricultural incentives that are based\n\n\ndirectly on estimates of NRAs and CTEs. We then exploit the Agricultural Distortion database\n\n\nrecently compiled by the World Bank to generate the first set of estimates of consistent indexes\n\n\nfor the agricultural sector for both developing and high-income countries over the past half\n\n\ncentury.\n\n\nThe new World Bank global panel dataset contains comparable estimates of annual\n\n\nNRAs and CTEs for a wide range of agricultural products for around 75 countries that together\n\n\naccount for all but one-tenth of the world’s population, GDP and agricultural production\n\n\n(Anderson and Valenzuela 2008). Applying our indexes to these data takes us much closer to\n\n\nunderstanding the true trade and welfare effects of policies without needing a formal model of\n\n\nglobal markets or even price elasticity estimates.\n\n\nThe remainder of the paper is structured as follows. After a brief literature review the\n\n\nnext section presents the theory for estimating welfare- and trade-reduction index numbers in the\n\n\nimport-competing sub-sector. This is extended to cover the exportables sub-sector in the\n\n\nfollowing section. The World Bank’s Agricultural Distortions database is then discussed,\n\nMention: \"Agricultural Distortions database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The paper describes this database and extends its coverage.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004064:4:0", "occurrence_id": "9cf005836abcafc054ef7dfa", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: performances. In a context of increasing pressure of globalization, _ICA_ surveys can be seen as\nan instrument for identifying key obstacles to firms’ productivity and competitiveness. They\ncan be used as a support to policy reforms for an increased economic growth.\n\n\nDrawing on the World Bank firm-surveys, this paper analyzes the relationship between\ninvestment climate and firm-level productivity for a large number of developing countries (22\namong which 5 MENA economies) and eight manufacturing industries. We first propose\ndifferent measures of firms’ productive performances, such as Labor Productivity ( _LP_ ), Total\nFactor Productivity ( _TFP_ ), and Technical Efficiency ( _TE_ ) using a production frontier\napproach. These indicators are compared to each others as well as across countries, in order to\nposition MENA manufacturing amongst a wide range of firms from other regions. We reveal\nthat enterprises in MENA perform in average poorly, compared to other countries of the\nsample. The exception is _Morocco_, whose various measures of firms’ productive performance\nalways rank close to the one of the most productive firms in the sample. An originality of our\napproach has also been to generate a few composite indicators of investment climate using\nPrincipal Component Analysis ( _PCA_ ), which summarizes well the key dimensions of the\ninvestment climate. This has allowed as well tackling the problem of multicolinearity when\nexplaining firm productive performances with a wide range of correlated _IC_ variables. We\ndefine four dimensions of the investment climate: the Quality of Infrastructure ( _Infra_ ), the\nBusiness-Government Relations ( _Gov_ ), the Human Capacity ( _H_ ), and the Financing\nConstraints ( _Fin_ ). We use, as well, city or region-sector averages to reduce the endogeneity\nproblem underlying the investment climate ( _IC_ ) variables. The analysis finally shows that\ninvestment climate matters for firms’ productive performances. This has been done by\nestimating an efficiency\n\nMention: \"the World Bank firm-surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Investment climate and productivity are linked using these firm data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004065:16:1", "occurrence_id": "c90dedb9b4231307d1a571c9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  In fact,\nMozambique has the lowest share of professionals and of those with more than 12 years\nof education in the manufacturing workforce.\n\n\n**_Table 2: Professionals and Tertiary Educated Workers in Manufacturing Firms_**\n\n\n\nAverage Share of\nProfessionals in Total\n\nWorkers\n\n\n\nAverage Percentage of\nWorkers with Tertiary\n\nEducation\n\n\n\n**Mozambique** **2.7%** **1.6%**\nMauritius 2.8% 3.9%\nEthiopia 3.5%\nEritrea 4.1%\nNigeria 4.3%\nSouth Africa 4.6% 11.5%\nMalawi 4.6%\nKenya 4.6% 15.0%\nZambia 5.2% 14.7%\nMadagascar 6.1% 12.0%\nUganda 6.9% 35.4%\nSenegal 6.9%\nMali 7.3% 11.5%\nBenin 8.1% 15.6%\n<u>Tanzania</u> <u>11.9%</u>\nSource: Investment Climate Surveys.\nNote: The surveys were conducted for different countries in different years.\n\n\nA more disaggregated comparison, at the level of manufacturing sub-sectors, is\npossible with South Africa. Even though the definitions of the skill categories are not the\nsame across the two countries, the differences in skill intensity in each of the sub-sectors\nshown in Table 3 are stark.\n\n\n15\n\nMention: \"Investment Climate Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Country shares are reported from these surveys, conducted in different years.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004049:3:0", "occurrence_id": "a61f59336cf0d02b9bedb1ce", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **I.** **INTRODUCTION**\n\n\nCompetition is the single most important factor to contain pubic procurement costs and\n\n\nperhaps discourage collusive bidding behavior and corrupt practices. However, many\n\n\ninfrastructure projects seem to continue faced with considerable difficulties to intensify\n\n\ncompetition in the procurement process (Estache and Iimi, 2008a). This may be partly\n\n\nunavoidable because infrastructure projects are normally extremely valuable and highly\n\ncustomized. <sup>1</sup> However, there are still several simple measures that could potentially extend\n\n\nmarket competition but sometimes have not been exploited carefully.\n\n\nThe current paper revisits the problem of endogenous participation in auctions and the\n\n\ndivision into procurement lots in the water supply and sewage industry. In this sector, it is\n\n\nalways of particular concern that the observed degree of competition has been limited. In the\n\n\npublic-private partnership (PPP) context, less than five multinational water enterprises may\n\nhave been dominant over the world (Foster, 2005). <sup>2</sup> [^2: According to the Private Participation in Infrastructure (PPI) database, the top 10 percent largest firms—\ndefined by the number of transactions that each company obtained—were awarded about half of total\ninfrastructure PPP contracts. In the water industry, especially, a multinational French water service operator,\nVeolia Environnement (former Vivendi Environnement, or Compagnie Générale des Eaux), was alone awarded] Even in the traditional public\n\n\nprocurement context, the degree of competition is in general limited (Estache and Iimi,\n\n\n2008a).\n\n\nThere is a tradeoff between competition in auctions and size of contracts. Not many firms can\n\n\napply for highly valuable contracts, because of financial and experiential constraints. Highly\n\n\nspecific works (which are normally valuable as well) would also limit competition. On the\n\n\nother hand, infrastructure projects are normally expected to exhibit economies of scale and\n\n\npossibly economies of scope in preparation and implementation. Therefore, larger lot works\n\n\ncould be procured at relatively low costs per unit, though fewer applicants may participate.\n\n\n1 Estache and Iimi (\n\nMention: \"Private Participation in Infrastructure (PPI) database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It shows the ten percent largest firms winning most transactions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004067:12:0", "occurrence_id": "7abe3497a40f0c4424f63c2e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Whom South Africa, which reports all South African companies with operations in other African\n\n\ncountries at the end of 2006. This publication enumerates the subsidiaries and operations of each\n\n\nSouth African multinational and reports partial information on the date of initial investment, hold\n\nings, and number of employees in each subsidiary. From it, all rms listed in the JSE, and with\n\n\noperations or subsidiaries in African HIPC eligible countries during the period of the initiatives\n\n\n(1995-2006) were selected, to obtain a sample of 35 companies with 187 subsidiaries in 26 coun\n\ntries. This information was complemented and checked with data from United Nations Conference\n\n\non Trade and Development (1993, 2004), Graham and Whiteside (1997-2004), Lexis-Nexis (2007),\n\n\nbusiness-press reports relating parent and subsidiaries in Lexis-Nexis and Factiva-Newsplus, and in\n\nformation requests directly sent to South African companies identi\u001ced as having a\u001eliates in eligible\n\nAfrican HIPC countries. <sup>9</sup>\n\n\nThe date of initial investment in subsidiaries located in African HIPC countries reveals whether\n\n\na parent company is a\u001bected by a speci\u001cc debt relief event. For instance, only companies with active\n\n\nsubsidiaries in African HIPC-eligible countries in June 1996 are considered as a\u001bected by the original\n\n\nHIPC initiative announcement. Table 2 presents the list of parent companies with investments in\n\n\nAfrican countries that have been eligible for the HIPC initiative since 1996 and the median size\n\n\nin assets and sales of these companies at the end of 2006. The nal row of the table displays the\n\n\naverage of each measure across companies. Both assets and sales of parents are at about US$ 2 bn,\n\n\nand the average parent company has about 4 subsidiaries. Parent companies with interests in HIPC\n\n\ncountries are also homogeneously distributed across industries, as shown in Table 3. Utilities are\n\n\nthe only industry where no South\n\nMention: \"data from United Nations Conference\n\n\non Trade and Development\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its records complement and check the company subsidiary data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004067:12:0", "occurrence_id": "f3893e6b40499515419803c5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Whom South Africa, which reports all South African companies with operations in other African\n\n\ncountries at the end of 2006. This publication enumerates the subsidiaries and operations of each\n\n\nSouth African multinational and reports partial information on the date of initial investment, hold\n\nings, and number of employees in each subsidiary. From it, all rms listed in the JSE, and with\n\n\noperations or subsidiaries in African HIPC eligible countries during the period of the initiatives\n\n\n(1995-2006) were selected, to obtain a sample of 35 companies with 187 subsidiaries in 26 coun\n\ntries. This information was complemented and checked with data from United Nations Conference\n\n\non Trade and Development (1993, 2004), Graham and Whiteside (1997-2004), Lexis-Nexis (2007),\n\n\nbusiness-press reports relating parent and subsidiaries in Lexis-Nexis and Factiva-Newsplus, and in\n\nformation requests directly sent to South African companies identi\u001ced as having a\u001eliates in eligible\n\nAfrican HIPC countries. <sup>9</sup>\n\n\nThe date of initial investment in subsidiaries located in African HIPC countries reveals whether\n\n\na parent company is a\u001bected by a speci\u001cc debt relief event. For instance, only companies with active\n\n\nsubsidiaries in African HIPC-eligible countries in June 1996 are considered as a\u001bected by the original\n\n\nHIPC initiative announcement. Table 2 presents the list of parent companies with investments in\n\n\nAfrican countries that have been eligible for the HIPC initiative since 1996 and the median size\n\n\nin assets and sales of these companies at the end of 2006. The nal row of the table displays the\n\n\naverage of each measure across companies. Both assets and sales of parents are at about US$ 2 bn,\n\n\nand the average parent company has about 4 subsidiaries. Parent companies with interests in HIPC\n\n\ncountries are also homogeneously distributed across industries, as shown in Table 3. Utilities are\n\n\nthe only industry where no South\n\nMention: \"Graham and Whiteside\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its compilation over this window serves the same checking role.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004067:12:0", "occurrence_id": "73e7acca3a2b29b88ba1dcd3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Whom South Africa, which reports all South African companies with operations in other African\n\n\ncountries at the end of 2006. This publication enumerates the subsidiaries and operations of each\n\n\nSouth African multinational and reports partial information on the date of initial investment, hold\n\nings, and number of employees in each subsidiary. From it, all rms listed in the JSE, and with\n\n\noperations or subsidiaries in African HIPC eligible countries during the period of the initiatives\n\n\n(1995-2006) were selected, to obtain a sample of 35 companies with 187 subsidiaries in 26 coun\n\ntries. This information was complemented and checked with data from United Nations Conference\n\n\non Trade and Development (1993, 2004), Graham and Whiteside (1997-2004), Lexis-Nexis (2007),\n\n\nbusiness-press reports relating parent and subsidiaries in Lexis-Nexis and Factiva-Newsplus, and in\n\nformation requests directly sent to South African companies identi\u001ced as having a\u001eliates in eligible\n\nAfrican HIPC countries. <sup>9</sup>\n\n\nThe date of initial investment in subsidiaries located in African HIPC countries reveals whether\n\n\na parent company is a\u001bected by a speci\u001cc debt relief event. For instance, only companies with active\n\n\nsubsidiaries in African HIPC-eligible countries in June 1996 are considered as a\u001bected by the original\n\n\nHIPC initiative announcement. Table 2 presents the list of parent companies with investments in\n\n\nAfrican countries that have been eligible for the HIPC initiative since 1996 and the median size\n\n\nin assets and sales of these companies at the end of 2006. The nal row of the table displays the\n\n\naverage of each measure across companies. Both assets and sales of parents are at about US$ 2 bn,\n\n\nand the average parent company has about 4 subsidiaries. Parent companies with interests in HIPC\n\n\ncountries are also homogeneously distributed across industries, as shown in Table 3. Utilities are\n\n\nthe only industry where no South\n\nMention: \"Lexis-Nexis\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Business press reports of parents and subsidiaries are drawn from it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004070:4:0", "occurrence_id": "8f2f572c307ebc5fc81b0692", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: periods. Because of limitations in the data on human capital related to education (HS), the\nintangible capital residual (HR) is used as a proxy measure of human capital in this study.\nHenceforth, HR would be referred in this paper as human and intangible capital. The different\ntypes of capital are determined by a number of factors, such as household income, community\ninfrastructures and strength of institutions (see Appendix 1). In addition, we include the\nmagnitude of natural disasters as a determinant in order to examine its impacts on capital. The\nnumber of occurrences of natural disasters and the extent of damages are described below.\n\n**III.** **Natural disasters**\n\n\nFour types of natural disasters are considered in the study: droughts, earthquakes, floods, and\nhurricanes/storms. Data on the number of natural disasters are obtained for 196 countries from\nthe Emergency Events Database (EM-DAT) <sup>1</sup> [^1: http://www.emdat.be/] . _Drought_ is characterized by a shortage in a\nregion's water supply as a result of constantly below average precipitation. _Earthquake_ is\ncharacterized by the shaking and displacement of ground due to seismic waves. This variable\nrefers to the occurrences of earthquakes only without secondary effects. _Flood_ is defined by a\nsignificant rise of water level in a stream, lake, reservoir or coastal region. _Storm_ is represented\nby wind with a speed between 48 and 55 knots.\n\n\nTable 1 shows a maximum of 196 countries for which data are available, with a total of 55\nevents of drought, 82 events of earthquake, 447 events of flood and 303 events of storm that\nstarted during the three given periods. A natural disaster, such as drought, may last for several\nyears but the data presented here refer explicitly on the tally of events in a single year, i.e., the\nyear that a natural disaster started. Table 2 presents the countries with more than 3 reported\nevents of\n\nMention: \"Emergency Events Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Disaster counts across countries are obtained from this register.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004072:24:0", "occurrence_id": "933b3c553c91397f05005d53", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Another route by which health affects schooling is orphanhood. This has received a great deal of\nattention in the literature on the economic effects of AIDS. If orphans receive less education,\nthen the intergenerational transmission of human capital can be interrupted, with important, and\npotentially disastrous, long-term effects (Bell, Devarajan, and Gersbach 2003). Case, Paxson,\nand Ableidinger (2002) use demographic and health surveys across 10 Sub-Saharan countries to\nexamine the impact of orphanhood and find that orphans are significantly less likely than other\nchildren to be enrolled in school. In this study, however, the repeated cross-sectional nature of\nthe data means that the interpretation of the results is not without ambiguity. Gertler, Levine, and\nAmes (2004) use panel data from Indonesia and find that a parental death doubles the probability\nthat a child will drop out of school the same year. Neither of these two studies finds a gender\neffect, either at the parent or child level. Other studies find little impact of parental death on\nschooling, possibly because members of extended families take on the parenting function\n(Ainsworth, Beegle, and Koda 2002; Kamali and others 1996; Lloyd and Blanc 1996). <sup>8</sup>\n\nConsistent with this view, Fortson (2006: 26) reports that children in areas in southern Africa\nwith high HIV prevalence are “less likely to attend school, [are] less likely to complete primary\nschool, and progress more slowly through school.” Fortson shows that more than half of this\nimpact on schooling can be attributed to the expectation of a shorter life of the parent and not to\norphanhood itself; all children do badly when adults expect to die sooner. Reductions in adult\nmortality might lead to greater investment in children’s education, because of higher demand\neither by parents or by children themselves, who expect to reap the future returns for longer\n\nMention: \"demographic and health surveys\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Ten countries' rounds support the orphanhood and schooling analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004072:24:0", "occurrence_id": "b0628d76a9c8955b91ebd826", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Another route by which health affects schooling is orphanhood. This has received a great deal of\nattention in the literature on the economic effects of AIDS. If orphans receive less education,\nthen the intergenerational transmission of human capital can be interrupted, with important, and\npotentially disastrous, long-term effects (Bell, Devarajan, and Gersbach 2003). Case, Paxson,\nand Ableidinger (2002) use demographic and health surveys across 10 Sub-Saharan countries to\nexamine the impact of orphanhood and find that orphans are significantly less likely than other\nchildren to be enrolled in school. In this study, however, the repeated cross-sectional nature of\nthe data means that the interpretation of the results is not without ambiguity. Gertler, Levine, and\nAmes (2004) use panel data from Indonesia and find that a parental death doubles the probability\nthat a child will drop out of school the same year. Neither of these two studies finds a gender\neffect, either at the parent or child level. Other studies find little impact of parental death on\nschooling, possibly because members of extended families take on the parenting function\n(Ainsworth, Beegle, and Koda 2002; Kamali and others 1996; Lloyd and Blanc 1996). <sup>8</sup>\n\nConsistent with this view, Fortson (2006: 26) reports that children in areas in southern Africa\nwith high HIV prevalence are “less likely to attend school, [are] less likely to complete primary\nschool, and progress more slowly through school.” Fortson shows that more than half of this\nimpact on schooling can be attributed to the expectation of a shorter life of the parent and not to\norphanhood itself; all children do badly when adults expect to die sooner. Reductions in adult\nmortality might lead to greater investment in children’s education, because of higher demand\neither by parents or by children themselves, who expect to reap the future returns for longer\n\nMention: \"panel data from Indonesia\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Parental death doubles dropout probability in this panel.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004073:27:1", "occurrence_id": "a53c50d367a77a80f9800460", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  South Africa in 2006_\n\n\nUS$ & Rand,\nMTN, Ltd. Telecoms 3,467.9 3 tranches\n\n\n\n5 yrs. (2\ntr.); 3 yrs.\n(1 tr.)\n\n\n\nLIBOR+65BP;\nLIBOR+90BP;\nLIBOR+60BP\n\n\n\nRand, 1\nVodacom Group Telecoms 1,137.6 tranche 5 years Undisclosed\n\n\n\nTrans African\nConcessions\n\n\n\nConstruction Rand, 2\nof toll road 475.3 tranches\n\n\n\n1 yr & 20\nyrs. Undisclosed 3 local banks 475.3\n\n\n\nSource: Dealogic Loanware.\na. Where this is disclosed. Note that actual participation by each bank in a given loan transaction is not always provided to Dealogic given\nvariance in disclosure practices across bank syndicates.\n\n\nThe proportion of syndicated loans for infrastructure denominated in local currencies has been rising\nin the focus countries since 2000. Sixty-six percent of the total syndicated lending to infrastructure in\n2006, as tracked by Loanware, was denominated in U.S. dollars ($773 million). <sup>22</sup> [^22: A large share of foreign-currency-denominated or foreign-indexed debt has been identified as a characteristic that\ncan render an emerging market economy vulnerable to capital flow reversals and financial crisis (see, for example,\nGoldstein and Turner 1996).] The remaining $404.8\nmillion (just over one-third) was denominated in local currency. In contrast, none of the syndicated loans\narranged in 2000 for borrowers in the 23 countries was denominated in local currency, and only 17\npercent ($77.2 million, for a loan to cellular-phone company Celtel Nigeria) in 2005 was denominated in\nlocal currency. None of the syndicated loans for infrastructure purposes transacted in 2006 for borrowers\nin Chile and Malaysia was denominated in local currency. These last mentioned loans were U.S.-dollardenominated with the exception of a $343.5 million U.S.-dollar–equivalent loan denominated in\nSingapore dollars.\n\n\nA few of the recent syndicated loans to the focus countries have been specially structured to reduce\nthe risks of currency\n\nMention: \"Dealogic Loanware\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Loan tranches, tenors and syndicate members are compiled in this deal database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004075:21:0", "occurrence_id": "3e44eb54ab8e0f0f56fc7d82", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 8: Frequency distribution of average annual growth rates in service coverage**\n\n**(rural)**\n\n\n\n50%\n\n\n40%\n\n\n30%\n\n\n20%\n\n\n10%\n\n\n0%\n\n\n\n100%\n\n\n80%\n\n\n60%\n\n\n40%\n\n\n20%\n\n\n0%\n\n\n\nPiped\nWater\n\n\n\nFlush\nToilet\n\n\n\nElectricity Landline\n\n\n\nAbove demographic\n\ngrowth\n\n\n\n<0% pa 0-4% pa 4-8% pa 8-12% pa >12% pa\n\n\n\nBelow demographic\n\ngrowth\n\n\n\nPiped water Flush toilet Electricity Landline\n\n\nSource: Authors, using AICD DHS/MICS Survey Database, 2007.\n\n\n**Figure 9: Estimated year of universal coverage under business as usual**\n\n\n80%\n\n\n60%\n\n\n40%\n\n\n20%\n\n\n0%\n\nBy 2050 2050-2100 2100-2200 After 2200\n\n\nPiped Water Flush Toilet Electricity Landline\n\n\nSource: Authors, using AICD DHS/MICS Survey Database, 2007.\n\n\n\n20\n\nMention: \"AICD DHS/MICS Survey Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Coverage of water, sanitation and electricity is extrapolated from these harmonised surveys.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004076:9:0", "occurrence_id": "c4e50ebf3a32010134abd6bc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: quintile of wealth to which the household belongs, and the leave-out take-up rate in the primary\n\n\nsampling unit where the household lives.\n\n\nThe index of wealth is estimated using factor analysis because we do not have household\n\n\nincome or expenditure data in the DHS. The variables used for the factor analysis are allowed to\n\n\ndiffer between countries depending on the data available in each survey so as to maximize the\n\n\ninformation used. In practice, the variables used include housing variables, variables on the\n\n\naccess to various types of provision for basic infrastructure services (there is a slight issue of\n\n\nendogeneity here, since we are modeling take-up of utility services, but it is minor given the\n\n\nmany other variables included in the index), and variables on a range of assets owned.\n\n\nThe regressions on take-up of service are estimated only on the samples of households\n\n\nwho live in neighborhoods where there is access, and the estimation follows a simple probit\n\n\nprocedure. The regressions are not presented here, as there are many of them, but they are rather\n\n\nstraightforward. The leave-out access rate is meant to capture the general conditions of the\n\n\nneighborhood (including factors such as the average distance from the water pipes or electic\n\n\nlines), while the wealth index quintiles are used to deal with the affordability issue.\n\n\nOnce the regressions have been estimated, we simulate what the access rate would be if\n\n\nall households living in areas where there is access would be lifted in terms of wealth from\n\n\nwherever they are in the distribution of wealth to the top wealth quintile. That is, we simulate\n\n\nwhat the take-up rate would be for all households living in primary sampling units where there is\n\n\naccess based on what the behavior of the households would be if they were in the top quintile,\n\n\nwhich corresponds implicitly to an assumption of no affordability problem, since the households\n\n\nin the top quintile should be able to afford the cost of piped water and electricity services. When\n\n\naggregating the results for\n\nMention: \"DHS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Lacking income data here, wealth is built by factor analysis instead.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004077:16:1", "occurrence_id": "da4d96f20045b857e550ac18", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  which consists of disbursements of\nloans made on concessional terms (net of repayments of principal) and grants by official agencies of the\nmembers of the Development Assistance Committee (DAC), by multilateral institutions, and by non-DAC\ncountries to promote economic development and welfare in countries and territories in the DAC list of\nODA recipients. Data source for the variable is World Development Indicators, World Bank.\n\n\n15\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Growth and welfare for aid-recipient countries come from this compilation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004078:6:1", "occurrence_id": "beac808e800194d65de8b290", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  consider the fertility response to natural\n\n\ndisasters often account for risk (Cain 1981; Cain 1983; Cain 1986; Pörtner 2008): the seasonality\n\n\nof flooding and hurricanes makes the concept of risk the key instrument. In the case of\n\n\nearthquakes, which are neither seasonal nor predictable in magnitude and epicenter, the concept\n\n\nof a shock is the dominant instrument.\n\n\nIn India, according to the EM-DAT data, the 2001 Gujarat earthquake was the largest recorded in\n\n\nterms of casualties since data are available in 1900. The second largest was in 1905 in Kangra,\n\n\nsome 1000km north of Gujarat along the Himalya that killed 20,000 people. Gujarat is subject to\n\n\nseismic activity, but it is not along the Himalaya mountain range which is the region most\n\n\nvulnerable to seismic activity. The last large earthquake in Gujarat was in 1956, more than a\n\n\n5\n\nMention: \"the EM-DAT data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It ranks the earthquake as the largest by casualties since records begin.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004079:20:2", "occurrence_id": "b229bd32408523fde5aeb427", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .86\nEconomic empowerment 0.04 -0.39 0.68 -0.39 0.85 -0.34 0.88 -0.43 0.53\nPolitical participation 0.05 0.03 0.08 0.03 0.09 0.04 0.13 0.03 0.05\n**_Household Consumption, Income (level), and Assets (change)_**\nSpending p.c. (Rs/year) 6951 6871 7072 6325 6741 6848 7241 6851 7169\nIncome p.c. (Rs/year) 5858 6795 5462 5843 5163 6104 5434 6168 5673\nAsset (Rs) 2532 2278 4406 2806 2502 2806 2502 3253 1685\n**_Nutritional Intake per capita (Level)_**\nEnergy (cals/day) 2052 2031 2104 1956 2075 2031 2097 2046 2095\nProtein intake (g/day) 47 46 49 44 49 45 50 46 48\n<u>Number of observations</u> <u>6340</u> <u>3824</u> <u>2516</u> <u>583</u> <u>656</u> <u>1149</u> <u>502</u> <u>1446</u> <u>1358</u>\n_Source:_ Own computation from CESS impact evaluation survey.\n\n\n18\n\nMention: \"CESS impact evaluation survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Sample sizes by treatment arm are computed from this evaluation survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004086:5:0", "occurrence_id": "5497feb82b6f7f3fb47d4002", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: provides a succinct description of the so-called Second Great Migration including information\n\n\nabout the historical context, the structure of immigrant flows by country of origin, and the most\n\n\nsalient measures taken by the US immigration regulators in order to shape migration according to\n\n\ntheir policy objectives.\n\n\nA wide scale phenomenon influenced by complex factors and having equally complex\n\n\nimplications, the US immigration story became the topic of a very rich economic literature\n\n\nattempting to answer the endless list of questions posed by migration researchers. In what\n\nfollows I will focus on the part related to the performance of male immigrants <sup>3</sup> and the\n\n\ndeterminants of their performance. Extensive surveys of the existing literature are available in\n\n\nBorjas (1994) and LaLonde and Topel (1997).\n\n\nThe research focusing on immigrants’ assimilation is an arena of strong debates. One\n\n\nexplanation for the lack of consensus on many topics in the literature can be found in the quality\n\n\nof data. No single source provides comprehensive information on immigrants, therefore authors\n\n\ncompensate by imposing various assumptions and implementing different methodologies which\n\n\noften lead to conclusions that significantly differ from each other. The US Census, although most\n\nwidely used <sup>4</sup> because of the richness in information, fails to distinguish between legal and illegal\n\n\nimmigrants, doesn’t account for return migration, and formulates questions essential for\n\n\nmigration research ambiguously. To benefit from yearly flows of data, some authors use instead\n\n\nthe Current Population Survey (CPS), while warning readers about the little reliability of their\n\nsmall sample-based estimates. <sup>5</sup> Alternatively, Immigration and Naturalization Service statistics\n\n\nare more accurate with respect to indicators of legal status, but fail to provide a clear picture of\n\nthe individual ones, such as immigrants’ income. <sup>6</sup> Longitudinal studies, although most conducive\n\nto comprehensive research, have often a restricted coverage, <sup>7</\n\nMention: \"US Census\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Rich in information but unable to separate the categories of interest.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004086:5:0", "occurrence_id": "9579e62c1d3a78a0b934860f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: provides a succinct description of the so-called Second Great Migration including information\n\n\nabout the historical context, the structure of immigrant flows by country of origin, and the most\n\n\nsalient measures taken by the US immigration regulators in order to shape migration according to\n\n\ntheir policy objectives.\n\n\nA wide scale phenomenon influenced by complex factors and having equally complex\n\n\nimplications, the US immigration story became the topic of a very rich economic literature\n\n\nattempting to answer the endless list of questions posed by migration researchers. In what\n\nfollows I will focus on the part related to the performance of male immigrants <sup>3</sup> and the\n\n\ndeterminants of their performance. Extensive surveys of the existing literature are available in\n\n\nBorjas (1994) and LaLonde and Topel (1997).\n\n\nThe research focusing on immigrants’ assimilation is an arena of strong debates. One\n\n\nexplanation for the lack of consensus on many topics in the literature can be found in the quality\n\n\nof data. No single source provides comprehensive information on immigrants, therefore authors\n\n\ncompensate by imposing various assumptions and implementing different methodologies which\n\n\noften lead to conclusions that significantly differ from each other. The US Census, although most\n\nwidely used <sup>4</sup> because of the richness in information, fails to distinguish between legal and illegal\n\n\nimmigrants, doesn’t account for return migration, and formulates questions essential for\n\n\nmigration research ambiguously. To benefit from yearly flows of data, some authors use instead\n\n\nthe Current Population Survey (CPS), while warning readers about the little reliability of their\n\nsmall sample-based estimates. <sup>5</sup> Alternatively, Immigration and Naturalization Service statistics\n\n\nare more accurate with respect to indicators of legal status, but fail to provide a clear picture of\n\nthe individual ones, such as immigrants’ income. <sup>6</sup> Longitudinal studies, although most conducive\n\nto comprehensive research, have often a restricted coverage, <sup>7</\n\nMention: \"Current Population Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its yearly flows are used despite unreliable small-sample estimates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004086:5:0", "occurrence_id": "b4d0d117acbe2f4606f35140", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: provides a succinct description of the so-called Second Great Migration including information\n\n\nabout the historical context, the structure of immigrant flows by country of origin, and the most\n\n\nsalient measures taken by the US immigration regulators in order to shape migration according to\n\n\ntheir policy objectives.\n\n\nA wide scale phenomenon influenced by complex factors and having equally complex\n\n\nimplications, the US immigration story became the topic of a very rich economic literature\n\n\nattempting to answer the endless list of questions posed by migration researchers. In what\n\nfollows I will focus on the part related to the performance of male immigrants <sup>3</sup> and the\n\n\ndeterminants of their performance. Extensive surveys of the existing literature are available in\n\n\nBorjas (1994) and LaLonde and Topel (1997).\n\n\nThe research focusing on immigrants’ assimilation is an arena of strong debates. One\n\n\nexplanation for the lack of consensus on many topics in the literature can be found in the quality\n\n\nof data. No single source provides comprehensive information on immigrants, therefore authors\n\n\ncompensate by imposing various assumptions and implementing different methodologies which\n\n\noften lead to conclusions that significantly differ from each other. The US Census, although most\n\nwidely used <sup>4</sup> because of the richness in information, fails to distinguish between legal and illegal\n\n\nimmigrants, doesn’t account for return migration, and formulates questions essential for\n\n\nmigration research ambiguously. To benefit from yearly flows of data, some authors use instead\n\n\nthe Current Population Survey (CPS), while warning readers about the little reliability of their\n\nsmall sample-based estimates. <sup>5</sup> Alternatively, Immigration and Naturalization Service statistics\n\n\nare more accurate with respect to indicators of legal status, but fail to provide a clear picture of\n\nthe individual ones, such as immigrants’ income. <sup>6</sup> Longitudinal studies, although most conducive\n\nto comprehensive research, have often a restricted coverage, <sup>7</\n\nMention: \"Immigration and Naturalization Service statistics\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"More accurate on legal status but silent on the flows themselves.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004094:5:2", "occurrence_id": "f60a4363e0f36ef29acfd410", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  followed in each country\ncovered in the survey, can be found in the EU-SILC user database description. <sup>3</sup> All 25 EU member\ncountries were covered in the 2006 EU-SILC round. <sup>4</sup> In addition to several key variables of interest\nnoted above, the EU-SILC dataset also includes two variables of particular interest for our analysis:\n\n\n3 UDB description ver 2005-4 from 15-09-07; <u>UDB description ver 2006-1 from 01-03-08, Eurostat, Directorate F: Social</u>\nStatistics and Information Society, Unit F-3: Living conditions and social protection statistics, Luxembourg.\n4 Bulgaria and Romania were not yet members of the European Union in 2006. In addition, Germany and Malta were not\npart of the database we received from Eurostat, so are excluded from the analysis.\n\n\n5\n\nMention: \"EU-SILC dataset\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Two variables of particular interest come alongside the key ones in this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004094:5:2", "occurrence_id": "40c3a5823ec4a79172af9939", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  followed in each country\ncovered in the survey, can be found in the EU-SILC user database description. <sup>3</sup> All 25 EU member\ncountries were covered in the 2006 EU-SILC round. <sup>4</sup> In addition to several key variables of interest\nnoted above, the EU-SILC dataset also includes two variables of particular interest for our analysis:\n\n\n3 UDB description ver 2005-4 from 15-09-07; <u>UDB description ver 2006-1 from 01-03-08, Eurostat, Directorate F: Social</u>\nStatistics and Information Society, Unit F-3: Living conditions and social protection statistics, Luxembourg.\n4 Bulgaria and Romania were not yet members of the European Union in 2006. In addition, Germany and Malta were not\npart of the database we received from Eurostat, so are excluded from the analysis.\n\n\n5\n\nMention: \"database we received from Eurostat\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Two countries absent from the received extract are dropped from the sample.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004103:9:0", "occurrence_id": "42c4d8b2632b6280d7e8f094", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 8\n\n\nThe module provides detailed information on migration histories of all current and former\n\n\nhousehold members, as well as information on the household migration networks abroad. Lastly,\n\n\nthe survey supplies community-level data, which include information on prices, access to\n\n\nservices and infrastructure in the locality.\n\n\nOur sample of interest is 9,565 individuals in the working age category of [15-64] in\n\n2005, excluding return migrants that have been in Albania less than 12 months. <sup>5</sup> The ALSMS05\n\n\ncollected data on _initial_ and _2005_ employment outcomes according to the three digit codes from\n\n\nthe 1988 International Standard Classification of Occupations (ISCO). The three digit codes can\n\nbe classified under 10 major occupational groups <sup>6</sup>, which can be collapsed into 5 distinct\n\n\ncategories. These categories are namely (i) agricultural (major group 6), (ii) low-skilled blue\n\n\ncollar (major groups 8 and 9), (iii) high-skilled blue collar (major group 7), (iv) low-skilled white\n\n\ncollar (major groups 4 and 5), and (v) high-skilled white collar (major groups 1, 2 and 3)\n\noccupations. <sup>7</sup> Although agricultural occupations are traditionally included as part of the high\n\n\nskilled blue collar category under the ISCO definition, we found it appropriate to create a unique\n\n\ncategory for them, as agriculture still remains the principal source of employment in Albania.\n\n\n_3.1. Occupational Ranking_\n\n\nAs our objective is to test the hypothesis of upward occupational mobility that is induced\n\n\nby past migration experience, it is essential to first rank occupational groups in a reasonable and\n\n\nsystematic manner. Following Sicherman and Galor (1990), we rank occupational groups\n\n\naccording to the average level of human capital necessary to be employed in each of them. The\n\n\n5 In our sample, there were\n\nMention: \"ALSMS05\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Initial and current employment outcomes are coded at three digits here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004108:16:0", "occurrence_id": "547147e6291b3fb0828895b3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: PAYING THE PRICE FOR UNRELIABLE POWER SUPPLIES:\nGENERATION OF ELECTRICITY BY PRIVATE FIRMS IN AFRICA\n\n\nshare of own generation. The share of manufacturing in GNI was positively associated with the degree of\nown generation. <sup>17</sup> [^17: The estimated coefficients reflect correlations, not causalities. Further analysis of the estimated relationship is\nimpeded by the simultaneity of the choice of own-generation and the explanatory variables.]\n\n\nThe World Bank’s enterprise surveys offer clues to why firms elect to generate their own power.\nBecause deficiencies in the provision of public power impede business operations, firms install their own\ngenerating equipment to protect themselves against power outages. <sup>18</sup> [^18: Statistics from the Enterprise Surveys on power outages and the adoption of own-generation appear in appendix 1,\ntables 2a to 2f.]\n\n\nPower outages have a significant effect on business operations. Firms that experienced few power\noutages (fewer than 30 days per annum) tended to declare that electricity was not an obstacle or was only\na minor obstacle for the operation and growth of their businesses. On the other hand, firms that suffered\nfrequent outages (more than 60 days per annum) tended to believe that electricity was a major or very\nsevere obstacle for the operation and growth of their businesses.\n\n\nFrequent power outages result in significant losses for enterprises in terms of forgone sales and\ndamaged equipment. Firms experiencing frequent power outages (more than 60 days per year) lose 10–12\npercent of their sales (figure 17). This is about twice as high as the figure for firms that have few outages\n(less than 15 days per year). Equipment damage traceable to power outages is about twice as high for\nfirms that suffer frequent outages as for firms that suffer few outages (1 percent of sales compared to 0.5\npercent of sales) (figure 18).\n\n\n\n**Figure 17 Sales lost owing to power outages**\n\n\nPercent\n\n\n**14.00**\n\n\n**12.00*\n\nMention: \"enterprise surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"They explain why firms choose to generate their own power.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004122:61:0", "occurrence_id": "34fa3bcf4b90b2c047ed8cca", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <u>Table A1.4: Japan; Numbers of tariff headings subject to SSG Actions, 1995-2007</u>\n\n|Fiscal<br>Yeara|Volume-basedb|Col3|Price -based|\n|---|---|---|---|\n||(5-digit<br>headings)|(additional<br>8-digit lines)|(8-digit lines)|\n|1995||5|3|\n|1996||71|0|\n|1997||5|0|\n|1998||3|2|\n|1999|1|2|8|\n|2000|1|6|4|\n|2001||10|0|\n|2002||0|8|\n|2003|6|1|13|\n|2004||0|9|\n|2005||4|6|\n|2006||2|16|\n|2007||5|10|\n\n\n\nSource: Tabulated from national notifications of SSG actions, available on the\nWTO documents website.\n<u>Notes</u>\na The fiscal year runs from 1 April of the year indicated until 31 March of the\nfollowing calendar year; e.g. FY 1995 covers 1 April 1995 – 31 March 1996.\nb In 1996, the large number of tariff lines were for live swine and swine meat.\n1999 and 2000: the 5-digit heading was Inulin. 2003: the 5-digit headings were\nfor butter and other dairy fats and for Maize (corn) starch and other vegetable\nstarches and products.\nThe principal products covered were Milk and products, rice, beans, wheat and\nrice flours and starches, swine and swine meat, raw silk.\n\n\n54\n\nMention: \"national notifications of SSG actions\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Support measures by year are tabulated from these member notifications.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004127:10:0", "occurrence_id": "d92743bcebe0a264992341b5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Table 3. Comparison of propensity score matched (PSM) and treatment regression**\n**<u>(TRE) of average participation effect on the participating households in GMA (ATT)</u>**\n\nPSM <u>Treatment Regression</u>\n<u>Outcome: per capita consumption</u> <u>Estimates</u> <u>Estimates</u> <u></u>\n<u>(1)</u> <u>(2)</u> <u>(3)</u>\nBangweulu -0.074 ** 0.858 *** -0.56 ***\nKafue -0.240 -0.286 0.13\nLower Zambezi -0.118 0.494 0.37\nLuangwa 0.034 0.530 *** -0.42 **\n<u>Overall</u> <u>0.083</u> <u>0.438</u> <u>***</u> <u>-0.30</u> <u>*</u>\nSignificance: *=Significance at 10%; **=Significance at 5%; ***=Significance at 1%\nSource: Authors’ calculations, Data from the IGMAW survey, 2006\n\nAmong the households living in GMAs, some households participate in CRB/VAGs. If\nCRB/VAGs provide broad social infrastructures from the hunting licensing revenue, we\nexpect all households in the GMA to gain in welfare, on average. However, if active\nparticipants in these institutions restrict benefit sharing to themselves, we would expect\nparticipants to gain more in welfare relative to non-participants.\n\nTable 3 shows the impact of participation in CRB/VA\n\nMention: \"IGMAW survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Household participation in community groups is measured from this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004132:8:0", "occurrence_id": "a682321747b84a813e667d2e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 1: Entrepreneurship around the world**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nSource: 2008 World Bank Group Entrepreneurship Survey.\n\n\nIn addition to these indicators, the 2008 _WBGES_ provides enough historical data to\n\n\nestablish the trends in the number of new and total businesses in 100 countries during the\n\n\neight year-period 2000 to 2008. In this regard, it is interesting to note that developing\n\n\ncountries have a higher trend in both the number of total and new businesses, with some\n\n\nof the countries doubling their numbers during the period, whereas industrialized\n\n\ncountries show little variation during the same period (Figure 2). Interestingly, the\n\n\nmajority of the surveyed countries with a negative business growth were those that\n\n\nsuffered a war, civil unrest or remarkable political instability during the eight-year\n\nperiod. <sup>2</sup>\n\n\n2 Such as Lebanon, Bosnia and Herzegovina and Haiti.\n\n\n7\n\nMention: \"2008 World Bank Group Entrepreneurship Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Global new-business counts and their historical trend come from this edition.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004140:1:0", "occurrence_id": "1ca1a1a12ea22d48cdcc90b3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Policy Research Working Paper 4945\n\n### **Abstract**\n\nSince 1990 migration flows from Albania have been\nmassive, relative to the size of the country and its\npopulation, but they have also fluctuated over time. This\npaper presents and discusses various descriptive trends,\nmainly in graphical form. The data come from the\nAlbanian Living Standards Measurement Survey, 2005\nround, and cover the period 1990–2004. The resulting\n\n\n\nobserved trends reflect changing push and pull factors\nin Albania and the two main host countries, Greece\nand Italy. The paper also presents a hazard approach to\nmodeling Albanian emigration and return migration.\nThis analysis highlights, among other things, the\nrelevance of networks in Albanian migration dynamics,\nboth to promote emigration and to delay return.\n\n\n\nThis paper—a product of the Poverty Team, Development Research Group—is part of a larger effort in the department\nto study the impact of migration on poverty and other outcomes in source countries. Policy Research Working Papers\nare also posted on the Web at http://econ.worldbank.org. The authors may be contacted at gcarletto@worldbank.org and\ncazzarri@worldbank.org.\n\n\n_The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development_\n_issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the_\n_names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those_\n_of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and_\n_its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent._\n\n\nProduced by the Research Support Team\n\nMention: \"the\nAlbanian Living Standards Measurement Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Descriptive trends over this fourteen-year period are drawn from this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004141:47:0", "occurrence_id": "57bc3954f2f0518ca5a6dbfc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |Col1|Col2|Col3|luggage, saddlery, harness,<br>footwear|\n|---|---|---|---|\n|Other Manufacturing<br>|Polyester buttons, tobacco,<br>camping equipment, coffee<br>roasting and grinding,<br>plastic products<br>manufacturing, tannery<br>(sheep, goats), printing,<br>non-metal products (glass,<br>rubber), wire and nails,<br>cement production, PP Bag<br>production, and firms<br>classified as “Others” in<br>the manufacturing survey|Other<br>manufacturing<br>and electronics|<br>Other manufacturing**c **|\n\n\n\n**_Constraints_**\n\n\nThe investment climate surveys ask very detailed questions about constraints. These questions overlap\n\nconsiderably, but not perfectly, across the different surveys. To construct constraints measures that are\n\ncomparable across rural and urban enterprises, we categorised individual constraints into coarser\n\ngroups of constraints, documented in the Tables below. A drawback to this procedure is that the\n\ndifferent constraints categories do not contain an equal number of items. Moreover, due to imperfect\n\noverlap, they may not measure exactly the same constructs across rural and urban areas.\n\n\n\n\n\n\n\n\n\n\n\n\n|Most important constraint:    Table B.2: Construction of the “Most Important Constraint” Variable|Col2|Col3|\n|---|---|---|\n|<br> <br>**Table B.2: Construction of the “Most Important Constraint” Variable**<br> <br> <br>|<br> <br>**Table B.2: Construction of the “Most Important Constraint” Variable**<br> <br> <br>|<br> <br\n\nMention: \"manufacturing survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its classification defines the manufacturing categories compared.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004141:47:0", "occurrence_id": "50b440ff12dd4b01b71254ec", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |Col1|Col2|Col3|luggage, saddlery, harness,<br>footwear|\n|---|---|---|---|\n|Other Manufacturing<br>|Polyester buttons, tobacco,<br>camping equipment, coffee<br>roasting and grinding,<br>plastic products<br>manufacturing, tannery<br>(sheep, goats), printing,<br>non-metal products (glass,<br>rubber), wire and nails,<br>cement production, PP Bag<br>production, and firms<br>classified as “Others” in<br>the manufacturing survey|Other<br>manufacturing<br>and electronics|<br>Other manufacturing**c **|\n\n\n\n**_Constraints_**\n\n\nThe investment climate surveys ask very detailed questions about constraints. These questions overlap\n\nconsiderably, but not perfectly, across the different surveys. To construct constraints measures that are\n\ncomparable across rural and urban enterprises, we categorised individual constraints into coarser\n\ngroups of constraints, documented in the Tables below. A drawback to this procedure is that the\n\ndifferent constraints categories do not contain an equal number of items. Moreover, due to imperfect\n\noverlap, they may not measure exactly the same constructs across rural and urban areas.\n\n\n\n\n\n\n\n\n\n\n\n\n|Most important constraint:    Table B.2: Construction of the “Most Important Constraint” Variable|Col2|Col3|\n|---|---|---|\n|<br> <br>**Table B.2: Construction of the “Most Important Constraint” Variable**<br> <br> <br>|<br> <br>**Table B.2: Construction of the “Most Important Constraint” Variable**<br> <br> <br>|<br> <br\n\nMention: \"investment climate surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Their detailed constraint questions overlap but do not match perfectly.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004095:8:0", "occurrence_id": "17d7e6032d356541cd1b8e72", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: According to INSTAT data, private sector employment increased by only 23,000 between 2001 and\n\n\n2004, adding only 2.5% to total employment in this period (WB, 2006). Two main implications of this\n\n\nsituation are the migration of a lot of young men to work abroad and the large withdrawals of women\n\nfrom the labor market <sup>9</sup> [^9: For an in-depth analysis of the mass Albanian migration since 1990, see King, et al. (2005).] . The consequential under-valuation of women’s time has resulted in significant\n\n\ndifferences in the time male and female groups allocate to paid and unpaid work (with women spending\n\n\nan inefficiently high proportion of their time in household production and caring activities, while men\n\n\noverspecialize in labor market activities) (Paci, 2002). While female represent at least half of the\n\n\npopulation in Albania, they account for the 40 percent of the total labor force and face higher rates of\n\nunemployment than men (especially the young) (ILO, 2001). <sup>10</sup> [^10: Between 1980 and 2004, female labor force participation in Albania has not increased much (from 39% to 42%) (see WB\nGenderStats at http://devdata.worldbank.org/genderstats). This rather stable trend challenges the argument that female labor\nsupply in migrant households may be explained by the aggregate national shift toward more working women in the economy.]\n\n\nDriven by economic hardships and geographic proximity, Albania has developed strong migration ties\n\n\nwith other labor markets, in particular Greece and Italy, and remittances play a significant role in\n\n\nAlbanian economy (Coulon and Piracha, 2005; Carletto et al. 2007). Much of migration from Albania\n\n\nshows a stable and common pattern in that it has traditionally been temporary in nature (particularly the\n\n\nflow to neighboring Greece), whether seasonal or circular. The limited empirical evidence available\n\n\nseems to suggest the existence of a “migration cycle”,\n\nMention: \"INSTAT data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Private sector employment growth over this window is measured here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004095:8:0", "occurrence_id": "c60b635c46f819f47840d165", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: According to INSTAT data, private sector employment increased by only 23,000 between 2001 and\n\n\n2004, adding only 2.5% to total employment in this period (WB, 2006). Two main implications of this\n\n\nsituation are the migration of a lot of young men to work abroad and the large withdrawals of women\n\nfrom the labor market <sup>9</sup> [^9: For an in-depth analysis of the mass Albanian migration since 1990, see King, et al. (2005).] . The consequential under-valuation of women’s time has resulted in significant\n\n\ndifferences in the time male and female groups allocate to paid and unpaid work (with women spending\n\n\nan inefficiently high proportion of their time in household production and caring activities, while men\n\n\noverspecialize in labor market activities) (Paci, 2002). While female represent at least half of the\n\n\npopulation in Albania, they account for the 40 percent of the total labor force and face higher rates of\n\nunemployment than men (especially the young) (ILO, 2001). <sup>10</sup> [^10: Between 1980 and 2004, female labor force participation in Albania has not increased much (from 39% to 42%) (see WB\nGenderStats at http://devdata.worldbank.org/genderstats). This rather stable trend challenges the argument that female labor\nsupply in migrant households may be explained by the aggregate national shift toward more working women in the economy.]\n\n\nDriven by economic hardships and geographic proximity, Albania has developed strong migration ties\n\n\nwith other labor markets, in particular Greece and Italy, and remittances play a significant role in\n\n\nAlbanian economy (Coulon and Piracha, 2005; Carletto et al. 2007). Much of migration from Albania\n\n\nshows a stable and common pattern in that it has traditionally been temporary in nature (particularly the\n\n\nflow to neighboring Greece), whether seasonal or circular. The limited empirical evidence available\n\n\nseems to suggest the existence of a “migration cycle”,\n\nMention: \"WB\nGenderStats\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Female participation over this span is reported in this gender series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004158:13:0", "occurrence_id": "be540511398aae77c6077710", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _3.2. Economic Impact Estimation Methodology: Social Accounting Matrix_\n\nVarious methods can be used to estimate higher-order effect of disasters based on\n\ndamages and/or losses data, including input-output (IO) table, social accounting matrix\n\n(SAM), and computable general equilibrium (CGE) model. <sup>15</sup> [^15: A summary and discussion of methodologies on impact estimation can be seen in a companion paper,\n“Critical Review of Methodologies on Disaster Impact Estimation.”] Since the cases for\n\nglobal aggregate include a large group of countries in different years, the data\n\navailability of method becomes one of the key issues for the selection of methodology.\n\nSAM is employed in this study because of its data availability of construction and the\n\n\n\nfamiliarity of use in international development community <sup>16</sup> [^16: Please see Appendix 1 for detailed description of SAM, and a companion paper, ‘Impact Estimation\nMethodology: Case Studies.’]\n\n\n\n.\n\n\n\nSocial accounting matrix (SAM) has been utilized to examine the higher-order\n\neffects across different socio-economic agents, activities, and factors. Notable studies\n\nusing a SAM or one of its variants include Cole (1995, 1998, and 2004) among others.\n\nLike IO models, the SAM approach has rigid coefficients and tends to provide upper\n\nbounds of impact estimates. On the other hand, the SAM framework with certain\n\ndisaggregation, as well as extended IO model and CGE model, can derive the\n\ndistributional impacts of a disaster in order to evaluate equity considerations for public\n\npolicies against disasters. In this paper, SAMs were constructed in an aggregated\n\n\n\nDue to the\n\n\n\nversion for each country and each decade, based on the World Bank data.\n\n\n\n17\n\n\n\nlarge number of SAMs that needed to be constructed, and in order to maintain the\n\nconsistency of the structure and features among them, the SAMs were constructed in the\n\nmost aggregated way—one sector (one value) for each principal account (see the Figure\n\n3-1). This simple structure is also\n\nMention: \"World Bank data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The aggregate matrices for each country and decade are built from these data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004160:14:0", "occurrence_id": "b63b5a8e03807d0b79447a59", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The underlying road database is based on the Digital Chart of the World (DCW),\n\nwhich was expanded using attributes from Michelin 2004 regional maps to distinguish\n\n\n\namong primary, secondary, and tertiary roads. <sup>4</sup> [^4: Thomas (2007) assigns travel along primary roads to be 60km/hr, secondary roads 40km/hr, and tertiary\nroads 20km/hr.]\n\n\n\nIn addition, further refinements to these\n\n\n\nmaps were made through a Uganda national roads dataset in order to allow for more\n\nprecise analysis. This database was then layered with various GPS data in order to extract\n\ntravel time and distance from major cities.\n\n<u>Power supply: A dummy variable showing the presence of an electric grid in</u>\n\ndistrict is used to proxy for access to power supply. This is a crude measure, but we could\n\nnot obtain data on actual usage or reliability of power supply.\n\n<u>Localization: There are several ways of measuring localization. These include</u>\n\nown-industry employment in the region, own-industry establishments in the region, or an\n\nindex of concentration, which reflects disproportionately high concentration of the\n\nindustry in the region in comparison to the nation. We use own-industry establishments\n\nas the measure of choice as we want to test if there are gain from locating in areas that are\n\nalready specialized in the firm’s chosen line of business. Also, when location “shopping”\n\nentrepreneurs are more likely to observe the density of establishments in an area,\n\ncompared to the number of people employed in them. Using the number of firms in the\n\nsame sector has also been commonly used in empirical work, with the underlying premise\n\nthat localization economies come from the absolute volume of similar activities in the\n\nneighborhood.\n\nFor the localization measure, we consider the number of establishments in the\n\nsame 2 digit industry sector who are within 20 kilometers of each firm under\n\n\n\nAs discussed earlier, we only use establishments that have been in\n\n\n\nconsideration.\n\n\n\n5\n\n\n\nbusiness for more than 5 years\n\nMention: \"Digital Chart of the World\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The underlying road network is taken from this chart.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004160:14:0", "occurrence_id": "45124ef00976c8d146956341", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The underlying road database is based on the Digital Chart of the World (DCW),\n\nwhich was expanded using attributes from Michelin 2004 regional maps to distinguish\n\n\n\namong primary, secondary, and tertiary roads. <sup>4</sup> [^4: Thomas (2007) assigns travel along primary roads to be 60km/hr, secondary roads 40km/hr, and tertiary\nroads 20km/hr.]\n\n\n\nIn addition, further refinements to these\n\n\n\nmaps were made through a Uganda national roads dataset in order to allow for more\n\nprecise analysis. This database was then layered with various GPS data in order to extract\n\ntravel time and distance from major cities.\n\n<u>Power supply: A dummy variable showing the presence of an electric grid in</u>\n\ndistrict is used to proxy for access to power supply. This is a crude measure, but we could\n\nnot obtain data on actual usage or reliability of power supply.\n\n<u>Localization: There are several ways of measuring localization. These include</u>\n\nown-industry employment in the region, own-industry establishments in the region, or an\n\nindex of concentration, which reflects disproportionately high concentration of the\n\nindustry in the region in comparison to the nation. We use own-industry establishments\n\nas the measure of choice as we want to test if there are gain from locating in areas that are\n\nalready specialized in the firm’s chosen line of business. Also, when location “shopping”\n\nentrepreneurs are more likely to observe the density of establishments in an area,\n\ncompared to the number of people employed in them. Using the number of firms in the\n\nsame sector has also been commonly used in empirical work, with the underlying premise\n\nthat localization economies come from the absolute volume of similar activities in the\n\nneighborhood.\n\nFor the localization measure, we consider the number of establishments in the\n\nsame 2 digit industry sector who are within 20 kilometers of each firm under\n\n\n\nAs discussed earlier, we only use establishments that have been in\n\n\n\nconsideration.\n\n\n\n5\n\n\n\nbusiness for more than 5 years\n\nMention: \"Michelin 2004 regional maps\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Attributes from these maps classify roads into three classes.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004160:14:0", "occurrence_id": "56c8093553d6b1fb99d51117", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The underlying road database is based on the Digital Chart of the World (DCW),\n\nwhich was expanded using attributes from Michelin 2004 regional maps to distinguish\n\n\n\namong primary, secondary, and tertiary roads. <sup>4</sup> [^4: Thomas (2007) assigns travel along primary roads to be 60km/hr, secondary roads 40km/hr, and tertiary\nroads 20km/hr.]\n\n\n\nIn addition, further refinements to these\n\n\n\nmaps were made through a Uganda national roads dataset in order to allow for more\n\nprecise analysis. This database was then layered with various GPS data in order to extract\n\ntravel time and distance from major cities.\n\n<u>Power supply: A dummy variable showing the presence of an electric grid in</u>\n\ndistrict is used to proxy for access to power supply. This is a crude measure, but we could\n\nnot obtain data on actual usage or reliability of power supply.\n\n<u>Localization: There are several ways of measuring localization. These include</u>\n\nown-industry employment in the region, own-industry establishments in the region, or an\n\nindex of concentration, which reflects disproportionately high concentration of the\n\nindustry in the region in comparison to the nation. We use own-industry establishments\n\nas the measure of choice as we want to test if there are gain from locating in areas that are\n\nalready specialized in the firm’s chosen line of business. Also, when location “shopping”\n\nentrepreneurs are more likely to observe the density of establishments in an area,\n\ncompared to the number of people employed in them. Using the number of firms in the\n\nsame sector has also been commonly used in empirical work, with the underlying premise\n\nthat localization economies come from the absolute volume of similar activities in the\n\nneighborhood.\n\nFor the localization measure, we consider the number of establishments in the\n\nsame 2 digit industry sector who are within 20 kilometers of each firm under\n\n\n\nAs discussed earlier, we only use establishments that have been in\n\n\n\nconsideration.\n\n\n\n5\n\n\n\nbusiness for more than 5 years\n\nMention: \"Uganda national roads dataset\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"It refines the network so analysis can be more precise.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004175:8:0", "occurrence_id": "20877fd835c0a0dc14eb7b2e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 2 shows the relative difference in the number of savings accounts in\n\n\nmunicipalities with Azteca, vs. municipalities without Azteca. Data on savings accounts\n\n\ncomes from the Mexican bank supervisory body, the Comisión Nacional Bancaria y de\n\n\nValores (CNBV). There is a clear rapid growth in savings accounts in municipalities with\n\nAzteca after December 2002 <sup>8</sup> [^8: Unfortunately, we cannot perform the same regression analysis for loans since we do not have date on loans\npre- and post-Azteca opening at the municipality level.] . These effects are also significant in a regression framework\n\nthat controls for municipality-specific time trends <sup>9</sup> [^9: Available from the authors upon request.] .\n\n\nThus, the opening of Banco Azteca had a non-trivial impact on availability of\n\n\nfinancial services in municipalities where the branches were present. More details on the\n\n\nbank opening are available in the Appendix.\n\n\n**II. Identification Strategy**\n\n\nIn order to identify the effect of Banco Azteca opening on economic activity, we\n\n\nexploit the cross-municipality and cross-time variation in Azteca branches. As discussed\n\n\nabove, Banco Azteca opened in October 2002, only in the municipalities that had pre-exiting\n\n\nGrupo Elektra stores. This allows us to estimate the following difference-in-difference\n\n\nregression, which compares municipalities with and without Banco Azteca before and after\n\n\nAzteca opening:\n\n_yict_    _c_  _t_                    - _Aztecact_                    - _Z_ _ict_  _ict_,\n\n\nwhere _i_ denotes individuals, _c_ denotes municipalities, and _t_ denotes quarters. ß is a set of\n\n\nmunicipality fixed effects, γ is a set of quarter fixed effects and Z is a matrix of individual\n\n\ncontrol variables. Our main variable of interest is the _Azteca_ dummy, which is the interaction\n\n\nbetween a post third quarter of 2002 dummy and a dummy that equals one for municipalities\n\n\nthat had at least one Azteca branch in the fourth quarter of 2002\n\nMention: \"Mexican bank supervisory body\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Savings account counts by municipality come from this supervisor.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004187:17:1", "occurrence_id": "40f21e4e0a0a61ec120286d5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: </u> <u>5</u> <u>62</u> <u>104</u> <u>206</u>\n\n\n\nGreenfield project 971 589 376 226 2162\nConcession 103 10 594 241 948\n<u>Divestiture</u> <u>478</u> <u>203</u> <u>65</u> <u>26</u> <u>772</u>\n<u>Total</u> <u>1587</u> <u>807</u> <u>1097</u> <u>597</u> <u>4088</u>\n\n\n\n<u>Source: PPI database.</u>\n\n8 A private sponsor builds a new facility largely at its own risk, transfers ownership to the government, leases\nthe facility from the government and operates it at its own risk up to the expiry of the lease.\n\n\n\n9 A private sponsor builds a new facility at its own risk, operates the facility at its own risk, and then transfers\nthe facility to the government at the end of the contract period. The private sponsor may or may not have the\nownership of the assets during the contract period.\n\n10 A private sponsor builds a new facility at its own risk, then owns and operates the facility at its own risk.\n\n11 A private sponsor builds a new facility in a liberalized market in which the government provides no revenue\nguarantees. The private developer assumes construction, operating, and market risk for the project.\n\n12 Electricity utilities or governments rent mobile power plants from private sponsors for periods ranging from 1\nyear to 15 years. A private sponsor places a new facility at its own risk, owns and operates the facility at its own\nrisk during the contract period.\n\nMention: \"PPI database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Transaction counts by sponsor type and region are tabulated from it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004196:32:1", "occurrence_id": "2ef10367ebbf3075955f430c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  GIEK from NOK 50b1n to 11Obin. In addition, its financial institution \"Eksportfinans\"<br>will lend banks NOK 50bln over the next two years to be used for medium to long-term export credits.|\n|||\n|**Pakistan**|11/18/2008:<br>**Plan**: The Central Bank will provide 100% refinancing to banks against export finance provided by them to<br>exporters under Part I of the Export Finance Scheme (EFS). Earlier, the State Bank was providing export<br>finance to the banks up to 70 percent. Export finance already provided by banks under Part I of EFS from<br>own sources at the ratio of 30 percent and outstanding as on Oct. 31, 2008 will also be refinanced by the<br>State Bank for the remaining period of individual loans|\n||<br>06/12/2009 **Announcement:** The IFC, Habib Bank Limited (HBL) and Citibank (Citi) announced that their<br>joint execution of a structured trade finance deal to support the import of equipment for a new power<br>generation plant**in Pakistan**. IFC and Citi worked together to structure a deal with coverage of up to Ђ73<br>million (PKR 8 Billion) to a Pakistani power company. HBL, Pakistan’s largest privately owned bank,<br>established the original letter of credit, which was confirmed by Citi utilizing IFC’s Global Trade Finance<br>program for 50% risk coverage.<br>|\n|||\n|**Poland**|Source: Berne Union, Compiled in May 2009<br>**Action:**Poland’s Export Credit Insurance Corporation (**KUKE Poland**), a joint-\n\nMention: \"Berne Union\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Export credit and guarantee arrangements are compiled from this association.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004198:13:1", "occurrence_id": "f0ba89a7aa9e4da916532fd2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: , and a\n\nsizeable share of aid may go directly from donors to NGOs. For these recipients, donors’ use of\n\nPFM systems as a share of total aid would be a misleading indicator of trust in country systems.\n\nIn any case, results reported below are unchanged if we replace government sector aid with total\n\naid in the denominator of _PFM_ .\n\nSome aid projects may be provided in support of program-based approaches, even if they\n\nare not part of an agreement with governments. Hence the DAC monitoring indicators measure\n\n_PBA_ and _DBS_ (a subset of _PBA_ ) as a share of total aid, not as a share of aid to the government\n\nsector. Again, however, our empirical findings are not sensitive to this choice of denominator.\n\nWe classify aid delivered through program-based approaches as recipient-managed\n\nbecause the programs are led by government and reflect its priorities, and include processes for\n\nharmonizing donor procedures and using some country systems (see Appendix 2). Subject to\n\nthose conditions, aid delivered in the form of projects can qualify as program-based, even if it\n\ndoes not use country PFM systems. As the case of _PBA_ illustrates, the distinction between\n\ndonor-managed and recipient-managed aid is more accurately depicted on a continuum than as a\n\n\n12\n\nMention: \"the DAC monitoring indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"They measure budget support as a share of total aid rather than of government aid.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004153:14:1", "occurrence_id": "e39c232296759709329a79b1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  It is also currently the main contributor to GDP (Table\nA-5, Panel 3). The analysis of the latest household survey (EICV 2) showed that poverty has\nfallen the most during the previous years in households that depend on agriculture. The low use\nof modern inputs (such as fertilizers and improved seeds), the lack of extension services as well\nas poor water management, however, hamper the development of the agriculture sector. The poor\ncondition of the necessary infrastructure in addition limits the extent of growth and\ntransformation in the sector.\n\nThis scenario assumes that additional donor financing will be spent on agriculture, investing and\nraising growth in its capital stock, again accompanied by increased growth in real recurrent\nexpenditure (operations and maintenance). A possible example for such an investment would be\nthe World Bank project on land husbandry, water harvesting and irrigation that is currently under\npreparation. The larger agricultural capital stock has a positive impact on productivity growth in\nthe agricultural sector. MDG and education indicators may improve due to indirect effects via\nhigher household incomes. However, in the model there is no direct effect on MDGs and\neducational indicators from spending on agriculture (as opposed to spending on infrastructure).\n\nThe simulation outcome in terms of macroeconomic effects is very similar to that of AID-INF,\nexcept for an even stronger bias of government spending on investment at the expense of\ngovernment consumption (Table 5). Annual GDP growth averages 6.5 percent, with the\nagriculture sector consequently growing faster than in the previous scenarios, whereas growth\nrates of the industry and services sector are between those of AID-HD and AID-INF (Table A-5).\n\nThe development of the MDG and education indicators is less positive than in AID-INF and AIDHD, but nevertheless more positive than in the baseline. The poverty headcount declines to 32\npercent in 2020 similar to AID-INF (Table 6), highlighting the indirect positive impact of\n\n\n11\n\nMention: \"EICV 2\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Poverty fell most among agriculture-dependent households in this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004166:13:1", "occurrence_id": "ecf5e2afba033586ad8eadb4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 4**\n(11.8) (11.8)\nNumber of secondary educated in household 18.6 17.9\n(20.5) (20.5)\nNumber of tertiary educated in household 1.7 0.8\n(76.1) (76.0)\nNumber of children below age 5 in household -69.0*** -69.2***\n(15.7) (15.7)\nNumber of males above age 15 in household 73.2*** 73.5***\n(18.8) (18.7)\nNumber of pre-flood migrants from household -6.0 -6.1\n(15.9) (15.9)\nConstant 180.8 174.1\n(219.7) (219.2)\nObservations 469 469\n<u>R-squared</u> <u>0.41</u> <u>0.41</u>\n<u>Standard errors in brackets</u>\n\n- significant at 10%; ** significant at 5%; *** significant at 1%\nSource: Authors estimations based on household survey in Bangladesh conducted by the\nInternational Food Policy Research Institute in 1998-99 (see del Ninno et al. 2001).\n\n\n**_4.3 Ex-ante preparedness of remittance-receiving households for disasters in Ghana_**\n**_and Burkina Faso_**\n\n\nIn this section, we explore whether households in Ghana and Burkina Faso that\nreceive remittances ex-ante better prepared against natural disasters compared to other\nhouseholds. West Africa in general and the Sahel region in particular are characterized by\n\nhouseholds that don’t receive any remittances). This appears to be lower than the average propensity to\nconsume likely because of the use of remittances for reconstruction after the flood.\n\n\n12\n\nMention: \"International Food Policy Research Institute\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its household survey underlies the preparedness estimates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004200:11:0", "occurrence_id": "9eb443a64a9d0f97e732ff29", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **_(in percent)_**\n\n**Employment (payroll)**\n\n1.5\n\n\n1.0\n\n\n0.5\n\n\n0.0\n\n\n-0.5\n\n\n-1.0\n\n\n-1.5\n\n\n-2.0\n\n\n-2.5\n\n\n-3.0\n\n\n\n**Figure 4. Macroeconomic indicators**\n\n\n**Industrial production**\n\n1.0\n\n0.5\n\n0.0\n\n-0.5\n\n-1.0\n\n-1.5\n\n-2.0\n\n-2.5\n\n-3.0\n\n-3.5\n\n-4.0\n\n\n\nJun-06 Dec-06 Jun-07 Dec-07 Jun-08 Dec-08\n\n\n\nAug-06 Feb-07 Aug-07 Feb-08 Aug-08 Feb-09\n\n\n\n1.5\n\n1.0\n\n0.5\n\n0.0\n\n-0.5\n\n-1.0\n\n-1.5\n\n-2.0\n\n-2.5\n\n-3.0\n\n-3.5\n\n-4.0\n\n\n\n**Overall macroeconomic factor**\n\n\n\n1.0\n\n\n0.5\n\n\n0.0\n\n\n-0.5\n\n\n-1.0\n\n\n-1.5\n\n\n-2.0\n\n\n-2.5\n\n\n-3.0\n\n\n-3.5\n\n\n\n**Export**\n\n\n\nAug-06 Feb-07 Aug-07 Feb-08 Aug-08 Feb-09\n\n\n\nJun-06 Dec-06 Jun-07 Dec-07 Jun-08 Dec-08\n\n\n\nSource: Bloomberg, Datastream, ILO, UK Statistical Office\nNote: The overall macroeconomic index is derived as the predicted common factor in a factor analysis of twelve variables\n(transformed into Y-o-Y growth in all cases): US index of industrial production, Germany index of industrial production (incl.\nconstruction), UK index of production (all industries), Japan index of production (all industries), US total exports, Germany exports\nf.o.b., UK exports, Japan exports, US nonfarm total employed, Germany persons\n\nMention: \"Bloomberg\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Market series feeding the macro index come from this vendor.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004200:11:0", "occurrence_id": "47b1f3471bba9dcc6b96cddf", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **_(in percent)_**\n\n**Employment (payroll)**\n\n1.5\n\n\n1.0\n\n\n0.5\n\n\n0.0\n\n\n-0.5\n\n\n-1.0\n\n\n-1.5\n\n\n-2.0\n\n\n-2.5\n\n\n-3.0\n\n\n\n**Figure 4. Macroeconomic indicators**\n\n\n**Industrial production**\n\n1.0\n\n0.5\n\n0.0\n\n-0.5\n\n-1.0\n\n-1.5\n\n-2.0\n\n-2.5\n\n-3.0\n\n-3.5\n\n-4.0\n\n\n\nJun-06 Dec-06 Jun-07 Dec-07 Jun-08 Dec-08\n\n\n\nAug-06 Feb-07 Aug-07 Feb-08 Aug-08 Feb-09\n\n\n\n1.5\n\n1.0\n\n0.5\n\n0.0\n\n-0.5\n\n-1.0\n\n-1.5\n\n-2.0\n\n-2.5\n\n-3.0\n\n-3.5\n\n-4.0\n\n\n\n**Overall macroeconomic factor**\n\n\n\n1.0\n\n\n0.5\n\n\n0.0\n\n\n-0.5\n\n\n-1.0\n\n\n-1.5\n\n\n-2.0\n\n\n-2.5\n\n\n-3.0\n\n\n-3.5\n\n\n\n**Export**\n\n\n\nAug-06 Feb-07 Aug-07 Feb-08 Aug-08 Feb-09\n\n\n\nJun-06 Dec-06 Jun-07 Dec-07 Jun-08 Dec-08\n\n\n\nSource: Bloomberg, Datastream, ILO, UK Statistical Office\nNote: The overall macroeconomic index is derived as the predicted common factor in a factor analysis of twelve variables\n(transformed into Y-o-Y growth in all cases): US index of industrial production, Germany index of industrial production (incl.\nconstruction), UK index of production (all industries), Japan index of production (all industries), US total exports, Germany exports\nf.o.b., UK exports, Japan exports, US nonfarm total employed, Germany persons\n\nMention: \"Datastream\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Additional macro series come from this vendor.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004200:11:0", "occurrence_id": "c8fea58e0c0c60c5cb2f8b58", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **_(in percent)_**\n\n**Employment (payroll)**\n\n1.5\n\n\n1.0\n\n\n0.5\n\n\n0.0\n\n\n-0.5\n\n\n-1.0\n\n\n-1.5\n\n\n-2.0\n\n\n-2.5\n\n\n-3.0\n\n\n\n**Figure 4. Macroeconomic indicators**\n\n\n**Industrial production**\n\n1.0\n\n0.5\n\n0.0\n\n-0.5\n\n-1.0\n\n-1.5\n\n-2.0\n\n-2.5\n\n-3.0\n\n-3.5\n\n-4.0\n\n\n\nJun-06 Dec-06 Jun-07 Dec-07 Jun-08 Dec-08\n\n\n\nAug-06 Feb-07 Aug-07 Feb-08 Aug-08 Feb-09\n\n\n\n1.5\n\n1.0\n\n0.5\n\n0.0\n\n-0.5\n\n-1.0\n\n-1.5\n\n-2.0\n\n-2.5\n\n-3.0\n\n-3.5\n\n-4.0\n\n\n\n**Overall macroeconomic factor**\n\n\n\n1.0\n\n\n0.5\n\n\n0.0\n\n\n-0.5\n\n\n-1.0\n\n\n-1.5\n\n\n-2.0\n\n\n-2.5\n\n\n-3.0\n\n\n-3.5\n\n\n\n**Export**\n\n\n\nAug-06 Feb-07 Aug-07 Feb-08 Aug-08 Feb-09\n\n\n\nJun-06 Dec-06 Jun-07 Dec-07 Jun-08 Dec-08\n\n\n\nSource: Bloomberg, Datastream, ILO, UK Statistical Office\nNote: The overall macroeconomic index is derived as the predicted common factor in a factor analysis of twelve variables\n(transformed into Y-o-Y growth in all cases): US index of industrial production, Germany index of industrial production (incl.\nconstruction), UK index of production (all industries), Japan index of production (all industries), US total exports, Germany exports\nf.o.b., UK exports, Japan exports, US nonfarm total employed, Germany persons\n\nMention: \"US index of industrial production\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"One of twelve variables entering the factor-analysis index.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004200:11:0", "occurrence_id": "cd8ef30e09cbf680563b8948", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **_(in percent)_**\n\n**Employment (payroll)**\n\n1.5\n\n\n1.0\n\n\n0.5\n\n\n0.0\n\n\n-0.5\n\n\n-1.0\n\n\n-1.5\n\n\n-2.0\n\n\n-2.5\n\n\n-3.0\n\n\n\n**Figure 4. Macroeconomic indicators**\n\n\n**Industrial production**\n\n1.0\n\n0.5\n\n0.0\n\n-0.5\n\n-1.0\n\n-1.5\n\n-2.0\n\n-2.5\n\n-3.0\n\n-3.5\n\n-4.0\n\n\n\nJun-06 Dec-06 Jun-07 Dec-07 Jun-08 Dec-08\n\n\n\nAug-06 Feb-07 Aug-07 Feb-08 Aug-08 Feb-09\n\n\n\n1.5\n\n1.0\n\n0.5\n\n0.0\n\n-0.5\n\n-1.0\n\n-1.5\n\n-2.0\n\n-2.5\n\n-3.0\n\n-3.5\n\n-4.0\n\n\n\n**Overall macroeconomic factor**\n\n\n\n1.0\n\n\n0.5\n\n\n0.0\n\n\n-0.5\n\n\n-1.0\n\n\n-1.5\n\n\n-2.0\n\n\n-2.5\n\n\n-3.0\n\n\n-3.5\n\n\n\n**Export**\n\n\n\nAug-06 Feb-07 Aug-07 Feb-08 Aug-08 Feb-09\n\n\n\nJun-06 Dec-06 Jun-07 Dec-07 Jun-08 Dec-08\n\n\n\nSource: Bloomberg, Datastream, ILO, UK Statistical Office\nNote: The overall macroeconomic index is derived as the predicted common factor in a factor analysis of twelve variables\n(transformed into Y-o-Y growth in all cases): US index of industrial production, Germany index of industrial production (incl.\nconstruction), UK index of production (all industries), Japan index of production (all industries), US total exports, Germany exports\nf.o.b., UK exports, Japan exports, US nonfarm total employed, Germany persons\n\nMention: \"Germany index of industrial production\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"A second member of the twelve-variable set.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004200:11:0", "occurrence_id": "ec02f91970d9b635bc4323bb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **_(in percent)_**\n\n**Employment (payroll)**\n\n1.5\n\n\n1.0\n\n\n0.5\n\n\n0.0\n\n\n-0.5\n\n\n-1.0\n\n\n-1.5\n\n\n-2.0\n\n\n-2.5\n\n\n-3.0\n\n\n\n**Figure 4. Macroeconomic indicators**\n\n\n**Industrial production**\n\n1.0\n\n0.5\n\n0.0\n\n-0.5\n\n-1.0\n\n-1.5\n\n-2.0\n\n-2.5\n\n-3.0\n\n-3.5\n\n-4.0\n\n\n\nJun-06 Dec-06 Jun-07 Dec-07 Jun-08 Dec-08\n\n\n\nAug-06 Feb-07 Aug-07 Feb-08 Aug-08 Feb-09\n\n\n\n1.5\n\n1.0\n\n0.5\n\n0.0\n\n-0.5\n\n-1.0\n\n-1.5\n\n-2.0\n\n-2.5\n\n-3.0\n\n-3.5\n\n-4.0\n\n\n\n**Overall macroeconomic factor**\n\n\n\n1.0\n\n\n0.5\n\n\n0.0\n\n\n-0.5\n\n\n-1.0\n\n\n-1.5\n\n\n-2.0\n\n\n-2.5\n\n\n-3.0\n\n\n-3.5\n\n\n\n**Export**\n\n\n\nAug-06 Feb-07 Aug-07 Feb-08 Aug-08 Feb-09\n\n\n\nJun-06 Dec-06 Jun-07 Dec-07 Jun-08 Dec-08\n\n\n\nSource: Bloomberg, Datastream, ILO, UK Statistical Office\nNote: The overall macroeconomic index is derived as the predicted common factor in a factor analysis of twelve variables\n(transformed into Y-o-Y growth in all cases): US index of industrial production, Germany index of industrial production (incl.\nconstruction), UK index of production (all industries), Japan index of production (all industries), US total exports, Germany exports\nf.o.b., UK exports, Japan exports, US nonfarm total employed, Germany persons\n\nMention: \"UK index of production\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"A third member of the twelve-variable set.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004200:11:0", "occurrence_id": "3b3f6fd70ed5e429ffdea1a0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **_(in percent)_**\n\n**Employment (payroll)**\n\n1.5\n\n\n1.0\n\n\n0.5\n\n\n0.0\n\n\n-0.5\n\n\n-1.0\n\n\n-1.5\n\n\n-2.0\n\n\n-2.5\n\n\n-3.0\n\n\n\n**Figure 4. Macroeconomic indicators**\n\n\n**Industrial production**\n\n1.0\n\n0.5\n\n0.0\n\n-0.5\n\n-1.0\n\n-1.5\n\n-2.0\n\n-2.5\n\n-3.0\n\n-3.5\n\n-4.0\n\n\n\nJun-06 Dec-06 Jun-07 Dec-07 Jun-08 Dec-08\n\n\n\nAug-06 Feb-07 Aug-07 Feb-08 Aug-08 Feb-09\n\n\n\n1.5\n\n1.0\n\n0.5\n\n0.0\n\n-0.5\n\n-1.0\n\n-1.5\n\n-2.0\n\n-2.5\n\n-3.0\n\n-3.5\n\n-4.0\n\n\n\n**Overall macroeconomic factor**\n\n\n\n1.0\n\n\n0.5\n\n\n0.0\n\n\n-0.5\n\n\n-1.0\n\n\n-1.5\n\n\n-2.0\n\n\n-2.5\n\n\n-3.0\n\n\n-3.5\n\n\n\n**Export**\n\n\n\nAug-06 Feb-07 Aug-07 Feb-08 Aug-08 Feb-09\n\n\n\nJun-06 Dec-06 Jun-07 Dec-07 Jun-08 Dec-08\n\n\n\nSource: Bloomberg, Datastream, ILO, UK Statistical Office\nNote: The overall macroeconomic index is derived as the predicted common factor in a factor analysis of twelve variables\n(transformed into Y-o-Y growth in all cases): US index of industrial production, Germany index of industrial production (incl.\nconstruction), UK index of production (all industries), Japan index of production (all industries), US total exports, Germany exports\nf.o.b., UK exports, Japan exports, US nonfarm total employed, Germany persons\n\nMention: \"Japan index of production\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"A fourth member of the twelve-variable set.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004201:6:0", "occurrence_id": "43621d514e34fb2b65579d9b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: (34) and Bulgaria (14). The track I projects are located in Germany, Hungary, Romania, New Zealand,\nFrance and Ukraine. <sup>10</sup> [^10: The data come for the UNFCCC website and the UNEP Risoe CDM pipeline.]\n\n\nThe number of JI projects is dwarfed by those in the CDM pipeline, which in June 2009 contained 4,417\nprojects. The disparity in the numbers between JI and CDM projects is partially attributable to the fact\nthat the CDM enjoyed a “prompt start” authorized by the Conference of the Parties (COP) to the\nUNFCCC which allowed the registration of projects starting in 2001. This means that the CDM’s Certified\nEmissions Reductions (CERs) obtained during the period between 2000 and 2008 can be used to assist in\nachieving compliance during the first commitment period. JI on the other hand can only generate\nEmission Reduction Units (ERUs) during the first commitment period, namely between 2008 and 2012.\nWhile the CDM is managed or administered on the international level by a governance structure that is\nlargely independent from host country performance, administrative capacities and GHG accounting\nframeworks, JI depends on emission limitation targets, allocation of AAUs, registries and the functioning\nof almost all other elements of the infrastructure of the Kyoto Protocol. Carbon credits generated by JI\nare nothing but AAUs in disguise, authorized for private trade, but generated by the host country which\nconverts AAUs into ERUs based on the evidence of verified emission reductions. As a result JI is\nenvironmentally more robust than the CDM, but institutionally more vulnerable. It is also highly\ndependent on the functioning of domestic institutions: in order to participate in JI, countries have to\nfulfill a number of accounting requirements and compliance has to continue throughout the\nimplementation of a JI project. Countries have to elaborate project approval procedures and be able to\naccount for ERUs as part of their AAUs. Participation in JI also depends on political will. Until June 2009\nthe Russian Federation has not approved a single JI\n\nMention: \"the UNEP Risoe CDM pipeline\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"It counts the clean development mechanism projects in the pipeline.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004202:5:1", "occurrence_id": "f18fc6e54a348c6a4090e2dc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  well, aside from the expansion of the Panama Canal which closed\nin Q42008. <sup>7</sup> [^7: Projectware Database, Dealogic, as of April 2009.] The expectation is that PPI levels will drop considerably through the rest of 2009\nin LAC as they have in every other region of the world. <sup>8</sup> [^8: See also Izaguirre (2009) which identifies a modest decrease in quarter to quarter PPI projects in LAC\ncoming to closure from 2007 Q4 to 2008 Q4.]\n\n\n**Figure 1: Private Financing to LAC Infrastructure, Quarterly from 2004 to 2008**\n\n\n9\n\n\n\n8\n\n\n7\n\n\n6\n\n\n5\n\n\n4\n\n\n3\n\n\n2\n\n\n1\n\n\n0\n\n\n\n\n|LAC US$ Mill (excluding commodity-oriented<br>infra and Panama)|Col2|Col3|Col4|Col5|Col6|Col7|Col8|Col9|Col10|Col11|Col12|Col13|Col14|Col15|Col16|Col17|Col18|Col19|Col20|Col21|Col22|Col23|Col24|Col25|Col26|Col27|Col28|Col29|Col30|Col31|Col32|Col33|Col34|Col35|Col36|Col37|\n|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|\n|LAC US$ Mill (excluding commodity-oriented<br>~\n\nMention: \"Projectware Database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Infrastructure deal records as of this date come from it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004203:6:1", "occurrence_id": "9d6b44c6e9bb69a5e9dccd31", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  world.\n\nAccording to the International Labour Organization (2002), 246 million children and\nteenagers between 5 and 17 years old are engaged in child labor around the world.\nFurthermore, 75.0% of those children work for their own family activities. Asia, Africa and\nLatin America are the continents with the most the child labor in the world. Asia has the\nhighest number of children in terms of volume but Africa is the leader relative to the total size\nof the work force.\n\nIn Brazil, data from PNAD (IBGE, 2007) has shown that from a 44.7 million population\nbetween 5 and 17 years old, 10.8% (4.8 million) are directly involved in child labor. The\nworst region of Brazil regarding this subject is the Northeast, with 13.4% of the 5-17 year old\npopulation working; Southeast holds the lowest average (7.9%). These statistics represent a\npositive evolution over the last 4 years (see Figure 3 below).\n\n\n5\n\nMention: \"data from PNAD\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its figures give the working share of children aged five to seventeen.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004205:14:0", "occurrence_id": "75aa0c7079090eb58205aeb0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: lines (US$ 2, US$ 3 and US$ 4 per person per day). The rest of the variables used in the\nregressions are taken from Loayza _et al_ . (2002), except for the education variables, which\nare from Barro and Lee (2001).\n\n\nOur regressions are conducted using an unbalanced panel of non-overlapping fiveyear periods spanning the years 1960-2000. Income and inequality data pertain to the latest\navailable year within each given period. The original income data is from the PWT 6.1\nwhereas the original inequality data comes from Dollar and Kraay’s (2002) database on\ninequality. <sup>19</sup>\n\n\nTable 1 presents summary statistics for income, inequality and the constructed\npoverty measures. <sup>20</sup> The table shows the wide range of per capita income levels in the\nsample – from less than $500 (Tanzania in the mid 1990s) to almost $35,000 (Luxembourg\nin the mid 1990s). The median observation corresponds to Mexico in the mid 1970s, with\nper capita income about $5,500. <sup>21</sup> Regarding inequality, the Gini indices range from a low\n0.17 (the Slovak Republic in the early 1990s) to a high 0.76 (Namibia in the mid 1990s),\nwith a median of 0.38. Regarding the poverty figures, by construction they must rise with\nthe poverty line and decline as the poverty measure changes from P0 to P2 (i.e., as one\nconsiders more bottom-sensitive measures). Table 1 shows that, depending on the poverty\nline used, median headcount poverty ranges from 2.4 percent (using US$2 per day as the\npoverty line) to about 13 percent (with US$4 per day), whereas the\n\nMention: \"PWT 6.1\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The original income data come from this version.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004205:14:0", "occurrence_id": "69648d0777500e3d493cf625", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: lines (US$ 2, US$ 3 and US$ 4 per person per day). The rest of the variables used in the\nregressions are taken from Loayza _et al_ . (2002), except for the education variables, which\nare from Barro and Lee (2001).\n\n\nOur regressions are conducted using an unbalanced panel of non-overlapping fiveyear periods spanning the years 1960-2000. Income and inequality data pertain to the latest\navailable year within each given period. The original income data is from the PWT 6.1\nwhereas the original inequality data comes from Dollar and Kraay’s (2002) database on\ninequality. <sup>19</sup>\n\n\nTable 1 presents summary statistics for income, inequality and the constructed\npoverty measures. <sup>20</sup> The table shows the wide range of per capita income levels in the\nsample – from less than $500 (Tanzania in the mid 1990s) to almost $35,000 (Luxembourg\nin the mid 1990s). The median observation corresponds to Mexico in the mid 1970s, with\nper capita income about $5,500. <sup>21</sup> Regarding inequality, the Gini indices range from a low\n0.17 (the Slovak Republic in the early 1990s) to a high 0.76 (Namibia in the mid 1990s),\nwith a median of 0.38. Regarding the poverty figures, by construction they must rise with\nthe poverty line and decline as the poverty measure changes from P0 to P2 (i.e., as one\nconsiders more bottom-sensitive measures). Table 1 shows that, depending on the poverty\nline used, median headcount poverty ranges from 2.4 percent (using US$2 per day as the\npoverty line) to about 13 percent (with US$4 per day), whereas the\n\nMention: \"database on\ninequality\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The original inequality data come from this companion compilation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004208:11:0", "occurrence_id": "1ffb27bcfc33ef475c75befb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: To evaluate global imbalances we examine the trade balance as a share of GDP\n\n\nacross regions and income groups, and also whether the countries tend to be surplus or\n\n\ndeficit countries. Figure 7 and 8 show movements in the aggregate trade deficit and\n\n\nsurplus for international borrowers and lenders, respectively. Specifically, Figure 7 is the\n\n\naggregate trade deficit relative to GDP over time of all the countries that had a deficit\n\n\nbefore the downturn. <sup>8</sup> Similarly, Figure 8 is the aggregate trade surplus relative to GDP\n\n\nover time of all the countries that had a surplus before the downturn. On aggregate, there\n\n\nis only a temporary improvement, and the position quickly returns to where it was before\n\n\nthe episode.  This suggests that there is not an overall rebalancing between pre-downturn\n\n\nsurplus countries and pre-downturn deficit countries.\n\n\nHowever, the pictures are very different when we look across regions (Figures 9\n\n\nand 10) or income groups (Figures 11 and 12). We find that in Latin America, East Asia,\n\n\nEurope and the Middle East and North Africa there is a tendency to reduce sharp deficits\n\n\nfollowing downturns. These are regions where the imbalances were deemed dangerous,\n\n\nand the government put policies in place to ensure they did not reemerge. These included\n\n\nsavings incentives and maintaining an undervalued exchange rate to strengthen the BOP\n\n\nposition. In addition, weakened firm fundamentals during a global downturn may induce\n\n\na drop in investment.\n\n\nIn contrast, North America tends to show a relatively stable account that improves\n\n\nslightly during the downturn, but worsens in its aftermath. This may reflect the flexibility\n\n\nof the U.S. economy and the safety of dollar assets. This may also be related to the\n\n\n8 Specifically, we use the value of deficit or surplus four years before the episode to characterize country as\na surplus or deficit country. Data are from the IMF BOP Statistics for a balanced sample of countries\nreporting.\n\n\n10\n\nMention: \"the IMF BOP Statistics\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Balance of payments positions before each episode are classified from these statistics.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004216:3:0", "occurrence_id": "d7a139f9d95c39a72faec66c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 1. **_Introduction_**\n\n\nThe US subprime crisis, and the likelihood that the attendant downturn will spread from\n\n\ndeveloped to developing countries has left policy-makers and academics concerned about the\n\n\n“human” costs of the crisis for poor countries. These human costs could in principle take a variety\n\n\nof forms—lower income and consumption levels, higher unemployment, lower investments in\n\n\neducation by families, and worse health and nutrition outcomes. Of particular concern are\n\n\nchanges that could have irreversible consequences for the wellbeing of children throughout their\n\n\nlives—for example, school dropout or impaired child development at critical, early ages. No\n\n\noutcome is more dramatic than infant death, and in no region is infant mortality higher than in\n\nSub-Saharan Africa (henceforth Africa, for short). <sup>1</sup> In this note we assess the likely impact of the\n\n\ncurrent financial crisis on infant mortality in Africa. To do this, we first use data from all\n\n\navailable Demographic and Health Surveys (DHS) for the region to construct retrospective birth\n\n\nand death histories, and assess the impact of past crises on infant mortality in Africa. We then use\n\n\nthe most recent (April 2009) IMF growth projections to estimate the likely impact of the current\n\n\ndownturn on infant mortality.\n\n\nThe basic messages of our analysis are two. First, although severe, the consequences of\n\n\nthe current economic crisis for infant mortality in Africa are likely to be much smaller than many\n\nof the estimates that have been frequently cited on blogs or in the popular press. <sup>2</sup> Specifically, our\n\n\nestimates suggest approximately 30,000 to 50,000 excess infant deaths in Africa in 2009. Second,\n\n\nhowever, we find that virtually all of these excess deaths take place to girls, not boys.\n\n\nFurthermore, these asymmetries in the impact of the crisis by gender are particularly apparent for\n\n\nsharp (rather than modest) contractions in per capita GDP—a finding that is important if the\n\nMention: \"Demographic and Health Surveys (DHS)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Retrospective birth and death histories are built from all regional rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004217:17:9", "occurrence_id": "272371eab213d6e0930da400", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _<br>_MENA_<br>_S._<br>_Asia_|_SSA_|_SSA_|_SSA_|_SSA+_<br>_25%_|_SSA+_<br>_25%_|_SSA+_<br>_25%_|_Low_<br>_Inc._|_Low_<br>_Inc._|_Low_<br>_Inc._|_EA&P_|_EA&P_|_EA&P_|_LAC_|_LAC_|_LAC_|_MENA_|_MENA_|_MENA_|_S._<br>_Asia_|_S._<br>_Asia_|_S._<br>_Asia_|_S._<br>_Asia_|\n|_GDI/GDP_|_18.9_|_18.9_|_18.9_|_29.6_|_29.6_|_29.6_|_21.1_|_21.1_|_21.1_|_24.2_|_24.2_|_24.2_|_23_|_23_|_23_|_19.4_|_19.4_|_19.4_|_27.4_|_27.4_|_27.4_|_27.4_|\n\n\nSource: World Bank, World Development Indicators. Note: SSA + 25% = Gross domestic investment achieved\nfollowing repatriation of 25 percent of the stock of capital flight.\n\nAnother important channel of growth effects of flight capital repatriation is through accelerated human\ncapital accumulation arising from higher funding for education. Given the low\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Investment rates under the repatriation scenarios are computed from these series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004218:16:0", "occurrence_id": "40c6bc818b07d5cc0d9d6053", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ~\nln( _yit_ )   _it_  _Tt_   ln( _N_ _it_ )  ln( _X_ _it_  )  _I_ _it_  _it_ (5)\n\n**IV.** **Empirical Results and Prospects for Convergence**\n\nFollowing estimation of parameters in the first-stage regression, the model specified by (5) is fitted to\na panel of 24 countries. Included in this panel are all the six GCC countries and a sub-sample of 18\ncountries in SSA. Data sources are World Bank, World Development Indicators and World\nEconomic Outlook. The time series data covers the period 1970—2007 as the study focuses on longterm changes and not cyclical variations in output. Per capita GDP, public and private investments,\nforeign direct investment, net interest income, changes in real effective exchange rate and money\nsupply data in the conditioning vector are from the World Development Indicators database. So is\nthe natural resource variable _N_ _it_, which is proxied by country exports, assuming that natural\nresources continue to account for the lion’s share of exports in these countries. <sup>37</sup> [^37: For most countries, including those which have entered the path of diversification, natural resources\ncontinue to account for over 90% of total exports. In part, this reflects the fact that the diversification process\nhas simply reduced the marginal propensity to imports in most countries which continue to operate far off the\nglobal production possibility frontier as a result of their productivity and competitiveness gap.] Data on corruption\nindex, political stability and political risk rating are obtained from International Country Risk Guide.\n\nThis section assesses the prospects for income convergence between GCC and SSA countries when\ninstitutional continuity is controlled for. Counterfactual simulations investigating plausible policies\nthat could enhance income convergence between the two regions are undertaken. However, in order\nto infer on the convergence parameters, it is foremost important to derive an empirical model that\nbest estimates the dynamics of output growth in these economies. Such an underlying model is\nprovided\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The long panel series for both region groups come from this compilation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004218:16:0", "occurrence_id": "a5ea09764aa5489b67c680e7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ~\nln( _yit_ )   _it_  _Tt_   ln( _N_ _it_ )  ln( _X_ _it_  )  _I_ _it_  _it_ (5)\n\n**IV.** **Empirical Results and Prospects for Convergence**\n\nFollowing estimation of parameters in the first-stage regression, the model specified by (5) is fitted to\na panel of 24 countries. Included in this panel are all the six GCC countries and a sub-sample of 18\ncountries in SSA. Data sources are World Bank, World Development Indicators and World\nEconomic Outlook. The time series data covers the period 1970—2007 as the study focuses on longterm changes and not cyclical variations in output. Per capita GDP, public and private investments,\nforeign direct investment, net interest income, changes in real effective exchange rate and money\nsupply data in the conditioning vector are from the World Development Indicators database. So is\nthe natural resource variable _N_ _it_, which is proxied by country exports, assuming that natural\nresources continue to account for the lion’s share of exports in these countries. <sup>37</sup> [^37: For most countries, including those which have entered the path of diversification, natural resources\ncontinue to account for over 90% of total exports. In part, this reflects the fact that the diversification process\nhas simply reduced the marginal propensity to imports in most countries which continue to operate far off the\nglobal production possibility frontier as a result of their productivity and competitiveness gap.] Data on corruption\nindex, political stability and political risk rating are obtained from International Country Risk Guide.\n\nThis section assesses the prospects for income convergence between GCC and SSA countries when\ninstitutional continuity is controlled for. Counterfactual simulations investigating plausible policies\nthat could enhance income convergence between the two regions are undertaken. However, in order\nto infer on the convergence parameters, it is foremost important to derive an empirical model that\nbest estimates the dynamics of output growth in these economies. Such an underlying model is\nprovided\n\nMention: \"World\nEconomic Outlook\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"It supplies the companion macro series over the same span.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004218:16:0", "occurrence_id": "d9923de43ddfc36c5a49e0ba", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ~\nln( _yit_ )   _it_  _Tt_   ln( _N_ _it_ )  ln( _X_ _it_  )  _I_ _it_  _it_ (5)\n\n**IV.** **Empirical Results and Prospects for Convergence**\n\nFollowing estimation of parameters in the first-stage regression, the model specified by (5) is fitted to\na panel of 24 countries. Included in this panel are all the six GCC countries and a sub-sample of 18\ncountries in SSA. Data sources are World Bank, World Development Indicators and World\nEconomic Outlook. The time series data covers the period 1970—2007 as the study focuses on longterm changes and not cyclical variations in output. Per capita GDP, public and private investments,\nforeign direct investment, net interest income, changes in real effective exchange rate and money\nsupply data in the conditioning vector are from the World Development Indicators database. So is\nthe natural resource variable _N_ _it_, which is proxied by country exports, assuming that natural\nresources continue to account for the lion’s share of exports in these countries. <sup>37</sup> [^37: For most countries, including those which have entered the path of diversification, natural resources\ncontinue to account for over 90% of total exports. In part, this reflects the fact that the diversification process\nhas simply reduced the marginal propensity to imports in most countries which continue to operate far off the\nglobal production possibility frontier as a result of their productivity and competitiveness gap.] Data on corruption\nindex, political stability and political risk rating are obtained from International Country Risk Guide.\n\nThis section assesses the prospects for income convergence between GCC and SSA countries when\ninstitutional continuity is controlled for. Counterfactual simulations investigating plausible policies\nthat could enhance income convergence between the two regions are undertaken. However, in order\nto infer on the convergence parameters, it is foremost important to derive an empirical model that\nbest estimates the dynamics of output growth in these economies. Such an underlying model is\nprovided\n\nMention: \"International Country Risk Guide\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Corruption, stability and risk ratings are taken from this guide.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004220:10:0", "occurrence_id": "65a516deb5f2eb41b396d07f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **7**\n\n\n**6**\n\n\n**5**\n\n\n**4**\n\n\n**3**\n\n\n**2**\n\n\n**1**\n\n\n**0**\n\n\n\nFigure 3: Unemployment Risks in Brazil\n\n\n\n\n\n\n\n_Source:_ Authors’ calculations based on PME.\n\n###### 3 The dynamic stochastic behavioral model\n\n\nWe are interested in formalizing the effect of the social insurance system on three\n\neconomic decisions: (i) the level of savings; (ii) efforts to preserve/find jobs in the “formal sector”\n\n(defined by access to social security); and (iii) retirement decisions. We use as our starting point\n\nthe standard life-cycle utility maximization framework and introduce uncertainty in employment\n\nstatus and life expectancy. Clearly, the assumptions of this framework are controversial;\n\nindividuals are usually not fully rational and do not have perfect foresight. If they were, there\n\nwould not be a need for a social security system in the first place. The model/framework is\n\nnonetheless useful as a benchmark to understand the _direction of change_ in certain behaviors as\n\na response to change in the rules of the social security system. Moreover, in our application, we\n\nexplore a large range of possible behavioral responses to policy changes. Behaviors that are\n\nmore like to have generated actual observations receive a higher weight. In essence, we use the\n\nmodel as a data generation mechanism and from this point of view it is not very different from\n\nlinear single-equation econometric models.\n\n\nThe dynamic stochastic problem that representative individuals are assumed to solve is\n\nformally given by:\n\n\n9\n\nMention: \"PME\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Unemployment risk in the metropolitan labour market is computed from this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004224:26:1", "occurrence_id": "fb98f916b2835fd91dccbc18", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: >1.0<br>5.0<br>2.0<br>1.0<br>10.0<br>1.0<br>1.0|\n|**Total**|179<br>70|1 - 15|3.0|\n\n\n(*) See Footnote in Table A-1.\n\n\n**Table A - 6 : Bureaucratic Red Tape on Imports:**\n**Number of Man-days per Year Spent in Resolving Problems**\n\n**with Customs and Tax Administration**\n\n\n\n\n\n\n\n\n\n\n\n|Country|2008 Survey|Col3|Col4|2001 Survey|\n|---|---|---|---|---|\n|**Country**|**% of total**<br>**respondents**|**Range(*)**<br>**(days)**|**Average(*)**<br>**(days)**|**Average (days)**|\n|**Egypt**<br>**Jordan**<br>**Lebanon**<br>**Morocco**<br>**Saudi Arabia**<br>**Syria**<br>**Tunisia**<br>**UAE**<br>**Yemen**|84.6<br>46.8<br>88.5<br>81.8<br>89.5<br>96.2<br>57.7<br>75<br>92.6|25 - 60<br> 25 - 42<br> 20 - 50<br> 20 - 56<br> 14 - 24<br> 20 - 42<br> 18 -\n\nMention: \"2008 Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Days lost resolving customs and tax problems are reported in this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004224:26:1", "occurrence_id": "56ddf9e9caad569e49ec6477", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: >1.0<br>5.0<br>2.0<br>1.0<br>10.0<br>1.0<br>1.0|\n|**Total**|179<br>70|1 - 15|3.0|\n\n\n(*) See Footnote in Table A-1.\n\n\n**Table A - 6 : Bureaucratic Red Tape on Imports:**\n**Number of Man-days per Year Spent in Resolving Problems**\n\n**with Customs and Tax Administration**\n\n\n\n\n\n\n\n\n\n\n\n|Country|2008 Survey|Col3|Col4|2001 Survey|\n|---|---|---|---|---|\n|**Country**|**% of total**<br>**respondents**|**Range(*)**<br>**(days)**|**Average(*)**<br>**(days)**|**Average (days)**|\n|**Egypt**<br>**Jordan**<br>**Lebanon**<br>**Morocco**<br>**Saudi Arabia**<br>**Syria**<br>**Tunisia**<br>**UAE**<br>**Yemen**|84.6<br>46.8<br>88.5<br>81.8<br>89.5<br>96.2<br>57.7<br>75<br>92.6|25 - 60<br> 25 - 42<br> 20 - 50<br> 20 - 56<br> 14 - 24<br> 20 - 42<br> 18 -\n\nMention: \"2001 Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The earlier round of the same constraint question.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004225:7:1", "occurrence_id": "c49cd8ef08f53be99d98c80b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: best” case against which to compare other countries.\n\n\nThird, while most studies look at the effects of individual dimensions of the\n\n\ninvestment climate, our focus is on regressions that combine multiple dimensions in a single\n\n\nregression. This not only addresses potential bias in the estimates, it allows us to test\n\n\ndirectly which dimensions have the biggest impact on firm performance.\n\n\nA number of recent studies make use of the World Business Environment Survey\n\n\n(WBES) <sup>5</sup> —a firm-level dataset with 4,000 firms across 54 countries—to study the effect of\n\n\n4 A few studies assess differences in business environment conditions within countries. Besley and Burgess\n(2004) and Ahsan and Pagés (2009) exploit differences in labor market regulations across Indian states to\nassess how different labor market regulations affect economic performance across Indian states. Almeida and\nCarneiro (2009) assess the effect of variation of enforcement in labor regulations within Brazil.\n5 Despite their similar names, the World Business Environment Survey (WBES), a one off survey in 1999, is\nnot part of the larger and ongoing World Bank firm survey effort, the Enterprise Surveys (ES). The earlier\n\n\n6\n\nMention: \"World Business Environment Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Four thousand firms across fifty-four countries in this one-off survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004225:7:1", "occurrence_id": "79b3382fa032fdbd63c9cc24", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: best” case against which to compare other countries.\n\n\nThird, while most studies look at the effects of individual dimensions of the\n\n\ninvestment climate, our focus is on regressions that combine multiple dimensions in a single\n\n\nregression. This not only addresses potential bias in the estimates, it allows us to test\n\n\ndirectly which dimensions have the biggest impact on firm performance.\n\n\nA number of recent studies make use of the World Business Environment Survey\n\n\n(WBES) <sup>5</sup> —a firm-level dataset with 4,000 firms across 54 countries—to study the effect of\n\n\n4 A few studies assess differences in business environment conditions within countries. Besley and Burgess\n(2004) and Ahsan and Pagés (2009) exploit differences in labor market regulations across Indian states to\nassess how different labor market regulations affect economic performance across Indian states. Almeida and\nCarneiro (2009) assess the effect of variation of enforcement in labor regulations within Brazil.\n5 Despite their similar names, the World Business Environment Survey (WBES), a one off survey in 1999, is\nnot part of the larger and ongoing World Bank firm survey effort, the Enterprise Surveys (ES). The earlier\n\n\n6\n\nMention: \"Enterprise Surveys\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"The predecessor survey is distinguished from this ongoing programme.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004226:1:0", "occurrence_id": "788c50ac84eb0747ab9a7487", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Policy Research Working Paper 5033\n\n#### **Abstract**\n\nSatisfaction surveys offer a potentially convenient\nand cost-effective means for measuring the quality of\nservices. However, concerns about subjectivity and\nselection bias impede greater use of satisfaction data.\nThis paper analyzes satisfaction data about health and\neducational services from the 2006 second round\nof the Governance and Decentralization Survey in\nIndonesia to assess whether satisfaction data can serve\nas reliable indicators of quality, despite dubiously\nhigh levels of reported satisfaction. The authors use\nan expectation disconfirmation model that posits that\na user’s satisfaction with a facility improves with the\n(positive) difference between the actual quality of the\n\n\n\nfacility and the household’s expected standard for quality,\nwhich is influenced by its socioeconomic characteristics.\nThe findings show that, after taking into account the\nexpectations of households, reported satisfaction does\nvary significantly with objective indicators of quality. The\nanalysis also checks for possible selection bias affecting\nthe results by using a two-stage selection model. The\nmodel yields policy-relevant insights into the aspects of\nservice delivery that most affect satisfaction, highlights\ndifferences across rich and poor districts, and shows that\nonce the role of expectations has been factored in, the\nvariation in user satisfaction can be highly informative for\npolicymakers and researchers alike.\n\n\n\nThis paper—a product of the Poverty Reduction Group, Poverty Reduction and Economic Management Network —is\npart of a larger effort in the group to analyze poverty and monitor and evaluate the effectiveness of poverty reduction\nprograms. Policy Research Working Papers are also posted on the Web at http://econ.worldbank.org. The author may be\ncontacted at eskoufias@worldbank.org.\n\n\n_The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development_\n_issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the_\n_names of the authors\n\nMention: \"Governance and Decentralization Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its second round measures satisfaction with health and education services.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004229:74:0", "occurrence_id": "d42fb87b2297216c334a8ecb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Annex 9: Methodology to calculate agricultural potential**\n\n\nTo calculate the agricultural potential of the districts in Uganda, data collected by\nthe Food and Agricultural Organization and the International Institute for Applied\nSystems Analysis, called the Global Agro-Ecological Zones (GAEZ), was used. GAEZ\nutilizes information on climate, soil, terrain, and assumptions about inputs to construct a\nmodel of the agricultural potential of an area for 20 different crop types. The two outputs\nproduced by the GAEZ are the potential area and the potential yield. The potential area,\nin hectares, is the amount of land that can be devoted to growing the crop, and the\npotential yield is the output, in kilograms, per hectare. To calculate the total potential\noutput of each crop in each district, the sum of the potential area in a district was\nmultiplied by the average potential output of each hectare in the district. <sup>42</sup> [^42: The crops are assumed to be exclusive, such that only one crop can be grown at a time.] The resulting\nfigure is the potential output of that crop for that district, which now allows comparison\nof agricultural potential across districts. <sup>43</sup> [^43: Assumptions and generalizations were made when calculating the GAEZs, for example, the outputs used\nin this report assumes a high rainfall. The GAEZ is presented as an approximation, limitations and\nshortcomings are outlined on the website <u>http://www.iiasa.ac.at/Research/LUC/GAEZ/index.htm.</u>]\n\nThough the GAEZ is calculated for 20 crops, not all of these crops are widely\ngrown in Uganda. The literature has shown that involvement in export corps decreases\nthe likelihood of Ugandan households being poor (Balat et al. 2008), therefore, three of\nthe four crops analyzed here were chosen because of their significance as exports. The\nfourth crop, maize, was chosen for its status as\n\nMention: \"Global Agro-Ecological Zones\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Climate, soil, terrain and input assumptions are combined in this agro-ecological zoning.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004230:29:1", "occurrence_id": "284a8b1fa9e67de843af97c3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: /AIDS Epidemic: A Survey of the Households Hit by HIV/AIDS in South Africa.”\nHenry J Kaiser Family Foundation. Washington DC.\n\n\nUNICEF 2004. _The State of the World’s Children 2005: Childhood Under Threat_ .\nAccessible at <u>www.unicef.org</u>\n\nWojcicki, Janet Maia. 2005. “Socioeconomic Status as a Risk Factor for HIV Infection in\nWomen in East, Central and Southern Africa: A Systematic Review.” _Journal of_\n_Biosocial Science_ 37(1): 1-36.\n\nWooldridge, Jeffrey. 2001. _Econometric Analysis of Cross Section and Panel Data_ . The\nMIT Press.\n\nWorld Bank. 2004. “User’s Guide to the Kagera Health and Development Survey\nDatasets.” mimeo.\n\nWorld Bank. 1997. _Confronting AIDS: Public Priorities in a Global Epidemic_ . World\nBank Policy Research Report, Washington, DC.\n\n\n28\n\nMention: \"Kagera Health and Development Survey\nDatasets\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"A user's guide documents these longitudinal health datasets.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004235:5:0", "occurrence_id": "c40af75f705c67e1343377df", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The remainder of the paper is organized as follows. The next section focuses on empirical analysis\nand particularly uses a nonparametric method to infer on the nature of the association between the\ndistribution of external debt and capital flight. Section III discusses a methodological approach to\ninfer on the causality between capital flight and external debt, with emphasis on error correction\nGranger causality models. Section IV discusses empirical results and policy implications. The last\nsection concludes.\n\n**II.** **Data Analysis**\n\nThe external debt series is obtained from the World Bank Debt Table. The estimated capital flight\nseries has been kindly provided by Ndikumana and Boyce (2008). <sup>8</sup> [^8: This new series is based on a more refined methodology which adjusts the change in external debt to account\nfor debt write-offs within the Enhanced HIPC initiative. In addition, this revised series is adjusted for\nunderreporting of remittances. For further details on estimation, see Ndikumana and Boyce (2008).] This last series is constructed\nfrom three different sources: the IMF’s Trade Statistics, World Bank World Tables and IMF Balance\nof Payments Statistics. The need to reconcile these different sources to construct the capital flight\nseries arises from the difficulty of gathering accurate information on capita flight as a result of\nwidespread use of trade misinvoicing. The recourse to trade misinvoicing is partly motivated by the\ndesire to evade customs duties, import restrictions and capital controls, and has, to certain extent,\nreinforced the illicit nature of flight capital in the landscape of global capital flows. Although laborintensive, the integration of different sources of information reduces the discrepancy between\nrecorded inflows and outflows of foreign exchange. <sup>9</sup> [^9: As a result of this discrepancy, capital flight is often recorded as errors and omissions in the Balance of\nPayments statistics [IMF (2006)].]\n\nThe resulting capital flight series is very comprehensive in its coverage; it includes low and middleincome, HIPC and non-HIPC countries\n\nMention: \"World Bank Debt Table\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The external debt series is obtained from this table.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004235:5:0", "occurrence_id": "54a7e4d2f4aa5012c6f9ca3e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The remainder of the paper is organized as follows. The next section focuses on empirical analysis\nand particularly uses a nonparametric method to infer on the nature of the association between the\ndistribution of external debt and capital flight. Section III discusses a methodological approach to\ninfer on the causality between capital flight and external debt, with emphasis on error correction\nGranger causality models. Section IV discusses empirical results and policy implications. The last\nsection concludes.\n\n**II.** **Data Analysis**\n\nThe external debt series is obtained from the World Bank Debt Table. The estimated capital flight\nseries has been kindly provided by Ndikumana and Boyce (2008). <sup>8</sup> [^8: This new series is based on a more refined methodology which adjusts the change in external debt to account\nfor debt write-offs within the Enhanced HIPC initiative. In addition, this revised series is adjusted for\nunderreporting of remittances. For further details on estimation, see Ndikumana and Boyce (2008).] This last series is constructed\nfrom three different sources: the IMF’s Trade Statistics, World Bank World Tables and IMF Balance\nof Payments Statistics. The need to reconcile these different sources to construct the capital flight\nseries arises from the difficulty of gathering accurate information on capita flight as a result of\nwidespread use of trade misinvoicing. The recourse to trade misinvoicing is partly motivated by the\ndesire to evade customs duties, import restrictions and capital controls, and has, to certain extent,\nreinforced the illicit nature of flight capital in the landscape of global capital flows. Although laborintensive, the integration of different sources of information reduces the discrepancy between\nrecorded inflows and outflows of foreign exchange. <sup>9</sup> [^9: As a result of this discrepancy, capital flight is often recorded as errors and omissions in the Balance of\nPayments statistics [IMF (2006)].]\n\nThe resulting capital flight series is very comprehensive in its coverage; it includes low and middleincome, HIPC and non-HIPC countries\n\nMention: \"IMF’s Trade Statistics\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"One of three sources reconciled to build the capital flight series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004235:5:0", "occurrence_id": "6ca2f8bfca294905149ed734", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The remainder of the paper is organized as follows. The next section focuses on empirical analysis\nand particularly uses a nonparametric method to infer on the nature of the association between the\ndistribution of external debt and capital flight. Section III discusses a methodological approach to\ninfer on the causality between capital flight and external debt, with emphasis on error correction\nGranger causality models. Section IV discusses empirical results and policy implications. The last\nsection concludes.\n\n**II.** **Data Analysis**\n\nThe external debt series is obtained from the World Bank Debt Table. The estimated capital flight\nseries has been kindly provided by Ndikumana and Boyce (2008). <sup>8</sup> [^8: This new series is based on a more refined methodology which adjusts the change in external debt to account\nfor debt write-offs within the Enhanced HIPC initiative. In addition, this revised series is adjusted for\nunderreporting of remittances. For further details on estimation, see Ndikumana and Boyce (2008).] This last series is constructed\nfrom three different sources: the IMF’s Trade Statistics, World Bank World Tables and IMF Balance\nof Payments Statistics. The need to reconcile these different sources to construct the capital flight\nseries arises from the difficulty of gathering accurate information on capita flight as a result of\nwidespread use of trade misinvoicing. The recourse to trade misinvoicing is partly motivated by the\ndesire to evade customs duties, import restrictions and capital controls, and has, to certain extent,\nreinforced the illicit nature of flight capital in the landscape of global capital flows. Although laborintensive, the integration of different sources of information reduces the discrepancy between\nrecorded inflows and outflows of foreign exchange. <sup>9</sup> [^9: As a result of this discrepancy, capital flight is often recorded as errors and omissions in the Balance of\nPayments statistics [IMF (2006)].]\n\nThe resulting capital flight series is very comprehensive in its coverage; it includes low and middleincome, HIPC and non-HIPC countries\n\nMention: \"World Bank World Tables\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"A second source in that reconciliation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004235:5:0", "occurrence_id": "287e9422f7103141fb2124d8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The remainder of the paper is organized as follows. The next section focuses on empirical analysis\nand particularly uses a nonparametric method to infer on the nature of the association between the\ndistribution of external debt and capital flight. Section III discusses a methodological approach to\ninfer on the causality between capital flight and external debt, with emphasis on error correction\nGranger causality models. Section IV discusses empirical results and policy implications. The last\nsection concludes.\n\n**II.** **Data Analysis**\n\nThe external debt series is obtained from the World Bank Debt Table. The estimated capital flight\nseries has been kindly provided by Ndikumana and Boyce (2008). <sup>8</sup> [^8: This new series is based on a more refined methodology which adjusts the change in external debt to account\nfor debt write-offs within the Enhanced HIPC initiative. In addition, this revised series is adjusted for\nunderreporting of remittances. For further details on estimation, see Ndikumana and Boyce (2008).] This last series is constructed\nfrom three different sources: the IMF’s Trade Statistics, World Bank World Tables and IMF Balance\nof Payments Statistics. The need to reconcile these different sources to construct the capital flight\nseries arises from the difficulty of gathering accurate information on capita flight as a result of\nwidespread use of trade misinvoicing. The recourse to trade misinvoicing is partly motivated by the\ndesire to evade customs duties, import restrictions and capital controls, and has, to certain extent,\nreinforced the illicit nature of flight capital in the landscape of global capital flows. Although laborintensive, the integration of different sources of information reduces the discrepancy between\nrecorded inflows and outflows of foreign exchange. <sup>9</sup> [^9: As a result of this discrepancy, capital flight is often recorded as errors and omissions in the Balance of\nPayments statistics [IMF (2006)].]\n\nThe resulting capital flight series is very comprehensive in its coverage; it includes low and middleincome, HIPC and non-HIPC countries\n\nMention: \"IMF Balance\nof Payments Statistics\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"A third source in that reconciliation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004236:19:2", "occurrence_id": "7a0d2b0870bea11c690bd50d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: sup>_d_</sup>\n\n_jf_ <sup>,</sup> <sup>and</sup>\n\n\n\n_e_ <sup>_d_</sup>\n\n\n\n_<u>jz</u>_\n_e_ <sup>_d_</sup>\n\n_jf_\n\n\n\native factor endowments (capital/labor), which we use to proxy _<u>jz</u>_\n\n_e_ <sup>_d_</sup>\n\n_jf_ <sup>,</sup> <sup>and</sup>\nthe same variables for technology trading partners. The latter variables are\n\nweighted by the inverse of the distance between trading countries. To test the\n\nrobustness of our specification, we add factor price ratio, namely the ratio of\n\nthe manufacturing wages over bank lending interest rates, and national TFP\nlevels to control for Hicks-neutral technological differences. Our proxy for _Rt_ <sup>_c_</sup>\n\nis the first component in the principal component analysis of the variables\nemployed to construct variable _Rt_ <sup>_c_</sup> <sup>.</sup>\n\n\n\n_t_ <sup>_c_</sup> <sup>.</sup>\n\n\n\n**5.2.2** **Data**\n\n\n**Factor** **endowments**\n\n\nData on capital stocks come from Serven and Calder´ _o_ n[33], who extend the\n\nseries provided by the Penn World Tables. The labor force is from the In\nternational Labor Organization (ILO), and it refers to economically active\n\npopulation defined as the 25-64 age group. To calculate endowments of high\nand low\n\nMention: \"Data on capital stocks\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Extended national series provide the capital endowment variable.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004236:19:2", "occurrence_id": "ff653c7367ec040f6c50f09b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: sup>_d_</sup>\n\n_jf_ <sup>,</sup> <sup>and</sup>\n\n\n\n_e_ <sup>_d_</sup>\n\n\n\n_<u>jz</u>_\n_e_ <sup>_d_</sup>\n\n_jf_\n\n\n\native factor endowments (capital/labor), which we use to proxy _<u>jz</u>_\n\n_e_ <sup>_d_</sup>\n\n_jf_ <sup>,</sup> <sup>and</sup>\nthe same variables for technology trading partners. The latter variables are\n\nweighted by the inverse of the distance between trading countries. To test the\n\nrobustness of our specification, we add factor price ratio, namely the ratio of\n\nthe manufacturing wages over bank lending interest rates, and national TFP\nlevels to control for Hicks-neutral technological differences. Our proxy for _Rt_ <sup>_c_</sup>\n\nis the first component in the principal component analysis of the variables\nemployed to construct variable _Rt_ <sup>_c_</sup> <sup>.</sup>\n\n\n\n_t_ <sup>_c_</sup> <sup>.</sup>\n\n\n\n**5.2.2** **Data**\n\n\n**Factor** **endowments**\n\n\nData on capital stocks come from Serven and Calder´ _o_ n[33], who extend the\n\nseries provided by the Penn World Tables. The labor force is from the In\nternational Labor Organization (ILO), and it refers to economically active\n\npopulation defined as the 25-64 age group. To calculate endowments of high\nand low\n\nMention: \"Penn World Tables\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The underlying capital series before extension.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004244:14:2", "occurrence_id": "3270b776f4a8d304ab10dd28", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  to 6-digit HS Comtrade 2007 import data available via\n\nWITS. Intra-EC trade is excluded.\n\n\nNot surprisingly, table 4 first reveals that the two economies with the largest _levels_ of imports\n\n\npotentially affected by the remedies are the EC ($8.0 billion) and the U.S. ($7.5 billion). However, the\n\n\n_relative_ importance of the targeted imports is fairly small, as only 0.4 per cent of 2007 imports in these\n\n\n13\n\nMention: \"HS Comtrade 2007 import data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Six-digit import flows accessed through the trade portal are used here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004254:15:0", "occurrence_id": "048c4ce83874c521a4ac62cc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **3. Conclusions**\n\n\n1. All international and global inequality concepts show a higher degree of\n\n\ninequality with the new 2005 PPP data than with the extrapolation based on the\n\n\nprevious round of the ICP. This is very significant since the new data include\n\n\nChina (first time ever) and India (first time since 1985).\n\n\n2. After 20 years of mean-income (GDP per capita) divergence, GDPs per capita of\n\n\nthe countries of the world have begun a process of convergence since 2001. This\n\n\nis due to the pick-up of growth in Africa, post-Communist countries and Latin\n\n\nAmerica. It is unclear how the global crisis will affect this process.\n\n\n3. Population-weighted (Concept 2) inequality, which has gone down during the last\n\n20 years of the 20 <sup>th</sup> century thanks to the high growth rates of China, shows, after\n\n\nyear 2000, a decline even when China is excluded. This is due to the high growth\n\n\nrates of India. Uncertainly as to the effects of the crisis remains there too although\n\n\nto the extent that the “locomotives” of Concept 2 decline, China and India\n\n\ncontinue to expand at a higher (per capita) growth rate than the rest of the world,\n\nConcept 2 may go on with its downward trend. <sup>16</sup> [^16: Up to a point, quite remote now (and made remoter after the 2005 ICP results) when China’s growth—\ndue to its having become a relatively rich country—turn to be globally disequalizing.]\n\n\n4. Global inequality between individuals amounts to more than 70 rather than 65\n\n\nGini points as when calculated with the “old” PPPs. This is a significant upward\n\n\nrevision. The trend, over the period 1988-2002, however, is absent as the global\n\n\nGini bounces between 68.4 and almost 71 Gini points. Its further evolution will,\n\n\nof course, depend on growth rates of China, India and the US (the “triangle”\n\nwhich\n\nMention: \"ICP\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Its newer round adds two large economies absent from the extrapolation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004255:11:1", "occurrence_id": "7926e7317852e7d3b0c28ab0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  to add, such\n\n\nknowledge may only be useful if it can be applied. For example, a mother may not be able to\n\n\nput her knowledge about nutritional requirements into practice, as her and the child’s diet is\n\ndetermined by largely unchangeable agricultural and economic conditions. These two aspects\n\n\nmay partly explain the variation in the findings on the effects of mother’s education.\n\nAnother socio-economic factor of particular interest in this paper is material well-being,\n\n\ntypically measured by consumption, income, and/or wealth. In fact, as almost all the reviewed\n\n\nstudies rely on DHS data, researchers are forced to resume to an asset indicator as a proxy for\n\nmaterial wealth. This asset indicator is typically computed using principal components\n\nanalysis on the basis of a list of assets, as suggested by Filmer and Pritchett (2001). When\n\n\nincluded, wealth, as expected, tends to increase the chances of child survival (for example\n\nHill et al., 2001) or infant survival (for example Ssewanyana and Younger, 2007) –\n\nindependent of other factors included in the regression. Yet, for infant mortality however, this\n\n\ndoes not turn to be out a universal finding. The meta-study by Charmarbagwala et al. (2005)\n\nsuggests that, although there can be little doubt that household income is a crucial factor in\n\n\ndetermining child health, it appears that income is not a significant determinant of infant\n\n\nmortality in the majority of cases. This can partly be explained by the fact that as mortality\n\nfalls, the bulk of under-five-mortality is infant rather than child death, and these deaths are\n\n\n9\n\nMention: \"DHS data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Almost all reviewed studies rely on it, forcing use of an asset proxy.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004263:9:0", "occurrence_id": "966e9d4ab9a6d787a37c9ebd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Distributional Impact Analysis of Climate Variability in Rural Indonesia\n\n\nconsumption quintile. These 12 datasets were then combined into a single dataset consisting of 118,500\nhouseholds.\n\n**Both datasets were matched to local daily rainfall data.** Rainfall data is taken from the NOAA\n(National Oceanic and Atmospheric Administration) Global Summary of the Day combined with\nadditional data obtained from the Indonesian Meteorological Agency ( _Badan Meteorologi dan Geofisika,_\n_BMG_ ). Imputed values for the missing values in the Global Summary of the Day data were provided by\nCEREGE, _Centre Europeen de Recherche et d’Enseignement des Geosciences de l’Environnement_ .\nRainfall data contains daily rainfall data for 52 stations, of which 36 stations match with the IFLS data\nand 49 stations match with the SUSENAS data. In the IFLS data households were matched with the\nclosest weather station at the community level whereas in the SUSENAS they were matched at the\nkabupaten (district) level.\n\n**The definition of monsoon onset is based on daily rainfall data.** The start of the monsoon is defined as\nthe number of days past August 1 that cumulative rainfall exceeds 200 mm, following Naylor _et al._\n(2007). For each station, we calculated the start date of monsoon defined as the number of days past\nAugust 1 when cumulative rainfall exceeds 200 mm. Onset was then standardized using each station’s\n“leave-out” mean and standard deviation across years. In other words, data from the onset year was\nexcluded when calculating the mean and standard deviation used to standardize each year’s onset.\n\n**Monsoon onset is sufficiently variable across Indonesia to allow for detailed analysis.** Annex 2\ndisplays district-level maps of Indonesia that indicate the timing of monsoon onset from 1990 to 1999.\nThese maps show that the 36 rain stations used to match the IFLS data generate meaningful\n\nMention: \"Global Summary of the Day\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Daily rainfall readings come from this station summary.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004263:9:0", "occurrence_id": "34e57ce5ff78acdc63772f78", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Distributional Impact Analysis of Climate Variability in Rural Indonesia\n\n\nconsumption quintile. These 12 datasets were then combined into a single dataset consisting of 118,500\nhouseholds.\n\n**Both datasets were matched to local daily rainfall data.** Rainfall data is taken from the NOAA\n(National Oceanic and Atmospheric Administration) Global Summary of the Day combined with\nadditional data obtained from the Indonesian Meteorological Agency ( _Badan Meteorologi dan Geofisika,_\n_BMG_ ). Imputed values for the missing values in the Global Summary of the Day data were provided by\nCEREGE, _Centre Europeen de Recherche et d’Enseignement des Geosciences de l’Environnement_ .\nRainfall data contains daily rainfall data for 52 stations, of which 36 stations match with the IFLS data\nand 49 stations match with the SUSENAS data. In the IFLS data households were matched with the\nclosest weather station at the community level whereas in the SUSENAS they were matched at the\nkabupaten (district) level.\n\n**The definition of monsoon onset is based on daily rainfall data.** The start of the monsoon is defined as\nthe number of days past August 1 that cumulative rainfall exceeds 200 mm, following Naylor _et al._\n(2007). For each station, we calculated the start date of monsoon defined as the number of days past\nAugust 1 when cumulative rainfall exceeds 200 mm. Onset was then standardized using each station’s\n“leave-out” mean and standard deviation across years. In other words, data from the onset year was\nexcluded when calculating the mean and standard deviation used to standardize each year’s onset.\n\n**Monsoon onset is sufficiently variable across Indonesia to allow for detailed analysis.** Annex 2\ndisplays district-level maps of Indonesia that indicate the timing of monsoon onset from 1990 to 1999.\nThese maps show that the 36 rain stations used to match the IFLS data generate meaningful\n\nMention: \"IFLS data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Households here are matched to the nearest station at community level.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004263:9:0", "occurrence_id": "273de8196adc9ebe88054236", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Distributional Impact Analysis of Climate Variability in Rural Indonesia\n\n\nconsumption quintile. These 12 datasets were then combined into a single dataset consisting of 118,500\nhouseholds.\n\n**Both datasets were matched to local daily rainfall data.** Rainfall data is taken from the NOAA\n(National Oceanic and Atmospheric Administration) Global Summary of the Day combined with\nadditional data obtained from the Indonesian Meteorological Agency ( _Badan Meteorologi dan Geofisika,_\n_BMG_ ). Imputed values for the missing values in the Global Summary of the Day data were provided by\nCEREGE, _Centre Europeen de Recherche et d’Enseignement des Geosciences de l’Environnement_ .\nRainfall data contains daily rainfall data for 52 stations, of which 36 stations match with the IFLS data\nand 49 stations match with the SUSENAS data. In the IFLS data households were matched with the\nclosest weather station at the community level whereas in the SUSENAS they were matched at the\nkabupaten (district) level.\n\n**The definition of monsoon onset is based on daily rainfall data.** The start of the monsoon is defined as\nthe number of days past August 1 that cumulative rainfall exceeds 200 mm, following Naylor _et al._\n(2007). For each station, we calculated the start date of monsoon defined as the number of days past\nAugust 1 when cumulative rainfall exceeds 200 mm. Onset was then standardized using each station’s\n“leave-out” mean and standard deviation across years. In other words, data from the onset year was\nexcluded when calculating the mean and standard deviation used to standardize each year’s onset.\n\n**Monsoon onset is sufficiently variable across Indonesia to allow for detailed analysis.** Annex 2\ndisplays district-level maps of Indonesia that indicate the timing of monsoon onset from 1990 to 1999.\nThese maps show that the 36 rain stations used to match the IFLS data generate meaningful\n\nMention: \"SUSENAS data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"District-level matching is applied for this survey instead.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004263:9:0", "occurrence_id": "4191225448bf5d2b5f69ed46", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Distributional Impact Analysis of Climate Variability in Rural Indonesia\n\n\nconsumption quintile. These 12 datasets were then combined into a single dataset consisting of 118,500\nhouseholds.\n\n**Both datasets were matched to local daily rainfall data.** Rainfall data is taken from the NOAA\n(National Oceanic and Atmospheric Administration) Global Summary of the Day combined with\nadditional data obtained from the Indonesian Meteorological Agency ( _Badan Meteorologi dan Geofisika,_\n_BMG_ ). Imputed values for the missing values in the Global Summary of the Day data were provided by\nCEREGE, _Centre Europeen de Recherche et d’Enseignement des Geosciences de l’Environnement_ .\nRainfall data contains daily rainfall data for 52 stations, of which 36 stations match with the IFLS data\nand 49 stations match with the SUSENAS data. In the IFLS data households were matched with the\nclosest weather station at the community level whereas in the SUSENAS they were matched at the\nkabupaten (district) level.\n\n**The definition of monsoon onset is based on daily rainfall data.** The start of the monsoon is defined as\nthe number of days past August 1 that cumulative rainfall exceeds 200 mm, following Naylor _et al._\n(2007). For each station, we calculated the start date of monsoon defined as the number of days past\nAugust 1 when cumulative rainfall exceeds 200 mm. Onset was then standardized using each station’s\n“leave-out” mean and standard deviation across years. In other words, data from the onset year was\nexcluded when calculating the mean and standard deviation used to standardize each year’s onset.\n\n**Monsoon onset is sufficiently variable across Indonesia to allow for detailed analysis.** Annex 2\ndisplays district-level maps of Indonesia that indicate the timing of monsoon onset from 1990 to 1999.\nThese maps show that the 36 rain stations used to match the IFLS data generate meaningful\n\nMention: \"SUSENAS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Its coarser geographic matching limits precision.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004269:27:1", "occurrence_id": "2fbbf6bc88f6eab1674e92a9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  some elements of _x_ are binary. Robust z-stat in parentheses.\nOmitted dummies are no education, female, single and catholic. The type of housing materials\n(soft roof and house, mixed or hard roof and house) serves as a proxy for wealth. The indicator\nfor mobility takes the value 1 if the individual goes more than once a year outside the country.\nThe sample size decreases when age dummies are included since at very old age nobody is HIV\npositive.\n_Source_ : MRC General Population Cohort 1989/90 (round 1).\n\n\n\n26\n\nMention: \"MRC General Population Cohort\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The first round of this cohort gives age-specific infection prevalence.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004274:7:0", "occurrence_id": "57623333bb64acb732ddd63f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: we try to assess how much of this concentration is driven by technological dispersion versus\n\n\ndemand. Eaton, Kortum, and Kramarz (2008) also relate trade patterns to productivity and\n\n\ndemand shocks. But while they dissect trading patterns only for French …rms, regardless\n\n\nof which products each …rm exports (there could be more than one product per …rm), we\n\n\nanalyze trade at the product level for many countries. <sup>6</sup> [^6: The distribution of exports across products is similar to what they …nd for French …rms.]\n\n\nIn the next section we document concentration and distributions of exports for 151\n\n\ncountries in the product-destination dimension and perform preliminary analysis. In sec\n\ntion 3 we estimate the contribution of technology versus demand to the distribution and\n\n\nconcentration of exports. Section 4 concludes.\n\n###### 2 Empirical Facts\n\n\nOur main data source is the UN Comtrade database. The U.N. classi…es exported commodi\n\nties and manufactured products by source and destination at the six-digit level (roughly\n\n\n5000 categories). We use the 1992 Harmonized System classi…cation (HS1992) for the year\n\n\n2000, to maximize the available bilateral trade pairs. Using a less disaggregated classi…ca\n\ntion might have lead to better coverage of countries (say, 4-digit SITC), but would miss the\n\n\nextreme concentration within …nely de…ned products. <sup>7</sup> [^7: An analysis of the distribution of product-destination export ‡ows at the 4-digit SITC level reveals\nsimilar patterns, but lower levels of concentration, as one might expect.]\n\n\nWe restrict our sample to manufactured categories, i.e. we drop from the sample all\n\n\nagriculture and commodities exports. Our focus on manufactured products stems from our\n\n\ninterest on exports that are not dependent on country-speci…c natural endowments, and\n\n\ncould potentially be produced everywhere in the world. We basically exclude products that\n\n\nrely directly on natural resources. Natural resources create strong comparative advantage\n\n\nfor extracables\n\nMention: \"the UN Comtrade database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Six-digit commodity flows by source and destination are the main data here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004275:36:1", "occurrence_id": "123081e294846030819f149b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  (-0.75) (0.57) (2.64) (-1.02)\n6-7 years later 0.152* 0.162* -0.003 0.102* 0.166** -0.717\n(2.32) (2.47) (-0.08) (2.30) (2.63) (-0.32)\n8-10 years later 0.306*** 0.331*** 0.116* 0.069 0.215** 3.157\n(3.74) (4.01) (2.31) (1.23) (2.72) (1.26)\n11 + years later 0.301** 0.319** 0.04 0.146* 0.279** -5.773\n(3.02) (3.18) (0.65) (2.14) (2.90) (-1.76)\n\nSample size 6253 6253 6253 5822 6252 5150\n\nSource: Linked ICS-ENIA panel data\nNote:   (1) The regression model included indicator variables for location in the capital region,\nfirm size, and 12 year dummy variables.\n(2) ***, ** and * denote statistical significance at the 1, 5 and 10 percent level\n\n\n\n35\n\nMention: \"ICS-ENIA panel data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Firms are linked across rounds for the capital-region and size regressions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004285:10:1", "occurrence_id": "59fde6f9974537652f9391be", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  El Alto -0.27 -10.54 0.000 639\nOruro -0.19 -5.94 0.000 547\nPuerto Suarez 0.10 2.38 0.018 354\nRiberalta 0.03 1.26 0.207 557\nRoboré -0.06 -1.45 0.148 467\nRurrenabaque -0.02 -0.80 0.427 557\nSan Ignacio de Velasco 0.22 5.90 0.000 458\nSan Joaquin -0.01 -0.44 0.658 448\nSanta Cruz / El Trompillo 0.04 1.16 0.247 557\nSucre -0.28 -5.65 0.000 321\nTarija -0.11 -1.71 0.089 318\nTrinidad -0.05 -1.75 0.081 561\n<u>Yacuiba</u> <u>-0.14</u> <u>-3.34</u> <u>0.001</u> <u>527</u>\n\n_Source:_ Authors’ estimation based on data from the NCDC’s Monthly Climatic Data for the\nWorld.\n\nNone of the stations in Bolivia get even close to having observations for all the 721 months\nin the 1948-2008 period, but some stations have reported more consistently than others. If\nwe limit ourselves to the 15 stations that have at least 400 observations, we find that three\nshow significant warming, four show significant cooling, and eight show no significant\ntrend. A similar distribution is found if we limit ourselves to the 11 stations that have at\nleast 500 observations.\n\n\n9\n\nMention: \"Monthly Climatic Data for the\nWorld\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Station-month climate observations are drawn from this publication.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004286:43:0", "occurrence_id": "284e961e6bcd30daf3101ac2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **ANNEX**\n\n**Annex Figure I: Map of Treated and Control Districts**\n\n\nNotes: The lighter regions represent lower poverty levels, and darker regions represent higher poverty\nlevels.\nSource: Author’s calculations using national poverty estimates\n\n\n42\n\nMention: \"national poverty estimates\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Subnational rates are shaded on the map from these estimates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004290:9:3", "occurrence_id": "6900841799aab33348df5d9c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  0.5 0.8 0.7 1.7 1.1\nOwnership of dwellings 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.1 0.0\n<u>Government services</u> <u>1.0</u> <u>-0.3</u> <u>0.2</u> <u>-0.5</u> <u>-0.2</u> <u>-1.0</u> <u>0.1</u> <u>0.0</u> <u>0.2</u> <u>0.2</u> <u>0.1</u> <u>0.3</u> <u>0.6</u>\n_Source: NSO 1994 series of accounts. GDP is from 2002 series back-adjusted using the 1994 series rates of growth._\n\n**Maize Production**\nAgricultural production and policy is dominated by maize. Although there has been\nsome diversification and a notable increase in smallholder tobacco production, the area\nunder maize cultivation is still roughly equivalent to that of all other crops combined.\nOver 60 percent of national calorie consumption derives from maize; 97 percent of\nfarmers grow maize; and over half of households grow no other crop <sup>4</sup> [^4: Integrated Household Survey, 2005 and Poverty and Vulnerability Assessment, World Bank, 2006.] . Given Malawi’s\nvulnerability to drought, the maize harvest is thus central to the welfare of the population,\nand one of the most important political and social variables. This is reflected by the\nprominence of food security in public policy. Through the Fertilizer Subsidy Program,\nthe government has made increasing maize\n\nMention: \"NSO 1994 series of accounts\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Growth rates here back-adjust the newer output series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004290:9:3", "occurrence_id": "b57f971a78f6e41fe9f6ed37", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  0.5 0.8 0.7 1.7 1.1\nOwnership of dwellings 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.1 0.0\n<u>Government services</u> <u>1.0</u> <u>-0.3</u> <u>0.2</u> <u>-0.5</u> <u>-0.2</u> <u>-1.0</u> <u>0.1</u> <u>0.0</u> <u>0.2</u> <u>0.2</u> <u>0.1</u> <u>0.3</u> <u>0.6</u>\n_Source: NSO 1994 series of accounts. GDP is from 2002 series back-adjusted using the 1994 series rates of growth._\n\n**Maize Production**\nAgricultural production and policy is dominated by maize. Although there has been\nsome diversification and a notable increase in smallholder tobacco production, the area\nunder maize cultivation is still roughly equivalent to that of all other crops combined.\nOver 60 percent of national calorie consumption derives from maize; 97 percent of\nfarmers grow maize; and over half of households grow no other crop <sup>4</sup> [^4: Integrated Household Survey, 2005 and Poverty and Vulnerability Assessment, World Bank, 2006.] . Given Malawi’s\nvulnerability to drought, the maize harvest is thus central to the welfare of the population,\nand one of the most important political and social variables. This is reflected by the\nprominence of food security in public policy. Through the Fertilizer Subsidy Program,\nthe government has made increasing maize\n\nMention: \"Integrated Household Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"It shows nearly all farmers grow maize and most grow nothing else.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004294:11:2", "occurrence_id": "0e4ff6b0c929ad1ee4cc6f86", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: KOR<br>BRA<br>IND<br>CHN<br>V.A. per Worker (Korea=100)- product 371<br>0<br>50<br>100<br>150<br>200<br>1980<br>1985<br>1990<br>1995<br>2000<br>2005<br>KOR<br>BRA<br>IND<br>CHN<br>V.A. per Worker (Current '000 US$/worker) - product 371<br>Product: Iron and steel<br>Source: UNIDO Industrial Statistics Database, Rev. 2 and 3|\n||0<br>100<br>200<br>300<br>400<br>1980<br>1985<br>1990<br>1995<br>2000<br>2005<br>KOR<br>BRA<br>IND<br>CHN<br>V.A. per Worker (Korea=100)- product 322<br>0<br>10<br>20<br>30<br>40<br>1980<br>1985<br>1990<br>1995<br>2000<br>2005<br>KOR<br>BRA<br>IND<br>CHN<br>V.A. per Worker (Current '000 US$/worker) - product 322<br>Product: Wearing apparel, except footwear<br>Source: UNIDO Industrial Statistics Database, Rev. 2 and 3||0<br>20<br>40<br>60<br>80<br>100<br>1980<br>1985<br>1990<br>1995<br>2000<br>2005<br>KOR<br>BRA<\n\nMention: \"UNIDO Industrial Statistics Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Value added per worker by product across these decades comes from it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004296:10:1", "occurrence_id": "00bfd0008cf272bb3f39c9b0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  to detect the true effect of aggregate growth on\n\n\npoverty. There are also long-standing concerns about the rising gap between aggregate\n\n\nhousehold consumption as measured from the NSS and the “private consumption” component of\n\n\ndomestic absorption in the national accounts (NAS). New problems also emerge in the post-1991\n\n\nperiod, including changes in survey design that we address later.\n\n\nNotwithstanding these issues, we believe there is now sufficient data for the “post\n\nreform” period to revisit the question of whether India’s higher growth rates have delivered the\n\n\n19 Equation (6.3) need not be estimated separately since the parameters can be inferred from the\nestimates of (6.1) and (6.2) and (4) using the adding-up restriction.\n20 For example, the first survey (1992) available in the post-reform period indicated a substantial\nincrease in poverty and this fuelled much debate about the wisdom of reforms. We questioned this\ninference at the time arguing that the 1992 survey was deceptive about trends (Datt and Ravallion, 1997).\n\n9\n\nMention: \"NSS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Household consumption measured here diverges from the accounts.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004296:10:1", "occurrence_id": "39b88cfad00273c1e48d0a2e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  to detect the true effect of aggregate growth on\n\n\npoverty. There are also long-standing concerns about the rising gap between aggregate\n\n\nhousehold consumption as measured from the NSS and the “private consumption” component of\n\n\ndomestic absorption in the national accounts (NAS). New problems also emerge in the post-1991\n\n\nperiod, including changes in survey design that we address later.\n\n\nNotwithstanding these issues, we believe there is now sufficient data for the “post\n\nreform” period to revisit the question of whether India’s higher growth rates have delivered the\n\n\n19 Equation (6.3) need not be estimated separately since the parameters can be inferred from the\nestimates of (6.1) and (6.2) and (4) using the adding-up restriction.\n20 For example, the first survey (1992) available in the post-reform period indicated a substantial\nincrease in poverty and this fuelled much debate about the wisdom of reforms. We questioned this\ninference at the time arguing that the 1992 survey was deceptive about trends (Datt and Ravallion, 1997).\n\n9\n\nMention: \"NAS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its private absorption component is the benchmark for that gap.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004302:4:1", "occurrence_id": "edb93261d810e4be7c6c44ed", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  more pressing. This paper contributes to that task by assessing PIL with\nempirical data.\n\n\nThe next section of this paper analyzes the argument that PIL constitutes a case of\njudicial overreach. The contention that PIL weakens policy formulation and\nimplementation in the legislative and executive branches is typically “dressed up” as a\nseparation of powers concern, but a more apt framework involves an assessment of the\nimpact of PIL on sectoral governance, which is fundamentally an empirical matter, not a\ndoctrinal one. The following section describes the charge that PIL favors middle class\ninterests rather than the concerns of the poor and marginalized. That section than presents\nestimates, based on original data taken from Supreme Court records an online legal\ndatabase, to assess that claim.\n\n\n_Public Interest Litigation: Is the Judiciary Overreaching?_\n\n\nAn old-fashioned view of legal rights holds that most social and economic matters\ndo not involve genuine rights because they require positive actions, not merely restraint,\nand have no single, identifiable duty holder. Positive obligations, moreover, entail\nsignificant expenditures that are the purview of the other branches of government. Courts,\ntherefore, should steer clear of the social, economic, and environmental concerns at the\nheart of PIL. More contemporary views (Holmes & Sunstein, 1999; Shue, 1996) hold that\n“for their fulfillment all rights require restraint, protection, and aid from the entity from\n\n\n3\n\nMention: \"Supreme Court records\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Original case data are compiled from these court records.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004302:4:1", "occurrence_id": "25b35885b09f869d0a1c5876", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  more pressing. This paper contributes to that task by assessing PIL with\nempirical data.\n\n\nThe next section of this paper analyzes the argument that PIL constitutes a case of\njudicial overreach. The contention that PIL weakens policy formulation and\nimplementation in the legislative and executive branches is typically “dressed up” as a\nseparation of powers concern, but a more apt framework involves an assessment of the\nimpact of PIL on sectoral governance, which is fundamentally an empirical matter, not a\ndoctrinal one. The following section describes the charge that PIL favors middle class\ninterests rather than the concerns of the poor and marginalized. That section than presents\nestimates, based on original data taken from Supreme Court records an online legal\ndatabase, to assess that claim.\n\n\n_Public Interest Litigation: Is the Judiciary Overreaching?_\n\n\nAn old-fashioned view of legal rights holds that most social and economic matters\ndo not involve genuine rights because they require positive actions, not merely restraint,\nand have no single, identifiable duty holder. Positive obligations, moreover, entail\nsignificant expenditures that are the purview of the other branches of government. Courts,\ntherefore, should steer clear of the social, economic, and environmental concerns at the\nheart of PIL. More contemporary views (Holmes & Sunstein, 1999; Shue, 1996) hold that\n“for their fulfillment all rights require restraint, protection, and aid from the entity from\n\n\n3\n\nMention: \"online legal\ndatabase\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"It supplies the companion case information.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004303:26:1", "occurrence_id": "9c9d3aea2bd74022cb3705ad", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: -2006), Urban Agglomeration\n3. World Gazzetteer, relevant year (1960-2006), City Proper\n4. World Gazzetteer, relevant year (1960-2006), Urban Agglomeration\n5. City Population, relevant year (1960-2006), City Proper\n6. City Population, relevant year (1960-2006), Urban Agglomeration\n\n<u>Variable Construction:</u>\nDue to the greater scope, and likely reliability of UN data over other online population sources,\npriority was given to UNDYB census data. Where these values were missing, WG and CP sites\nwere consulted. Encouragingly, the data provided by these two sites very often conferred with\nvalues for UNDYB data, indicating they are drawn from similar data sources and may at least\nprovide some similar level of valid measurement of urban population. When constructing the\ndataset, we gathered data on city size which would reflect the area which incidents of urban\nviolence/protest would be recorded, and which best reflect the size and scale of the city. In many\ncases, the metropolitan area would certainly underestimate the size of the area and population\nwhich would be reported for. A good example of this is Manila. In many cases also, estimates for\nurban agglomeration would overreport the area and population, such as in Tokyo, whose\nagglomeration with Yokohama stands at around 37 million, considerably larger than the 12\nmillion in the Tokyo agglomeration. For some cities measures of the metropolitan area and urban\nagglomeration from different sources were used. This was done to increase the number of\n\n\n25\n\nMention: \"World Gazzetteer\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its city and agglomeration counts fill gaps in the official series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004303:26:1", "occurrence_id": "b864a1ca38ffd889daf1b080", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: -2006), Urban Agglomeration\n3. World Gazzetteer, relevant year (1960-2006), City Proper\n4. World Gazzetteer, relevant year (1960-2006), Urban Agglomeration\n5. City Population, relevant year (1960-2006), City Proper\n6. City Population, relevant year (1960-2006), Urban Agglomeration\n\n<u>Variable Construction:</u>\nDue to the greater scope, and likely reliability of UN data over other online population sources,\npriority was given to UNDYB census data. Where these values were missing, WG and CP sites\nwere consulted. Encouragingly, the data provided by these two sites very often conferred with\nvalues for UNDYB data, indicating they are drawn from similar data sources and may at least\nprovide some similar level of valid measurement of urban population. When constructing the\ndataset, we gathered data on city size which would reflect the area which incidents of urban\nviolence/protest would be recorded, and which best reflect the size and scale of the city. In many\ncases, the metropolitan area would certainly underestimate the size of the area and population\nwhich would be reported for. A good example of this is Manila. In many cases also, estimates for\nurban agglomeration would overreport the area and population, such as in Tokyo, whose\nagglomeration with Yokohama stands at around 37 million, considerably larger than the 12\nmillion in the Tokyo agglomeration. For some cities measures of the metropolitan area and urban\nagglomeration from different sources were used. This was done to increase the number of\n\n\n25\n\nMention: \"City Population\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"A second fallback source for the same two definitions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004303:26:1", "occurrence_id": "905741b9c722b4606fb31917", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: -2006), Urban Agglomeration\n3. World Gazzetteer, relevant year (1960-2006), City Proper\n4. World Gazzetteer, relevant year (1960-2006), Urban Agglomeration\n5. City Population, relevant year (1960-2006), City Proper\n6. City Population, relevant year (1960-2006), Urban Agglomeration\n\n<u>Variable Construction:</u>\nDue to the greater scope, and likely reliability of UN data over other online population sources,\npriority was given to UNDYB census data. Where these values were missing, WG and CP sites\nwere consulted. Encouragingly, the data provided by these two sites very often conferred with\nvalues for UNDYB data, indicating they are drawn from similar data sources and may at least\nprovide some similar level of valid measurement of urban population. When constructing the\ndataset, we gathered data on city size which would reflect the area which incidents of urban\nviolence/protest would be recorded, and which best reflect the size and scale of the city. In many\ncases, the metropolitan area would certainly underestimate the size of the area and population\nwhich would be reported for. A good example of this is Manila. In many cases also, estimates for\nurban agglomeration would overreport the area and population, such as in Tokyo, whose\nagglomeration with Yokohama stands at around 37 million, considerably larger than the 12\nmillion in the Tokyo agglomeration. For some cities measures of the metropolitan area and urban\nagglomeration from different sources were used. This was done to increase the number of\n\n\n25\n\nMention: \"UNDYB census data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Given greater reliability, these values take priority over online sources.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004307:9:0", "occurrence_id": "e08e84c1fde69e0839f32fec", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Breaking the Waves? Education, Youth Bulges, and Armed Conflict\n\n\n_3.1 IIASA Human Capital Data_\n\n\nEducation data originate from a new dataset compiled by researchers at IIASA (Lutz et al. 2007). The dataset\n\n\ncontains information from 120 countries for the 1970-2000 period, based on individual-level educational attainment\n\n\ndata from Demographic Health Surveys and national censuses, back projected using multi-state demographic\n\n\nmodeling. The dataset measures educational attainment as opposed to more commonly used education measures like\n\n\neducational spending or enrollment rates. It covers a great number of developing countries, including many\n\n\ncountries with recent conflict experiences.\n\n\nThe IIASA data consists of population estimates by age, gender and highest educational attainment (4\n\n\ncategories: no schooling, some primary, completed lower secondary, completed tertiary), covering the period from\n\n\n1970 to 2000, for 120 countries/territories (discrepancies between this figure and others below arise from the fact\n\n\nthat more countries have been added since the publication of Lutz et al. (2007) and the introduction of splits and\n\n\nmergers during data matching). It is discussed in some detail here because of the novel methodology it is based on\n\n\nand the implications this has for the present analysis.\n\n\nThe dataset corrects a number of deficiencies of existing education statistics for this period that have to this\n\n\ndate prevented the kind of analysis conducted in this study. The principal existing source for historical international\n\n\neducation statistics has been UNESCO, which has been collating national education statistics since the 1960s.\n\n\nUnfortunately, this database suffers from at least four problems. Firstly, there are many gaps, reflecting years in\n\n\nwhich no statistics were provided by particular countries. Secondly, not all countries had the capacity to collect\n\n\nreliable data, especially in earlier years. Thirdly, the statistics are not fully consistent across countries due to\n\n\ndifferent definitions. Fourthly they are not consistent over time, as definitions, incl. internationally agreed ones, such\n\n\nas the _International Standard Classification of Education\n\nMention: \"Demographic Health Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Individual attainment records here are back-projected across 120 countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004307:9:0", "occurrence_id": "d54271f7a01901e1953fbd88", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Breaking the Waves? Education, Youth Bulges, and Armed Conflict\n\n\n_3.1 IIASA Human Capital Data_\n\n\nEducation data originate from a new dataset compiled by researchers at IIASA (Lutz et al. 2007). The dataset\n\n\ncontains information from 120 countries for the 1970-2000 period, based on individual-level educational attainment\n\n\ndata from Demographic Health Surveys and national censuses, back projected using multi-state demographic\n\n\nmodeling. The dataset measures educational attainment as opposed to more commonly used education measures like\n\n\neducational spending or enrollment rates. It covers a great number of developing countries, including many\n\n\ncountries with recent conflict experiences.\n\n\nThe IIASA data consists of population estimates by age, gender and highest educational attainment (4\n\n\ncategories: no schooling, some primary, completed lower secondary, completed tertiary), covering the period from\n\n\n1970 to 2000, for 120 countries/territories (discrepancies between this figure and others below arise from the fact\n\n\nthat more countries have been added since the publication of Lutz et al. (2007) and the introduction of splits and\n\n\nmergers during data matching). It is discussed in some detail here because of the novel methodology it is based on\n\n\nand the implications this has for the present analysis.\n\n\nThe dataset corrects a number of deficiencies of existing education statistics for this period that have to this\n\n\ndate prevented the kind of analysis conducted in this study. The principal existing source for historical international\n\n\neducation statistics has been UNESCO, which has been collating national education statistics since the 1960s.\n\n\nUnfortunately, this database suffers from at least four problems. Firstly, there are many gaps, reflecting years in\n\n\nwhich no statistics were provided by particular countries. Secondly, not all countries had the capacity to collect\n\n\nreliable data, especially in earlier years. Thirdly, the statistics are not fully consistent across countries due to\n\n\ndifferent definitions. Fourthly they are not consistent over time, as definitions, incl. internationally agreed ones, such\n\n\nas the _International Standard Classification of Education\n\nMention: \"IIASA data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Population by age, sex and schooling in four categories comes from this dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004307:9:0", "occurrence_id": "aff49fc9b2ba9d565161ce9a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Breaking the Waves? Education, Youth Bulges, and Armed Conflict\n\n\n_3.1 IIASA Human Capital Data_\n\n\nEducation data originate from a new dataset compiled by researchers at IIASA (Lutz et al. 2007). The dataset\n\n\ncontains information from 120 countries for the 1970-2000 period, based on individual-level educational attainment\n\n\ndata from Demographic Health Surveys and national censuses, back projected using multi-state demographic\n\n\nmodeling. The dataset measures educational attainment as opposed to more commonly used education measures like\n\n\neducational spending or enrollment rates. It covers a great number of developing countries, including many\n\n\ncountries with recent conflict experiences.\n\n\nThe IIASA data consists of population estimates by age, gender and highest educational attainment (4\n\n\ncategories: no schooling, some primary, completed lower secondary, completed tertiary), covering the period from\n\n\n1970 to 2000, for 120 countries/territories (discrepancies between this figure and others below arise from the fact\n\n\nthat more countries have been added since the publication of Lutz et al. (2007) and the introduction of splits and\n\n\nmergers during data matching). It is discussed in some detail here because of the novel methodology it is based on\n\n\nand the implications this has for the present analysis.\n\n\nThe dataset corrects a number of deficiencies of existing education statistics for this period that have to this\n\n\ndate prevented the kind of analysis conducted in this study. The principal existing source for historical international\n\n\neducation statistics has been UNESCO, which has been collating national education statistics since the 1960s.\n\n\nUnfortunately, this database suffers from at least four problems. Firstly, there are many gaps, reflecting years in\n\n\nwhich no statistics were provided by particular countries. Secondly, not all countries had the capacity to collect\n\n\nreliable data, especially in earlier years. Thirdly, the statistics are not fully consistent across countries due to\n\n\ndifferent definitions. Fourthly they are not consistent over time, as definitions, incl. internationally agreed ones, such\n\n\nas the _International Standard Classification of Education\n\nMention: \"UNESCO\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"The long-standing collation of national education statistics is the comparator.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004310:12:0", "occurrence_id": "568c6157c97618822edea76e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: shock pushes households’ utility below this level, they enter poverty. Conversely, if they are\nlifted above this level of utility, they are no longer in poverty.\n\n\nThe framework of Hertel et al (2004, 2009b), and that which this paper adopts, uses the\nAIDADS system to represent consumer preferences. This choice is based on AIDADS’ strength in\ncapturing food expenditure patterns across the income spectrum (Verma et al, 2009), and for its\nability to perform well out of sample when compared to other demand systems (see Cranfield et\nal, 2002, 2003) <sup>8</sup> [^8: Please see Appendix C for details of AIDADS formulation and parameterization.] . Reflecting its suitability for poverty analysis is that AIDADS devotes two‐thirds\nof its parameters to characterizing consumer behavior at very low levels of income. Estimation\nof this demand system is undertaken using the 80 country, per capita consumption data set\noffered by Version 6.1 of the GTAP database, also following Hertel et al (2004). For each\ncommodity, we have estimates of subsistence quantities of consumption, from which we may\ninfer (for average prices), budget shares at the subsistence level of income.\n\n\nThe poverty line in Tanzania is set to match the observed national poverty headcount\nratio reported by the World Bank (2006) and this in turn dictates the poverty level of utility in\nthe initial equilibrium. So, in the wake of a change in climate, commodity prices and wages will\nadjust, household incomes will change, as will the consumption profile of households at the\npoverty line, thereby resulting in new utility level. If household utility rises above the poverty\nlevel of utility, then it is lifted out of poverty. Conversely, if the household utility level falls below\nthe poverty utility threshold, then it has become impoverished.\n\n\nEquations (2) to (4) describe how the model can then be used to predict the change in\nthe national poverty rate – the percentage of the population\n\nMention: \"GTAP database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Per-capita consumption for eighty countries in this version drives the demand system.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004312:19:0", "occurrence_id": "1115b0b2c8375f151c3c382b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: the proportion of the adult population in each employment category. <sup>22</sup> Registered and\n\nunregistered wage earners, self-employed, and non-employed individuals, together account for\n\n87% of the adult population. <sup>23</sup> Therefore, in the empirical work we focus on these four groups.\n\nInformal employment and self-employment are considered two separate categories, as\n\nemphasized in the recent literature (Maloney, 2004, Fields, 1990, 2005). Finally, we have also\n\ncomputed some measures of past informality, poverty and inequality in the city using the 1980\n\nBrazilian census. <sup>24</sup> In 2002, there are 5,513 cities in Brazil.\n\nThird, we use detailed information on other city level characteristics from two statistical\n\nand research institutes in Brazil - _Instituto de Pesquisa Economica Aplicada_ (IPEA), and\n\n_Instituto Brasileiro de Geografia e Estatistica_ (IBGE). <sup>25</sup> In particular, we collect information on\n\nthe city‟s GDP per capita (2000), total number of firms (2000), average firm size (2000), share of\n\nagriculture in GDP (2000), share of manufacturing in GDP (2000), share of services in GDP\n\n(2000), geographical city characteristics (including geographical area, altitude, longitude and\n\nlatitude), city transportation costs (1995), total federal transfers to each city (1990), the city head\n\ncount poverty index and the city Theil inequality index. The total number of firms (2002) in each\n\ncity comes from the _Cadastro Central de Empresas_, collected by _IBGE_, which only includes\n\nformal firms. We also use past city level variables published by IPEA for the years 1970, 1980,\n\nand 1991, including city population, per\n\nMention: \"1980\n\nBrazilian census\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Historic informality and inequality in the city are computed from it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004317:3:0", "occurrence_id": "75313e754d9a850dd8a43e01", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **1. Labor market performance in the Philippines**\n\n\nDespite decent economic growth, labor market performance in the Philippines was less than\nsatisfactory in the past five years. Real wages slightly declined and new job creation lagged\nbehind the rapid increase in working-age population. Compared with other countries in East Asia,\nthe employment rate is relatively low in the Philippines.\n\n\n_1.1_ _Changes over time_\n\n\nBased on the Labor Force Survey (LFS) data for 2003-2007, the labor force participation rate,\nthe employment rate, and the real wage rate all declined over the period, while the working age\npopulation rate increased marginally (Table 1). <sup>1</sup> In the context of a slight increase in workingage population, a 2 percent decline in the participation rate and the employment rate suggest a\nworrisome trend of limited job creation for a rapidly growing population. A 5 percent decline in\nreal wages stands in contrast with the rapid economic growth. On average, 66 percent of the\npopulation was in the working age range, of which 62 percent participated in the labor force and\n58 percent were employed. The mean daily wage was 195 PHP in 2000 constant terms.\n\n\n\n**<u>Table 1: Labor Market Indicators in the Philippines (2003-2007)</u>**\n**Working Age** **Labor force** **Employment rate to** **NSO unemployment**\n\n\n\n**Working Age** **Labor force** **Employment rate to** **NSO unemployment** **Daily wage**\n\n**Population** **participation rate** **working-age population** **rate to participation** **(2000 PHP)**\n\n2003 65.5% 63.0% 59.2% 6.2% 201\n2004 65.9% 62.2% 58.0\n\nMention: \"the Labor Force Survey (LFS)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Participation, employment and real wages over this window come from it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004318:11:1", "occurrence_id": "30a1f461f6802c79e60abf66", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Rc_ + _DummiesR_ + _ec_ (2)\n\n\nwhere the _β_ s are parameters to be estimated. _Exp/popc_ are exports per capita in country\n\n\n_c_, and _Bud/popc_ is the budget of the EPA per capita in country _c_ . _GDP/popc_ is GDP per\n\n\ncapita measured as the average for the period 2000-2004 in 2005 constant U.S. dollars from the\n\n\nWorld Bank’s World Development Indicators. _Tc_ is an index of trade restrictiveness imposed\n\n\nby country _c_ on its imports from the rest of the world, _MAc_ is an index of market access\n\n\nrestrictions imposed by the rest of the world on exports of country _c_, and both are borrowed\n\n\nfrom Kee, Nicita and Olarreaga (2009). _V olc_ is the volatility of the exchange rate in country\n\n\n_c_, measured by the coefficient of variation of the dollar to local currency exchange rate during\n\n\nthe period 2000-2004 obtained from the World Development Indicators. _Regc_ is the number\n\n\n9\n\nMention: \"the\n\n\nWorld Bank’s World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Output per head averaged over this window in constant dollars comes from it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004326:16:2", "occurrence_id": "b3ed74ea97409163acb423a4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: potential,<br>cash crop|<br>3|5.4|13.7|1.4|1.0|0.5|22.0|\n|Agronomic <br>potential,<br>cash crop|2|10.3|4.1|9.3|2.1|0.2|26.0|\n|Agronomic <br>potential,<br>cash crop|1 (best)|1.2|7.6|10.0|15.1|2.2|36.0|\n|Agronomic <br>potential,<br>cash crop|subtotal|21.0|29.4|25.3|20.6|3.7|100.0|\n\n\nData Source: GIS+HDM4 data, IFPRI SPAM, and the IHS2\n\n\n\n\n\n\n\n**Table 8: Tobacco production, by agronomic potential and travel time to urban centers**\n\n\n\n\n\n\n\n|Tobacco production,<br>share, %|Col2|travel time to nearest 4 major cities|Col4|Col5|Col6|Col7|Col8|\n|---|---|---|---|---|---|---|---|\n|Tobacco production,<br>share, %|Tobacco production,<br>share, %|5 <br>(longest)|4|3|2|1 <br>(shortest)|subtotal|\n|Agronomic <br>potential,<br>cash crop|5 (lowest)|1.0|3.5|2.9|0.0|0.0|7.5|\n|Agronomic <br>potential,<br>cash crop|<\n\nMention: \"GIS+HDM4 data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Agonomic potential and travel time are computed from this linked set.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004326:16:2", "occurrence_id": "3d6a4688a176415fb981a922", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: potential,<br>cash crop|<br>3|5.4|13.7|1.4|1.0|0.5|22.0|\n|Agronomic <br>potential,<br>cash crop|2|10.3|4.1|9.3|2.1|0.2|26.0|\n|Agronomic <br>potential,<br>cash crop|1 (best)|1.2|7.6|10.0|15.1|2.2|36.0|\n|Agronomic <br>potential,<br>cash crop|subtotal|21.0|29.4|25.3|20.6|3.7|100.0|\n\n\nData Source: GIS+HDM4 data, IFPRI SPAM, and the IHS2\n\n\n\n\n\n\n\n**Table 8: Tobacco production, by agronomic potential and travel time to urban centers**\n\n\n\n\n\n\n\n|Tobacco production,<br>share, %|Col2|travel time to nearest 4 major cities|Col4|Col5|Col6|Col7|Col8|\n|---|---|---|---|---|---|---|---|\n|Tobacco production,<br>share, %|Tobacco production,<br>share, %|5 <br>(longest)|4|3|2|1 <br>(shortest)|subtotal|\n|Agronomic <br>potential,<br>cash crop|5 (lowest)|1.0|3.5|2.9|0.0|0.0|7.5|\n|Agronomic <br>potential,<br>cash crop|<\n\nMention: \"IFPRI SPAM\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its spatial allocation model informs the crop potential categories.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004326:16:2", "occurrence_id": "9d02f0ff7d19b297a1f96f9a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: potential,<br>cash crop|<br>3|5.4|13.7|1.4|1.0|0.5|22.0|\n|Agronomic <br>potential,<br>cash crop|2|10.3|4.1|9.3|2.1|0.2|26.0|\n|Agronomic <br>potential,<br>cash crop|1 (best)|1.2|7.6|10.0|15.1|2.2|36.0|\n|Agronomic <br>potential,<br>cash crop|subtotal|21.0|29.4|25.3|20.6|3.7|100.0|\n\n\nData Source: GIS+HDM4 data, IFPRI SPAM, and the IHS2\n\n\n\n\n\n\n\n**Table 8: Tobacco production, by agronomic potential and travel time to urban centers**\n\n\n\n\n\n\n\n|Tobacco production,<br>share, %|Col2|travel time to nearest 4 major cities|Col4|Col5|Col6|Col7|Col8|\n|---|---|---|---|---|---|---|---|\n|Tobacco production,<br>share, %|Tobacco production,<br>share, %|5 <br>(longest)|4|3|2|1 <br>(shortest)|subtotal|\n|Agronomic <br>potential,<br>cash crop|5 (lowest)|1.0|3.5|2.9|0.0|0.0|7.5|\n|Agronomic <br>potential,<br>cash crop|<\n\nMention: \"IHS2\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Household production shares are taken from this second survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004331:22:1", "occurrence_id": "ffc7c5f3ec91729d04677b60", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  (LCs and short-term loans) as\nreported by most banks interviewed (see next section). Therefore, any drop in LCs\nshould be interpreted with caution.\n\n\n\n**III.** **Banks’ perspective on scarcity of trade finance: Weak demand**\n**and heightened risk aversion**\n\nThe World Bank survey also included interviews of a total of 78 banks and financial\ninstitutions in 13 developing countries (excl. Indonesia) to assess any changes and trends\nin trade credit volumes post-September 2008. This is particularly useful in light of the\nabsence of data on trade finance. Comprehensive statistical compilation of flows is\ndifficult to extract from balance of payments due to the high cost of information,\ntechnology needed, and the difficulty to guarantee a high level of accuracy to very short\nterm capital movements <sup>10</sup> . From the countries surveyed, only two of them (Brazil, and\nTurkey) had macro trade finance data available overt time as described below. Moreover,\na large part of trade finance transactions takes place between buyers and suppliers which\nare not required to report their transactions.\nAgainst this background, the bank surveys aimed to produce qualitative results and were\nconducted through phone or in-person interviews. The bank questionnaire covers bank\nlending activities and practices pertaining to trade loans, trade credits, and short and\nmedium-term financing for export and import transactions from September 2008 to April\n\n\n10 At present, the only available and reliable source of statistics comes from the Berne Union database, which provides\ndata on the amount of business of export credit agencies.\n\n21\n\nMention: \"World Bank survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Interviews with seventy-eight banks across thirteen countries are in it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004331:22:1", "occurrence_id": "1a091e9a04ac64edae4e3cc5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  (LCs and short-term loans) as\nreported by most banks interviewed (see next section). Therefore, any drop in LCs\nshould be interpreted with caution.\n\n\n\n**III.** **Banks’ perspective on scarcity of trade finance: Weak demand**\n**and heightened risk aversion**\n\nThe World Bank survey also included interviews of a total of 78 banks and financial\ninstitutions in 13 developing countries (excl. Indonesia) to assess any changes and trends\nin trade credit volumes post-September 2008. This is particularly useful in light of the\nabsence of data on trade finance. Comprehensive statistical compilation of flows is\ndifficult to extract from balance of payments due to the high cost of information,\ntechnology needed, and the difficulty to guarantee a high level of accuracy to very short\nterm capital movements <sup>10</sup> . From the countries surveyed, only two of them (Brazil, and\nTurkey) had macro trade finance data available overt time as described below. Moreover,\na large part of trade finance transactions takes place between buyers and suppliers which\nare not required to report their transactions.\nAgainst this background, the bank surveys aimed to produce qualitative results and were\nconducted through phone or in-person interviews. The bank questionnaire covers bank\nlending activities and practices pertaining to trade loans, trade credits, and short and\nmedium-term financing for export and import transactions from September 2008 to April\n\n\n10 At present, the only available and reliable source of statistics comes from the Berne Union database, which provides\ndata on the amount of business of export credit agencies.\n\n21\n\nMention: \"Berne Union database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The only reliable statistics on export credit agency business.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004336:31:1", "occurrence_id": "a6e14d20d321cc8af8e7a5b6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ) 0.24 0.43 1437\nAge 37.87 7.56 1437\nGirls School Present in Birth Village by Age 7 0.55 0.50 1437\nTime Spent on Children’s General Needs (Minutes/Day) 94.47 141.29 1437\nTime Spent Helping Child With Schoolwork 5.05 22.77 1437\nWas Any Time Spent Helping Child With Schoolwork?\n0.06 0.23 1437\n(Y/N) (fraction)\n\n\n\n34.72\n\n\n\nPaid Work (Minutes/Day) 40.67 131.71 1437\nAny Paid Work (Y/N) (fraction) 0.11 0.31 1437\nTime Outside the House 99.33 157.58 1437\n<u>Time Inside the House</u> <u>501.88</u> <u>225.67</u> <u>1437</u>\n_Notes_ : Data from the LEAPS survey of households in 2004. “Paid Tutorials” are expenditures on private tuitions outside school. All\nexpenditures are reported in Pakistani Rupees (Rs.) (at the time of the survey, $1 was approximately Rs.65).\n\n\n\n30\n\nMention: \"the LEAPS survey of households\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Expenditure on private tutorials and time inside the house come from this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004337:26:0", "occurrence_id": "a0618719f4c4264c6933d841", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Annex 1: The Sensitivity of Central Government Revenues and Expenditures to the**\n**Business Cycle** <sup>**15**</sup> [^15: We are thankful to Catalina Delgado and Conrado Garcia for excellent research assistance provided for\nthis section.]\n\nBased on monthly data for the period from January 1997 to January 2008 <sup>16</sup> [^16: The data sample selection has been determined by data constraints because consistent information on\nrelevant budget aggregates is not available for a long enough time series. The source is the National\nTreasury and IPEA data (www.ipea.gov.br)] the sensitivity\nof federal revenues and expenditures to the business cycle is assessed through the\nestimation of closed-form elasticities using the following specification:\n\n\n\n\n\n\n### Re vBKi    a 0   12 a 1 j log   Re\n\nRe _vi_  _t_ _j_  1  Re _v_\n\n\n### vBKi    a 0   12 a 1 j log   Re vBKi\n\n_vi_  _t_ _j_  1  Re _vi_\n\n\n\n_<u>vi</u>_\n\n\n\n_<u>i</u>_\n\n\n\n <u>Re</u> _<u>vi</u>_ \n\n <sup></sup> _BK_  <sup></sup>\n Re _vi_  _t_\n\n\n\n_j_  0 2 _j_  _IPI_ _iBK_  _t_  _j_\n\n\n\n <u>Re</u>\n\nlog  <sup></sup>\n\n\n\n_BK_\n_i_  _t_\n\n\n\n_j_ log _<u>i</u>_\n\n <sup></sup>\n\n\n\n <u>Re</u> _<u>vi</u>_ \n\nMention: \"IPEA data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Budget aggregates from this institute support the cyclicality assessment.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004340:15:0", "occurrence_id": "ac810e38db0301a3379c7e9f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Within the Northwestern areas dominated by Jats, the other caste groups show lower\nevidence of son preference than Jats. This corresponds with the fact that the Jats effectively\nowned the place, owning most if not all of the land and extending their lineage-based\nadministrative system to the village as whole ─ the other castes were heavily dependent on\nthe Jats for access to their livelihood, through patron-client relationships (Pradhan 1966, Das\nGupta 1995). The stronger son preference of Jats has been noted from early censuses (Hutton\n1933: 200-204). Interestingly, Visaria (2007) found that the district of Gujarat with the\nlowest child sex ratio in the 2001 census had a heavy concentration of people who migrated\nthere from Haryana about 200 years ago, and still follow Jat customs.\n\nIn China, the Han with their rigid patrilineages constitute the overwhelming majority of the\npopulation, but as discussed above patrilineal corporate structures were weaker in the North\nthan in the South. Some of the non-Han minorities have not yet adopted the Han kinship\nsystem. For example, the southern provinces bordering Myanmar and Laos have minorities\nsuch as the Na who practice matrilineal kinship, and the Lahu who practice bilateral kinship\n(Hua 2001, Du 2002). In Western Chinese provinces such as Xinjiang, the Uyghurs prefer to\nmarry within their own village and cooperate economically with both maternal and paternal\nkin, so daughters are not lost to their parents (Rudelson 1997). Anderson and Siver (1995:\nTable 7) find the 1990 census data show clear ethnic differentials in the sex ratios at birth in\nXinjiang province: the Uyghurs had a normal ratio of 105, while the Han had an elevated\nratio of 109.\n\nWestern Chinese provinces such as Tibet and Qinghai have large Tibetan populations, who\ntraditionally practice fraternal polyandry (Child 2003, Levine 1988). A 1958 census of one of\nthese areas shows high\n\nMention: \"2001 census\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its district-level child sex ratios identify the extreme case.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004340:15:0", "occurrence_id": "9dfcf8d4b2bc4de07404f834", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Within the Northwestern areas dominated by Jats, the other caste groups show lower\nevidence of son preference than Jats. This corresponds with the fact that the Jats effectively\nowned the place, owning most if not all of the land and extending their lineage-based\nadministrative system to the village as whole ─ the other castes were heavily dependent on\nthe Jats for access to their livelihood, through patron-client relationships (Pradhan 1966, Das\nGupta 1995). The stronger son preference of Jats has been noted from early censuses (Hutton\n1933: 200-204). Interestingly, Visaria (2007) found that the district of Gujarat with the\nlowest child sex ratio in the 2001 census had a heavy concentration of people who migrated\nthere from Haryana about 200 years ago, and still follow Jat customs.\n\nIn China, the Han with their rigid patrilineages constitute the overwhelming majority of the\npopulation, but as discussed above patrilineal corporate structures were weaker in the North\nthan in the South. Some of the non-Han minorities have not yet adopted the Han kinship\nsystem. For example, the southern provinces bordering Myanmar and Laos have minorities\nsuch as the Na who practice matrilineal kinship, and the Lahu who practice bilateral kinship\n(Hua 2001, Du 2002). In Western Chinese provinces such as Xinjiang, the Uyghurs prefer to\nmarry within their own village and cooperate economically with both maternal and paternal\nkin, so daughters are not lost to their parents (Rudelson 1997). Anderson and Siver (1995:\nTable 7) find the 1990 census data show clear ethnic differentials in the sex ratios at birth in\nXinjiang province: the Uyghurs had a normal ratio of 105, while the Han had an elevated\nratio of 109.\n\nWestern Chinese provinces such as Tibet and Qinghai have large Tibetan populations, who\ntraditionally practice fraternal polyandry (Child 2003, Levine 1988). A 1958 census of one of\nthese areas shows high\n\nMention: \"1990 census data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Ethnic differentials in sex ratios at birth are visible here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004347:27:1", "occurrence_id": "e22738b4b71691d19e8dfbc5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .\nStern (ed) _Services in the International Economy_, University of Michigan Press, USA.\n\nde Janvry, A. and E. Sadoulet (2000) “Rural Poverty in Latin America: Determinants and Exit\nPaths,” _Food Policy_ . 24(4), pp 389‐409.\n\nDeardorff, A.V. (1994) “The Possibility of Factor Price Equalization, Revisited”. _Journal of_\n_International Economics_ . 36, pp. 167‐175.\n\nDeardorff, A.V. (2001a) “Fragmentation in Simple Trade Models,” _North American Journal of_\n_Economics and Finance._ Vol. 12:121‐137.\n\nDeardorff, A.V. (2001b) “Fragmentation across Cones,” in S.W. Arndt and H. Kierzkowski (eds.),\n_Fragmentation: New Production Patterns in the World Economy_, Oxford University Press, 2001,\npp. 35‐51.\n\nDimaranan, B.V. (ed) (2006). Global Trade, Assistance, and Production: The GTAP 6 Data Base,\nCenter for Global Trade Analysis, Purdue University, IN.\n\nFeenstra, R.C. and G.H. Hanson (1996) “Globalization, Outsourcing, and Wage\nInequality,” _American Economic Review_ . Vol. 86, No. 2, pp. 240‐245.\n\nFinancial Express (2006) “Indian pay rises topping charts in Asia”, December 1, 2006. Available:\n<u>http://www.financialexpress.com/old/latest_full_story.php?content_id=147790</u>\n\nGATS (2006) http://www.wto.org/english/tratop_e/serv_e/gatsintr\n\nMention: \"GTAP 6 Data Base\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"A bibliography entry citing this supply, assistance and production database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004351:28:0", "occurrence_id": "a9fcf0436db17556a535f30e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: wage rigidity. The estimates also imply that the larger is the share of permanent workers\n\n\nand/or the larger is the EPL index, the stronger is this effect. These results are in line with our\n\n\nexpectations, since the existence of permanent contracts complemented with strict labour\n\n\nregulations gives workers more leeway in wage negotiations, which in turn should lead to\n\n\ngreater wage rigidity. In particular, it is harder for firms to cut workers’ wages if the threat of\n\n\nlay-off is more difficult to implement. Thus, permanent contracts impose greater wage rigidity\n\n\nthan temporary contracts as long as permanent workers are more protected by labour\n\n\nregulations. As a consequence, the effect of permanent contracts on wage rigidity should be\n\n\nmore significant in countries with stricter employment protection.\n\n\nThe WDN survey contains information on the structure of agreements applicable for a given\n\n\nfirm. Managers were asked if a collective wage agreement exists and if so, whether it is a\n\n\nfirm-level agreement or a binding agreement that was negotiated at a level outside the firm\n\n\n(e.g. national, sector level, etc). We use this information to analyse the implications that the\n\n\nunion contracts negotiated at different levels have on wage rigidity. For this purpose, we\n\n\nconstruct three non-nested dummy variables that characterise the type of union contract(s)\n\n\napplying to the firm; the first indicating the existence of only a firm-level agreement, the\n\n\nsecond signifying only an outside agreement, and the third being equal to one if a firm has\n\n\nboth firm-level and outside agreements.\n\n\nAppendix 7 gives an overview of the cross-country differences in the incidence of union\n\n\ncontracts negotiated at different levels. This comparison reveals striking contrasts in the\n\n\ntendency of different types of union contracts across the sampled countries. In particular,\n\n\nthere is a group of countries (Austria, Belgium, Spain, France, Italy and Slovenia) where\n\n\nalmost all firms have union contracts and also display a very high incidence of higher-level\n\n\nbargaining agreements. On the other hand most\n\nMention: \"WDN survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Managers describe the wage agreements applicable in their firm here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004369:57:1", "occurrence_id": "9e491c6b3ee93aa715b4dd79", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \nAdvances and Research Challenges.” _World Bank Policy Research Working Paper No._\n_4460_ . The World Bank.\n\nWindmeijer, Frank (2005). “A finite sample correction for the variance of linear efficient\ntwo-step GMM estimators.” _Journal of Econometrics_ 126 (1): 25–51.\n\nThe World Bank (2003). “Trade, Investment, and Development in the Middle East and\nNorth Africa.” The World Bank.\n\n_____ (2008). “Public-Private Partnership and Infrastructure Subsidy and Tariff\nReform.” Unpublished technical note. The World Bank.\n\nWorld Development Indicators (2007), <u>www.worldbank.org.</u>\n\n\n56\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"A reference-list entry for this compilation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004373:23:0", "occurrence_id": "bc3e0561fdfa655f0512399e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <u>Table</u> <u>3:</u> <u>Mass</u> <u>media</u> <u>bias</u> <u>in</u> <u>public</u> <u>opinion</u>\nMedia bias,\nmagnitude\nTV Press\n(η1i, η2i)\n\n\n\n0.08 - 0.12 Japan, Mexico, India, Slovenia, Cyprus,\nEthiopia\n\n\n0.04 - 0.08 Sweden, Switzerland, Brazil, Turkey, Peru,\nMoldova, Indonesia, Vietnam, Serbia, Egypt,\nAndorra, Burkina Faso, Zambia, France\n\n\n\nThailand, Cyprus\n\n\nJordan, Malaysia\n\n\n\n0.02-0.04 China USA, Mexico,\nBrazil, Romania,\nEgypt\n\n\n\n≈ 0\n\n\n\nArgentina, Australia, Bulgaria, Chile, Taiwan, Colombia, Finland,\nGermany, Ghana, Italy, Republic of Korea, Mali, Morocco, Netherlands, Poland, Russian Federation, South Africa, Spain, Trinidad\nand Tobago, United Kingdom\n\n\n\n<u>-0.08</u> <u>-</u> <u>-0.02</u> <u>Ukraine,</u> <u>Rwanda</u> <u>Indonesia</u>\n\n\nNote: First column - ranges for point OLS estimates are reported. For each range, only the\ncountries, for which coefficients are different from 0 at a significance level less than 5%, are\nreported. ”≈ 0” range - countries with no significant TV or press bias. Sample of the countries\nused - WVS wave 3 (except Iran, Iraq, Hong Kong, New Zealand, where questions about mass\nmedia were not asked)\n\n\ncritical towards the government) to 0.12, which combined with on average 75% of\nrespondents watching TV or reading newspaper, may lead for some countries to a\ndecrease in our estimates\n\nMention: \"WVS wave 3\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its country sample defines where media bias is estimated.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004374:28:0", "occurrence_id": "c68ed0c52d4fcb653cdd608e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: earthquake mortality prevention than rich countries such that they respond less to an elevated\n\n\nearthquake propensity than rich countries. <sup>12</sup>\n\n\nModel 3 interacts the quake propensity variable with dummy variables for democracies\n\n\nand autocracies. Consistent with the argument that the prospect of elections should make\n\n\ndemocratic rulers more sensitive to earthquake propensity, the estimated coefficient for quake\n\n\npropensity in democracies is 2/3 larger than the respective estimated coefficient in autocracies\n\n\n(-.0044 versus -.0029), although the difference between the two estimated coefficients is not\n\n\nstatistically significant. A one standard deviation increase in quake propensity is estimated to\n\n\ndecrease expected mortality by 49.4 per cent in democracies and by 42.2 per cent in\n\n\nautocracies, if other variables are held at mean values. The statistically significant coefficient\n\n\nof the linear democracy term indicates that democratic governments suffer significantly fewer\n\n\nearthquake deaths at all levels of quake risk.\n\n\nModel 4 further distinguishes among both democratic and non-democratic regimes. The\n\n\nspecification distinguishes among autocracies according to whether the age of the ruling party\n\n\nat the time the current leader came to power – ruling-party institutionalization – is above or\n\n\nbelow the median for that variable (see Gehlbach and Keefer 2009). Democracies are\n\n\ndistinguished according to whether they are above or below the median years of continuous\n\n\ndemocratic experience. <sup>13</sup> Since the Database of Political Institutions starts in 1975, and is\n\n\n12 Given the strong correlation of high income with good governance, it is difficult to exclude an alternative\n\n\nexplanation for this finding, namely that poor countries are also generally governed by rulers who are less\n\n\nsensitive to the broad public interest and, therefore, less responsive to increases in earthquake risk. However, this\n\n\ninterpretation of the risk-income interaction is also consistent with our theoretical arguments.\n\n\n13 Since we use two different operationalizations for democracies on the one hand and autocracies on the\n\n\nother, there is no\n\nMention: \"the Database of Political Institutions\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its coverage beginning in that year constrains the democratic experience measure.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004381:26:0", "occurrence_id": "aec445dadae7a948be186617", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: well as in three high income comparator countries, Canada, Finland, and the Republic of\n\n\nKorea.\n\n\nTo understand the dispersion of both wealth and cognitive achievement in these\n\n\ncountries, Figure 1 characterizes wealth and reading competency inequality in these\n\n\ncountries. For each country, the vertical dashed line denoting PISA wealth and the\n\n\nvertical solid line denoting PISA reading competency start at the country’s 10% quantile\n\n\nlevel and terminate at the country’s 90% quantile level of wealth and reading competency\n\n\nrespectively. The country mean levels of wealth and reading are denoted by the squares.\n\n\nArgentina displays the largest inequality in reading competency with 317 points\n\n\nseparating the top and bottom 10%; this is followed by Uruguay with a difference of 311\n\n\npoints. The difference between the top and bottom 10% of the PISA wealth index for\n\n\nthese countries is 2.2 and 2.4, respectively. In general, the six Latin American countries\n\n\nshow larger dispersions in both reading competency and in the wealth index than the\n\n\nthree high income countries. But Mexico stands out as the country most able to mitigate\n\n\nwealth inequality in Latin America: it exhibits the widest difference in wealth between\n\n\nthe top and bottom 10%, 2.9 points, but exhibits at the same time the narrowest difference\n\n\nin competency between the top and bottom 10% at 245 points.\n\n\nIn order to understand the relationship between wealth and competency, Figure 2 displays\n\n\nthe differences in PISA reading competency between the top and bottom quartiles of the\n\n\nwealth index. The heights of the grey columns denote the sizes of differences in reading\n\n\n25\n\nMention: \"the PISA wealth index\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Top-to-bottom gaps in this asset index are computed for six countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004390:15:0", "occurrence_id": "1299d488ed5bc501960e752f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 2. Total Debt in Spain**\n\n\nSource: Banking Supervision Report 2001-2008 and Statistical Bulletin, Bank of Spain.\n\n\n**Figure 3. Nominal Credit and GDP Growth**\n\n\n**(%)**\n\n\nSource: Bank of Spain and National Institute of Statistic\n\n13\n\nMention: \"Banking Supervision Report 2001-2008\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Total debt and credit growth are plotted from these supervisory reports.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004391:25:1", "occurrence_id": "000f52298dbad0607bc62d3a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: >Wall of dwelling (% with<br>cement / CI sheet)<br>Roof of dwelling (% with<br>cement / CI sheet)<br>Safe latrine use (%)<br>Electricity connection (%)<br>TV ownership (%)<br>Phone ownership (%)|4280<br>5281<br>35.2<br>40.3<br>37.7<br>55.2<br>76.4<br>89.9<br>52.0<br>69.3<br>31.2<br>44.2<br>15.8<br>26.5<br>1.5<br>12.2|3222<br>4653<br>33.6<br>43.3<br>21.4<br>39.5<br>68.1<br>84.2<br>35.2<br>55.6<br>14.6<br>25.4<br>3.6<br>10.1<br>0.1<br>1.5|2623<br>3919<br>31.6<br>42.5<br>17.4<br>33.9<br>64.5<br>81.6<br>29.4<br>50.0<br>10.0<br>20.2<br>1.8<br>6.7<br>0.0<br>0.9|\n|_Source_: HIES 2000, 2005<br>|_Source_: HIES 2000, 2005<br>|_Source_: HIES 2000, 2005<br>|_\n\nMention: \"HIES\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Two rounds support the household characteristics tabulated.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004394:39:4", "occurrence_id": "6435ad892bbff903c0c7637f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  2.75 9.88 0.41 0.30 0.34 0.66 0.17 0.45 0.37 0.62 0.03 0.04 0.23 0.08 0.54 0.27\nSweden 51 0.39 2.11 10.09 0.54 0.36 0.48 0.51 0.39 0.32 0.29 0.53 0.09 0.02 0.26 0.09 0.22 0.68\n<u>United Kingdom</u> <u>51</u> <u>0.36</u> <u>2.35</u> <u>10.32</u> <u>0.69</u> <u>0.31</u> <u>0.82</u> <u>0.18</u> <u>0.30</u> <u>0.37</u> <u>0.30</u> <u>0.52</u> <u>0.07</u> <u>0.02</u> <u>0.26</u> <u>0.13</u> <u>0.76</u> <u>0.05</u>\n\n\nNote: Data used to create dummy variables – urban and rural - are not available for the Netherlands and Slovenia.\n\nSource: EU Statistics on Income and Living Conditions (SILC) database, 2007\n\n\n2\n\nMention: \"EU Statistics on Income and Living Conditions\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its 2007 database gives incomes, though urban flags are missing for two countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004396:6:1", "occurrence_id": "12ae3b0763a33070ea3a04bf", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  individual _i_ is characterized exclusively by a\n\nset of attributes { _yi_, _Ci_, _ei_ }, with _y_ denoting an advantage, **_C_** denoting a vector of\n\ncircumstance characteristics, and _e_ denoting an effort level. Let us also follow Roemer\n\n(1998) in treating effort as a continuous variable, while the vector **_Ci_** consists of _J_\n\n\nrespondents to the Life in Transition Survey of 2006 felt that the “gap between the rich and the poor was\ntoo high” in Turkey, and when asked what was the “main reason why there are some people in need in our\ncountry today?” 63% choose “injustice in society” as their answer.\n6 This section is based on and summarizes the more comprehensive discussion in Ferreira and Gignoux\n(2008).\n\nMention: \"the Life in Transition Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its respondents judge whether the rich-poor gap is too high.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004399:7:0", "occurrence_id": "90013cfc09f63f95513d0275", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The remainder of the paper is organized as follows. Section 2 presents the\n\nmethodology and data used for estimating the district-level prevalence in the absence of\n\nnationally-representative population surveys. Section 3 presents the regression results and\n\ndiscusses how the district-level estimates of HIV prevalence obtained from the\n\napplication of the SAE methodology to the DHS 2000 survey compare with the sentinel\n\ndata estimates for 2001 and the “gold-standard” prevalence rates derived directly from\n\nthe 2004 DHS Survey. Section 4 concludes by reviewing the main findings.\n\n#### **2. Methodology and Data**\n\n\n**2a. The basic idea**\n\n\nThe methodology used in this paper draws on the recently developed small area\n\nestimation techniques applied to welfare measurement (Elbers et al., 2003). <sup>12</sup> [^12: The important distinction is that while in the welfare measurement the household survey is representative\nof the total population of households, in the prevalence measurement ANC data are representative of only\nthe respective sub-population of all population of women. The paper discusses in a greater detail below the\npossible ways to deal with this caveat.] The idea is\n\nstraightforward. Let p _i_ be an indicator of individual HIV infection found in the sentinel\n\ndata. We first model p _i_ as a function of individual level variables (such as age and\n\neducation) found in the 2001 sentinel (ANC) survey, and of selected commune/district\n\nmeans derived from the 1998 Malawi National Census data (Benson, 2002), the 2001\n\nHealth Facilities data, 2001 GIS data, and the 2000 Demographic and Health Survey\n\n\n\n(DHS) data. <sup>13</sup> [^13: We potentially could use the 2004 DHS data instead of the 2000 DHS data, but since the reliability of\nestimates depends on the various data sources been closer to each other in time, we use the 2000 DHS\nwhich is more consistent with the 2001 ANC survey, 2001 GIS data, and the 1998 Census.]\n\n\n\nSuch commune/district means are useful\n\nMention: \"DHS 2000 survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Small-area estimation applied here yields district prevalence.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004399:7:0", "occurrence_id": "3de676367f9c88c1b8832462", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The remainder of the paper is organized as follows. Section 2 presents the\n\nmethodology and data used for estimating the district-level prevalence in the absence of\n\nnationally-representative population surveys. Section 3 presents the regression results and\n\ndiscusses how the district-level estimates of HIV prevalence obtained from the\n\napplication of the SAE methodology to the DHS 2000 survey compare with the sentinel\n\ndata estimates for 2001 and the “gold-standard” prevalence rates derived directly from\n\nthe 2004 DHS Survey. Section 4 concludes by reviewing the main findings.\n\n#### **2. Methodology and Data**\n\n\n**2a. The basic idea**\n\n\nThe methodology used in this paper draws on the recently developed small area\n\nestimation techniques applied to welfare measurement (Elbers et al., 2003). <sup>12</sup> [^12: The important distinction is that while in the welfare measurement the household survey is representative\nof the total population of households, in the prevalence measurement ANC data are representative of only\nthe respective sub-population of all population of women. The paper discusses in a greater detail below the\npossible ways to deal with this caveat.] The idea is\n\nstraightforward. Let p _i_ be an indicator of individual HIV infection found in the sentinel\n\ndata. We first model p _i_ as a function of individual level variables (such as age and\n\neducation) found in the 2001 sentinel (ANC) survey, and of selected commune/district\n\nmeans derived from the 1998 Malawi National Census data (Benson, 2002), the 2001\n\nHealth Facilities data, 2001 GIS data, and the 2000 Demographic and Health Survey\n\n\n\n(DHS) data. <sup>13</sup> [^13: We potentially could use the 2004 DHS data instead of the 2000 DHS data, but since the reliability of\nestimates depends on the various data sources been closer to each other in time, we use the 2000 DHS\nwhich is more consistent with the 2001 ANC survey, 2001 GIS data, and the 1998 Census.]\n\n\n\nSuch commune/district means are useful\n\nMention: \"2004 DHS Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its direct rates are the gold standard for comparison.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004399:7:0", "occurrence_id": "57d7c174964d4fcee6196be2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The remainder of the paper is organized as follows. Section 2 presents the\n\nmethodology and data used for estimating the district-level prevalence in the absence of\n\nnationally-representative population surveys. Section 3 presents the regression results and\n\ndiscusses how the district-level estimates of HIV prevalence obtained from the\n\napplication of the SAE methodology to the DHS 2000 survey compare with the sentinel\n\ndata estimates for 2001 and the “gold-standard” prevalence rates derived directly from\n\nthe 2004 DHS Survey. Section 4 concludes by reviewing the main findings.\n\n#### **2. Methodology and Data**\n\n\n**2a. The basic idea**\n\n\nThe methodology used in this paper draws on the recently developed small area\n\nestimation techniques applied to welfare measurement (Elbers et al., 2003). <sup>12</sup> [^12: The important distinction is that while in the welfare measurement the household survey is representative\nof the total population of households, in the prevalence measurement ANC data are representative of only\nthe respective sub-population of all population of women. The paper discusses in a greater detail below the\npossible ways to deal with this caveat.] The idea is\n\nstraightforward. Let p _i_ be an indicator of individual HIV infection found in the sentinel\n\ndata. We first model p _i_ as a function of individual level variables (such as age and\n\neducation) found in the 2001 sentinel (ANC) survey, and of selected commune/district\n\nmeans derived from the 1998 Malawi National Census data (Benson, 2002), the 2001\n\nHealth Facilities data, 2001 GIS data, and the 2000 Demographic and Health Survey\n\n\n\n(DHS) data. <sup>13</sup> [^13: We potentially could use the 2004 DHS data instead of the 2000 DHS data, but since the reliability of\nestimates depends on the various data sources been closer to each other in time, we use the 2000 DHS\nwhich is more consistent with the 2001 ANC survey, 2001 GIS data, and the 1998 Census.]\n\n\n\nSuch commune/district means are useful\n\nMention: \"ANC data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Antenatal clinic surveillance covers only women, not all households.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004399:7:0", "occurrence_id": "e888420d6680ae472ccced64", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The remainder of the paper is organized as follows. Section 2 presents the\n\nmethodology and data used for estimating the district-level prevalence in the absence of\n\nnationally-representative population surveys. Section 3 presents the regression results and\n\ndiscusses how the district-level estimates of HIV prevalence obtained from the\n\napplication of the SAE methodology to the DHS 2000 survey compare with the sentinel\n\ndata estimates for 2001 and the “gold-standard” prevalence rates derived directly from\n\nthe 2004 DHS Survey. Section 4 concludes by reviewing the main findings.\n\n#### **2. Methodology and Data**\n\n\n**2a. The basic idea**\n\n\nThe methodology used in this paper draws on the recently developed small area\n\nestimation techniques applied to welfare measurement (Elbers et al., 2003). <sup>12</sup> [^12: The important distinction is that while in the welfare measurement the household survey is representative\nof the total population of households, in the prevalence measurement ANC data are representative of only\nthe respective sub-population of all population of women. The paper discusses in a greater detail below the\npossible ways to deal with this caveat.] The idea is\n\nstraightforward. Let p _i_ be an indicator of individual HIV infection found in the sentinel\n\ndata. We first model p _i_ as a function of individual level variables (such as age and\n\neducation) found in the 2001 sentinel (ANC) survey, and of selected commune/district\n\nmeans derived from the 1998 Malawi National Census data (Benson, 2002), the 2001\n\nHealth Facilities data, 2001 GIS data, and the 2000 Demographic and Health Survey\n\n\n\n(DHS) data. <sup>13</sup> [^13: We potentially could use the 2004 DHS data instead of the 2000 DHS data, but since the reliability of\nestimates depends on the various data sources been closer to each other in time, we use the 2000 DHS\nwhich is more consistent with the 2001 ANC survey, 2001 GIS data, and the 1998 Census.]\n\n\n\nSuch commune/district means are useful\n\nMention: \"2001 sentinel (ANC) survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Individual age and education variables come from this sentinel round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004399:7:0", "occurrence_id": "f6a9e89483303d164360bb52", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The remainder of the paper is organized as follows. Section 2 presents the\n\nmethodology and data used for estimating the district-level prevalence in the absence of\n\nnationally-representative population surveys. Section 3 presents the regression results and\n\ndiscusses how the district-level estimates of HIV prevalence obtained from the\n\napplication of the SAE methodology to the DHS 2000 survey compare with the sentinel\n\ndata estimates for 2001 and the “gold-standard” prevalence rates derived directly from\n\nthe 2004 DHS Survey. Section 4 concludes by reviewing the main findings.\n\n#### **2. Methodology and Data**\n\n\n**2a. The basic idea**\n\n\nThe methodology used in this paper draws on the recently developed small area\n\nestimation techniques applied to welfare measurement (Elbers et al., 2003). <sup>12</sup> [^12: The important distinction is that while in the welfare measurement the household survey is representative\nof the total population of households, in the prevalence measurement ANC data are representative of only\nthe respective sub-population of all population of women. The paper discusses in a greater detail below the\npossible ways to deal with this caveat.] The idea is\n\nstraightforward. Let p _i_ be an indicator of individual HIV infection found in the sentinel\n\ndata. We first model p _i_ as a function of individual level variables (such as age and\n\neducation) found in the 2001 sentinel (ANC) survey, and of selected commune/district\n\nmeans derived from the 1998 Malawi National Census data (Benson, 2002), the 2001\n\nHealth Facilities data, 2001 GIS data, and the 2000 Demographic and Health Survey\n\n\n\n(DHS) data. <sup>13</sup> [^13: We potentially could use the 2004 DHS data instead of the 2000 DHS data, but since the reliability of\nestimates depends on the various data sources been closer to each other in time, we use the 2000 DHS\nwhich is more consistent with the 2001 ANC survey, 2001 GIS data, and the 1998 Census.]\n\n\n\nSuch commune/district means are useful\n\nMention: \"1998 Malawi National Census data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Commune and district means are derived from this enumeration.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004399:7:0", "occurrence_id": "4c6d058a532ad9d340f06a70", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The remainder of the paper is organized as follows. Section 2 presents the\n\nmethodology and data used for estimating the district-level prevalence in the absence of\n\nnationally-representative population surveys. Section 3 presents the regression results and\n\ndiscusses how the district-level estimates of HIV prevalence obtained from the\n\napplication of the SAE methodology to the DHS 2000 survey compare with the sentinel\n\ndata estimates for 2001 and the “gold-standard” prevalence rates derived directly from\n\nthe 2004 DHS Survey. Section 4 concludes by reviewing the main findings.\n\n#### **2. Methodology and Data**\n\n\n**2a. The basic idea**\n\n\nThe methodology used in this paper draws on the recently developed small area\n\nestimation techniques applied to welfare measurement (Elbers et al., 2003). <sup>12</sup> [^12: The important distinction is that while in the welfare measurement the household survey is representative\nof the total population of households, in the prevalence measurement ANC data are representative of only\nthe respective sub-population of all population of women. The paper discusses in a greater detail below the\npossible ways to deal with this caveat.] The idea is\n\nstraightforward. Let p _i_ be an indicator of individual HIV infection found in the sentinel\n\ndata. We first model p _i_ as a function of individual level variables (such as age and\n\neducation) found in the 2001 sentinel (ANC) survey, and of selected commune/district\n\nmeans derived from the 1998 Malawi National Census data (Benson, 2002), the 2001\n\nHealth Facilities data, 2001 GIS data, and the 2000 Demographic and Health Survey\n\n\n\n(DHS) data. <sup>13</sup> [^13: We potentially could use the 2004 DHS data instead of the 2000 DHS data, but since the reliability of\nestimates depends on the various data sources been closer to each other in time, we use the 2000 DHS\nwhich is more consistent with the 2001 ANC survey, 2001 GIS data, and the 1998 Census.]\n\n\n\nSuch commune/district means are useful\n\nMention: \"2001\n\nHealth Facilities data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Facility characteristics enter the district-level predictors.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004399:7:0", "occurrence_id": "2c45c3144c197d28d5e7abb3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The remainder of the paper is organized as follows. Section 2 presents the\n\nmethodology and data used for estimating the district-level prevalence in the absence of\n\nnationally-representative population surveys. Section 3 presents the regression results and\n\ndiscusses how the district-level estimates of HIV prevalence obtained from the\n\napplication of the SAE methodology to the DHS 2000 survey compare with the sentinel\n\ndata estimates for 2001 and the “gold-standard” prevalence rates derived directly from\n\nthe 2004 DHS Survey. Section 4 concludes by reviewing the main findings.\n\n#### **2. Methodology and Data**\n\n\n**2a. The basic idea**\n\n\nThe methodology used in this paper draws on the recently developed small area\n\nestimation techniques applied to welfare measurement (Elbers et al., 2003). <sup>12</sup> [^12: The important distinction is that while in the welfare measurement the household survey is representative\nof the total population of households, in the prevalence measurement ANC data are representative of only\nthe respective sub-population of all population of women. The paper discusses in a greater detail below the\npossible ways to deal with this caveat.] The idea is\n\nstraightforward. Let p _i_ be an indicator of individual HIV infection found in the sentinel\n\ndata. We first model p _i_ as a function of individual level variables (such as age and\n\neducation) found in the 2001 sentinel (ANC) survey, and of selected commune/district\n\nmeans derived from the 1998 Malawi National Census data (Benson, 2002), the 2001\n\nHealth Facilities data, 2001 GIS data, and the 2000 Demographic and Health Survey\n\n\n\n(DHS) data. <sup>13</sup> [^13: We potentially could use the 2004 DHS data instead of the 2000 DHS data, but since the reliability of\nestimates depends on the various data sources been closer to each other in time, we use the 2000 DHS\nwhich is more consistent with the 2001 ANC survey, 2001 GIS data, and the 1998 Census.]\n\n\n\nSuch commune/district means are useful\n\nMention: \"2001 GIS data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Geographic covariates for the districts come from it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004399:7:0", "occurrence_id": "621427adcf2fae61a38fd8d5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The remainder of the paper is organized as follows. Section 2 presents the\n\nmethodology and data used for estimating the district-level prevalence in the absence of\n\nnationally-representative population surveys. Section 3 presents the regression results and\n\ndiscusses how the district-level estimates of HIV prevalence obtained from the\n\napplication of the SAE methodology to the DHS 2000 survey compare with the sentinel\n\ndata estimates for 2001 and the “gold-standard” prevalence rates derived directly from\n\nthe 2004 DHS Survey. Section 4 concludes by reviewing the main findings.\n\n#### **2. Methodology and Data**\n\n\n**2a. The basic idea**\n\n\nThe methodology used in this paper draws on the recently developed small area\n\nestimation techniques applied to welfare measurement (Elbers et al., 2003). <sup>12</sup> [^12: The important distinction is that while in the welfare measurement the household survey is representative\nof the total population of households, in the prevalence measurement ANC data are representative of only\nthe respective sub-population of all population of women. The paper discusses in a greater detail below the\npossible ways to deal with this caveat.] The idea is\n\nstraightforward. Let p _i_ be an indicator of individual HIV infection found in the sentinel\n\ndata. We first model p _i_ as a function of individual level variables (such as age and\n\neducation) found in the 2001 sentinel (ANC) survey, and of selected commune/district\n\nmeans derived from the 1998 Malawi National Census data (Benson, 2002), the 2001\n\nHealth Facilities data, 2001 GIS data, and the 2000 Demographic and Health Survey\n\n\n\n(DHS) data. <sup>13</sup> [^13: We potentially could use the 2004 DHS data instead of the 2000 DHS data, but since the reliability of\nestimates depends on the various data sources been closer to each other in time, we use the 2000 DHS\nwhich is more consistent with the 2001 ANC survey, 2001 GIS data, and the 1998 Census.]\n\n\n\nSuch commune/district means are useful\n\nMention: \"2000 Demographic and Health Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Chosen over the later round for temporal consistency with other sources.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004399:7:0", "occurrence_id": "cb1cacf20208ece76bc1f9a6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The remainder of the paper is organized as follows. Section 2 presents the\n\nmethodology and data used for estimating the district-level prevalence in the absence of\n\nnationally-representative population surveys. Section 3 presents the regression results and\n\ndiscusses how the district-level estimates of HIV prevalence obtained from the\n\napplication of the SAE methodology to the DHS 2000 survey compare with the sentinel\n\ndata estimates for 2001 and the “gold-standard” prevalence rates derived directly from\n\nthe 2004 DHS Survey. Section 4 concludes by reviewing the main findings.\n\n#### **2. Methodology and Data**\n\n\n**2a. The basic idea**\n\n\nThe methodology used in this paper draws on the recently developed small area\n\nestimation techniques applied to welfare measurement (Elbers et al., 2003). <sup>12</sup> [^12: The important distinction is that while in the welfare measurement the household survey is representative\nof the total population of households, in the prevalence measurement ANC data are representative of only\nthe respective sub-population of all population of women. The paper discusses in a greater detail below the\npossible ways to deal with this caveat.] The idea is\n\nstraightforward. Let p _i_ be an indicator of individual HIV infection found in the sentinel\n\ndata. We first model p _i_ as a function of individual level variables (such as age and\n\neducation) found in the 2001 sentinel (ANC) survey, and of selected commune/district\n\nmeans derived from the 1998 Malawi National Census data (Benson, 2002), the 2001\n\nHealth Facilities data, 2001 GIS data, and the 2000 Demographic and Health Survey\n\n\n\n(DHS) data. <sup>13</sup> [^13: We potentially could use the 2004 DHS data instead of the 2000 DHS data, but since the reliability of\nestimates depends on the various data sources been closer to each other in time, we use the 2000 DHS\nwhich is more consistent with the 2001 ANC survey, 2001 GIS data, and the 1998 Census.]\n\n\n\nSuch commune/district means are useful\n\nMention: \"2001 ANC survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its proximity in time to the census supports the chosen round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004401:5:0", "occurrence_id": "0bd9e3f46ddb5ce5a1728b4b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: health services and access to labor opportunities. In the paper, we look at all three channels\n\nwith a focus on the third one. We assume that access to roads has a different impact on\n\nincomes, depending on the type of activity carried out. We underline that poverty reduction\n\ndepends on the type of activity the household is involved in. Some activities could be\n\nconsidered as a poverty trap. Our particular interest lies in the difference between agricultural\n\nand non-farming activities since agriculture represented 61% of total employment in\n\nCameroon in 2001 and constitutes the major activity in which poor households are involved.\n\n\nOur second contribution lies in the use of data from the second Cameroonian national\n\nhousehold surveys (\"Deuxième Enquête Camerounaise Auprès des Ménages\", ECAM II)\n\nfrom which the infrastructure side has been poorly (if not) used in the literature. It covers all\n\nthe dimensions of poverty and introduces the basis of a monitoring and evaluation system of\n\nhouseholds' livelihoods. The three national households surveys (the first was in 1996 and the\n\nlast one in 2007) are also at the core of the reduction poverty plan for Cameroon, since\n\npoverty in Cameroon is of major concern since the 80s <sup>4</sup> . Economic growth began to recover\n\nfrom 1994 but the first national household surveys revealed that 50.5% of the Cameroon\n\npopulation lived under the poverty line in 1996. In 1998/99, private consumption was stagnant\n\nthus highlighting the fact that economic growth did not reach population and more\n\n\n\nimportantly the poorest part of it <sup>5</sup>\n\n\n\n.\n\n\n\nFinally, unlike the majority of cross-section analyses of poverty and road access, our results\n\ndo not call for huge, widespread investments in roads to fight against poverty. We\n\ndemonstrate that road access, proxied by the time (in hours) needed to reach the closest tarred\n\nroad, has no direct impact on consumption expenditures at the\n\nMention: \"Cameroonian national\n\nhousehold surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The second round supplies the infrastructure dimensions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004401:5:0", "occurrence_id": "a4dda75058952a3e1445c7ec", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: health services and access to labor opportunities. In the paper, we look at all three channels\n\nwith a focus on the third one. We assume that access to roads has a different impact on\n\nincomes, depending on the type of activity carried out. We underline that poverty reduction\n\ndepends on the type of activity the household is involved in. Some activities could be\n\nconsidered as a poverty trap. Our particular interest lies in the difference between agricultural\n\nand non-farming activities since agriculture represented 61% of total employment in\n\nCameroon in 2001 and constitutes the major activity in which poor households are involved.\n\n\nOur second contribution lies in the use of data from the second Cameroonian national\n\nhousehold surveys (\"Deuxième Enquête Camerounaise Auprès des Ménages\", ECAM II)\n\nfrom which the infrastructure side has been poorly (if not) used in the literature. It covers all\n\nthe dimensions of poverty and introduces the basis of a monitoring and evaluation system of\n\nhouseholds' livelihoods. The three national households surveys (the first was in 1996 and the\n\nlast one in 2007) are also at the core of the reduction poverty plan for Cameroon, since\n\npoverty in Cameroon is of major concern since the 80s <sup>4</sup> . Economic growth began to recover\n\nfrom 1994 but the first national household surveys revealed that 50.5% of the Cameroon\n\npopulation lived under the poverty line in 1996. In 1998/99, private consumption was stagnant\n\nthus highlighting the fact that economic growth did not reach population and more\n\n\n\nimportantly the poorest part of it <sup>5</sup>\n\n\n\n.\n\n\n\nFinally, unlike the majority of cross-section analyses of poverty and road access, our results\n\ndo not call for huge, widespread investments in roads to fight against poverty. We\n\ndemonstrate that road access, proxied by the time (in hours) needed to reach the closest tarred\n\nroad, has no direct impact on consumption expenditures at the\n\nMention: \"Deuxième Enquête Camerounaise Auprès des Ménages\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The full name of the round whose infrastructure side is underused.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004401:5:0", "occurrence_id": "dfdc18b31f9c193407aa1bc5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: health services and access to labor opportunities. In the paper, we look at all three channels\n\nwith a focus on the third one. We assume that access to roads has a different impact on\n\nincomes, depending on the type of activity carried out. We underline that poverty reduction\n\ndepends on the type of activity the household is involved in. Some activities could be\n\nconsidered as a poverty trap. Our particular interest lies in the difference between agricultural\n\nand non-farming activities since agriculture represented 61% of total employment in\n\nCameroon in 2001 and constitutes the major activity in which poor households are involved.\n\n\nOur second contribution lies in the use of data from the second Cameroonian national\n\nhousehold surveys (\"Deuxième Enquête Camerounaise Auprès des Ménages\", ECAM II)\n\nfrom which the infrastructure side has been poorly (if not) used in the literature. It covers all\n\nthe dimensions of poverty and introduces the basis of a monitoring and evaluation system of\n\nhouseholds' livelihoods. The three national households surveys (the first was in 1996 and the\n\nlast one in 2007) are also at the core of the reduction poverty plan for Cameroon, since\n\npoverty in Cameroon is of major concern since the 80s <sup>4</sup> . Economic growth began to recover\n\nfrom 1994 but the first national household surveys revealed that 50.5% of the Cameroon\n\npopulation lived under the poverty line in 1996. In 1998/99, private consumption was stagnant\n\nthus highlighting the fact that economic growth did not reach population and more\n\n\n\nimportantly the poorest part of it <sup>5</sup>\n\n\n\n.\n\n\n\nFinally, unlike the majority of cross-section analyses of poverty and road access, our results\n\ndo not call for huge, widespread investments in roads to fight against poverty. We\n\ndemonstrate that road access, proxied by the time (in hours) needed to reach the closest tarred\n\nroad, has no direct impact on consumption expenditures at the\n\nMention: \"ECAM II\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its infrastructure module covers dimensions ignored in the literature.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004402:23:1", "occurrence_id": "89b9e1bff942464acc32f9c1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 2\n\nCentral urban 5.5 3.0 2.5 0.0 0.0 -0.2 0.6\nEastern urban 16.9 6.4 6.4 0.0 0.0 0.6 0.2\nNorthern urban 38.3 6.0 0.9 0.0 5.0 0.7 0.3\nWestern urban 9.3 0.6 0.6 -0.4 0.0 -0.9 0.0\n\nCentral rural 20.9 3.3 3.3 -0.2 -0.6 -1.2 -0.1\nEastern rural 37.5 0.5 1.0 -0.2 -0.1 -1.9 -0.2\nNorthern rural 57.6 8.3 0.6 0.0 7.7 0.8 0.0\nWestern rural 21.4 1.0 1.4 0.8 0.1 -0.6 -0.2\n\n<u>IDP</u> <u>77.9</u> <u>0.9</u> <u>0.1</u> <u>0.0</u> <u>-0.2</u> <u>0.6</u> <u>-0.4</u>\n<u>Source: Author’s simulations based on 2005–06 UNHS</u>\nNote: * = includes IDP camps\n\n\n\nBase\n<u>(2005–06)</u>\n\n\n\n<u>All 5</u>\n<u>foods</u> <u>Maize</u> <u>Matooke</u> <u>Cassava</u>\n\n\n\n<u>Impact of observed price increases</u>\n<u>(percentage point change\n\nMention: \"2005–06 UNHS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Simulations of food price effects run on this survey round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004404:9:0", "occurrence_id": "8d538fff853e6a001639489a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: for the U.S. and 7 cents/mile for the U.K. From Carisma and Lowder(2008) we\ntake values of deadweight loss European countries based on an INFRAS/IWW 2000\nstudy. Transforming the 0.68% deadweight loss value for U.K. into 7 cents/mile, we\ncompute the values in cents/mile for these other countries. From Carisma and Lowder(2008) we also obtain results for most of the other countries which the authors\ncompile from separate studies. The Korea Transport Institute’s latest estimate of\ncongestion costs is of 2.97% of GDP in 2004 while the Ministry of Land, Infrastructure and Transport in Japan estimated congestion costs for this country at around\n2% of GDP in 2000. For Sao Paulo, a cost of 2.4% of GDP was estimated by\nWilloughby (2000); for Mexico City, a cost of 2.65% of GDP by Ochoa and Radian\n(1997). We use the results for these cities for Brazil and Mexico. For India, we\nassume the numbers to be similar to those in Bangkok and Kuala Lumpur, 2.1 and\n1.8% of GDP respectively according to a study on sustainable infrastructure in Asia\ncompleted by the United Nations Economic and Social Commission for Asia and\nthe Pacific (2007). Creutzig and He (2009) estimate the congestion cost for China\nis 3.35% of GDP and the Russian government has estimated that congestion costs\nfor Russia are about 3% of GDP. <sup>7</sup> [^7: http://www.latimes.com/news/nationworld/world/la-fg-russia-roads15-2009jul15,0,6188164.story] We set the value of congestion cost at 2.15% of\nGDP. Finally, from Carisma and Lowder(2008) we obtain the cost of congestion for\nU.K. as a percent of GDP to be 3% from which we then transform the other figures\ninto cents/\n\nMention: \"INFRAS/IWW 2000\nstudy\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its deadweight loss estimates for European countries are carried over.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004405:20:2", "occurrence_id": "74d6eb388537a1a63ae7f5f6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: u>**\n\n\n\n**<u>Real</u>** **<u>annual</u>**\n**sales growth,**\n**<u>%</u>**\n\n\n\n**<u>Type of firm</u>**\n\n\n\n**Return** **on**\n**<u>Equity, %</u>**\n\n\n\n**EBIT to interest**\n**<u>expenses, x</u>**\n\n\n\n**Current to total**\n**<u>liabilities, %</u>**\n\n\n\n**<u>Total debt to</u>**\n**common equity,**\n**<u>%</u>**\n\n\n\n**Total liabilities**\n**<u>to Equity, %</u>**\n\n\n\n**Paid-in to total**\n**<u>capital, %</u>**\n\n\n\n**<u>Type of firm</u>**\n\n\n\n**<u>Retained</u>**\n**earnings to**\n**<u>equity, %</u>**\n\n\n\nPrivate 33.5 4.4 27.7 84.1 74.5 36.1\n<u>Former SOE</u> <u>37.0</u> <u>6.0</u> <u>27.1</u> <u>90.3</u> <u>104.2</u> <u>30.1</u>\n\n<u>Source: Coface; World Bank staff analysis</u>\n\nSince most privatizations occurred in the consumer staples industry, the comparison could be\ninfluenced by industry particulars. Moreover, former SOEs\n\nMention: \"Coface\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Firm performance by ownership type is drawn from this insurer's data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004410:3:0", "occurrence_id": "d3867377835922b485dd8685", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **1.** **Introduction**\n\nWith the development spotlight increasingly focused on Sub-Saharan Africa (SSA),\n\n\n\nthere is renewed interest in understanding the constraints to higher growth in the region <sup>1</sup>\n\n\n\n.\n\n\n\nRecent findings have emphasized the role of the broader investment climate in shaping the\n\n\nopportunities and incentives for enterprises to invest, create jobs and expand (World Bank\n\n\nWDR 2004). This paper uses detailed enterprise-level data to examine how different\n\n\ndimensions of the investment climate –availability of infrastructure, access to finance, the\n\n\nregulatory environment and corruption -- affect patterns of employment growth. In addition\n\n\nto testing which dimensions are most associated with higher growth, there remains a question\n\n\nwhether the priorities for reform are common across countries. Thus, the aim is to compare\n\n\nthe impact of the different dimensions of the investment climate in SSA with those in other\n\n\nregions, particularly to test whether the top constraints faced by entrepreneurs in SSA have a\n\n\ncommon impact in low-income countries in other regions.\n\n\nTo examine the impact of the investment climate on employment growth, the paper\n\n\nuses data from over 17,000 enterprises in 31 SSA countries included in the World Bank’s\n\n\nEnterprise Surveys (ES). To put the experience of SSA into a larger global context, results\n\n\nare compared with those of 48 other low or lower-middle income countries (LOW) and 26\n\n\n\nThese datasets are\n\n\n\nupper-middle or high income countries (HIGH) in other regions.\n\n\n\n2\n\n\n\ncollected with a comparable instrument and sampling strategy, providing a unique source of\n\n\ndisaggregated information on many dimensions of interest of the investment climate.\n\n\nCovering the full range of enterprise sizes, they also allow for an analysis of how conditions\n\n\nvary across enterprise types that proves to be important in highlighting priorities for reform.\n\n\n1 See the Africa Action Plan, commitments of the G8 to focus on SSA development, the Stern Commission for\nAfrica, the World Bank’s Six Strategic Pillars, among others.\n\n2\n\nMention: \"Enterprise Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Over seventeen thousand firms across the region are covered here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004413:11:0", "occurrence_id": "77bd24fc278dca9c9b706f7f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 10. Import Content of Exports by Manufacture Sectors, Argentina, 1997**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n_Source_ : Author’s calculations based on INDEC data.\n\n\n**3.** **TRADE PERFORMANCE**\n\n\n**Structural transformation in Argentina has been scarce, as evidenced by the fact that the**\n**Argentinean basket and structure of export products has not changed dramatically over the past**\n**decade.** This results in a relatively low and stagnated sophistication of its exports, and specialization\nin activities that appear to offer relatively little scope for technological catch-up to the world frontier. <sup>7</sup> [^7: Sanchez, G. and Butler, I. (2008), Competitiveness and Growth in Argentina: Appropriability, Misallocation or\nDisengagement? IERAL - Fundación Mediterranea.]\n\n\n**At the same time, Argentina’s exports are well diversified compared to the average in LAC, even**\n**though product concentration has increased over the past decade.** In 2007 almost 30 percent of\ntotal exports were concentrated in ten products, up from 25 percent in 1997 (Table 1). The product\nconcentration index (Figure 11) has been stable over the last decade, with a slight increase in 20012003 (clearly due to a decrease in exports diversification due to the crisis) and then a decrease in 20032006. The recent decrease is probably related to a revival in manufacturing after the crisis related to\nthe commodity boom and import substitution following the devaluation of the peso and the rebound in\ndomestic demand. Compared to LAC’s average and other competitors like Chile and Russia,\nArgentina’ product concentration is fairly low. However, Argentina’s product concentration is high\ncompared to other important competitors such as Brazil. Moreover, product concentration started\nincreasing again in 2007, a sign that the country may need to reconsider boosting export\ndiversification in its trade policy.\n\n\n**Figure 11. Product Concentration Index for Argentina and Main Competitors 1990-2007**\n\n\n\n\n\n\n\n\n\n\n\n_Source\n\nMention: \"INDEC data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Export shares by manufacturing sector are computed from these statistics.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004414:35:0", "occurrence_id": "5ca447309efd3ebd57a7287d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |Definition and construction Source|Standard deviation of log difference of GDP per capita Authors’ construction using World Bank WDI Herfindahl Index HS-4 Digit World Bank diversification database Share of 5 largest products in overall exports World Bank diversification database Share of 10 largest products in overall exports World Bank diversification database First principal component of 3 product diversification measures Authors’ construction Herfindahl Index for destination markets Authors’ construction with UN COMTRADE Share of 5 largest destination markets in overall exports Authors’ construction with UN COMTRADE Share of 10 largest destination markets in overall exports Authors’ construction with UN COMTRADE First principal component of 3 market diversification measures Authors’ construction Total trade (exports + imports) divided by GDP Chang et al. (2009) Restrictions on cross-border financial transactions Chinn & Ito (2008) Capital flows to the region as a percentage of GDP Caldero´n et al. (2005) Standard deviation of growth rate of main trading partners Chang et al. (2009) Standard deviation of terms of trade (ToT) Chang et al. (2009) Standard deviation of real effective exchange rate (REER) Authors’ construction based on IMF IFS Standard deviation of annual log differences of CPI Chang et al. (2009) Number of years a country underwent a banking crisis Chang et al. (2009) Log of the value of GDP per capita in the initial period World Bank WDI Log difference of GDP per capita in constant USD World Bank WDI Initial value of ratio of secondary gross enrollment World Bank WDI Standard deviation of government expenditures Chang et al. (2009) Portion of years with natural disasters Chang et al. (2009) Index constructed from ICRG indicators Chang et al. (2009) Index of civil liberties and political rights Freedom House (2009) Number of years a country experienced political iterregna MarshallJaggers:2008 Number of years a country experienced conflict Sarkees & Schafer (2000) Number of attempted\n\nMention: \"World Bank WDI Herfindahl Index\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Concentration at four-digit level is constructed from this index.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004414:35:0", "occurrence_id": "bb533067e449bdcacb57ffe8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |Definition and construction Source|Standard deviation of log difference of GDP per capita Authors’ construction using World Bank WDI Herfindahl Index HS-4 Digit World Bank diversification database Share of 5 largest products in overall exports World Bank diversification database Share of 10 largest products in overall exports World Bank diversification database First principal component of 3 product diversification measures Authors’ construction Herfindahl Index for destination markets Authors’ construction with UN COMTRADE Share of 5 largest destination markets in overall exports Authors’ construction with UN COMTRADE Share of 10 largest destination markets in overall exports Authors’ construction with UN COMTRADE First principal component of 3 market diversification measures Authors’ construction Total trade (exports + imports) divided by GDP Chang et al. (2009) Restrictions on cross-border financial transactions Chinn & Ito (2008) Capital flows to the region as a percentage of GDP Caldero´n et al. (2005) Standard deviation of growth rate of main trading partners Chang et al. (2009) Standard deviation of terms of trade (ToT) Chang et al. (2009) Standard deviation of real effective exchange rate (REER) Authors’ construction based on IMF IFS Standard deviation of annual log differences of CPI Chang et al. (2009) Number of years a country underwent a banking crisis Chang et al. (2009) Log of the value of GDP per capita in the initial period World Bank WDI Log difference of GDP per capita in constant USD World Bank WDI Initial value of ratio of secondary gross enrollment World Bank WDI Standard deviation of government expenditures Chang et al. (2009) Portion of years with natural disasters Chang et al. (2009) Index constructed from ICRG indicators Chang et al. (2009) Index of civil liberties and political rights Freedom House (2009) Number of years a country experienced political iterregna MarshallJaggers:2008 Number of years a country experienced conflict Sarkees & Schafer (2000) Number of attempted\n\nMention: \"World Bank diversification database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Shares of the largest export products come from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004414:35:0", "occurrence_id": "e9cf7c8597d054be0bfad9dd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |Definition and construction Source|Standard deviation of log difference of GDP per capita Authors’ construction using World Bank WDI Herfindahl Index HS-4 Digit World Bank diversification database Share of 5 largest products in overall exports World Bank diversification database Share of 10 largest products in overall exports World Bank diversification database First principal component of 3 product diversification measures Authors’ construction Herfindahl Index for destination markets Authors’ construction with UN COMTRADE Share of 5 largest destination markets in overall exports Authors’ construction with UN COMTRADE Share of 10 largest destination markets in overall exports Authors’ construction with UN COMTRADE First principal component of 3 market diversification measures Authors’ construction Total trade (exports + imports) divided by GDP Chang et al. (2009) Restrictions on cross-border financial transactions Chinn & Ito (2008) Capital flows to the region as a percentage of GDP Caldero´n et al. (2005) Standard deviation of growth rate of main trading partners Chang et al. (2009) Standard deviation of terms of trade (ToT) Chang et al. (2009) Standard deviation of real effective exchange rate (REER) Authors’ construction based on IMF IFS Standard deviation of annual log differences of CPI Chang et al. (2009) Number of years a country underwent a banking crisis Chang et al. (2009) Log of the value of GDP per capita in the initial period World Bank WDI Log difference of GDP per capita in constant USD World Bank WDI Initial value of ratio of secondary gross enrollment World Bank WDI Standard deviation of government expenditures Chang et al. (2009) Portion of years with natural disasters Chang et al. (2009) Index constructed from ICRG indicators Chang et al. (2009) Index of civil liberties and political rights Freedom House (2009) Number of years a country experienced political iterregna MarshallJaggers:2008 Number of years a country experienced conflict Sarkees & Schafer (2000) Number of attempted\n\nMention: \"IMF IFS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its price series give the inflation volatility measure.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004414:35:0", "occurrence_id": "e60f4e2bab25a8a9aee98405", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |Definition and construction Source|Standard deviation of log difference of GDP per capita Authors’ construction using World Bank WDI Herfindahl Index HS-4 Digit World Bank diversification database Share of 5 largest products in overall exports World Bank diversification database Share of 10 largest products in overall exports World Bank diversification database First principal component of 3 product diversification measures Authors’ construction Herfindahl Index for destination markets Authors’ construction with UN COMTRADE Share of 5 largest destination markets in overall exports Authors’ construction with UN COMTRADE Share of 10 largest destination markets in overall exports Authors’ construction with UN COMTRADE First principal component of 3 market diversification measures Authors’ construction Total trade (exports + imports) divided by GDP Chang et al. (2009) Restrictions on cross-border financial transactions Chinn & Ito (2008) Capital flows to the region as a percentage of GDP Caldero´n et al. (2005) Standard deviation of growth rate of main trading partners Chang et al. (2009) Standard deviation of terms of trade (ToT) Chang et al. (2009) Standard deviation of real effective exchange rate (REER) Authors’ construction based on IMF IFS Standard deviation of annual log differences of CPI Chang et al. (2009) Number of years a country underwent a banking crisis Chang et al. (2009) Log of the value of GDP per capita in the initial period World Bank WDI Log difference of GDP per capita in constant USD World Bank WDI Initial value of ratio of secondary gross enrollment World Bank WDI Standard deviation of government expenditures Chang et al. (2009) Portion of years with natural disasters Chang et al. (2009) Index constructed from ICRG indicators Chang et al. (2009) Index of civil liberties and political rights Freedom House (2009) Number of years a country experienced political iterregna MarshallJaggers:2008 Number of years a country experienced conflict Sarkees & Schafer (2000) Number of attempted\n\nMention: \"ICRG indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"An index of political risk is built from these indicators.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004418:25:1", "occurrence_id": "fabbdf3f2da2e8bada782552", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \ni\n\n\n\n= 1 � s <sup>Q</sup> L;i\n\n\n\n\n<sup>Q</sup> L;i <sup>�</sup> <sup>sQ</sup>\n\n\n\n\n<sup>j</sup> L <sup>+</sup> <sup>b\"j</sup>\n\n\n\nchoice.\n\nwhere <sup>b</sup> \n\n\njK <sup>+</sup> <sup>b\"j</sup> L\n\n\n\nM <sup>is</sup> <sup>the</sup> <sup>degree</sup> <sup>of</sup> <sup>returns</sup> <sup>to</sup> <sup>scale</sup> <sup>in</sup> <sup>sector</sup> <sup>j.</sup> <sup>The</sup> <sup>result</sup> <sup>are</sup> <sup>robust</sup> <sup>to</sup> <sup>this</sup>\n\n\n\nchoice.\n\n\n\nj = b\"j\n\n\n##### 8 Results\n\nWe will discuss now the empirical results obtained when the production function is estimated on\n\n\nthe …rm level data contained in Amadeus for Belgium, France, Great Britain, Italy, Spain over the\n\n\nperiod 1998-2005. To avoid overburdening the reader, we report results for selected estimators\n\n\n(OLS, System GMM, and Olley and Pakes) for only one of our countries, Belgium.\n\n\nThe estimation results\n\nMention: \"rm level data contained in Amadeus\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The production function is estimated on firm records across five countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004423:3:1", "occurrence_id": "46de2c928977bd8ebe7f1986", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  within India by caste and tribal status have received greater attention over\nthe last few decades (see Murthi et al, 1996; Dyson and Moore, 1993; Dreze and Gazdar, 1997;\nMaharatna, 1998; 2000). The preoccupation of such studies on child mortality has however been\non correlates of mortality and the slow decline in child mortality in recent years. There is a good\nreason for this as India looks likely to falter on achieving this important Millennium\nDevelopment Goal – a fear expressed in 2000 by Claeson et al.\n\nThe anthropological literature has also focused on access to health for India’s Adivasis. Excess\nmortality of tribal children during monsoons in particular has been analyzed by activists more\nrecently through a food security lens. The Right to Food movement and the recent Supreme\nCourt judgments exhort governments to take more aggressive action to address malnutrition\nnationally but more so in tribal areas. The evidence from the grassroots highlights the peculiar\nproblems in tribal areas because of which poverty is higher, and health and education outcomes\nare lower. There is for instance, wide acknowledgment that excess mortality in childhood for\ntribals is partly due to poverty and partly to poor access to services. Our own previous analysis\nof mortality using data from India’s Reproductive and Child Health Surveys, 2005 (see World\nBank, 2006) and a recent study on Orissa also shows that once poverty levels are controlled for,\nthe effect of tribal (or indeed Dalit or Scheduled Caste (SC)) status goes away. By and large,\ntherefore, much of the analytical work on Scheduled Tribes (STs) in India has confirmed what\nwe already know about child mortality and its correlates. This paper is an attempt to drill deeper\n\n\n1 The authors are grateful to Julie McLaughlin and Rinku Murgai for comments.\n2 We use the terms Scheduled Tribes, tribes and the self-preferred term Adivasis interchangeably in this paper.\n\n\n2\n\nMention: \"India’s Reproductive and Child Health Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its mortality data support the poverty and access analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004431:16:0", "occurrence_id": "261a1979b9858fd84364876d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **<u>Table 2. CDM Pipeline Inventory of Projects, by Project Type</u>**\n\n\n\nAnnual CERs\nNumber of Projects\n\n\n\nAnnual CERs Annual CERs\n\n<u>2008-2012</u> <u>2008-2020</u>\n\n\n\n<u>2008-2012</u> <u>2008-2020</u>\n\n\n4,631 2,786,791 7,416,430\n\n\n\n<u>Project type</u> Share of Total (%)\nAfforestation 0.11 0.01 0.18\nAgriculture 0.02 0.01 0.01\nBiomass energy 14.29 7.02 6.49\nCement 0.63 1.13 0.79\nCO2 capture 0.06 0.01 0.01\nCoal bed/mine methane 1.45 4.62 4.71\nEE households 0.45 0.14 0.14\nEE industry 3.13 0.78 0.61\nEE own generation 9.78 9.27 8.68\nEE service 0.37 0.03 0.03\nEE supply side 1.32 1.39 2.70\nEnergy distribution 0.19 0.35 0.34\nFossil fuel switch 2.33 6.37 6.13\nFugitive gas 0.50 1.92 1.66\nGeothermal 0.32 0.60 0.58\nHFCs 0.48 17.10 14.83\nHydro 26.97 17.23 20.75\nLandfill gas 5.94 7.58 6.78\nMethane avoidance 11.34 4.25 3.73\nN2O 1.45 8.95 8.44\nPFCs and SF6 0.30 0.46 0.54\nReforestation 0.95 0.49 0.81\nSolar 0.78 0.08 0.10\nTidal 0.02 0.04 0.05\nTransport 0.26 0.18 0.18\n\nMention: \"CDM Pipeline Inventory of Projects\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Project counts and annual credited reductions by type are tabulated here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004432:25:0", "occurrence_id": "d44599fb1f54b4019138d51f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 5. **_We argue that medium term fiscal policy should usefully be guided by an_**\n**_approach similar to “permanent” or sustained fiscal adjustment scenario_**\n**_outlined here._** This approach results in a feasible, sustained fiscal adjustment,\nreturning fiscal policy on a long-term sustainable path within a reasonable amount\nof time. But the transition to that scenario will require coordinated and _sustained_\nrevenue, expenditure, and financing measures. The size of adjustment in each year\nwill also have to be realistic in order to avoid backsliding.\n\n\n**<u>Table 4: Assumptions for fiscal adjustment scenarios</u>**\n\n\n\n\n\n\n\n\n\n\n\n\n\n_Source: World Bank staff estimates. Data for 2008 is based on the data from the Ministry of Finance._\n\n\n**How Fast Does Russia Accumulate Net Debt Under Alternative Scenarios?**\n\n\n**_With different fiscal strategies specified, we run the fiscal sustainability model to derive_**\n**_illustrative baseline projections for net public debt throughout the projection period._**\nNote that as explained in the annex, this framework also incorporates the dynamics of the\noil fund assets and consequently, an oil fund rule to deal with the large but volatile oil\nrevenue in Russia. In Figure 19, we present results of fiscal sustainability analysis for the\nfiscal strategies outlined in the previous section. The diagram shows the net debt to GDP\nratio, including the Oil Stabilization Fund stock as a negative item, under the various\nspending programs. The starting point is a negative number: the Oil Fund and other\nforeign exchange assets exceed the gross public debt by a substantial margin reflecting\nRussia’s de facto _net creditor_ position.\n\n\n24\n\nMention: \"the Ministry of Finance\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The baseline year's fiscal data come from this ministry.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004409:16:1", "occurrence_id": "c4c121e725220e0b01dc4373", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: t_ is time-period e\u001bect.\n\nFigure 1 shows the estimated coe\u001ecients over the number of interactions corresponding to a\n\nrolling window of 60 observations. <sup>9</sup> This preliminary evidence shows that, in fact, the semi-elasticity\n\n\nof R&D over GDP with respect to educational attainment is generally positive, but it is clearly a\n\n\nnon-linear function. The relationship between R&D and human capital is unstable and rising with\n\n\nthe rank of the IPR index. Furthermore, the changes in the semi-elasticity seem to be discrete\n\n\nand unpredictable. It is zero in the samples with the worst levels of IPRs, then abruptly rises in\n\n\nthe middle of sample, and stabilizes towards the end of the sample. These abrupt changes in the\n\n\nrelevant semi-elasticity are not due to abrupt changes in the IPR index as we move up the rankings\n\n\nof IPRs. Considering that the each iteration involves a set of observations with increasing IPR\n\n\nindex, the slope of the curve in Figure 1 corresponds approximately to the cross derivative of R&D\n\n\nshare with respect to human capital and the IPR rank. Thus, we expect that this cross derivative\n\n\ncould be positive on average for the whole sample. In any case, we discuss the results from our\n\n\ntwo-stage estimations further below.\n\n\n9We excluded one observation from the data, namely for El Salvador in 1980, as the Lederman and Saenz data\nhad a value of 2.27% of GDP. This data point is consistent with the RICyT data, but it is impossibly high for a\npoor developing economy, and there were no data points within ve years of this observation. Estimations with this\nobservation also yielded notable unpredictable non-linearities. The corresponding graph is available from the authors\nupon request. We are grateful to Bill Maloney and Edwin Goni for pointing out this outlier.\n\n\n14\n\nMention: \"RICyT data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"The implausible transport cost share is checked against this source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004436:10:0", "occurrence_id": "0f4e3a2dea73dcbaa30426fe", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: computerization to that of a reduction in stamp duties can ideally provide a bound on the magnitude of the\n\n\nassociated reduction in transaction cost.\n\n\n**3. Data and estimation strategy**\n\n\nData on credit disbursed by commercial banks and the number of total registered transactions for 1997\n\n2007 point towards significant increases in supply of credit as well as land sales. To assess whether any of\n\n\nthese increases can be attributed to computerization, we use the roll-out of this intervention across SROs\n\n\nto identify the impact of computerization on credit supply and transaction frequency in a way that can be\n\n\ncompared to that of stamp duty reductions. Below we discuss the approach as well as the assumptions\n\n\nrequired to hold in order to be able to do so.\n\n\n**3.1 Data and descriptive evidence**\n\n\nWe have quarterly data on credit disbursed by all scheduled commercial banks to retail customers for a 11\n\nyear period in all of AP’s 1064 _taluks_ (blocks) from India’s Central Bank (Reserve Bank of India, RBI). <sup>5</sup> [^5: Scheduled commercial banks include all public and private sector banks with the exception of cooperative banks who\ncontributed less than 10 percent to total lending from 1997 to 2006 (RBI 2008).]\n\n\nThis provides a total of 44 observations per _taluk_, from Q1 of 1997 to Q4 of 2007, i.e at least 7 quarters\n\n\nbefore the first batch of registry offices was computerized to 2 years after the last of the state’s SROs had\n\n\nshifted to operate on a fully computerized basis. The quality and reliability of these data derives from the\n\n\nfact that they are based on routine reports which all scheduled commercial banks are required to submit to\n\n\nRBI at regular intervals. To match them to information on computerization, we aggregate up from the\n\n\nbranch to the _taluk_ level. This information is combined with annual data on the total number of registered\n\n\nland transactions (incl. sales and mortgages separately) for the period of 1997-2007 from the Andhra\n\n\nPradesh Department of Stamps\n\nMention: \"Data on credit disbursed by commercial banks\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Bank lending and registered transaction counts over this decade show supply growth.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004438:26:0", "occurrence_id": "c83d9e1a516f4cc460ab0d1d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **TABLE 1. Description of Household Surveys**\n\n\nCountry Name of Survey Survey years Obs.\n\n\nArgentina Encuesta Permanente de Hogares (EPH‐C) semestre II 2004, 2005 167,770\nBrazil Pesquisa Nacional por Amostra de Domicilios (PNAD) 2002, 2003, 2004 1,169,598\nChile Encuesta de Caracterización Socioeconómica Nacional (CASEN) 2000, 2003 509,825\n\n\nColombia Encuesta Continua de Hogares (ECH) 2001, 2003, 2004 315,528\nCosta Rica Encuesta de Hogares de Propósitos Múltiples (HPM) 2001, 2002, 2003, 2004 173,403\nDominican Rep. Encuesta Nacional de Fuerza de Trabajo (ENFT) onda Octubre 2000, 2001, 2002, 2003, 2004, 2005, 2006 184,611\n\n\nEcuador Encuesta de Empleo, Desempleo y Subempleo (ENEMDU) 2003, 2004, 2005 242,410\nEl Salvador Encuesta de Hogares de Propósitos Múltiples (EHPM) 2000, 2001, 2002, 2003, 2004, 2005 409,093\nGuatemala Encuesta Nacional de Empleo e Ingresos (ENEI) 2002, 2003, 2004 91,343\nHonduras Encuesta Permanente de Hogares de Propósitos Múltiples (EPHPM) 2001, 2004, 2005, 2006 206,868\n\n\nMexico Encuesta Nacional de Ingresos y Gastos de los Hogares (ENIGH) 2000, 2002, 2004, 2005, 2006 384,168\n\n\n\nNicaragua Encuesta Nacional de Hogares sobre Medición de Nivel de Vida\n(EMNV)\n\n\n\n2001, 2005 59,424\n\n\n\nPanama Encuesta de Hogares (EH) 2001, 2002, 2003, 2004\n\nMention: \"Encuesta Permanente de Hogares\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its 167,770 observations over these years enter the pooled household sample.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004438:26:0", "occurrence_id": "cec1bb929b7aef688625a32a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **TABLE 1. Description of Household Surveys**\n\n\nCountry Name of Survey Survey years Obs.\n\n\nArgentina Encuesta Permanente de Hogares (EPH‐C) semestre II 2004, 2005 167,770\nBrazil Pesquisa Nacional por Amostra de Domicilios (PNAD) 2002, 2003, 2004 1,169,598\nChile Encuesta de Caracterización Socioeconómica Nacional (CASEN) 2000, 2003 509,825\n\n\nColombia Encuesta Continua de Hogares (ECH) 2001, 2003, 2004 315,528\nCosta Rica Encuesta de Hogares de Propósitos Múltiples (HPM) 2001, 2002, 2003, 2004 173,403\nDominican Rep. Encuesta Nacional de Fuerza de Trabajo (ENFT) onda Octubre 2000, 2001, 2002, 2003, 2004, 2005, 2006 184,611\n\n\nEcuador Encuesta de Empleo, Desempleo y Subempleo (ENEMDU) 2003, 2004, 2005 242,410\nEl Salvador Encuesta de Hogares de Propósitos Múltiples (EHPM) 2000, 2001, 2002, 2003, 2004, 2005 409,093\nGuatemala Encuesta Nacional de Empleo e Ingresos (ENEI) 2002, 2003, 2004 91,343\nHonduras Encuesta Permanente de Hogares de Propósitos Múltiples (EPHPM) 2001, 2004, 2005, 2006 206,868\n\n\nMexico Encuesta Nacional de Ingresos y Gastos de los Hogares (ENIGH) 2000, 2002, 2004, 2005, 2006 384,168\n\n\n\nNicaragua Encuesta Nacional de Hogares sobre Medición de Nivel de Vida\n(EMNV)\n\n\n\n2001, 2005 59,424\n\n\n\nPanama Encuesta de Hogares (EH) 2001, 2002, 2003, 2004\n\nMention: \"Pesquisa Nacional por Amostra de Domicilios\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its 1.17 million observations over these years enter the pooled household sample.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004438:26:0", "occurrence_id": "9a7493801a2e088da016bff2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **TABLE 1. Description of Household Surveys**\n\n\nCountry Name of Survey Survey years Obs.\n\n\nArgentina Encuesta Permanente de Hogares (EPH‐C) semestre II 2004, 2005 167,770\nBrazil Pesquisa Nacional por Amostra de Domicilios (PNAD) 2002, 2003, 2004 1,169,598\nChile Encuesta de Caracterización Socioeconómica Nacional (CASEN) 2000, 2003 509,825\n\n\nColombia Encuesta Continua de Hogares (ECH) 2001, 2003, 2004 315,528\nCosta Rica Encuesta de Hogares de Propósitos Múltiples (HPM) 2001, 2002, 2003, 2004 173,403\nDominican Rep. Encuesta Nacional de Fuerza de Trabajo (ENFT) onda Octubre 2000, 2001, 2002, 2003, 2004, 2005, 2006 184,611\n\n\nEcuador Encuesta de Empleo, Desempleo y Subempleo (ENEMDU) 2003, 2004, 2005 242,410\nEl Salvador Encuesta de Hogares de Propósitos Múltiples (EHPM) 2000, 2001, 2002, 2003, 2004, 2005 409,093\nGuatemala Encuesta Nacional de Empleo e Ingresos (ENEI) 2002, 2003, 2004 91,343\nHonduras Encuesta Permanente de Hogares de Propósitos Múltiples (EPHPM) 2001, 2004, 2005, 2006 206,868\n\n\nMexico Encuesta Nacional de Ingresos y Gastos de los Hogares (ENIGH) 2000, 2002, 2004, 2005, 2006 384,168\n\n\n\nNicaragua Encuesta Nacional de Hogares sobre Medición de Nivel de Vida\n(EMNV)\n\n\n\n2001, 2005 59,424\n\n\n\nPanama Encuesta de Hogares (EH) 2001, 2002, 2003, 2004\n\nMention: \"Encuesta de Caracterización Socioeconómica Nacional\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its 509,825 observations over these years enter the pooled household sample.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004438:26:0", "occurrence_id": "6bdede7739f8b7a1fc033a69", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **TABLE 1. Description of Household Surveys**\n\n\nCountry Name of Survey Survey years Obs.\n\n\nArgentina Encuesta Permanente de Hogares (EPH‐C) semestre II 2004, 2005 167,770\nBrazil Pesquisa Nacional por Amostra de Domicilios (PNAD) 2002, 2003, 2004 1,169,598\nChile Encuesta de Caracterización Socioeconómica Nacional (CASEN) 2000, 2003 509,825\n\n\nColombia Encuesta Continua de Hogares (ECH) 2001, 2003, 2004 315,528\nCosta Rica Encuesta de Hogares de Propósitos Múltiples (HPM) 2001, 2002, 2003, 2004 173,403\nDominican Rep. Encuesta Nacional de Fuerza de Trabajo (ENFT) onda Octubre 2000, 2001, 2002, 2003, 2004, 2005, 2006 184,611\n\n\nEcuador Encuesta de Empleo, Desempleo y Subempleo (ENEMDU) 2003, 2004, 2005 242,410\nEl Salvador Encuesta de Hogares de Propósitos Múltiples (EHPM) 2000, 2001, 2002, 2003, 2004, 2005 409,093\nGuatemala Encuesta Nacional de Empleo e Ingresos (ENEI) 2002, 2003, 2004 91,343\nHonduras Encuesta Permanente de Hogares de Propósitos Múltiples (EPHPM) 2001, 2004, 2005, 2006 206,868\n\n\nMexico Encuesta Nacional de Ingresos y Gastos de los Hogares (ENIGH) 2000, 2002, 2004, 2005, 2006 384,168\n\n\n\nNicaragua Encuesta Nacional de Hogares sobre Medición de Nivel de Vida\n(EMNV)\n\n\n\n2001, 2005 59,424\n\n\n\nPanama Encuesta de Hogares (EH) 2001, 2002, 2003, 2004\n\nMention: \"Encuesta Continua de Hogares\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its 315,528 observations over these years enter the pooled household sample.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004438:26:0", "occurrence_id": "4fa113f40348908b37bf705a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **TABLE 1. Description of Household Surveys**\n\n\nCountry Name of Survey Survey years Obs.\n\n\nArgentina Encuesta Permanente de Hogares (EPH‐C) semestre II 2004, 2005 167,770\nBrazil Pesquisa Nacional por Amostra de Domicilios (PNAD) 2002, 2003, 2004 1,169,598\nChile Encuesta de Caracterización Socioeconómica Nacional (CASEN) 2000, 2003 509,825\n\n\nColombia Encuesta Continua de Hogares (ECH) 2001, 2003, 2004 315,528\nCosta Rica Encuesta de Hogares de Propósitos Múltiples (HPM) 2001, 2002, 2003, 2004 173,403\nDominican Rep. Encuesta Nacional de Fuerza de Trabajo (ENFT) onda Octubre 2000, 2001, 2002, 2003, 2004, 2005, 2006 184,611\n\n\nEcuador Encuesta de Empleo, Desempleo y Subempleo (ENEMDU) 2003, 2004, 2005 242,410\nEl Salvador Encuesta de Hogares de Propósitos Múltiples (EHPM) 2000, 2001, 2002, 2003, 2004, 2005 409,093\nGuatemala Encuesta Nacional de Empleo e Ingresos (ENEI) 2002, 2003, 2004 91,343\nHonduras Encuesta Permanente de Hogares de Propósitos Múltiples (EPHPM) 2001, 2004, 2005, 2006 206,868\n\n\nMexico Encuesta Nacional de Ingresos y Gastos de los Hogares (ENIGH) 2000, 2002, 2004, 2005, 2006 384,168\n\n\n\nNicaragua Encuesta Nacional de Hogares sobre Medición de Nivel de Vida\n(EMNV)\n\n\n\n2001, 2005 59,424\n\n\n\nPanama Encuesta de Hogares (EH) 2001, 2002, 2003, 2004\n\nMention: \"Encuesta de Hogares de Propósitos Múltiples\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its 173,403 observations over these years enter the pooled household sample.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004438:26:0", "occurrence_id": "e1651118646fe203898048a0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **TABLE 1. Description of Household Surveys**\n\n\nCountry Name of Survey Survey years Obs.\n\n\nArgentina Encuesta Permanente de Hogares (EPH‐C) semestre II 2004, 2005 167,770\nBrazil Pesquisa Nacional por Amostra de Domicilios (PNAD) 2002, 2003, 2004 1,169,598\nChile Encuesta de Caracterización Socioeconómica Nacional (CASEN) 2000, 2003 509,825\n\n\nColombia Encuesta Continua de Hogares (ECH) 2001, 2003, 2004 315,528\nCosta Rica Encuesta de Hogares de Propósitos Múltiples (HPM) 2001, 2002, 2003, 2004 173,403\nDominican Rep. Encuesta Nacional de Fuerza de Trabajo (ENFT) onda Octubre 2000, 2001, 2002, 2003, 2004, 2005, 2006 184,611\n\n\nEcuador Encuesta de Empleo, Desempleo y Subempleo (ENEMDU) 2003, 2004, 2005 242,410\nEl Salvador Encuesta de Hogares de Propósitos Múltiples (EHPM) 2000, 2001, 2002, 2003, 2004, 2005 409,093\nGuatemala Encuesta Nacional de Empleo e Ingresos (ENEI) 2002, 2003, 2004 91,343\nHonduras Encuesta Permanente de Hogares de Propósitos Múltiples (EPHPM) 2001, 2004, 2005, 2006 206,868\n\n\nMexico Encuesta Nacional de Ingresos y Gastos de los Hogares (ENIGH) 2000, 2002, 2004, 2005, 2006 384,168\n\n\n\nNicaragua Encuesta Nacional de Hogares sobre Medición de Nivel de Vida\n(EMNV)\n\n\n\n2001, 2005 59,424\n\n\n\nPanama Encuesta de Hogares (EH) 2001, 2002, 2003, 2004\n\nMention: \"Encuesta Nacional de Fuerza de Trabajo\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its 184,611 observations over these years enter the pooled household sample.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004438:26:0", "occurrence_id": "df7c42204af8cc80457ec798", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **TABLE 1. Description of Household Surveys**\n\n\nCountry Name of Survey Survey years Obs.\n\n\nArgentina Encuesta Permanente de Hogares (EPH‐C) semestre II 2004, 2005 167,770\nBrazil Pesquisa Nacional por Amostra de Domicilios (PNAD) 2002, 2003, 2004 1,169,598\nChile Encuesta de Caracterización Socioeconómica Nacional (CASEN) 2000, 2003 509,825\n\n\nColombia Encuesta Continua de Hogares (ECH) 2001, 2003, 2004 315,528\nCosta Rica Encuesta de Hogares de Propósitos Múltiples (HPM) 2001, 2002, 2003, 2004 173,403\nDominican Rep. Encuesta Nacional de Fuerza de Trabajo (ENFT) onda Octubre 2000, 2001, 2002, 2003, 2004, 2005, 2006 184,611\n\n\nEcuador Encuesta de Empleo, Desempleo y Subempleo (ENEMDU) 2003, 2004, 2005 242,410\nEl Salvador Encuesta de Hogares de Propósitos Múltiples (EHPM) 2000, 2001, 2002, 2003, 2004, 2005 409,093\nGuatemala Encuesta Nacional de Empleo e Ingresos (ENEI) 2002, 2003, 2004 91,343\nHonduras Encuesta Permanente de Hogares de Propósitos Múltiples (EPHPM) 2001, 2004, 2005, 2006 206,868\n\n\nMexico Encuesta Nacional de Ingresos y Gastos de los Hogares (ENIGH) 2000, 2002, 2004, 2005, 2006 384,168\n\n\n\nNicaragua Encuesta Nacional de Hogares sobre Medición de Nivel de Vida\n(EMNV)\n\n\n\n2001, 2005 59,424\n\n\n\nPanama Encuesta de Hogares (EH) 2001, 2002, 2003, 2004\n\nMention: \"Encuesta de Empleo, Desempleo y Subempleo\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its 242,410 observations over these years enter the pooled household sample.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004438:26:0", "occurrence_id": "aca548f079c157c924468aa3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **TABLE 1. Description of Household Surveys**\n\n\nCountry Name of Survey Survey years Obs.\n\n\nArgentina Encuesta Permanente de Hogares (EPH‐C) semestre II 2004, 2005 167,770\nBrazil Pesquisa Nacional por Amostra de Domicilios (PNAD) 2002, 2003, 2004 1,169,598\nChile Encuesta de Caracterización Socioeconómica Nacional (CASEN) 2000, 2003 509,825\n\n\nColombia Encuesta Continua de Hogares (ECH) 2001, 2003, 2004 315,528\nCosta Rica Encuesta de Hogares de Propósitos Múltiples (HPM) 2001, 2002, 2003, 2004 173,403\nDominican Rep. Encuesta Nacional de Fuerza de Trabajo (ENFT) onda Octubre 2000, 2001, 2002, 2003, 2004, 2005, 2006 184,611\n\n\nEcuador Encuesta de Empleo, Desempleo y Subempleo (ENEMDU) 2003, 2004, 2005 242,410\nEl Salvador Encuesta de Hogares de Propósitos Múltiples (EHPM) 2000, 2001, 2002, 2003, 2004, 2005 409,093\nGuatemala Encuesta Nacional de Empleo e Ingresos (ENEI) 2002, 2003, 2004 91,343\nHonduras Encuesta Permanente de Hogares de Propósitos Múltiples (EPHPM) 2001, 2004, 2005, 2006 206,868\n\n\nMexico Encuesta Nacional de Ingresos y Gastos de los Hogares (ENIGH) 2000, 2002, 2004, 2005, 2006 384,168\n\n\n\nNicaragua Encuesta Nacional de Hogares sobre Medición de Nivel de Vida\n(EMNV)\n\n\n\n2001, 2005 59,424\n\n\n\nPanama Encuesta de Hogares (EH) 2001, 2002, 2003, 2004\n\nMention: \"Encuesta Nacional de Empleo e Ingresos\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its 91,343 observations over these years enter the pooled household sample.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004438:26:0", "occurrence_id": "2c6a99cbc4dd0ef94df4e1e0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **TABLE 1. Description of Household Surveys**\n\n\nCountry Name of Survey Survey years Obs.\n\n\nArgentina Encuesta Permanente de Hogares (EPH‐C) semestre II 2004, 2005 167,770\nBrazil Pesquisa Nacional por Amostra de Domicilios (PNAD) 2002, 2003, 2004 1,169,598\nChile Encuesta de Caracterización Socioeconómica Nacional (CASEN) 2000, 2003 509,825\n\n\nColombia Encuesta Continua de Hogares (ECH) 2001, 2003, 2004 315,528\nCosta Rica Encuesta de Hogares de Propósitos Múltiples (HPM) 2001, 2002, 2003, 2004 173,403\nDominican Rep. Encuesta Nacional de Fuerza de Trabajo (ENFT) onda Octubre 2000, 2001, 2002, 2003, 2004, 2005, 2006 184,611\n\n\nEcuador Encuesta de Empleo, Desempleo y Subempleo (ENEMDU) 2003, 2004, 2005 242,410\nEl Salvador Encuesta de Hogares de Propósitos Múltiples (EHPM) 2000, 2001, 2002, 2003, 2004, 2005 409,093\nGuatemala Encuesta Nacional de Empleo e Ingresos (ENEI) 2002, 2003, 2004 91,343\nHonduras Encuesta Permanente de Hogares de Propósitos Múltiples (EPHPM) 2001, 2004, 2005, 2006 206,868\n\n\nMexico Encuesta Nacional de Ingresos y Gastos de los Hogares (ENIGH) 2000, 2002, 2004, 2005, 2006 384,168\n\n\n\nNicaragua Encuesta Nacional de Hogares sobre Medición de Nivel de Vida\n(EMNV)\n\n\n\n2001, 2005 59,424\n\n\n\nPanama Encuesta de Hogares (EH) 2001, 2002, 2003, 2004\n\nMention: \"Encuesta Permanente de Hogares de Propósitos Múltiples\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its 206,868 observations over these years enter the pooled household sample.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004438:26:0", "occurrence_id": "5f0bcb399ffbe0d5fc499549", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **TABLE 1. Description of Household Surveys**\n\n\nCountry Name of Survey Survey years Obs.\n\n\nArgentina Encuesta Permanente de Hogares (EPH‐C) semestre II 2004, 2005 167,770\nBrazil Pesquisa Nacional por Amostra de Domicilios (PNAD) 2002, 2003, 2004 1,169,598\nChile Encuesta de Caracterización Socioeconómica Nacional (CASEN) 2000, 2003 509,825\n\n\nColombia Encuesta Continua de Hogares (ECH) 2001, 2003, 2004 315,528\nCosta Rica Encuesta de Hogares de Propósitos Múltiples (HPM) 2001, 2002, 2003, 2004 173,403\nDominican Rep. Encuesta Nacional de Fuerza de Trabajo (ENFT) onda Octubre 2000, 2001, 2002, 2003, 2004, 2005, 2006 184,611\n\n\nEcuador Encuesta de Empleo, Desempleo y Subempleo (ENEMDU) 2003, 2004, 2005 242,410\nEl Salvador Encuesta de Hogares de Propósitos Múltiples (EHPM) 2000, 2001, 2002, 2003, 2004, 2005 409,093\nGuatemala Encuesta Nacional de Empleo e Ingresos (ENEI) 2002, 2003, 2004 91,343\nHonduras Encuesta Permanente de Hogares de Propósitos Múltiples (EPHPM) 2001, 2004, 2005, 2006 206,868\n\n\nMexico Encuesta Nacional de Ingresos y Gastos de los Hogares (ENIGH) 2000, 2002, 2004, 2005, 2006 384,168\n\n\n\nNicaragua Encuesta Nacional de Hogares sobre Medición de Nivel de Vida\n(EMNV)\n\n\n\n2001, 2005 59,424\n\n\n\nPanama Encuesta de Hogares (EH) 2001, 2002, 2003, 2004\n\nMention: \"Encuesta de Hogares\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its the pooled sample observations over these years enter the pooled household sample.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004439:4:1", "occurrence_id": "54def8521a8df277db7c1196", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  were $46 (Figure Two).\n\n\n5 Kenny, 2009.\n6 This for a two-lane road between 6-8 meters wide with a bituminous surface, for countries where we have\nfour or more estimates based on individual project data Data from the ROCKs database\n<u>http://www.worldbank.org/transport/roads/rd_tools/rocks_main.htm</u>\n<u>7 The results suggest that an increase in the CPI (reflecting reduced corruption) from Pakistan’s score to</u>\nGhana’s would be associated with a drop in Pakistan’s road construction prices of around 20 percent. The\nequation is Cost = 51.9 – 5.9*(CPI), with the constant and CPI entering at one percent, R=0.23, N=28.\nIncluding GDP per capita reduces the coefficient on CPI to 4.5 and the significance to ten percent (GDP per\ncapita does not enter significantly). For a range of World Bank financed infrastructure goods and works\ncontracts, Kenny and Musatova (2009) find that aid dependency and weak institutions lead to fewer bidders\nand a higher risk of price escalations.\n\n3\n\nMention: \"the ROCKs database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Project-level road cost estimates are pooled in this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004442:3:1", "occurrence_id": "50fe413de96e770038b8f850", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  dynamic model of the welfare\neffects of real exchange rate undervaluation and to assess the desirability of reserve\naccumulation and real exchange rate undervaluation as a second-best instrument to\ninternalize such learning-by-doing externalities. <sup>5</sup>\n\nOur setup is based on the notion that reserve accumulation and undervaluation\nsolve a targeting problem: In a first-best world with a full set of instruments, government would like to subsidize investment to induce agents to internalize their learningby-doing externalities. However, if policymakers face difficulties in targeting productive investment opportunities, an alternative mechanism is needed. By lending to for\n\n1China, for example, was sitting atop of USD 2.4 trillion of official foreign reserves by early 2010\n\n- a whopping 49% of its GDP, and experienced growth of more than 9% on average over the past\ndecade (data from the People’s Bank of China). Its performance on both measures was followed\nclosely by other Asian tiger economies such as Taiwan and South Korea.\n\n2See e.g. Aizenman and Marion (2003), Durdu et al. (2009), Mendoza et al. (2009) or Carroll\nand Jeanne (2009) for proponents of this view.\n\n3This is discussed e.g. in Jeanne and Ranci`ere (2009). However, see Carroll and Jeanne (2009)\nfor a more positive assessment.\n\n4Mercantilism was a widespread view among economic thinkers in Europe during the period of\n1500 - 1750 and is still a frequent argument in the public discourse among non-economists.\n5For the purposes of this paper, we equate the term “reserve accumulation” to “real exchange\nrate undervaluation,” since reserve accumulation in our model of closed capital accounts requires\nthat more tradable goods are exported, which makes tradable goods in the domestic economy scarcer\nand depreciates the real exchange rate.\n\n\n2\n\nMention: \"the People’s Bank of China\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Credit relative to output and growth over the decade come from this central bank.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004455:20:0", "occurrence_id": "dd52f00e024901bc6f4671f0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: OECD) (OECD 2003: Table 2.1a; OECD 2007: Table 6.1a). That poses interesting questions\nabout what caused the “shift” of the student score distribution. While extension of compulsory\ncomprehensive education can explain higher performance for low-achievers who were mostly in\nvocational tracks, it is more complicated to explain higher performance of top-achievers. The\nquestion is whether the introduction of lower secondary schools could have an impact on\nstudents in former general secondary schools and what was in the reform that increased scores so\nsignificantly.\n\n\nFigure 4: Change in reading literacy distribution between PISA 2000 and 2006\n\n\nFigure 5: Change in reading literacy distribution between PISA 2003 and 2006\n\n\n\n\n\n\n\n<u>Estimates of score change for students in different tracks</u>\n\nResults for differences-in-differences propensity score matching estimates of the effect of\nabolishing tracking system for 15-year-olds are presented in Tables 3, 4 and 5. Table 3 contains\nestimates of factual and counterfactual mean scores for all students in PISA 2000, 2003 and\n2006. In addition, results for students in vocational and non-vocational tracks are presented.\nFactual scores were weighted by survey weights provided in the official PISA datasets.\nCounterfactual scores were constructed using matching methods with survey weights taken into\naccount in the manner described above.\n\n\n18\n\nMention: \"PISA datasets\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Official survey weights from these datasets weight the factual and counterfactual scores.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004460:12:0", "occurrence_id": "5bf7d94986cffe6bc08a1938", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Once an applicant is selected in the random ballot, they must provide a valid job offer in\n\n\nNew Zealand (unskilled jobs suffice) within six months in order to have their application to\n\n\nimmigrate approved. These job offers are increasingly arranged by large employers visiting\n\n\nSamoa after the annual ballot results are announced. Once a job offer is filed along with a\n\n\nresidence application, it typically takes three to nine months for an applicant to receive a decision\n\n\nand they are then given up to one year to move. If they are successful, their immediate family\n\n\n(spouse and dependent children up to age 24) can also move to New Zealand with them. This\n\n\nrule specifying which family members can and cannot accompany the successful migrant,\n\n\ncoupled with the random selection amongst Samoan Quota applicants, is key to being able to\n\n\novercome the selectivity issues raised in the previous section.\n\n\nConceptually, we can estimate the impact of migration on family members left behind by\n\n\ncomparing outcomes for the group of households in Samoa that sent SQ emigrants to those for\n\n\nthe group with unsuccessful ballots who would not be eligible to move their entire household to\n\n\nNew Zealand had their Principal Applicant been chosen in the ballot. We next discuss the data\n\n\nthat has been collected to allow us to implement this estimation approach.\n\n\n**3.3 The Samoa Labour Mobility Survey**\n\n\nThe data used in this paper are from the Samoa Labour Mobility Survey (SLMS) which\n\n\nwas designed by the authors and implemented by the Samoa Bureau of Statistics in late 2008.\n\n\nThe survey is based on a self-weighting sample of 622 households in 90 villages, drawn from all\n\n\nregions of Samoa. Out of these surveyed households, 78 had current or former members that\n\nwere successful applicants to the Samoan Quota in the previous five years. <sup>6</sup> A further 121\n\n\nhouseholds contained individuals that had applied at least once to the Samoan Quota in the\n\n\nprevious five years and had never been successful. The remaining 423 households in the sample\n\n\ndid\n\nMention: \"the Samoa Labour Mobility Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Designed by the authors and fielded by the national statistics office.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004461:12:0", "occurrence_id": "8e9de7800e79431a8a9a0879", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Among the 204 firms suffering tsunami damage, 168 (82 percent) repaired or\n\n\nreplaced some damaged enterprise assets within three months of the tsunami. On average,\n\n\nowners suffering damage reported spending US$111 on enterprise recovery in the three\n\n\nmonths following the tsunami. Excluding those spending nothing in this period, the\n\n\naverage expenditure was $135. While a substantial sum relative to pre-tsunami income\n\n\nlevels, the amount represents less than 15 percent of the assets lost or damaged by the\n\n\ntsunami. The largest initial effort went to equipment and working capital: an average of\n\n\nUS$39 was spent on equipment, representing 25 percent of the losses in this category.\n\n\nRelative to the losses suffered, the smallest investment was in land and buildings, where\n\n\nowners reported spending US$29 compared with losses of US$362. <sup>9</sup> [^9: The data on assets lost in this paragraph differ slightly from those reported for microenterprises in Table\n1, because the data here come from the baseline survey while those in Table one come from the October]\n\n\nWhat were the sources of funds which enterprises used to replace lost or damaged\n\n\nequipment? On average, just over half of the funds spent in the first three months came\n\n\nfrom own savings (51 percent). An additional 15 percent were obtained through loans\n\n\nfrom family members (9 percent) or friends (6 percent). One-fifth of the resources (20\n\n\npercent) came from grants or loans from tsunami relief agencies. The remaining 14\n\n\npercent was spread among credit from suppliers (6 percent), loans from microfinance\n\n\norganizations (2 percent), moneylenders (2 percent), banks (less than 1 percent),\n\n\nremittances from relatives abroad (1 percent) and other sources. The largest spenders\n\n\nrelied more heavily on loans from family members and credit from suppliers, while those\n\n\nspending the least relied more on own savings. As a percentage of the total amount spent\n\n\nby all 176 entrepreneurs with complete data, own savings represents only 36 percent of\n\n\nexpenditures\n\nMention: \"baseline survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Funding sources for firms come from this round rather than the later one.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004464:9:1", "occurrence_id": "ec2c03f32079b8cd65a69769", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  of\ncommunity literacy rates, and the primary source of water in the community is a\nstatistically significant determinant of the average level of access to an improved water\nsource.\n\nCommunities with higher levels of literacy, higher rates of subsidized medical care,\nand better access to water and sanitation are less likely to produce migrants. In all\nspecifications, the literacy and health variables are significant and correctly signed,\nindicating that lack of access to these services increases the likelihood of migration. The\naverage level of education in a community is also a potentially important determinant of\nmigration, but community education and literacy are highly collinear in the GLSS data\nand only one variable is a significant determinant of migration when both are present.\nAccess to improved drinking water sources and improved sanitation facilities in the\norigin communities is correctly signed but insignificant, but only because availability of\nthese services is highly correlated with community literacy and health outturns. If the\nlatter variables are removed from the regression (column 3 of Table 2), both the water\nand sanitation variables become significant. Interestingly, once these community-level\ncharacteristics are controlled for, the rural dummy is no longer significant in explaining\nthe likelihood of migration. Another potentially important community variable for\ndetermining migration—distance to nearest market—was not significant in any of the\nspecifications, but this could be because this variable was only available for rural\ncommunities.\n\n\n11 Literacy is defined as one’s ability to read a sentence in either English or at least one of the Ghanaian\nlanguages. Health insurance coverage is defined as 1 if an employer, the government, or a health insurer\npays the greatest portion of a patient’s medical expenses, and 0 otherwise. Access to improved water is\ndefined as having access to pipe-borne water. Access to improved sanitation is defined as access to a flush\ntoilet or KVIP (Ghana Water Directorate 2008).\n\n\n8\n\nMention: \"the GLSS data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Community education and literacy are collinear in these data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004482:22:0", "occurrence_id": "5900549b46bef65e4aa394b0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: yields biased estimates. <sup>23</sup> In the estimation, we instrument for the lagged growth rate, ∆ _xciht−_ 1,\n\n\nwith the second lag of the log level of exports, _ln_ ( _xciht−_ 2) if _xciht−_ 2 _>_ 1 and a value of zero if the\n\nsecond lag of the level of exports is less than 1. <sup>24</sup>\n\n\nBy utilizing 6-digit HS product fixed effects in (5) we control for changes in production costs or\n\n\ntechnology that imply that a particular good _h_ will have a growth rate for exports that is higher\n\n\nor lower than average. Note that commodities with very high average growth rates also tend to be\n\n\nthose most likely to be targeted for antidumping measures. As in equations (2) and (3) we use year\n\n\ndummies to control for all aggregate variation in China and country _i_ over time.\n\n\nFor estimating equation (5), we calculate annual export growth of China’s exports to 38 different\n\n\ncountries _i_ listed in table 1, excluding the U.S., EU and India.\n\n\n**5.2** **Instrumental** **variables** **estimates** **of** **trade** **deflection** **and** **trade** **depres-**\n\n\n**sion**\n\n\nTable 10 presents our estimates of trade deflection and trade depression from a panel of Chinese\n\n\nexports to 38 countries. Our finding of a chilling effect of U.S. antidumping duties from the difference\n\nin-difference equation (2) discussed in section 4.1 appears to be robust across models. Although we\n\n\nfind no evidence of chilling in specification (14) which regresses the growth of the volume of Chinese\n\n\ntrade on\n\nMention: \"panel of Chinese\n\n\nexports\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Deflection and depression across thirty-eight destinations are estimated on this panel.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004446:7:1", "occurrence_id": "5ee305cd471a91849660f83b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  becomes important to identify which those reforms are. Yet\n\n\nwe find that quite small changes in the empirical proxies for outcomes that we consider lead to wild\n\n\nfluctuations in the set of variables that are identified as important for those outcomes.\n\n\nThe rest of this paper proceeds as follows. In Section 2 we provide details on the Global\n\n\nIntegrity and Doing Business datasets that we use in our empirical analysis, and we justify our selection\n\n\nof outcome variables corresponding to these datasets. In Section 3 we explain the Bayesian Model\n\n\nAveraging methodology. Section 4 contains the results and Section 5 discusses the robustness of the\n\n\nresults and caveats. Section 6 offers conclusions.\n\n\n6\n\nMention: \"the Global\n\n\nIntegrity and Doing Business datasets\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Both datasets supply the empirical measures and justify the outcome selection.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004483:6:0", "occurrence_id": "629e0304dd1e8d8bbfbd1958", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: than 2 million beneficiaries and, if the rural pensions program is included, the number of\n\n\nbeneficiaries exceeds 8 million. <sup>6</sup> [^6: We are grateful to Rafael Rofman for sharing with us his data on coverage.] Even with high rate of coverage poverty will only be\n\n\neradicated if benefits are high enough and the family structure is not too burdensome. See\n\n\nTable 1.\n\n\n\n**<u>Table 1.  Pension Policies in Latin American Countries</u>**\n**Country** **Type of** **Year of** **Coverage of** **Coverage**\n**pension** **pension** **the** **of**\n\n\n\n**Coverage**\n**of elderly**\n\n\n\n**Coverage of**\n\n**the**\n**economically**\n\n\n\n**Coverage**\n\n**of**\n**employed**\n\n\n\n**system**\n\n\n\n**Year of**\n**pension**\n\n**reform**\n\n\n\n**active**\n**<u>population</u>**\n\n\n\n**persons**\n\n\n\n**<mark>Argentina</mark>** <mark>mixed</mark> <mark>1994</mark> <mark>40.9</mark> <mark>44.6</mark> <mark>70.5</mark>\n**Bolivia** private 1997 12.5 13.1 89.8\n**<mark>Brazil</mark>** <mark>public</mark> <mark>-</mark> <mark>48.1</mark> <mark>51.7</mark> <mark>85.3</mark>\n**Chile** private 1981 62.7 67.3 75.5\n**<mark\n\nMention: \"Rafael Rofman\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Coverage data for social programme beneficiaries were shared by this source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004484:34:0", "occurrence_id": "3a5e125a929257be0e104dab", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ###### **Appendices**\n\n**Appendix 1: Exports by Region (as Percentage of GDP), 1970-2007**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nSource: Own illustration. Data: World Bank, World Development Indicators, National Accounts (1970-2000).\n\n*2006 figure.\n\n\n33\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Regional aggregates over this span are drawn from these national accounts.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004489:3:1", "occurrence_id": "744b7288594e03754330c0b7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: - political, fiscal and administrative into composite indexes for a worldwide sample of 182 countries (see Ivanyna and Shah, 2010,\nfor details on the underlying data and their aggregation).\n\nThe paper further utilizes improved measures on the incidence of corruption.\nWe use three different datasets for this purpose, two of which have not been used\nbefore in similar studies. As a base measure of corruption we use Transparency\nInternational’s Corruption Perception Index, which is an aggregate index based on\nthe opinion of country experts. Further, with the help of Global Corruption Barometer Survey, which is also conducted by the Transparency International, we measure\ncorruption - its frequency and amount - at the household level for 71 countries with\nmore than 55 thousands households covered. Our third dataset is the World Bank\nEnterprises Survey, which covers 99 countries, and provides micro-level data on corruption - both frequency and amount of bribery - for more than 80,000 firms in\ndifferent industries. Therefore, our research on corruption and decentralization extends both to micro- and macro-level data, and encompasses both households and\nbusinesses.\n\nThe paper is organized as follows. Section 2 presents a brief review of conceptual\nand empirical literature. The dataset on decentralization is described in Section 3,\nand various measures of corruption, which we use in this paper, are presented in\nSection 7. This is followed by a discussion on empirical model and variables in\n\n\n2\n\nMention: \"Corruption Perception Index\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The base corruption measure, aggregated from expert opinion.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004489:3:1", "occurrence_id": "c95bf3af4e73705d81003913", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: - political, fiscal and administrative into composite indexes for a worldwide sample of 182 countries (see Ivanyna and Shah, 2010,\nfor details on the underlying data and their aggregation).\n\nThe paper further utilizes improved measures on the incidence of corruption.\nWe use three different datasets for this purpose, two of which have not been used\nbefore in similar studies. As a base measure of corruption we use Transparency\nInternational’s Corruption Perception Index, which is an aggregate index based on\nthe opinion of country experts. Further, with the help of Global Corruption Barometer Survey, which is also conducted by the Transparency International, we measure\ncorruption - its frequency and amount - at the household level for 71 countries with\nmore than 55 thousands households covered. Our third dataset is the World Bank\nEnterprises Survey, which covers 99 countries, and provides micro-level data on corruption - both frequency and amount of bribery - for more than 80,000 firms in\ndifferent industries. Therefore, our research on corruption and decentralization extends both to micro- and macro-level data, and encompasses both households and\nbusinesses.\n\nThe paper is organized as follows. Section 2 presents a brief review of conceptual\nand empirical literature. The dataset on decentralization is described in Section 3,\nand various measures of corruption, which we use in this paper, are presented in\nSection 7. This is followed by a discussion on empirical model and variables in\n\n\n2\n\nMention: \"Global Corruption Barometer Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"It measures the frequency and magnitude of corruption experiences.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004489:3:1", "occurrence_id": "065a04560cf7cd017c3d2967", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: - political, fiscal and administrative into composite indexes for a worldwide sample of 182 countries (see Ivanyna and Shah, 2010,\nfor details on the underlying data and their aggregation).\n\nThe paper further utilizes improved measures on the incidence of corruption.\nWe use three different datasets for this purpose, two of which have not been used\nbefore in similar studies. As a base measure of corruption we use Transparency\nInternational’s Corruption Perception Index, which is an aggregate index based on\nthe opinion of country experts. Further, with the help of Global Corruption Barometer Survey, which is also conducted by the Transparency International, we measure\ncorruption - its frequency and amount - at the household level for 71 countries with\nmore than 55 thousands households covered. Our third dataset is the World Bank\nEnterprises Survey, which covers 99 countries, and provides micro-level data on corruption - both frequency and amount of bribery - for more than 80,000 firms in\ndifferent industries. Therefore, our research on corruption and decentralization extends both to micro- and macro-level data, and encompasses both households and\nbusinesses.\n\nThe paper is organized as follows. Section 2 presents a brief review of conceptual\nand empirical literature. The dataset on decentralization is described in Section 3,\nand various measures of corruption, which we use in this paper, are presented in\nSection 7. This is followed by a discussion on empirical model and variables in\n\n\n2\n\nMention: \"World Bank\nEnterprises Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Micro-level corruption frequency and amounts across ninety-nine countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004489:3:1", "occurrence_id": "ef6f7474cdcfef34d407b8fd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: - political, fiscal and administrative into composite indexes for a worldwide sample of 182 countries (see Ivanyna and Shah, 2010,\nfor details on the underlying data and their aggregation).\n\nThe paper further utilizes improved measures on the incidence of corruption.\nWe use three different datasets for this purpose, two of which have not been used\nbefore in similar studies. As a base measure of corruption we use Transparency\nInternational’s Corruption Perception Index, which is an aggregate index based on\nthe opinion of country experts. Further, with the help of Global Corruption Barometer Survey, which is also conducted by the Transparency International, we measure\ncorruption - its frequency and amount - at the household level for 71 countries with\nmore than 55 thousands households covered. Our third dataset is the World Bank\nEnterprises Survey, which covers 99 countries, and provides micro-level data on corruption - both frequency and amount of bribery - for more than 80,000 firms in\ndifferent industries. Therefore, our research on corruption and decentralization extends both to micro- and macro-level data, and encompasses both households and\nbusinesses.\n\nThe paper is organized as follows. Section 2 presents a brief review of conceptual\nand empirical literature. The dataset on decentralization is described in Section 3,\nand various measures of corruption, which we use in this paper, are presented in\nSection 7. This is followed by a discussion on empirical model and variables in\n\n\n2\n\nMention: \"dataset on decentralization\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The paper describes this dataset as its own construction.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004490:8:0", "occurrence_id": "1e53fea05f83641784af3851", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The 16 industries consist of 6 R&D-intensive industries and 10 low R&D\n\nintensity industries. <sup>3</sup> The R&D intensity of the 16 industries is based on US data. An\n\n\nindustry’s R&D intensity was calculated as R&D expenditures divided by the value\n\n\nadded of that industry. They are grouped in low (10 industries) and high (6 industries)\n\n\nR&D intensity industries. <sup>4</sup>\n\n\nThe TFP index is calculated as the difference between value added and factor\n\n\nincome, with inputs weighted by their income shares, i.e.,\n\n\n_lnTFP_  _lnY_  _ln_ _L_  _(_ 1  _)ln_ _K_, with  equal to the year-country-sector specific\n\n\nlabor’s share. Labor share equal to wages over value added. Value added, wages, and\n\n\nfixed capital formation are reported in current US dollars, and all are deflated using US\n\n\nGDP deflators (1990=100). The capital stocks are derived from the deflated fixed capital\n\n\nformation series using the perpetual inventory model with a 5% depreciation rate.\n\n\nThe R&D flow data are taken from the ANBERD 2000 (OECD) database\n\n\n(DSTI/EAS Division). The database covers 15 OECD countries from 1973 to 2002 at\n\n\neither the two-, three- or four-digit level. <sup>5</sup> R&D flows cover all intramural business\n\n\nenterprise expenditures, converted into US dollars, and then deflated using US GDP\n\n\n3 The 10 low R&D-intensity industries are: (1) 31-Food, Beverage & Tobacco; (2) 32-Textiles, Apparel &\nLeather; (3) 33-Wood Products & Furniture; (4) 34-Paper, Paper Products &\n\nMention: \"ANBERD 2000 (OECD) database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Research flows for fifteen countries over this span come from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004491:17:0", "occurrence_id": "e08060f201ba99a985d7a137", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: potential new trade barriers to other complementary databases (e.g., on trade flows) to\n\n\ndetermine how economically important any given policy action would be.\n\n\nSuch an approach was also adopted by a new initiative launched in June 2009\n\n\ncalled the _Global Trade Alert_ . <sup>16</sup> The Global Trade Alert (GTA) was designed to go\n\n\nbeyond the trade remedy data provided in the Global Antidumping Database to catalogue\n\n\nand detail dozens of other types of trade and trade-related measures – e.g., changes in\n\n\ntariffs, import licensing, government procurement, imposition of subsidies, and changes\n\n\nin technical and sanitary or phytosanitary standards among others. Many interpreted the\n\n\nGTA’s contributions throughout the global economic crisis as being intended to ‘name\n\n\nand shame’ government policymakers by raising the public profile of trade policy\n\n\nactions. Perhaps government officials would think twice about how they structured new\n\n\npolicy initiatives if they were concerned that the GTA would flag their policies with a red\n\n\ntriangle that indicated ‘the measure has been implemented and almost certainly\n\n\ndiscriminates against foreign commercial interests.’ Nevertheless, the publicity around\n\n\nthe GTA has resulted in critics of its own. Some focused on the utility of the information\n\n\nbeing provided for understanding the scale of the spread of protectionism taking hold\n\n\nduring the global crisis, given the lack of comparable pre-crisis data available to use as a\n\n\nbenchmark. Others argued that the GTA was overstating the incidence of protectionism\n\n\nwhich could provide fodder and cover for other policymakers interested in increasing\n\n\n16 For a discussion see Evenett (2009). According to the Global Trade Alert website, it is coordinated by the\n\n\nCentre for Economic Policy Research (London), and its funders include The World Bank, UK’s Trade\n\n\nPolicy Unit (combining the Department for International Development and the Department for Business,\n\n\nInnovation and Skills), the German Marshall Fund of the United States, Canada’s Centre for International\n\n\nGovernance Innovation, and Canada’s International Development Research\n\nMention: \"Global Antidumping Database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"The newer monitoring system was designed to go beyond this remedy dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004495:38:2", "occurrence_id": "4fdf4d8ff3d119732f4a4083", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |12.47|0.3|14.6|111|791|21|923|\n|Argentina|4.58|9.47|1.7|15.7|366|739|118|1223|\n|Cuba|6.46|8.52|2.6|17.6|218|270|93|581|\n|Brazil|4.92|12.98|1.1|19.0|180|476|35|691|\n||||||||||\n|Source: Social Database ECLAC.|Source: Social Database ECLAC.|Source: Social Database ECLAC.|||||||\n\nMention: \"Social Database ECLAC\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Social spending and coverage figures by country come from this regional database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004496:8:3", "occurrence_id": "c91b06cd7b638bfa96bf903e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  concern corporate taxes. Furthermore and in addition to consistency problems\nassociated with reporting on collection time, government data on revenues and expenditures are\nnot sufficiently disaggregated to allow a fair comparison across countries.\n\nThe empirical analysis and estimation are based on a sample of 14 countries selected across SubSaharan Africa on the basis of data quality. <sup>14</sup> Operational and structural budget deficit are\ncalculated for each country, and the results are expressed as a percentage of GDP to allow crosscountry comparisons. Real GDP per capita, public and private investment variables are taken from\nthe World Bank World Development Indicator database. The data on public finance are taken\nfrom Government Financial Statistics (GFS) and World Economic Outlook (WEO).\n\nTable 1 provides the results averaged over the period 1980-2007 (a long-term trend of the fiscal\nadjustment and growth nexus is also assessed from Figure 1 below). Columns 1 and 2 provide\nestimates of operational and structural balances averaged over the sample period for each country.\nThe following columns (3 and 5) provide a nonparametric measure of correlation assessing the\n\n\n14 This set of sampled countries include: Benin, Botswana, Central African Republic, Cote d’Ivoire, Cameroon,\nEthiopia, Gabon, Gambia, Ghana, Kenya, Mozambique, Nigeria, Rwanda and Senegal.\n\n\n7\n\nMention: \"World Bank World Development Indicator database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Output per head and investment shares are taken from this compilation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004496:8:3", "occurrence_id": "970185cd1e8bbbd23c468b84", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  concern corporate taxes. Furthermore and in addition to consistency problems\nassociated with reporting on collection time, government data on revenues and expenditures are\nnot sufficiently disaggregated to allow a fair comparison across countries.\n\nThe empirical analysis and estimation are based on a sample of 14 countries selected across SubSaharan Africa on the basis of data quality. <sup>14</sup> Operational and structural budget deficit are\ncalculated for each country, and the results are expressed as a percentage of GDP to allow crosscountry comparisons. Real GDP per capita, public and private investment variables are taken from\nthe World Bank World Development Indicator database. The data on public finance are taken\nfrom Government Financial Statistics (GFS) and World Economic Outlook (WEO).\n\nTable 1 provides the results averaged over the period 1980-2007 (a long-term trend of the fiscal\nadjustment and growth nexus is also assessed from Figure 1 below). Columns 1 and 2 provide\nestimates of operational and structural balances averaged over the sample period for each country.\nThe following columns (3 and 5) provide a nonparametric measure of correlation assessing the\n\n\n14 This set of sampled countries include: Benin, Botswana, Central African Republic, Cote d’Ivoire, Cameroon,\nEthiopia, Gabon, Gambia, Ghana, Kenya, Mozambique, Nigeria, Rwanda and Senegal.\n\n\n7\n\nMention: \"Government Financial Statistics\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Public finance series come from this fund compilation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004496:8:3", "occurrence_id": "91fd8ddfb8061d3ad2440659", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  concern corporate taxes. Furthermore and in addition to consistency problems\nassociated with reporting on collection time, government data on revenues and expenditures are\nnot sufficiently disaggregated to allow a fair comparison across countries.\n\nThe empirical analysis and estimation are based on a sample of 14 countries selected across SubSaharan Africa on the basis of data quality. <sup>14</sup> Operational and structural budget deficit are\ncalculated for each country, and the results are expressed as a percentage of GDP to allow crosscountry comparisons. Real GDP per capita, public and private investment variables are taken from\nthe World Bank World Development Indicator database. The data on public finance are taken\nfrom Government Financial Statistics (GFS) and World Economic Outlook (WEO).\n\nTable 1 provides the results averaged over the period 1980-2007 (a long-term trend of the fiscal\nadjustment and growth nexus is also assessed from Figure 1 below). Columns 1 and 2 provide\nestimates of operational and structural balances averaged over the sample period for each country.\nThe following columns (3 and 5) provide a nonparametric measure of correlation assessing the\n\n\n14 This set of sampled countries include: Benin, Botswana, Central African Republic, Cote d’Ivoire, Cameroon,\nEthiopia, Gabon, Gambia, Ghana, Kenya, Mozambique, Nigeria, Rwanda and Senegal.\n\n\n7\n\nMention: \"World Economic Outlook\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The companion fiscal series come from this fund outlook.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004502:18:0", "occurrence_id": "ce1ca72a58121485edc5e626", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: enforcement of taxes, the use of non-monetary payments and growing arrears, laxity in\n\n\nbankruptcy enforcement and a parochial attitude toward public procurement…..” <sup>19</sup>\n\n\n3.2 Capital Inflows\n\n\nNotwithstanding the macro-micro “disconnect”, the money kept flowing in, attracted by\n\n\nRussia’s “asset play”. This sub-section provides only a quick sketch of the capital flow numbers\n\n\nin Table 4 and the main events surrounding them; the underlying economics and a more detailed\n\n\nchronology follow in sub-sections 3.3 to 3.6 below. Box 1 provides an overview of balance-of\n\npayments accounting as a guide to Table 4.\n\n\nThe impact of improving market sentiment combined with liberalized access to the GKO\n\n\nmarket is evident: portfolio flows into the GKO market just in the first quarter of 1997 were more\n\n\nthan thrice the amount for the whole of 1996 as shown in Table 4. Reserves rose rapidly and the\n\n\nstock market boomed as captured by the RTS index reported in the table. And ‘Other capital\n\n\nflows, net’, which includes capital flight as well as purchases of dollars by households, dropped\n\n\nduring the first two quarters of 1997.\n\n\nThere was an equally dramatic reversal in the fourth quarter of 1997 as the Asian crisis\n\n\nspilled over. Inflows into the GKO market, which had already fallen sharply in the third quarter,\n\n\nfell precipitously in the fourth while capital flight resumed and reserves and the stock market both\n\n\nplunged.\n\n\n19 Buckberg and Pinto (1997). In retrospect, the statement about macroeconomic discipline was optimistic.\n\n\n\n17\n\nMention: \"the RTS index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The stock market boom is captured by this exchange index.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004503:22:0", "occurrence_id": "13f77a388bb45628c85429a0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 21\n\n\ncorresponding to the degree to which it is an obstacle to firm performance. <sup>45</sup> [^45: Ayyagari et al. (2005) present the mean reported values for a number of investment climate variables in a sample\nof over 6,000 firms in 80 countries. In the overall sample, taxes and regulation, political instability, inflation, and\nfinancing are reported as being the greatest obstacles to firm growth.] High mean reported\nvalues for particular dimensions of the investment climate are then interpreted as evidence of the\nseverity of obstacles to growth.\n\nHowever, this may not be the case. Perceptions of the degree to which some investment climate\nvariables differ from the actual effect of these variables on firm productivity, business\nperformance, or firm growth. Despite their intimate knowledge of their business processes and\noperating environment, firms may not fully recognize the true origin of their main problems and\nmistakenly identify as a constraint something which is in fact a symptom of another less obvious\nproblem. Because of these shortcomings, investment climate constraints are increasingly\ncomplemented by the World Bank “Doing Business” indicators, which are based on expert\nsurveys (not just firm-level perceptions) and provide a more comparable cross-country\nperspective across a detailed range of regulation.\n\nThe problem remains, as survey results often vary depending on whether respondents are asked\nto rate their most important constraints, or to rank them. While ranking appears to be favored by\nresearchers who have examined different methodologies, as it forces stronger expression and\nrelationships (Alvin and Krosnick 1985), it may not be entirely reliable: firms or experts who are\nasked to rank constraints may not have a good basis for determining whether their top-ranked\nconstraint is serious or not. Ranking without a solid and meaningful benchmark against which\nlocal firms in a given country can rate the severity of a particular constraint may not provide\nuseful information. In addition, there are instances where picking any single quantitative\ncriterion could be misleading, as firms often face several constraints simultaneously. Ranking all\nof them\n\nMention: \"World Bank “Doing Business” indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Expert-survey based measures complement firm perceptions of constraints.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004506:16:0", "occurrence_id": "e50f8faf8a45a2d70cd57327", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 14\n\n\n**Figure 7. Latin America and East Asia: Gross Domestic Savings (1970-2008)**\n\n(Percent of GDP)\n\n\n\n\n\n_Note: East Asia is the un-weighted average of Hong Kong, Singapore, South Korea, Taiwan,_\n_Indonesia, Thailand and Malaysia, while LAC covers the whole Latin American and Caribbean_\n_region._\n_Source: World Development Indicators (2010)._\n\n\n**Figure 8. Latin American Capital Markets Reform Implementation (1990-2002)**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n_Note: Percentage of Latin American countries having implemented reforms._\n_Source: De la Torre and Schmukler (2007)._\n\nMention: \"World Development Indicators (2010)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Regional comparisons and reform series are drawn from this edition.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004507:53:2", "occurrence_id": "76386fe097ed0e2b9b71ae42", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  rescaled such that it takes\nvalues between 0 and 1, with higher numbers implying higher quality of telecommunication\nservices.\n\n\n\nElectricity Generating Capacity (in MW per 1000 workers). The variable was expressed in\nlogs.\n\n\n\nAuthors' construction using Canning (1999), United Nation's\nEnergy Statistical Yearbook, and national sources where available.\n\n\n\nQuality of Electricity Services Electric Power Transmission and Distribution Losses (as percentage of electricity output). The\nvariable was rescaled such that it takes values between 0 and 1, with higher numbers\nimplying higher quality of telecommunication services.\n\n\nRoad Network (a) Length of total road network, and (b) Length of paved road network. Both variables are\nmeasured in kilometers per sq. km. of surface area of the country, and then expressed in logs\n\n\n\nThe World Bank's World Development Indicators and national\nsources where available.\n\n\n\nRoad Network (a) Length of total road network, and (b) Length of paved road network. Both variables are International Road Federation's World Road Statistics, the World\nmeasured in kilometers per sq. km. of surface area of the country, and then expressed in logs Bank's World Development Indicators, and national sources where\n\navailable.\nQuality of the Road Network Share of paved roads in the overal road network. This variables takes values between 0 and International Road Federation's World Road Statistics, the World\n1, with higher numbers implying higher quality of the road network. Bank's World Development Indicators, and national sources where\navailable.\nSynthetic Index of Infrastructure First principal component of the three different dimensions of infrastructure considered in Author's calculations.\nQuantity our analysis: telecommunications, electricity and roads. IK1 comprises information on main\n\n\n\nAuthor's calculations.\n\n\n\nSynthetic Index of Infrastructure\nQuality\n\n\n\nFirst principal component of the three different dimensions of infrastructure considered in\nour analysis: telecommunications, electricity and roads. IK1 comprises information on main\nlines, EGC, and total roads; IK2 on main lines,\n\nMention: \"Energy Statistical Yearbook\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Electricity service variables are built using this yearbook and national sources.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004507:53:2", "occurrence_id": "9230f2acad0cd393dc660176", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  rescaled such that it takes\nvalues between 0 and 1, with higher numbers implying higher quality of telecommunication\nservices.\n\n\n\nElectricity Generating Capacity (in MW per 1000 workers). The variable was expressed in\nlogs.\n\n\n\nAuthors' construction using Canning (1999), United Nation's\nEnergy Statistical Yearbook, and national sources where available.\n\n\n\nQuality of Electricity Services Electric Power Transmission and Distribution Losses (as percentage of electricity output). The\nvariable was rescaled such that it takes values between 0 and 1, with higher numbers\nimplying higher quality of telecommunication services.\n\n\nRoad Network (a) Length of total road network, and (b) Length of paved road network. Both variables are\nmeasured in kilometers per sq. km. of surface area of the country, and then expressed in logs\n\n\n\nThe World Bank's World Development Indicators and national\nsources where available.\n\n\n\nRoad Network (a) Length of total road network, and (b) Length of paved road network. Both variables are International Road Federation's World Road Statistics, the World\nmeasured in kilometers per sq. km. of surface area of the country, and then expressed in logs Bank's World Development Indicators, and national sources where\n\navailable.\nQuality of the Road Network Share of paved roads in the overal road network. This variables takes values between 0 and International Road Federation's World Road Statistics, the World\n1, with higher numbers implying higher quality of the road network. Bank's World Development Indicators, and national sources where\navailable.\nSynthetic Index of Infrastructure First principal component of the three different dimensions of infrastructure considered in Author's calculations.\nQuantity our analysis: telecommunications, electricity and roads. IK1 comprises information on main\n\n\n\nAuthor's calculations.\n\n\n\nSynthetic Index of Infrastructure\nQuality\n\n\n\nFirst principal component of the three different dimensions of infrastructure considered in\nour analysis: telecommunications, electricity and roads. IK1 comprises information on main\nlines, EGC, and total roads; IK2 on main lines,\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Network density measures are divided by national surface area from this source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004507:53:2", "occurrence_id": "d41b721f0f008732c17908f5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  rescaled such that it takes\nvalues between 0 and 1, with higher numbers implying higher quality of telecommunication\nservices.\n\n\n\nElectricity Generating Capacity (in MW per 1000 workers). The variable was expressed in\nlogs.\n\n\n\nAuthors' construction using Canning (1999), United Nation's\nEnergy Statistical Yearbook, and national sources where available.\n\n\n\nQuality of Electricity Services Electric Power Transmission and Distribution Losses (as percentage of electricity output). The\nvariable was rescaled such that it takes values between 0 and 1, with higher numbers\nimplying higher quality of telecommunication services.\n\n\nRoad Network (a) Length of total road network, and (b) Length of paved road network. Both variables are\nmeasured in kilometers per sq. km. of surface area of the country, and then expressed in logs\n\n\n\nThe World Bank's World Development Indicators and national\nsources where available.\n\n\n\nRoad Network (a) Length of total road network, and (b) Length of paved road network. Both variables are International Road Federation's World Road Statistics, the World\nmeasured in kilometers per sq. km. of surface area of the country, and then expressed in logs Bank's World Development Indicators, and national sources where\n\navailable.\nQuality of the Road Network Share of paved roads in the overal road network. This variables takes values between 0 and International Road Federation's World Road Statistics, the World\n1, with higher numbers implying higher quality of the road network. Bank's World Development Indicators, and national sources where\navailable.\nSynthetic Index of Infrastructure First principal component of the three different dimensions of infrastructure considered in Author's calculations.\nQuantity our analysis: telecommunications, electricity and roads. IK1 comprises information on main\n\n\n\nAuthor's calculations.\n\n\n\nSynthetic Index of Infrastructure\nQuality\n\n\n\nFirst principal component of the three different dimensions of infrastructure considered in\nour analysis: telecommunications, electricity and roads. IK1 comprises information on main\nlines, EGC, and total roads; IK2 on main lines,\n\nMention: \"World Road Statistics\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Total and paved road lengths are taken from this compilation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004512:2:0", "occurrence_id": "66f28dd2db22f26434926495", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ### **Deals versus Rules:** **Policy Implementation Uncertainty and Why Firms Hate It**\n\n##### Mary Hallward-Driemeier Gita Khun-Jush Lant Pritchett *\n\n**Key words:** Uncertainty, Corruption, Enterprise Surveys, Doing Business, Policy\nImplementation\n**JEL codes:** H32, O17, D81, J23, K2\n\n\n- We would like to thank the NBER for support via the Africa Project. We would like to thank Bill\nBattaile, Asli Demirguc-Kunt, Ahmed Galal, Ricardo Hausmann, Marialisa Motta, Stephen O’Connell, Rita\nRamalho, James Robinson, Dani Rodrik, Luis Serven, Andrei Shleifer, Sylvia Solf and David Weil for\nconversations that were, perhaps unbeknownst to them, helpful and to Dan Willey for motivation on this\ntopic **.** Mamadou Bady Balde provided the materials on Rwanda. The views expressed here do not\nnecessarily represent the views of the World Bank or its Executive Directors. Corresponding author: Mary\nHallward-Driemeier, mhallward@worldbank.org.\n\nMention: \"Enterprise Surveys\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Listed among the paper's key words as a data source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004514:3:0", "occurrence_id": "91da03027a36f0607ff4854e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: #### **The Political, Regulatory and Market Failures That** **Caused the US Financial Crisis**\n\n_“From the current handwringing, you’d think that the banks came up with_\n\n\n_the idea of looser underwriting standards on their own, with regulators_\n\n\n_just asleep on the job. In fact, it was the regulators who relaxed these_\n\n\n_standards—at the behest of community groups and ‘progressive political_\n\n\n_forces.’”_\n\n\nProfessor Stan Leibowitz, University of Texas <sup>1</sup>\n\n\n**I. Summary and Literature Review**\n\n\n**Summary**\n\n\nIn 2010, the US is in the worst financial crisis since the Great Depression. The core of the crisis\n\n\nis that 44 percent of all home mortgages (or 25 million mortgages) are default prone, a figure that\n\n\nis unprecedented in US history. <sup>2</sup> Why did financial institutions and homeowners acquire so\n\n\nmany mortgages that are in default or in danger of wider default? I argue that the crisis is a\n\n\nresult of regulatory failure, market failure and, most of all, political failure.\n\n\nFirst, the seeds of the crisis were sown while Congress was appropriately imposing\n\n\ntougher regulation on banks and savings and loan associations in the early 1990s in response to\n\n\nthe Savings and Loan crisis. Congress made a grave error: it agreed to avoid real regulation of\n\n\nthe two Government Sponsored Enterprises (GSEs) commonly known as Fannie Mae and\n\n1 New York Post, February 5, 2008. See Leibowitz (2008) for an elaboration of his views where he argues that “in\nan attempt to increase home ownership, virtually every [relevant] branch of government undertook an attack on\nunderwriting standards.”\n2 Testimony of Ed Pinto (2008, p. 8), former chief credit officer of Fannie Mae. The principal database of the New\nYork Fed under reports default prone mortgages\n\nMention: \"the New\nYork Fed\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its principal database understates mortgages prone to default.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004516:34:1", "occurrence_id": "c835bea0b91c0d43e143f1a1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  for additional background information to\nhelp consumers understand the financial services and products they purchase. In Azerbaijan, the most\ncommon complaint consumers submitted to the central bank was that they did not understand why their\ncommercial banks failed to grant loans to them. In Croatia, consumers complained to banking supervisors\nthat “We don’t understand what we signed”. The Diagnostic Reviews found that consumers are often\nconfused about the terms and conditions of their financial contracts and largely unaware of the full extent\nof the risks inherent in the financial product or service on offer or actually acquired. The EU Consumer\nCredit Directive (now implemented in most EU new member states) provides a clear algebraic expression\nfor the calculation of the annual percentage rate of charge (APR). Yet consumers often do not understand\nhow the final rate is calculated. For example, the Diagnostic Review for Latvia noted that one of the most\ncommon forms of complaint submitted to the consumer protection authority was over calculation of the\ninterest rate on consumer loans. Numerous consumers complained that banks unilaterally increased the\ninterest rate on their mortgage loans. Yet investigation of the terms for those contracts showed that the\ncontract clearly stated that the interest rate was set at a specified margin over a widely quoted reference\nrate, such as LIBOR or the Riga inter-bank offered rate. In such cases, the consumer had clearly not\nunderstood the terms of the contract.\n\n**The complaints office should be able to access programs of financial education.** Where a complaint is\nan inquiry rather than a dispute, the office should provide consumers with access to reliable financial\neducation resources. Responding to consumer inquiries and providing reliable information to deal with\nconsumers’ immediate needs thus provides an excellent forum for consumer education on financial\nmatters.\n\n**Statistics on consumer complaints should be analyzed and published—and used to identify future**\n**improvements in the consumer protection framework.** In none of the countries of the Diagnostic\nReviews were statistics on financial consumer complaints aggregated or analyzed to identify trends.\n\nMention: \"Statistics on consumer complaints\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"A recommendation that these complaint counts be analysed and published.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004517:15:0", "occurrence_id": "26d645cda1414043885af669", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: and ability by frontline providers in health and education via the observation schedules and hours\ntaught and via the teacher and medical tests. We propose to measure effective budget and delays\nin salaries as a proxy for the effort exerted by higher level authorities. Finally, physical resources\nare measured by infrastructure and material indicators. Obviously we do not observe effort or\nability directly so our approach is an indirect way of getting at the key arguments in equation (1).\nMoreover, the indicators, by construction, do not speak to the important question of how to affect\nthese arguments because we cannot directly measure the incentive constraints influencing staff\nbehavior, and even if we could we would not be able to causally link them. In fact, the cause of the\nproblem may not be closely connected to the appropriate solution, since the institutional system\nhas, over time, influenced norms and thus behavior. For instance, it may well be the case that low\npay, or inadequate supply of inputs, has eroded (non-pecuniary) motivation to exert high effort and\ncreated norms that do not reward high effort. Then increasing pay, or increasing supply, may just\nresult in better paid and better supplied workers that still lack motivation to serve the public.\n\n**2.5 Outcomes**\n\nTo avoid making structural assumptions on the link between inputs, performance and outcomes,\nwe do not suggest that outcomes should be part of the index. However, when the various subindicators and the final index are presented, it also makes sense to report, separately, on\noutcomes. In health there are measures for many countries at the national level, such as U5\nmortality rates. Quantity outcomes in education are also available (various measures of flows and\nstock of schooling) for a large subset of countries. However, on quality there is no comparable\n(for many countries at least) data available. Thus, we suggest that student learning achievement\nis collected as part of the survey in education. This in itself is\n\nMention: \"U5\nmortality rates\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Named as an example of nationally available health outcome measures.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004520:16:1", "occurrence_id": "4e51f1bf1258289a0ebd4d64", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \nWe draw on information collected as part of a global automotive supplier benchmarking study of the\nInternational Motor Vehicle Program (IMVP), which aims to compare and contrast practices, capabilities\nand performance of automobile suppliers around the world. The international assessment, focusing on\nseats, exhaust systems and brakes, will cover plants in China, South Korea, Japan, Europe and North\nAmerica. We have already collected data in plants in China, Japan, and a in few European plants. Further\ninformation of this project can be found in Brandt and Van Biesebroeck (2008).\n\nThe identities of the firms we interviewed cannot be revealed, but given that the differences fell largely\nalong national lines we will discuss the two strategies in general by contrasting Asian and Western\nautomakers. The first strategy can be referred to as “cautious localization.” The Japanese and Korean\nproducers interviewed favor this approach. Vehicles are produced in China in large volumes but entirely\ndesigned overseas. Most first-tier suppliers are joint ventures (JVs) between a local Chinese firm and a\nforeign partner that is responsible for manufacture and often the design of the part back in the home\ncountry. Some modules are supplied by wholly owned foreign subsidiaries (WOS), which are allowed in\nChina for parts but not for final assembly. The lead firm has to give explicit approval to use domestic\nfirms, which tend to have a large cost advantage, even as second- or third-tier suppliers. As a result, a\nmajority of second-tier suppliers also tend to be either JVs or WOS.\n\nThis centralized GVC organization facilitates product quality but raises costs since parts cannot be altered\nor easily outsourced to take advantage of lower cost, lower quality manufacturing. In the end, vehicles\nmay cost too much to appeal to a large number of buyers, and it is more difficult to introduce products\nspecifically aimed at the local market. While average income levels in the Chinese economy are rising\nrapidly, relatively wealthy customers have led the\n\nMention: \"the\nInternational Motor Vehicle Program (IMVP)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its supplier benchmarking study compares capabilities and performance.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004524:39:30", "occurrence_id": "0f239ac7e0241777a2e9a26a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: br>0.91<br>_0.85_<br>_0.85_<br>_0.85_|\n|JPN|JPN|JPN|JPN|JPN|JPN|JPN|JPN|JPN|JPN|\n|CAN|CAN|CAN|CAN|CAN|CAN|CAN|CAN|CAN|CAN|\n|USA|USA|USA|USA|USA|USA|USA|USA|USA|USA|\n|EU15|EU15|EU15|EU15|EU15|EU15|EU15|EU15|EU15|EU15|\n|OEUR|OEUR|OEUR|OEUR|OEUR|OEUR|OEUR|OEUR|OEUR|OEUR|\n|RUS|RUS|RUS|RUS|RUS|RUS|RUS|RUS|RUS|RUS|\n\n\n\nSource: Authors' calculations via COMTRADE and GTAP6 Databases\nNotes: PR is the average of each exporter’s unit value divided by the world average export unit value from 2000 to 2004. BIPR equals PR divided by the\n2001 weighted average import price for each importing region. Italicized values ( _0.85_ ) have been truncated for purposes of incorporation into the model,\nfor which this represents trigger point for the P-SSM.\n\n\n38\n\nMention: \"COMTRADE\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Unit values underlying the price ratios are computed from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004524:39:30", "occurrence_id": "a8e32b2dc44b56855b9d16ca", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: br>0.91<br>_0.85_<br>_0.85_<br>_0.85_|\n|JPN|JPN|JPN|JPN|JPN|JPN|JPN|JPN|JPN|JPN|\n|CAN|CAN|CAN|CAN|CAN|CAN|CAN|CAN|CAN|CAN|\n|USA|USA|USA|USA|USA|USA|USA|USA|USA|USA|\n|EU15|EU15|EU15|EU15|EU15|EU15|EU15|EU15|EU15|EU15|\n|OEUR|OEUR|OEUR|OEUR|OEUR|OEUR|OEUR|OEUR|OEUR|OEUR|\n|RUS|RUS|RUS|RUS|RUS|RUS|RUS|RUS|RUS|RUS|\n\n\n\nSource: Authors' calculations via COMTRADE and GTAP6 Databases\nNotes: PR is the average of each exporter’s unit value divided by the world average export unit value from 2000 to 2004. BIPR equals PR divided by the\n2001 weighted average import price for each importing region. Italicized values ( _0.85_ ) have been truncated for purposes of incorporation into the model,\nfor which this represents trigger point for the P-SSM.\n\n\n38\n\nMention: \"GTAP6 Databases\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The companion database used in the same computations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004532:19:0", "occurrence_id": "42e8aa236a7ab495c46d5884", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: why we have not been able to employ such information. Its entries have been extracted\n\n\nfrom country reports by the Asian Disaster Reduction Center. The reports generally\n\n\nfocus on summary information rather than specific information for emergency\n\n\npreparedness by type of disaster (e.g., floods, droughts). In the case of Japan, for\n\n\nexample, much of the $34 billion expenditure is clearly for earthquake-related measures.\n\n\nThe listed funds for Bangladesh are more than twice China’s and four times Indonesia’s,\n\n\nand they include both emergency food assistance and disaster management. Much of the\n\n\nIndonesian fund undoubtedly relates to geologic disasters (earthquakes, volcanic\n\n\neruptions, tsunamis) as well as weather-related disasters.\n\n\nMost reports do not provide time series information, nor do they go beyond reporting for\n\nsingle funds or national-level agencies. <sup>9</sup> [^9: Our thanks to Tim Essam for his help with gathering this information.] Local expenditures are not included.\n\n\nIn summary, the available data are far too spotty, non-specific, non-standardized, and\n\n\ntemporally limited to permit estimation of cost functions that could be used for\n\n\nprojection.  In addition, they cover relatively few countries. There is simply no way to\n\n\nconstruct a reasonable cost analysis from such information.\n\n\n**7.2 The Education Alternative**\n\nSince direct cost measures cannot be derived from the available data, we turn to an\n\n\nindirect approach. As the panel results in Table 6 show, improvements in female\n\n\neducation are powerfully associated with reductions in disaster risks once changes in\n\n\nweather and income are accounted for. In this section, we exploit this relationship to\n\n\naddress the adaptation cost question indirectly.\n\n\nOur approach applies straightforward algebra to equation (1). Given an anticipated\n\n\nchange in precipitation, we calculate the increase in education that will be just sufficient\n\n\nto restore the risk level prior to the precipitation change. With subscripts B for the\n\n\nbaseline case and N for the risk-neutralizing case, we impose the following constraint on\n\n\nthe relevant elements of equation (1)\n\nMention: \"the Asian Disaster Reduction Center\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its country reports supply the entries, though only at summary level.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004533:40:2", "occurrence_id": "e2a58cac9869c62fd9f5a82e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Bloomberg for 2005|\n|Korea|1970.Q1‐2005.Q1|Bank of Korea|\n|Malaysia|1988.Q1‐2004.Q4|IMF’s International Financial Statistics|\n|Mexico|1980.Q1‐2005.Q1|Instituto Nacional de Estadística Geografía e Informática (INEGI,<br>http://www.inegi.gob.mx/)|\n|New Zealand|1982.Q2‐2005.Q1|Statistics New Zealand (SNZ,http://www.stats.govt.nz/)|\n|Peru|1979.Q1‐2005.Q1|Central Reserve Bank of Peru (www.bcrp.gob.pe)|\n|Portugal|1970.Q1‐2005.Q1|IMF’s IFS and Bloomberg for 2005|\n|Singapore|1975.Q1‐2005.Q1|Singapore Department of Statistics (http://www.singstat.gov.sg/)|\n|Spain|1970.Q1‐2005.Q1|IMF’s IFS and Bloomberg for 2005|\n|Sweden|1970.Q1‐2005.Q1|IMF’s IFS and Bloomberg for 2005|\n|Taiwan|1970.Q1‐2005.Q1|National Statistics, Republic of China (http://eng.stat.gov.tw/).|\n|Thailand|1993.Q1‐2005.Q1|National Economic and Social Information Board<br>(http://www.nesdb\n\nMention: \"IMF’s International Financial Statistics\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The country's quarterly series in this table come from this compilation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004535:16:0", "occurrence_id": "819466ba55e3eeaa0566be8c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 15\n\n\nchange. If one is willing to assume that the marginal responses of supply and demand\n\nto a price change are the same in aggregate, then _a_ = _b_ =0.5. <sup>6</sup> [^6: With linear demand and supply curves for a country’s economy, this equates to an assumption that the\naggregate demand and supply curves have the same slope, so that each side of the economy contributes\nequally to the ITRI.]\n\n\nThus from a practical viewpoint, _Bj_ can be computed with all the information\n\n\navailable for calculating NRAs and CTEs (or the PSEs and CSEs generated by the\n\n\nOECD), provided two assumptions are made: (1) equal domestic price elasticities of\n\n\nsupply across products within a country (and the same for domestic price elasticities\n\n\nof demand); and (2) equal responsiveness of aggregate supply and demand to price\n\n\nchanges for the set of products of concern for an economy. Ideally policy analysts\n\n\nwould incorporate elasticity estimates and information on responsiveness of aggregate\n\n\nsupply and demand where available. However, where they are not available, estimates\n\n\nof the indexes _Bj_, _RBj_ and _SBj_ are nonetheless superior to existing widely-used\n\n\nagricultural policy measures of trade distortions. Analogous assumptions can be made\n\n\nfor the domestic measures derived in this paper.\n\n\n**The Distortions to Agricultural Incentives database**\n\n\nA new database (Anderson and Valenzuela 2008), generated by the World Bank’s\n\n\nDistortions to Agricultural Incentives research project using a methodology\n\n\nsummarized in Anderson et al. (2008), provides a timely opportunity to estimate\n\n\nindexes of the trade- and welfare-reducing effects of different policy instruments.\n\n\nThe database includes estimates of different agricultural policy instruments for 75\n\n\ncountries that together account for over 90 percent of the world’s population,\n\n\nfarmers, agricultural GDP and total GDP. The estimates in the database are\n\n\nconsistent estimates of annual NRAs and CTEs over the years 1955 to 2007. The\n\n\ncountry coverage is most complete for the years 1960 to 2004 so only that period is\n\n\nreported in this manuscript\n\nMention: \"Distortions to Agricultural Incentives database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its new compilation supplies the domestic policy measures derived here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004536:3:2", "occurrence_id": "6e0309a17c2b935b893b4fba", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  in all sectors.\n5 Information is missing for various reasons. A sizeable fraction of the respondents refuse, forget or fail to answer\nsome questions. In other cases, even well-trained interviewers may neglect to ask some questions. Sometimes\nrespondents just say they do not have the information available to them or they do not know the answer to the\nquestion. Some questions are simply not applicable to some respondents (see Allison, 2001). All of these cases may\nbe applicable to IC data.\n6 The number of observations available in the complete case decreases as we consider more and more investment\nclimate variables. If we consider all the variables included in the survey, the complete case due to missing cells is\n\n2\n\nMention: \"IC data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Several methodological caveats may apply to this survey's respondents.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004541:23:0", "occurrence_id": "625a0cc4ffc4f25dd9b89e3b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Table 1: Marriage and Education by Birth Cohort in China**\n\n<u>Male Cohorts Born</u> <u>Female Cohorts Born</u>\n\n1935- 1945- 1955\n<u>1935-1945</u> <u>1945-1955</u> <u>1955-1965</u> <u>1945</u> <u>1955</u> <u>1965</u>\n\nSex Ratio of Cohorts 1.13 1.08 1.04\n\nTotal cohort size 569,588 873,505 1,024,910 504,251 809,753 987,090\n\nCohort Share Never Married by Education (%)\n\nTotal share never married 5.64 5.41 3.95 0.20 0.28 0.38\n\nLess than Primary 12.41 13.21 27.74 0.21 0.24 0.64\n\nPrimary 4.00 3.84 8.11 0.13 0.22 0.18\n\nSecondary 1.82 1.21 2.15 0.51 0.98 0.40\n\nCollege and Higher 0.79 0.66 1.30 0.67 2.30 1.56\n\nEducational Attainment by Cohort (%)\n\nLess than Primary 22.41 20.23 2.12 57.81 42.83 8.28\n\nPrimary 67.60 67.45 22.00 37.61 50.57 35.48\n\nSecondary 7.48 11.64 69.45 3.64 6.32 52.81\n\nCollege and Higher 2.51 0.68 6.43 0.94 0.27 3.43\n\n\n**_Source_** **:** China 1982 Census (1% sample), China 1990 Census (1% sample), China 2000 census (0.1%\nsample).\n\n\n**_Notes_** **\n\nMention: \"China 1982 Census\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its one percent sample gives the earliest schooling distribution.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004541:23:0", "occurrence_id": "1537995bb0ef1c40d3ba9864", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Table 1: Marriage and Education by Birth Cohort in China**\n\n<u>Male Cohorts Born</u> <u>Female Cohorts Born</u>\n\n1935- 1945- 1955\n<u>1935-1945</u> <u>1945-1955</u> <u>1955-1965</u> <u>1945</u> <u>1955</u> <u>1965</u>\n\nSex Ratio of Cohorts 1.13 1.08 1.04\n\nTotal cohort size 569,588 873,505 1,024,910 504,251 809,753 987,090\n\nCohort Share Never Married by Education (%)\n\nTotal share never married 5.64 5.41 3.95 0.20 0.28 0.38\n\nLess than Primary 12.41 13.21 27.74 0.21 0.24 0.64\n\nPrimary 4.00 3.84 8.11 0.13 0.22 0.18\n\nSecondary 1.82 1.21 2.15 0.51 0.98 0.40\n\nCollege and Higher 0.79 0.66 1.30 0.67 2.30 1.56\n\nEducational Attainment by Cohort (%)\n\nLess than Primary 22.41 20.23 2.12 57.81 42.83 8.28\n\nPrimary 67.60 67.45 22.00 37.61 50.57 35.48\n\nSecondary 7.48 11.64 69.45 3.64 6.32 52.81\n\nCollege and Higher 2.51 0.68 6.43 0.94 0.27 3.43\n\n\n**_Source_** **:** China 1982 Census (1% sample), China 1990 Census (1% sample), China 2000 census (0.1%\nsample).\n\n\n**_Notes_** **\n\nMention: \"China 1990 Census\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its one percent sample gives the middle year.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004541:23:0", "occurrence_id": "b6dcbb203eead5b399ffd1b9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Table 1: Marriage and Education by Birth Cohort in China**\n\n<u>Male Cohorts Born</u> <u>Female Cohorts Born</u>\n\n1935- 1945- 1955\n<u>1935-1945</u> <u>1945-1955</u> <u>1955-1965</u> <u>1945</u> <u>1955</u> <u>1965</u>\n\nSex Ratio of Cohorts 1.13 1.08 1.04\n\nTotal cohort size 569,588 873,505 1,024,910 504,251 809,753 987,090\n\nCohort Share Never Married by Education (%)\n\nTotal share never married 5.64 5.41 3.95 0.20 0.28 0.38\n\nLess than Primary 12.41 13.21 27.74 0.21 0.24 0.64\n\nPrimary 4.00 3.84 8.11 0.13 0.22 0.18\n\nSecondary 1.82 1.21 2.15 0.51 0.98 0.40\n\nCollege and Higher 0.79 0.66 1.30 0.67 2.30 1.56\n\nEducational Attainment by Cohort (%)\n\nLess than Primary 22.41 20.23 2.12 57.81 42.83 8.28\n\nPrimary 67.60 67.45 22.00 37.61 50.57 35.48\n\nSecondary 7.48 11.64 69.45 3.64 6.32 52.81\n\nCollege and Higher 2.51 0.68 6.43 0.94 0.27 3.43\n\n\n**_Source_** **:** China 1982 Census (1% sample), China 1990 Census (1% sample), China 2000 census (0.1%\nsample).\n\n\n**_Notes_** **\n\nMention: \"China 2000 census\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its thousandth sample gives the latest year.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004546:41:0", "occurrence_id": "6e42d10215b7125de7f3e1ab", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Source: United Nations Statistical Database. Available on line at\nhttp://unstats.un.org/unsd/snaama/dnllist.asp.\n\n**TOTAL POPULATION AGES 15 TO 64** : It corresponds to total population ages 15-64.\nSource: World Bank staff estimates from various sources including census reports, the United\nNations Population Division's World Population Prospects, national statistical offices,\nhousehold surveys conducted by national agencies, and Macro International. For Taiwan,\nChina the data comes from the National Statistical Office.\n\n**POPULATION TOTAL** : Population, total. Total population is based on the de facto\ndefinition of population, which counts all residents regardless of legal status or citizenship-except for refugees not permanently settled in the country of asylum, who are generally\nconsidered part of the population of their country of origin. The values shown are midyear\nestimates.\nSource: World Bank staff estimates from various sources including census reports, the United\nNations Population Division's World Population Prospects, national statistical offices,\nhousehold surveys conducted by national agencies, and Macro International.\n\n**TOTAL LABOR FORCE** : Labor force, total. Total labor force comprises people ages 15\nand older who meet the International Labor Organization definition of the economically\nactive population: all people who supply labor for the production of goods and services during\na specified period. It includes both the employed and the unemployed. While national\npractices vary in the treatment of such groups as the armed forces and seasonal or part-time\nworkers, in general the labor force includes the armed forces, the unemployed and first-time\njob-seekers, but excludes homemakers and other unpaid caregivers and workers in the\ninformal sector. Source: International Labor Organization, using World Bank population\nestimates.\n\n\n40\n\nMention: \"United Nations Statistical Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"National accounts population series are downloaded from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004546:41:0", "occurrence_id": "74718e388c820a0ba9f217a1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Source: United Nations Statistical Database. Available on line at\nhttp://unstats.un.org/unsd/snaama/dnllist.asp.\n\n**TOTAL POPULATION AGES 15 TO 64** : It corresponds to total population ages 15-64.\nSource: World Bank staff estimates from various sources including census reports, the United\nNations Population Division's World Population Prospects, national statistical offices,\nhousehold surveys conducted by national agencies, and Macro International. For Taiwan,\nChina the data comes from the National Statistical Office.\n\n**POPULATION TOTAL** : Population, total. Total population is based on the de facto\ndefinition of population, which counts all residents regardless of legal status or citizenship-except for refugees not permanently settled in the country of asylum, who are generally\nconsidered part of the population of their country of origin. The values shown are midyear\nestimates.\nSource: World Bank staff estimates from various sources including census reports, the United\nNations Population Division's World Population Prospects, national statistical offices,\nhousehold surveys conducted by national agencies, and Macro International.\n\n**TOTAL LABOR FORCE** : Labor force, total. Total labor force comprises people ages 15\nand older who meet the International Labor Organization definition of the economically\nactive population: all people who supply labor for the production of goods and services during\na specified period. It includes both the employed and the unemployed. While national\npractices vary in the treatment of such groups as the armed forces and seasonal or part-time\nworkers, in general the labor force includes the armed forces, the unemployed and first-time\njob-seekers, but excludes homemakers and other unpaid caregivers and workers in the\ninformal sector. Source: International Labor Organization, using World Bank population\nestimates.\n\n\n40\n\nMention: \"World Population Prospects\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its estimates are one input among census reports and national offices.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004546:41:0", "occurrence_id": "c480e1f58066b47415de0f31", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Source: United Nations Statistical Database. Available on line at\nhttp://unstats.un.org/unsd/snaama/dnllist.asp.\n\n**TOTAL POPULATION AGES 15 TO 64** : It corresponds to total population ages 15-64.\nSource: World Bank staff estimates from various sources including census reports, the United\nNations Population Division's World Population Prospects, national statistical offices,\nhousehold surveys conducted by national agencies, and Macro International. For Taiwan,\nChina the data comes from the National Statistical Office.\n\n**POPULATION TOTAL** : Population, total. Total population is based on the de facto\ndefinition of population, which counts all residents regardless of legal status or citizenship-except for refugees not permanently settled in the country of asylum, who are generally\nconsidered part of the population of their country of origin. The values shown are midyear\nestimates.\nSource: World Bank staff estimates from various sources including census reports, the United\nNations Population Division's World Population Prospects, national statistical offices,\nhousehold surveys conducted by national agencies, and Macro International.\n\n**TOTAL LABOR FORCE** : Labor force, total. Total labor force comprises people ages 15\nand older who meet the International Labor Organization definition of the economically\nactive population: all people who supply labor for the production of goods and services during\na specified period. It includes both the employed and the unemployed. While national\npractices vary in the treatment of such groups as the armed forces and seasonal or part-time\nworkers, in general the labor force includes the armed forces, the unemployed and first-time\njob-seekers, but excludes homemakers and other unpaid caregivers and workers in the\ninformal sector. Source: International Labor Organization, using World Bank population\nestimates.\n\n\n40\n\nMention: \"World Bank population\nestimates\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Labour organisation calculations use these population denominators.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004549:15:1", "occurrence_id": "f68d5b63a9618a07a0e56042", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: sup> For the Chinese EAF producer, the tax is equal to only 4 percent (compared to 3\npercent for U.S. firms using EAF). For Brazilian producers, this rate is much lower than their\nChinese counterparts at 11 percent (BOF) and 2 percent (EAF). Notably, this means that\nBrazilian EAF steel producers would be taxed by a lower amount than their U.S. counterparts\naccording to OECD carbon footprint estimates.\n\n\nImposing differentiated border taxes on imported steel does satisfy some of the policy\nconstraints. It likely would result in support from domestic firms (Constraint 1a). It also would\nprovide incentives for foreign firms to use the less intensive carbon technique (Constraints 2). It\nwould plausibly be consistent with WTO national treatment rules (individual domestic and\nforeign firms are both charged according to their carbon emissions). And since the U.S. therefore\nmight win a WTO dispute, foreign firms may be loath to ask their government for a formal\ncomplaint at the dispute body.\n\n\nThe real problem rests with Constraint 4. Unfortunately, one cannot look at the steel coil\nand know if it is produced by one technique or the other. <sup>21</sup> [^21: There are certain steel types that currently cannot be produced economically using EAF processes (e.g. exterior grade\nauto steel) because of the impurities present in recycled steel. For many other products, including flat rolled coil used\nin many situations, EAF and BOF processes could both be used. Moreover, the range of items where EAF produced\nsteel is used has expanded dramatically in recent years; a few decades ago, minimills produced only for the low-end\nlong product market such as rebar.] One cannot perform tests on the steel\nitself since the CO2 is not contained in the steel but instead is emitted in the production process.\nSimilar problems occur for requiring importers to buy emission permits as envisioned under\nvarious U.S. legislative proposals. Without cooperation by the Chinese and Brazilian\n\nMention: \"OECD carbon footprint estimates\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Their estimates imply a lower tax burden for the comparison producers.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004551:22:0", "occurrence_id": "2dbffca0518047a8ede4d657", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: performance of individuals “treated” by particular programs. Clearly, not all trade-related\nprograms can be amenable to such “treatment-effect” methodologies. The easiest are exportpromotion programs that target individual firms.. <sup>27</sup> [^27: See Volpe and Carballo (2008) for an evaluation of Peru’s export-promotion program.]\n\n\nEven in the limited areas where their application is relatively straightforward, applications of\n“clinical” impact evaluation methods to trade-related programs have so far been limited. They\nprovide no evaluation of spillover effects—even though spillovers are key to the justification of\npublic intervention. Moreover, like all clinical impact evaluations, they have uncertain “external\nvalidity”, as what works in one setting may not work in another. Notwithstanding these and other\ncaveats that have been extensively discussed in the literature (Rodrik 2008), they offer a valuable\ntool to understand what works and what does not. In particular, when carefully thought out, they\ncan help identify which _components_ of assistance programs work best. That is, beyond their\ncontribution to general accountability, they have the power to generate useful knowledge to\nrenew the factual basis on which to base policy advice and donor practice.\n\n\nInvestment in data and analysis should span work at both the macro and micro levels. An agenda\nin this area must center on a framework for rigorous evaluation of aid-for trade-projects,\nempirical research on aid impact evaluation, drawing on macro datasets from the CRS OECD\ndatabases and micro-data from projects that are implemented by development agencies. Country\nand/or regional analyses of aid for trade effectiveness are needed to assess how types of aid for\ntrade funds are spent in relation to their returns, as measured by increased trade flows, lower\ntrade costs, etc. Data on trade costs could be collected from a variety of sources, including trade\nsupport institutions, customs authorities, and international transport companies. Detailed data\nwill be needed to assess policies related to specific aid for trade interventions – e.g. support for\nindustrial upgrading, certification\n\nMention: \"CRS OECD\ndatabases\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Macro aid series for the evaluation literature come from these databases.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004553:7:1", "occurrence_id": "c0a766936395c4024c86cf63", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  (2) and defined above.\n\n_D_ and _α0i_ denote time effects and bank-level fixed effects, respectively. As in most papers, our\n\n\nestimations impose the restrictions of symmetry and degree one homogeneity in the price of\n\ninputs. <sup>6</sup>\n\n\nWe compute the H-statistic and the Lerner index using bank-level balance sheet and\n\n\nincome statement data from Bankscope for the period 1994-2008. The sample for MENA\n\n\nincludes data on 250 commercial and Islamic banks, operating in 12 MENA countries, namely:\n\n\nAlgeria, Bahrain, Egypt, Jordan, Kuwait, Lebanon, Morocco, Oman, Qatar, Saudi Arabia,\n\n\nTunisia, and United Arab Emirates. Table 1 shows the number of banks and observations for\n\n\neach country in our sample.\n\n\nWe compare competition in MENA to that in other regions using similar Bankscope data\n\n\nfor 3,785 banks operating in 122 countries across 6 regions. Table 2 lists the countries included\n\n\nin each region and reports the number of banks and observations that go into the calculations of\n\n\nthe H-statistic and the Lerner index for each region.\n\n\n**3.** **Empirical results on the non-structural measures of competition**\n\n\n<u>3.a. H- Statistic</u>\n\n\nTable 3 shows H-statistics for individual countries in MENA over two periods: 1994-2001 and\n\n\n2002-2008. Also, this table reports p-values for tests of whether H is equal to 0 and the market\n\n\ncan be best characterized as a monopoly or H is equal to 1 and the market operates under perfect\n\n\ncompetition.\n\n\n6 However, the results do not change if we drop these constraints.\n\n\n6\n\nMention: \"data on 250 commercial and Islamic banks\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The regional sample covers a quarter thousand banks across twelve countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004553:7:1", "occurrence_id": "00add28ccbb9845349a047bd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  (2) and defined above.\n\n_D_ and _α0i_ denote time effects and bank-level fixed effects, respectively. As in most papers, our\n\n\nestimations impose the restrictions of symmetry and degree one homogeneity in the price of\n\ninputs. <sup>6</sup>\n\n\nWe compute the H-statistic and the Lerner index using bank-level balance sheet and\n\n\nincome statement data from Bankscope for the period 1994-2008. The sample for MENA\n\n\nincludes data on 250 commercial and Islamic banks, operating in 12 MENA countries, namely:\n\n\nAlgeria, Bahrain, Egypt, Jordan, Kuwait, Lebanon, Morocco, Oman, Qatar, Saudi Arabia,\n\n\nTunisia, and United Arab Emirates. Table 1 shows the number of banks and observations for\n\n\neach country in our sample.\n\n\nWe compare competition in MENA to that in other regions using similar Bankscope data\n\n\nfor 3,785 banks operating in 122 countries across 6 regions. Table 2 lists the countries included\n\n\nin each region and reports the number of banks and observations that go into the calculations of\n\n\nthe H-statistic and the Lerner index for each region.\n\n\n**3.** **Empirical results on the non-structural measures of competition**\n\n\n<u>3.a. H- Statistic</u>\n\n\nTable 3 shows H-statistics for individual countries in MENA over two periods: 1994-2001 and\n\n\n2002-2008. Also, this table reports p-values for tests of whether H is equal to 0 and the market\n\n\ncan be best characterized as a monopoly or H is equal to 1 and the market operates under perfect\n\n\ncompetition.\n\n\n6 However, the results do not change if we drop these constraints.\n\n\n6\n\nMention: \"Bankscope data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The comparator covers thousands of banks across 122 countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004558:9:1", "occurrence_id": "0f8586b681129dfca65cb25b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  never be market entry by new\n\nexporters as a result of price changes (zero trade flows will always remain unchanged at\n\nzero).\n\n\nThis paper extends the standard trist model described in Brenton _et_ _al._ (2009)\n\nin two ways. First, we advance the static simulations associated with a single year\n\nto include sequential simulations that take into account multi-year tariff liberalization\n\nscenarios. Second, we include scenarios that allow for variations in other categories of\n\nborder-related charges, such as the spending tax and the revenue tax, as well as the\n\nimplementation of a VAT. It is important to remember, however, that TRIST has been\n\ndesigned with the specific task of providing policy makers with important insights into\n\nthe short-term effects of trade reform. It has not been designed for making longer-term\n\npredictions about the economy wide impact of trade reform; TRIST only looks at the\n\nimport side of the economy whereas trade reform will also have an impact on exporting\n\nsectors by reducing bias against exporting.\n\n#### **3 Analysis of Trade-Related Adjustment Costs**\n\n###### **3.1 Description of the data**\n\n\nWe use proprietary data provided by Syrian Customs, which were collected via the newly\nimplemented Automated System for Customs Data (asycuda) system, for Syrian cus\ntoms posts employing the system over the period between January and end-July 2009.\n\nThe data set comprises information on all import transactions at the 8-digit harmonized\n\nsystem (HS) level, including information on trading partners, c.i.f. import values, col\n\n8\n\nMention: \"Syrian Customs\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Proprietary transaction data collected through the automated customs system.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004564:13:0", "occurrence_id": "be989a4bcf53aa81272a389a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Note that, if the normalized citation curves intersect we may still be able to establish a\n\n\nrobust ordering in terms of average influence by imposing declining marginal influence. By a\n\n\nsimilar argument, the corresponding second-order dominance condition for this case is that\n\n_GA_ ( _c_ )  _GB_ ( _c_ ) for all _c_, with strict inequality for some _c_ . If this holds then one can conclude that\n\n\nthe publication record of A has greater average influence per publication for any influence\n\n\nfunction satisfying both our core and extra assumptions.\n\n\n**Empirical examples**\n\n\nWe illustrate these ideas using comparisons of the economics publications records of\n\n\nselected universities and the World Bank.\n\n\nWe rely on <u>Google Scholar</u> for citation counts. Compared to other bibliographic\n\n\ndatabases, Google Scholar casts a broad net, including citations by books, working papers,\n\n\nconference proceedings, open-access journals, new and less well-established journals. Google\n\n\nScholar is also more “global” in its reach, as it includes research outputs from everywhere in the\n\n\nworld and all languages. Google Scholar is also timelier than the bibliographic databases.\n\n\nNo time limit was set for the data on citations, since all are old institutions; in practice the\n\n\nrecords we use date back to 1964, but only five percent of them are before 1980. We searched in\n\n\nSCOPUS for articles where any author was affiliated with one of the institutions we consider;\n\n\nmultiple records are therefore possible in the dataset where collaboration has taken place across\n\n\ntwo or more of the institutions. Only articles in the discipline “Economics, Econometrics and\n\n\nFinance” were included in our analysis. Google Scholar citation data were then obtained for each\n\n\narticle in our database.\n\n\nTable 1 gives the summary statistics on citations for Berkeley, Harvard, London School\n\n\nof Economics (LSE), Oxford, Princeton and Yale, plus the World Bank. The _h_ -index ranking is\n\n\nHarvard (\n\nMention: \"Google Scholar citation data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Citation counts are obtained for each article in the bibliometric database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004565:24:3", "occurrence_id": "b541f131ac9dbcddac152d14", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: br>-24%<br>-24%<br>22%<br>**1%**<br>97%<br>-1%<br>28%<br>-25%<br>-25%<br>23%<br>**11%**<br>98%<br>10%<br>26%<br>-23%<br>-24%<br>22%<br>**3%**<br>97%<br>4%<br>28%<br>-25%<br>-24%<br>20%|\n\n\nSource: Authors’ calculations based on data from PROCOMER.\n\n<u>Table 7. The Contribution of Product Turn Over among Incumbent Firms in the Short Run</u>\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n|Col1|Growth Share Growth<br>cont. cont. cont. Entry value Exit value<br>firms firm_prod firm_prod Entry gap Exit gap|\n|---|---|\n|1998<br>1999<br>2000<br>2001<br>2002<br>2003<br>2004<br>2005<br>2006<br>2007|**26%**<br>94%<br>30%<br>50%<br>-45%<br>-41%<br>34%<br>**-7%**<br>94%<br>-1%<br>52%<br>-49%<br>-51%<br>42%<br>**-4%**<br>97%<br>-4%<br>54%<br>-51%<br>-49%<br>46\n\nMention: \"PROCOMER\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Firm export and turnover dynamics are computed from this trade agency's records.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004568:41:0", "occurrence_id": "0f8aaaf271500993879d58cd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 1990 0.032\n1991 0.008\n1992 0.274\n1994 0.038 0.007 0.071\n1996 0.045\n1997 0.062 0.010 0.177\n1998 0.133 0.011\n1999 0.046 0.009\n2000 0.079\n2001 0.036 0.011\n2002\n2003 0.035 0.043\n2004 0.011 0.078 0.023\n2005 0.023 0.014\n2006 0.010 0.011\n2007 0.017 0.171 0.019\n_Change in Variance_ **_-0.021_** **_0.002_** **_0.046_** **_-0.263_** **_-0.006_** **_0.007_** **_0.008_**\n**_<u>(-0.028)</u>_**\n\n\n\n*The industry variable in the 1994 Sakernas is at the one-digit level; ** Lower secondary and above for\nCambodia; *** Upper secondary and above for Cambodia.\n\n<u>Table 14: Evolution of Skill Premiums by Sector and Country</u>\nIndonesia Philippines Thailand Vietnam Cambodia China Mongolia\n_<u>Senior Sec. or higher</u>_\n1988\n\n - Agriculture/Mining 0.192\n\n - Manufacturing 0.343\n\n - Trade/Services 0.539\n1990\n\n - Agriculture/Mining 1.812\n\n - Manufacturing 0.642\n\n - Trade/Services 0.969\n1991\n\n - Agriculture/Mining 0.186\n\n - Manufacturing 0.322\n\n - Trade/Services 0.560\n\n1992 -0.586\n\n - Agriculture/Mining -0.102\n\n - Manufacturing -0.237\n\n - Trade/Services\n1994\n\n - Agriculture/Mining 0.809 0.161 1.148\n\n - Manufacturing 0.419 0.301 0.665\n\n - Trade/Services 0.\n\nMention: \"the 1994 Sakernas\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its industry coding is only at the one-digit level.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004569:22:1", "occurrence_id": "32487f45f4b568d636ea0bb0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: January 18, 2005</u> <u><mark>March 3, 2005</mark></u> <u>1½ month</u>\n\n\n<u>Average</u> <u>2,250</u> <u>13 months</u>\n\n_Source: Default information is based on \"Sovereign Defaults And Rating Transition Data, 2009 Update\", Standard and Poors, March_\n_17, 2010; sovereign spreads from J.P. Morgan EMBI Global_\n\n(*) In the case of Venezuela, there was a debate at the time among credit agencies; evidently investors did not react to S&P downgrade\n\n\nWe run a set of regressions with interactions between the systematic component of\nsovereign spreads (estimated from equation 5) and our country-specific crisis dummies using:\n\n\n��,��� ���,� ��� <sup>�</sup> ��� ��� <sup>�</sup> �� ��� <sup>�</sup> �� ��� <sup>�</sup> ��� <sup>�</sup> ����� �� <sup>�</sup> ������ ����       (7)\n\nwhere:  ����� ����,� �� <sup>�</sup>       - <sup>�</sup> ���, and the results are reported in Table 5.\n\n\nAs seen from Table 5, the estimated coefficient SSR is positive and statically significant,\neven with the presence of domestic macroeconomic variables in the equation explaining the\n\ndeterminants of private corporate bond market spreads in emerging economies. Interacting SSR\nwith the country crisis dummy variable provides a measure of the degree to which sovereign risk\naffects private external borrowing capital costs during times of sovereign debt distress and\nfinancial crises.\n\nMention: \"Standard and Poors\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Sovereign default and rating transition records come from this update.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004570:15:0", "occurrence_id": "b2c5519a0c8a7cf39b3a21c5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: - The baseline scenario fixes the primary balance at its historical value and takes the growth rate\nprojections from the IMF’s _World Economic Outlook_ (WEO) database. This WEO database\nalready assumes that after the crisis the growth rates are permanently lower than before the crisis.\n\n- <u>Scenario 1</u> calculates the required primary balance if the countries try to reach a debt target.\nThese debt targets are: (i) to reduce its debt to 40 percent of GDP by 2020 if debt stock is above\n40 percent of GDP at end-2009, or (ii) to permanently stabilize its debt-to-GDP ratio if the debt is\nalready below 40 percent of GDP at end-2009. <sup>20</sup> [^20: The 40 percent of GDP target corresponds to the sample median of the 2004-07 average public debt level of the MICs, see also\nIMF (2010b).] Comparing the required primary balances going\nforward to the country’s historical values shows the extent to which a country needs to adjust its\nprimary balance to reach its desired debt target.\n\n- <u>Scenario 2</u> is similar to scenario 1, but it assumes that the crisis will be prolonged further and\nthese countries will need to continue to implement expansionary policies as they did in 2007-09\nperiod, thereby leading to additional debt accumulation. It then estimates the required primary\nbalance if a country still aims to achieve the same debt target as under scenario 1 but this time by\nthe year 2020.\n\n- <u>Scenario 3</u> examines a more gradual approach towards adjusting the primary balances in each\ncountry so as to reach their debt target (as specified under scenario 1). This may be necessary for\nsome countries if the required fiscal adjustment under the earlier two scenarios may be very large\nand/or might not be politically feasible to implement (or might fuel further instability and,\nperhaps, a new downturn because of insufficient aggregate demand without such stimulus\nprograms).\n\n\n\n**_Baseline_*\n\nMention: \"World Economic Outlook\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its projected growth rates feed the debt dynamics simulation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004578:23:0", "occurrence_id": "6bb0998d5e7977176ca00405", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Appendix 1. Fedesarrollo’s Longitudinal Social Survey,** **_Encuesta Longitudinal Social_** **, ELS**\n\n\nSince September 1999, Fedesarrollo _,_ a Colombian think-tank, with the support of the Bogota, Cali,\nand Bucaramanga Chambers of Commerce, has carried out its social survey in these three cities\n(ELS, 2008). This survey, entitled the _Fedesarrollo_ Longitudinal Social Survey since 2004 (Stage\nIX), has become the only household panel for Colombia. It monitors a group of households for four\nyears through a rotating panel. The method employed corresponds to probability-based sampling at\nall stages, stratified with random unit selection at each stage and final selection of the informant\naccording to a prior recount of the number of households in the assigned city block. The sample is\nstratified and taken on a multistage basis in each city through the selection of primary sample units\n(censual sectors), followed by secondary units (city blocks) and final sample units (the selected\nhousehold). The sample framework is built around the cartographic inventory and lists of dwellings,\nhouseholds, and persons obtained from information gathered by the 1993 _Censo Nacional de_\n_Población y Vivienda_ (National Census of Population and Housing) with the latest updates for 2004\nobtained from each city’s planning office and the DANE. The new households that replaced the\nsample ones refusing to cooperate with the survey—as well as the 10 additional cities that now form\na part of the survey—were selected on the basis of the cartographic inventory and the list of blocks\nand dwellings updated to the year 2005 (the most recent year) provided by _Servinformación_, an\nentity supported by the DANE and the planning offices of the cities involved.\n\n\nThe 2008 stage has seen a substantial increase in the sample with the introduction of 10 more cities\nthat\n\nMention: \"cartographic inventory\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"It underpins the sample framework alongside dwelling and person lists.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004578:23:0", "occurrence_id": "12f3097ca530c1cb9c6b9d2c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Appendix 1. Fedesarrollo’s Longitudinal Social Survey,** **_Encuesta Longitudinal Social_** **, ELS**\n\n\nSince September 1999, Fedesarrollo _,_ a Colombian think-tank, with the support of the Bogota, Cali,\nand Bucaramanga Chambers of Commerce, has carried out its social survey in these three cities\n(ELS, 2008). This survey, entitled the _Fedesarrollo_ Longitudinal Social Survey since 2004 (Stage\nIX), has become the only household panel for Colombia. It monitors a group of households for four\nyears through a rotating panel. The method employed corresponds to probability-based sampling at\nall stages, stratified with random unit selection at each stage and final selection of the informant\naccording to a prior recount of the number of households in the assigned city block. The sample is\nstratified and taken on a multistage basis in each city through the selection of primary sample units\n(censual sectors), followed by secondary units (city blocks) and final sample units (the selected\nhousehold). The sample framework is built around the cartographic inventory and lists of dwellings,\nhouseholds, and persons obtained from information gathered by the 1993 _Censo Nacional de_\n_Población y Vivienda_ (National Census of Population and Housing) with the latest updates for 2004\nobtained from each city’s planning office and the DANE. The new households that replaced the\nsample ones refusing to cooperate with the survey—as well as the 10 additional cities that now form\na part of the survey—were selected on the basis of the cartographic inventory and the list of blocks\nand dwellings updated to the year 2005 (the most recent year) provided by _Servinformación_, an\nentity supported by the DANE and the planning offices of the cities involved.\n\n\nThe 2008 stage has seen a substantial increase in the sample with the introduction of 10 more cities\nthat\n\nMention: \"National Census of Population and Housing\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Its enumeration with later updates defines the sample units.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004589:5:0", "occurrence_id": "caf82fa34e94734dd34d8974", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: debt portfolio and the terms of new borrowing to estimate default probabilities. These\ninclude the effective interest rate and the average time to maturity of the portfolio, both\ndefined below.\n\n##### **3. Data and stylized facts**\n\n\nWe use a database mirroring an extensive subset of the choice of variables in Manasse et al.\n(2003) and Fioramanti (2008). <sup>2</sup> [^2: We are very grateful to Marco Fioramanti for kindly sharing his dataset with us.] The dataset contains information on 16 variables which\ncover representative regressors from the thematic groups defined above (see Table 2 for a\ndescription of the covariates used). In all cases, the frequency is annual. We construct an\nunbalanced panel of 46 EMCs in the period 1980‐2004. Table 3 presents a list with all\ncountries included in the dataset.\n\nThe dependent variable is a binary variable taking value one if a given country is defined to\nbe in default at a given year. A country/year is defined to be in a debt crisis if it is classified\nas being in default by Standard & Poor’s or if it receives a large non‐concessional\nInternational Monetary Fund (IMF) loan defined as access in excess of 100 percent of the\ncountry’s IMF quota. The default episodes are listed in Standard & Poor’s (2004), while non‐\nconcessional loan are drown from IMF _International Financial Statistics_ (IFS) database. This\ndefinition corresponds to the criterion used in Manasse et al. (2003) to identify debt crises.\n\n\nOn the regressors’ side, we include proxies for the most important determinants of\nsovereign defaults considered in the literature as cited in Table 1. Explanatory variables are\nsourced from IMF‘s _World Economic Outlook_ (WEO) and IFS databases, and the World\nBank‘s _Global Development Finance_ (GDF) database. The Country policy and Institutional\nAssessment (CPIA) was obtained from the World Bank.\n\nMention: \"Country policy and Institutional\nAssessment (CPIA)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its policy and institutional ratings are obtained from the bank.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004592:29:0", "occurrence_id": "3c0b84106182626990c3a9ae", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 29\n\n\nFigure 3: Decomposition of indices by policy instrument, 1980-84 and 2000-04\n\n(percent)\n(a) Decomposition of WRI, border measure only components\n\n1980–84 2000–04\n\n\n\n250\n\n\n200\n\n\n150\n\n\n100\n\n\n50\n\n\n0\n\n\n\n250\n\n\n200\n\n\n150\n\n\n100\n\n\n50\n\n\n0\n\n\n\nANZ NA ECA EU EFTA Ja/Ko\n\n\n\nANZ NA ECA EU EFTA Ja/Ko\n\n\n\n\n\n350\n\n\n300\n\n\n250\n\n\n200\n\n\n150\n\n\n100\n\n\n50\n\n\n0\n\n\n\n350\n\n\n300\n\n\n250\n\n\n200\n\n\n50\n\n\n\nDecoupled payments\n\n\nNon‐product specific (NPS) support\n\n\n\nDecoupled payments\n\n\nNon‐product specific (NPS) support\n\n\n\nSource: Authors’ calculations based on data in Anderson and Valenzuela (2008).\n\nMention: \"Anderson and Valenzuela (2008)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Support category estimates are computed from this dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004593:29:1", "occurrence_id": "d249e17870f2b8cb755e94f0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  with the ease of starting or closing a business;\n\n  - Trade Freedom: Rises as the levels of tariff and nontariff barriers are reduced;\n\n  - Fiscal Freedom: Increases as marginal tax rates on personal and corporate income and\n\ntotal taxes as a fraction of GDP fall;\n\n  - Freedom from Government: Rises as the importance of state-owned enterprises and\n\ngovernment expenditure as a share of GDP are reduced;\n\n  - Monetary Freedom: Rises as inflation rates fall and as the imposition of price controls are\n\nreduced;\n\n  - Investment Freedom: Rises with quality of investment climate for both foreign and\n\ndomestic investors;\n\n  - Financial Freedom: Rises as regulations on banking and other financial institutions are\n\neased;\n\n  - Property Rights: Rises with legislation and enforcement that protect individual property;\n\n  - Freedom from Corruption: Rises with the country’s score on Transparency International’s\n\nCorruption Perception Index.\n\n\n28\n\nMention: \"Transparency International’s\n\nCorruption Perception Index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The freedom from corruption dimension rises with this score.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004595:31:1", "occurrence_id": "079d945a9436415c01cc6f8a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  exported was\n\n\nthe same as the entrant’s main product during the year of entry into the new product. The\n\n\n24 The _D_ category refers to the new export goods for which there is a demand in the domestic market. The _X_\ncategory includes new export goods that are mainly produced for exportation. The grouping of new goods\nin terms of these two categories was based on information on domestic demand obtained from the Ministry\nof Agriculture (mainly for fresh produce), the National Institute of Statistics and the website of a\nsupermarket (E.Wong).\n\n\n30\n\nMention: \"website of a\nsupermarket\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Retail prices are collected alongside official statistics from this site.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004604:8:0", "occurrence_id": "7ae8ee82d52ac098d42a7d32", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: to crime. <sup>8</sup> [^8: We have also tried specifications including the share of losses of sales due to power outage and the\naverage level of schooling of firm employees. The results including these additional controls are similar.\nHowever, several countries do not have these measures. To preserve as many SADC countries as possible\nin our sample, we opted to report the results without including them.] The country level controls, _Z_, include logged GDP per capita, the ICRG index\n\n\nof government efficiency, the firing costs in years of salary, and the interest spread (as a\n\n\nmeasure of banking competitiveness). All the country-level variables come from the\n\n\nWorld Development Indicators data base of the World Bank. Since the regressions are at\n\n\nthe firm level, while we have country-level explanatory variables, to avoid exaggerating\n\n\nestimation precision, we cluster the standard errors at the country level (Moulton 1986).\n\n\nTable 3 reports the results.\n\n\nThe most important finding is that foreign ownership has a strong and significant\n\n\nrelationship with both sales growth and TFP. Increasing foreign ownership by 10\n\n\npercentage points would increase sales growth by 1.2 percentage points, and TFP by 1.8\n\n\npercentage points—effects that are non-trivial. To put this in perspective, Table A1 in\n\n\nthe appendix provides the comparison of SADC with the East Asia and Pacific region\n\n\n(consisting of China, Indonesia, Malaysia, and Thailand, EAP hereafter) and all non\nSADC countries with similar income level as SADCs. <sup>9</sup> [^9: That is, with income less than 4600 U.S. dollars of 2000, the level of South Africa.] Foreign ownership is\n\n\nsignificantly higher in SADC (17.5 percent) than in EAP (13.5 percent) and the similar\n\nincome sample (10.8 percent). In light of the positive effect of foreign ownership and\n\n\nSADC's advantage in this measure, foreign ownership has reduced the disadvantage of\n\n\nSADC toward other regions\n\nMention: \"World Development Indicators data base\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"All country-level regressors including the competition measure come from it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004606:6:0", "occurrence_id": "216b01f96377bfb603eb6101", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: underlines a positive and significant association between preparation, supervision and post-conflict\n\n\nprojects. Again, our analysis seems to be a lower bound.\n\n\nThe IEG data provide information on the outcome of the project, but also on the characteristics of\n\n\nthe project. The characteristics of project j, _Pj_, do not vary overtime. They include the main sectors\n\n\nof intervention, whether the project is an IDA or an IBRD project, whether it is an investment project\n\nor not, as well as the quality of preparation and the supervision efforts assessed by IEG. <sup>2</sup> All projects\n\n\nconsidered in the analysis have been evaluated and therefore have been closed. We deliberately used\n\n\nthe original closing date of the project to define its duration in order to avoid potential endogeneity\n\n\nlinked to the revised closing date _vis-`a-vis_ the success of the project. Projects started during war\n\n\nare dropped from our sample. Indeed, we want to distinguish post-conflict projects from projects in\n\n\ncountries at peace.\n\n\nIn order to define the civil war episodes, we used the PRIO version 4-2007 database and chose the\n\nhigh intensity criterion defined as at least 1,000 war related deaths per year. <sup>3</sup> This, of course, inevitably\n\n\nleads us to different results than if we used a lower level of casualties. Indeed, wars are shorter using\n\n\nthis definition and our database includes more post-conflict episodes and relapses to civil war than if we\n\n\nused a total of 1,000 deaths for the conflict as a threshold. It also implies that the country reported as\n\n\nbeing in a post-conflict period or at peace, for example in the few years preceding the collapse into war,\n\n\nmay have been considered at war using a lower intensity criteria, thus impacting on the results. How\n\never, the high intensity criterion allows defining clear episodes when a lower threshold could have led to\n\n\nunclear\n\nMention: \"IEG data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Project outcomes and characteristics are recorded in these evaluations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004606:6:0", "occurrence_id": "beb7b2d9ea766f7bff148bda", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: underlines a positive and significant association between preparation, supervision and post-conflict\n\n\nprojects. Again, our analysis seems to be a lower bound.\n\n\nThe IEG data provide information on the outcome of the project, but also on the characteristics of\n\n\nthe project. The characteristics of project j, _Pj_, do not vary overtime. They include the main sectors\n\n\nof intervention, whether the project is an IDA or an IBRD project, whether it is an investment project\n\nor not, as well as the quality of preparation and the supervision efforts assessed by IEG. <sup>2</sup> All projects\n\n\nconsidered in the analysis have been evaluated and therefore have been closed. We deliberately used\n\n\nthe original closing date of the project to define its duration in order to avoid potential endogeneity\n\n\nlinked to the revised closing date _vis-`a-vis_ the success of the project. Projects started during war\n\n\nare dropped from our sample. Indeed, we want to distinguish post-conflict projects from projects in\n\n\ncountries at peace.\n\n\nIn order to define the civil war episodes, we used the PRIO version 4-2007 database and chose the\n\nhigh intensity criterion defined as at least 1,000 war related deaths per year. <sup>3</sup> This, of course, inevitably\n\n\nleads us to different results than if we used a lower level of casualties. Indeed, wars are shorter using\n\n\nthis definition and our database includes more post-conflict episodes and relapses to civil war than if we\n\n\nused a total of 1,000 deaths for the conflict as a threshold. It also implies that the country reported as\n\n\nbeing in a post-conflict period or at peace, for example in the few years preceding the collapse into war,\n\n\nmay have been considered at war using a lower intensity criteria, thus impacting on the results. How\n\never, the high intensity criterion allows defining clear episodes when a lower threshold could have led to\n\n\nunclear\n\nMention: \"PRIO version 4-2007 database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its high-intensity criterion defines the civil war episodes.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004607:6:0", "occurrence_id": "b2cfee5e63b7a222f3e310de", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Information instruments have diverse forms, including reports of measured pollution,\n\n\nenvironmental accident reports, and environmental performance ratings. In the US, for example,\n\n\nthe Toxic Release Inventory (TRI) discloses toxic chemical releases and waste management\n\n\nactivities by significant toxic polluters and federal facilities. In developing countries, however,\n\n\nweak regulatory institutions may have much greater difficulty in implementing such emissions\n\n\ninventories. In addition, despite an emerging literature on stakeholders’ role in improving firms’\n\n\nenvironmental performance (Arora and Cason, 1998; Blackman and Bannister, 1998; Pargal and\n\n\nWheeler, 1996; Wheeler et al., 1997), concerns remain about the public’s ability to understand\n\n\nand utilize complex emissions reports. For example, Bui and Mayer (2003) find that the release\n\n\nof TRI’s highly-detailed information on facilities’ toxic emissions has virtually no effect on\n\n\nhousing prices in neighboring areas, even when the release is unexpected. The dual problems of\n\n\nemissions inventories in developing countries – technical feasibility and public understanding –\n\n\nhave led to a preference for programs that condense complex information into environmental\n\n\nperformance ratings that are disclosed to the public.\n\n\nResearch on the effectiveness of performance rating and disclosure (PRD) programs suggests\n\n\nthat that have a significant, positive impact on regulatory compliance (Afash et al., 1997;\n\n\nDasgupta et al., 2006; Garcia et al., 2007, 2009; Wang et al., 2004). Dasgupta et al. (2006)\n\n\nsummarize the changes in compliance rates for several PRD programs in Asia: One and two\n\n\nyears after the inception of a PRD, compliance rates increased from 37% to 61% in Indonesia,\n\n\n8% to 58% in the Philippines, 10% to 24% in Vietnam, 75% to 85% in Zhenjiang, China and\n\n\n23% to 62% in Hohhot, China.\n\n\nSeveral empirical studies also find that PRD programs have improved firms’ environmental\n\n\nperformance in Indonesia (Afash et al., 1997\n\nMention: \"Toxic Release Inventory\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It discloses releases and waste management by major polluters in one country.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004614:8:0", "occurrence_id": "240896fcc2f8c2907d68eb52", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Human capital_ is approximated by the initial gross rate of secondary schooling (in logs) and the\n\n\ndata is obtained from Barro and Lee (2001). <sup>5</sup> [^5: This “flow” measure captures more closely current policies on schooling and human capital investment than\n“stock” measures related with educational attainment of the adult population or life expectancy (Loayza et al.\n2005).] _Financial development_ is measured by the ratio of domestic\n\n\ncredit to the private sector to GDP and the data is collected from Beck, Demirgüç‐Kunt and Levine\n\n\n(2000), Beck and Demirgüç‐Kunt (2009), and updated using data from the IMF’s International Financial\n\n\nStatistics and the World Bank’s WDI. For the sake of robustness, we use other proxies of financial\n\n\ndevelopment: domestic credit provided by domestic money banks, and liquid liabilities of the financial\n\n\nsector. Both variables are expressed as a percentage of GDP and in logs. _Institutional quality_, on the\n\n\nother hand, comprises different dimensions such as absence of corruption, rule of law, enforcement of\n\n\ncontracts, quality of the bureaucracy, democratic accountability, among others. We use the ICRG index\n\n\nof political risk as our indicators of institutional quality. The data is published in the International Risk\n\n\nCountry Guide (ICRG) by the Political Risk Services (PRS) Group. The _lack of price stability_ is\n\n\napproximated by the average CPI inflation rate. This variable typically reflects the quality of monetary\n\n\nand fiscal policies and is directly related to other indicators of poor macroeconomic management. The\n\n\ndata on the inflation rate is gathered from the IMF’s International Financial Statistics.\n\n\n_Infrastructure_ is a multi‐dimensional concept; however, most empirical studies have focused on\n\n\na single‐sector approach partly due to: (i) the difficulty of capturing the multiple dimensions of\n\n\ninfrastructure in a simple way, and (ii) the high correlation often found among indicators of different\n\n\ntypes of infrastructure assets (Calderón and Servén, 2004). <sup\n\nMention: \"International Financial\n\n\nStatistics\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its series update the earlier bank structure measures.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004614:8:0", "occurrence_id": "642051302e85b0db9a5cef26", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Human capital_ is approximated by the initial gross rate of secondary schooling (in logs) and the\n\n\ndata is obtained from Barro and Lee (2001). <sup>5</sup> [^5: This “flow” measure captures more closely current policies on schooling and human capital investment than\n“stock” measures related with educational attainment of the adult population or life expectancy (Loayza et al.\n2005).] _Financial development_ is measured by the ratio of domestic\n\n\ncredit to the private sector to GDP and the data is collected from Beck, Demirgüç‐Kunt and Levine\n\n\n(2000), Beck and Demirgüç‐Kunt (2009), and updated using data from the IMF’s International Financial\n\n\nStatistics and the World Bank’s WDI. For the sake of robustness, we use other proxies of financial\n\n\ndevelopment: domestic credit provided by domestic money banks, and liquid liabilities of the financial\n\n\nsector. Both variables are expressed as a percentage of GDP and in logs. _Institutional quality_, on the\n\n\nother hand, comprises different dimensions such as absence of corruption, rule of law, enforcement of\n\n\ncontracts, quality of the bureaucracy, democratic accountability, among others. We use the ICRG index\n\n\nof political risk as our indicators of institutional quality. The data is published in the International Risk\n\n\nCountry Guide (ICRG) by the Political Risk Services (PRS) Group. The _lack of price stability_ is\n\n\napproximated by the average CPI inflation rate. This variable typically reflects the quality of monetary\n\n\nand fiscal policies and is directly related to other indicators of poor macroeconomic management. The\n\n\ndata on the inflation rate is gathered from the IMF’s International Financial Statistics.\n\n\n_Infrastructure_ is a multi‐dimensional concept; however, most empirical studies have focused on\n\n\na single‐sector approach partly due to: (i) the difficulty of capturing the multiple dimensions of\n\n\ninfrastructure in a simple way, and (ii) the high correlation often found among indicators of different\n\n\ntypes of infrastructure assets (Calderón and Servén, 2004). <sup\n\nMention: \"WDI\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It supplies the companion updates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004614:8:0", "occurrence_id": "aeb62ccf1323c940ec8b301e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Human capital_ is approximated by the initial gross rate of secondary schooling (in logs) and the\n\n\ndata is obtained from Barro and Lee (2001). <sup>5</sup> [^5: This “flow” measure captures more closely current policies on schooling and human capital investment than\n“stock” measures related with educational attainment of the adult population or life expectancy (Loayza et al.\n2005).] _Financial development_ is measured by the ratio of domestic\n\n\ncredit to the private sector to GDP and the data is collected from Beck, Demirgüç‐Kunt and Levine\n\n\n(2000), Beck and Demirgüç‐Kunt (2009), and updated using data from the IMF’s International Financial\n\n\nStatistics and the World Bank’s WDI. For the sake of robustness, we use other proxies of financial\n\n\ndevelopment: domestic credit provided by domestic money banks, and liquid liabilities of the financial\n\n\nsector. Both variables are expressed as a percentage of GDP and in logs. _Institutional quality_, on the\n\n\nother hand, comprises different dimensions such as absence of corruption, rule of law, enforcement of\n\n\ncontracts, quality of the bureaucracy, democratic accountability, among others. We use the ICRG index\n\n\nof political risk as our indicators of institutional quality. The data is published in the International Risk\n\n\nCountry Guide (ICRG) by the Political Risk Services (PRS) Group. The _lack of price stability_ is\n\n\napproximated by the average CPI inflation rate. This variable typically reflects the quality of monetary\n\n\nand fiscal policies and is directly related to other indicators of poor macroeconomic management. The\n\n\ndata on the inflation rate is gathered from the IMF’s International Financial Statistics.\n\n\n_Infrastructure_ is a multi‐dimensional concept; however, most empirical studies have focused on\n\n\na single‐sector approach partly due to: (i) the difficulty of capturing the multiple dimensions of\n\n\ninfrastructure in a simple way, and (ii) the high correlation often found among indicators of different\n\n\ntypes of infrastructure assets (Calderón and Servén, 2004). <sup\n\nMention: \"ICRG index\n\n\nof political risk\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its components measure contract enforcement, bureaucracy and accountability.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004620:25:0", "occurrence_id": "7c3340b8aad6af4656606177", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: as number 1 in Newsweek‘s index falls to 17 if I put all the weight on health; Australia‘s rank at\n\n\nnumber 4 falls to 13 if one put all the weight on education. In exploring the website for\n\n\nNewsweek‘s mashup, the most dramatic impact of re-weighting appears to be for China; if one\n\n\nputs all the weight on ―economic dynamism‖ China‘s rank rises from 66 to 13.\n\n\nNone of their websites make it easy for users to properly assess the sensitivity of these\n\n\nmashup indices to changing weights. Yet it would be relatively easy to program the required\n\n\nflexibility into the current web sites, so users can customize the index with their preferred\n\n\nweights, to see what difference it makes. The only example I know of to date is the OECD‘s\n\n\n[Social Institutions and Gender Index. Their interactive website, ―My Gender Index,‖ allows](http://my.genderindex.org/)\n\n\nusers to vary the composition and weights of the index, and immediately gives the corresponding\n\n\ncountry rankings and maps them. There are also some useful graphical tools for assessing\n\n\nrobustness from the work of Foster et al. (2009). A careful assessment of robustness using such\n\n\ntools would be a more open approach than encouraging users to think that the data have been\n\n\naggregated in the one uniquely optimal way.\n\n\nFew of the mashups of development data have said much about data quality, including\n\n\ninternational comparability. Data constraints are often mentioned, but most of the time the\n\n\nmashups take their data as given with little or no critical attention to the problems; the data often\n\ncome from others who can be blamed for its inadequacies. <sup>35</sup> [^35: An exception is the DBI, which relies on primary data collected by the team.] Under certain circumstances,\n\n\nforming a mashup index may actually help reduce data concerns, notably when averaging across\n\n\nindicators reduces overall errors. This may have bearing on the choice of indicators, though one\n\nMention: \"Social Institutions and Gender Index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Its interactive site lets users vary weights, the sole such example cited.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004622:21:0", "occurrence_id": "5fb917f2f7b102550ef8e58c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: research projects meant to improve the stress testing along this line are under way at\nmany authorities. Among the central banks that included a loss distribution into their\nstress testing is the Austrian Central Bank. <sup>25</sup> [^25: Sveriges Riksbank also considers a credit loss distribution (Foglia, 2009).] In the Austrian stress test, an overall systemwide credit loss distribution is estimated by using a portfolio model, the Credit Risk Plus,\nwhich combines model-estimated sectoral default rates with individual borrowers’ default\nprobabilities from the central credit register. <sup>26</sup> [^26: See Foglia (2009), for a description of other approaches and research projects meant to incorporate a loss\ndistribution instead of a point estimate of the loss.]\n\nAn alternative approach to estimating the expected loss is the estimation of the LLPs; an\nincrease in LLPs accordingly diminishes net earnings. The Croatian Central Bank\nassumes that the LLPs will have the same growth as NPLs, which are model-estimated.\nLLPs are also estimated by the Romanian Central Bank (bank-level LLPs) and the Polish\nCentral Bank (aggregate level LLPs). <sup>27</sup> [^27: See section 4.1.]\n\n**6.2. Capital Adequacy Ratio (CAR)**\n\nOwing to the focus of the Basel Accords on capital adequacy, financial stability\nauthorities express the outputs of their stress tests in terms of the number of banks with\nCAR under 8% (undercapitalized banks) <sup>28</sup> [^28: The general 8% minimum CAR could substituted by a respective country-specific minimum CAR, and\nCSEECBs often use 10% or 12% as the minimum CAR.] and accordingly the system-wide need for\nrecapitalization. The capital adequacy ratio has in the numerator the expected loss\nsubtracted from the buffer of capital and eventually profit, and in the denominator the\nrisk-weighted assets (RWA). All CSEECBs express the output of the stress testing in\nterms of CAR\n\nMention: \"the central credit register\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Individual borrower default probabilities are taken from this register.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004626:3:1", "occurrence_id": "ffd46bbf2035a43fb565af58", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  access to credit.\n\n\nAccording to the 2006 Doing Business Report, Malawi ranked 109 out of 129 countries\n\n\nin terms of private credit to GDP, a frequently-used measure of financial development. Malawi\n\n\nalso gets the lowest marks in the “depth of credit information index” which proxies for the\n\n\namount and quality of information about borrowers available to lenders. Using more micro data,\n\n1 The following quote from 1973 by Robert McNamara when he was the World Bank president exemplifies this\nview: “The miracle of the Green Revolution may have arrived, but for the most part, the poor farmer has not been\nable to participate in it. He simply cannot afford to pay for the irrigation, the pesticide, the fertilizer… For the small\nholder operating with virtually no capital, access to capital is crucial.”\n\nMention: \"2006 Doing Business Report\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its ranking on credit to output places the country low.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004628:43:2", "occurrence_id": "5fc2aba4730a1bf41ad9b024", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 0.000 -0.061 0.024 -0.067* -0.120** 0.017\n(0.031) (0.042) (0.028) (0.038) (0.046) (0.040)\n\n\nDistrict: Muzaffarabad 0.078** -0.014 0.016 0.027 0.003 -0.069\n(0.036) (0.045) (0.034) (0.047) (0.056) (0.051)\n\n\nConstant 0.375*** 0.426*** 0.249*** 0.374*** 0.458*** 0.225***\n(0.066) (0.077) (0.053) (0.077) (0.089) (0.075)\n\nObservations 4670 4670 4670 4670 4670 4670\n<u>R-squared</u> <u>0.025</u> <u>0.036</u> <u>0.031</u> <u>0.135</u> <u>0.111</u> <u>0.082</u>\n_Notes:_ Estimated coefficients are reported from specifications that correlate trust in outsiders with distance to the fault-line and\nother geographic controls, controlling for our measure of foreign assistance. Distance from the fault-line and the epicenter is\ncalculated using the 2009 District Household Census and digital maps of the Himalayan Frontal Thrust provided by NESPAK. The\nslope is the mean slope of the Union Council of the village from digital maps provided by NESPAK. Other variables from the\nDistrict Household Survey 2009. Omitted Dummy variables are Asset Index: Rich, District: Abbottabad. The regressions also\nincluded a full set of age indicator variables (co\n\nMention: \"the 2009 District Household Census\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"District locations here give distances to the fault line and epicentre.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004630:37:0", "occurrence_id": "5d7cb7aa7c983311d6f1f31b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Table A.11 Matrix of correlation coefficients between household variables.\n\n\nNote: variable abbreviation: pedu0: Proportion of members without vocational training\npedu1: Proportion of members with vocational training\npedu2: Proportion of members with college/university\npwork: Proportion of working members\n\n\n_Source: Authors’ estimation from VHLSS_\n\n\n36\n\nMention: \"VHLSS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Working-member shares by schooling are estimated from this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004632:11:1", "occurrence_id": "b0f5f81b729a72c5c3f30292", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  a small increase in the excess reserves over total deposits ratio. <sup>13</sup> The\nsecond panel reports a correlation matrix for a number of control variables. The correlation is high among some of these variables. Our approach is therefore to check the\nrobustness of our results to the inclusion of alternative sets of control variables.\n\nWe go on to investigate the determinants of the trend in …nancial intermediation by\nestimating equation (2). The main results are reported in Table 2. The speci…cations\nshown include control variables lagged by two periods <sup>14</sup> (i.e. half a year), in addition to\nthe full set of country …xed e¤ects, country-speci…c trends, year e¤ects and seasonal e¤ects.\n\n\n9We also experimented with a broader measure of in‡ows including foreign direct investment, portfolio\n‡ows and debt liabilities. This variable tended not to be signi…cant.\n\n10Banking crises are identi…ed using the Caprio-Klingebiel World Bank dataset (2003) at\nwww1.worldbank.org/…nance/html/database_sfd.html. According to this source, banking crises\nocourred in the following countries (with crisis inception years in brackets): Bulgaria (1996), Latvia\n(1995), Poland (1993), Lithuania (1994).\n\n11Bulgaria and Romania joined the European Union in 2007. All other sample countries joined in 2004.\nSlovenia has - in 2007 - also joined the euro.\n\n12The series used are not available for a su¢ciently long period prior to reform in Poland and the Czech\nRepublic (notably quarterly GDP).\n\n13Credit to the private sector includes credit in both domestic and foreign currency. Payments technology may a¤ect the use of domestic currency transactions accounts by residents\n\nMention: \"Caprio-Klingebiel World Bank dataset\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Banking crisis episodes are identified using this dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004633:14:1", "occurrence_id": "6185a19ebb7e5af0ca94dea3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: %, with an\n\n\naverage of 44%. In our smaller sample, Islamic banks are significantly smaller than conventional\n\n\nbanks, have higher fixed assets, but lower non-interest earning assets.\n\n\nWe would also like to control for the ownership structure of banks, but face the problem\n\n\nthat the ownership dummy is not well populated in Bankscope and only reflects the current\n\n\nownership structure. We therefore control for the ownership structure on the country level, using\n\n\ndata from Barth, Caprio, and Levine (2008). The foreign bank share ranges from zero to 100%,\n\n\nwhile the government ownership share ranges from zero to 98%. On average, the foreign\n\n\n(government) ownership share is 35% (17%).\n\n\nIn some of our regressions, we also include several country-level indicators. Specifically,\n\n\nwe include GDP per capita as indicator of economic development, GDP growth, Private Credit to\n\n\nGDP as measure of financial development and the three-bank concentration ratio. A large\n\n\nliterature has related bank concentration to both the efficiency and stability of financial systems\n\n\n13\n\nMention: \"Barth, Caprio, and Levine (2008)\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Foreign and government bank ownership shares come from this source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004634:18:2", "occurrence_id": "9b0ab7de2bc01ba5ee574aca", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ‐3%|\n|**Senegal**<br>**(single units)**|350|29|29|29|29|29|29|\n|**Tanzania**|59|1|3%|14%|N/A|N/A|N/A|\n\n\nSource: UN COMTRADE via WITS and author’s calculations based on data from individual country EPZ\nauthorities\n\nGiven that many of the African zone programs were established only recently, could the low levels of\nexports simply be a function of time? Certainly, this is part of the story. Data on growth of exports\nduring the period 2000 to 2008 do indicate that the African zones, by and large, grew relatively quickly\n(although in most cases from very small bases) during the decade. A closer look at the growth patterns,\nhowever, raises some concerns over many of the African zones programs. As can be seen in Table 2,\nmany of the African programs experienced rapid growth during the period 2000 to 2004, but slower\ngrowth or even decline since then. This is partly attributable to the base on which they are growing in\neach of those periods. But there may also be a structural factor at play. This is competition from Asia in\ngeneral, and specifically, garment sector competition in the post‐MFA environment. In the period 2000\nto 2004, African EPZs (particularly Kenya and Lesotho) benefited enormously from trade preferences to\nthe US market granted under the AGOA program. At the same time, they faced limited competition in\nthe US market from Asian producers, who continued to confront quota restrictions until the Multifiber\nAgreement expired at the end of 2004. At least in the cases of Kenya and Lesotho, declining\ncompetitiveness in the aftermath of the MFA is a well‐documented source of export stagnation and\nemployment losses since 2005.\n\nMore broadly, if we look at the typical growth path of successful\n\nMention: \"UN COMTRADE\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Zone trade values are drawn from this database through the trade portal.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004634:18:2", "occurrence_id": "dc0e19b60f6b43db6425e2ea", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ‐3%|\n|**Senegal**<br>**(single units)**|350|29|29|29|29|29|29|\n|**Tanzania**|59|1|3%|14%|N/A|N/A|N/A|\n\n\nSource: UN COMTRADE via WITS and author’s calculations based on data from individual country EPZ\nauthorities\n\nGiven that many of the African zone programs were established only recently, could the low levels of\nexports simply be a function of time? Certainly, this is part of the story. Data on growth of exports\nduring the period 2000 to 2008 do indicate that the African zones, by and large, grew relatively quickly\n(although in most cases from very small bases) during the decade. A closer look at the growth patterns,\nhowever, raises some concerns over many of the African zones programs. As can be seen in Table 2,\nmany of the African programs experienced rapid growth during the period 2000 to 2004, but slower\ngrowth or even decline since then. This is partly attributable to the base on which they are growing in\neach of those periods. But there may also be a structural factor at play. This is competition from Asia in\ngeneral, and specifically, garment sector competition in the post‐MFA environment. In the period 2000\nto 2004, African EPZs (particularly Kenya and Lesotho) benefited enormously from trade preferences to\nthe US market granted under the AGOA program. At the same time, they faced limited competition in\nthe US market from Asian producers, who continued to confront quota restrictions until the Multifiber\nAgreement expired at the end of 2004. At least in the cases of Kenya and Lesotho, declining\ncompetitiveness in the aftermath of the MFA is a well‐documented source of export stagnation and\nemployment losses since 2005.\n\nMore broadly, if we look at the typical growth path of successful\n\nMention: \"data from individual country EPZ\nauthorities\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Zone-level counts come from each country's authority.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004635:14:0", "occurrence_id": "017489269cde88081a27dec9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: from a failed business (Sullivan, et al., 1989). While the personal guarantee of a firm’s owner\n\n\nmight encourage a level of financial discipline, in countries without a personal bankruptcy\n\n\nframework, a single business failure could doom an owner to a lifetime of outstanding debt\n\n\n(Uttamchandani and Menezes, 2010) and effectively prevent them from re-entering the market as\n\n\nseasoned entrepreneurs (Armour and Cumming, 2005).\n\n\n**3.** **Bankruptcy Regimes during the Crisis**\n\n\n**_3.1_** **_Bankruptcy Regime as One of the Main Tools for Entrepreneurship Recovery_**\n\n\nThe 2008 global financial crisis is causing a sharp increase in bankruptcies around the world.\n\n\nEnsuring that viable companies can continue to operate as going concerns and preserving jobs\n\n\nhas become especially important. In response, policymakers are debating whether existing\n\n\nbankruptcy regimes adequately address current business’ demands.\n\n\nIneffective procedures for dealing with bankruptcy can deepen and prolong a crisis. Low\n\n\ndemand may make firms reluctant to restructure and regain their health, affecting the health of\n\n\nthe banking sector and further restricting credit to firms that may be recovering. Distress among\n\n\nfinancial institutions can also reduce the incentives for firms to repay their loans. The\n\n\nbankruptcy processes – which are generally already under strain during normal times – can be\n\n\ncompletely overwhelmed (Demirgüç-Kunt, 2009).\n\n\nDuring 2009, the number of corporate bankruptcies in Japan was 13,306 (up 4.9% from\n\n\n12,681 in 2008, Teikoku Databank, 2010 <sup>vi</sup> ), in Great Britain 94,135 (5.88% growth compare to\n\n\n2008, Ministry of Justice, 2010 <sup>vii</sup> ), and in Germany, the number of corporate bankruptcies was\n\n\n32,687, which represents 11% annual increase, whereas the volume of debt\n\nMention: \"Teikoku Databank\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Corporate bankruptcy counts for the country come from this data bank.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004636:8:0", "occurrence_id": "af82ce29a1953523301d1bb3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: the country. Section 3 examines the recent evolution of commonly-used measures of bank\n\n\nconcentration such as the share of assets held by the top five banks and the Herfindahl\n\nHirschman (HH) index and compares these measures for Russia to those for Brazil, China, and\n\n\nIndia. Section 4 reviews the regulatory and supervisory framework governing bank entry and\n\n\noperations as well as bank insolvency and exit in Russia. The objective of this section is to\n\n\ndetermine whether the regulatory and supervisory framework promotes contestability. Section 5\n\n\ndescribes the methodologies and data used to obtain the Panzar and Rosse H-statistic and the\n\n\nLerner index of market power. Section 6 reports the estimates for the Lerner Index and the H\n\nstatistic. Here we compare Russia to Brazil, China, and India and also investigate differences\n\n\nacross groups of banks within Russia. Finally, Section 7 concludes and discusses some policy\n\n\nrecommendations.\n\n\n**2.** **The structure of the Russian banking sector**\n\n\nThe Russian banking sector consists of 1,007 banks. <sup>**5**</sup> Despite the large number of\n\n\ninstitutions operating in the country, the Russian banking sector is fairly concentrated and\n\n\ndominated by state-owned banks. The largest 20 banks control approximately 70 percent of total\n\n\nbank assets and six government-controlled banks account for 52 percent of total bank assets and\n\n\n60 percent of deposits. The largest government-owned bank, Sberbank, is also the largest credit\n\n\ninstitution in the country and accounts for 25 percent of total assets of the Russian banking\n\n\nsystem.\n\n\n**5** This number and most of the statistics in this section come from the following publications produced by the\nCentral Bank of Russia: “Review of the Banking Sector of the Russian Federation,” February 2010, and the\n“Bulletin of Banking Statistics”, January, 2010.\n\n\n7\n\nMention: \"Bulletin of Banking Statistics\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Banking sector figures appear in this issue of the central bank bulletin.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004638:19:1", "occurrence_id": "843f0480760d3ebc2fe7407f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \ndistributed lag model (ARDL) yields a short-run coefficient of the changes in growth of close to 0.3,\n\nimplying that a drop of 2.3 percentage points in growth (from 7 to 4.7 percent) would be associated\n\nwith an increase in unemployment of about 0.7 percentage points (annex 4). <sup>15</sup> Given the relatively\n\nlow value of the “Okun coefficient”, as well as the low value of the elasticity of employment to\n\noutput, we expect a jobless recovery.\n\n\nThe labor market is a cause of concern given the lower job creation rates, higher job destruction\n\nand increases in the size of the informal labor market. In addition to the effect of the sudden stop on\n\njob flows, the World Bank rapid survey (World Bank, 2009c) shows that large manufacturing firms\n\nare reducing employment more than small firms, making it likely that informal employment will\n\nincrease. In emerging economies, the informal labor market expands as economic activity contracts,\n\nand hence it is countercyclical in nature (Perry et al. 2008). The expected increase in informality is\n\n\n13 El-Ehwany and EL-Megharbel, 2008 found that a change of 1 percentage point in economic growth would lead to a\nchange of 0.53 percentage points in employment in the same direction.\n\n14 The 0.3 elasticity of employment with respect to output is estimated by means of an Autoregressive Distributed Lag\n(ARDL) model. Using both yearly data for 1980-2009, or quarterly data for 2003-2009, the long run elasticity decreases\nin value and turns statistically insignificant when a deterministic trend is introduced, and the short run elasticity is 0.1.\nIn the quarterly model, the long run elasticity is 0.3, with insignificant short run elasticity. This elasticity is lower than\nthat reported in the previous footnote and generally used in studies of the labor market in Egypt.\n\n15\n\nMention: \"World Bank rapid survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"It shows large manufacturers cutting employment more than small ones.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004646:8:0", "occurrence_id": "57fb70e500fe770df257a366", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Here, the question of whether commodity price volatility has increased over time is\n\n\naddressed with analytical rigor and at a disaggregated level. We use a unique, newly\n\n\ncompiled dataset of 45 individual commodity prices and five commodity price indices at\n\n\na monthly frequency. Two econometric tests are applied for identifying structural breaks\n\n\nin GARCH-type processes in order to provide a robustness check and these are described\n\n\nbelow.\n\n##### **3. Data**\n\n\nFor this paper we use a newly compiled dataset which covers 45 individual commodities\n\n\nand 5 commodity price indexes. We use a monthly unbalanced panel of observations\n\n\nfrom Global Financial Data (GFD) covering the period 1784 – 2009. We take a pragmatic\n\n\napproach as to what constitutes a “commodity” and take all “commodities” included in\n\n\nthe GFD dataset as our subject of study. We use data from GFD as it is the most\n\n\ncomprehensive source, verifying that the price data is consistent with that from\n\n\nalternative sources (World Bank, IMF, and UNCTAD) for any overlapping commodity\n\n\nand time period. The details of the dataset can be found in Table 1.\n\n\n7\n\nMention: \"Global Financial Data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The long monthly panel of commodity prices comes from this provider.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004646:8:0", "occurrence_id": "6576026b1ceaaed84dcc11f2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Here, the question of whether commodity price volatility has increased over time is\n\n\naddressed with analytical rigor and at a disaggregated level. We use a unique, newly\n\n\ncompiled dataset of 45 individual commodity prices and five commodity price indices at\n\n\na monthly frequency. Two econometric tests are applied for identifying structural breaks\n\n\nin GARCH-type processes in order to provide a robustness check and these are described\n\n\nbelow.\n\n##### **3. Data**\n\n\nFor this paper we use a newly compiled dataset which covers 45 individual commodities\n\n\nand 5 commodity price indexes. We use a monthly unbalanced panel of observations\n\n\nfrom Global Financial Data (GFD) covering the period 1784 – 2009. We take a pragmatic\n\n\napproach as to what constitutes a “commodity” and take all “commodities” included in\n\n\nthe GFD dataset as our subject of study. We use data from GFD as it is the most\n\n\ncomprehensive source, verifying that the price data is consistent with that from\n\n\nalternative sources (World Bank, IMF, and UNCTAD) for any overlapping commodity\n\n\nand time period. The details of the dataset can be found in Table 1.\n\n\n7\n\nMention: \"GFD dataset\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"All commodities it includes define the subject of study.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004646:8:0", "occurrence_id": "7d3c1414bd0cd75cabffba62", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Here, the question of whether commodity price volatility has increased over time is\n\n\naddressed with analytical rigor and at a disaggregated level. We use a unique, newly\n\n\ncompiled dataset of 45 individual commodity prices and five commodity price indices at\n\n\na monthly frequency. Two econometric tests are applied for identifying structural breaks\n\n\nin GARCH-type processes in order to provide a robustness check and these are described\n\n\nbelow.\n\n##### **3. Data**\n\n\nFor this paper we use a newly compiled dataset which covers 45 individual commodities\n\n\nand 5 commodity price indexes. We use a monthly unbalanced panel of observations\n\n\nfrom Global Financial Data (GFD) covering the period 1784 – 2009. We take a pragmatic\n\n\napproach as to what constitutes a “commodity” and take all “commodities” included in\n\n\nthe GFD dataset as our subject of study. We use data from GFD as it is the most\n\n\ncomprehensive source, verifying that the price data is consistent with that from\n\n\nalternative sources (World Bank, IMF, and UNCTAD) for any overlapping commodity\n\n\nand time period. The details of the dataset can be found in Table 1.\n\n\n7\n\nMention: \"data from GFD\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Chosen for comprehensiveness, with prices checked against other sources.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004647:7:0", "occurrence_id": "a86841d98fe51f42beb97dd4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: regression. So if we …nd a negative e¤ect of land restrictions in OLS regressions that omit land\n\n\nproductivity controls but include human capital controls, this can be taken as particularly strong\n\n\nevidence.\n\n\nThe OLS regressions reported later provide strong evidence of a signi…cant negative e¤ect\n\n\nof land restrictions on equilibrium wage and suggest that the magnitude of this negative e¤ect\n\n\ndecreases with distance. While the OLS estimates may be acceptable as credible evidence of\n\n\na negative e¤ect of land restrictions on equilibrium wage, they cannot give us an estimate of\n\n\nthe magnitude of this e¤ect because of the possible downward bias from omitted productivity\n\n\nheterogeneity. We provide estimates of the causal e¤ect of land restrictions on the wage using an\n\n\ninstrumental variables approach that corrects for the possible downward bias.\n\n\nWhile the unusual positive land productivity correlation in Sri Lanka is useful for our empir\n\nical analysis, there is a second and equally important advantage of Sri Lanka as a case study to\n\n\nunderstand the e¤ects of land restrictions. The history of malaria infestation from 16th to early\n\n\n20th centuries, and the unique role it played in the land policy provides a credible source of exoge\n\nnous variations in the LDO restrictions. The variations in the historical malaria prevalence across\n\n\nDSDs are useful especially because a nationwide malaria eradication program was implemented as\n\n\nearly as 1947 (Brown, 1986, Lucas, 2010), long before the data used for this study were collected\n\n\n(we use HIES 2002). We develop an instrumental variables approach that exploits this exogenous\n\n\nsource of variations to identify and estimate the e¤ect of LDO restrictions on wages. To ensure the\n\n\nvalidity of the exclusion restrictions, we control for a set of covariates including current malaria\n\n\nparasite infection rate, individual and DSD level education, age (to capture cohort e¤ects), and\n\n\ndistrict …xed e¤ects. In a recent paper, Lucas (\n\nMention: \"HIES 2002\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its household data support the instrumental variables design.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004651:14:1", "occurrence_id": "78e85c580a3f05080572f010", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  team of at least three persons. Children are not allowed\n\n\nto bring any materials or stationery (other than pens) into the testing area. The grades selected for\n\n\nthe QAT are tested simultaneously in the same testing area and the seating for the students is\n\n\nmixed. All test materials provided to the test takers are collected back. All unused test materials\n\n\nare retained by the testing agency. ADU also reports that test item recycling across QAT rounds\n\n\nis presently rare given the size of its test item bank.\n\n\nThe QAT serves as the main source of data for the analysis of learning dynamics in\n\n\nprogram schools. While PEF has conducted nine QATs since the inception of the FAS program\n\n\n(QAT 9, the latest QAT, was administered in March 2010), this study uses data from QAT\n\n\n4−QAT 8 which spans November 2007–November 2009, a two-year period, covering, in whole\n\n\nor in part, three academic years. QAT 4 is the first QAT round in the analysis as it represents the\n\n\n13\n\nMention: \"data from QAT\n\n\n4−QAT 8\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Five quarterly rounds spanning this period form the study sample.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004665:37:0", "occurrence_id": "15dec647540d7eb973bfc040", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: where _dij_ is the Euclidean distance between the geographic centers of prefectures _i_ and _j_, _diport_\nis the Euclidean distance between the center of region _i_ and that of the nearest prefecture\nwhich has a major international port, and _area_ is a prefecture's land area (in km <sup>2</sup> ). The data\non land areas is taken from _Regional Economic Statistical Yearbook 2008_ .  Eight major\ninternational ports are considered: Shanghai, Ningbo, Tianjin, Guangzhou, Qingdao City,\nDalian City, Qinhuangdao City and Shenzhen City. These correspond to the eight Chinese\nports which had the largest throughput (measured in million tonnes of traffic) in ISL (2006).\nPre-multiplication of a variable by **W** - **W** represents the second-order spatial lag of that\nvariable.\n\n_Additional parameters used in the numerical solution of the theoretical model:_\n\nAs outlined in section 3.2, to obtain a numerical solution for the system of equations [3] – [7],\nwhich describe the short-run equilibrium of our theoretical model, we also require measures\nfor the parameters , ,  . For 2007, the measures for  and , which correspond to the\nregional shares of the national urban and rural workforces respectively, are constructed using\ndata on \"urban employed persons\" and \"total employed persons\" from the _Regional Economic_\n_Statistical Yearbook 2008_ . For , rural employment is calculated as total employed persons\nminus urban employed persons <sup>35</sup> [^35: In the various provincial statistical yearbooks, the numbers of urban and rural employed persons exactly sum\nto the number of total employed persons at the provincial level.] According to the _China Statistical Yearbook 2008_,\nemployed persons refer to \" _people aged 16 and over who are engaged in gainful employment_\n_and thus receive remuneration payment or earn business income_ \". , the national share of\nurban workers in\n\nMention: \"Regional Economic Statistical Yearbook 2008\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Prefecture land areas are taken from this yearbook.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004670:23:0", "occurrence_id": "7d04a3d9a04bb4c9f1987dbf", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 3: Weights on life expectancy in the old and new HDI**\n\n\n\n.007\n\n\n.006\n\n\n.005\n\n\n.004\n\n\n.003\n\n\n.002\n\n\n.001\n\n\n\n|Col1|Old HDI<br>New HDI|Col3|Col4|Col5|Col6|Col7|Col8|\n|---|---|---|---|---|---|---|---|\n|||||||||\n|||||||||\n|||||||||\n|||||||||\n|||||||||\n|||||||||\n\n\n5 6 7 8 9 10 11 12\n\n\nGross national income ($ per person per year; log scale)\n\n\n\n<u>Source (this figure and all following ones): Author’s calculation from data for 2008 provided in the 2010 HDR. The</u>\nfitted line is a locally smoothed (nonparametric) regression.\n\n\n**Figure 4: Implicit valuations in HDI and marginal costs of an extra year of life expectancy**\n\n\n\n9,000\n\n\n8,000\n\n\n7,000\n\n\n6,000\n\n\n5,000\n\n\n4,000\n\n\n3,000\n\n\n2,000\n\n\n1,000\n\n\n0\n\n\n\n|Col1|Qatar<br>Liechtenstein<br>Implicit valuation<br>in HDI<br>Marginal cost<br>(Dowrick et al.)<br>Zimbabwe|Col3|Col4|Col5|Col6|Col7|Col8|\n|---|---|---|---|---|---|---|---|\n|||||||||\n|||||||||\n\nMention: \"2010 HDR\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Underlying values for that year are extracted from this report.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004671:4:0", "occurrence_id": "db72983fa6bdba56da235232", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Africa, among firms that are credit constrained, and among firms that perceive access to finance as\nan obstacle. Our analysis shows that firm size and age are significantly correlated with firm growth.\nDistinguishing between different types of ownership, we find that firms tend to have higher growth\nif they are an exporter, are part of entities with multiple establishments, are foreign owned, or are\nprivately owned.\n\n\nThe paper is organized as follows. The next section reviews the literature and shows how this paper\nrelates to it. Section 3 presents an overview of the data and describes the sample used. Section 4\nexamines the most binding constraint of the business environment. Section 5 examines the effect of\naccess to finance on employment growth in the full sample and by region, while section 6 looks at\ndeterminants of access to finance. Section 7 investigates differences in the effect of financial access\nvariables on employment growth, and section 8 looks at the relationship between firm size and\nfinancial constraints. The last section concludes.\n\n##### **2. Literature Review**\n\n\nUnderstanding the firm growth process is important for designing appropriate policies for job\ncreation and pro-poor growth. Many studies are devoted to understanding the determinants of firm\ngrowth, especially employment growth. Most of these studies focus on the manufacturing sector and\nlarge firms (Evans 1987a, 1987b; Hall 1987; Dunne, Roberts, and Samuelson 1989). They find that\nfirm age and firm size are important in the analysis of firm growth.\n\n**2.1** **Effect of the Business Environment on Firm Growth**\n\nA number of recent studies use World Business Environment Survey data (a firm-level data set\ncovering 4,000 firms across 54 countries) to study the effect of the business environment on firm\ngrowth. Using subjective, firm-level data on the business environment, some of these studies show\nthe importance of finance, corruption, and property rights (Batra, Kaufmann, and Stone 2003;\nAyyagari\n\nMention: \"World Business Environment Survey data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Four thousand firms across fifty-four countries support the business environment literature.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004676:56:2", "occurrence_id": "4b059a92f57bae0544ee5bf1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <br>1.2<br>1.7<br>2.0<br>2.8<br>**8.6**<br>**12.2**|\n\n\nAuthors’ calculations using China’s Customs Statistics Data\n\n\n\n\n\n55\n\nMention: \"China’s Customs Statistics Data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Export values by category are computed from these customs records.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004649:12:0", "occurrence_id": "6adddcef70ac066a5ad27e92", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: that is statistically valid for rural areas in India.  After data cleaning, a total sample of 22,538 rural\n\n\nhouseholds and 12,325 urban households were included in this paper’s analysis. <sup>5</sup> [^5: Because samples were drawn as a fixed number of 15 households from each block, the analysis throughout the paper\nuses weights for making the analysis representative of the population.]\n\n\nThe energy questions that were asked in the IHDS survey are more extensive than those of\n\n\ncomparable surveys such as the Demographic and Health Surveys and the Living Standards\n\n\nMeasurement Studies. The energy questions in the IHDS survey cover use of fuels, cash\n\n\nexpenditures on fuels, the collection time for obtaining biomass fuels, and the types of stoves and\n\n\nelectric appliances in the household. For electricity, there are also questions on the reliability of the\n\n\npower supply and the source of electricity for the household such as whether they are serviced by\n\n\nthe local state electricity board or from a neighbor. These detailed energy questions allow for a\n\n\nmuch more detailed analysis of the welfare impact of the use of different types of energy. In fact, we\n\n\nnext turn to the demand for energy in India.\n\n\n**Patterns of Household Energy Use**\n\n\nOf the many factors that have constrained India’s development, energy usage figures prominently.\n\n\nEnergy use per capita in India lags far behind that of other countries around the world (World Bank\n\n\n2006). Moreover, there are wide disparities in energy use between urban and rural areas (where\n\n\nthree-fourths of the country’s people live). This section highlights India’s existing patterns of\n\n\nhousehold energy use, and estimates energy demand functions.\n\n\nThe patterns of urban and rural energy demand in India differ markedly (Table 1). In rural\n\n\nareas, patterns of energy use typically involve high reliance on traditional fuels, including wood,\n\n\ndung, and straw burned in inefficient stoves. In urban areas, only one-third of households use\n\n\nfuelwood, compared to nine-tenths of rural households.\n\nMention: \"IHDS survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its energy module is broader than comparable survey questionnaires.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004649:12:0", "occurrence_id": "85d19b6b05c22981112408a7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: that is statistically valid for rural areas in India.  After data cleaning, a total sample of 22,538 rural\n\n\nhouseholds and 12,325 urban households were included in this paper’s analysis. <sup>5</sup> [^5: Because samples were drawn as a fixed number of 15 households from each block, the analysis throughout the paper\nuses weights for making the analysis representative of the population.]\n\n\nThe energy questions that were asked in the IHDS survey are more extensive than those of\n\n\ncomparable surveys such as the Demographic and Health Surveys and the Living Standards\n\n\nMeasurement Studies. The energy questions in the IHDS survey cover use of fuels, cash\n\n\nexpenditures on fuels, the collection time for obtaining biomass fuels, and the types of stoves and\n\n\nelectric appliances in the household. For electricity, there are also questions on the reliability of the\n\n\npower supply and the source of electricity for the household such as whether they are serviced by\n\n\nthe local state electricity board or from a neighbor. These detailed energy questions allow for a\n\n\nmuch more detailed analysis of the welfare impact of the use of different types of energy. In fact, we\n\n\nnext turn to the demand for energy in India.\n\n\n**Patterns of Household Energy Use**\n\n\nOf the many factors that have constrained India’s development, energy usage figures prominently.\n\n\nEnergy use per capita in India lags far behind that of other countries around the world (World Bank\n\n\n2006). Moreover, there are wide disparities in energy use between urban and rural areas (where\n\n\nthree-fourths of the country’s people live). This section highlights India’s existing patterns of\n\n\nhousehold energy use, and estimates energy demand functions.\n\n\nThe patterns of urban and rural energy demand in India differ markedly (Table 1). In rural\n\n\nareas, patterns of energy use typically involve high reliance on traditional fuels, including wood,\n\n\ndung, and straw burned in inefficient stoves. In urban areas, only one-third of households use\n\n\nfuelwood, compared to nine-tenths of rural households.\n\nMention: \"Demographic and Health Surveys\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Their narrower energy coverage is the comparison point.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004649:12:0", "occurrence_id": "1418c061eb7ce4eca5e37457", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: that is statistically valid for rural areas in India.  After data cleaning, a total sample of 22,538 rural\n\n\nhouseholds and 12,325 urban households were included in this paper’s analysis. <sup>5</sup> [^5: Because samples were drawn as a fixed number of 15 households from each block, the analysis throughout the paper\nuses weights for making the analysis representative of the population.]\n\n\nThe energy questions that were asked in the IHDS survey are more extensive than those of\n\n\ncomparable surveys such as the Demographic and Health Surveys and the Living Standards\n\n\nMeasurement Studies. The energy questions in the IHDS survey cover use of fuels, cash\n\n\nexpenditures on fuels, the collection time for obtaining biomass fuels, and the types of stoves and\n\n\nelectric appliances in the household. For electricity, there are also questions on the reliability of the\n\n\npower supply and the source of electricity for the household such as whether they are serviced by\n\n\nthe local state electricity board or from a neighbor. These detailed energy questions allow for a\n\n\nmuch more detailed analysis of the welfare impact of the use of different types of energy. In fact, we\n\n\nnext turn to the demand for energy in India.\n\n\n**Patterns of Household Energy Use**\n\n\nOf the many factors that have constrained India’s development, energy usage figures prominently.\n\n\nEnergy use per capita in India lags far behind that of other countries around the world (World Bank\n\n\n2006). Moreover, there are wide disparities in energy use between urban and rural areas (where\n\n\nthree-fourths of the country’s people live). This section highlights India’s existing patterns of\n\n\nhousehold energy use, and estimates energy demand functions.\n\n\nThe patterns of urban and rural energy demand in India differ markedly (Table 1). In rural\n\n\nareas, patterns of energy use typically involve high reliance on traditional fuels, including wood,\n\n\ndung, and straw burned in inefficient stoves. In urban areas, only one-third of households use\n\n\nfuelwood, compared to nine-tenths of rural households.\n\nMention: \"Living Standards\n\n\nMeasurement Studies\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"A second comparator with less detailed energy questions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004677:5:0", "occurrence_id": "529cf334a6368917c6719f9e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 4\n\n\nof people exposed to drought. The regular occurrence of natural disasters together with the\navailability of three highly comparable nationally representative household living standard surveys\n(2002‐2004‐2006) render Vietnam particularly suitable to illustrate the approach advanced in this\npaper. The study constructs in particular disaggregated maps of the annual occurrence of droughts,\n(local and riverine) floods and storms during 2001‐2006 as well as the likelihood of their occurrence\nusing meteorological data from rainfall stations, storm tracks, and satellite images. The occurrence\nof these geo‐referenced disasters, further augmented with the likelihood of their occurrence derived\nfrom the historical weather records, is subsequently linked with household consumption and other\nhousehold and location characteristics from the repeated nationally representative household\nsurveys and secondary data sources to estimate the effect of extreme weather events on household\nwelfare.\n\n\nThe results suggest that the immediate losses from floods and hurricanes can be substantial,\nwith hurricanes causing most havoc (up to 52 percent consumption loss among households close to\nlarge urban centers). Households tend to cope well with droughts, largely through irrigation.\nFrequent exposure to disasters erodes the standard of living, but reduces the immediate effects of\nshocks as households become less exposed and better prepared. Households in frequently\ninundated areas have even been able to turn the floods into an advantage, as long as the flooding is\nnot too severe. There is however no adaptation to hurricanes, rather the contrary, with high\nfrequency of hurricanes exacerbating the losses from particular events. Finally, those further away\nfrom the large urban centers are not only poorer, but also tend to suffer less from disasters, likely\ndue to the adoption of less risky (but less remunerative) portfolios and a higher likelihood of\nreceiving disaster relief.\n\n\nThe next section outlines the approach used to obtain geographically disaggregated estimates\nof droughts, heavy rainfall, river floods and storms with hurricane force. Section three reviews the\nempirical methodology to estimate the welfare effects. Section four reviews the data used and\nsection five presents the empirical findings regarding the\n\nMention: \"nationally representative household living standard surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Three highly comparable rounds make the country suitable for the approach.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004677:5:0", "occurrence_id": "3a745eaa06a70dbb003b0b7e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 4\n\n\nof people exposed to drought. The regular occurrence of natural disasters together with the\navailability of three highly comparable nationally representative household living standard surveys\n(2002‐2004‐2006) render Vietnam particularly suitable to illustrate the approach advanced in this\npaper. The study constructs in particular disaggregated maps of the annual occurrence of droughts,\n(local and riverine) floods and storms during 2001‐2006 as well as the likelihood of their occurrence\nusing meteorological data from rainfall stations, storm tracks, and satellite images. The occurrence\nof these geo‐referenced disasters, further augmented with the likelihood of their occurrence derived\nfrom the historical weather records, is subsequently linked with household consumption and other\nhousehold and location characteristics from the repeated nationally representative household\nsurveys and secondary data sources to estimate the effect of extreme weather events on household\nwelfare.\n\n\nThe results suggest that the immediate losses from floods and hurricanes can be substantial,\nwith hurricanes causing most havoc (up to 52 percent consumption loss among households close to\nlarge urban centers). Households tend to cope well with droughts, largely through irrigation.\nFrequent exposure to disasters erodes the standard of living, but reduces the immediate effects of\nshocks as households become less exposed and better prepared. Households in frequently\ninundated areas have even been able to turn the floods into an advantage, as long as the flooding is\nnot too severe. There is however no adaptation to hurricanes, rather the contrary, with high\nfrequency of hurricanes exacerbating the losses from particular events. Finally, those further away\nfrom the large urban centers are not only poorer, but also tend to suffer less from disasters, likely\ndue to the adoption of less risky (but less remunerative) portfolios and a higher likelihood of\nreceiving disaster relief.\n\n\nThe next section outlines the approach used to obtain geographically disaggregated estimates\nof droughts, heavy rainfall, river floods and storms with hurricane force. Section three reviews the\nempirical methodology to estimate the welfare effects. Section four reviews the data used and\nsection five presents the empirical findings regarding the\n\nMention: \"meteorological data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Station rainfall, storm tracks and satellite imagery give flood and storm likelihood.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004677:5:0", "occurrence_id": "40cce1c8b01afbe5b190e6e0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 4\n\n\nof people exposed to drought. The regular occurrence of natural disasters together with the\navailability of three highly comparable nationally representative household living standard surveys\n(2002‐2004‐2006) render Vietnam particularly suitable to illustrate the approach advanced in this\npaper. The study constructs in particular disaggregated maps of the annual occurrence of droughts,\n(local and riverine) floods and storms during 2001‐2006 as well as the likelihood of their occurrence\nusing meteorological data from rainfall stations, storm tracks, and satellite images. The occurrence\nof these geo‐referenced disasters, further augmented with the likelihood of their occurrence derived\nfrom the historical weather records, is subsequently linked with household consumption and other\nhousehold and location characteristics from the repeated nationally representative household\nsurveys and secondary data sources to estimate the effect of extreme weather events on household\nwelfare.\n\n\nThe results suggest that the immediate losses from floods and hurricanes can be substantial,\nwith hurricanes causing most havoc (up to 52 percent consumption loss among households close to\nlarge urban centers). Households tend to cope well with droughts, largely through irrigation.\nFrequent exposure to disasters erodes the standard of living, but reduces the immediate effects of\nshocks as households become less exposed and better prepared. Households in frequently\ninundated areas have even been able to turn the floods into an advantage, as long as the flooding is\nnot too severe. There is however no adaptation to hurricanes, rather the contrary, with high\nfrequency of hurricanes exacerbating the losses from particular events. Finally, those further away\nfrom the large urban centers are not only poorer, but also tend to suffer less from disasters, likely\ndue to the adoption of less risky (but less remunerative) portfolios and a higher likelihood of\nreceiving disaster relief.\n\n\nThe next section outlines the approach used to obtain geographically disaggregated estimates\nof droughts, heavy rainfall, river floods and storms with hurricane force. Section three reviews the\nempirical methodology to estimate the welfare effects. Section four reviews the data used and\nsection five presents the empirical findings regarding the\n\nMention: \"historical weather records\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"They augment the geo-referenced disaster record.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004687:24:0", "occurrence_id": "e4c0c91d00420095ced2003c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Implications for Poverty Analysis_\n\n\nThe fact that both mean consumption and distributional measures vary substantially across\n\n\nthe modules implies that there will be complex comparability problems when poverty estimates\n\n\nfrom different modes of consumption measurement are combined. For example, consider a poverty\n\n\nmonitoring survey that switched from 14-day to 7-day recall. This switch would raise reported\n\n\nmean consumption (see Table 3) and lower measured inequality (see Table 7). In a setting where\n\n\nthe poverty line is set below median consumption, these effects will work in concert to make\n\n\npoverty look lower than it would if the method of consumption measurement had not changed (as\n\n\noccurred with changes in India’s NSS). Conversely, if a switch was made from a disaggregated\n\nrecall list to a collapsed list (as the Indonesian _Susenas_ does every 3 <sup>rd</sup> year), it would lower the\n\n\nmean (implying higher measured poverty) and lower inequality (lower measured poverty) with\n\n\nambiguous implications for poverty measures. Hence there are unlikely to be simple correction\n\n\nfactors that can be used to enforce comparability on different consumption modules when\n\n\nattempting to measure poverty consistently over either time or space. Similarly, the literature that\n\n\nreacts to possible survey understatement by combining survey measures of inequality with national\n\n\naccounts measures of mean consumption (such as Bhalla, 2000, and Sala-i-Martin and Pinkovskiy,\n\n\n2010) is likely to be misguided since survey design changes will also affect estimated inequality.\n\n\nWe attempt to quantify the implications of survey design for poverty analysis in a setting\n\n\nlike Tanzania by presenting aggregate poverty numbers based on an international poverty line, as\n\n\nwell as the characteristics of those households deemed poor by each module. Figure 1a presents the\n\n\npoverty headcount rate based on a common international poverty line of $1.25/person/day\n\nconverted into 2008 Tanzanian shillings on a PPP basis. <sup>25</sup> With this absolute poverty line\n\nMention: \"NSS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its change in consumption measurement method is the cautionary case.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004689:27:1", "occurrence_id": "c241caae6390cf789bb9981f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 40<br>11%<br>38%<br>70%<br>56%<br>30<br>47<br>31<br>48<br>777<br>-|8%<br>9%<br>- <br>-<br>59%<br>64%<br>70%<br>54%<br>3%<br>17%<br>13%<br>15%<br>60%<br>42%<br>3.6<br>5.55<br>3.45<br>5.61<br>32<br>33<br>7%<br>27%<br>74%<br>59%<br>32<br>30<br>33<br>31<br>458<br>1469|\n\n\nSource: Authors’ analysis of Haiti 2003 school census data.\n\n\n**Table 2: School Attendance Rates**\n\n\n% of children ages 6-14 attending school\n<u>1994/95</u> <u>2000</u> <u>2005</u>\n_Panel A: School attendance, by area of residence_\nRural 67% 83% 80%\n<u>Urban</u> <u>86%</u> <u>93%</u> <u>91%</u>\n_Panel B: School attendance, by gender_\nGirls 74% 87% 85%\n<u>Boys</u> <u>74%</u> <u>85%</u> <u>82%</u>\n<u>All children 6-14</u> <u>74%</u> <u>86%</u> <u\n\nMention: \"Haiti 2003 school census data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"School characteristics and attendance are analysed from this enumeration.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004697:28:2", "occurrence_id": "aec21c144ed0c32b31bc2864", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  trend NRA) where national trend NRA in each of the two sub-periods is\n\n\nobtained by ordinary least squares linear regression of the national NRA on time. Estimates\n\n\nshown are an unweighted average of national NRA deviations each year, averaged over the\n\n\nnumber of years in each period.\n\n\nSource: Authors’ compilation based on NRAs from Anderson and Valenzuela (2008).\n\nMention: \"Anderson and Valenzuela (2008)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Nominal rate assistance estimates averaged by period come from this compilation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004700:23:0", "occurrence_id": "36fe1973d5f3c414f3fee05c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <u>Table 5. Fee Structure in Selected Countries</u>\n\n\n\nFixed\n<u>commission</u>\n\n\n\nFees on\n<u>contributions</u>\n\n\n\nFees on\n\n\n\n<u>assets</u>\n\n\n\nFees on\n\n\n\nLimits on\n\n\n\n<u>commission</u> <u>contributions</u> <u>assets</u> <u>return</u> <u>fees</u>\n\nBolivia x x x\nColumbia x x\nChile x\nCosta Rica x x\nEl Salvador x x\nMexico x\nPeru x\nUruguay x x\n\nBulgaria x x x\nEstonia x x\nHungary x x x\nKazakhstan x x x\nLatvia x x\nPoland x x x x\nSlovakia x x x\n\nAustralia x x x\n<u>Sweden</u> <u>x</u> <u>x</u>\nSource: OECD, World Bank\n\nAs shown in Figure 2, a comparison among the fees charged by the different pension funds shows that at\nleast for the years 2006 and 2007 the fees charged by the PFR-VEB are considerably lower than the fees\ncharged by the NPFs. There is no correlation between the fees charged and the size of the funds.\n\nFigure 2 Fees charged by second pillar pension funds\n(% Assets)\n\n\n\n<u>return</u>\n\n\n\nSource FFMS\n\n\n\n22\n\nMention: \"FFMS\n\n\n\n22\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Pension fund fees as a share of assets are plotted from this regulator.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004706:22:0", "occurrence_id": "397912fb344a8e6715ec1752", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: - 21 \n\nsignificantly, as shown in Table 1. Though, the appliance ownership is still an important\n\n\ndeterminant of the residential demand for electricity. Thus, a number of dummy variables for\n\n\nvarious household appliances are included in _X_ 1 as independent variables.\n\n\n**Alternative energy choice.** Fourth, if there are any important alternative energy sources,\n\n\nsuch as liquefied petroleum gas (LPG) and firewood, those substitutes should be incorporated\n\n\nin the estimation model. Ideally, with sufficient data the model should be estimated as a\n\n\nsystem of multiple demand functions (e.g., Nauges and van den Berg, 2009). However, data\n\n\nare not always available. Even in a single equation model, the demand function can also\n\n\ninvolve prices of alternative sources (Filippini and Pachauri, 2004). An alternative approach\n\n\nmay be to introduce another equation for deciding whether or not to connect the service\n\n\n(Basani _et al_ ., 2008). Behind this approach, it is implicitly assumed that some alternatives\n\n\nexist. Given this specification, the conventional Hausman self-selection model can be used\n\n\nfor estimation.\n\n\nIn the case of Albania, there are few important alternative energy choices for households.\n\n\nThe country has already achieved almost universal access to electricity network, though the\n\n\nquality of supply still remains problematic. The access rate was 99.97 percent in 2005.\n\n\nPeople may use LPG for heating and especially for cooking, but the use of gas may be\n\nrelatively limited, in particular in urban areas. <sup>14</sup> By contrast, many firms are equipped with\n\ntheir own private backup facilities. <sup>15</sup> Therefore, no alternative choice is explicitly considered\n\n\nin our empirical model. The paper takes into account whether households are using electricity\n\n\nor LPG for major household activities, such as heating and cooking. This is expected to\n\n\n14 There is some difference between the LSMS data and people’s perception\n\nMention: \"the LSMS data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Measured energy use diverges from what people report.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004712:30:1", "occurrence_id": "05e7eaf2ce1d7d9e69889490", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: u> <u>Labor market shocks</u> <u>General economic distress</u> <u>Mobility intention</u> <u>Migration</u>\n\n\n\nDo your military\nservice (all years)\n\n\nGet married (live in\ncouple)\n\n\nHave a child (birth\nor adoption)\n\n\n\nReceive unemployment\nbenefits\n\n\nHave to do a job below\nyour qualifications\n\n\nHave to accept wage\ncuts/arrears\n\n\n\nTurn to relatives for\nfinancial assistance\n\n\n\nHave to move in with\nFind a better job\nrelatives\n\n\nHave to sell some of your\nhousehold assets\n\n\n\nWork for wages\n(for an employer)\n\n\n\nMove from\nurban to rural\narea\n\nMove from\nrural to urban\narea\n\n\n29\n\n\n\nHave to cut down on\nDivorce\nbasic food consumption\n\nOfficially retire\nDecided not to\n<u>work</u>\n\nSource: Life in Transition Survey data (Authors’ own calculation)\n\nMention: \"Life in Transition Survey data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Household coping strategies such as cutting food consumption are tabulated here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004713:29:0", "occurrence_id": "b41133407f4091ed522e0547", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 1: Variance of Fourth Graders’ Age against GDP per Capita (PPP)**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nNotes: Data sourced from Third International Mathematics and Science Study (TIMSS) 2007 and the\nWorld Development Indicators. 37 economies are included in the sample. The variance of age for United\nKingdom is calculated based on the weighted average of the figures of England and Scotland. The\nvariance of age for Canada is calculated by the weighted average of Alberta, British Columbia, Quebec,\nand Toronto provinces. GDP per capita (PPP) in 2003 is selected so that income matches the time that the\nfourth grade cohort in TIMSS 2007 commenced primary education.\n\n\n**Figure 2: Variance of Age against Average Test Score of Fourth Graders**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nNotes: Data sourced from TIMSS 2007 and the World Development Indicators. 37 economies are\nincluded in the sample. The test score for United Kingdom is calculated based on the weighted average of\nthe figures of England and Scotland. The test score for Canada is calculated by the weighted average of\nAlberta, British Columbia, Quebec, and Toronto provinces. GDP per capita (PPP) in 2003 is selected so\nthat income matches the time that the fourth grade cohort in TIMSS 2007 commenced primary education.\nAverage standardized score is the average of the standardized international scale mathematics and science\nscores.\n\n\n28\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Income per head for the cohort's schooling entry comes from this source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004713:29:0", "occurrence_id": "6e0d1eb84d9a51bfc741732b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 1: Variance of Fourth Graders’ Age against GDP per Capita (PPP)**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nNotes: Data sourced from Third International Mathematics and Science Study (TIMSS) 2007 and the\nWorld Development Indicators. 37 economies are included in the sample. The variance of age for United\nKingdom is calculated based on the weighted average of the figures of England and Scotland. The\nvariance of age for Canada is calculated by the weighted average of Alberta, British Columbia, Quebec,\nand Toronto provinces. GDP per capita (PPP) in 2003 is selected so that income matches the time that the\nfourth grade cohort in TIMSS 2007 commenced primary education.\n\n\n**Figure 2: Variance of Age against Average Test Score of Fourth Graders**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nNotes: Data sourced from TIMSS 2007 and the World Development Indicators. 37 economies are\nincluded in the sample. The test score for United Kingdom is calculated based on the weighted average of\nthe figures of England and Scotland. The test score for Canada is calculated by the weighted average of\nAlberta, British Columbia, Quebec, and Toronto provinces. GDP per capita (PPP) in 2003 is selected so\nthat income matches the time that the fourth grade cohort in TIMSS 2007 commenced primary education.\nAverage standardized score is the average of the standardized international scale mathematics and science\nscores.\n\n\n28\n\nMention: \"TIMSS 2007\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Fourth-grade test scores and age variance come from this assessment.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004704:36:0", "occurrence_id": "285b0e5a21696c3429431e10", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Table 1: Baseline characteristics of intervention and control districts, school feeding program impact assessment, northern Lao PDR, 2006.\n\n\n\nNo. of\n<u>villages</u>\n\n\n\nPoverty rate,\nrural\n<u>households</u> <u>Percent of villages with</u> <u>Vulnerability*</u>\n\n\n\n<u>District</u> <u>Province</u> <u>Type</u>\n\n\nPhongsaly Phongsaly Intervention\n\n\nKhua Phongsaly Intervention\n\n\nNhot ou Phongsaly Intervention\n\n\n\nNo. of\n<u>households</u>\n\n\n\n5,682 92 0.36 18% 70% 12% 72% Very vulnerable\n\n\n4,858 113 0.40 10% 80% 11% 73% Very vulnerable\n\n\n4,514 91 0.32 7% 70% 23% 65% Very vulnerable\n\n\n\nno road\n\n<u>access</u>\n\n\n\nno\nprimary\n\n<u>school</u>\n\n\n\ncomplete\n\nprimary\n\n<u>school</u>\n\n\n\nincomplete\n\nprimary\n\n<u>school</u>\n\n\n\n<u>Gnoi</u> <u>Luang Prabang</u> <u>Control</u> <u>6,931</u> <u>112</u> <u>0.42</u> <u>25%</u> <u>60%</u> <u>15%</u> <u>75%</u> <u>Vulnerable</u>\n\nSource: 2005 Census, Lao PDR; Lao Expenditure and Consumption Survey 2002-03; WFP 2005 Vulnerability Analysis. * Vulnerability score is from the WFP\n2005 Vulnerability Analysis.\n\nMention: \"2005 Census\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Population and household characteristics come from this enumeration.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004704:36:0", "occurrence_id": "cc84cf1b78835d7b92750e0f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Table 1: Baseline characteristics of intervention and control districts, school feeding program impact assessment, northern Lao PDR, 2006.\n\n\n\nNo. of\n<u>villages</u>\n\n\n\nPoverty rate,\nrural\n<u>households</u> <u>Percent of villages with</u> <u>Vulnerability*</u>\n\n\n\n<u>District</u> <u>Province</u> <u>Type</u>\n\n\nPhongsaly Phongsaly Intervention\n\n\nKhua Phongsaly Intervention\n\n\nNhot ou Phongsaly Intervention\n\n\n\nNo. of\n<u>households</u>\n\n\n\n5,682 92 0.36 18% 70% 12% 72% Very vulnerable\n\n\n4,858 113 0.40 10% 80% 11% 73% Very vulnerable\n\n\n4,514 91 0.32 7% 70% 23% 65% Very vulnerable\n\n\n\nno road\n\n<u>access</u>\n\n\n\nno\nprimary\n\n<u>school</u>\n\n\n\ncomplete\n\nprimary\n\n<u>school</u>\n\n\n\nincomplete\n\nprimary\n\n<u>school</u>\n\n\n\n<u>Gnoi</u> <u>Luang Prabang</u> <u>Control</u> <u>6,931</u> <u>112</u> <u>0.42</u> <u>25%</u> <u>60%</u> <u>15%</u> <u>75%</u> <u>Vulnerable</u>\n\nSource: 2005 Census, Lao PDR; Lao Expenditure and Consumption Survey 2002-03; WFP 2005 Vulnerability Analysis. * Vulnerability score is from the WFP\n2005 Vulnerability Analysis.\n\nMention: \"Lao Expenditure and Consumption Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Consumption levels underpin the vulnerability categories.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004714:22:0", "occurrence_id": "7d6a389f25965fd1d4e36348", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **<u>Table 1. Structure of the SABESP Dataset</u>**\n\n**SABESP Dataset 1: Full Sample**\n\n\n\n\n\n\n\n\n\n\n\n\n\n|Field|Description|\n|---|---|\n|Procurement method<br>indicator<br>|CP for_Concorrências_, TP for_Tomada de Preços_|\n|Bidding process<br>identification number<br>|Unique ID assigned to each bidding process|\n|Works description|Summary description of the nature of the works procured<br>|\n|Segment|Works projects are categorized by segment, based on<br>activity type. There are 16 project segments, including<br>water plants, water pipes, connections, and sewage plants<br>|\n|Date of opening of the first<br>envelope|For 2007 (prior to the reform), the first envelope contains<br>the technical, financial, legal, and fiscal qualification<br>documents. For 2009 (post reform), the first envelope<br>corresponds to the price proposal<br>|\n|Contract signature date|Date contract was signed<br>|\n|Estimated cost|Officially estimated project cost, as prepared by SABESP,<br>in Brazilian currency<br>|\n|Contract price|Actual size of the contract awarded as a result of the<br>procurement process, in Brazilian currency<br>|\n\n\nPage | 21\n\nMention: \"SABESP Dataset\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Procurement records for the water utility are structured in this dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004716:6:0", "occurrence_id": "5af6382659524de6388ebce2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: outpatient as well as inpatient care in their RCMS schemes (19), generating incentives to\nproviders to admit patients for ailments that could be treated more cheaply at home.\n\nOut-of-pocket payments as a share of medical bills remain significantly higher for outpatient\ncare compared to inpatient care in both urban and rural areas. The most recent National Health\nServices Survey (NHSS) (11) shows that in 2008, 33% of patients received partial\nreimbursement for outpatient care compared to 85% who received partial reimbursement (in the\naverage amount of 35% of the medical bills) for inpatient care. This is confirmed by field\nresearch. Citizens’ scorecard survey shows that out-of-pocket payments as a share of medical\nbills remain significantly higher for outpatient care compared to inpatient care, and reach 84% of\n_monthly_ per capita income among patients in the poorest quintile (compared to about 11% in the\nrichest quintile) for an average outpatient visit, and a staggering 140% of _annual_ per capita\nincome in the poorest quintile (compared to about 10% in the richest quintile) for a single\naverage hospitalization episode. (20)\n\n_Earmarking of vertical programs to promote equity may also be failing in some areas_\nTo implement national priorities, the central government has been circumventing the problem of\nresource allocation at local level by increasingly relying on vertical programs and earmarked\nspecific purpose transfers (that are conditional, to incentivize sub-national governments, as\nopposed to general-purpose transfers that are unconditional to provide general budget support).\nAs part of China’s HSR, six new public health initiatives (for hepatitis B vaccination of older\nchildren, folic acid supplementation, fluorosis prevention, cervical and breast cancer screening\nfor rural women, cataract treatment, and rural water and sanitation initiatives) have been recently\nadded to the hundreds of earmarked transfers already in existence in the health sector. Currently,\nthe entire central government budget for the health sector consists of earmarked transfers\n(accounting for\n\nMention: \"National Health\nServices Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Reimbursement shares for outpatient and inpatient care come from this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004716:6:0", "occurrence_id": "13a7881fe7dc824136889e21", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: outpatient as well as inpatient care in their RCMS schemes (19), generating incentives to\nproviders to admit patients for ailments that could be treated more cheaply at home.\n\nOut-of-pocket payments as a share of medical bills remain significantly higher for outpatient\ncare compared to inpatient care in both urban and rural areas. The most recent National Health\nServices Survey (NHSS) (11) shows that in 2008, 33% of patients received partial\nreimbursement for outpatient care compared to 85% who received partial reimbursement (in the\naverage amount of 35% of the medical bills) for inpatient care. This is confirmed by field\nresearch. Citizens’ scorecard survey shows that out-of-pocket payments as a share of medical\nbills remain significantly higher for outpatient care compared to inpatient care, and reach 84% of\n_monthly_ per capita income among patients in the poorest quintile (compared to about 11% in the\nrichest quintile) for an average outpatient visit, and a staggering 140% of _annual_ per capita\nincome in the poorest quintile (compared to about 10% in the richest quintile) for a single\naverage hospitalization episode. (20)\n\n_Earmarking of vertical programs to promote equity may also be failing in some areas_\nTo implement national priorities, the central government has been circumventing the problem of\nresource allocation at local level by increasingly relying on vertical programs and earmarked\nspecific purpose transfers (that are conditional, to incentivize sub-national governments, as\nopposed to general-purpose transfers that are unconditional to provide general budget support).\nAs part of China’s HSR, six new public health initiatives (for hepatitis B vaccination of older\nchildren, folic acid supplementation, fluorosis prevention, cervical and breast cancer screening\nfor rural women, cataract treatment, and rural water and sanitation initiatives) have been recently\nadded to the hundreds of earmarked transfers already in existence in the health sector. Currently,\nthe entire central government budget for the health sector consists of earmarked transfers\n(accounting for\n\nMention: \"Citizens’ scorecard survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Field research confirms higher out-of-pocket shares through user reports.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004720:3:1", "occurrence_id": "9000d837f3c9de662b33a542", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  of education (see Basu and Van [6] for example), therefore jeopardizing\nhuman capital growth. Yet, as emphasized by Baland and Robinson [4], the real\nissue is to better understand the determinants of child labor so as to evaluate its\nwelfare implications. More generally, it is crucial to jointly investigate the factors\ndriving schooling and child labor decisions.\n\nThis paper aims at analyzing the determinants of primary school\nenrollment, attendance and child labor in Bolivia from 1999 to 2007, identifying\nhow the substitution and complementary relationships among such activities\nevolve over time.\n\nThe unprecedented use of Bolivia’s national household survey MECOVI\nfor several years allows for an in-depth historical analysis of the recent trends of\nschooling and child labor. Due to the lack of empirical literature on this specific\nissue for Bolivia, this study represents a contribution that aims at filling the gap.\n\nResults at the descriptive level reveal that enrollment became\nprogressively more widespread in Bolivia. Nonetheless, the attendance figures are\ndiscouraging, as about 40 percent of the enrolled children did not go to school.\n\nTriprobit estimations show that the increase in enrollment is led by\nindigenous and children living in urban areas, whereas poverty and indigenous are\n\n\n1 GDP data are from the United Nations National Accounts Main Aggregate Database, data on\npoverty are from MECOVI 2007.\n2 More specifically, this represents the second Millennium Development Goal (MDG) as\nestablished by the United Nations.\n3 According to the MECOVI surveys employed, more than 50 percent of the total Bolivian\npopulation declare themselves belonging to indigenous groups.\n\n\n2\n\nMention: \"MECOVI\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its repeated use allows historical analysis of schooling trends.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004723:4:0", "occurrence_id": "e6cb3c9698ec656aecdb6a66", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Access 2009_ introduced statistics on the use of financial services in 139 countries and mapped a\nbroad range of policies and initiatives supporting financial inclusion, focusing on data collection\nand measurement challenges. Building on this, _Financial Access 2010_ updated statistics on the\nuse of financial services and analyzed changes that took place in 2009 – a turbulent year for the\nfinancial sector in most countries around the world. As more countries provided more and higher\nquality data after the 2009 report, the focus shifted to the analysis of financial inclusion policies\naround the world (see World Bank, 2008; CGAP and the World Bank, 2009 and 2010).\n\nThis paper uses the _Financial Access_ database to count the number of unbanked individuals in\nthe world, and analyze the change of access to formal financial services around the world. Next,\nit reviews the role of the main financial regulator in relation to the implementation of financial\ninclusion reforms and mandates to give a broad global picture of the state of financial inclusion\npolicies.\n\nOur findings indicate that 56 percent of adults worldwide are unbanked, although the numbers\ndiffer across high-income and developing countries (17 percent and 64 percent, respectively).\nOverall, we predict a slight improvement in access to financial services in 2009: approximately\n50 million more adults have accounts, although the adult population increased by approximately\n79 million. The results of the econometric analysis confirm the earlier findings in the literature\nthat higher deposit and/or loan correlation is associated with higher economic and financial\ndevelopment as measured by GDP per capita, the amount of electricity use, the availability of\nexplicit deposit insurance and better credit environment, etc. Although access to deposit services\nhas improved in 2009, the global financial crisis took its toll: volume of deposits and loans\nshrank. The world as a whole added 65 deposit accounts per 1,000 adults but the number of\noutstanding loans remained more or less the same. <sup>2</sup> At the same time, global retail networks\n\nMention: \"_Financial Access_ database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"It counts the unbanked worldwide and tracks changes in access.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004735:17:0", "occurrence_id": "69015541fcffc8bb59d2dcc3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: years of schooling), the education level of the head‘s parents, and the head‘s religion and\n\n\nethnicity.\n\n2. We then introduce locational dummies such as urban/rural, or regional, dummies to\n\nmeasure where the household was living at the time of the first round survey. Most\n\n\nmultipurpose surveys with a migration module would collect the information needed to\n\n\nallow these variables to be constructed, and even without a specific migration module, it\n\nis common to ask where households were living five years ago. <sup>12</sup> [^12: For example, Smith and Thomas (2003) find that Malaysian households can accurately recall migration histories,\nparticularly for moves which are not very local or very short in duration.]\n\n3. Next, ―community‖ variables are added, which can be obtained from community modules\n\nin most household surveys or perhaps population censuses. Once the retrospective\n\n\nlocation is identified (as per model 2), the use of such variables depends only on the\n\n\navailability of such auxiliary data, and not on further recall per se. In the case of\n\n\nIndonesia, these come from the community-level survey from 1997 and are inserted into\n\n\nboth the IFLS2 and IFLS3 household surveys. For Vietnam, unfortunately the community\n\n\nmodule only collects data on rural communes, which can reduce the estimation sample\n\n\nsize significantly. Thus we will use instead a household-level variable which indicates\n\n\nhousehold poverty status as classified by the government in the first survey round.\n\n4. We then add variables describing a household head‘s sector of work. At this point we\n\nclearly start to lean more heavily on our ability to explicitly insert round 1 values of these\n\n\nvariables into the round 2 data. However, information on these variables could probably\n\n\nbe easily collected on a retrospective basis. Indeed retrospective work histories have been\n\n\ncollected in a number of labor surveys.\n\n5. Further demographic variables that we force to be time-invariant are then added - such as\n\nhousehold size and the number of children aged under 5. These would\n\nMention: \"community-level survey from 1997\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its auxiliary infrastructure data are inserted into two household rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004735:17:0", "occurrence_id": "a4419454c23dfa1496c03476", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: years of schooling), the education level of the head‘s parents, and the head‘s religion and\n\n\nethnicity.\n\n2. We then introduce locational dummies such as urban/rural, or regional, dummies to\n\nmeasure where the household was living at the time of the first round survey. Most\n\n\nmultipurpose surveys with a migration module would collect the information needed to\n\n\nallow these variables to be constructed, and even without a specific migration module, it\n\nis common to ask where households were living five years ago. <sup>12</sup> [^12: For example, Smith and Thomas (2003) find that Malaysian households can accurately recall migration histories,\nparticularly for moves which are not very local or very short in duration.]\n\n3. Next, ―community‖ variables are added, which can be obtained from community modules\n\nin most household surveys or perhaps population censuses. Once the retrospective\n\n\nlocation is identified (as per model 2), the use of such variables depends only on the\n\n\navailability of such auxiliary data, and not on further recall per se. In the case of\n\n\nIndonesia, these come from the community-level survey from 1997 and are inserted into\n\n\nboth the IFLS2 and IFLS3 household surveys. For Vietnam, unfortunately the community\n\n\nmodule only collects data on rural communes, which can reduce the estimation sample\n\n\nsize significantly. Thus we will use instead a household-level variable which indicates\n\n\nhousehold poverty status as classified by the government in the first survey round.\n\n4. We then add variables describing a household head‘s sector of work. At this point we\n\nclearly start to lean more heavily on our ability to explicitly insert round 1 values of these\n\n\nvariables into the round 2 data. However, information on these variables could probably\n\n\nbe easily collected on a retrospective basis. Indeed retrospective work histories have been\n\n\ncollected in a number of labor surveys.\n\n5. Further demographic variables that we force to be time-invariant are then added - such as\n\nhousehold size and the number of children aged under 5. These would\n\nMention: \"IFLS2\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The earlier household round receiving the community module.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004735:17:0", "occurrence_id": "4b571f2b53fc0fb5a7416583", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: years of schooling), the education level of the head‘s parents, and the head‘s religion and\n\n\nethnicity.\n\n2. We then introduce locational dummies such as urban/rural, or regional, dummies to\n\nmeasure where the household was living at the time of the first round survey. Most\n\n\nmultipurpose surveys with a migration module would collect the information needed to\n\n\nallow these variables to be constructed, and even without a specific migration module, it\n\nis common to ask where households were living five years ago. <sup>12</sup> [^12: For example, Smith and Thomas (2003) find that Malaysian households can accurately recall migration histories,\nparticularly for moves which are not very local or very short in duration.]\n\n3. Next, ―community‖ variables are added, which can be obtained from community modules\n\nin most household surveys or perhaps population censuses. Once the retrospective\n\n\nlocation is identified (as per model 2), the use of such variables depends only on the\n\n\navailability of such auxiliary data, and not on further recall per se. In the case of\n\n\nIndonesia, these come from the community-level survey from 1997 and are inserted into\n\n\nboth the IFLS2 and IFLS3 household surveys. For Vietnam, unfortunately the community\n\n\nmodule only collects data on rural communes, which can reduce the estimation sample\n\n\nsize significantly. Thus we will use instead a household-level variable which indicates\n\n\nhousehold poverty status as classified by the government in the first survey round.\n\n4. We then add variables describing a household head‘s sector of work. At this point we\n\nclearly start to lean more heavily on our ability to explicitly insert round 1 values of these\n\n\nvariables into the round 2 data. However, information on these variables could probably\n\n\nbe easily collected on a retrospective basis. Indeed retrospective work histories have been\n\n\ncollected in a number of labor surveys.\n\n5. Further demographic variables that we force to be time-invariant are then added - such as\n\nhousehold size and the number of children aged under 5. These would\n\nMention: \"IFLS3\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The later household round receiving the community module.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004743:13:0", "occurrence_id": "0560fac9db465311ceb2ff7c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Box 1: Global gains from migration**\n\nEven though quantitative estimates of the direct gains from migration are difficult to obtain,\neconomic simulations suggest that an increase in South-North migration would produce\nsubstantial income gains in the long-run; these income gains could exceed those from\ncomprehensive trade liberalization; and the destination countries in the North would\ncapture one fifth the overall benefits of increased immigration (World Bank 2006, Winters\n_et al._ 2003, Anderson & Winters 2008, van der Mensbrugghe & Roland-Holst 2009).\nDocumented welfare gains from South-North migration work primarily though the increase\nin the available labor force. Ortega and Peri (2009) found that immigration increases\nemployment in the destination countries in the North one for one, implying no crowdingout of natives. This result implies that immigration increases the total GDP of the receiving\ncountry without affecting average wages or labor productivity.\n\nImmigration has also been observed to boost productivity through innovation and\nspecialization. Data from the United States show that one percent increase in the share of\nmigrant university graduates increase the number of patent applications and grants issued\nper capita (Chellaraj _et al._ 2008, Hunt & Gauthier-Loiselle 2008). However, burdensome\nregulatory requirements and procedures that foreign doctors, engineers, architects and\naccountants have to meet in order to practice in the destination country can impose\nsignificant financial and other costs on these highly skilled immigrants (Mattoo and Mishra\n2009). Also the less-educated immigrants increase labor productivity as they complement\nthe uneducated local labor force that, based on their knowledge of the local language and\ninstitutions, will be better able to specialize in more productive complementary tasks (Peri\n& Spaber 2009). Furthermore, immigrants are often willing to do jobs that locals no longer\nare interested in, such as care for the elderly (UNDP 2009, p. 85). Also, the availability of\nlow-cost childcare by the immigrants can enable young\n\nMention: \"Data from the United States\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Migrant graduate shares and patent counts come from these records.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004748:12:0", "occurrence_id": "0a5d2b1064713277b0521dc4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |Table 1: Enterprise Survey average value reported by firms by country regressed on<br>the DB value|Col2|Col3|Col4|Col5|Col6|Col7|Col8|\n|---|---|---|---|---|---|---|---|\n||Imports|Imports|Imports|Construction Permit|Construction Permit|Operating License|Operating License|\n|OLS regression|OLS regression|OLS regression|OLS regression|OLS regression|OLS regression|OLS regression|OLS regression|\n|**Constant**|**Constant**|5.99||52.49||36.22||\n|**Slope**|**Slope**|0.14|0.14|0.05|0.05|0.00|0.00|\n|**N **|**N **|91|91|74.00|74.00|87.00|87.00|\n|**R-Squared**|**R-Squared**|0.129|0.129|0.03|0.03|0.00|0.00|\n|**t-test, β=0**|**t-test, β=0**|0.000|0.000|0.07|0.07|0.49|0.49|\n|**t-test, β=1**|**t-test, β=1**|0.000|0.000|0.00|0.00|0.00|0.00|\n|**Quartic** <br>|**\n\nMention: \"Enterprise Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Firm-reported averages are regressed on the regulatory indicator values.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004751:11:0", "occurrence_id": "5a6554a399397b191b200055", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: fragility of the mechanisms governing public sector works and the difficulty in controlling for multiple\nconflict of interests.\n\n\nConcessionaires led by the Autostrade group try to block any government intervention aimed at the\nintroduction of a more harmonious tariff system. In 2006 Autostrade provocatively threatened to sue the\ngovernment and ask for compensation in case a new tariff law would be enacted. <sup>23</sup> [^23: <u>[http://archiviostorico.corriere.it/2006/ottobre/08/Autostrade_soci_rivolta_governo_ora_co_9_061008023.shtml](http://archiviostorico.corriere.it/2006/ottobre/08/Autostrade_soci_rivolta_governo_ora_co_9_061008023.shtml)</u>]\n\n\nNevertheless, statistical data on this subject are less univocal. As Figure 4 shows, the evolution of the tariff\nbefore and after 1999 (the dotted line) does not seem to show a significantly different trend, although from\n2003 onward tariffs seem to be growing more than inflation.\n\nHowever one possible explanation may be given for the “tariff rage” of both concessionaires and users.\nBefore 1999, tariffs were linked to the highest one imposed by Autostrade, so the whole network would have\na nationally-fixed tariff system, either directly or through shadow-tolls. After 1999, a new system was\nintroduced, which linked the tariff applied on each highway both to general aggregates (inflation) and to\nconcession-specific variables (quality and productivity). For this reason, the geography of the highway tariff\nhas evolved unevenly since 1999. This may also explain consumer aversion and the haste against the new\nconcession system.\n\n\nFigure 4: Tariff and Inflation Rate Evolution <sup>24</sup> [^24: Source: Italian statistical institute, Istat. The base year for the two series is 1995=100.]\n\n\n130\n125\n120\n115\n110\n105\n100\n95\n90\n85\n80\n\n\n1991 1993 1995 1997 1999 2001 2003 2005\n\n\nInflation Rate Tariff\n\n\n_Source: Italian Statistical Institute, ISTAT. Inflation refers to variation in the consumer price index (CPI);_\n_the base year for\n\nMention: \"consumer price index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Variation in this index defines the inflation measure with that base year.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004732:12:0", "occurrence_id": "74c6a2fa22e0ee587bb384c0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: words” in Spanish, Italian, and French as some of the project descriptions are in the\n\n\ndonor’s language.\n\n\nOur method of categorizing aid flows through “key words” also has some\n\n\ndrawbacks, starting from omitted “key words” to projects that include more than one key\n\n\nword from different categories. Nonetheless, the gains outweigh the drawbacks; using\n\n\ndetailed project information is a crucial element of our analysis as we set out to evaluate\n\n\nthe effectiveness of aid on its main objective.\n\n\nAfter employing our “key word” search process we are left with close to 41,000\n\n\naid flow observations. Most of the observations from the CRS dataset are dropped as they\n\n\ndo not match any of the seven categories of interest. Most aid flows are either for disaster\n\n\nrelief, humanitarian purposes (including health, water, and food), and/or for primary and\n\n\nsecondary educational projects. <sup>6</sup>\n\n\nThe distribution of all aid flows and the distributions of all firm perceptions\n\n\namongst the different obstacles are presented in Figure 1. <sup>7</sup> [^7: The distribution of firm perceptions and aid flows will change in each country, region, and year. Figure 1\nis intended to display the general data and is not meant to be used to reach any conclusion about the\nrelation between aid and firm perspectives.] The size of the bubbles\n\n\ndescribes the distribution of aid flows. We can clearly see that the _trade_ area has received\n\n\nmore money than the other areas as 32.17% of all disbursements in the sample target this\n\n\narea. This is mainly driven by transportation related disbursements such as projects\n\n\ndesigned to build/improve roads and ports which require sizeable investments and are\n\n\nincluded under the trade category. The other two areas which aid targets more intensely\n\n\nare _rule of law_ and economic and political _stability_ .\n\n\n6 We only include projects that target the education of the labor force such as vocational and professional\ntraining under our Labor category. Aid targeted for primary or secondary\n\nMention: \"the CRS dataset\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Most of its flow observations fail to match the categories of interest.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004754:10:1", "occurrence_id": "b3da0fd1073cc5229ce89360", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  <br> <br> <br> <br> <br> <br> <br> <br>1 <br>[n=74]<br>|\n\n\n*** indicates p<0.01.\n\n\n<u>C. Rule of Law</u>\n\n\nIn order to measure the strength of the rule of law we use the index developed by Kaufmann,\n\n\nKraay, and Zoido-Lobaton (1999, hereafter KKZ). The KKZ data are available for 1996, 1998,\n\n2000, and then annually from 2002 to 2008. <sup>5</sup> [^5: KKZ data set and its description are available at http://www.worldbank.org/wbi/governance/data.html. See\nHaselmann et al. (2010) on the effects of alternative legal changes on bank lending in transition economies.] We use the index values from 2004 because it is\n\n\nclose to coterminous with the 2003 BCL survey, and because it offers wider country coverage\n\n\nthan the 2002 survey. We note that our results are not sensitive to the index year chosen, because\n\nthe KKZ rule of law index does not vary dramatically over time. <sup>6</sup> [^6: The correlation between the 2004 data and those from other years runs from .91 to .98.]\n\n\n<u>D. Depositor Monitoring</u>\n\n\nWe use the level of deposit insurance to measure the intensity of depositor monitoring. More\n\n\nspecifically, we compute, for each country _i_, the inverse of the ratio of deposit insurance\n\n\ncoverage to per capita GDP:\n\n\n\n<u>1</u>\n\n\n\n<u>1</u>\n\n_Depositor_ _Monitoring_ _i_  ( _Coverage_ _Limiti_ / _GDP_ _Per_ _Capita_ _i_ ) (2)\n\n\n\n( _Coverage_ _Limiti_ / _GDP_ _Per_ _Capita_ _i_ )\n\n\n\n[5 KKZ data set and its description are available at http://www.worldbank.org/wbi/governance/data.html. See\n\nMention: \"KKZ data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Governance indices available for these years and annually thereafter.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004754:10:1", "occurrence_id": "751f25976d1831312a206a90", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  <br> <br> <br> <br> <br> <br> <br> <br>1 <br>[n=74]<br>|\n\n\n*** indicates p<0.01.\n\n\n<u>C. Rule of Law</u>\n\n\nIn order to measure the strength of the rule of law we use the index developed by Kaufmann,\n\n\nKraay, and Zoido-Lobaton (1999, hereafter KKZ). The KKZ data are available for 1996, 1998,\n\n2000, and then annually from 2002 to 2008. <sup>5</sup> [^5: KKZ data set and its description are available at http://www.worldbank.org/wbi/governance/data.html. See\nHaselmann et al. (2010) on the effects of alternative legal changes on bank lending in transition economies.] We use the index values from 2004 because it is\n\n\nclose to coterminous with the 2003 BCL survey, and because it offers wider country coverage\n\n\nthan the 2002 survey. We note that our results are not sensitive to the index year chosen, because\n\nthe KKZ rule of law index does not vary dramatically over time. <sup>6</sup> [^6: The correlation between the 2004 data and those from other years runs from .91 to .98.]\n\n\n<u>D. Depositor Monitoring</u>\n\n\nWe use the level of deposit insurance to measure the intensity of depositor monitoring. More\n\n\nspecifically, we compute, for each country _i_, the inverse of the ratio of deposit insurance\n\n\ncoverage to per capita GDP:\n\n\n\n<u>1</u>\n\n\n\n<u>1</u>\n\n_Depositor_ _Monitoring_ _i_  ( _Coverage_ _Limiti_ / _GDP_ _Per_ _Capita_ _i_ ) (2)\n\n\n\n( _Coverage_ _Limiti_ / _GDP_ _Per_ _Capita_ _i_ )\n\n\n\n[5 KKZ data set and its description are available at http://www.worldbank.org/wbi/governance/data.html. See\n\nMention: \"KKZ data set\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Its documentation is hosted at the governance programme site.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004754:10:1", "occurrence_id": "07c5192beaf9379a16568764", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  <br> <br> <br> <br> <br> <br> <br> <br>1 <br>[n=74]<br>|\n\n\n*** indicates p<0.01.\n\n\n<u>C. Rule of Law</u>\n\n\nIn order to measure the strength of the rule of law we use the index developed by Kaufmann,\n\n\nKraay, and Zoido-Lobaton (1999, hereafter KKZ). The KKZ data are available for 1996, 1998,\n\n2000, and then annually from 2002 to 2008. <sup>5</sup> [^5: KKZ data set and its description are available at http://www.worldbank.org/wbi/governance/data.html. See\nHaselmann et al. (2010) on the effects of alternative legal changes on bank lending in transition economies.] We use the index values from 2004 because it is\n\n\nclose to coterminous with the 2003 BCL survey, and because it offers wider country coverage\n\n\nthan the 2002 survey. We note that our results are not sensitive to the index year chosen, because\n\nthe KKZ rule of law index does not vary dramatically over time. <sup>6</sup> [^6: The correlation between the 2004 data and those from other years runs from .91 to .98.]\n\n\n<u>D. Depositor Monitoring</u>\n\n\nWe use the level of deposit insurance to measure the intensity of depositor monitoring. More\n\n\nspecifically, we compute, for each country _i_, the inverse of the ratio of deposit insurance\n\n\ncoverage to per capita GDP:\n\n\n\n<u>1</u>\n\n\n\n<u>1</u>\n\n_Depositor_ _Monitoring_ _i_  ( _Coverage_ _Limiti_ / _GDP_ _Per_ _Capita_ _i_ ) (2)\n\n\n\n( _Coverage_ _Limiti_ / _GDP_ _Per_ _Capita_ _i_ )\n\n\n\n[5 KKZ data set and its description are available at http://www.worldbank.org/wbi/governance/data.html. See\n\nMention: \"BCL survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"The index year is chosen to align closely with this survey round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004755:8:0", "occurrence_id": "8aa4f4dae460bf172cd368c7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: past eight years in over 100 countries with a consistent survey instrument. <sup>3</sup> [^3: See www.enterpriseseurveys.org for more details. Similar surveys were previously conducted under the\nleadership of the World Bank and other IFIs in Africa (RPED), and world-wide in 2000 (World Business\nEnvironment Survey). Enterprise Surveys still go under different names in some regions, e.g. BEEPS in the\nCentral Asia and Eastern European transition economies had first been launched in 1999 and was then modified\nto be comparable with the broader global initiative.] The surveys try to\n\n\ncapture business perceptions on the most important obstacles to enterprise operation and\n\n\ngrowth, but also include detailed information on management and financing arrangements of\n\n\ncompanies.  Sample sizes vary between 250 and 1,500 companies per country and data are\n\n\ncollected using either simple random or random stratified sampling. The sample includes\n\n\nformal and informal enterprises of all sizes, different ownership types and across nine sectors\n\n\nin manufacturing, construction, services and transportation. Firms from different locations,\n\n\nsuch as capital city, major cities and small towns, are included. Our empirical work relies on\n\n\ndata for formal enterprises from 37 African countries and 49 countries from the rest of the\n\n\nworld covering in total 35,000 firms during 2006 to 2009 (Appendix Table A1). <sup>4</sup> [^4: We lose 25 countries when using the share of working capital financed with external finance, mostly in Eastern\nand Central Europe and East Asia. There are a total of 95 surveys but some countries enter the dataset in\nmultiple years. Information collected is standardized across countries and covers formal establishments with 5\nor more employees. However, Africa, East Europe and Central Asia, and Latin America include establishments\nwith fewer than 5 employees. In Africa, there are an additional 25 standardized surveys covering micro-firms\nand targeting informal establishments.] In addition,\n\n\nwe focus on informal mostly micro-firms for a sample of 25 African countries.\n\n\nThe Enterprise Surveys offer\n\nMention: \"World Business\nEnvironment Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The worldwide predecessor round of the firm survey programme.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004755:8:0", "occurrence_id": "0774768b217f2613c5965bde", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: past eight years in over 100 countries with a consistent survey instrument. <sup>3</sup> [^3: See www.enterpriseseurveys.org for more details. Similar surveys were previously conducted under the\nleadership of the World Bank and other IFIs in Africa (RPED), and world-wide in 2000 (World Business\nEnvironment Survey). Enterprise Surveys still go under different names in some regions, e.g. BEEPS in the\nCentral Asia and Eastern European transition economies had first been launched in 1999 and was then modified\nto be comparable with the broader global initiative.] The surveys try to\n\n\ncapture business perceptions on the most important obstacles to enterprise operation and\n\n\ngrowth, but also include detailed information on management and financing arrangements of\n\n\ncompanies.  Sample sizes vary between 250 and 1,500 companies per country and data are\n\n\ncollected using either simple random or random stratified sampling. The sample includes\n\n\nformal and informal enterprises of all sizes, different ownership types and across nine sectors\n\n\nin manufacturing, construction, services and transportation. Firms from different locations,\n\n\nsuch as capital city, major cities and small towns, are included. Our empirical work relies on\n\n\ndata for formal enterprises from 37 African countries and 49 countries from the rest of the\n\n\nworld covering in total 35,000 firms during 2006 to 2009 (Appendix Table A1). <sup>4</sup> [^4: We lose 25 countries when using the share of working capital financed with external finance, mostly in Eastern\nand Central Europe and East Asia. There are a total of 95 surveys but some countries enter the dataset in\nmultiple years. Information collected is standardized across countries and covers formal establishments with 5\nor more employees. However, Africa, East Europe and Central Asia, and Latin America include establishments\nwith fewer than 5 employees. In Africa, there are an additional 25 standardized surveys covering micro-firms\nand targeting informal establishments.] In addition,\n\n\nwe focus on informal mostly micro-firms for a sample of 25 African countries.\n\n\nThe Enterprise Surveys offer\n\nMention: \"Enterprise Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Their coverage of micro-firms is uneven across the region.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004758:11:1", "occurrence_id": "10f8147f060c3525902d76de", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  within each coun\n\ntry and year. In happiness research this variable is often used in conjunction with income\n\n\nand/or income inequality to test the importance of the relative income position as opposed\n\n\nto absolute income in explaining satisfaction. Relative income has been found to have a\n\n\nsignificant and positive effect on satisfaction but the sign and significance of this variable\n\n\nmay be affected by collinearity with other variables such as income or income inequality.\n\n\nIt is important therefore to test how the inclusion and exclusion of this variable may alter\n\n\nresults for income inequality.\n\n\nCountries’ wealth ( _W_ ) is measured with GDP per capita for each country and year.\n\n\nAs already mentioned, this variable has been extracted from the IMF database and is used\n\n\nin real terms, USD and PPP values.\n\n\nWe also use a number of control variables ( _C_ ) as follows. A first set of variables\n\n\nmeasures _individual_ _and_ _family_ _attributes_ which are possible predictors of life-satisfaction.\n\n\nThese are being unemployed (dummy), sex (female), age (continuous with the addition of\n\n\nage squared) and a dummy for tertiary education and marriage status (dummy where one\n\n\nincludes: “ _married_ ” and “ _living_ _together_ _as_ _married_ ”). These are all variables which have\n\n\n11Note that for countries that changed currency during the period considered (adoption of the EURO or\nUSD or introduction of a new local currency) the IMF data use only the latest currency. This meant that\nwe had to transform first the income values from our database into the same currencies used by the IMF\nusing the appropriate exchange rates for each currency and each year and only then apply the constant,\nUSD and PPP transformations.\n\n12For the selection of the most appropriate Gini, we followed indications provided by Gruen and Klasen\n(2008). According to tests conducted by these authors on the Gini WIDER database: “ _Gini\n\nMention: \"IMF database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Income per capita by country and year is extracted in real dollar terms.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004758:11:1", "occurrence_id": "1fd3f5c00ee6ddc1496d2c40", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  within each coun\n\ntry and year. In happiness research this variable is often used in conjunction with income\n\n\nand/or income inequality to test the importance of the relative income position as opposed\n\n\nto absolute income in explaining satisfaction. Relative income has been found to have a\n\n\nsignificant and positive effect on satisfaction but the sign and significance of this variable\n\n\nmay be affected by collinearity with other variables such as income or income inequality.\n\n\nIt is important therefore to test how the inclusion and exclusion of this variable may alter\n\n\nresults for income inequality.\n\n\nCountries’ wealth ( _W_ ) is measured with GDP per capita for each country and year.\n\n\nAs already mentioned, this variable has been extracted from the IMF database and is used\n\n\nin real terms, USD and PPP values.\n\n\nWe also use a number of control variables ( _C_ ) as follows. A first set of variables\n\n\nmeasures _individual_ _and_ _family_ _attributes_ which are possible predictors of life-satisfaction.\n\n\nThese are being unemployed (dummy), sex (female), age (continuous with the addition of\n\n\nage squared) and a dummy for tertiary education and marriage status (dummy where one\n\n\nincludes: “ _married_ ” and “ _living_ _together_ _as_ _married_ ”). These are all variables which have\n\n\n11Note that for countries that changed currency during the period considered (adoption of the EURO or\nUSD or introduction of a new local currency) the IMF data use only the latest currency. This meant that\nwe had to transform first the income values from our database into the same currencies used by the IMF\nusing the appropriate exchange rates for each currency and each year and only then apply the constant,\nUSD and PPP transformations.\n\n12For the selection of the most appropriate Gini, we followed indications provided by Gruen and Klasen\n(2008). According to tests conducted by these authors on the Gini WIDER database: “ _Gini\n\nMention: \"IMF data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Its use of only the latest currency forces a conversion step.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004758:11:1", "occurrence_id": "a08ef13189e210533933e814", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  within each coun\n\ntry and year. In happiness research this variable is often used in conjunction with income\n\n\nand/or income inequality to test the importance of the relative income position as opposed\n\n\nto absolute income in explaining satisfaction. Relative income has been found to have a\n\n\nsignificant and positive effect on satisfaction but the sign and significance of this variable\n\n\nmay be affected by collinearity with other variables such as income or income inequality.\n\n\nIt is important therefore to test how the inclusion and exclusion of this variable may alter\n\n\nresults for income inequality.\n\n\nCountries’ wealth ( _W_ ) is measured with GDP per capita for each country and year.\n\n\nAs already mentioned, this variable has been extracted from the IMF database and is used\n\n\nin real terms, USD and PPP values.\n\n\nWe also use a number of control variables ( _C_ ) as follows. A first set of variables\n\n\nmeasures _individual_ _and_ _family_ _attributes_ which are possible predictors of life-satisfaction.\n\n\nThese are being unemployed (dummy), sex (female), age (continuous with the addition of\n\n\nage squared) and a dummy for tertiary education and marriage status (dummy where one\n\n\nincludes: “ _married_ ” and “ _living_ _together_ _as_ _married_ ”). These are all variables which have\n\n\n11Note that for countries that changed currency during the period considered (adoption of the EURO or\nUSD or introduction of a new local currency) the IMF data use only the latest currency. This meant that\nwe had to transform first the income values from our database into the same currencies used by the IMF\nusing the appropriate exchange rates for each currency and each year and only then apply the constant,\nUSD and PPP transformations.\n\n12For the selection of the most appropriate Gini, we followed indications provided by Gruen and Klasen\n(2008). According to tests conducted by these authors on the Gini WIDER database: “ _Gini\n\nMention: \"Gini WIDER database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Tests on this database assess the reliability of its coefficients.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004768:6:0", "occurrence_id": "7ae829c489a35b7d1594069c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Tij_ = _β0_ + _β1 Aij_ + _β2 Fij_ + _β3S ij_ + _β4Iij_ + _єij_ (2)\n\n\nwhere _β0_ to _β4_ are the coefficients to be estimated. The standard procedure for analyzing the\ndeterminants of the test score differences over time is to fit equations between test scores and\nobserved characteristics. The observed test score differential can be decomposed as:\n\n\n_T2006_          - _T2003_ = ( _X2006_          - _X2003_ )  _2006_ + _X2003_ (  _2006_          -  _2006_ ) (3)\n\n\nwhere _T_ is the standardized test score, _Xi_ is a vector of student, family, school, and institutional\ncharacteristics for the _i_ th individual,  is a vector of coefficients, and 2006 and 2003 subscripts\nare identifiers of the PISA test score in math in years 2003 and 2006 evaluated at 2006 prices.\nThe overall increase in test score can, therefore, be decomposed into two components. One is the\nportion attributable to differences in characteristics ( _X2006_ - _X2003_ ) evaluated at 2006 prices or to\nthe performance of the 2006 group of students (  _2006_ ), while the other portion is attributable to\ndifferences in the effects on performance _(_  _2006_ -  _2003_ ) of 2003 and 2006 students derived from\nthe same characteristics. This second (unexplained) component, while more difficult to interpret\nin the present context than an earnings gap decomposition framework, may have had more than\none explanation. The first and most obvious explanation is that the unexplained portion of the\ntest score increase may reflect certain unobserved family characteristics that are correlated with\nachievement over time, possibly related to household wealth. The second possible explanation\nmay be that\n\nMention: \"PISA test score\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Mathematics scores in these two years, valued at constant prices, are decomposed.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004769:13:1", "occurrence_id": "6e8b12061741e1a842ed4436", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  stood. In Annex 9 we have also computed the export intensity <sup>17</sup> of the\nmentioned sectors. According to those results, the automotive sector is the sector\n\n\n16 Production Linkages and Industrial Upgrading: Case Study of Indonesia‟s Automotive Industry, by Haryo\nAswicahyono and Pratiwi Kartika.\n17 Due to data limitations, we have used as proxy for export intensity the share of nominal exports over the sectoral\nvalue added. The sectoral value added is defined, according to the UNIDO ISIC Database, as the value of census\noutput less the value of census input.\n\n\n12\n\nMention: \"the UNIDO ISIC Database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Sectoral value added is defined here as census output less census input.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004776:9:0", "occurrence_id": "2dabaaad1266bda593a70171", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: reference cluster. The variables and their sources are described in Appendix 2. The regressors fall into\nthree broad categories:\n\n\n**Figure 2: Correlations between Growth and the Sources of Vulnerability**\n\n\n\n\n\n\n\n\n\nSources: IMF’s World Economic Outlook database and authors’ calculations.\nNotes: The lower growth and lower vulnerability is identified by L, the lower growth and higher vulnerability is identified by V, the higher\ngrowth and lower vulnerability is identified by G, and the higher growth and higher vulnerability is identified by H. The absence of higher\ngrowth, higher vulnerability data in 2000 and 2008 reflects the lack of countries in such cluster.\n\n\n- <u>Initial conditions. These comprise measures that may be susceptible to policies and institutional</u>\nchanges over a relatively long period. This category includes measures that reflect a country’s\neconomic development and its economic structure, as well as the degree to which an economy is\nopen to the flows of goods and capital. This includes the initial level of income and oil trade balance\nas a share of total exports plus imports. In addition, measures of trade (exports and imports as a\npercentage of GDP) and financial (FX assets plus liabilities as a percentage of GDP) openness are\nincluded and, in turn, these are accompanied by indicators that serve to qualify the type of\nopenness experienced by individual countries. Specifically, the difference between export and\nimport growth rates is used to assess the type of trade openness. Also, FDI (net, as a percent of\n\n\n7\n\nMention: \"IMF’s World Economic Outlook database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Growth and vulnerability series plotted here come from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004778:11:0", "occurrence_id": "732eebe7afdfe223d066f926", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: CÔTE D’IVOIRE’S INFRASTRUCTURE: A CONTINENTAL PERSPECTIVE\n\n\nc. Power\n\n\nd. Water\n\n\nSource: AICD Interactive Infrastructure Atlas for Côte d‘Ivoire downloadable from\nhttp://www.infrastructureafrica.org/aicd/system/files/civ_new_ALL.pdf\n\n\n8\n\nMention: \"AICD Interactive Infrastructure Atlas\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The country's power and water maps are downloadable from this atlas.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004780:10:0", "occurrence_id": "7a8a4a7d0166bf8bf073cb1d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: KENYA’S INFRASTRUCTURE: A CONTINENTAL PERSPECTIVE\n\n\n**Table 2. Kenya’s road indicators benchmarked against Africa’s low- and middle-income countries**\n\n\n\nUnit Low-income\ncountries\n\n\n\n**Kenya** Middleincome\ncountries\n\n\n\nPaved road density km/1000 km2 of arable land 86.6 **152** 507.4\n\nUnpaved road density km/1000 km2 of arable land 504.7 **930** 1,038.3\n\n\n\nGIS rural accessibility % of rural population within 2 km of allseason road\n\n\n\n21.7 **32** 59.9\n\n\n\nPaved road traffic Average annual daily traffic 1,049.6 **1,108** 2,786.0\n\nUnpaved road traffic Average annual daily traffic 62.6 **38** 12.0\n\nPaved network condition % in good or fair condition 80.0 **84** 79.0\n\nUnpaved network condition % in good or fair condition 57.6 **63** 58.3\n\n\n\nPerceived transport quality % firms identifying roads as major\nbusiness constraint\n\n\n\n23.0 **37** 10.7\n\n\n\nSource: Gwillliam and others 2009, derived from AICD national database downloadable from http://www.infrastructureafrica.org/aicd/tools/data.\n\n\n**Challenges**\n\n\nNevertheless, the country faces a huge rehabilitation backlog that must be addressed before the trunk\nnetwork can be considered to be in a maintainable condition (figure 4). As of 2006, levels of capital\nspending for the roads sector—at around 1 percent of GDP—were low by regional standards and fell\nsubstantially short of what would be needed to clear the rehabilitation backlog in a reasonable period of\ntime. There is a need for a one-time push on road sector investment to remedy this situation.\n\n\n**Figure 4. Kenya is not spending enough\n\nMention: \"AICD national database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Constraints cited as a major business obstacle are drawn from this national database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004781:4:0", "occurrence_id": "6fffdd48dba3dccb2709d9e3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: iberia’s 14-year civil war left much of the country’s infrastructure shambles. The country’s 170\nmegawatt power generation capacity and national grid were completely destroyed. In Monrovia,\njust 0.1 percent of households had access to electricity. According to the 2008 National Census,\n# L\naccess to piped water fell from 15 percent of the population in 1986 to less than 3 percent in 2008. The\nnational road network was left in severe disrepair.\n\n\nPeace brought many positive developments. The Freeport of Monrovia is now privately managed and\nhas resumed normal operations. Essential rehabilitation work has been carried out, and the port’s\nperformance now matches that of neighboring ports along the West African coast. Liberia has also\nsuccessfully liberalized its mobile telephone markets, with access surging to 40 percent in 2009, at some\nof the lowest prices in Africa.\n\n\nBut Liberia still faces critical infrastructure challenges. Perhaps the starkest lies in the power sector.\nThe country’s installed power generation capacity is little more than 2 megawatts per million people,\nbarely one-tenth of the benchmark level for other low-income countries in Africa. The cost of generating\npower is $0.77 per kilowatt-hour, exceptionally high by any standard. Furthermore, the power tariff of\n$0.43 per kilowatt-hour is about three times the average for Africa, which is already very high by global\nstandards. Investment in more cost-effective power generation is therefore an important strategic\nobjective for the country if electrification is to reach the wider population.\n\n\nAlthough good progress has been made in securing donor finance for road reconstruction, existing\ncapital spending is about half the level that would be required to complete the rehabilitation process\nwithin five years, and existing maintenance spending is about half what would be needed to preserve\nthese new investments into the future. Many countries have addressed this problem by establishing road\nfunds financed by user charges. That approach is being considered in Liberia.\n\n\nAddressing Liberia’s public infrastructure needs will require\n\nMention: \"the 2008 National Census\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Access to piped water and electricity is measured in this enumeration.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004786:6:0", "occurrence_id": "8e25e68087f5bb3579b91501", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: THE DEMOCRATIC REPUBLIC OF CONGO’S INFRASTRUCTURE: A CONTINENTAL PERSPECTIVE\n\n\n**Figure 1. Historic and potential future links between infrastructure and growth**\n\na. Historic changes in growth per capita\n\n\nEvidence from enterprise surveys suggests that infrastructure constraints throughout the region are\nresponsible for about 40 percent of the productivity handicap faced by Sub-Saharan firms, with the\nremainder being due to poor governance, bureaucratic red tape, and financing constraints. In many\ncountries, lack of affordable power is the infrastructure constraint that weighs most heavily on firms.\nWhile detailed enterprise survey evidence is not available for the DRC, power would likely emerge as a\nmajor concern here as well.\n\n\n3\n\nMention: \"enterprise surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Firm reports attribute a large share of the growth gap to infrastructure.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004788:32:2", "occurrence_id": "2decae60e465ec71192655e5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: </mark> <mark>(0.025)</mark> <mark>(0.024)</mark>\n\n**Pando** 0.0168 0.0067 0.0217 0.0443\n\n<mark>(0.073)</mark> <mark>(0.068)</mark> <mark>(0.078)</mark> <mark>(0.085)</mark>\n\n\n**<mark>Observations</mark>** <mark>303</mark> <mark>303</mark> <mark>303</mark> <mark>303</mark>\n\n**R2** 0.0913 0.0940 0.0918 0.116\n\n**<mark>*** p<0.01, ** p<0.05, * p<0.1</mark>**\n\nNote: Marginal effects reported; standard errors reported in parenthesis.\nSource: Authors from World Bank (2010) APER database.\n\n_Additional Controls_\n\nThis section reports the estimates from alternative specifications of Equation 2 that address two\npotentially important issues: scale effects and the inclusion of observable controls in the areas of\nmunicipal capacity and political issues. In all previous estimations, agricultural spending\nvariables are expressed in per capita terms, in order to get rid of scale effects—that is, the\npossibility of vulnerability impacts are being driven mainly by the magnitude of spending rather\nthan its efficiency. Estimations were also undertaken using total spending, thus capturing the\n\n\n31\n\nMention: \"APER database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Project-level estimates with standard errors are drawn from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004790:19:0", "occurrence_id": "52bbe3e386e4e0a2af4692b4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: theories and at least 145 proposed regressors as proxies. We turn to the standard growth literature\nto guide us in this regard and focus on a subset of well accepted growth determinants in our\nanalysis (see annex tables and data description for details).\n\nThe baseline empirical growth model consists, in addition to the variable of interest, of four\ndeterminants of economic growth - initial income level, rates of physical and human capital\naccumulation, trade openness and institutional quality. We use data on income per capita from\nWDI. To account for differences in human capital accumulation we use latest available data on\nyears of schooling from Barro and Lee (2010). <sup>11</sup> We use M2 (as percent of GDP) as a proxy for\nfinancial development from WDI. Finally, we use a measure of Rule of Law from World\nGovernance indicators. <sup>12</sup>\n\nWe use dynamic panel regressions based on three year panel data for each country. The\nregression specification is the following:\n\n(1)\n\nwhere is the natural logarithm of GDP per capita, is the level of GDP per capita at the\nbeginning of each three year period, is the dynamic service export sophistication\nmeasure, is the set of relevant control variables for growth determinants, represents time\ndummies, stands for country fixed effects and is the error term.\n\n\nResults of fixed effect and System GMM regression for three year periods are presented in Table\n3. <sup>13</sup> The convergence effect, human capital, and service EXPY seem to matter for growth only in\nthe last time period (2005-07). We add various controls for growth to our baseline regression for\nthe period. Introduction of the controls does not change the significance of the co-efficient on\n_SE_ . <sup>14</sup>\n\n\n11 Neoclassical growth model, since the seminal work of Lucas (1988), the concept\n\nMention: \"data on income per capita from\nWDI\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Output per head enters the growth regression as the outcome variable.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004790:19:0", "occurrence_id": "6c495e91213fe9fc5d519b44", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: theories and at least 145 proposed regressors as proxies. We turn to the standard growth literature\nto guide us in this regard and focus on a subset of well accepted growth determinants in our\nanalysis (see annex tables and data description for details).\n\nThe baseline empirical growth model consists, in addition to the variable of interest, of four\ndeterminants of economic growth - initial income level, rates of physical and human capital\naccumulation, trade openness and institutional quality. We use data on income per capita from\nWDI. To account for differences in human capital accumulation we use latest available data on\nyears of schooling from Barro and Lee (2010). <sup>11</sup> We use M2 (as percent of GDP) as a proxy for\nfinancial development from WDI. Finally, we use a measure of Rule of Law from World\nGovernance indicators. <sup>12</sup>\n\nWe use dynamic panel regressions based on three year panel data for each country. The\nregression specification is the following:\n\n(1)\n\nwhere is the natural logarithm of GDP per capita, is the level of GDP per capita at the\nbeginning of each three year period, is the dynamic service export sophistication\nmeasure, is the set of relevant control variables for growth determinants, represents time\ndummies, stands for country fixed effects and is the error term.\n\n\nResults of fixed effect and System GMM regression for three year periods are presented in Table\n3. <sup>13</sup> The convergence effect, human capital, and service EXPY seem to matter for growth only in\nthe last time period (2005-07). We add various controls for growth to our baseline regression for\nthe period. Introduction of the controls does not change the significance of the co-efficient on\n_SE_ . <sup>14</sup>\n\n\n11 Neoclassical growth model, since the seminal work of Lucas (1988), the concept\n\nMention: \"data on\nyears of schooling\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Latest attainment measures proxy human capital accumulation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004790:19:0", "occurrence_id": "9d1b48e499904414e9b4dc9b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: theories and at least 145 proposed regressors as proxies. We turn to the standard growth literature\nto guide us in this regard and focus on a subset of well accepted growth determinants in our\nanalysis (see annex tables and data description for details).\n\nThe baseline empirical growth model consists, in addition to the variable of interest, of four\ndeterminants of economic growth - initial income level, rates of physical and human capital\naccumulation, trade openness and institutional quality. We use data on income per capita from\nWDI. To account for differences in human capital accumulation we use latest available data on\nyears of schooling from Barro and Lee (2010). <sup>11</sup> We use M2 (as percent of GDP) as a proxy for\nfinancial development from WDI. Finally, we use a measure of Rule of Law from World\nGovernance indicators. <sup>12</sup>\n\nWe use dynamic panel regressions based on three year panel data for each country. The\nregression specification is the following:\n\n(1)\n\nwhere is the natural logarithm of GDP per capita, is the level of GDP per capita at the\nbeginning of each three year period, is the dynamic service export sophistication\nmeasure, is the set of relevant control variables for growth determinants, represents time\ndummies, stands for country fixed effects and is the error term.\n\n\nResults of fixed effect and System GMM regression for three year periods are presented in Table\n3. <sup>13</sup> The convergence effect, human capital, and service EXPY seem to matter for growth only in\nthe last time period (2005-07). We add various controls for growth to our baseline regression for\nthe period. Introduction of the controls does not change the significance of the co-efficient on\n_SE_ . <sup>14</sup>\n\n\n11 Neoclassical growth model, since the seminal work of Lucas (1988), the concept\n\nMention: \"World\nGovernance indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its rule of law measure captures institutional quality.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004799:9:0", "occurrence_id": "b514582c0fab3d7ea1da6e6c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **3.** **Data**\n\n\nWe are able to study the impacts of extreme weather events on rural households by\n\n\nmerging household and community level data from the Indonesian Family Life Survey (IFLS)\n\n\nwith daily rainfall data covering a 25 year period. The combined data set contains information on\n\n\nrainfall, household expenditures, household level socio-economic characteristics, and community\n\n\nlevel attributes.\n\n\nHousehold and community surveys were fielded from late June to the end of October\n\n\n2000 for IFLS3 and from August 1997 to January 1998 for IFLS2. The surveys include village\n\nlevel data which allows the determination of the extent to which access to better infrastructure or\n\n\nsocial programs increases resiliency. The consumption aggregate consists of food and nonfood\n\n\ncomponents. The food component consists of 37 food items (purchases and the value of own\n\n\nproduction or gifts) consumed within the last week. The nonfood component consists of\n\n\nfrequently purchased goods and services (utilities, personal toiletries, household items, domestic\n\n\nservices, recreation and entertainment, transport, sweepstakes), less frequent purchases and\n\n\ndurables (clothing, furniture, medical, ceremonies, tax), housing, and educational expenditures\n\n\nfor children living in the household. Transfers out of the household were excluded. All values are\n\n\nmonthly figures and are in real terms. To obtain real values, both temporal and spatial deflators\n\nwere used, using prices in December 2000 in Jakarta as the base. <sup>6</sup> [^6: The spatial deflator used is the ratio of the location (province, urban/rural area) poverty line (in December 2000\nprices) to the Jakarta poverty line. Thus the spatial deflator used converts the local December 2000 values into\nJakarta December 2000 values.]\n\n\nUsing daily rainfall data from 1979 to 2004, we calculated the 25 year mean and standard\n\n\ndeviations for monsoon onset and the amount of post-onset rainfall for 32 weather stations. The\n\n\nrainfall data from these weather stations were then matched to communities in IFLS. Weather\n\n\ndata were merged with household survey data at the community level\n\nMention: \"Indonesian Family Life Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its household and community layers are merged with daily rainfall over 25 years.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004803:15:1", "occurrence_id": "7e945576809463659db3d9fc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: * 0.326*** 0.274***\n(0.0864) (0.0860) (0.0297) (0.115) (0.114) (0.0460)\n\n\n\nEx-post\n\n\n\nEx-post\n\n\n\nEx-post\n<u>Exports</u>\n\n\n\n<u>VARIABLES</u> <u>FDI</u> <u>FDI</u> <u>Exports</u> <u>Exports</u>\n\n\n\n<u>FDI</u>\n\n\n\n<u>FDI</u>\n\n\n\nObservations 886 882 901 901 445 445 464 464\nR-squared 0.055 0.082 0.129 0.405 0.038 0.063 0.214 0.279\nNumber of\n<u>countries</u> <u>60</u> <u>60</u> <u>60</u> <u>60</u> <u>60</u> <u>60</u> <u>60</u> <u>60</u>\nStandard errors in parentheses\n*** p<0.01, ** p<0.05, * p<0.1\nAll estimations include period fixed effects and country fixed effects (demeaning approach).\n\nSource: Author’s calculations based on data from the IMF. See text for details.\n\nMention: \"data from the IMF\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The demeaned panel estimates are computed from these data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004806:26:1", "occurrence_id": "9765b48ed9e3c2f2064e2a10", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  which will\n\npresumably also be borne disproportionately by the non-poor.\n\n\nThus far we have considered the rural population in isolation. Next we bring urban\n\nhouseholds into the picture to examine the broader distributional implications of climate\n\nchange in India. Around 15 percent of households classified as urban in NSS61 (representing\n\n16 percent of the urban population) report that they own land apart from their homesteads.\n\nWe treat these (presumably) peri-urban households as though they were rural, performing\n\nthe same calculations as those summarized in Table 6 to obtain their %∆ <sup>�</sup> _c_ . For _bona_ _fide_\n\nurban households, who have _λ_ = 0, we assume zero wage effects. In other words, as a\n\nfirst approximation, we assume that urban and rural labor markets are segmented from each\nother. <sup>30</sup> As a consequence, these urban households will only experience the impact of higher\n\ncereal prices and, since these households are not agricultural producers, the welfare effects\n\n\n29Again, it comes down to _ϵw,π_ being substantially less than unity. An adaptation that mitigates the\ndecline in _π_ due to warming by 10 percentage points, only mitigates the decline in _w_ by around 3.5 percentage\npoints.\n\n30This also assumes no change in the amenity value of climate for urban residents (see fn. 9), or at least\nthat any change in the amenity value of climate is distributed equally in the population.\n\n\n24\n\nMention: \"NSS61\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Urban households reporting land ownership are identified in this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004809:7:0", "occurrence_id": "7fb9c8573cda49b63df56b34", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _nk_ _i_ _i_ _nk_\n###  1  xkt ln  xkt  where  kt  1  xkti (2)\n\n_nk_ _i_  1  _kt_   _kt_  _nk_ _i_  1\n\n\n\n_nk_ _i_ _i_ _nk_\n\n_<u>x</u>_  _<u>x</u>_  <u>1</u>\n\n\n\n_nk_ _i_ _i_ _nk_\n### Tkt  1  xkt ln  xkt  where  kt  1  x\n\n_nk_ _i_  1  _kt_   _kt_  _nk_ _i_  1\n\n\n\n_nk_ _i_ _i_\n### 1  xkt ln  xkt  where  kt  1\n\n\n### kt   kt ln  kt  where  kt   xkti\n\n_nk_ _i_  1  _kt_   _kt_  _nk_ _i_  1\n\n\n\n1 _kt_  _kt_  _k_ _i_  1\n\n\n\nIn order to explore action at the extensive margin, we also consider the simple\n\nnumber of exporters of good _k_ to OECD countries.\n\n\nOur data are COMTRADE import data for OECD countries (either taken as a bloc or\n\ndisaggregated by importer) at the product level. Our preferred product classifications\n\n\nare SITC4. The alternative, HS6, is more disaggregated (with 4,990 to 5,016 lines\n\ndepending\n\nMention: \"COMTRADE import data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Product-level flows by importer give the number of exporters.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004810:21:0", "occurrence_id": "8080b1e115552833435234db", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Measuring True Sales and Underreporting with Matched Firm-Level Survey and Tax office Data\n\n\nwhere is the sales reported in the survey, is a constant, and is the corresponding\n\n\nmeasurement error. The reported sales are expected to be positively correlated with true\n\n\nsales ( ). We specify the survey sales to be a function of the true sales because at the\n\n\ntime of the survey true sales are known to the firm. The log linear specification is flexible as\n\n\nit allows the survey sales to be either above or below the true sales.\n\n\nSummarizing, equations (7), (10) and (11) form a MIMIC model, where a firm’s true sales\n\n\n(S ) is the latent variable and reported sales to the tax office ( ) and in the survey ( ) are\n\n\nthe indicators (measures). We also include the restriction that reported sales to the tax\n\n\noffice do not exceed the true sales (equation 12):\n\n```\n                                      ⑺\n\n                                      ⑽\n                                      ⑾\n                                      ⑿\n\n```\n\nIn line with the MIMIC literature, we assume to be multivariate normally distributed\n\n\nwith mean zero but we allow for any possible correlation among these error terms. In the\n\n\nAppendix B we discuss the identification of the parameters of the model and derive the\n\n\nlikelihood function for estimation in the next section.\n\n\n**4 Estimated true sales versus reported sales**\n\n\nThe data for this study are from Mongolia - a land-locked country in East and Central Asia\n\n\nthat has gone through radical changes from central planning towards market economy in\n\n\n1990s. We use two data sources. The first source of data is the World Bank Productivity\n\n\n20\n\nMention: \"World Bank Productivity\n\n\n20\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"One of two sources on productivity during the transition decade.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004813:39:0", "occurrence_id": "d9ed1e2ab1ab8a7e8c2e2b9f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Table 5: Establishment Size and Growth – Large vs. Small Informal Sector**\nThe regressions estimated in this table are: Employment Growth/Sales Growth/Productivity Growth = a + b0 Size Dummy for 20-99 employees +\nb1Size Dummy for ≥100 employees + b2 Age +Country Dummies + Sector Dummies + Year Dummies + e. Employment Growth is defined as\nthe log difference in permanent, full-time employment over a two year period. Sales Growth is defined as the log difference in sales over a two\nyear period and Labor Productivity Growth is defined as the log difference in labor productivity (Sales/Employment) over a two year period. Cols\n1-3 present results for countries that have a large informal sector (above the median value) and cols. 4-6 present results for countries with a small\ninformal sector (below the median value) where informal sector is defined by the informal sector’s contribution to GDP in Schneider et al.\n(2010). All data is at the firm level from the World Bank Enterprise Surveys. All regressions are OLS regressions with standard errors clustered\nat the country level.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n38\n\nMention: \"the World Bank Enterprise Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Firm-level observations behind the informal share regressions come from here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004815:27:0", "occurrence_id": "371644acef8590f67236ff2c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 3: World Bank Food Price Index and price index weighted by poverty impacts**\n\n\n\n\n\n\n\n26\n\nMention: \"World Bank Food Price Index\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Commodity prices and their poverty-weighted counterpart are plotted from it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004816:32:0", "occurrence_id": "4e70342d1e6d6f4712f6a4d4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: b) More volatility of sales is associated with higher financial savings;\n\n\nc) Higher cost of capital is associated with higher financial savings;\n\n\nd) Increase in financial savings leads to increase in physical savings (i.e. investment) in\n\n\nabout 2 years.\n\n\nDespite the data limitations, two important findings are consistent across the two very\n\n\ndifferent datasets. The fact that these findings are consistent in both datasets reaffirms the\n\n\nusefulness and credibility of these results.\n\n\nFirst is the finding that larger firms invest more (physical saving), which is essential for\n\n\nmaintaining competitiveness and growth. The study points out that one important reason small\n\n\nfirms invest less is lack of access to credit. Thus, improving access to credit for small and\n\n\nmedium enterprises is likely to have a positive impact on investment.\n\n\nSecond, despite the financial deepening, the use of credit products has been declining\n\n\nduring the last decade. This result is observed in both datasets. Specifically, usage of credit\n\n\nproducts in ICA data has declined from 23% in 2003, to 21% in 2005 and 18% in 2007 surveys.\n\n\nAmong the listed firms, the debt levels relative to assets have declined from about 25% average\n\n\n(20% median) in 2001 to below 20% average (and below 15% median) in 2008. This trend raises\n\n\nconcern and more research is needed to understand the reasons for such “deleveraging.” The\n\n\nreasons could be supply-side constraints, with banks unwillingness to lend at rates attractive to\n\n\ncorporate sector or crowding out by the government sector, or demand-side weaknesses, such as\n\n\nlack of productive growth opportunities. Our data do not allow such analysis.\n\n\nLarge literature has established a solid link between access to finance, productive\n\n\ninvestment and growth on the macro and micro levels. The results of this study confirm that\n\n\nfirms in Egypt, especially small and medium firms, would benefit from increased access to\n\n\nfinance which is likely to support firm investment into productive assets (i.e. physical savings)\n\n\nand stimulate\n\nMention: \"ICA data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Credit product usage declines across these successive rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004825:12:0", "occurrence_id": "a5d50fc77df4e8725b06afda", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Page 9\n\n\nAlbanian (with the expected negative sign) was lack of cohesion within members of the community.\nStudents‟ excellence in math, however, was correlated with a larger set of factors. The proportion\nof students excelling in math was lower in schools with poor facilities. While teachers‟\ncharacteristics did not appear significant, teacher appraisal and transparency with respect to\nteachers‟ absence from school were; indeed, students‟ excellence in math was more likely in\nschools where teachers believed they would be penalized if they underperformed, and where a\nteacher absence register existed and was regularly used. Lack of parental involvement in the school\nalso correlated negatively, with students‟ performance in the final math exam. A father‟s education\nwas strongly and positively associated with children‟s performance in math. And the extent of\ndivide (lack of perceived cohesion) in the community appears to matter by reducing the proportion\nof students excelling in both math and language.\n\n\n**_1.c.4. Lab-type experiments and parents’ demand for accountability_**\n\n\n31. The low levels of parent participation in school governance structures could reflect lack of\ninformation or parents‟ reluctance to hold teachers accountable. As discussed earlier **,** Albania has\nmade concerted efforts to improve governance across the public sector. Its efforts in the education\nsector have been received positively by households. Seeking to complement governance\nimprovements in education compact and the supply of education services, the Government has been\ninvesting in shortening the route to accountability, by strengthening participatory accountability\nstructures (see again Figure 1). However, the data collected in the Albania School Stakeholder\nSurvey indicate that parental participation in these structures is still very low. Is this lack of\nparticipation owing to poor institutions, a lack of knowledge and understanding, or a preference not\nto participate?\n\n\n32. Parents appear seriously under-informed about accountability structures in their children‟s\nschools, and in many places these structures are not operating as they should. The findings\nsummarized previously,\n\nMention: \"teacher absence register\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Where such registers exist and are used, absence falls.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004825:12:0", "occurrence_id": "b1f5784e508728455f062b4a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Page 9\n\n\nAlbanian (with the expected negative sign) was lack of cohesion within members of the community.\nStudents‟ excellence in math, however, was correlated with a larger set of factors. The proportion\nof students excelling in math was lower in schools with poor facilities. While teachers‟\ncharacteristics did not appear significant, teacher appraisal and transparency with respect to\nteachers‟ absence from school were; indeed, students‟ excellence in math was more likely in\nschools where teachers believed they would be penalized if they underperformed, and where a\nteacher absence register existed and was regularly used. Lack of parental involvement in the school\nalso correlated negatively, with students‟ performance in the final math exam. A father‟s education\nwas strongly and positively associated with children‟s performance in math. And the extent of\ndivide (lack of perceived cohesion) in the community appears to matter by reducing the proportion\nof students excelling in both math and language.\n\n\n**_1.c.4. Lab-type experiments and parents’ demand for accountability_**\n\n\n31. The low levels of parent participation in school governance structures could reflect lack of\ninformation or parents‟ reluctance to hold teachers accountable. As discussed earlier **,** Albania has\nmade concerted efforts to improve governance across the public sector. Its efforts in the education\nsector have been received positively by households. Seeking to complement governance\nimprovements in education compact and the supply of education services, the Government has been\ninvesting in shortening the route to accountability, by strengthening participatory accountability\nstructures (see again Figure 1). However, the data collected in the Albania School Stakeholder\nSurvey indicate that parental participation in these structures is still very low. Is this lack of\nparticipation owing to poor institutions, a lack of knowledge and understanding, or a preference not\nto participate?\n\n\n32. Parents appear seriously under-informed about accountability structures in their children‟s\nschools, and in many places these structures are not operating as they should. The findings\nsummarized previously,\n\nMention: \"Albania School Stakeholder\nSurvey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"It shows parental participation in accountability structures remains low.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004802:8:0", "occurrence_id": "fcea313cf826411dd3c41090", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Child mortality as the primary outcome variable of interest in this paper is generally defined\n\n\nas the number of children dying before the age of five per 1000 children born alive. This\n\n\nnumber can either be computed as a cohort measure, which means that a given birth cohort is\n\n\nfollowed over a five year period, or as a period measure, in which age-specific mortality rates\n\n\nfrom different birth cohorts are combined for a given time window. Even though cohort\n\n\nmeasures may seem more intuitive, they are problematic from an analytical perspective, since\n\n\nthe estimated mortality rates reflect risk exposure over an extended period of time with\n\n\npotentially large changes in the underlying risk structure.\n\n\nTo minimize the time difference between observed water and sanitation infrastructure and\n\n\nmortality outcomes, we limit our analysis to mortality within the 12-months-window\n\n\npreceding the household interview. The dependent variable we construct is a binary variable\n\n\ncapturing the probability of a child between the age of 0 and 48 months dying over the 12\n\n\nmonths interval leading up to the interview. This mortality variable thus excludes children\n\n\nwho died more than 12 months before the respective DHS survey, children older than 48\n\n\nmonths one year prior to the interview, and children born less than 12 months prior to the\n\n\ninterview.\n\n\nEven though the Demographic and Health Surveys (DHS) standardize most variables in their\n\n\npublished data, this is unfortunately not true for variables on water and sanitation\n\n\ninfrastructure. Given the vast array of different technologies used across countries, the detail\n\n\nand structure of the collected information on water and sanitation infrastructure differs widely\n\n\nacross the 38 surveys used in this paper. In total, we found about 150 different sanitation and\n\n\n200 different water codes in our sample. To make the collected information comparable\n\n\nacross countries, we developed a few simple coding rules, which map the various types of\n\n\nwater and sanitation infrastructure into three broad categories. Since we want to explicitly\n\n\ndifferentiate between different types of water and sanitation technologies, we follow a coding\n\n\nsimilar to the water and sanitation ladder proposed by the WHO/UNICEF\n\nMention: \"DHS survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its birth histories define the eligible child mortality window.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004802:8:0", "occurrence_id": "c55a7b3ba7d966224488e9b7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Child mortality as the primary outcome variable of interest in this paper is generally defined\n\n\nas the number of children dying before the age of five per 1000 children born alive. This\n\n\nnumber can either be computed as a cohort measure, which means that a given birth cohort is\n\n\nfollowed over a five year period, or as a period measure, in which age-specific mortality rates\n\n\nfrom different birth cohorts are combined for a given time window. Even though cohort\n\n\nmeasures may seem more intuitive, they are problematic from an analytical perspective, since\n\n\nthe estimated mortality rates reflect risk exposure over an extended period of time with\n\n\npotentially large changes in the underlying risk structure.\n\n\nTo minimize the time difference between observed water and sanitation infrastructure and\n\n\nmortality outcomes, we limit our analysis to mortality within the 12-months-window\n\n\npreceding the household interview. The dependent variable we construct is a binary variable\n\n\ncapturing the probability of a child between the age of 0 and 48 months dying over the 12\n\n\nmonths interval leading up to the interview. This mortality variable thus excludes children\n\n\nwho died more than 12 months before the respective DHS survey, children older than 48\n\n\nmonths one year prior to the interview, and children born less than 12 months prior to the\n\n\ninterview.\n\n\nEven though the Demographic and Health Surveys (DHS) standardize most variables in their\n\n\npublished data, this is unfortunately not true for variables on water and sanitation\n\n\ninfrastructure. Given the vast array of different technologies used across countries, the detail\n\n\nand structure of the collected information on water and sanitation infrastructure differs widely\n\n\nacross the 38 surveys used in this paper. In total, we found about 150 different sanitation and\n\n\n200 different water codes in our sample. To make the collected information comparable\n\n\nacross countries, we developed a few simple coding rules, which map the various types of\n\n\nwater and sanitation infrastructure into three broad categories. Since we want to explicitly\n\n\ndifferentiate between different types of water and sanitation technologies, we follow a coding\n\n\nsimilar to the water and sanitation ladder proposed by the WHO/UNICEF\n\nMention: \"Demographic and Health Surveys\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Most variables are standardised, though not the ones needed here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004827:10:0", "occurrence_id": "3ec3f50c3f23776a04d6537e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The Cases of Rice and Wheat**\n\n\nThe two food commodities that have received the most attention because of price\n\n\nsurges are the key staples of wheat and rice. The length of their international price\n\n\nspikes around 1974 were broadly similar to those around 2008, but the height of the\n\n\nspike – particularly for rice – was greater in 1974. The recent price rises were more\n\n\ngradual except in the final months, so we consider an extra year in the lead-up to the\n\n\n2008 spike.\n\n\nEstimates of the Ti’s are available for all key rice and wheat countries, in the\n\n\nform of nominal assistance coefficients (NACs), from three sources. Anderson and\n\n\nValenzuela (2008) provide them to 2004 for developing countries and to 2007 for\n\n\nhigh-income countries (summarized in Anderson 2009). They are similarly available\n\n\nfor high-income countries for 2008 in OECD (2010). For developing countries,\n\n\nAnderson and Nelgen (2010b) provide estimates based on FAO and World Bank data\n\n\non producer and border prices, respectively, for 2005 to 2008. The most-recent\n\n\ndeveloping country estimates are less reliable than the NAC estimates in Anderson\n\n\nand Valenzuela (2008), for several reasons. One is that the coverage is not as\n\n\nextensive, because domestic prices are not available for some countries. Another is\n\n\nthat the FAO’s producer prices and World Bank international prices are not always as\n\n\nreliable as previously-used domestic and border prices from national statistical\n\n\nagencies. The FAO producer prices in US current dollars (FAO 2010) were converted\n\n\ninto an index set at 100 for 2004, and the 2004 US dollar prices in Anderson and\n\n\nValenzuela (2008) updated using the changes in these indexes through 2008.\n\n\nLikewise, the Thailand 5 percent broken rice and Canadian wheat prices from the\n\n\nWorld Bank were converted to indexes set at 100 for 2004, and the 2004 border\n\n\nprices in Anderson and Valenzuela (2008) updated using changes in those indexes\n\n\nthrough 2008.\n\n\nThese NAC estimates are reported\n\nMention: \"World Bank data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Border price estimates for developing countries build on these data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004827:10:0", "occurrence_id": "e1240620bba93b62f2b647ed", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The Cases of Rice and Wheat**\n\n\nThe two food commodities that have received the most attention because of price\n\n\nsurges are the key staples of wheat and rice. The length of their international price\n\n\nspikes around 1974 were broadly similar to those around 2008, but the height of the\n\n\nspike – particularly for rice – was greater in 1974. The recent price rises were more\n\n\ngradual except in the final months, so we consider an extra year in the lead-up to the\n\n\n2008 spike.\n\n\nEstimates of the Ti’s are available for all key rice and wheat countries, in the\n\n\nform of nominal assistance coefficients (NACs), from three sources. Anderson and\n\n\nValenzuela (2008) provide them to 2004 for developing countries and to 2007 for\n\n\nhigh-income countries (summarized in Anderson 2009). They are similarly available\n\n\nfor high-income countries for 2008 in OECD (2010). For developing countries,\n\n\nAnderson and Nelgen (2010b) provide estimates based on FAO and World Bank data\n\n\non producer and border prices, respectively, for 2005 to 2008. The most-recent\n\n\ndeveloping country estimates are less reliable than the NAC estimates in Anderson\n\n\nand Valenzuela (2008), for several reasons. One is that the coverage is not as\n\n\nextensive, because domestic prices are not available for some countries. Another is\n\n\nthat the FAO’s producer prices and World Bank international prices are not always as\n\n\nreliable as previously-used domestic and border prices from national statistical\n\n\nagencies. The FAO producer prices in US current dollars (FAO 2010) were converted\n\n\ninto an index set at 100 for 2004, and the 2004 US dollar prices in Anderson and\n\n\nValenzuela (2008) updated using the changes in these indexes through 2008.\n\n\nLikewise, the Thailand 5 percent broken rice and Canadian wheat prices from the\n\n\nWorld Bank were converted to indexes set at 100 for 2004, and the 2004 border\n\n\nprices in Anderson and Valenzuela (2008) updated using changes in those indexes\n\n\nthrough 2008.\n\n\nThese NAC estimates are reported\n\nMention: \"FAO producer prices\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its dollar prices are converted into an index based on that year.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004828:14:0", "occurrence_id": "f9e48dd6c7e24ece458ff009", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The problem with these data on disbursement lags however is that initial disbursement\n\n\nprojections are tentative and often revised over the life of the project. Thus, they are a kind of a\n\n\n“moving target” if we want to measure what actual disbursements delays are. In light of this, we\n\n\ninstead measure disbursement delays relative to typical disbursement rates for projects approved in the\n\n\nsame region, year, and sector, drawing from Kraay (2010). In particular, for each project we construct a\n\n\ntypical disbursement profile based on the actual average annual disbursement rate for all projects in the\n\n\nsame sector, approval year, and geographical region.  For each year of the project, we then calculate\n\n\nthe difference between the cumulative actual annual disbursement rate and the cumulative predicted\n\n\nannual disbursement rate. Finally, we take this cumulative disbursement lag over the first half of the\n\n\nproject, and consider it as a potential leading indicator of project success ( _DISBLAGH1_ ).\n\n\nTable 2 contains summary statistics on the main variables of interest. These are reported for\n\n\ntwo samples that are defined based on various data availability constraints. Our first sample begins in\n\n\n1983 (i.e. with projects evaluated in 1983), and contains 6,253 subsequent project evaluations (after\n\n\neliminating 548 observations with missing data on the relevant right-hand-side variables). In this\n\n\nsample, the outcome rating is a binary satisfactory/unsatisfactory, and we have data available on the\n\n\nindicated basic project characteristics as well as the ISR flags data.  There are a further 810 project\n\n\nevaluations between 1972 and 1982 but we do not include them as the ISR flag data, as well as the\n\n\npreparation and supervision costs data, is not available from this earlier period. Our second sample\n\n\nconsists of 3,887 evaluations performed since 1995. For this sample, we use the available six-point scale\n\n\nfor the outcomes, and also a richer set of control variables including the disbursement lags and a limited\n\n\nset task manager variables, that we\n\nMention: \"ISR flags data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Project characteristic flags are available for evaluations from that year on.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004829:61:1", "occurrence_id": "ac48a47aee8b9528f81abce0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  grades 1-4 of urban\nprimary schools, how many were enrolled in grades 1-4 of rural primary schools, how many were\nenrolled in grades 5-8 of urban primary schools, and how many were enrolled in grade 5-8 of\nrural primary schools. Beginning in 2007, under FUNDEB, the algorithm was complicated even\nmore, to depend (in addition to all these previous factors) on the number of children enrolled in\nurban vs. rural creches, preschools, secondary schools, and adult education (EJA). Fortunately,\nthe annual Census of Schools contains all of the required information, as it is the basis for the\ndistribution of FUNDEF funds. Minist´erio da Educa¸c˜ao (2008) describes the changes initiated\nin 2007.\n\nWhile the laws induced endogenous municipal responses, I can simulate the municipality’s\nvirtual tax revenue by using the exact algorithms of the laws in each year, but using pre-reform\n(1997) data on revenue and enrollments. To better predict how enrollments will expand over\nthe sample period without introducing endogenous information, I exploit the nation-wide rate\nat which municipal governments took over the state’s role in primary education during 19982008. Primary education is usually run by municipal governments, but the state government\nis involved in some municipalities more than in others. I wanted to account for the fact that\nsome municipalities had a greater potential to secure higher revenue through take-over than did\nothers. The federal rate of takeover is exogenous to pre-primary education policy in any given\nmunicipality. However, by taking this into account and thus allowing simulated revenue to grow\nmore in municipalities with heavy initial state and federal involvement in primary education in the\npre-period, I can better predict revenue without drawing on any one municipality’s endogenous\nresponse to the reforms.\n\nThe additional instruments are computed simply by interacting simulated revenue with the\nfollowing variables: year 2000 median per capita household income,\n\nMention: \"annual Census of Schools\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"It holds the school characteristics used to allocate education funds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004831:15:1", "occurrence_id": "bb9f361fb9c2bd710864bc30", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  is 0:15. Finally, data for\nthe gravity variables come from the CEPII database.\n\n###### 4.3 Descriptive Statistics\n\n\nAppendix 8:F reports some statistics at the …rm level for each exporting country.\n\"Risky\" products account for an important share of …rms total exports in all …ve\ncountries. Additionally, \"risky\" …rms–exporting at least one \"risky\" product–represent\naround half of the total …rm population in all …ve countries. Non \"risky\" …rms are …rms\nthat export no \"risky\" products at all. Appendices 8:G and 8:H report some statistics\nfor our survival data. Considering …rms in all countries the average spell duration is\nabout one year and four months and the median duration is only one year. Almost 40%\nof the spells are right censored, and 17% are left censored. Considering each country\nindividually, Senegal exhibits the highest average spell duration, and Ghana the lowest.\nA large proportion of spells, 56%, start with trade values lower than 10’000 dollars, 13%\nare initiated with trade values higher than 100’000 dollars and only 3% start with initial\ntrade values greater than 1’000’000 dollars. Dropping all spells with initial trade value\ninferior to 10’000 dollars (100’000, or 1’000’000 dollars) increase (even more) the average\nand median spell duration. The higher the initial trade value the higher the probability\nto survive. These results are in line with …ndings in previous empirical studies (Besedes\nand Prusa, 2006a, 2006b).\n\nWe now move on to characterize the duration of trade relationships non paramet\n\n14\n\nMention: \"CEPII database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The gravity regression variables are taken from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004834:40:1", "occurrence_id": "a35ad1d651fd6240e2041a3a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: )\nSquared poverty gap 6.66 8.03 7.69 7.92 8.01 7.57\n\n(0.06) (0.07) (0.07) (0.08) (0.08) (0.07)\n_Inequality_\n\nGini 55.7 57.0 56.7 56.9 57.0 56.6\n\n(0.002) (0.002) (0.002) (0.002) (0.002) (0.002)\n\n\n<u>Source: IBGE – National Household Income Survey (PNAD), 2006.</u>\nNote: Standard errors in parentheses.  Extreme and Moderate Poverty Line can be found in table 2 in the the annex.\nHeadcount, poverty gap and squared poverty gap measures are also known as FGT (0, 1 and 2) respectively (Foster et al. (1984))\n\n\n**<u>Table 9: Food Price Effects on Poverty and Inequality (All Brazil) – Alpha =1</u>**\n\n\n\nExpenditure\n\nand Bolsa\n<u>Familia Effect</u>\n\n\n\nExpenditure\n\nand BPC\n\n<u>Effect</u>\n\n\n\nExpenditure\nBaseline\n\nEffect\n\n\n\nExpenditure\n\nand Market\n\n<u>Income</u>\n\n\n\nTotal\n\n\n\n_Extreme Poverty_\n\nHeadcount 11.04 13.53 12.43 13.42 13.49 12.30\n\n(0.14) (0.11) (0.14) (0.14) (0.14) (0.14)\nPoverty gap 3.80 4.78 4.33 4.67 4.77 4.21\n\n(0.05) (0.07) (0.05) (0.06) (0.07) (0.06)\nSquared poverty gap 1.95 2.47 2.22 2.38 2.47 2.14\n\n(\n\nMention: \"National Household Income Survey (PNAD)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Poverty line coefficients with standard errors are estimated on this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004836:6:1", "occurrence_id": "34da53a9bebc0d517de6c88c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  log of real GDP and log of its area. Real\nGDP is measured in constant 2005 US dollars at PPP. I follow Dollar and Kraay (2003) and Alcala and\nCiccone (2004) to use the PPP for converting GDP values into US dollars. Presenting GDP in PPP is more\nappropriate than deflating with market exchange rates for cross-country analysis. As for the area of the\ncountry, studies like Rodrik (1998) and Frankel and Rose (2000) shows that area has a significantly\nnegative impact on openness. Large countries are less likely to trade because of relatively higher\ndomestic demand and higher transport costs of exporting abroad. Moreover small size limits the\ncountry’s possibilities to diversify production. Another explanatory variable is the remoteness of the\ncountry from the rest of the world. Anderson and Van Wincoop (2003) show that a country’s trade with\nany partner country depends on its average distance from the rest of the world. Following the method\nintroduced in Head (2003) I define a remoteness index. <sup>6</sup> [^6: Remoteness is defined as _remotei_  1/  _jGDPj_ / _Distij_  where Distij is the distance between two countries _i_ and _j_ .] Finally, I control for the past export growth\nperformance of the country. This variable can control for the positive steps taken in the past to spur\nexport performance such as implementation of trade liberalization policies, expansion of trade into new\nmarkets, exports of new products, or establishing trade agreements.\n\n\nExport performance of countries is measured by log of export sales in constant 2005 US dollars\nwhich is obtained from World Bank Development Indicators (WDI). By measuring exports in constant US\ndollars, I assume that exported goods’ prices are roughly equalized across countries. Endogeneity is a\nmajor concern in the estimation, especially for the relationship between GDP and export. Gravity\nmodels assume that the volume of trade between two trading countries is positively related to the GDPs\n\n\n6 Remoteness is defined as _remotei_  1/  _jGDPj_\n\nMention: \"World Bank Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Export sales in constant dollars are obtained from this compilation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004839:29:0", "occurrence_id": "e09a54bf6d04603b83fb7a58", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: and the synergy effect has a negative and significant coefficient, implying that\nfirms belonging to industries with high asset tangibility (essentially capitalintensive industries) are _less_ sensitive to the synergy effect. <sup>17</sup>\n\n\nGiven that these interactions have significant explanatory power, we combine them to address a potential omitted variable bias, and to compare there\ncoefficients in a joint multivariate regression. In columns (4) and (5) we\ncombine one regressors capturing each of the three hypothesis that these interactions try to capture: information, financial constraints or capabilities.\nWe find that typically the same signs, magnitudes and statistical significance\nlevels persist as the ones just discussed, and the interpretations from above\nare valid when we control for all these effects simultaneously.\n\n#### **5 Concluding remarks**\n\n\nIn spite of their growing interest for the profession, firm-level datasets are\nstill rare for low-income countries, and virtually inexistent for African countries. Our exploration of African customs data on firm-level exports revealed\na set of stylized facts that are consistent with evidence from previous studies\nanalyzing rich or middle-income countries. We showed that exporters in our\nset of African countries experiment a lot on export markets, at a low scale\nand with low survival rates, particularly in the first year. That is, they operate in a difficult environment characterized by very high “infant-mortality\nrates”. Therefore we investigate more in detail what determines if they survive beyond their first year.\n\n\nThe most striking finding coming out of our analysis —and which could\nnot be observed on the product-level data used by previous studies of export\nsurvival—is that exporters of similar goods to the same destination exert a\n_positive_ externality on new entrants. That is, the more they are, the higher\nthe survival probability of new entrants—although the effect is relatively\nsmall. This finding is at first sight surprising, as one might expect that\n\nMention: \"African customs data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Firm-level export records reveal a set of stylised facts.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004850:6:0", "occurrence_id": "fa3d4e4049157b616f4551fe", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Not only are health professionals not the majority of high-skilled emigrants, but they\n\n\nappear to have lower emigration rates on average than other skilled professionals. We compare\n\n\nthe overall brain drain rates for tertiary educated migrants who emigrated at age 22 or higher\n\n\n(Beine et al, 2007) to the medical brain drain rates in Bhargava et al. (2010). Across 161\n\n\ncountries, the median medical brain drain rate is 5.4 percent, compared to a median skilled brain\n\n\ndrain rate of 8.4 percent. The skilled brain drain rate exceeds the medical brain drain rate for 69\n\n\npercent of the countries in this sample.\n\n\nMattoo et al. (2010) note that there is often a concern that not all educated migrants end\n\n\nup working in skilled occupations after they have migrated— a phenomenon which they call\n\n\n―brain waste.‖ However, Table 1 shows that the most common occupations for educated\n\n\nmigrants are skilled occupations, particularly those in the so-called STEM fields (science,\n\n\ntechnology, engineering and mathematics). Moreover, using the same 2008 sample we calculate\n\n\nthat 79 percent of working migrants from developing countries with a bachelors‘ degree or more\n\n\nare working in occupations in the United States in which the majority of workers have post\n\nsecondary education, as are 90 percent of those with a masters degree or more, and 96 percent of\n\n\nthose with a Ph.D. The stereotype of foreign workers with Ph.D.s driving taxis is certainly the\n\n\nexception; only 2 out of 1,936 developing country migrants with Ph.D.s in the American\n\n\nCommunity Survey sample are taxi drivers. This is in line with Mattoo et al.‘s (2010) work using\n\n\nthe 2000 census, where they find it is mainly skilled migrants from non-English speaking\n\n\ncountries with poor quality education systems who struggle to find skilled work—a finding\n\n\nwhich might mean that the actual skill level of these migrants is lower than their education would\n\n\nsuggest.\n\n\n**Question 2:\n\nMention: \"American\n\n\nCommunity Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Occupational outcomes of developing-country doctoral migrants are observed here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004850:6:0", "occurrence_id": "1aa7b2377d26ddab6403cf06", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Not only are health professionals not the majority of high-skilled emigrants, but they\n\n\nappear to have lower emigration rates on average than other skilled professionals. We compare\n\n\nthe overall brain drain rates for tertiary educated migrants who emigrated at age 22 or higher\n\n\n(Beine et al, 2007) to the medical brain drain rates in Bhargava et al. (2010). Across 161\n\n\ncountries, the median medical brain drain rate is 5.4 percent, compared to a median skilled brain\n\n\ndrain rate of 8.4 percent. The skilled brain drain rate exceeds the medical brain drain rate for 69\n\n\npercent of the countries in this sample.\n\n\nMattoo et al. (2010) note that there is often a concern that not all educated migrants end\n\n\nup working in skilled occupations after they have migrated— a phenomenon which they call\n\n\n―brain waste.‖ However, Table 1 shows that the most common occupations for educated\n\n\nmigrants are skilled occupations, particularly those in the so-called STEM fields (science,\n\n\ntechnology, engineering and mathematics). Moreover, using the same 2008 sample we calculate\n\n\nthat 79 percent of working migrants from developing countries with a bachelors‘ degree or more\n\n\nare working in occupations in the United States in which the majority of workers have post\n\nsecondary education, as are 90 percent of those with a masters degree or more, and 96 percent of\n\n\nthose with a Ph.D. The stereotype of foreign workers with Ph.D.s driving taxis is certainly the\n\n\nexception; only 2 out of 1,936 developing country migrants with Ph.D.s in the American\n\n\nCommunity Survey sample are taxi drivers. This is in line with Mattoo et al.‘s (2010) work using\n\n\nthe 2000 census, where they find it is mainly skilled migrants from non-English speaking\n\n\ncountries with poor quality education systems who struggle to find skilled work—a finding\n\n\nwhich might mean that the actual skill level of these migrants is lower than their education would\n\n\nsuggest.\n\n\n**Question 2:\n\nMention: \"2000 census\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Comparable findings on skilled migrants come from this enumeration.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004848:72:0", "occurrence_id": "e94d12c7928fe0f2c8c6dfd7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Data Appendix**\n\n\n**1.** **Quarterly Public Finance Database:** This dataset includes quarterly time series for total government revenue, tax\nrevenue, revenues from social security, total expenditures (current and capital), interest payments and transfers. The\ndata refer to the consolidated general government or the consolidated central government (including social security),\ndepending on the existence of quarterly accounts for the general government in each country. Data are from national\nsources, with the exception of the UK and Canada from the OECD Economic Outlook and Belgium, Bulgaria, Czech\nRepublic, Cyprus, France, Germany, Hungary, Poland, Slovakia and Slovenia from Eurostat. Links to these sources\nare provided below. For the Latin American and Asian countries in the sample the data correspond to central\ngovernment accounts. For the remaining countries the data refers to the Consolidated General Government.\n\n\n**2.** **Annual Tax Revenue Breakdown:** This dataset contains the breakdown of tax revenues by their components:\nPersonal Income Tax Revenue, Corporate Income Tax Revenue, Value Added Tax Revenues (or Sales Tax Revenues,\nfor countries with no VAT) and Social Security Contribution Revenues. This database was constructed separately\nfrom the Quarterly Public Finance Dataset since these series are not always reported together and, many times, the tax\nrevenue breakdown is in annual frequency. For Latin American and Asian countries, where Central Government is\nbeing considered, only federal tax revenues are included. Data was taken from national sources, with the exception of\nArgentina, Chile, Colombia and Mexico from ECLAC; Canada, UK, France, Germany, Belgium, Czech Republic,\nPoland, Slovakia and Hungary from OECDstats and Cyprus from the IMF’s IFS.\n\n\n**3.** **Statutory Tax Rates:** The data related to the statutory tax rates, comprised by tax brackets and rates of Personal\nIncome Tax, Corporate Income Tax, Social Security and VAT, comes from the OECD\n\nMention: \"Quarterly Public Finance Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Quarterly government revenue and tax series are assembled in this dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004848:72:0", "occurrence_id": "fe4a267cbc8dcac76ccefa33", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Data Appendix**\n\n\n**1.** **Quarterly Public Finance Database:** This dataset includes quarterly time series for total government revenue, tax\nrevenue, revenues from social security, total expenditures (current and capital), interest payments and transfers. The\ndata refer to the consolidated general government or the consolidated central government (including social security),\ndepending on the existence of quarterly accounts for the general government in each country. Data are from national\nsources, with the exception of the UK and Canada from the OECD Economic Outlook and Belgium, Bulgaria, Czech\nRepublic, Cyprus, France, Germany, Hungary, Poland, Slovakia and Slovenia from Eurostat. Links to these sources\nare provided below. For the Latin American and Asian countries in the sample the data correspond to central\ngovernment accounts. For the remaining countries the data refers to the Consolidated General Government.\n\n\n**2.** **Annual Tax Revenue Breakdown:** This dataset contains the breakdown of tax revenues by their components:\nPersonal Income Tax Revenue, Corporate Income Tax Revenue, Value Added Tax Revenues (or Sales Tax Revenues,\nfor countries with no VAT) and Social Security Contribution Revenues. This database was constructed separately\nfrom the Quarterly Public Finance Dataset since these series are not always reported together and, many times, the tax\nrevenue breakdown is in annual frequency. For Latin American and Asian countries, where Central Government is\nbeing considered, only federal tax revenues are included. Data was taken from national sources, with the exception of\nArgentina, Chile, Colombia and Mexico from ECLAC; Canada, UK, France, Germany, Belgium, Czech Republic,\nPoland, Slovakia and Hungary from OECDstats and Cyprus from the IMF’s IFS.\n\n\n**3.** **Statutory Tax Rates:** The data related to the statutory tax rates, comprised by tax brackets and rates of Personal\nIncome Tax, Corporate Income Tax, Social Security and VAT, comes from the OECD\n\nMention: \"OECD Economic Outlook\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Two countries' fiscal series come from this publication instead.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004848:72:0", "occurrence_id": "4625754e2bda13076a961898", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Data Appendix**\n\n\n**1.** **Quarterly Public Finance Database:** This dataset includes quarterly time series for total government revenue, tax\nrevenue, revenues from social security, total expenditures (current and capital), interest payments and transfers. The\ndata refer to the consolidated general government or the consolidated central government (including social security),\ndepending on the existence of quarterly accounts for the general government in each country. Data are from national\nsources, with the exception of the UK and Canada from the OECD Economic Outlook and Belgium, Bulgaria, Czech\nRepublic, Cyprus, France, Germany, Hungary, Poland, Slovakia and Slovenia from Eurostat. Links to these sources\nare provided below. For the Latin American and Asian countries in the sample the data correspond to central\ngovernment accounts. For the remaining countries the data refers to the Consolidated General Government.\n\n\n**2.** **Annual Tax Revenue Breakdown:** This dataset contains the breakdown of tax revenues by their components:\nPersonal Income Tax Revenue, Corporate Income Tax Revenue, Value Added Tax Revenues (or Sales Tax Revenues,\nfor countries with no VAT) and Social Security Contribution Revenues. This database was constructed separately\nfrom the Quarterly Public Finance Dataset since these series are not always reported together and, many times, the tax\nrevenue breakdown is in annual frequency. For Latin American and Asian countries, where Central Government is\nbeing considered, only federal tax revenues are included. Data was taken from national sources, with the exception of\nArgentina, Chile, Colombia and Mexico from ECLAC; Canada, UK, France, Germany, Belgium, Czech Republic,\nPoland, Slovakia and Hungary from OECDstats and Cyprus from the IMF’s IFS.\n\n\n**3.** **Statutory Tax Rates:** The data related to the statutory tax rates, comprised by tax brackets and rates of Personal\nIncome Tax, Corporate Income Tax, Social Security and VAT, comes from the OECD\n\nMention: \"Eurostat\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Nine European countries' series are taken from this office.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004848:72:0", "occurrence_id": "0c49acf87dda3f79aa368c4b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Data Appendix**\n\n\n**1.** **Quarterly Public Finance Database:** This dataset includes quarterly time series for total government revenue, tax\nrevenue, revenues from social security, total expenditures (current and capital), interest payments and transfers. The\ndata refer to the consolidated general government or the consolidated central government (including social security),\ndepending on the existence of quarterly accounts for the general government in each country. Data are from national\nsources, with the exception of the UK and Canada from the OECD Economic Outlook and Belgium, Bulgaria, Czech\nRepublic, Cyprus, France, Germany, Hungary, Poland, Slovakia and Slovenia from Eurostat. Links to these sources\nare provided below. For the Latin American and Asian countries in the sample the data correspond to central\ngovernment accounts. For the remaining countries the data refers to the Consolidated General Government.\n\n\n**2.** **Annual Tax Revenue Breakdown:** This dataset contains the breakdown of tax revenues by their components:\nPersonal Income Tax Revenue, Corporate Income Tax Revenue, Value Added Tax Revenues (or Sales Tax Revenues,\nfor countries with no VAT) and Social Security Contribution Revenues. This database was constructed separately\nfrom the Quarterly Public Finance Dataset since these series are not always reported together and, many times, the tax\nrevenue breakdown is in annual frequency. For Latin American and Asian countries, where Central Government is\nbeing considered, only federal tax revenues are included. Data was taken from national sources, with the exception of\nArgentina, Chile, Colombia and Mexico from ECLAC; Canada, UK, France, Germany, Belgium, Czech Republic,\nPoland, Slovakia and Hungary from OECDstats and Cyprus from the IMF’s IFS.\n\n\n**3.** **Statutory Tax Rates:** The data related to the statutory tax rates, comprised by tax brackets and rates of Personal\nIncome Tax, Corporate Income Tax, Social Security and VAT, comes from the OECD\n\nMention: \"Annual Tax Revenue Breakdown\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Tax revenues are decomposed by component in this dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004848:72:0", "occurrence_id": "61c5a0f652d74aa2891f7c95", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Data Appendix**\n\n\n**1.** **Quarterly Public Finance Database:** This dataset includes quarterly time series for total government revenue, tax\nrevenue, revenues from social security, total expenditures (current and capital), interest payments and transfers. The\ndata refer to the consolidated general government or the consolidated central government (including social security),\ndepending on the existence of quarterly accounts for the general government in each country. Data are from national\nsources, with the exception of the UK and Canada from the OECD Economic Outlook and Belgium, Bulgaria, Czech\nRepublic, Cyprus, France, Germany, Hungary, Poland, Slovakia and Slovenia from Eurostat. Links to these sources\nare provided below. For the Latin American and Asian countries in the sample the data correspond to central\ngovernment accounts. For the remaining countries the data refers to the Consolidated General Government.\n\n\n**2.** **Annual Tax Revenue Breakdown:** This dataset contains the breakdown of tax revenues by their components:\nPersonal Income Tax Revenue, Corporate Income Tax Revenue, Value Added Tax Revenues (or Sales Tax Revenues,\nfor countries with no VAT) and Social Security Contribution Revenues. This database was constructed separately\nfrom the Quarterly Public Finance Dataset since these series are not always reported together and, many times, the tax\nrevenue breakdown is in annual frequency. For Latin American and Asian countries, where Central Government is\nbeing considered, only federal tax revenues are included. Data was taken from national sources, with the exception of\nArgentina, Chile, Colombia and Mexico from ECLAC; Canada, UK, France, Germany, Belgium, Czech Republic,\nPoland, Slovakia and Hungary from OECDstats and Cyprus from the IMF’s IFS.\n\n\n**3.** **Statutory Tax Rates:** The data related to the statutory tax rates, comprised by tax brackets and rates of Personal\nIncome Tax, Corporate Income Tax, Social Security and VAT, comes from the OECD\n\nMention: \"Quarterly Public Finance Dataset\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"The tax breakdown was built separately because series are not always reported together.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004848:72:0", "occurrence_id": "04004a5bf628eaad00a0bcf2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Data Appendix**\n\n\n**1.** **Quarterly Public Finance Database:** This dataset includes quarterly time series for total government revenue, tax\nrevenue, revenues from social security, total expenditures (current and capital), interest payments and transfers. The\ndata refer to the consolidated general government or the consolidated central government (including social security),\ndepending on the existence of quarterly accounts for the general government in each country. Data are from national\nsources, with the exception of the UK and Canada from the OECD Economic Outlook and Belgium, Bulgaria, Czech\nRepublic, Cyprus, France, Germany, Hungary, Poland, Slovakia and Slovenia from Eurostat. Links to these sources\nare provided below. For the Latin American and Asian countries in the sample the data correspond to central\ngovernment accounts. For the remaining countries the data refers to the Consolidated General Government.\n\n\n**2.** **Annual Tax Revenue Breakdown:** This dataset contains the breakdown of tax revenues by their components:\nPersonal Income Tax Revenue, Corporate Income Tax Revenue, Value Added Tax Revenues (or Sales Tax Revenues,\nfor countries with no VAT) and Social Security Contribution Revenues. This database was constructed separately\nfrom the Quarterly Public Finance Dataset since these series are not always reported together and, many times, the tax\nrevenue breakdown is in annual frequency. For Latin American and Asian countries, where Central Government is\nbeing considered, only federal tax revenues are included. Data was taken from national sources, with the exception of\nArgentina, Chile, Colombia and Mexico from ECLAC; Canada, UK, France, Germany, Belgium, Czech Republic,\nPoland, Slovakia and Hungary from OECDstats and Cyprus from the IMF’s IFS.\n\n\n**3.** **Statutory Tax Rates:** The data related to the statutory tax rates, comprised by tax brackets and rates of Personal\nIncome Tax, Corporate Income Tax, Social Security and VAT, comes from the OECD\n\nMention: \"OECDstats\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Several European countries' tax series come from this portal.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004848:72:0", "occurrence_id": "3212b36964f533fedc1f164a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Data Appendix**\n\n\n**1.** **Quarterly Public Finance Database:** This dataset includes quarterly time series for total government revenue, tax\nrevenue, revenues from social security, total expenditures (current and capital), interest payments and transfers. The\ndata refer to the consolidated general government or the consolidated central government (including social security),\ndepending on the existence of quarterly accounts for the general government in each country. Data are from national\nsources, with the exception of the UK and Canada from the OECD Economic Outlook and Belgium, Bulgaria, Czech\nRepublic, Cyprus, France, Germany, Hungary, Poland, Slovakia and Slovenia from Eurostat. Links to these sources\nare provided below. For the Latin American and Asian countries in the sample the data correspond to central\ngovernment accounts. For the remaining countries the data refers to the Consolidated General Government.\n\n\n**2.** **Annual Tax Revenue Breakdown:** This dataset contains the breakdown of tax revenues by their components:\nPersonal Income Tax Revenue, Corporate Income Tax Revenue, Value Added Tax Revenues (or Sales Tax Revenues,\nfor countries with no VAT) and Social Security Contribution Revenues. This database was constructed separately\nfrom the Quarterly Public Finance Dataset since these series are not always reported together and, many times, the tax\nrevenue breakdown is in annual frequency. For Latin American and Asian countries, where Central Government is\nbeing considered, only federal tax revenues are included. Data was taken from national sources, with the exception of\nArgentina, Chile, Colombia and Mexico from ECLAC; Canada, UK, France, Germany, Belgium, Czech Republic,\nPoland, Slovakia and Hungary from OECDstats and Cyprus from the IMF’s IFS.\n\n\n**3.** **Statutory Tax Rates:** The data related to the statutory tax rates, comprised by tax brackets and rates of Personal\nIncome Tax, Corporate Income Tax, Social Security and VAT, comes from the OECD\n\nMention: \"IFS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"One country's series are taken from this fund compilation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004855:11:0", "occurrence_id": "b6cab17f6c06b2e0d519c72c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 10\n\n\nsectional regressions are not always plausible. At the same time, previous studies have\n\n\ngenerally focused on democratization, interpreted broadly and typically using Polity\n\n\nindicators of democratization that reflect a range of _de facto_ performance characteristics of\n\n\ncountries (e.g., are executives constrained in fact, if not _de jure_ ?).\n\n\nGiven the fairly unsatisfactory state of the art in identifying the causal effects of\n\n\ninstitutional arrangements in cross-country comparisons, this paper is far less ambitious\n\n\nregarding documenting causal inference. However, it is more ambitious in its effort to\n\n\ndisaggregate the specific attributes of countries that favor development-friendly outcomes.\n\n\nRather than looking at broad and subjective indicators of democracy, such as those in Polity,\n\n\nit focuses on narrowly-defined and objectively-measured institutional arrangements.\n\n\nThe analysis is based on estimates of _yit_   0   1  ( _collective_ _actionit_ )   _it_  2  _it_,\n\n\nwhere _it_ indexes country-years; _y_ represents various development-relevant outcomes; _collective_\n\n\n_action_ represents a number of collective action variables, including competitive elections; and\n\n\n**X** a range of controls that might affect the costs to citizens of engaging in collective action\n\n\nindependent of the particular collective action variable that is the focus here. Yearly\n\n\nobservations exaggerate statistical significance, since most estimated effects are driven by\n\n\ncross-country differences that change little over time. To counter this effect, clustered\n\n\nstandard errors, which assume errors across observations within countries are not\n\n\nindependent, are reported. However, estimated effects from data averaged over country\n\nyears are similar to those based on country-year observations.\n\n\nThe standard strategy to minimize omitted variable bias in the estimates  1 of\n\n\ninstitutional effects on development outcomes relies on instrumental variables. The strategy\n\n\nis not viable in the current setting for two reasons. First, if one accepts the\n\nMention: \"Polity\n\n\nindicators of democratization\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"They capture de facto performance characteristics in the democratization literature.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004856:11:1", "occurrence_id": "a28c115ac2b31dd05ef4141d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Bureau. A key contribution of this\n\n\nstudy is the linking of firm level U.S. manufacturing data to industry level measures of EU product\n\n\nstandards and their relationship with international norms. This section outlines the main features\n\n\nof the datasets.\n\n\n**3.1** **U.S.** **Manufacturing** **firms** **Across** **Industries** **and** **Time**\n\n\nThe CM is conducted every five years and the empirical part of this paper makes use of CM\n\ninformation from 1992, 1997, and 2002. <sup>11</sup> The unit of observation for the Census is a manufacturing\n\n\nestablishment, or plant, and it contains detail information on inputs and output of all establish\nments. <sup>12</sup> For 1992 and 1997, plants are classified at the four-digit Standard Industrial Classification\n\n\nlevel (SIC4). In 2002, industry classification changed to the 6-digit North American Industry Clas\n\nones for which differentiation and variety is important.\n\n10See Bernard et al. (2009) for a complete description of the LFTTD and its construction. For an extensive\ndiscussion of the LBD see Jarmin and Miranda (2002).\n\n11Though CM data are available for earlier periods, I cannot use them because export information on the LFTTD\nis not available.\n\n12The CM imputes input usage data for small manufactures, referred to in the data as “administrative records”.\nAs it is customary in the U.S. microdata research —see Bernard, Redding and Schott (2010)—, these observations\nare excluded from the analysis.\n\n\n10\n\nMention: \"firm level U.S. manufacturing data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Firm records are linked to industry-level product standards.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004856:11:1", "occurrence_id": "42b9e14bbf1b4f6c6ce196b2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Bureau. A key contribution of this\n\n\nstudy is the linking of firm level U.S. manufacturing data to industry level measures of EU product\n\n\nstandards and their relationship with international norms. This section outlines the main features\n\n\nof the datasets.\n\n\n**3.1** **U.S.** **Manufacturing** **firms** **Across** **Industries** **and** **Time**\n\n\nThe CM is conducted every five years and the empirical part of this paper makes use of CM\n\ninformation from 1992, 1997, and 2002. <sup>11</sup> The unit of observation for the Census is a manufacturing\n\n\nestablishment, or plant, and it contains detail information on inputs and output of all establish\nments. <sup>12</sup> For 1992 and 1997, plants are classified at the four-digit Standard Industrial Classification\n\n\nlevel (SIC4). In 2002, industry classification changed to the 6-digit North American Industry Clas\n\nones for which differentiation and variety is important.\n\n10See Bernard et al. (2009) for a complete description of the LFTTD and its construction. For an extensive\ndiscussion of the LBD see Jarmin and Miranda (2002).\n\n11Though CM data are available for earlier periods, I cannot use them because export information on the LFTTD\nis not available.\n\n12The CM imputes input usage data for small manufactures, referred to in the data as “administrative records”.\nAs it is customary in the U.S. microdata research —see Bernard, Redding and Schott (2010)—, these observations\nare excluded from the analysis.\n\n\n10\n\nMention: \"CM\n\ninformation\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Conducted every five years, with these three rounds used here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004856:11:1", "occurrence_id": "d7a1d8afac527db91c9c9453", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Bureau. A key contribution of this\n\n\nstudy is the linking of firm level U.S. manufacturing data to industry level measures of EU product\n\n\nstandards and their relationship with international norms. This section outlines the main features\n\n\nof the datasets.\n\n\n**3.1** **U.S.** **Manufacturing** **firms** **Across** **Industries** **and** **Time**\n\n\nThe CM is conducted every five years and the empirical part of this paper makes use of CM\n\ninformation from 1992, 1997, and 2002. <sup>11</sup> The unit of observation for the Census is a manufacturing\n\n\nestablishment, or plant, and it contains detail information on inputs and output of all establish\nments. <sup>12</sup> For 1992 and 1997, plants are classified at the four-digit Standard Industrial Classification\n\n\nlevel (SIC4). In 2002, industry classification changed to the 6-digit North American Industry Clas\n\nones for which differentiation and variety is important.\n\n10See Bernard et al. (2009) for a complete description of the LFTTD and its construction. For an extensive\ndiscussion of the LBD see Jarmin and Miranda (2002).\n\n11Though CM data are available for earlier periods, I cannot use them because export information on the LFTTD\nis not available.\n\n12The CM imputes input usage data for small manufactures, referred to in the data as “administrative records”.\nAs it is customary in the U.S. microdata research —see Bernard, Redding and Schott (2010)—, these observations\nare excluded from the analysis.\n\n\n10\n\nMention: \"CM data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Earlier rounds are unusable because export information is missing.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004860:7:1", "occurrence_id": "9c36e7a1584f3dd5ba5eed77", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  still in school\n\n\n(Filmer and Schady, 2009; Glewwe and Olinto, 2004).\n\n\nSimilarly, the evidence of program effects on learning outcomes is limited and\n\n\nsomewhat mixed, making it difficult to draw conclusions. Improvements in cognitive\n\n2 The _Subsidios_ program was implemented in the late 2000´s in Bogotá and targeted vulnerable population classified as\n_Sisben_ level 1 and _Sisben_ 2 based on the proxy-means test constructed with information from a census of the poor. The\nprogram‘s transfers and conditions are similar to _Familias en Acción_, which is a nationwide and older program.\nHowever, the _Subsidios_ program offered two other treatments with a different structure of benefits. One is the ―savings\ntreatment,‖ which reserves one third of the bimonthly payment to be given to families as a lump sum at the end of the\nyear, just before enrollment into the subsequent grade level. The other is the ―tertiary treatment,‖ which also reserves\npart of the bimonthly payment, but then pays families a substantially larger amount if students graduate from high\nschool (eleventh grade). Students who continue to tertiary education are eligible to receive that amount one year earlier\nthan those who do not. These two additional treatments may change the way families respond to the program in two\nways: one is due to the reduction in the amount of cash families receive on a bimonthly basis, therefore limiting their\nliquidity; another is through the incentives created by linking cash transfers directly to grade progression and\nmatriculation in tertiary education.\n3 It is worth noting that the results of these two papers are not generalizable to the entire population covered by\n_Oportunidades_ given that the samples of analysis only contain non-migrant rural households. Migrants are expected to\nhave higher enrollment in college and better job opportunities, including higher salaries.\n\n\n6\n\nMention: \"census of the poor\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Its information builds the proxy means test that targets transfers.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004867:32:1", "occurrence_id": "a9f37dd23f223b6f2856b827", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  country figures as of 2005 from the power AICD database, downloadable from\n[http://www.infrastructureafrica.org/aicd/tools/data. For Benin, national access data were taken from World Bank (2009a) and are based on](http://www.infrastructureafrica.org/aicd/tools/data)\nestimates in 2006. Electricity consumption was taken from the CIA Factbook and is for 2007. Power outages were taken from World Bank\n(2009a) and are for 2009. Value lost to outages was taken from Enterprise Surveys 2009. Firms that indicate that power is a constraint and\nfirms that own generators taken from Enterprise surveys 2004. Collection rate is taken from World Bank 2009a and is for 2009. System losses\nare for 2009 and taken from World Bank 2009a. Tariff information is for 2009 and is from World Bank 2009a.\n\n- Represents information only for SBEE and does not include information for CEB; ** As of 2004\n\n\nThe power distribution utility in Benin, SBEE, is relatively efficient. Its system losses are modest\nrelative to African peers, and its collection rates are relatively high. Total system losses are estimated at\n21 percent, with 17 percent from distribution. While these losses are high relative to the 10 percent\ninternational benchmark, system losses are lower than in comparable low-income countries and even with\nthe level of middle-income countries in Sub Saharan Africa (table 11). In other West African countries\nlosses are higher: Togo has been burdened with transmission losses of around 30 percent (2004), Côte\nd’Ivoire around 23 percent (2008) (Tallapragada and others 2009). SBEE has been able to collect 96\npercent of its billing, above the average for low-income peers (table 11).\n\n\n**Challenges**\n\n\nImportant differences in access to electricity remain between urban and rural populations. Whereas 53\npercent of urban dwellers have access to electricity\n\nMention: \"CIA Factbook\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Electricity consumption for that year comes from this almanac.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004867:32:1", "occurrence_id": "27463f0d2bfb226a4a498a07", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  country figures as of 2005 from the power AICD database, downloadable from\n[http://www.infrastructureafrica.org/aicd/tools/data. For Benin, national access data were taken from World Bank (2009a) and are based on](http://www.infrastructureafrica.org/aicd/tools/data)\nestimates in 2006. Electricity consumption was taken from the CIA Factbook and is for 2007. Power outages were taken from World Bank\n(2009a) and are for 2009. Value lost to outages was taken from Enterprise Surveys 2009. Firms that indicate that power is a constraint and\nfirms that own generators taken from Enterprise surveys 2004. Collection rate is taken from World Bank 2009a and is for 2009. System losses\nare for 2009 and taken from World Bank 2009a. Tariff information is for 2009 and is from World Bank 2009a.\n\n- Represents information only for SBEE and does not include information for CEB; ** As of 2004\n\n\nThe power distribution utility in Benin, SBEE, is relatively efficient. Its system losses are modest\nrelative to African peers, and its collection rates are relatively high. Total system losses are estimated at\n21 percent, with 17 percent from distribution. While these losses are high relative to the 10 percent\ninternational benchmark, system losses are lower than in comparable low-income countries and even with\nthe level of middle-income countries in Sub Saharan Africa (table 11). In other West African countries\nlosses are higher: Togo has been burdened with transmission losses of around 30 percent (2004), Côte\nd’Ivoire around 23 percent (2008) (Tallapragada and others 2009). SBEE has been able to collect 96\npercent of its billing, above the average for low-income peers (table 11).\n\n\n**Challenges**\n\n\nImportant differences in access to electricity remain between urban and rural populations. Whereas 53\npercent of urban dwellers have access to electricity\n\nMention: \"Enterprise Surveys 2009\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Value lost to outages is reported by firms in this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004867:32:1", "occurrence_id": "07b15a88cf7385ca9c87d231", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  country figures as of 2005 from the power AICD database, downloadable from\n[http://www.infrastructureafrica.org/aicd/tools/data. For Benin, national access data were taken from World Bank (2009a) and are based on](http://www.infrastructureafrica.org/aicd/tools/data)\nestimates in 2006. Electricity consumption was taken from the CIA Factbook and is for 2007. Power outages were taken from World Bank\n(2009a) and are for 2009. Value lost to outages was taken from Enterprise Surveys 2009. Firms that indicate that power is a constraint and\nfirms that own generators taken from Enterprise surveys 2004. Collection rate is taken from World Bank 2009a and is for 2009. System losses\nare for 2009 and taken from World Bank 2009a. Tariff information is for 2009 and is from World Bank 2009a.\n\n- Represents information only for SBEE and does not include information for CEB; ** As of 2004\n\n\nThe power distribution utility in Benin, SBEE, is relatively efficient. Its system losses are modest\nrelative to African peers, and its collection rates are relatively high. Total system losses are estimated at\n21 percent, with 17 percent from distribution. While these losses are high relative to the 10 percent\ninternational benchmark, system losses are lower than in comparable low-income countries and even with\nthe level of middle-income countries in Sub Saharan Africa (table 11). In other West African countries\nlosses are higher: Togo has been burdened with transmission losses of around 30 percent (2004), Côte\nd’Ivoire around 23 percent (2008) (Tallapragada and others 2009). SBEE has been able to collect 96\npercent of its billing, above the average for low-income peers (table 11).\n\n\n**Challenges**\n\n\nImportant differences in access to electricity remain between urban and rural populations. Whereas 53\npercent of urban dwellers have access to electricity\n\nMention: \"Enterprise surveys 2004\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Generator ownership and constraint reports come from this earlier round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004869:10:1", "occurrence_id": "1ebdf75732f6049869f85f61", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 246 positive cross-border trade ‡ows, 2,548 zeros and\n\n\n826 instances of missing values. We therefore need to account for potential sample selection issues in our\n\n\nmodel and econometric methodology.\n\n\nWe use 2005 GDP measured in constant (2000) US Dollars from the World Development Indicators.\n\n\nAs will become clear, each country’s GDP share is an important component of our analysis. We calculate\n\n\neach country’s share of world GDP by dividing its GDP by world GDP, where world GDP is the sum of\n\n\nthe 116 countries in our sample.\n\n\nThe measure of bilateral distance that we use captures the internal distance in a country, accounts\n\n\nfor the distance from a number of major cities and is constructed by CEPII. <sup>9</sup> [^9: The distance measure used is distw and is described at http://www.cepii.fr/anglaisgraph/bdd/distances.htm] Border data are also\n\n\nsourced from CEPII. Our control variables, including dummies for whether or not two countries share\n\n\na common language, <sup>10</sup> [^10: We construct a dummy that is equal to one if two countries share either a common o¢ cial or common ethnic language.] a common colonizer or were once the same country, as well as a dummy for a\n\n\nlandlocked country, are sourced from CEPII. Additional variables for identi…cation as …xed export costs\n\n\nor instruments are taken from the World Bank Doing Business database. These data include the number\n\n\nof procedures needed to start a business, the number of days it takes to start a business and the cost of\n\n\nregistering to start a new business in a country. We also considered data on the number of documents\n\n\nrequired to export or to import goods. Finally, we took the newly recoded data on religious similarity\n\n\nused by Helpman, Melitz & Rubinstein (2008) from Elhanan Helpman’s website.\n\n\n7 The 2009 version of this paper used the full LPI. That paper also restricted\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Output for the reference year in constant dollars comes from this source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004869:10:1", "occurrence_id": "d1ed3f15b021d2fb471b1fd4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 246 positive cross-border trade ‡ows, 2,548 zeros and\n\n\n826 instances of missing values. We therefore need to account for potential sample selection issues in our\n\n\nmodel and econometric methodology.\n\n\nWe use 2005 GDP measured in constant (2000) US Dollars from the World Development Indicators.\n\n\nAs will become clear, each country’s GDP share is an important component of our analysis. We calculate\n\n\neach country’s share of world GDP by dividing its GDP by world GDP, where world GDP is the sum of\n\n\nthe 116 countries in our sample.\n\n\nThe measure of bilateral distance that we use captures the internal distance in a country, accounts\n\n\nfor the distance from a number of major cities and is constructed by CEPII. <sup>9</sup> [^9: The distance measure used is distw and is described at http://www.cepii.fr/anglaisgraph/bdd/distances.htm] Border data are also\n\n\nsourced from CEPII. Our control variables, including dummies for whether or not two countries share\n\n\na common language, <sup>10</sup> [^10: We construct a dummy that is equal to one if two countries share either a common o¢ cial or common ethnic language.] a common colonizer or were once the same country, as well as a dummy for a\n\n\nlandlocked country, are sourced from CEPII. Additional variables for identi…cation as …xed export costs\n\n\nor instruments are taken from the World Bank Doing Business database. These data include the number\n\n\nof procedures needed to start a business, the number of days it takes to start a business and the cost of\n\n\nregistering to start a new business in a country. We also considered data on the number of documents\n\n\nrequired to export or to import goods. Finally, we took the newly recoded data on religious similarity\n\n\nused by Helpman, Melitz & Rubinstein (2008) from Elhanan Helpman’s website.\n\n\n7 The 2009 version of this paper used the full LPI. That paper also restricted\n\nMention: \"World Bank Doing Business database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Procedure counts and days serve as fixed cost proxies and instruments.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004876:11:0", "occurrence_id": "18023749cce00f0efe7ff28b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: NIGER’S INFRASTRUCTURE: A CONTINENTAL PERSPECTIVE\n\n\n**Figure 3. Niger’s population, poorest areas, and mineral resources are concentrated in the south**\n\n\n**a. Population** **b. Poverty**\n\n\n**c. Topography** **d. Natural resources**\n\n\n_Source_ : AICD Interactive Infrastructure Atlas for Niger (www.infrastructureafrica.org).\n\n\n6\n\nMention: \"AICD Interactive Infrastructure Atlas for Niger\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Population, poverty, terrain and resource panels are drawn from this atlas.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004878:3:0", "occurrence_id": "8f073816e08b31cdd3bc3856", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **1. Introduction**\n\n\nThis paper uses a unique and rich micro dataset on pension funds to shed new light on\n\n\nhow incentives might affect institutional investors in their portfolio allocation decisions. In\n\n\nparticular, we study two aspects discussed in the literature but still relatively unexplored: (i) how\n\n\ninstitutional investors trade in different types of assets and (ii) what incentives managers at\n\n\nvarious layers of the financial industry face when implementing their investment strategies.\n\n\nBecause of the interest of the literature in herding behavior, we focus on herding statistics. This\n\n\nkind of analysis helps understand more broadly the role of institutional investors on capital\n\n\nmarket activity and the services they effectively provide as financial intermediaries.\n\n\nFurthermore, it offers new evidence on the importance of several factors often discussed in the\n\n\nliterature on the behavior of institutional investors, including information, liquidity, incentives\n\n\nrelated to organizational aspects of the financial industry, and the regulatory framework in which\n\n\nmanagers operate.\n\n\nInstitutional investors are interesting to analyze not only because they have become very\n\n\nlarge, but also because detailed asset-level portfolios over time (unavailable at the household or\n\n\nretail-investor level) are sometimes accessible. In particular, institutional investors are\n\n\nincreasingly relevant for both asset management and the development of financial systems. In\n\n\nfact, institutional investors are likely to be among the most important conduits of private and\n\n\npublic savings, intermediating funds and supplying capital for firms and countries to grow. As\n\n\ninstitutional investors became prevalent, research flourished trying to understand how they\n\n\ninvest. Many papers in the literature focus on equity mutual funds, for which portfolio data are\n\n\nmostly publicly available, and study their investment patterns (Grinblatt et al., 1995; Wermers,\n\n\n1999; and Kacperczyk et al., 2005). Others analyze general data on institutional investors (Sias\n\n\n2\n\nMention: \"micro dataset on pension funds\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"A unique rich dataset on these institutional investors underlies the study.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004878:3:0", "occurrence_id": "51a14517b9426ab5277365b0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **1. Introduction**\n\n\nThis paper uses a unique and rich micro dataset on pension funds to shed new light on\n\n\nhow incentives might affect institutional investors in their portfolio allocation decisions. In\n\n\nparticular, we study two aspects discussed in the literature but still relatively unexplored: (i) how\n\n\ninstitutional investors trade in different types of assets and (ii) what incentives managers at\n\n\nvarious layers of the financial industry face when implementing their investment strategies.\n\n\nBecause of the interest of the literature in herding behavior, we focus on herding statistics. This\n\n\nkind of analysis helps understand more broadly the role of institutional investors on capital\n\n\nmarket activity and the services they effectively provide as financial intermediaries.\n\n\nFurthermore, it offers new evidence on the importance of several factors often discussed in the\n\n\nliterature on the behavior of institutional investors, including information, liquidity, incentives\n\n\nrelated to organizational aspects of the financial industry, and the regulatory framework in which\n\n\nmanagers operate.\n\n\nInstitutional investors are interesting to analyze not only because they have become very\n\n\nlarge, but also because detailed asset-level portfolios over time (unavailable at the household or\n\n\nretail-investor level) are sometimes accessible. In particular, institutional investors are\n\n\nincreasingly relevant for both asset management and the development of financial systems. In\n\n\nfact, institutional investors are likely to be among the most important conduits of private and\n\n\npublic savings, intermediating funds and supplying capital for firms and countries to grow. As\n\n\ninstitutional investors became prevalent, research flourished trying to understand how they\n\n\ninvest. Many papers in the literature focus on equity mutual funds, for which portfolio data are\n\n\nmostly publicly available, and study their investment patterns (Grinblatt et al., 1995; Wermers,\n\n\n1999; and Kacperczyk et al., 2005). Others analyze general data on institutional investors (Sias\n\n\n2\n\nMention: \"general data on institutional investors\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Broader investor data characterise the related literature.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004880:31:2", "occurrence_id": "530a014643eb04b9d910768e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  industry could translate\ninto $5.4 billion in additional exports—a 233 percent increase.\n\nEven rough employment estimates suggest the potential gains in manufacturing jobs. Africa‘s\npopulation (north and south of the Sahara) is 1 billion, slightly less than India‘s 1.15 billion. In\n2009 manufacturing value added was 16 percent of GDP in India, 13 percent in Sub-Saharan\nAfrican countries, and 16 percent in North African countries such as Egypt, Morocco, and\nTunisia. <sup>24</sup> India‘s employment in manufacturing was 8.7 million in 2009. So it is reasonable to\nassume that total manufacturing employment in Africa is at most 10 million. This suggests that\nrelocation of even a small share of China‘s 85 million labor-intensive manufacturing jobs would\n\n\n24 Data from the World Bank, World Development Indicators database.\n\nMention: \"World Development Indicators database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Employment in labour-intensive manufacturing is estimated using these series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004891:24:2", "occurrence_id": "c703f8f985d56f80c2c5bedd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: >0.01726<br>0.03574<br>0.21274|**Real**<br>**PPP**<br>_with migration_<br>0.01667<br>0.02288<br>0.17114<br>0.03513<br>0.05369<br>0.27763<br>0.01059<br>0.02088<br>0.14169<br>0.02682<br>0.04824<br>0.22984<br>0.00569<br>0.01261<br>0.11994<br>0.01726<br>0.03574<br>0.21274|\n\n\n\n        - Authors’ calculations, using data from the World Development Indicators. Index gen\n\nerated from first principal component of (1)–(5), variously excluding and including (4).\n\n\nReal GDP and PPP-adjusted indicate growth rates calculated from GDP data in real\n\n\n2000 U.S. dollars and constant 2005 international dollars.\n\n_†_ Migration flows are calculated for 1985, 1995, and 2005, respectively, the years where\n\n\ndata are available.\n\n\n**5.2** **Toward** **a** **multipolar** **world**\n\n\nWhat do the changing polarities mean for the distribution of economic influence in the global\n\n\neconomy as a whole? To the extent that growth polarity is an accurate measure of such influence,\n\n\nthe results in the previous section suggest that multipolarity increased steadily through the end\n\n\nof the Cold War, fell during the final decade of the 20th century, and finally rose again in the\n\n\nfirst decade of the 21st century; indeed, over the past decade, the world has attained some of\n\nthe most diverse distributions since 1968\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"A principal component index is generated from these variables.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004894:41:4", "occurrence_id": "a5a0f064a20daa055e557e12", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 2046.93<br>-1400.03<br>-1422.65<br>-3737.19<br>-3591.42<br>-3580.85<br>-2015.00<br>0.43<br>0.43<br>0.42<br>0.44<br>0.44<br>0.44<br>0.43|8765.83<br>13636.63<br>8767.11<br>8935.05<br>13818.89<br>8948.42<br>606.98<br>513.99<br>614.50<br>-4354.92<br>-6790.31<br>-4353.55<br>0.15<br>0.29<br>0.15|\n|Observations<br>RDHS data from<br>Sample|1504<br>1029<br>1030<br>2823<br>2713<br>2704<br>1483<br>2000<br>2000<br>2000<br>2000<br>2000<br>2000<br>2000<br>Middle age<br>women<br>whose first<br>child was<br>born before<br>genocide<br>Middle age<br>women<br>whose first<br>son was born<br>before<br>genocide<br>Middle age<br>women<br>whose first<br>daughter was<br>born before<br>genocide<br>Middle age<br>women<br>whose oldest<br>sibling was<br>born before<br>genocide<br>Middle\n\nMention: \"RDHS data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Estimates are reported with observation counts from this survey round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004895:25:0", "occurrence_id": "c408b66af521d03793dd3a7e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: the various groups studied have the lowest overall mortality rate in both high and low HIV\n\nprevalence countries. On the other hand, men with less than primary education have the largest\n\n\nincrease in mortality. In high HIV-prevalence countries these men have mortality rates that are\n\n\nsubstantially higher than men with more education, and than women.\n\n\nOne group for which the higher mortality among men does not hold is among young\n\n\nwomen (ages 15-24) especially those with less education. For this group, mortality rates are\n\n\nhigher than all other groups of the same age. This is consistent with the age profile of\n\n\nHIV/AIDS, but is also consistent with findings by others that younger more educated women\n\n\nappear to be engaging in more “protective” behaviors as knowledge about HIV and AIDS is\n\n\ndisseminated.\n\n\nLast, perhaps unsurprisingly, countries that have experienced conflict or episodes of\n\n\ninternationally recognized genocide, display heterogeneous patterns in the time trends and\n\n\nsocioeconomic patterns in mortality. The experience from several countries suggests that male\n\n\nmortality is typically more responsive to these events—and in several cases it is urban or more\n\n\neducated males for whom mortality increases most. At the same time, it is important to\n\n\nrecognize that increased mortality during these episodes of conflict is not restricted to men—in\n\n\nall cases women’s mortality is affected as well.\n\n\n**References**\n\nBicego, George. 1997. “Estimating Adult Mortality Rates in the Context of the AIDS Epidemic\n\nin Sub-Saharan Africa: Analysis of DHS Sibling Histories.” _Health Transition Review_\n7(Suppl 2): 7-22.\n\nCase, Anne and Christina Paxson. 2010. “The Changing Role of Education in the AIDS Crisis”.\n\nCenter for Health and Wellbeing. Princeton University. Mimeo. August 2010.\n\nConley, Dalton, and Rebecca Glauber. 2005. “Sibling similarity and differences in\n\nsocioeconomic status: life course and family resource effects.” NBER Working Paper\n11320.\n\nDahan, Momi\n\nMention: \"DHS Sibling Histories\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"A bibliography title describing adult mortality analysis from these histories.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004898:24:0", "occurrence_id": "b951dede3dabd9b423e1f13c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: will be near zero in those countries. In high income countries more workers, particularly among\n\n\nthe more able, can afford to trade off income against moral or other psychic considerations when\n\n\nchoosing a job. The morality-ability correlation will be positive in those countries.\n\n\nWe address these endogeneity concerns in several ways. First, as explained in more detail be\n\nlow, we exploit time series variation in the data, by modifying the sample or alternatively by\n\n\nmodifying the specification. Subsequently, we instrument for the morality variables using 2SLS\n\n\nand limited-information maximum likelihood (LIML) in Table 5. Coefficients for the morality\n\nability correlation variable remain positive and highly significant in these tests, while results for\n\n\nthe morality proxy are mixed.\n\n\nIncome is measured for 2008, while the morality variables are measured from surveys con\n\nducted between 1995 and 2008 but with variation among countries. The potential for estimates to\n\n\nreflect reverse causation from income to morality is reduced somewhat by measuring morality\n\n\nfurther back in time, e.g. for 1995 instead of for 2008. Accordingly, in equation 4-1 of Table 4\n\n\nwe drop the 19 observations for which morality is measured using only the 5 <sup>th</sup> WVS survey wave\n\n\n(2005-2008). This change reduces the mean survey year represented in the data by two years,\n\n\nfrom 2002 to 2000. The model specification is from equation 2-4, based on RST but with the\n\n\nmorality variables added. The coefficient for the morality proxy in equation 4-1 is insignificant\n\n\nand only one-third as large as in equation 2-4, consistent with the view that the latter estimate\n\n\nmostly reflects reverse causation from income to morality. The coefficient on the morality-ability\n\n\ncorrelation remains positive and highly significant, however, in equation 4-1. Moreover, its coef\n\nficient is one-third larger in equation 4-1 than in equation 2-4.\n\n\nIn equation 4-2 we drop\n\nMention: \"WVS survey wave\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Observations relying only on this wave are dropped in the robustness check.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004903:12:0", "occurrence_id": "d7e72ae3db4248306568e83f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: to determine this geographic area. Using GIS, we overlaid flood maps with population\n\n\nmaps from the Gridded Population of the World v3 (CIESIN-CIAT 2005) to obtain\n\n\nestimates of the population exposed to each event. Population grid maps are available in 5\n\nyear intervals since 1990. We calibrated an exponential curve for the remaining years to\n\n\ncomplete the panel. We used the resulting variable in the flood fatality and magnitude\n\n\nmodels (equations (2)-(4)) instead of national population, which was used in the flood\n\n\nfrequency model (equation (1)) as observations in the latter model were defined as country\n\nyears.\n\n\nThe resulting estimates of population exposure to a flood are more accurate than\n\n\nstatistics based on country-level population statistics, which previous studies on natural\n\n\ndisasters have used. In our sample, mean population density in areas affected by floods is\n\n\n394.46 persons/km <sup>2</sup>, while mean population density at the country level is 154.53\n\n\npersons/km <sup>2</sup> . This difference reflects the fact that people have tended to concentrate in\n\n\nflood plains for millenia. Increased populations in flood plains is one explanation for the\n\n\nobserved growth in number of floods (Figure 1; Freeman et al. 2003; IPCC 2007a, Chapter\n\n\n3).\n\n\n_3.3 Vulnerability: Socioeconomic and governance indicators_\n\n\nThe indicator of income is GDP per capita converted to constant 2005 international dollars\n\n\nusing purchasing power parity rates. It comes from the World Development Indicators\n\n\n(WDI 2010). Although GNI per capita more accurately measures the income of a country‘s\n\n\npopulation, it was available for slightly few countries and years than GDP per capita. The\n\n\nchoice between GDP per capita and GNI per capita is inconsequential in our models: we\n\n\nestimated all the models using both measures, and results were virtually indistinguishable.\n\n\n11\n\nMention: \"Gridded Population of the World v3\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Overlaying flood maps on these grids yields exposed population counts.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004903:12:0", "occurrence_id": "95912dfa4310fca3e35e23f0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: to determine this geographic area. Using GIS, we overlaid flood maps with population\n\n\nmaps from the Gridded Population of the World v3 (CIESIN-CIAT 2005) to obtain\n\n\nestimates of the population exposed to each event. Population grid maps are available in 5\n\nyear intervals since 1990. We calibrated an exponential curve for the remaining years to\n\n\ncomplete the panel. We used the resulting variable in the flood fatality and magnitude\n\n\nmodels (equations (2)-(4)) instead of national population, which was used in the flood\n\n\nfrequency model (equation (1)) as observations in the latter model were defined as country\n\nyears.\n\n\nThe resulting estimates of population exposure to a flood are more accurate than\n\n\nstatistics based on country-level population statistics, which previous studies on natural\n\n\ndisasters have used. In our sample, mean population density in areas affected by floods is\n\n\n394.46 persons/km <sup>2</sup>, while mean population density at the country level is 154.53\n\n\npersons/km <sup>2</sup> . This difference reflects the fact that people have tended to concentrate in\n\n\nflood plains for millenia. Increased populations in flood plains is one explanation for the\n\n\nobserved growth in number of floods (Figure 1; Freeman et al. 2003; IPCC 2007a, Chapter\n\n\n3).\n\n\n_3.3 Vulnerability: Socioeconomic and governance indicators_\n\n\nThe indicator of income is GDP per capita converted to constant 2005 international dollars\n\n\nusing purchasing power parity rates. It comes from the World Development Indicators\n\n\n(WDI 2010). Although GNI per capita more accurately measures the income of a country‘s\n\n\npopulation, it was available for slightly few countries and years than GDP per capita. The\n\n\nchoice between GDP per capita and GNI per capita is inconsequential in our models: we\n\n\nestimated all the models using both measures, and results were virtually indistinguishable.\n\n\n11\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Gross national income per head in constant dollars comes from this edition.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004904:9:0", "occurrence_id": "554fa1450ab620e3a3b08a8c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Our dataset differs from ones used in the earlier survey based work by Ng et al. (1999),\n\n\nGiannetti et al. (2011) and others in several important ways. First, we use data from actual trade\n\n\ncredit contracts rather than data based on survey responses, thereby mitigating the usual\n\n\nmisreporting concerns associated with survey based data. Importantly, our dataset consists of\n\n\ntrade credit contracts signed, while papers using the National Survey of Small Business Finance\n\n\n(NSSBF) database use trade credit contracts offered. Our dataset also covers a broader set of\n\n\nindustries that includes technology firms, allowing for an analysis of trade credit contract terms\n\n\nacross a broader range of industries, and includes trade credit terms not only for U.S. firms but\n\n\nalso for international firms, allowing for a comparison of trade credit terms across different\n\n\njurisdictions. Moreover, our dataset covers suppliers of all sizes, and not just small firms as in\n\n\nthe SSBF. This is an important difference because the credit terms offered by large firms could\n\n\nbe very different than those by small firms.\n\n\nWe begin by summarizing the main characteristics of buyers, suppliers, and contracts.\n\n\nPanels A and B of Table 1 show summary statistics of buyer and supplier characteristics. First,\n\n\nthe buyers in our sample are very large – 33 out of 56 (or 59%) of buyers have over US$ 10\n\n\nbillion in sales and only one buyer has less than US$ 2 billion in sales.\n\n\nThe buyers are also creditworthy as measured by whether or not they are investment\n\n\ngrade – about 75% of buyers in the dataset are investment grade. That we have mostly large,\n\n\ninvestment-grade buyers will make it harder to find evidence in this dataset of a financing motive\n\n\nfor trade credit. We should, therefore, treat our results with appropriate caution.\n\n\nBuyers are active in a range of industries, with the majority in retail industries. The\n\n\nsectoral distribution of buyers is: 16% in auto manufacturing, 13% in diversified retail, 29\n\nMention: \"National Survey of Small Business Finance\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"It records contracts offered rather than signed, unlike this study's data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004904:9:0", "occurrence_id": "3b842a3e3100ae4337e06c0d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Our dataset differs from ones used in the earlier survey based work by Ng et al. (1999),\n\n\nGiannetti et al. (2011) and others in several important ways. First, we use data from actual trade\n\n\ncredit contracts rather than data based on survey responses, thereby mitigating the usual\n\n\nmisreporting concerns associated with survey based data. Importantly, our dataset consists of\n\n\ntrade credit contracts signed, while papers using the National Survey of Small Business Finance\n\n\n(NSSBF) database use trade credit contracts offered. Our dataset also covers a broader set of\n\n\nindustries that includes technology firms, allowing for an analysis of trade credit contract terms\n\n\nacross a broader range of industries, and includes trade credit terms not only for U.S. firms but\n\n\nalso for international firms, allowing for a comparison of trade credit terms across different\n\n\njurisdictions. Moreover, our dataset covers suppliers of all sizes, and not just small firms as in\n\n\nthe SSBF. This is an important difference because the credit terms offered by large firms could\n\n\nbe very different than those by small firms.\n\n\nWe begin by summarizing the main characteristics of buyers, suppliers, and contracts.\n\n\nPanels A and B of Table 1 show summary statistics of buyer and supplier characteristics. First,\n\n\nthe buyers in our sample are very large – 33 out of 56 (or 59%) of buyers have over US$ 10\n\n\nbillion in sales and only one buyer has less than US$ 2 billion in sales.\n\n\nThe buyers are also creditworthy as measured by whether or not they are investment\n\n\ngrade – about 75% of buyers in the dataset are investment grade. That we have mostly large,\n\n\ninvestment-grade buyers will make it harder to find evidence in this dataset of a financing motive\n\n\nfor trade credit. We should, therefore, treat our results with appropriate caution.\n\n\nBuyers are active in a range of industries, with the majority in retail industries. The\n\n\nsectoral distribution of buyers is: 16% in auto manufacturing, 13% in diversified retail, 29\n\nMention: \"SSBF\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its coverage of small firms only limits comparison with all-size suppliers.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004905:20:2", "occurrence_id": "99a473b15452d138d817b481", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  0.50 0.37\nMedium soil quality 0.36 0.36 0.37 0.37 0.34 0.41\nFlat land 0.71 0.70 0.72 0.73 0.69 0.73\nGently sloped land 0.24 0.24 0.23 0.23 0.23 0.25\nIrrigated land 0.05 0.05 0.04 0.04 0.03 0.01\n**Crop choice:**\nTeff 0.32 0.29 0.36 0.35 0.48 0.29\nWheat/barely 0.21 0.20 0.22 0.21 0.24 0.26\nSorghum/maize 0.30 0.34 0.23 0.23 0.20 0.31\nPulses 0.13 0.12 0.14 0.16 0.04 0.10\n<u>Oilseeds</u> <u>0.05</u> <u>0.05</u> <u>0.04</u> <u>0.04</u> <u>0.04</u> <u>0.03</u>\n<u>Number of observations</u> <u>5443</u> <u>3360</u> <u>2083</u> <u>1728</u> <u>238</u> <u>102</u>\n_Source:_ Own computation from AU/WB/UG land tenure survey.\n_Note:_ Only plots with cereals, pulses and oilseeds included.\n\n\n\n19\n\nMention: \"AU/WB/UG land tenure survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Plot-level cereal, pulse and oilseed results are computed from this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004910:30:0", "occurrence_id": "34327645035b0c5951103f75", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: On the basis of 50 subsamples of 1012 observations, we get the results for _g_, numbers of\nmodels visited and model size presented in Figure 10. Results are now a bit different from\nthose in the growth applications, as a result of the much larger value of _n_ . In particular,\nthe Maruyama-George, Forte _et_ _al._ and Zellner-Siow priors as well as the case with fixed\n_g_ = _n_ lead to much higher posterior medians of _g_, a smaller number of models visited in\nthe chain and smaller mean model size than in the previous applications. Of course, all of\nthese priors depend on _n_, but it is interesting that the F-Z prior with _a_ = 2 + 2 _/n_ and the\nhyper- _g/n_ and horseshoe _/n_ priors are less affected. Of course, the benchmark priors and\nthe Bottolo-Richardson prior depend on _n_ through max _{n, k_ <sup>2</sup> _}_, which is here of the same\norder of magnitude as for the FLS data.\n\n\n\nlog ( g ) : Random θ, m = 7\n\n\n\nNumber of Models: Random θ, m = 7\n\n\n\nModel Size: Random θ, m = 7\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n1 2 3 4 5 6 7 8 9 11 13 15\n\n\n\n1 2 3 4 5 6 7 8 9 11 13 15\n\n\n\n1 2 3 4 5 6 7 8 9 11 13 15\n\n\n\n**Fig.** **10.** Tobias-Li data: log posterior medians of _g_, number of models visited\n\nand posterior mean model size.\n\n\nPrediction based on prediction subsamples of 178 observations each is summarized in the\nlower panel of Figure 7. Even though there is some variability in the posterior inclusion\nprobabilities of the regressor\n\nMention: \"FLS data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"The estimator's order of magnitude matches that found on these data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004911:21:0", "occurrence_id": "df5cc7fbfafc7d2e501ab422", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 20\n\n\nget more or less allocations), and budget implementation (whether the allocations\ntranslated into better services or outcomes) appears to be almost entirely missing.\n\nThere _is_ evidence on a specific type of budget information which was\npioneered in the education sector in Uganda. Purposefully designed survey\ninstruments—named the Public Expenditure Tracking Surveys (PETS)—were used\nto measure discrepancies between budget allocations to schools in Uganda, and\nwhat actually reached the schools (Reinikka and Svensson, 2004). One study argues\nthat the publication in mass media of estimated “leakage” in the flow of funds led to\ndramatic improvements, with most of the intended funds reaching the schools\nsubsequent to the information campaign (Reinikka and Svensson, 2005). In this\nargument, accountability for the flow of funds improved because of addressing\ninformation gaps among parents in schools, and thus enabling them to exercise their\nclient power _vis a vis_ district authorities responsible for transferring funds to\nschools.\n\nFollowing the celebrated success of the Uganda case there has been a\nproliferation of PETS activities, and external investments in strengthening the\ncapacity of civil society organizations to undertake such surveys to uncover and\npublicize “leakage” and thereby stem it. More qualitative evaluations of these efforts\nsuggest that they can be frequently successful in getting the funds to reach the\nintended destinations, although to a lesser degree than the estimates in Uganda\n(McGee and Gaventa, 2010).\n\nHowever, alternate arguments about the drivers of success in the Uganda\ncase indicate that the client-power channel was only part of the story. Hubbard\n(2007) examines larger changes in school funding policies in the Ugandan Ministry\nof Education, and suggests that the information that activated client-power may\nhave been only one element in increasing the amount of resources available in\nschools; that other, perhaps larger, elements consisted of strengthening of the\ncompact between the Ministry and the districts following President Museveni’s\n1996 election campaign in which free and universal primary education\n\nMention: \"Public Expenditure Tracking Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Purpose-built instruments measured gaps between allocations and receipts.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004913:11:0", "occurrence_id": "04d7ee93044da652e741454f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: decisions and careers in order to prevent their access to the sources of rents (e.g. natural\nresources) when political checks and balances are low and electoral competition is absent.\n\nAs in Keefer and Knack (2007) we adopt the measure checks and balances cum elections\nfrom the Database of Political Institutions (DPI). <sup>16</sup> [^16: As an alternative measure we adopt a laxer version of checks and balances cum elections.] This measure is a function of the number of\nparties in the government coalition (for parliamentary systems), whether the president's party has\na majority in the legislature (presidential systems), whether elections are governed by closed-list\nor open-list rules (with the former granting more authority to the heads of parties), and of the\nDPI's legislative index of electoral competitiveness.\n\nThe use of formal checks and balances presents, however, some limitations. Since it is\nonly one of many possible political determinants of low institutional quality, the results are\nunlikely to be as strong as for quality of governance. Moreover, this measure draws on a fixed\nview of organization of government and some checks and balances may be hard to observe.\n\nWe would expect both quality of governance and checks and balances to show little\nvariation over time and great variation across countries. However, the relatively wide time span\nallows the within standard deviation to be almost half of the across standard deviation (2.6\nagainst 5.6) for quality of governance <sup>17</sup> [^17: Within/across relative variation is even higher for quality of governance wide (4 against 7.5).] and nearly the same for checks and balances (1.3 against\n1.0), justifying the panel analysis.\nA set of other factors is likely to impact the quantity and the volatility of public\ninvestment. As observed in Knack and Keefer (2007), left-leaning governments might prove\nmore prone to intervene in the economy and\n\nMention: \"Database of Political Institutions\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its checks and balances with elections measure is adopted here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004914:77:1", "occurrence_id": "d64dcc75533d3be63dc995d1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Introduction.‖ _Journal of Economic Surveys_ 10:\n\n367–421.\n\nWorld Bank. 2005a. Project Appraisal Document: Rural Access Improvement and\n\nDecentralization Project. Report No. 31624-NP. The World Bank.\n\nWorld Bank. 2005b. _Infrastructure and the World Bank: A Progress Report_ . World Bank:\n\nWashington, DC.\n\nWorld Bank. 2007. Nepal Public Finance Management Review: Managing Public Finances\n\nfor a New Nepal (No. 43384-NP).\n\nWorld Bank. 2008a. _Procurement under World Bank-Financed Projects: FY07 Annual_\n\n_Report_ . Operations Policy and Country Services. Washington, DC: World Bank.\n\nWorld Bank. 2009. Project Paper on Nepal Rural Access Improvement and Decentralization\n\nProject, 2009 (No. 50766-NP).\n\nWorld Bank. 2009. Enterprise Survey: 2009 Nepal. Available at\n\n<u>[http://www.enterprisesurveys.org/ExploreEconomies/?economyid=136&year=2009](http://www.enterprisesurveys.org/ExploreEconomies/?economyid=136&year=2009)</u>\n\nMention: \"Enterprise Survey: 2009 Nepal\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"A reference-list entry for this country survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004916:53:1", "occurrence_id": "713a784acba0bbc572eadced", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: u> <u>2,198</u> <u>3,621</u>\n\n<u>Panel B: Women</u>\n**Industry of main employment**\n<u>RCA=0</u> <u>RCA=1</u> <u>Total</u>\nAgriculture, Hunting and Forestry 48.2 56.28 53.04\nEducation 16.92 12.64 14.35\nSale, maintenance and repair of motor vehicles and 8.92 8.53 8.69\nmotorcycles, retail sales of automotive fuel\n\n\n\n8.85\n\n\n\n3.37\n\n\n\nSale, maintenance and repair of motor vehicles and 8.92 8.53 8.69\nmotorcycles, retail sales of automotive fuel\n\nHealth and social work 10.08 6.74 8.08\nPublic administration and defense 2.67 3.41 3.11\nManufacture of textiles 2.55 1.32 1.81\nHotels and restaurants 1.85 1.78 1.81\nManufacture of wearing apparel; dressing and dyeing of fur 0.81 1.4 1.16\n<u>Manufacture of food products and beverages</u> <u>1.27</u> <u>0.93</u> <u>1.07</u>\n<u>Sub-total</u> <u>93.27</u> <u>93.03</u> <u>93.12</u>\n<u>N obs (total)</u> <u>863</u> <u>1,290</u> <u>2,153</u>\nSource: Author’s calculations using TLSS (2007). The industry is included in the table if it had more than 1% of\ntotal employment.\n\n\n\n8.92\n\n\n\n8.53\n\n\n\n52\n\nMention: \"TLSS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Employment shares by industry above one percent are computed from this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004917:15:5", "occurrence_id": "3f0999974ef975e042f0010b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  - [0.272] 0.161\n\n_NBRt_ . <sup>_T_</sup> _t_ - [0.282] -0.055\n\n\n\n\n- -0.030\n\n[0.509]\n\n- -0.183\n\n[0.649]\n\n\n\n-3.426***\n\n\n\n-3.524***\n\n\n\n-3.154***\n\n\n\n_vit_\n\n\n\nconstant\n\n\n\n\n[0.125] [0.158] [0.252] [0.092] [0.200] [0.169]\n\nSigma-squared 0.180 0.172 0.099 0.190 0.207 1.28E-01\n\n\n\n\n[0.125]\n\n0.180\n\n[0.011]\n\n\n\n\n[0.158]\n\n0.172\n\n[0.013]\n\n\n\n\n[0.252]\n\n0.099\n\n[0.012]\n\n\n\n\n[0.092]\n\n0.190\n\n[0.009]\n\n\n\n0.207\n\n[0.0207]\n\n\n\n\n[0.200]\n\n\n\nSigma-squared 0.180 0.172 0.099 0.190 0.207 1.28E-01\n\n[0.011] [0.013] [0.012] [0.009] [0.0207] [0.011]\n\nWald chi2 15408.11 4980.84 21248.55 19794.51 1728.23 7729.43\nProb > chi2 0.000 0.000 0.000 0.000 0.000 0.000\nLog likelihood 57.275 27.675 71.755 56.894 15.317 59.712\nNumber of obs 244 144 100 244 144 100\n1Description in natural logs.\nSource: Authors’ calculation from the World Bank’s World Development Indicators (WDI).\n\n\n\n14\n\nMention: \"World Development Indicators (WDI)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The logged variables in the table are calculated from this compilation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004919:35:0", "occurrence_id": "191fd319455f94cb76dd31c1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 3: Conflict Intensity across Districts of Nepal**\n\n\n**Author’s calculations using electronic casualties data files provided by the Informal Sector Service (INSEC). District terciles**\n\n\n**are defined by the distribution of total district casualties per 1000 inhabitants. District population figures are based on the 1996**\n\n\n**population census** .\n\n\n**Source: Deaths and abductions are taken from electronic data files provided by the Informal Sector Service (INSEC). District**\n\n**population figures are based on the 1996 population census.**\n\n\n34\n\nMention: \"electronic casualties data files\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"District conflict intensity is calculated from these casualty records.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004919:35:0", "occurrence_id": "8c8947964a12586c183d0a22", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 3: Conflict Intensity across Districts of Nepal**\n\n\n**Author’s calculations using electronic casualties data files provided by the Informal Sector Service (INSEC). District terciles**\n\n\n**are defined by the distribution of total district casualties per 1000 inhabitants. District population figures are based on the 1996**\n\n\n**population census** .\n\n\n**Source: Deaths and abductions are taken from electronic data files provided by the Informal Sector Service (INSEC). District**\n\n**population figures are based on the 1996 population census.**\n\n\n34\n\nMention: \"1996 population census\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"District population figures normalise the casualty counts.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004921:16:0", "occurrence_id": "e8fbc6518199023c74e8eed1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: (0.032) (0.010) (0.032)\n\nAged 11-14 0.829*** 0.856*** -0.028\n\n(0.031) (0.011) (0.030)\n\nAged 15-17 0.569*** 0.743*** -0.174***\n\n<u>(0.039)</u> <u>(0.016)</u> <u>(0.037)</u>\n\n\nNote: Y1 is the observed schooling rate of children whose parents have\ndisabilities. Y0 is the counterfactual schooling rate, that is their enrollment if their\nparent had not been disabled.  ATT=Y1–Y0 is the effect of parental disability on\nchildren’s schooling rate.\n\n           - significant at 10%; ** significant at 5%; *** significant at 1%\n\nFigures in parentheses are standard errors. Standard errors are corrected for\nsampling weights and estimated using bootstrap (non-parametric) with 500\nreplications.\n\n<u>Source: Estimation from the 2006 VHLSS.</u>\n\n\nTable A.5 in the Appendix presents the ordered probit regressions of academic\n\n\nrecords. We tried both ordered logit with and without instruments for parental disability.\n\n\nEstimates of the parental variables are very small and not statistically significant in all the\n\n\nregressions. This means that once attending school, children with disabled parents do not\n\n\nhave worse educational performance than children with non-disabled parents given that\n\n\nthe estimation procedure controls for other observed characteristics.\n\n\nFinally, the regressions of school enrollment and academic records show some\n\n\ninteresting findings on other factors associated with education. Girls have higher school\n\n\nenrollment as well as better educational performance than boys even after other observed\n\n\nvariables are controlled. Gender equality in children‟s education is very important, since\n\n\nit can contribute to gender equality in wage and other social economic activities when\n\nMention: \"the 2006 VHLSS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Ordered probit estimates of academic records with bootstrap replications use this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004940:8:0", "occurrence_id": "2bf780fa377204aa3b46b0e9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 2: Map of Budget Execution Rates**\n\n**of Capital Expenditures, 2009**\n\n\nSource: Authors’ calculations based on MEF data.\n\n\nWhile capacity and political economy considerations have been widely studied, the\n\n\nextent to which the budget size and allocation process, as well as needs, affect spending has\n\n\nbeen less investigated, in part because these features are more specific to the design of the\n\n\nPeruvian decentralization reform. Although they do relate to some extent to capacity, they\n\n\nhave relevance on their own: in fact, our empirical analysis suggests that budget-related issues,\n\n\nin particular the adequacy of the transferred budget with respect to the local capacity, are\n\n\namong the most important determinants of spending ability. This is of particular relevance for\n\n\nPeru, since capacity building and planning are long term investments, but the recent boom in\n\n\ncommodity prices and generous revenue sharing agreements have spurred transfers to some\n\n\nmunicipalities that may simply not be able to manage the sudden increases in revenues.\n\n\nBecause of some peculiar features of the Peruvian decentralization reform,\n\n\n7\n\nMention: \"MEF data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Capital expenditure execution rates are computed from these finance ministry records.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004946:15:1", "occurrence_id": "7817e75ca7f9d80894379218", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  In 2008, monetary policy\nwas tightening, the banks' lending interest rates remained almost unchanged, and the deposits\ninterest rates were increasing in order to attract additional sources after low deposit growth at the\ninitial stage of the world crisis. In 2009, the tightening of the monetary policy was only partially\nfollowed by the banks' interest rates. Like in 2008, the stronger reaction was again shown by the\nincrease of deposit interest rates (by 1.2 percent) relative to the increase of the average lending\ninterest rate (by 0.4 percent). In both years, the banks were reducing the interest rate spread,\n\n\n11 Different analyses within the NBRM for different analyzed periods have shown that only 2 to 10 percent of the\nmain monetary policy interest rate change was transmitted over average lending interest rates on the banks' denar\ncredits, in the same month or with a one month time lag. The accommodation of the short-run dynamics of the\nbanks' lending interest rates to their long-term equilibrium usually is slow.\n\n\n13\n\nMention: \"the NBRM\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Analyses within this bank estimate limited policy pass-through.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004950:31:1", "occurrence_id": "c4b8e5cd717feecb5fc08737", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  March 2006. In\n\n\norder to minimize exposure to the violence, and due to the security situation, the TLSS 2007 was\n\n\ninterrupted and resumed in January 2007. All households interviewed previously were re\n\ninterviewed. It is, however, still possible that some biases may remain. In particular, it is possible\n\n\nthat some of the results we obtain are not due to exposure to the 1999 violence but to exposure to\n\n\nthe civil upheaval in 2006.\n\n\nIn order to control for this potential exposure to the violence in 2006, we explore a variable\n\n\nin the 2007 dataset that captures whether an individual has been absent from home in the past 12\n\nmonths for security reasons. <sup>27</sup> Our calculations show that individuals who have been absent from\n\n\nhome for security reasons in 2006 are all residing in Dili. Therefore, we believe that this dummy\n\n\nreliably captures quite well the level of exposure to the 2006 violence.\n\n\n27 Those that declare being absent from home for security reason represents the 2.7 % of our sample.\n\n30\n\nMention: \"TLSS 2007\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Fieldwork was interrupted by the security situation and resumed the following January.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004918:9:0", "occurrence_id": "675d85907cd4469ebea7a34c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: formality is associated with a higher wage premium and higher returns to education for formal labor in\n2006.\n\n\nTable 4 – Percentage of Formal vs. Informal Labor\n\n|Col1|1998|2006|\n|---|---|---|\n|Informal|88.66|64.12|\n|Formal|11.34|35.88|\n\n\n\n_Source_ : Authors’ calculations using ELMS 1998 and ELMPS 2006.\n\n\nTable 5 - Formality and Firm Size - 2006\n\n\n\n|Firm Size|informal Labor|Formal Labor|\n|---|---|---|\n|1 to 4|90.8%|9.2%|\n|5 to 49|85.6%|14.4%|\n|50+|35.8%|64.2%|\n|dont-know/miss|15.8%|84.2%|\n\n\n_Source_ : Authors’ calculations using ELMPS 2006.\n\n\n\n\n|Firm Size|informal|Formal|\n|---|---|---|\n|1to4|69.9%|12.7%|\n|5to49|19.9%|6.0%|\n|50+|2.5%|8.1%|\n|dont-know/miss|7.7%|73.2%|\n\n\n\nMost of the informal labor (79.5 percent in 1998 and 70 percent in 2006) works in micro enterprises (1\nto 4 employees), while most of the formal labor works in the government sector (Table 5). The majority\nof the informal workers are clustered in the private sector. Therefore, most of the informal labor is\nclustered in private micro enterprise, while most of the formal labor is in the government sector. The\nlargest share of informal labor in 1998 and 2006 are illiterate, while the largest share of\n\nMention: \"ELMS 1998\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The earlier round giving the informal employment share.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004918:9:0", "occurrence_id": "8c5ea7fe84f11690f2a0ba9b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: formality is associated with a higher wage premium and higher returns to education for formal labor in\n2006.\n\n\nTable 4 – Percentage of Formal vs. Informal Labor\n\n|Col1|1998|2006|\n|---|---|---|\n|Informal|88.66|64.12|\n|Formal|11.34|35.88|\n\n\n\n_Source_ : Authors’ calculations using ELMS 1998 and ELMPS 2006.\n\n\nTable 5 - Formality and Firm Size - 2006\n\n\n\n|Firm Size|informal Labor|Formal Labor|\n|---|---|---|\n|1 to 4|90.8%|9.2%|\n|5 to 49|85.6%|14.4%|\n|50+|35.8%|64.2%|\n|dont-know/miss|15.8%|84.2%|\n\n\n_Source_ : Authors’ calculations using ELMPS 2006.\n\n\n\n\n|Firm Size|informal|Formal|\n|---|---|---|\n|1to4|69.9%|12.7%|\n|5to49|19.9%|6.0%|\n|50+|2.5%|8.1%|\n|dont-know/miss|7.7%|73.2%|\n\n\n\nMost of the informal labor (79.5 percent in 1998 and 70 percent in 2006) works in micro enterprises (1\nto 4 employees), while most of the formal labor works in the government sector (Table 5). The majority\nof the informal workers are clustered in the private sector. Therefore, most of the informal labor is\nclustered in private micro enterprise, while most of the formal labor is in the government sector. The\nlargest share of informal labor in 1998 and 2006 are illiterate, while the largest share of\n\nMention: \"ELMPS 2006\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The later round showing formality and firm size.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004935:4:1", "occurrence_id": "1127d9481f7dbeeaeac4db41", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  fertilizer use on rice yields, showing that the\nrelationship is positive up to a point, beyond which the impact of additional use of fertilizer on yields is\nnegative. We also show the relative inefficiency of the subsidy in increasing rice production, since\nfertilizer subsidies are significantly higher than the value of the increase in production achieved through\nthe fertilizer subsidy. In the fifth section we discuss the rationale for subsidizing agricultural inputs and\nreview some of the international experience to date. Finally, section six concludes and provides some\npolicy recommendations.\n\n\nFor this research we use two different surveys by the National Statistics Office (BPS), the Agricultural\nCensus 2003 and the Rice Household Survey 2008. The first comprises a random sample of over 46,000\nrice farmers in 29 provinces, while the latter uses a smaller sample of over 11,000 rice farmers from the\n15 largest rice-producing provinces in Indonesia. The results of the benefit incidence analysis is very\nsimilar using both surveys; there is little targeting in the subsidy, with the largest farmers capturing a\ndisproportionate share of the subsidy.\n\n\n**3**\n\nMention: \"Agricultural\nCensus 2003\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"A random sample of over forty-six thousand rice farmers is enumerated here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004935:4:1", "occurrence_id": "c5b43472ff055018e954e5b3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  fertilizer use on rice yields, showing that the\nrelationship is positive up to a point, beyond which the impact of additional use of fertilizer on yields is\nnegative. We also show the relative inefficiency of the subsidy in increasing rice production, since\nfertilizer subsidies are significantly higher than the value of the increase in production achieved through\nthe fertilizer subsidy. In the fifth section we discuss the rationale for subsidizing agricultural inputs and\nreview some of the international experience to date. Finally, section six concludes and provides some\npolicy recommendations.\n\n\nFor this research we use two different surveys by the National Statistics Office (BPS), the Agricultural\nCensus 2003 and the Rice Household Survey 2008. The first comprises a random sample of over 46,000\nrice farmers in 29 provinces, while the latter uses a smaller sample of over 11,000 rice farmers from the\n15 largest rice-producing provinces in Indonesia. The results of the benefit incidence analysis is very\nsimilar using both surveys; there is little targeting in the subsidy, with the largest farmers capturing a\ndisproportionate share of the subsidy.\n\n\n**3**\n\nMention: \"Rice Household Survey 2008\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The companion household survey of rice producers.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004953:23:0", "occurrence_id": "19b0d4471475be44fa5e6b07", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: controlling for this trend leaves no variation that can be exploited to identify these coefficients. <sup>8</sup>\n\n\nOur main industry-level female intensity variable ( _FL_ ) is computed as the average share of female\n\n\nworkers in each sector in a sample of some 20 developed and developing countries. We favor this\n\n\nmeasure because it is most representative of the female intensity of sectors in an average country.\n\n\nHowever, a drawback of our main measure is that it does not include non-manufacturing sectors. To\n\n\ncheck robustness of our results, we implement an alternative strategy, of constructing a measure of\n\n\n_FL_ based on data for a single country - the U.S.. We thus use the Labor Force Statistics database\n\n\nof the U.S. Bureau of Labor Statistics (BLS). Using data from the Current Population Survey,\n\n\nthe BLS publishes “Women in the Labor Force: A Databook” on an annual basis since 2005. It\n\n\ncontains information on total employment and the female share of employment in each industry\n\n\ncovered by the Census. The data are available at the 4-digit U.S. Census 2007 classification (262\n\n\ndistinct sectors, including both manufacturing and non-manufacturing). In order to construct the\n\n\nshare of female workers in total employment in sector _i_, _FLi_, we take the mean of this value across\n\n\nthe years for which the data on the female share of employment are available (2004-2009). The\n\n\nresulting sample includes 78 manufacturing and 15 non-manufacturing sectors. Appendix Table A1\n\n\nreports the values of _FLi_ for the top and bottom 5 sectors according to U.S. data. These sectors\n\n\nare similar to what we find in our baseline measure: the least female-intensive sectors tend to be\n\n\nin heavy machinery, while the most female-intensive sectors in textiles and apparel. Appendix\n\n\nTable A1 presents the basic summary statistics of the U.S.-based\n\nMention: \"Labor Force Statistics database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its series give the single-country labour measure.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004953:23:0", "occurrence_id": "c5b05031d2b5ff96f4c20b32", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: controlling for this trend leaves no variation that can be exploited to identify these coefficients. <sup>8</sup>\n\n\nOur main industry-level female intensity variable ( _FL_ ) is computed as the average share of female\n\n\nworkers in each sector in a sample of some 20 developed and developing countries. We favor this\n\n\nmeasure because it is most representative of the female intensity of sectors in an average country.\n\n\nHowever, a drawback of our main measure is that it does not include non-manufacturing sectors. To\n\n\ncheck robustness of our results, we implement an alternative strategy, of constructing a measure of\n\n\n_FL_ based on data for a single country - the U.S.. We thus use the Labor Force Statistics database\n\n\nof the U.S. Bureau of Labor Statistics (BLS). Using data from the Current Population Survey,\n\n\nthe BLS publishes “Women in the Labor Force: A Databook” on an annual basis since 2005. It\n\n\ncontains information on total employment and the female share of employment in each industry\n\n\ncovered by the Census. The data are available at the 4-digit U.S. Census 2007 classification (262\n\n\ndistinct sectors, including both manufacturing and non-manufacturing). In order to construct the\n\n\nshare of female workers in total employment in sector _i_, _FLi_, we take the mean of this value across\n\n\nthe years for which the data on the female share of employment are available (2004-2009). The\n\n\nresulting sample includes 78 manufacturing and 15 non-manufacturing sectors. Appendix Table A1\n\n\nreports the values of _FLi_ for the top and bottom 5 sectors according to U.S. data. These sectors\n\n\nare similar to what we find in our baseline measure: the least female-intensive sectors tend to be\n\n\nin heavy machinery, while the most female-intensive sectors in textiles and apparel. Appendix\n\n\nTable A1 presents the basic summary statistics of the U.S.-based\n\nMention: \"Current Population Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The annual gender databook is published from this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004953:23:0", "occurrence_id": "8a30857d62f691bb07620048", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: controlling for this trend leaves no variation that can be exploited to identify these coefficients. <sup>8</sup>\n\n\nOur main industry-level female intensity variable ( _FL_ ) is computed as the average share of female\n\n\nworkers in each sector in a sample of some 20 developed and developing countries. We favor this\n\n\nmeasure because it is most representative of the female intensity of sectors in an average country.\n\n\nHowever, a drawback of our main measure is that it does not include non-manufacturing sectors. To\n\n\ncheck robustness of our results, we implement an alternative strategy, of constructing a measure of\n\n\n_FL_ based on data for a single country - the U.S.. We thus use the Labor Force Statistics database\n\n\nof the U.S. Bureau of Labor Statistics (BLS). Using data from the Current Population Survey,\n\n\nthe BLS publishes “Women in the Labor Force: A Databook” on an annual basis since 2005. It\n\n\ncontains information on total employment and the female share of employment in each industry\n\n\ncovered by the Census. The data are available at the 4-digit U.S. Census 2007 classification (262\n\n\ndistinct sectors, including both manufacturing and non-manufacturing). In order to construct the\n\n\nshare of female workers in total employment in sector _i_, _FLi_, we take the mean of this value across\n\n\nthe years for which the data on the female share of employment are available (2004-2009). The\n\n\nresulting sample includes 78 manufacturing and 15 non-manufacturing sectors. Appendix Table A1\n\n\nreports the values of _FLi_ for the top and bottom 5 sectors according to U.S. data. These sectors\n\n\nare similar to what we find in our baseline measure: the least female-intensive sectors tend to be\n\n\nin heavy machinery, while the most female-intensive sectors in textiles and apparel. Appendix\n\n\nTable A1 presents the basic summary statistics of the U.S.-based\n\nMention: \"U.S. Census 2007\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Employment is available at its four-digit classification of 262 sectors.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004956:1:0", "occurrence_id": "a7f80dce99cb0faad6052d9c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Policy Research Working Paper 5780\n\n### **Abstract**\n\nUsing a cross-section of more than 40,000 manufacturing\nand services firms in 79 developing countries from the\nWorld Bank’s Enterprise Surveys Database, this paper\nassesses how firm location determines the likelihood and\nextent of exporting in developing countries. Descriptive\nstatistics confirm higher export participation (but not\nintensity) for firms in core versus non-core regions,\ndespite the finding that firms in the core assess many\naspects of the investment climate more negatively.\nResults from a probit model show that, in addition to\nfirm-specific characteristics, both regional investment\nclimate and agglomeration factors have a significant\nimpact on export participation. Specifically, customs\nclearance and electricity quality matter for export\nparticipation for manufacturing firms. Although\nlocalization economies and export spillovers are\nassociated with increased exporting, the opposite is found\n\n\n\nfor urbanization economies for both manufacturing\nand services firms. The analysis finds that firm-level\ndeterminants of exporting matter more for firms located\nin non-core regions, while regional determinants and\nagglomeration economies play a larger role in core\nregions.\nThe findings point to the presence of congestion costs\nin the core, and suggest that policy interventions to target\nexport participation are likely to have a greater impact if\nthey are focused on core regions over non-core regions,\nwhere firm-specific factors predominate. Moreover,\nthe importance of export spillovers and localization\neconomies highlights the potential value of efforts to\nremove barriers to natural agglomeration both in core\nand non-core regions, for example through investments\nin infrastructure, the provision of social services, and\nregional integration arrangements.\n\n\n\nThis paper is a product of the International Trade Department, Poverty Reduction and Economic Management Network. It\nis part of a larger effort by the World Bank to provide open access to its research and make a contribution to development\npolicy discussions around the world. Policy Research Working Papers are also posted on the Web at http://econ.worldbank.\norg. The authors may be contacted at t\n\nMention: \"Enterprise Surveys Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Over forty thousand firms across 79 countries test how location shapes exporting.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004960:24:1", "occurrence_id": "7d34f53d910943641da99a6f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  at the onset of the\nfinancial crisis (October 2008 – April 2009). BEEPS 2009 provides data on 9,098 firms from the\n21 ECA countries covered in this background paper. The sample provides a representative\nsample of firms for each of these countries. <sup>7</sup> [^7: The survey covers all 29 countries in which the EBRD is operational, with the exception of Turkmenistan. See\n<u>[http://www.ebrd.com/country/sector/econo/surveys/beeps.htm for detailed information on BEEPS 2009.](http://www.ebrd.com/country/sector/econo/surveys/beeps.htm)</u>] By contrast, databases that offer financial statement\ninformation, such as Worldscope or Amadeus, typically cover only the larger, formal enterprises\nin the region.\n\nWe use three indicators of the use of bank credit by firms. Our first indicator _Bank loan_\nis a dummy variable measuring whether the firm had a bank loan or credit line at the time of the\nsurvey (end of 2008, beginning of 2009). Likewise the indicator _Overdraft_ is a dummy variable\nmeasuring whether the firm had access to an overdraft facility at the time of the survey. Finally,\nthe indicator _Leverage_ captures the share of assets purchased in 2007 that were financed with\nbank credit (only measured for those firms which actually purchased assets in 2007). Table 4.1\nprovides an overview of these three indicators of bank finance by country, whereby we\ndistinguish between EU 10 countries, (potential) EU candidate countries and EU neighborhood\ncountries. The table shows that half of the firms in Emerging Europe use bank loans or an\noverdraft facility, while on average only one-quarter of recent firm investment is financed\nthrough a bank loan. The use of bank credit differs strongly across regions. In the EU member\n\n\n6 With BEEPS Brown et al. (2011a), and Popov and Udell (2010) analyse credit availability,\n\nMention: \"BEEPS 2009\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Nine thousand firms across twenty-one countries are covered in this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004961:18:0", "occurrence_id": "fa101cb0b8ea2c2b49dcd2bf", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: An adult individual can get SS through his or her formal job, through his or her spouse‘s job or\n\nthrough parents if he or she is a dependant) <sup>9</sup> [^9: In Mexico, ss through a formal job covers spouse, parents living in household, and children up to 15, or up to 26 if\nstudent or disabled and living in the same house.] . The decision of spouses to move sectors based on SS\n\n\nstatus is likely to be taken at the household level. If it is a joint decision, ignoring the spouse‘s\n\n\ndecision is an omitted variable; but including it introduces endogeneity.\n\n\nWe address this potential concern by estimating the model at the household level. This abstracts\n\n\nfrom whether one or more members in the household switch between formal and informal jobs to\n\n\nlook at the status of the household as a whole. It limits the effects to those switches that affect the\n\n\nhousehold‘s source of health insurance.\n\n\nAn important concern is that the rollout of _SP_ could be correlated with the informality trends in\n\n\nmunicipalities. This would affect our estimates because the error term would be correlated with\n\n\nour estimate of interest. The results presented in Table 2 however indicate there is little\n\n\ncorrelation between the share of formality in a given period and municipality and the probability\n\n\nthat municipality receives _SP_ in the following period. The probability of _SP_ being rolled out is\n\n\nmeasured with a dummy variable that takes a value of 1 when a municipality shows at least one\n\n\naffiliate in the _SP_ registry, and zero otherwise. This is true whether the share of formality is\n\n\nlagged one or two years (Columns (1) – (4)). The rate of change of formality (measured as a\n\n\npercentage of households or of individuals in SS) in a municipality is likewise uncorrelated with\n\n\nthe timing of _SP_ ‘s rollout (Columns (5) and (6)). These results limit concerns of the possible\n\nendogeneity of the deployment of _SP_ across municipalities. <sup\n\nMention: \"_SP_ registry\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"A municipality is flagged formal if it lists at least one affiliate here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004981:21:1", "occurrence_id": "cd6c3bfc237646b1e3a74b5b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: >\n\n\n\n<mark>Securities market</mark>\n<mark>capitalization</mark>\n\n\n\n<mark>Standard & Poor’s; Bank</mark>\n<mark>of International</mark>\n<mark>Settlements</mark>\n\n\n\n<mark>Stock market capitalization plus Domestic private bond market</mark>\n<mark>capitalization as % of GDP.</mark>\n\n\n\n**<u><mark>Financial structure</mark></u>**\n<u><mark>Financial structure ratio</mark></u> <u><mark>Authors’ calculations</mark></u> <u><mark>Bank private credit / Stock value traded.</mark></u>\n\n\n\n<mark>Expected financial</mark>\n<mark>Authors’ calculations</mark>\n<mark>structure ratio</mark>\n\n\n\n<mark>The expected ratio is derived by regressing the Financial</mark>\n<mark>structure ratio on log real GDP per capita, legal origin, distance</mark>\n<mark>to the equator, population size and density, and natural</mark>\n<mark>resource as % of exports using OLS regression on annual</mark>\n<mark>OECD data.</mark>\n\n\n\n<mark>Log absolute value of the difference between the expected and</mark>\n\n<mark>Financial structure gap</mark> <mark>Authors’ calculations</mark> <mark>the actual Financial structure ratio. The residuals were first</mark>\n\n<mark>deflated by the prediction error.</mark>\n<mark>Fin. structure</mark>\n\n<mark>Actual Financial structure ratio / Expected Financial structure</mark>\n\n<mark>ratio/Optimal fin.</mark> <mark\n\nMention: \"OECD data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Regressions on annual values from this source generate the residuals.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004982:9:0", "occurrence_id": "9babe5d61e1f4885bc558812", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: being surveyed in larger economies and is chosen to make each country’s sample representative\n\n\nof the non-agricultural private economy.\n\n\nFrom the Enterprise Survey we construct the following four access to and use of financial\n\n\nservices indicators: (i) _access to finance_ is an indicator variable ranging from 1-5 with 1\n\n\nindicating access to finance is “no obstacle” to the operation of firm to 5 indicating a “very\n\n\nsevere obstacle”; (ii) _account_ is a dummy variable equal to one if the firm has an account at the\n\n\ntime of the survey and zero otherwise; (iii) _overdraft_ is a dummy variable equal to one if the firm\n\n\nhas an overdraft facility at the time of the survey and zero otherwise; (iv) _loan_ is a dummy\n\n\nvariable equal to one if the firm has a line of credit or loan from a financial institution at the time\n\n\nof survey and zero otherwise.\n\n\nWe match the two samples by building a cross-sectional dataset that matches the firm\n\n\ncharacteristics with the average of the available data from the FSAP reports. Maximum country\n\n\noverlap between the two data sources is 54 countries with over 25,000 firm level observations.\n\n\nAppendix Table A1 lists the countries in our sample, a breakdown of the firm distribution by\n\n\ncountry, and by-country summary statistics of the FSAP variables we will use in the subsequent\n\n\nanalysis. Table 1 provides descriptive statistics and Table 2 correlations on the country-level.\n\n\nThe descriptive statistics in Table 1 show that over 90% of firms in our sample have an\n\n\naccount. This percentage, however, varies significantly across countries. While in the Slovak\n\n\nRepublic, 20.8% of firms have an account, 99.8% do so in Croatia. Almost 50% of firms have\n\n\nan overdraft facility and 45% have a loan. Behind this average, however, are again large cross\n\ncountry variations. While only 1.3% of firms have an overdraft facility and 3.\n\nMention: \"Enterprise Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Four financial access indicators are constructed from this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004982:9:0", "occurrence_id": "23d86cd75d78f13bfcd8fa3f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: being surveyed in larger economies and is chosen to make each country’s sample representative\n\n\nof the non-agricultural private economy.\n\n\nFrom the Enterprise Survey we construct the following four access to and use of financial\n\n\nservices indicators: (i) _access to finance_ is an indicator variable ranging from 1-5 with 1\n\n\nindicating access to finance is “no obstacle” to the operation of firm to 5 indicating a “very\n\n\nsevere obstacle”; (ii) _account_ is a dummy variable equal to one if the firm has an account at the\n\n\ntime of the survey and zero otherwise; (iii) _overdraft_ is a dummy variable equal to one if the firm\n\n\nhas an overdraft facility at the time of the survey and zero otherwise; (iv) _loan_ is a dummy\n\n\nvariable equal to one if the firm has a line of credit or loan from a financial institution at the time\n\n\nof survey and zero otherwise.\n\n\nWe match the two samples by building a cross-sectional dataset that matches the firm\n\n\ncharacteristics with the average of the available data from the FSAP reports. Maximum country\n\n\noverlap between the two data sources is 54 countries with over 25,000 firm level observations.\n\n\nAppendix Table A1 lists the countries in our sample, a breakdown of the firm distribution by\n\n\ncountry, and by-country summary statistics of the FSAP variables we will use in the subsequent\n\n\nanalysis. Table 1 provides descriptive statistics and Table 2 correlations on the country-level.\n\n\nThe descriptive statistics in Table 1 show that over 90% of firms in our sample have an\n\n\naccount. This percentage, however, varies significantly across countries. While in the Slovak\n\n\nRepublic, 20.8% of firms have an account, 99.8% do so in Croatia. Almost 50% of firms have\n\n\nan overdraft facility and 45% have a loan. Behind this average, however, are again large cross\n\ncountry variations. While only 1.3% of firms have an overdraft facility and 3.\n\nMention: \"FSAP reports\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Their averaged data supply the country-level financial characteristics.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004989:15:0", "occurrence_id": "e05bbd8ab6abad5f60604fc5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ANGOLA’S INFRASTRUCTURE: A CONTINENTAL PERSPECTIVE\n\n\nPower\n\n\n**Achievements**\n\n\nAngola has been making substantial investments in the power sector since 2002 to restore and\nreconstruct the infrastructure that was destroyed during the civil war. Recent estimates from Empresa\nNacional de Electricidade (ENE)—Angola’s major power-generating company—indicate that these\ninvestments led to an increase in generation capacity from around 830 MW in 2002 to over 1,200 MW in\n2008 (World Bank 2010a) (table 2). In terms of per capita generating capacity, Angola fares better than\nthe average African fragile state or resource-rich country. Angola has 70 MW per million people\ncompared to resource-rich or low-income fragile countries, which have only 43 or 46 MW per million\npeople, respectively (table 1).\n\n\nFurthermore, a relatively high share of Angola’s generation capacity is actually operational. In 2008\nalmost 1,000 MW or 80 percent of the installed power generation capacity was operational. This is\nhigher, on average, than resource-rich countries, where roughly 66 percent was operational and generating\npower. Thus, Angola has been able to rapidly ramp up its power production over the past decade. This\nincreased generation and operational capacity facilitated a 13 percent average annual growth in power\nproduction between 1999 and 2008. As of 2008 around 4,133 GWh of power was produced, a steep\nincrease compared to 1999 production levels of 1,295 GWh (World Bank 2010a).\n\n\nThe delay involved in obtaining an electricity connection has fallen dramatically since 2007.\nInvestment climate surveys in 2007 reported that firms encountered delays of over two months in Luanda\nand over six months in other parts of Angola in order to obtain a new electricity connection (World Bank\n2007b). But Angola has achieved tremendous progress in this area—firms in 2010 recorded only a sevenday delay, on average, in obtaining a power connection (\n\nMention: \"Investment climate surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Connection delays reported by firms fall sharply after this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004990:16:3", "occurrence_id": "fe145ec85f51da04075f3efc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Khartoum–Juba</u> <u>Road and river</u> <u>13.5</u> <u>127</u>\n<u>Road</u> <u>7.5</u>\n<u>Malakal–Juba</u> <u>River</u> <u>2</u> <u>270</u>\n\n\n\nMode of\n<u>transport</u> <u>Time (days)</u>\n\n\n\n_Source:_ AICD calculations based on data from Yoshino and others (2009); UNJLC and\n<u>FAO (2005); and Keer-MISC ( http://www.keer-misc.com)</u>\n\nMention: \"Yoshino and others (2009)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Transport times by mode are drawn from this source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004998:31:1", "occurrence_id": "6d8a521c3d12084aef5c7243", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: tools/data)\n_Note_ : Access figures from Demographic and Health Surveys (1991 and 1998) and Multiple Indicators Cluster Survey (2006) as reported by\nWHO 2010a and 2010b. A country is considered to have non-scarce water resources if renewable internal freshwater resources per capita are\ngreater than 3,000 mm.\n\n- = data not available.\n\n\nOpen defecation has also declined, if slowly, with the expansion of traditional latrines (figure 15b).\nThe practice declined from 9 percent to 7 percent between 1998 and 2006, both figures being only about a\nquarter of the typical level of open defecation in resource-rich countries (table 9). Over this period, use of\ntraditional latrines increased from 33 percent of the population in 1998 to 35 percent in 2006. But because\nthe overall level of access to improved sanitation in 2006 remained just 47 percent, the country was far\nbehind the goal of achieving the Millennium Development Goal of 74 percent of population with access\nto improved sanitation (AMCOW 2010b).\n\n\n26\n\nMention: \"Demographic and Health Surveys\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Access figures from these two rounds feed the water scarcity classification.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004998:31:1", "occurrence_id": "4a79fdab2084f831bf606e33", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: tools/data)\n_Note_ : Access figures from Demographic and Health Surveys (1991 and 1998) and Multiple Indicators Cluster Survey (2006) as reported by\nWHO 2010a and 2010b. A country is considered to have non-scarce water resources if renewable internal freshwater resources per capita are\ngreater than 3,000 mm.\n\n- = data not available.\n\n\nOpen defecation has also declined, if slowly, with the expansion of traditional latrines (figure 15b).\nThe practice declined from 9 percent to 7 percent between 1998 and 2006, both figures being only about a\nquarter of the typical level of open defecation in resource-rich countries (table 9). Over this period, use of\ntraditional latrines increased from 33 percent of the population in 1998 to 35 percent in 2006. But because\nthe overall level of access to improved sanitation in 2006 remained just 47 percent, the country was far\nbehind the goal of achieving the Millennium Development Goal of 74 percent of population with access\nto improved sanitation (AMCOW 2010b).\n\n\n26\n\nMention: \"Multiple Indicators Cluster Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The later round contributing to the same access figures.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005006:1:0", "occurrence_id": "f69e77cc7325015813fe7c0c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Policy Research Working Paper 5830\n\n### **Abstract**\n\nWhen deciding who should receive welfare benefits with\nthe aim to increase household well-being, it is necessary\nto understand the effects of the distribution of power\nwithin the households at which the aid is directed. Two\nprimary household models have been used to study intrahousehold bargaining and decision making: the unitary\nmodel and the collective model. The unitary model seems\nto fit Pakistan’s context because the prevailing traditional\nculture positions the male head as the household\ndecision maker. However, using a set of direct measures\n\n\n\nof decision-making power from the Pakistan Social and\nLiving Standard Measurement Survey, this study finds\nthat even in a country where men seem to have more\npower than women, the collective household bargaining\nmodel applies. This study also finds that, in Pakistan,\nwhen women have more decision-making power at home,\nhouseholds tend to spend more on women’s preferred\ngoods (such as clothing and education), family members\neat more non-grain food items, and children, particularly\ngirls, are more likely to be enrolled in school.\n\n\n\nThis paper is a product of the Human Development Social Protection Unit, South Asia Region. It is part of a larger effort\nby the World Bank to provide open access to its research and make a contribution to development policy discussions\naround the world. Policy Research Working Papers are also posted on the Web at http://econ.worldbank.org. The author\nmay be contacted at xhou@worldbank.org.\n\n\n_The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development_\n_issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the_\n_names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those_\n_of the authors. They\n\nMention: \"the Pakistan Social and\nLiving Standard Measurement Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Direct measures of household decision-making power come from this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005014:7:0", "occurrence_id": "69c45c21434858c59e3d3aeb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: declining trend, in particular among the EU candidate and EU neighborhood countries, and in\nboth of these groups a decline in national savings ensued.\n\n\nThe case of emerging Europe also appears to be consistent with what economic theory\nwould predict: capital flows to high-growth countries (Figure 3; left chart). Hence, the allocation\npuzzle—that, paradoxically, capital flows to low-growth countries instead of high-growth\ncountries—does not appear true in Europe either. These figures also suggest that the underlying\nstrength of the link between foreign savings and growth might vary across groups. Still, an\nincrease in current account deficits (i.e., capital inflows that happen to be absorbed) and an\nincrease in growth are common among two of these country groups in emerging Europe (Figure\n3; right chart). The opposite is the case, however, among emerging markets outside Europe—that\nis, high growth in incomes occurs jointly with an increase in current account surpluses (i.e., the\nallocation puzzle holds).\n\n\n**Figure 3. Current Account Balances (% GDP) and Growth in Emerging Economies**\n\n\n\n\n\n\n\nSource: Authors’ calculations based on World Bank and IMF data (2011). Note: Period 1997-2008.\n\n\nIn terms of the composition, capital flows take the form of FDI, portfolio, and banking\n(or other) flows. Clearly, FDI in emerging Europe played a key role; it is particularly large across\nall three groups of emerging European countries and is positively correlated with growth (Figure\n4). This is even more evident in the period immediately preceding the crisis. But the most\ndistinctive feature is the large role played by banking flows, which comprise both wholesale\nfunding and financial FDI. The latter relates to the relationship between parent banks and their\nsubsidiaries. Indeed, no other region appears to experience such a degree of ―ownership-driven‖\ncapital integration. Financial FDI may well combine the best aspects of alternative financing\nsources; namely\n\nMention: \"World Bank and IMF data (2011)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"External balances and growth over this period are computed from these sources.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005015:27:1", "occurrence_id": "b8f94f3a2fd2d4ced8c680b2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: boreholes* % pop 15.3 49.0 41.5\n\nAccess to surface water* % pop 30.3 23.7 32.7\n\nFlush toilets* % pop 5.3 11.2 9.8\n\nImproved latrines* % pop 15.1 6.4 9.2\n\nTraditional latrines* % pop 69.8 54.8 52.4\n\nOpen defecation* % pop 9.5 27.6 34.2\n\nDomestic water consumption** liter/capita/day 25 78.9 102.9\n\nCollection ratio** % sales 88.0 69.7 94.4\n\nDistribution losses** % production 45 43.9 33.4\n\nCost recovery of tariffs** % total costs 51.6 59.3 44.2\n\nTariff (US cents per m3) ** Republic of Congo Scarce water resources Other developing regions\n\nResidential tariff (at 30 m3) ** 48.6 60.26\n3.0–60.0\nNon-residential tariff (at100 m3/mo** 48.6 120.74\n\nSources: * Household survey, 2001; refers to national access. ** AICD database, 2005; refers only to SNDE.\n\n\n23\n\nMention: \"AICD database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Utility-specific access figures for that year come from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:004969:10:0", "occurrence_id": "f837c2f3ed6b8e6ab1abccce", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 6: Annual growth in modern and traditional services, 2000-08.**\n\n\n_Notes:_ Modern Services include communication, insurance, finance, computer & information, royalties and\nlicense fees and other business services. Traditional service are Transport, Travel, Construction, Personal, cultural\nand recreational services.\n\n\n**2.5** **Europe’s comparative advantage in service exports**\n\nNext we study the revealed comparative advantage (RCA), as defined by the share of a\nparticular export in a country‘s export basket compared to the share of that export in the global\nexport basket, in the three European groups. <sup>4</sup> EU-15 has RCA in services and the gap between\ngoods and services is increasing. NMS has also a comparative advantage in services (see Figure\n7). The gap to goods was large in the 1990s but has now decreased to the same level as in EU15. ACC did have a comparative advantage in services until early 2000s when it dropped and\nhas since had a comparative disadvantage in services.\n\n**Figure 7: Revealed comparative advantage in services and goods, for EU-15, NMS and ACC, 1990-007.**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n_Source_ : Authors‘ calculation using Balance of Payments, IMF. 2011.\n\n\n4 Formally, the RCA index for country _C_ and product _J_ is defined as: _RCA_   _ECJ_ / _ECT_  / _EWJ_ / _EWT_ , where _E_ denotes exports, _T_\n\ndenotes all products, and _W_ denotes the world.\n\n\n9\n\nMention: \"Balance of Payments, IMF\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Regional flow series are computed from this publication.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005017:21:0", "occurrence_id": "192785f135eb5d59086d6dd2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Transition matrices from panel data_\n\nPrevious figures describe net job flows of employment and unemployment. However, these net\nflows only give a partial description of the workings of the labor market. Net job flows can be\ndecomposed into four components, as follows:\n\n\nthat is, the change in total employment between final ( ) and initial period ( ) equals the growth\nof employment due to transitions from inactivity (and from out of the labor force) ( ) and\nfrom unemployment ( ), minus those who lost employment and became inactive ( )\nor unemployed ( ). Similar decompositions can be done for net flows in unemployment:\n\n\nand in inactivity\n\n\nThese components can be observed in the cells of a transition matrix:\n\n|Col1|Inactive<br>f|Unemployed<br>f|Employed<br>f|\n|---|---|---|---|\n|Inactive **i **|-|=|=|\n|Unemployed **i**|=|-|=|\n|Employed **i**|=|=|-|\n\n\n\nThis transition matrix provides more nuanced information about job flows and reveals how changes\nin employment status are the consequence of gross flows to and from unemployment and inactivity.\nThese flows also indicate the likelihood of moving from one status to another. The Mexican\nOccupation and Employment Survey ( _Encuesta Nacional de Ocupación y Empleo_ —ENOE) has a\nrotating panel that interviews the same household for five consecutive quarters. This implies that we\ncan observe the same household and their individuals, and therefore its employment dynamics, for\n\n\n20\n\nMention: \"Mexican\nOccupation and Employment Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its rotating panel re-interviews the same households, giving status flows.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005024:15:0", "occurrence_id": "1311d4d571ae62ff8b7718f9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: (e.g. Bates, 1981; Krueger, Schiff and Valdes, 1988; van de Walle, 2001; Anderson and\n\nMasters, 2009; Bates and Block, 2010). <sup>41</sup>\n\n\nFollowing recommendations by the World Bank and the International Monetary Fund\n\n(Badiane et al., 2002), African cotton sectors have however seen their share of reforms. <sup>42</sup>\n\n\nFigure 5, which uses the database constructed by Delpeuch and Leblois (2011), depicts the\n\n\nproportion of competitive cotton markets in SSA countries in 1990 and in 2008. On the one\n\n\nhand, this figure shows that an increasing number of markets have become competitive; but\n\n\non the other, it also points to the fact that, in many countries, markets remain non\n\ncompetitive. Indeed, while many countries have undertaken reforms over the past two\n\n\ndecades, the nature of these reforms has widely varied across regions, ranging from far\n\nreaching market and price liberalizations, to only marginal adjustments.\n\n\nSchematically, reforms have come in two waves: a first series of far-reaching reforms in ESA\n\n\nin the mid-1990s and a second series of more modest reforms in WCA in the course of the\n\n2000s. <sup>43</sup> As illustrated in figure 6, a number of countries in ESA have privatized their\n\nparastatals, liberalized prices and allowed for competition. <sup>44</sup> Private stakeholders have\n\n\nhowever reacted to these reforms in different ways. While the Tanzanian cotton market is\n\n\nconsidered to be relatively highly competitive (Poulton, 2009); markets in Uganda, Zambia\n\n\nand Zimbabwe are said to exhibit lower degrees of competition (Brambilla and Porto, 2009;\n\n\nPoulton and Hanyami-Mlambo, 2009).\n\n\nResistance to market reforms has been much stronger in WCA, where the reforms\n\n\nimplemented in the\n\nMention: \"Delpeuch and Leblois (2011)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its constructed database gives the share of competitive cotton markets.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005028:42:0", "occurrence_id": "f3b4ea5d805ee344b387fb9c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Appendix B: Data Sour.ces and Unit Root Tests\n\n\nThe data sources employed of the series used in the paper are the\nfollowing.\n\n\nCHILE:\nConsumption: Quarterly index of total consumption. Data constructed by the\nauthors based on an interpolation of annual data using a regression (annual\nfrequency) of consumption on the manufacturing industry physical sales index and\nthe real value of imports of consumer goods. Raw data from Banco Central de Chile\n\n(1989), Indicadores Econ6micos y Sociales 1960-1988, except data on imports that\nare from P. Meller and M. Cabezas (1989), \"Estimaci6n de las elasticidades precio\ne ingreso de las importaciones chilenas 1974-87, Colecci6n Estudios CIEPLAN 28).\nMoney: Ml series (currency plus demand deposits), Banco Central de Chile, op.\ncit.\nReal GDP: From P. Meller and Cabezas, op. cit.\n\nConsumption Deflator: National CPI. For 1979-89, Banco Central de Chile, op. cit.\nFor 1974-78, R. Cortazar and J. Marshall (1980), \"Indice de Precios al\nConsumidor\", Colecci6n Estudios CIEPLAN 4.\nNominal Interest Rate: Effective monthly interest rate paid on 30-89 days\ndeposits. The basic data is available monthly (average of month). Because the\npublished return is the average of the month, we need to take a convention to\ncompute the end-of-quarter returns. For example, we compute the nominal quarterly\nreturn for 1975.1 by composing the monthly returns of March, April and June of\n1975. (Therefore the quarterly return is approximately the quarter return from\nMarch 15 to June 15.)\n\nPopulation: IMF, International Financial Statistic. The year was assigned to\nJune. The other months were computed by geometric interpolation between two\n\nconsecutive years, and the\n\nMention: \"manufacturing industry physical sales index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Consumption is regressed on this volume index to interpolate annual data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005028:42:0", "occurrence_id": "89285e20feaeab326905d066", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Appendix B: Data Sour.ces and Unit Root Tests\n\n\nThe data sources employed of the series used in the paper are the\nfollowing.\n\n\nCHILE:\nConsumption: Quarterly index of total consumption. Data constructed by the\nauthors based on an interpolation of annual data using a regression (annual\nfrequency) of consumption on the manufacturing industry physical sales index and\nthe real value of imports of consumer goods. Raw data from Banco Central de Chile\n\n(1989), Indicadores Econ6micos y Sociales 1960-1988, except data on imports that\nare from P. Meller and M. Cabezas (1989), \"Estimaci6n de las elasticidades precio\ne ingreso de las importaciones chilenas 1974-87, Colecci6n Estudios CIEPLAN 28).\nMoney: Ml series (currency plus demand deposits), Banco Central de Chile, op.\ncit.\nReal GDP: From P. Meller and Cabezas, op. cit.\n\nConsumption Deflator: National CPI. For 1979-89, Banco Central de Chile, op. cit.\nFor 1974-78, R. Cortazar and J. Marshall (1980), \"Indice de Precios al\nConsumidor\", Colecci6n Estudios CIEPLAN 4.\nNominal Interest Rate: Effective monthly interest rate paid on 30-89 days\ndeposits. The basic data is available monthly (average of month). Because the\npublished return is the average of the month, we need to take a convention to\ncompute the end-of-quarter returns. For example, we compute the nominal quarterly\nreturn for 1975.1 by composing the monthly returns of March, April and June of\n1975. (Therefore the quarterly return is approximately the quarter return from\nMarch 15 to June 15.)\n\nPopulation: IMF, International Financial Statistic. The year was assigned to\nJune. The other months were computed by geometric interpolation between two\n\nconsecutive years, and the\n\nMention: \"Banco Central de Chile\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The raw series behind the reconstruction come from this central bank.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005028:42:0", "occurrence_id": "4a6c7575e1210f0560a3f90d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Appendix B: Data Sour.ces and Unit Root Tests\n\n\nThe data sources employed of the series used in the paper are the\nfollowing.\n\n\nCHILE:\nConsumption: Quarterly index of total consumption. Data constructed by the\nauthors based on an interpolation of annual data using a regression (annual\nfrequency) of consumption on the manufacturing industry physical sales index and\nthe real value of imports of consumer goods. Raw data from Banco Central de Chile\n\n(1989), Indicadores Econ6micos y Sociales 1960-1988, except data on imports that\nare from P. Meller and M. Cabezas (1989), \"Estimaci6n de las elasticidades precio\ne ingreso de las importaciones chilenas 1974-87, Colecci6n Estudios CIEPLAN 28).\nMoney: Ml series (currency plus demand deposits), Banco Central de Chile, op.\ncit.\nReal GDP: From P. Meller and Cabezas, op. cit.\n\nConsumption Deflator: National CPI. For 1979-89, Banco Central de Chile, op. cit.\nFor 1974-78, R. Cortazar and J. Marshall (1980), \"Indice de Precios al\nConsumidor\", Colecci6n Estudios CIEPLAN 4.\nNominal Interest Rate: Effective monthly interest rate paid on 30-89 days\ndeposits. The basic data is available monthly (average of month). Because the\npublished return is the average of the month, we need to take a convention to\ncompute the end-of-quarter returns. For example, we compute the nominal quarterly\nreturn for 1975.1 by composing the monthly returns of March, April and June of\n1975. (Therefore the quarterly return is approximately the quarter return from\nMarch 15 to June 15.)\n\nPopulation: IMF, International Financial Statistic. The year was assigned to\nJune. The other months were computed by geometric interpolation between two\n\nconsecutive years, and the\n\nMention: \"Indicadores Econ6micos y Sociales\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"This publication over that span supplies the raw data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005028:42:0", "occurrence_id": "3b177cb26687a33f2edf8fa0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Appendix B: Data Sour.ces and Unit Root Tests\n\n\nThe data sources employed of the series used in the paper are the\nfollowing.\n\n\nCHILE:\nConsumption: Quarterly index of total consumption. Data constructed by the\nauthors based on an interpolation of annual data using a regression (annual\nfrequency) of consumption on the manufacturing industry physical sales index and\nthe real value of imports of consumer goods. Raw data from Banco Central de Chile\n\n(1989), Indicadores Econ6micos y Sociales 1960-1988, except data on imports that\nare from P. Meller and M. Cabezas (1989), \"Estimaci6n de las elasticidades precio\ne ingreso de las importaciones chilenas 1974-87, Colecci6n Estudios CIEPLAN 28).\nMoney: Ml series (currency plus demand deposits), Banco Central de Chile, op.\ncit.\nReal GDP: From P. Meller and Cabezas, op. cit.\n\nConsumption Deflator: National CPI. For 1979-89, Banco Central de Chile, op. cit.\nFor 1974-78, R. Cortazar and J. Marshall (1980), \"Indice de Precios al\nConsumidor\", Colecci6n Estudios CIEPLAN 4.\nNominal Interest Rate: Effective monthly interest rate paid on 30-89 days\ndeposits. The basic data is available monthly (average of month). Because the\npublished return is the average of the month, we need to take a convention to\ncompute the end-of-quarter returns. For example, we compute the nominal quarterly\nreturn for 1975.1 by composing the monthly returns of March, April and June of\n1975. (Therefore the quarterly return is approximately the quarter return from\nMarch 15 to June 15.)\n\nPopulation: IMF, International Financial Statistic. The year was assigned to\nJune. The other months were computed by geometric interpolation between two\n\nconsecutive years, and the\n\nMention: \"National CPI\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"It serves as the consumption deflator for these years.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005028:42:0", "occurrence_id": "41d9628cdbc6e2a06768b750", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Appendix B: Data Sour.ces and Unit Root Tests\n\n\nThe data sources employed of the series used in the paper are the\nfollowing.\n\n\nCHILE:\nConsumption: Quarterly index of total consumption. Data constructed by the\nauthors based on an interpolation of annual data using a regression (annual\nfrequency) of consumption on the manufacturing industry physical sales index and\nthe real value of imports of consumer goods. Raw data from Banco Central de Chile\n\n(1989), Indicadores Econ6micos y Sociales 1960-1988, except data on imports that\nare from P. Meller and M. Cabezas (1989), \"Estimaci6n de las elasticidades precio\ne ingreso de las importaciones chilenas 1974-87, Colecci6n Estudios CIEPLAN 28).\nMoney: Ml series (currency plus demand deposits), Banco Central de Chile, op.\ncit.\nReal GDP: From P. Meller and Cabezas, op. cit.\n\nConsumption Deflator: National CPI. For 1979-89, Banco Central de Chile, op. cit.\nFor 1974-78, R. Cortazar and J. Marshall (1980), \"Indice de Precios al\nConsumidor\", Colecci6n Estudios CIEPLAN 4.\nNominal Interest Rate: Effective monthly interest rate paid on 30-89 days\ndeposits. The basic data is available monthly (average of month). Because the\npublished return is the average of the month, we need to take a convention to\ncompute the end-of-quarter returns. For example, we compute the nominal quarterly\nreturn for 1975.1 by composing the monthly returns of March, April and June of\n1975. (Therefore the quarterly return is approximately the quarter return from\nMarch 15 to June 15.)\n\nPopulation: IMF, International Financial Statistic. The year was assigned to\nJune. The other months were computed by geometric interpolation between two\n\nconsecutive years, and the\n\nMention: \"Indice de Precios al\nConsumidor\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"An earlier study supplies the price series for the earlier years.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005029:16:1", "occurrence_id": "6e2a9441201f826563721838", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: >0.44<br>0.44<br>0.50<br>0.09<br>0.29<br>0.27<br>0.44<br>0.18<br>0.38<br>0.02<br>0.14<br>0.44<br>0.50<br> <br> <br>0.12<br>0.32<br>0.37<br>0.48<br>0.26<br>0.44<br>0.17<br>0.38<br>0.03<br>0.18<br>0.04<br>0.19<br>0.12<br>0.32<br>0.18<br>0.38|\n\n\n\nSource: Own computations using June 2010 (EFSA II) surveys. Note:\nThe sample is restricted to the households residing in the six strata that\ncover areas directly affected by the earthquake. See Table 16 for a\ncomplete description of variables in directly affected and non directly\naffected areas.\n\n\n_(e)_ _Pre-earthquake conditions_\n\n\nTable 5 presents summary statistics by quintile of pre-earthquake wealth. The poorest\nhouseholds lived mostly in the East of the directly affected area (Léogane, Gressier, Jacmel, Petit\nGoâve, Grand Goâve and Croix-des-Bouquets). They were mostly agricultural households: 66%\nof them were practicing agriculture, compared to only 5% among the wealthiest, who were\nmostly concentrated in Port-au-Prince or Pétionville. The poorest households lived in houses\nwith no electricity and no toilets, and do not have access privately to water. They had no car and\n\n\n15\n\nMention: \"June 2010 (EFSA II) surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Household statistics for the six strata are computed from this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005030:23:0", "occurrence_id": "52306a326a34c2085ca99a21", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: of a foreign competitor, given the controls in our framework. <sup>32</sup> [^32: While Ayyagari et al. (2007) find that the presence of a foreign competitor matters for innovation, their\nspecification is quite different from ours: they test for the correlation of different business environment variables\nseparately rather than controlling for a number of key variables simultaneously as we do.] This empirical result is not\nunexpected -- in light of our discussion above of the logically ambiguous and countervailing\neffects of the degree of active competition on R&D investment and the resulting innovation.\n\n\n**_National-level market competitiveness variables_**\n\nTo explore the importance of national-level measures of the openness to competition of\nthe business environment, we follow the spirit of Loayza, Oviedo and Serven (2010) and\nDjankov, McLiesh and Ramalho (2006) by examining the associations between our innovation\nvariables and the aggregate national rank across regulations that affect all 10 stages of the life of\na business, as covered by the Doing Business Indicators (DB) (see the discussion in Section 3,\nabove).\n\n\nRather than replace the country-level fixed effects with the DB rank order variables as\nadditional explanatory variables in the existing firm-level regressions, we examine the rank\ncorrelation between the country-level fixed effects from the firm-level regressions and the DB\nvariables. A key advantage of this approach is that it overcomes the challenging interpretation\nproblem, both economically and statistically, of including a rank order variable in regressions\nhaving a mix of discrete and continuous variables. This approach also allows the relative\nimportance of the 10 constituent DB sub-indices to be examined, as the collinearity of the\nindividual DB variables does not conflate the econometric results when correlations are\nexamined one-by-one. We report correlation results only for innovation outcomes, as the fixed\neffects from employment growth and TFP reflect a broader and more various range of factors –\nincluding important cross-country differences in macroeconomic facets of growth in the case of\nemployment growth, and firm-specific knowledge and other assets that have built up over time in\nthe\n\nMention: \"Doing Business Indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Regulations across ten business stages are ranked using these indicators.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005034:15:0", "occurrence_id": "2ba9836edd2175ceb4390f0d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 4: Firm growth across age and size classes**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n_Source: Interim Enterprise Survey, 2009/10; Own calculations_\n\n\n**The effect of the size of the locality**\n**Growth has been geographically asymmetrical in South Asia.** Urban areas in the South Asia have\nproven to be the bearers of growth. This indicates that SAR countries switch their sectoral composition\naway from agriculture into industry and as technological advances in domestic agriculture release labor\nfrom agriculture to migrate to cities (Henderson, 2003) <sup>9</sup> . This leaves ‘lagging areas’ with higher poverty\nrates: the poverty headcount is related to distance, and the mass of poverty is related to density. Aging\nareas tend to have a higher proportion of poor residents, and the leading areas tend to contain a higher\nshare of the country’s poor people, because of the dense population in leading areas (World\nDevelopment Report 2009).\n\n\n**Urbanization is also reflected in employment growth.** South Asia’s economic and social transformation\nhas been complemented by rapid urbanization. In the following, we define urban areas as either the\ncapital city or as a city with at least one million inhabitants. Descriptive statistics (see table below)\nestablish a difference in relative employment growth of approximately five percentage points across the\n\n\n9 The dataset allows for aggregate data, and not for a deeper discussion of agglomeration effects and congestion\ncosts of urbanization.\n\n\n14\n\nMention: \"Interim Enterprise Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Firm growth by age and size class is computed from this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005037:7:0", "occurrence_id": "57fcb7513cf72f5ff0670b94", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: —Insert Figure 1b—\n\n\nAlthough debt relief has brought the debt of HIPCs to sustainable levels and allowed these\ncountries to gain access to external financing, the evidence suggests that such countries remain\nsomehow credit constrained (for instance vis-à-vis other IDA countries that, despite their higher\ndebt ratios, could build a track record of creditworthiness). Furthermore, the new borrowing from\nprivate creditors to HIPC has come at high interest rates. The default risk premium that postMDRI countries have to pay on new debt can be measured by the spreads between the yields on\nthe dollar bonds issued by Ghana and Senegal and the yield on US Treasuries of the same maturity.\nFigure 2a shows that the yield spread on the Ghana bond has remained above 4% while the spread\non the Senegal bond has been even higher at 6.5%.\n\n\n—Insert Figure 2a—\n\n\nFurther evidence of the relative high interest rates at which post-MDRIs borrow comes from the\nGDF database. Figure 2b shows the simple mean for post-MDRI countries of the average interest\nrates that they pay on new public and publicly guaranteed debt from private creditors and from\noffi cial creditors. <sup>6</sup> [^6: Country data on interest rates are weighted averages of the interest rates on new debt commitments taken from\nthe GDF database.] While interest rates on concessional offi cial lending has remained below 2% over\nthe whole period, interest rates on private lending has fluctuated around 4% reaching 6% in 2007\nwhen new borrowing became available after the implementation of MDRI relief. <sup>7</sup> [^7: Evidence of higher borrowing costs would emerge from excluding bank credit covered by guarantees of exportcredit agencies.] More importantly,\nthere is no evidence that debt relief initiatives have brought about a reduction in the interest rate\non new private debt despite the fall in debt levels.\n\n\n—Insert Figure 2b—\n\n\nThis brings us to the critical question of our paper: Would it have been possible to design the\ndebt relief initiative so that markets could distinguish\n\nMention: \"GDF database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Average borrowing rates for relieved countries are shown in this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005038:7:0", "occurrence_id": "d17f755fc37d8b772ac3067f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: External Wealth of Nations Mark II database developed by Lane and Milesi-Ferretti (2007) as\n\n\nthe ratio of the sum of total financial assets and financial liabilities to GDP. Finally, we relied on\n\nCalvo and Reinhart (2000) to construct banking crisis and currency crisis. <sup>2</sup> [^2: The period from 2000 through 2005 represents a time of stable economic growth in most countries in our sample,\nalso known as the ―Great Moderation‖ (Stock and Watson (2002)). The exception to this is Argentina, where a crisis\nlasted from 1999 through 2002.]\n\n\nTable 1 [about here]\n\n\nTable 1 shows summary statistics for financial sector real value added growth rates\n\n\n(calculated as log differences of dollar amounts at 1995 prices) for each country. The mean\n\n\ngrowth rate for each country is positive; however, the series for 17 out of the 28 countries exhibit\n\n\nnegative skewness. These countries, mostly emerging markets, are: Argentina; Bolivia; Chile;\n\n\nColombia; Costa Rica; Denmark; Hong Kong SAR, China; India; Indonesia; Italy; Korea;\n\n\nMexico; Philippines; Singapore; Sweden; Thailand; United States; and Venezuela. In contrast, as\n\n\ncan be seen from Table 1, countries with positively skewed financial sector growth series include\n\n\nmostly OECD economies.\n\n\nFigure 1 [about here]\n\n\nFurthermore, the average level of kurtosis is higher for countries with negative skew in\n\n\nthe series (6.619 compared to 5.527). Figure 1 illustrates the significance of the difference in the\n\n\nskew via a probability plot. The difference is seen in a greater degree of deviation from the\n\n\nnormal distribution (Gaussian bell curve) in the tails of the distribution of financial sector growth\n\n\nrates for the subsample of countries with negative skew. The difference is substantial: for\n\n\ninstance, an annual growth rate of minus 20 percent in financial sector real value added has 0.0\n\n\nprobability of occurrence under normality, while the empirical probability in countries with\n\n\npositive skew is 0\n\nMention: \"External Wealth of Nations Mark II database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Net foreign asset positions are measured as financial assets less liabilities here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005042:40:0", "occurrence_id": "1fbbc70b056b222ccbef0af2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Table 10: Household meal-consumption status during non-monga by membership status in safety-net program**\n\n**and receipt of support during non-monga (%)**\n\n\n\n\n\n\n\n16.4\n\n\n51.3\n\n\n32.4\n\n\n\n10.5\n\n\n59.0\n\n\n30.5\n\n\n\n8.5\n\n\n50.7\n\n\n40.8\n\n\n\n8.5\n\n\n50.9\n\n\n40.6\n\n\n\n**Total**\n\n\nStarvation\n\n\nMeal rationing\n\n\nFull meals\n\n\n\n6.4\n\n\n52.8\n\n\n40.8\n\n\n\n**No. observations** **38,476** **7,143** **695** **434,604**\n\n\n_Source_ : InM survey data (2006–07).\n\n\n39\n\nMention: \"InM survey data (2006–07)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Large-sample household observations come from this survey round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005043:25:0", "occurrence_id": "b67b2dd8f6f63c451a6aee4a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The relationship between sector and type of economic activity among household\n\n\nmembers, and household welfare in Uganda has been found in other analyses as well.\n\n\nConsumption regressions based on data from previous UNHS also show that type of\n\n\nemployment is a significant independent predictor of household welfare, with\n\n\nemployment in non-farm sectors having a greater effect on welfare (World Bank, 2006) **.**\n\n\nAlthough correlations between the growth of non-farm income opportunities and lower\n\n\npoverty, either over time in Uganda or at the household level in repeated cross\n\n\nsections do not necessarily imply causality, they do suggest _that the expansion of the_\n\n\n_non-farm enterprises in the informal sector was indeed good structural transformation_\n\n\n_for Uganda as it seeks to alleviate poverty_ .\n\n\n**6.** **Discussion**\n\n\nDespite the importance of non-farm household enterprise incomes in increasing\n\n\nhousehold welfare, it has to be recognized that most economic activities in non-farm\n\n\nenterprises use low capital and rudimentary technology. They are usually services, or\n\n\nthe production of low quality goods. Therefore, labor productivity growth in this sector\n\n\nhas limits and the risk of failure is also high. Although some hire external labor, many\n\n\ndepend only on family labor input. Nearly 80 percent did not have a hired worker in\n\n\n2005/06 while only 6 percent had three or more workers. When HEs are started as a\n\n\nsecondary source of income, employment was not always full time **.** Around 44 percent\n\n\nof HEs (in operation for at least one year) reported operating for less than 12 months\n\n\nin a year, with 21 percent operating for less than six months in 2005/06 for example.\n\n\nIn rural areas, it is common for households with non-farm enterprises to devote some\n\n\nweeks entirely to agriculture. About half of household enterprises seem to fail within\n\n\n24\n\nMention: \"UNHS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Consumption regressions show employment type predicts welfare.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005045:11:0", "occurrence_id": "d34d368de2328a187122165f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Contribution of prices and volumes**\n\n\nIn this section, the issue of the contribution of prices and volumes in the recent high growth in Peru‟s\nexports is explored. Due to limitations with the data available, the empirical analysis looks at this question\nseparately for traditional and non-traditional exports.\n\n\nFor the case of traditional exports, between 1994 and 1999 prices hardly varied, with an annual average\ngrowth of 0.7%, while volumes grew by 10.6%. <sup>11</sup> In contrast, between 2000 and 2009, the prices of these\nexports grew by an annual average of 13.7% and volumes by 5.3% (see Table 3). In the first period, the\nhigh growth of volumes reflects the growing extraction of minerals and hydrocarbons, following changes\nin the governing laws for both activities, introducing different types of incentives (e.g. related to taxes and\nexchange rates), the privatization of state mines and the access to agreements on tax stability, which\nhave contributed to boosting the production and the supply of this type of products, despite the low price.\nIn the second period, with the continuous price boom – except at the end of 2008 and the beginning of\n2009 due to the international financial crisis -, volumes grew by 5.3%, as a result of the lower grade of the\nmineral extracted from existing mines and few important new mines coming on stream. This has given\nrise to a relative stagnation of the volumes in recent years, despite the significant price increase (see\nFigure 5, mobile average of 6 months on the black line).\n\n\n11 / The analysis can only be performed with data starting in 1994, as the data from the Central Bank of Peru has\ndisaggregated data on the prices and volumes of traditional and non-traditional exports as from that date. A study\non this same topic is that of Banco Central de Peru (2008).\n\n\n10\n\nMention: \"the Central Bank of Peru\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its disaggregated export prices and volumes begin in that year.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005051:13:1", "occurrence_id": "18e70062dde678cd5c154d90", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Targeted|Targeted|Targeted|Targeted|Targeted|Targeted|Targeted|Targeted|Targeted|Targeted|\n|||||||||||||||\n|||||||||||||||\n|||||||||||||||\n\n\n\nSource: Own calculations based on SEDLAC databases\n\n\n10\n\nMention: \"SEDLAC databases\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Programme participation and welfare statistics are computed from these databases.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005052:6:3", "occurrence_id": "b814aaa72f89cc834d80e3d0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ss database includes a panel of firms. The panel feature allows us to tract\nfirms over time. It is also among the few databases that records annual information on firms’ product\n\n\n4 A formal proof of this relationship is presented in Klette and Kortum (2004).\n\n\n5\n\nMention: \"ss database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its firm panel allows tracking the same enterprises over time.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005054:7:1", "occurrence_id": "8d2a171a30bfabc36b1ff4f3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  the\n\n\ncommunity head to the nearest secondary school. The distance is self-reported by the\n\ncommunity head in the Service Availability Roster of the IFLS. <sup>5</sup>\n\n\nTable 2 presents descriptive statistics on the main variables used in our analysis. It\n\n\nshows that individuals with upper secondary or higher levels of education have, on\n\n\naverage, 108% higher wages than those with lower education. They also have 7.778 more\n\n\nyears of schooling. The respondents with an upper secondary education are younger than\n\n4 It is possible to estimate a non-parametric selection model with multiple levels of schooling but the data\nrequirements to do it are very strong. In particular, one needs one instrumental variable for each transition.\nIt is not feasible to pursue this with our dataset.\n5 We would have liked to use instead the distance between the community of residence in childhood and\nthe nearest school in childhood. Our hope is that current residence and current school availability are good\napproximations (as in Card, 1995). We show below that this measure of distance to school is a good\npredictor of upper secondary school attendance.\n\n\n6\n\nMention: \"Service Availability Roster of the IFLS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Community heads report distances to the nearest secondary school here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005055:14:1", "occurrence_id": "8823603929b4deb6e702b0fc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |Participants<br>of EIT in<br>(Math)|Share,<br>%|Participants<br>of EIT in<br>(English)|Share,<br>%|\n|---|---|---|---|---|---|---|---|---|---|\n|2009|391,257|329,839|84.0|215,080|55.0|175,493|44.9|70,315|18.0|\n|2010|363,751|284,550|78.0|133,549|34.1|183,682|46.9|46,678|11.9|\n\n\nSource: based on State Statistics Service of Ukraine statistics bulletins on general secondary institutions in the relevant years and\nofficial EITs reports of the Ukrainian Centre for Education Quality Monitoring.\n\n\n- EIT is the main but not the only way for entering the higher education system. According to\n\nthe Ministry of Education and Science of Ukraine in 2008 overall 91.2% of students were\n\n\nadmitted to the higher education institutions of ІІІ-ІV level of accreditation on the basis of\n\nthe EIT certificates <sup>21</sup> .\n\n\nSo far there has been little analysis of the EIT data. Besides the simple descriptive statistics in\n\n\nthe annual EIT report, there are three studies that use EIT data. Kovtunets et al (undated) provide\n\n\nsome analysis of whether the EIT exam results predict performance during the first year at\n\n\nuniversity. Muravyev and Talavera (2010) use the EIT to see how an announced (but not\n\n\nimplemented) language policy change has affected the subject choice and performance of\n\n\nstudents at minority language schools. The closest to the current paper is Coupé et al (2010) who\n\n\nuse a sample\n\nMention: \"EIT data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Certificate-based records remain little analysed beyond descriptive statistics.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005058:72:2", "occurrence_id": "2f4e2d0dac264d3607e61f61", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: >239<br>16.4<br>11.6<br>14.4<br>7.9<br>84<br>62<br>94<br>9.2<br>6.8<br>10.3<br>147<br>140<br>137<br>10.1<br>9.6<br>9.4<br>2.0<br>2.4<br>1.6<br>18<br>16<br>7 <br>2.0<br>1.7<br>0.8<br>29<br>29<br>16<br>2.0<br>1.9<br>1.1<br> <br> <br> <br> <br> <br>42.4<br>41.9<br>367<br>547<br>62.2<br>59.9<br>798<br>775<br>54.7<br>53.2|\n|Total|544|<br>912<br> <br>1456<br>|\n\n\n_Note: Multiple answers possible_\n\n_Source: Calculations based on the FinScope Survey 2009 data._\n\n\n\n\n\n\n\n\n\n**<u>Table A1.15: Distribution of Credit Sources, by Gender, 2006 (percent)</u>**\n\n|1.8.2|Credit sources, by gender|\n|---|---|\n|**_1.8.2 _**<br> <br>|Male<br>Female<br>Total|\n|<br>Relative or friend<br>Rotating saving & credit group (UPATU)<br>Saving and credit co-\n\nMention: \"the FinScope Survey 2009 data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Credit source distributions by gender are computed from this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005061:11:0", "occurrence_id": "cec7abd306677ddb0c628e13", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: MOZAMBIQUE’S INFRASTRUCTURE: A CONTINENTAL PERSPECTIVE\n\n\n**Figure 2. Mozambique’s population, income, and mineral resources are concentrated in the center and south**\n\n\na. Population b. Poverty\n\n\nc. Topography d. Natural resources\n\n\n_Source_ : AICD Interactive Infrastructure Atlas for Mozambique (www.infrastructureafrica.org).\n\n\n6\n\nMention: \"AICD Interactive Infrastructure Atlas for Mozambique\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Population, poverty, terrain and resource panels come from this atlas.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005067:10:1", "occurrence_id": "57a302856c847f0953e83e10", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _  1 _w_  1 _z_  1\n\n\n### ln Yit   z ln X itz  5.0  zz ln X itw ln X itz  it  it (1)\n\n_z_  1 _w_  1 _z_  1\n\n\n\n_zz_ _it_ _it_ _it_ _it_\n\n_w_  1 _z_  1\n\n\n\n 1 _w_  1 _z_  1\n\n\n\n1\n\n\n\n1\n\n\n\nwhere Y is real output and the inputs _X_ _z_ are labor, real materials, electricity and the\n\n\ncapital stock,  _it_ is a productivity shock known to the firm but unobserved by the\n\n9 We use the words plant and firm interchangeably, although plants are the unit on which the ENIA survey\ncollects data. Between 1997 and 2003 only 8.3% of Chilean plants are part of a multi-plant firm (Fernandes\nand Paunov, 2011).\n10 Due to a reorganization of the Chilean territory during our sample period, our final sample includes\nsector- _comunas_ present in the first and last sample years – 1992 and 2004 - as well as in an intermediate\nsample year 1998.\n11 Statistical tests based on OLS estimates indicate that the translog functional form is more appropriate\nthan the Cobb-Douglas functional form for the Chilean industries.\n\n\n8\n\nMention: \"the ENIA survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Plants rather than firms are its unit of observation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005070:10:1", "occurrence_id": "ec3f84bc62ebde5424093a72", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  regulation (such as, the number of documents required to\n\n\n4 We also experimented with using an overall (weighted) tariff rate, average values over 2001 to 2005 and taken\nfrom World Development Indicators, World Bank. However, our main results did not change much on adding the\ntariff measure interacted with _Documents_ to any of the specifications discussed in the paper. In fact, our results for\nthe main interaction terms were strengthened by using the weighted tariff measure stated.\n\n\n9\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Average weighted tariff rates over this window come from this source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005079:18:0", "occurrence_id": "ad59a89603312950e084e9b7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figures and Tables**\n\n\nFigure 1. Volatility of Aggregate Price Indices using IMF Commodity Indices\n\n\n\n\n\n\n\nNote: The figure shows the evolution of the annualized standard deviations of Hodrick-Prescott\nfiltered price series. The aggregate price indices for all primary, non-fuel primary and energy goods\nare from IMF Primary Commodity Price Tables (2005=100). The aggregate price indices for import\nand export manufactured goods are from the Bureau of Labor Statistics (2000=100). The latter data is\navailable using the Standard International Trade Classification from 1993 to 2005 and available using\nNorth American Industry Classification System from 2005 to 2010. We constructed an extended\nseries throughout the period 1993 to 2010 by setting the same index value for December 2005 in\nthose two available series.\n\n\n17\n\nMention: \"IMF Commodity Indices\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Volatility of aggregate price indices is computed from these indices.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005079:18:0", "occurrence_id": "5e0e27bc2fe022bb3d4ea5a9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figures and Tables**\n\n\nFigure 1. Volatility of Aggregate Price Indices using IMF Commodity Indices\n\n\n\n\n\n\n\nNote: The figure shows the evolution of the annualized standard deviations of Hodrick-Prescott\nfiltered price series. The aggregate price indices for all primary, non-fuel primary and energy goods\nare from IMF Primary Commodity Price Tables (2005=100). The aggregate price indices for import\nand export manufactured goods are from the Bureau of Labor Statistics (2000=100). The latter data is\navailable using the Standard International Trade Classification from 1993 to 2005 and available using\nNorth American Industry Classification System from 2005 to 2010. We constructed an extended\nseries throughout the period 1993 to 2010 by setting the same index value for December 2005 in\nthose two available series.\n\n\n17\n\nMention: \"IMF Primary Commodity Price Tables\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Aggregate indices for primary and energy goods come from these tables.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005082:12:0", "occurrence_id": "019211165401ea009df0a79b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 5. Household Vulnerability after Economic Shocks**\n\n**Panel A. Interest Rate Shock**\n1. Arbitrary Threshold 2. FM Threshold\n\n\n\n\n\n\n\n**Panel B. Exchange Rate Shock**\n1. Arbitrary Threshold 2. FM Threshold\n\n\n\n\n\n\n\n\n\n\n\n**Panel C. Unemployment Shock**\n1. Arbitrary Threshold 2. FM Threshold\n\n\n\n\n\n\n\n\n\n\n\n**Panel D. Combined and Independent Shocks**\n1. Arbitrary Threshold 2. FM Threshold\n\n\n\n\n\n\n\n\n\n\n\nSource: Croatia HBS and authors’ calculations.\n\n\n\n% of Indebted Households\n% of Households at Risk, prior to the shock\n% of Households at Risk, due to the shock\nDebt Service, % of Income, prior to the shock\n<mark>Debt Service, % of Income, due to the shock</mark>\n\n\n11\n\nMention: \"Croatia HBS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Indebtedness and risk exposure are calculated from this household survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005083:20:1", "occurrence_id": "cc6f89f38f37dc443ca46fa0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: </sup>\n\npolygons. All road lengths were calculated in meters for each individual VDC. These road lengths\nwere calculated by using the Spatial Analyst Zonal method.\n\n###### **4.2 Household Survey Data**\n\n\nThe primary economic data we use come from a repeated cross-section of the Nepal Living Standards\nSurvey (NLSS) which is a nationally-representative survey of households and communities conducted\nby the Nepal Central Bureau of Statistics with assistance from the World Bank. The first round of\nthe household survey, NLSS-I covered 3,373 households in 274 wards sampled from four geographic\nand ecological strata: Mountains, Urban Hills, Rural Hills and Terai. Between 12 and 16 households\nwere interviewed in each ward between June 1995 and May 1996.\n\nThe NLSS-II, on the other hand, was conducted between April 2003 and April 2004 and its\ncross-sectional sample was constructed using a stratified design similar to NLSS-I. It covered 73\ndistricts of Nepal (excluding the Rasuwa and Mustang districts) and enumerated 3912 households in\n326 wards. Both rounds collected detailed information on household consumption of a wide range\nof food and non-food items, data on socio-demographic composition of the interviewed households,\nlabor status of household members, health and educational status, information on various sources\n\n\n12Kappa values in this range are considered ‘excellent’.(Monserud and Leemans (1992))\n13Village Development Committee is administrative name for villages in Nepal\n\n\n19\n\nMention: \"Nepal Living Standards\nSurvey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"A repeated cross-section of households and communities is the primary data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005083:20:1", "occurrence_id": "9a0c42d90e5104bf0240d107", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: </sup>\n\npolygons. All road lengths were calculated in meters for each individual VDC. These road lengths\nwere calculated by using the Spatial Analyst Zonal method.\n\n###### **4.2 Household Survey Data**\n\n\nThe primary economic data we use come from a repeated cross-section of the Nepal Living Standards\nSurvey (NLSS) which is a nationally-representative survey of households and communities conducted\nby the Nepal Central Bureau of Statistics with assistance from the World Bank. The first round of\nthe household survey, NLSS-I covered 3,373 households in 274 wards sampled from four geographic\nand ecological strata: Mountains, Urban Hills, Rural Hills and Terai. Between 12 and 16 households\nwere interviewed in each ward between June 1995 and May 1996.\n\nThe NLSS-II, on the other hand, was conducted between April 2003 and April 2004 and its\ncross-sectional sample was constructed using a stratified design similar to NLSS-I. It covered 73\ndistricts of Nepal (excluding the Rasuwa and Mustang districts) and enumerated 3912 households in\n326 wards. Both rounds collected detailed information on household consumption of a wide range\nof food and non-food items, data on socio-demographic composition of the interviewed households,\nlabor status of household members, health and educational status, information on various sources\n\n\n12Kappa values in this range are considered ‘excellent’.(Monserud and Leemans (1992))\n13Village Development Committee is administrative name for villages in Nepal\n\n\n19\n\nMention: \"NLSS-I\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its first round covered over three thousand households in 274 wards.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005083:20:1", "occurrence_id": "2d7669475db8f30c8a2d0f90", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: </sup>\n\npolygons. All road lengths were calculated in meters for each individual VDC. These road lengths\nwere calculated by using the Spatial Analyst Zonal method.\n\n###### **4.2 Household Survey Data**\n\n\nThe primary economic data we use come from a repeated cross-section of the Nepal Living Standards\nSurvey (NLSS) which is a nationally-representative survey of households and communities conducted\nby the Nepal Central Bureau of Statistics with assistance from the World Bank. The first round of\nthe household survey, NLSS-I covered 3,373 households in 274 wards sampled from four geographic\nand ecological strata: Mountains, Urban Hills, Rural Hills and Terai. Between 12 and 16 households\nwere interviewed in each ward between June 1995 and May 1996.\n\nThe NLSS-II, on the other hand, was conducted between April 2003 and April 2004 and its\ncross-sectional sample was constructed using a stratified design similar to NLSS-I. It covered 73\ndistricts of Nepal (excluding the Rasuwa and Mustang districts) and enumerated 3912 households in\n326 wards. Both rounds collected detailed information on household consumption of a wide range\nof food and non-food items, data on socio-demographic composition of the interviewed households,\nlabor status of household members, health and educational status, information on various sources\n\n\n12Kappa values in this range are considered ‘excellent’.(Monserud and Leemans (1992))\n13Village Development Committee is administrative name for villages in Nepal\n\n\n19\n\nMention: \"NLSS-II\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The second round conducted across that year.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005084:33:1", "occurrence_id": "204a6c6e867f2c80ea72c50c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: /sup> New export opportunities for manufacturing enterprises in Kinshasa or\nthe planned export processing zone in the vicinity of Kinshasa will critically depend on less costly\nlinkages with regional markets and secure access to deep water ports, of which Pointe Noire is\ncurrently the closest.\n\n\n**5.1** **General Cross-Border Transport and Trade**\n\nWe first formulate some policy recommendations that concern DRC policy in general, and are thus\napplicable in the Western region just as they are elsewhere in the country.\n\n_Regulatory and Administrative Measures_\n\n**First and foremost, our analysis points to the importance of customs reform** . This has been\nspelt out very clearly in the recent DTIS (World Bank, 2010c). A central element of such a reform\nare the systematic implementation of single clearing and payment points for traders (“guichet\nunique”). A corollary of this is a significant reduction in the number of public or semi-public\nagencies (or even private agents) active at border posts and allowed (or at least tolerated) to collect\nfees from traders. The Presidential decree of 28 March, 2002, stating that only four agencies are\nallowed to operate at border posts, should be enforced.\n\n**An area for reform of particular interest to the Western region is the organization of transit**\n**trade.** Leakage and fraud in transit trade should be minimized. Transit could be further facilitated\nby more intensive collaboration between DRC and RC customs offices, and the modalities of\ncharging (and reimbursing) turnover tax (ICA), or the new value added tax that will replace the\nICA, should be improved. <sup>29</sup> [^29: A law for creating a value added tax system has been passed in 2010.]\n\n\n28 Somewhat soberingly, the problems identified and policy recommendations made in World Bank (2010) are very\nsimilar to those of a comparable analysis carried out two decades earlier, in the final period of the Mobutu régime\n(Word Bank, 1991)\n\nMention: \"DTIS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its analysis sets out the case for customs reform.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005085:15:3", "occurrence_id": "4d58645adaa39e5426c6edfb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  form, All coefficient statistically significant.|\n\n\nSource: Author‘s calculations based on LFS data sets from years 2003-2008.\n\n\nThe results presented in Table 5 are not surprising. At first the relative risk of transition to the\nsame state as in previous year is always much (even thousands in some cases) higher than to\nfull time/open term employment. Such high coefficients result from very low probability of\ntransition from full time/open term contract to any form of contract and also from low\nprobabilities of transitions from any other state to full time/open term contract.\n\n\nIn addition, the relative risk of transition to non-employment increases for any type contract\nother than a full time/open term one. What is most interesting is that this relative risk is the\nhighest for the self-employed. Partially it is a result of very low probability of transition from\nself-employment to full time/open term employment, but on the other hand it also means that\nif one treats the open term/full time employment as the ―ideal‖ state, engagement in selfemployment is in relative terms the ―worse‖ possible option with the highest risk to end up\nwithout any employment next year instead of reaching the goal. So it proves the earlier\nobservations drawn from the analysis of actual labor market transitions (Table 4).\n\nMention: \"LFS data sets\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Coefficients across this window are computed on these labour rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005086:15:0", "occurrence_id": "bf51b688091bd357a1fe3963", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 14\n\n\nprescribed order. The response rates in 1999, for example, vary from country to country with\n\n\nan average response rate of around 60 percent. Because EVS poses an identical set of\n\n\nquestions to individuals in various European countries, the survey provides a unique\n\n\nopportunity to empirically examine tax morale changes over time. EVS has been designed as\n\n\na wide-ranging survey, thereby reducing the danger of framing effects when compared with\n\n\nmany other surveys that focus entirely on a specific question. This avoids the possibility of\n\n\nrespondents reporting what they believe are the answers required by the researchers (Torgler\n\n\n& García-Valiñas, 2007). We will focus on the 10 Eastern European countries that have\n\n\nacceded to the European Union in 2004 or 2007. The general question to assess the level of\n\n\ntax morale in a society is:\n\n\nPlease tell me for each of the following statements whether you think it can always be\n\n\njustified, never be justified, or something in between: (...) Cheating on tax if you have the\n\n\nchance (% “never justified” = code 1 on a ten-point scale where 1=never and 10=always).\n\n\n_Table 1_ presents mean tax morale values for all countries in 1999 and 2008, based on a\n\n\nscale from 0 to 3, were 3 is the highest tax morale (tax evasion is never justifiable) while\n\n\nvalue 0 integrates values 4 to 10. In general, extracting only a single question from the EVS\n\n\nalso reduces problems of complexity inherent with the construction of an index, especially\n\n\nregarding the measurement procedure or a low correlation between the items. Even so, one\n\n\nshould recognize that there are some good reasons to use a multi-item index instead of a\n\n\nsingle question to measure tax morale. Tax morale is likely to be a multi-dimensional concept,\n\n\nwhich may require a multi-item measurement tool, as in psychometric studies. In this context,\n\n\na single-item measure like ours has some disadvantages compared to\n\nMention: \"EVS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"A single question is extracted to avoid index construction complexity.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005088:35:0", "occurrence_id": "b44a5935d1205562d435f8d9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |Norway|6.44|2.50|3.94|\n|---|---|---|---|\n|**Poland**|n.r.|<br>9.60|<br>n.r.|\n|**Portugal**|n.r.|<br>8.10|<br>n.r.|\n|**Romania**|n.a.|<br>6.40|<br>n.a.|\n|**Slovak Republic**|16.53|<br>11.10|<br>5.43|\n|**Slovenia**|10.90|<br>4.90|<br>6.00|\n|**Spain**|24.00|<br>8.30|<br>15.70|\n|**Sweden**|6.88|<br>6.10|<br>0.78|\n|**United Kingdom**|n.r.|<br>5.30|<br>n.r.|\n\n\n_Source: Eurostat: European Union-Statistics on Income and Living Conditions (EU-SILC), own calculations_\n_Note: n.a. – data not available, n.r. – data not reliable due to divergent development in time or too low value_\n_compared with the unemployment rate._\n\n**_Informal employment in Europe: share of labor force working in small firms or being self-_**\n\n**_employed (%), 2006-2007_**\n\n|Col1|Workers in firms with less<br>than 10 employees<br>(% of labor force)|Col3|Self-employers<br>(\n\nMention: \"European Union-Statistics on Income and Living Conditions\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Country figures with reliability flags are computed from this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005092:36:2", "occurrence_id": "05efbba5084d5378d902f356", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  J., and G. Sedlacek (1985). Heterogeneity, aggregation, and market wage functions: and\n\n\nempirical model of self-selection in the labor market. _Journal of Political Economy_ 93 (6):\n\n\n1077–1125.\n\n\nHeckman, J. and C. Pagés (2004). _Law and Employment: Lessons from Latin America and the_\n\n\n_Caribbean._ The University of Chicago Press: Chicago.\n\n\n[Heritage Foundation (2011). The 2011 Index of Economic Freedom, http://www.heritage.org/index/.](http://www.heritage.org/index/)\n\n\n35\n\nMention: \"2011 Index of Economic Freedom\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"A reference-list entry for this index publication.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005095:56:1", "occurrence_id": "e2169adeb55d058dcc43d959", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 65%|151.00|4,008,095.95|284,592.68|7.43|\n|**Developing**<br>**countries**<br>**average**|47.52%|9.90%|26.73%|17.82%|48.51%|37.62%|141.19|9,150,287.27|168,779.94|7.44|\n|**OIC-GCC**<br>**average**|79.41%|8.82%|8.82%|14.71%|61.76%|52.94%|155.33|3,837,600.94|138,565.15|7.43|\n|**Low-income**<br>**countries**<br>**average**|58.62%|3.45%|20.69%|6.90%|72.41%|51.72%|66.43|1,018,377.27|72,356.47|5.58|\n|**OECD**|30.00%|86.67%|53.33%|93.33%|33.33%|50.00%|274.00|56,203,661.64|2,080,697.75|24.47|\n\n\nSource: CGAP financial access report data 2010\n\n\nNote: The CGAP Financial Access Database covers questionnaires sent to 151 economies: 13 in East Asia and the Pacific, 27 in Europe and Central Asia, 20 in\nLatin America and the Caribbean, 14 in the Middle East and North Africa, 6 in South Asia, 40 in Sub- Saharan Africa, and 23 in the high-income OECD\n\nMention: \"CGAP financial access report data 2010\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Access indicators across many economies come from this report's data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005095:56:1", "occurrence_id": "9d3156dc1859f111f14dc3b0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 65%|151.00|4,008,095.95|284,592.68|7.43|\n|**Developing**<br>**countries**<br>**average**|47.52%|9.90%|26.73%|17.82%|48.51%|37.62%|141.19|9,150,287.27|168,779.94|7.44|\n|**OIC-GCC**<br>**average**|79.41%|8.82%|8.82%|14.71%|61.76%|52.94%|155.33|3,837,600.94|138,565.15|7.43|\n|**Low-income**<br>**countries**<br>**average**|58.62%|3.45%|20.69%|6.90%|72.41%|51.72%|66.43|1,018,377.27|72,356.47|5.58|\n|**OECD**|30.00%|86.67%|53.33%|93.33%|33.33%|50.00%|274.00|56,203,661.64|2,080,697.75|24.47|\n\n\nSource: CGAP financial access report data 2010\n\n\nNote: The CGAP Financial Access Database covers questionnaires sent to 151 economies: 13 in East Asia and the Pacific, 27 in Europe and Central Asia, 20 in\nLatin America and the Caribbean, 14 in the Middle East and North Africa, 6 in South Asia, 40 in Sub- Saharan Africa, and 23 in the high-income OECD\n\nMention: \"CGAP Financial Access Database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Questionnaires to 151 economies form the coverage of this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005087:36:0", "occurrence_id": "57ab45fa94e12ab0b7e06da2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **_Perceived discrimination._** In a number of countries, estimated proportion of informal\n\n\nemployees who consider themselves belonging to a discriminated group, is well above similar\n\n\nproportion among both formal employees and informal self-employed (Figure 15), once again\n\n\nsupporting the hypothesis that exclusion or poverty escape motive plays an important role in\n\n\nthe way how employees end up working without a contract. Table 7 provides evidence that\n\n\nwithin European regions the differences between informal and formal employees are\n\n\nstatistically significant (at the country level the relevant sub-samples are too small).\n\n\n\n\n\n\n\n**Figure 15 Perceived discrimination among formal and informal workers,**\n\n**by country. 2004-2009**\n_Note:_ The sample size of employees without contracts is less than 30 for IT, SE and FI; between 50 and 90\n\nfor CZ, RO, SK, HU, EE, BE, FR, DE and CH; between 100 and 745 for other countries.\n\n_Source:_ Calculation with ESS data\n\n\n**_Union experience._** Another indicator which helps to understand where the informal\n\n\nworkers are coming from is trade-union membership (current or past). Informal workers with\n\n\nunion experience have clearly been working in the formal sector some time ago (and those\n\n\nwith current membership probably plan to return). Figure 16 compares union coverage\n\n\n(current or past) among formal and informal workers. It appears that transition from formal to\n\n\ninformal sector is quite common in Nordic countries, Cyprus, Israel, Belgium, Ireland, UK,\n\n\nGermany, and Austria, where from 35% to 80% of informal employees (and, except for\n\n\nCyprus, similar fractions for informal self-employed) come from the unionized sector. In\n\n\nIsrael, Greece, Portugal, and France these proportions are lower (around 20%), but still\n\n\nsimilar to the ones found among formal workers, again suggesting that the two sectors are not\n\n\nisolated from each other.\n\n\n35\n\nMention: \"ESS data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Country samples of varying size underpin the calculated indicator.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005099:4:0", "occurrence_id": "b5ece1048b94ac052eba6716", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: growth rates, albeit significantly below pre-crisis projected growth averages (IMF, 2009a). <sup>5</sup> [^5: According to IMF’s estimates published in the World Economic Outlook Database, growth rates averaged 2.47% across Sub-Saharan Africa in\n2009, significantly below the rates enjoyed in the pre-crisis years (about 5.5%) between 2004 and 2008.] These potential\nbenefits of debt relief for borrowing countries (in terms of economic growth) suggest that most highly\nindebted poor countries within the region might indeed have been located on the wrong side of the debt\nLaffer curve, meaning that higher debt has been hurting economic growth.\n\n\nHowever, in addition to the exposure to external shocks, post-HIPC growth rates have also suffered from\nexcess growth volatility, reflecting the rather anemic nature of the growth process. The observed volatility is\nlikely to affect long-run growth and, more generally, prospects for meeting the first millennium development\ngoals of halving poverty by 2015 in the region. At the same time, external debt figures have shown a\nrenewed increasing trend in a number of countries, with a rapid deterioration of HIPCs’ debt-threshold\nindicators (World Bank, 2006). This trend has raised concerns about the risks of another sovereign debt\ncrisis in the post-HIPC across the region.\n\n\nThis paper fills an important gap in the literature by proposing a quantitative macroeconomic model that\ncaptures the links between growth, foreign aid, and external debt in the post-HIPC era. The structural model\nallows one to better understand the various channels through which foreign aid and external debt may\nimpact growth. One of the main features of the model is the production function that incorporates the\neffects of public capital, and the productivity of private capital and labor on output. Also significant is the\nadjustment of the balance of payment equation to account for excessive reliance on external resources to\nbridge financing gaps.\n\n\nThe objective of this paper is to quantify the HIPC dividends across Sub-Saharan Africa and assess the\npotential risks associated with the renewed deterioration of external debt\n\nMention: \"the World Economic Outlook Database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Regional average growth for that year comes from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005101:7:0", "occurrence_id": "54237d800c69af70e4c9aea3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: nutrition outcomes (Baird et al. 2007; Ferreira and Schady 2008; Paxon and Schady 2005).\n\n\nThere are, however, a few studies that analyze the link between income shocks and short-term\n\n\nhealth-related coping strategies. For example, Conceição et al. (2009) acknowledge that an income\n\n\nshock reduces the household‘s ability to pay for maintaining or improving health generally\n\n\ndecreases and private spending on food, medicine and health care also decreases. Cutler et al.\n\n\n(2002) find that during the economic crisis in Mexico, out-of-pocket health expenditures on\n\n\nhospitalization, doctors visit and dental care declined (3.9 to 3.1 percent of GDP) between 1994\n\n\nand 1995. The decline was more evident among families with elderly members and that led to\n\n\nhigher mortality rates in post crisis Mexico. Frankenberg et al. (1999) report a significant decline in\n\n\nthe proportion of household incomes spent on health in Indonesia. They show that public\n\n\nhealthcare use declined sharply and the percentage of children under five visiting community\n\n\nprimary health care centers fell by half.\n\n#### 3 Data\n\n\nThis paper analyzes data from five Crisis Response Surveys (CRS) conducted in Armenia,\n\n\nBulgaria, Montenegro, Romania, and Turkey. All the surveys were conducted between mid-2009\n\n\nand early-2010. These CRSs are shorter versions of regular household surveys (such as Living\n\n\nStandards Measurement Surveys) but special modules focus on income shocks and coping\n\n\nstrategies. Although the survey contains information on whether incomes rose, fell or stayed the\n\n\nsame, a key drawback of the surveys is that they have no measure of the intensity of the effect on\n\nany of the income components. <sup>6</sup> [^6: Income modules are often difficult to implement and data from those modules are often suspect because households\nare reluctant to provide accurate information.] CRS for four countries were launched in rural and urban locations,\n\n\nbut the Turkey CRS only surveyed urban households.\n\n\nHouseholds affected by income shocks are those households that report that any one of the\n\n\nfollowing applies\n\nMention: \"Crisis Response Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Five country rounds track primary health care use through the transition.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005106:9:0", "occurrence_id": "c48371fe83222b647cee6bc0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <u>4.1 Statistical source and variables</u>\n\nThe statistical source used for this analysis is the World Bank‘s World Development Indicators\n[(WDI). This database provides more than 800 development indicators, with time series for 209](http://web.worldbank.org/WBSITE/EXTERNAL/DATASTATISTICS/0,,contentMDK:20523397~menuPK:64133163~pagePK:64133150~piPK:64133175~theSitePK:239419~isCURL:Y,00.html)\n[countries and 18 country groups from 1960 to 2007. From the World Bank‘s World Development](http://web.worldbank.org/WBSITE/EXTERNAL/DATASTATISTICS/0,,contentMDK:20523404~menuPK:64133163~pagePK:64133150~piPK:64133175~theSitePK:239419~isCURL:Y,00.html)\nIndicators (WDI), we have temporal observations (T=10) for 16 Latin American countries for the\nperiod 1996–2006. We are able to form a balanced panel of data. (See descriptive statistics of the\nvariables in Table 1.)\n\n\n[INSERT TABLE 1]\n\n\nTable 2 presents a summary of the key variables used to empirically validate the combined\nstochastic-inefficiency model (and the control variables used in the second step analysis).\n\n\n[INSERT TABLE 2]\n\n\n<u>4.2 Empirical results</u>\n\n###### The indices of TEC, TP, SEC and  0 7 a  14 and the average of G are calculated using equations \n\nthese indices for the selected period (2001-2003) is presented in Table 3.\n\n\n[INSERT TABLE 3]\n\nTable 3 shows that the\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its eight hundred indicators across 209 economies are the statistical source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005114:29:0", "occurrence_id": "8da2bc57b4a224e9d2f977bc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: turning points, suggesting that the FAO data and the country reporting might not be that way off\nas the case in Mozambique and other African countries.\n\n\n\n\n\n**Figure A2.3: Mozambique Cashew Exports**\n\n\nFAO data shows almost no increase in the value of exports especially around the time of\nthe reforms in early 1990s. The partner data on the other hand show almost doubling of the\nexport values during this period. FAO data, which is used by almost all the researchers, have the\nsame production volumes (58 thousand tons) for 7 years (1999 to 2005) when the raw cashew\nexport prices fluctuated by more than 100 percent down and up. This is very unlikely. Actually\ndata from other sources such as Agricultural Development Strategy report prepared by WB in\n2005 show that the production during 2000 to 2002 were 46, 47 and 44 thousand tons\nrespectively. 2000 is a year of floods as well but we do not know what their impact on the cashew\nproduction was.\n\n\nSimilarly, export prices for processed cashews reported by FAO do not change for 4 years\n(1999 to 2002) while processed export prices in India decline by almost 50 percent. Thus, the\ndata supplied by Mozambique and then used by FAO should not be taken at its face value.\nINCAJU also supplies export and production data which is also different from all other numbers\nbut have the same problems outlined above.\n\n\n28\n\nMention: \"FAO data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Its figures and national reporting appear broadly consistent for the case.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005114:29:0", "occurrence_id": "8e6d5a434423a9a3ddc83727", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: turning points, suggesting that the FAO data and the country reporting might not be that way off\nas the case in Mozambique and other African countries.\n\n\n\n\n\n**Figure A2.3: Mozambique Cashew Exports**\n\n\nFAO data shows almost no increase in the value of exports especially around the time of\nthe reforms in early 1990s. The partner data on the other hand show almost doubling of the\nexport values during this period. FAO data, which is used by almost all the researchers, have the\nsame production volumes (58 thousand tons) for 7 years (1999 to 2005) when the raw cashew\nexport prices fluctuated by more than 100 percent down and up. This is very unlikely. Actually\ndata from other sources such as Agricultural Development Strategy report prepared by WB in\n2005 show that the production during 2000 to 2002 were 46, 47 and 44 thousand tons\nrespectively. 2000 is a year of floods as well but we do not know what their impact on the cashew\nproduction was.\n\n\nSimilarly, export prices for processed cashews reported by FAO do not change for 4 years\n(1999 to 2002) while processed export prices in India decline by almost 50 percent. Thus, the\ndata supplied by Mozambique and then used by FAO should not be taken at its face value.\nINCAJU also supplies export and production data which is also different from all other numbers\nbut have the same problems outlined above.\n\n\n28\n\nMention: \"Agricultural Development Strategy report\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its production figures for those years corroborate the alternative series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005114:29:0", "occurrence_id": "7081af006056712fee11929d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: turning points, suggesting that the FAO data and the country reporting might not be that way off\nas the case in Mozambique and other African countries.\n\n\n\n\n\n**Figure A2.3: Mozambique Cashew Exports**\n\n\nFAO data shows almost no increase in the value of exports especially around the time of\nthe reforms in early 1990s. The partner data on the other hand show almost doubling of the\nexport values during this period. FAO data, which is used by almost all the researchers, have the\nsame production volumes (58 thousand tons) for 7 years (1999 to 2005) when the raw cashew\nexport prices fluctuated by more than 100 percent down and up. This is very unlikely. Actually\ndata from other sources such as Agricultural Development Strategy report prepared by WB in\n2005 show that the production during 2000 to 2002 were 46, 47 and 44 thousand tons\nrespectively. 2000 is a year of floods as well but we do not know what their impact on the cashew\nproduction was.\n\n\nSimilarly, export prices for processed cashews reported by FAO do not change for 4 years\n(1999 to 2002) while processed export prices in India decline by almost 50 percent. Thus, the\ndata supplied by Mozambique and then used by FAO should not be taken at its face value.\nINCAJU also supplies export and production data which is also different from all other numbers\nbut have the same problems outlined above.\n\n\n28\n\nMention: \"data supplied by Mozambique\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its values, passed to the international agency, should not be taken at face value.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005133:26:1", "occurrence_id": "3f163366b8b911aa7b8987c2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Number of observations</u> <u>1296</u> <u>1296</u> <u>1296</u> <u>1263</u> <u>1263</u> <u>1263</u>\n_P-value testing CCT = UCT:_ 0.019 0.002\n_P-value testing CTF = CTM:_ 0.140 0.790\n_P-value testing CCTF = CCTM:_ 0.354 0.497\n_P-value testing CCTF = UCTF:_ 0.100 0.003\n_P-value testing CCTF = UCTM:_ 0.547 0.076\n_P-value testing CCTM = UCTF:_ 0.003 0.001\n_P-value testing CCTM = UCTM:_ 0.102 0.128\n_<u>P-value testing UCTF = UCTM:</u>_ <u>0.281</u> <u>0.190</u>\nNotes: Robust standard errors in brackets, clustered at the village level. *significant at 10%; **\nsignificant at 5%; *** significant at 1%. All regressions are restricted to children under 60\nmonths old and include child age fixed effects and child gender dummies. The treatment arms\nare abbreviated as CCT (conditional cash transfer), UCT (unconditional cash transfer), CTF\n(cash transfer given to fathers), CTM (cash transfer given to mothers), CCT-Father (conditional\ncash transfer given to fathers), CCT-Mother (conditional cash transfer given to mothers), UCTFather (unconditional cash transfer given to fathers), and UCT-Mother (unconditional cash\ntransfer given to mothers). Data source: Nahouri Cash Transfer Pilot Project (\n\nMention: \"Nahouri Cash Transfer Pilot Project\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Arm-level outcomes for conditional and unconditional transfers come from this pilot.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005098:22:2", "occurrence_id": "608c80da5a81dcdaa97d63fe", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |\n|Netherlands|32.1%|<br>37.5%|<br>45.1%|\n|Italy|31.5%|<br>36.7%|<br>45.8%|\n|**Slovak Republic**|**31.4%**|<br>**33.2%**|<br>**38.8%**|\n|Bosnia - Republika Srpska (1)|31.1%|<br>32.8%|<br>34.5%|\n|Macedonia (1)|28.5%|<br>30.9%|<br>33.2%|\n|Portugal|28.1%|<br>30.3%|<br>37.3%|\n|Spain|28.0%|<br>29.3%|<br>38.0%|\n|Norway|27.5%|<br>31.1%|<br>37.5%|\n|Japan|26.0%|<br>27.3%|<br>29.5%|\n|United States|22.6%|<br>25.7%|<br>30.1%|\n|United Kingdom|19.9%|<br>26.6%|<br>32.8%|\n|Switzerland|15.9%|<br>26.9%|<br>31.7%|\n|Ireland|7.8%|<br>14.0%|<br>27.0%|\n\n\n\n1. Values refer to 2009.\n_Source:_ OECD Tax and Benefit model.\n\n\n\n19\n\nMention: \"OECD Tax and Benefit model\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Country replacement rates for that year are computed with this model.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005112:71:0", "occurrence_id": "73e27188386f788b650cd500", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 111. The current account balance of a country reflects its savings-investment balance. The\nabove analysis explains China‘s large current account surplus. The reasons for the large and\ngrowing trade imbalance between the U.S. and China were, however, mainly the relocation of\nmany of the low value-added labor-intensive products from other East Asian economies to\nChina. The U.S. has long had a trade deficit with East Asia because of imports of labor-intensive\nproducts. This relocation concentrates the trade surplus of East Asia with the U.S. on China,\nwhich can be observed in the changing patterns of the United States‘ trade deficits with Japan\nand China from 1985 to 2009 (Figure 53).\n\n\n**Figure 53: Share of US-Japan and US-China Trade Deficits in Total US trade**\n\n**Deficit**\n\n50%\n\n\n45%\n\n\n40%\n\n\n35%\n\n\n30%\n\n\n25%\n\n\n\n20%\n\n\n15%\n\n\n10%\n\n\n5%\n\n\n0%\n\n\n\nChina\n\n\nJapan\n\n\n\nSource: United Nations Comtrade, World Integrated Trade Solution database; authors' calculations.\n\n\n112. This relocation as well as the regional integration through production networks also\ncaused China‘s trade deficit with other East Asian economies. Figure 54 illustrates the rising\nimportance of China in U.S. imports from East Asia. Related to this, Figure 29 above shows how\nChina‘s trade balance with other East Asian economies deteriorated, as China increasingly\nimported raw materials and semi-finished products from these countries. Many subcomponents\nof these commodities are produced in advanced countries and assembled in China, as a result of\nrelocation of labor-intensive production to China. Figure 55 shows the decomposition of the\nproduction of an iPhone assembled in China, which illustrates the small share of value added in\nChina versus inputs from other countries, notably Japan, Korea, Germany and the United States.\n\n\n70\n\nMention: \"World Integrated Trade Solution database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Share series over this span are computed from this portal's data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005117:24:1", "occurrence_id": "e6f81efaa1d0927b27b5ff8b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: >40</sup>\nBased on the estimated coefficient from the air passenger STRI in column 2, liberalizing aviation policies\nsuch that the index score falls from 50 to 25 is associated with a 20 percent increase in the number of\nflights. <sup>41</sup> Looking at the set of conditioning variables, attractiveness as a tourist destination, economic\nsize and income per capita all exert a positive and significant effect of flights, as expected. Overall the\nmodel fits the data very well, explaining about 86 percent of the cross-country variation in the number\nof international flights. When we instrument for aviation policies with institutional accountability, the\nmagnitude of the coefficient on policy increases substantially. The partial effect of liberalization, as\ninferred from the model in column 3, such that the index score falls from 50 to 25 would increase the\ntotal number of flights by some 42 percent.\n\n\nThe same qualitative results obtain when we look at total seat capacity rather than flights. These\nestimations are therefore not shown to conserve space but are available upon request. In both cases—\n\n\n39 For instance, Dresner et al. (2002) show that constrained access to gates may constitute a barrier to entry (and increase the\ncost of airline service for incumbents). Similarly, Brueckner (2002) uses a Cournot duopoly model to show how incumbent\nduopolists may restrict runway capacity such that no third party can enter the market. These studies strongly suggest that\nairport infrastructure matters.\n\n40 The results are robust to alternative weights with which the cross-border and commercial presence part are combined in the\nSTRI; specifically, the ALI component (i.e. BASA provisions) may assume any weight in the 60-90% band without materially\naffecting the results.\n\n41 Following the log-linear functional form, exp{(-.70)(-.25)} = 1.1912.\n\n\n23\n\nMention: \"air passenger STRI\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Its estimated coefficient translates index liberalisation into gains.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005120:12:0", "occurrence_id": "859c9f543852196e56a3cf1c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: In the sections that follow, we will use the latest version of the WVS-EVS to ad\n\ndress the two questions stated earlier: the definition and relation between bonding and\n\n\nbridging indicators; and the relation between these indicators and life satisfaction in\n\n\nthe context of world regions characterized by different histories, cultures and levels of\n\n\ndevelopment.\n\n\n[TABLE 1]\n\n##### **3 Data, model and variables**\n\n\nThe data we use include all rounds of the WVS-EVS carried out between 1980 and 2008.\n\n\nThe data set was created by combining the 1981-2004 World and European integrated\n\n\ndata file (version 2006.04.23) and the last round of the World Value Survey following the\n\n\ninstructions provided by the World Values Survey organization (www.worldvaluessurvey.org).\n\n\nThe resulting dataset includes five rounds of the World Values Surveys and three rounds\n\n\nof the European Values Surveys for a total of 98 countries, 1639 regions, 24 recorded\n\n\nyears and 355,298 observations. This is the largest dataset of its kind to date.\n\n\nOur final objective is to assess the impact of bonding and bridging attitudes on life\n\n\nsatisfaction worldwide. For this purpose we use what can be described as a standard\n\n\nmodel when working on happiness cross-country and longitudinally with the WVS-EVS.\n\n\nThe model is as follows:\n\n\n_Hi_ = _α_ + _βBi_ + _γXi_ + _ηCc_ + _δYt_ + _µi_ (1)\n\n\nwhere _H_ stands for happiness, _B_ is a vector of variables representing bonding and\n\n\nbridging attitudes, _X_ is a vector of control variables representing individual and house\n\nhold characteristics, _C_ is a vector of countries dummies, _Y_ is a vector of years dummies,\n\n\n_α_, _β_, _γ_, _η_, _δ_, are the parameters to be estimated and\n\nMention: \"WVS-EVS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The latest integrated version addresses the study's two questions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005120:12:0", "occurrence_id": "920fa15b71aa1c245c687a65", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: In the sections that follow, we will use the latest version of the WVS-EVS to ad\n\ndress the two questions stated earlier: the definition and relation between bonding and\n\n\nbridging indicators; and the relation between these indicators and life satisfaction in\n\n\nthe context of world regions characterized by different histories, cultures and levels of\n\n\ndevelopment.\n\n\n[TABLE 1]\n\n##### **3 Data, model and variables**\n\n\nThe data we use include all rounds of the WVS-EVS carried out between 1980 and 2008.\n\n\nThe data set was created by combining the 1981-2004 World and European integrated\n\n\ndata file (version 2006.04.23) and the last round of the World Value Survey following the\n\n\ninstructions provided by the World Values Survey organization (www.worldvaluessurvey.org).\n\n\nThe resulting dataset includes five rounds of the World Values Surveys and three rounds\n\n\nof the European Values Surveys for a total of 98 countries, 1639 regions, 24 recorded\n\n\nyears and 355,298 observations. This is the largest dataset of its kind to date.\n\n\nOur final objective is to assess the impact of bonding and bridging attitudes on life\n\n\nsatisfaction worldwide. For this purpose we use what can be described as a standard\n\n\nmodel when working on happiness cross-country and longitudinally with the WVS-EVS.\n\n\nThe model is as follows:\n\n\n_Hi_ = _α_ + _βBi_ + _γXi_ + _ηCc_ + _δYt_ + _µi_ (1)\n\n\nwhere _H_ stands for happiness, _B_ is a vector of variables representing bonding and\n\n\nbridging attitudes, _X_ is a vector of control variables representing individual and house\n\nhold characteristics, _C_ is a vector of countries dummies, _Y_ is a vector of years dummies,\n\n\n_α_, _β_, _γ_, _η_, _δ_, are the parameters to be estimated and\n\nMention: \"World and European integrated\n\n\ndata file\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"One of the two components combined to build the dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005120:12:0", "occurrence_id": "cb5a97404d4fbda59d293139", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: In the sections that follow, we will use the latest version of the WVS-EVS to ad\n\ndress the two questions stated earlier: the definition and relation between bonding and\n\n\nbridging indicators; and the relation between these indicators and life satisfaction in\n\n\nthe context of world regions characterized by different histories, cultures and levels of\n\n\ndevelopment.\n\n\n[TABLE 1]\n\n##### **3 Data, model and variables**\n\n\nThe data we use include all rounds of the WVS-EVS carried out between 1980 and 2008.\n\n\nThe data set was created by combining the 1981-2004 World and European integrated\n\n\ndata file (version 2006.04.23) and the last round of the World Value Survey following the\n\n\ninstructions provided by the World Values Survey organization (www.worldvaluessurvey.org).\n\n\nThe resulting dataset includes five rounds of the World Values Surveys and three rounds\n\n\nof the European Values Surveys for a total of 98 countries, 1639 regions, 24 recorded\n\n\nyears and 355,298 observations. This is the largest dataset of its kind to date.\n\n\nOur final objective is to assess the impact of bonding and bridging attitudes on life\n\n\nsatisfaction worldwide. For this purpose we use what can be described as a standard\n\n\nmodel when working on happiness cross-country and longitudinally with the WVS-EVS.\n\n\nThe model is as follows:\n\n\n_Hi_ = _α_ + _βBi_ + _γXi_ + _ηCc_ + _δYt_ + _µi_ (1)\n\n\nwhere _H_ stands for happiness, _B_ is a vector of variables representing bonding and\n\n\nbridging attitudes, _X_ is a vector of control variables representing individual and house\n\nhold characteristics, _C_ is a vector of countries dummies, _Y_ is a vector of years dummies,\n\n\n_α_, _β_, _γ_, _η_, _δ_, are the parameters to be estimated and\n\nMention: \"World Value Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its last round supplies the second component of the file.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005120:12:0", "occurrence_id": "8f0f40b99e4d5a8392a5a2cc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: In the sections that follow, we will use the latest version of the WVS-EVS to ad\n\ndress the two questions stated earlier: the definition and relation between bonding and\n\n\nbridging indicators; and the relation between these indicators and life satisfaction in\n\n\nthe context of world regions characterized by different histories, cultures and levels of\n\n\ndevelopment.\n\n\n[TABLE 1]\n\n##### **3 Data, model and variables**\n\n\nThe data we use include all rounds of the WVS-EVS carried out between 1980 and 2008.\n\n\nThe data set was created by combining the 1981-2004 World and European integrated\n\n\ndata file (version 2006.04.23) and the last round of the World Value Survey following the\n\n\ninstructions provided by the World Values Survey organization (www.worldvaluessurvey.org).\n\n\nThe resulting dataset includes five rounds of the World Values Surveys and three rounds\n\n\nof the European Values Surveys for a total of 98 countries, 1639 regions, 24 recorded\n\n\nyears and 355,298 observations. This is the largest dataset of its kind to date.\n\n\nOur final objective is to assess the impact of bonding and bridging attitudes on life\n\n\nsatisfaction worldwide. For this purpose we use what can be described as a standard\n\n\nmodel when working on happiness cross-country and longitudinally with the WVS-EVS.\n\n\nThe model is as follows:\n\n\n_Hi_ = _α_ + _βBi_ + _γXi_ + _ηCc_ + _δYt_ + _µi_ (1)\n\n\nwhere _H_ stands for happiness, _B_ is a vector of variables representing bonding and\n\n\nbridging attitudes, _X_ is a vector of control variables representing individual and house\n\nhold characteristics, _C_ is a vector of countries dummies, _Y_ is a vector of years dummies,\n\n\n_α_, _β_, _γ_, _η_, _δ_, are the parameters to be estimated and\n\nMention: \"World Values Surveys\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Five of its rounds enter the resulting pooled dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005120:12:0", "occurrence_id": "7c06030061cefefb76dd6384", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: In the sections that follow, we will use the latest version of the WVS-EVS to ad\n\ndress the two questions stated earlier: the definition and relation between bonding and\n\n\nbridging indicators; and the relation between these indicators and life satisfaction in\n\n\nthe context of world regions characterized by different histories, cultures and levels of\n\n\ndevelopment.\n\n\n[TABLE 1]\n\n##### **3 Data, model and variables**\n\n\nThe data we use include all rounds of the WVS-EVS carried out between 1980 and 2008.\n\n\nThe data set was created by combining the 1981-2004 World and European integrated\n\n\ndata file (version 2006.04.23) and the last round of the World Value Survey following the\n\n\ninstructions provided by the World Values Survey organization (www.worldvaluessurvey.org).\n\n\nThe resulting dataset includes five rounds of the World Values Surveys and three rounds\n\n\nof the European Values Surveys for a total of 98 countries, 1639 regions, 24 recorded\n\n\nyears and 355,298 observations. This is the largest dataset of its kind to date.\n\n\nOur final objective is to assess the impact of bonding and bridging attitudes on life\n\n\nsatisfaction worldwide. For this purpose we use what can be described as a standard\n\n\nmodel when working on happiness cross-country and longitudinally with the WVS-EVS.\n\n\nThe model is as follows:\n\n\n_Hi_ = _α_ + _βBi_ + _γXi_ + _ηCc_ + _δYt_ + _µi_ (1)\n\n\nwhere _H_ stands for happiness, _B_ is a vector of variables representing bonding and\n\n\nbridging attitudes, _X_ is a vector of control variables representing individual and house\n\nhold characteristics, _C_ is a vector of countries dummies, _Y_ is a vector of years dummies,\n\n\n_α_, _β_, _γ_, _η_, _δ_, are the parameters to be estimated and\n\nMention: \"European Values Surveys\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Three of its rounds enter the resulting pooled dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005135:13:0", "occurrence_id": "f3cf86791facf4bb4717696c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _V.II_ _Results: Sensitivity Analysis of poverty and inequality impact_\n\n\nAmong the modifications in the assumptions tested for robustness are the following: (a) allowing\n\n\nlayoffs in the formal sector but no wage reductions, and in the informal sector, allowing only\n\n\nwage reductions but no layoffs; (b) allowing formal sector layoffs and wage reductions but only\n\n\nwage reductions in the informal sector; (c) allowing layoffs in both the informal and formal sector\n\n\nand endogenously determining the wage reduction; (d) same as (b) but with no unemployment\n\n\nbenefits paid out; (e1) same as (b) but with a proportional tax to pay for the increased\n\n\nunemployment benefits; and (e2) same as (b) but with a proportional tax on households in\n\n\nquintile 4 and 5 to pay for the increased unemployment benefits. Based on discussions with\n\n\nvarious stakeholders in Latvia, model (b) was chosen as the preferred model.\n\n\nThe sensitivity tests concluded that the increase in poverty from 14.4 to 20.2 is very robust to\n\n\ndifferent scenarios (Figure 4) with one exception. If post-layoff incomes are not comparable to\n\n\nunemployment benefits that workers would be eligible to receive, the impact on poverty can be as\n\n\nhigh as 23 percent.\n\n\n**Figure 4: Sensitivity analysis of poverty simulations for various assumptions**\n\n\n\n\n\n\n\n\n\nSource: Authors calculations based on 2006 EU-SILC.\n\n\n\n12\n\nMention: \"2006 EU-SILC\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Poverty simulations under alternative assumptions run on this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005146:6:1", "occurrence_id": "59bb1b6e8d79969f66500e5f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  since the ENE only includes basic background characteristics for\n\n\neach individual: age, gender, marital status, education, whether or not the individual is a head of\n\n\nhousehold, and whether or not the individual is a migrant (defined as living in a state that is\n\n\ndifferent from the state where the person was born). The reason for using ENE data in this paper\n\n\nis that it is high frequency data with broad geographic coverage that allows me to identify\n\n\ninformal business owners and to track them over time. These features are essential for the impact\n\n\nanalysis performed in later sections of this paper.\n\n\n4 After 2004-IV, the ENE was changed to a new survey, the ENOE, and some of the questions used to\ndefine the variables in this paper where modified, limiting comparability across the two surveys.\n5 Bruhn (2008) includes a detailed description of how I classified individuals into different occupation groups.\n\n\n4\n\nMention: \"ENE data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its high frequency and broad geographic coverage suit the migration measure.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005150:11:0", "occurrence_id": "35838be0d49a9380f508a4c3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: MFIs is attributable to informal insurance and risk-sharing mechanisms, as 16 percent cite\n\n\ncommercial banks as their potential go-to source for business funds.\n\n\n**Figure 1: Sources of Start-up Financing, by Country (average 2009-10)**\n\n\n\n\n\n\n\n\n\n\n\n<u>Source: Gallup World Poll 2009 and 2010.</u>\n\n\nIt would be quite natural to wonder at this point as to why this distinction is important.\n\n\nAfter all, shouldn‟t the source of a loan be immaterial as long as the poor have someone to\n\n\nborrow from when the need arises? However, the literature draws a clear advantage of access to\n\n\nformal lending institutions for a number of reasons. For instance, Collins et al. (2009) observed\n\n\nthat microcredit is far more reliable, in terms of both availability and price, than one‟s informal\n\n\n10\n\nMention: \"Gallup World Poll 2009 and 2010\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Average shares of start-up financing sources by country come from these rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005152:31:2", "occurrence_id": "6c161cb8712cc8528c69dfd0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: u>**\n\n**Agriculture** **30%** **36%** **35%** **34%** **32%** **41%** **41%** **30%**\n- Central 10% 14% 17% 15% 13% 21% 19% 15%\n- Sub national 20% 22% 17% 19% 18% 19% 22% 15%\n**Irrigation** **40%** **29%** **32%** **28%** **22%** **26%** **20%** **15%**\n- Central 33% 21% 25% 22% 17% 20% 12% 10%\n- Sub national 7% 8% 6% 6% 5% 6% 8% 5%\n**<u>Subsidies</u>** **<u>30%</u>** **<u>35%</u>** **<u>34%</u>** **<u>38%</u>** **<u>46%</u>** **<u>33%</u>** **<u>39%</u>** **<u>55%</u>**\nSource: World Bank staff calculations based on data from MoF and MoA.\n\nNote:\n\n1. Agriculture Spending consists of spending on sub-sector: (1) Food crops, (2) Estate/horticulture crops, and (3) Livestock.\n2. Agricultural Spending at Central Level comprises of spending on sector: (1) Agriculture, (2) Irrigation, and (3) Agricultural\n\nMention: \"MoF and MoA\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Agriculture spending shares are calculated from these two ministries' data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005157:10:0", "occurrence_id": "31d9d4714c7764e205444049", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ernment consumption minus social security contribution and capital consumption\n\nallowances. Government investment spending is de…ned as development expendi\nture from the Ministry of Finance government spending accounts. The data on\n\ntax revenues, Tt, are also taken from the same source, and is de…ned as the sum\nof total indirect taxes, direct taxes on households, social security contributions\nand other capital transfers received. Finally Yt is the cyclical component of real\noutput extracted by a Hodrick-Prescott …lter. Following the de…nition laid out in\nStock and Watson (2002), growth below trend is classi…ed as a growth recession. <sup>14</sup>\n\n\nThe real GDP data is taken from Abeysinghe and Gulasekaran (2004).\n\nThe ordering of the variables in Xt assumes that shocks in tax revenues and\nreal output have no contemporaneous e¤ect on government spending. As argued\n\nin Blanchard and Perotti (2002), this identifying minimum-delay assumption may\n\nbe a sensible description of how government spending operates given that in the\n\nshort-run government spending may be unable to adjust to spending in response\n\nto changes in the …scal and macroeconomic conditions. On the other hand, Barro\n\nand Redlick (2011) argue that the government spending shock in a structural VAR\n\nis likely to be endogenous, as a higher GDP leads to higher taxes and therefore to\n\nmore government spending. However, Barro and Redlick (2011) use yearly data\n\nand their argument, thus, is unlikely to hold at the quarterly frequency. Due to\n\ndecision lags, contemporaneous discretionary government spending is unlikely to\n\nrespond within a quarter to any news about the economy.\n\n\nThe identi…cation of tax shocks is more problematic in the framework outlined\n\nin this paper. As noted in Blanchard and Perotti (2002) identi…cation of tax shocks\n\ndepend on purging innovations in revenues of automatic responses to output.\n\nMention: \"data on\n\ntax revenues\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Indirect and direct tax totals are taken from the same spending accounts.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005167:21:0", "occurrence_id": "77e09949516bfae3ae3e952a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Table 1. The share of persons insured by the mandatory health insurance in the total\npopulation\n\n<u>1998.</u> <u>2000.</u> <u>2001.</u> <u>2003.</u> <u>2005.</u> <u>2007.</u> <u>2008.</u> <u>2009.</u>\nShare of population\ncovered by\n\n90.5% 91.5% 95.0% 96.7% 97.49% 98.31% 98.48% 98.07%\n\nmandatory health\n<u>insurance</u>\n\nSource: HZZO\n\n\n\n20\n\nMention: \"HZZO\n\n\n\n20\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Mandatory health insurance coverage rates come from this insurer.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005168:40:2", "occurrence_id": "95a6d11769190fcd714f0454", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 0.339</u> <u>0.286</u> <u>0.318</u>\n\n**Note:** - significant at 10%, ** significant at 5%, *** significant at 1%\n\nSource: Worldwide Governance Indicators (WGI) project, Political Institutions Database DPI (2009) and Reinhart & Rogoff (2009).\n\n\n39\n\nMention: \"Worldwide Governance Indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Governance measures enter the regression alongside the institutional controls.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005168:40:2", "occurrence_id": "d334199c235d60b70afc498f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 0.339</u> <u>0.286</u> <u>0.318</u>\n\n**Note:** - significant at 10%, ** significant at 5%, *** significant at 1%\n\nSource: Worldwide Governance Indicators (WGI) project, Political Institutions Database DPI (2009) and Reinhart & Rogoff (2009).\n\n\n39\n\nMention: \"Political Institutions Database DPI\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its political variables are the companion source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005170:10:0", "occurrence_id": "da1d2e5ff6a5ae18d1b16373", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: upper) <sup>3</sup> [^3: Due to low response rates, actual or imputed rents are not included. Previous analysis showed that the\nconsumption aggregates constructed from LiTS I compare well with conventional measures based on national\naccounts and household surveys (see Zaidi et al, 2009).] . We find that across the transition region, the poorest third of the population in each\n\n\ncountry is more likely to report that their household was adversely impacted by the crisis (see\n\n\nTable 4). Similarly, household asset ownership (ownership of a car, a secondary residence or a\n\n\nbank account) is negatively correlated with crisis impact. The greater subjective impact of the\n\n\ncrisis on poorer households is consistent with the fact that they are much more likely than richer\n\nhouseholds to have experienced job losses. <sup>4</sup> [^4: The greater impact of the crisis on the poor as well as the correlation between job loss and low\nconsumption could be due to greater vulnerability of initially poor households to job loss, or to lower consumption\ncaused by the job loss experienced by the household. The data do not enable a comparison between the two\nhypotheses as only current consumption is measured in LiTS II.]\n\n\nThe subjective impact on the poorest households is greatest in Turkey, the western CIS (Belarus\n\nMoldova-Ukraine) sub-region, the new EU member states and the south Caucasus. However, in\n\n\nRussia, Central Asia and south-eastern Europe, the LiTS II data do not suggest a differential\n\n\nsubjective impact for the poorest third of households relative to richer households. Lastly,\n\nfemale-headed households <sup>5</sup> [^5: Because the LiTS data measure crisis impact at the household level, it is not possible to analyze differences\nat individual level. The lower likelihood that female-headed households were affected is consistent with the crisis\nimpact on sectors that employed higher proportions of males (for example, construction).] and the elderly (or pensioners) do not report a significantly greater\n\n\ncrisis\n\nMention: \"the LiTS II data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Its second round shows no differential subjective impact for the poorest.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005172:8:0", "occurrence_id": "43a333249b0ee71fb9fd9fc4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: commencement. The trial began in early 2008 and lasted one year, with STI testing at baseline, 4, 8, and\n\n\n12 months.\n\n\n_Participants_\n\n\nInclusion criteria consisted of males and females, aged 18-30 (and spouses ages 16 or over),\n\n\nresiding in one of 10 study villages within the Kilombero/Ulanga districts of the Ifakara Health and\n\n\nDemographic Surveillance System (IHDSS) in south-west Tanzania (Schellenberg et al., 2002). The\n\n\nvillages consisted of 8 rural villages and 2 semi-urban neighborhoods in Ifakara town, with participants\n\n\nevenly distributed across the villages. On average across the 10 villages approximately 20% of the 18-30\n\n\nold residents were enrolled in the study. There were three exclusion criteria: being pregnant at the time of\n\n\nregistration, having the intention to permanently migrate out of the IHDSS area within the next year, and\n\n\nunwillingness to participate if assigned to the control arm. Participants testing HIV-positive at baseline\n\n\nwere eligible for enrollment.\n\n\n_Interventions_\n\n\nThe intervention arm was divided into two sub-arms – a low-value cash award arm eligible for up\n\n\nto $30 over the course of the study (Tsh 10,000 or approximately $10 at each of the three 4-month testing\n\n\nintervals testing round), and a high-value CCT arm eligible for up to $60 (Tsh 20,000 or approximately\n\n\n$20 per testing round). Those amounts were determined based on focus-group discussions in neighboring\n\n\nvillages conducted before the intervention started, balancing sufficient incentive levels against concerns\n\n\nabout scalability and potential coercion. All participants were tested for STIs at baseline and then every 4\n\n\nmonths for one year. Participants in the two intervention arms were eligible to receive cash award\n\n\nincentive payments if they tested negative for curable STIs (excluding mycoplasma genitalium) at the 4,\n\n\n8, and 12-month testing rounds. STIs tested\n\nMention: \"Ifakara Health and\n\n\nDemographic Surveillance System\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Ten study villages across two districts are observed in this surveillance system.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005173:18:0", "occurrence_id": "e8616a0858d4dba2f68912d0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **5.** **VF Financial Information and Performance**\n\n\nThe discussion of VF performance begins with an examination of the financial position of Village Funds, as\n\nreported in the 2010 survey and summarized in Table 10. Information relevant to the balance sheet is given\n\nin the first two panels, and income and expenditure data near the bottom of the table. The first two columns\n\nof numbers show the median and mean amounts per village fund, _for those reporting values for these variables_ . The\n\nworking assumption is that when values are not reported, they are really zeros; thus, for instance, while the\n\nmean donation received by the 50 village funds that got funding from NGOs was 101,000 baht, this\n\nrepresents an average of just 2,000 baht per village fund when averaged over all VFs.\n\n\nBy far the most important source of capital is the initial contribution of a million baht by the government,\n\nwhich mainly occurred in 2002-2004. In the first couple of years of the VF, some Village Funds received\n\nadditional working capital from the government, which pushed the mean contribution up to 1.165 million\n\nbaht (for those Village Funds reporting). Two other sources of funding are relatively important: the first is\n\nborrowing from banks such as the BAAC. Over one in five Village Funds have borrowed from the\n\nestablished banks, and the median amount borrowed by these VFs was a million baht. The other important\n\nsource of capital is savings deposits by VF members, which averaged 161,000 baht, or about a tenth of total\n\nloanable funds.\n\n\nMost capital is used to lend to VF members, although at the time of the survey – May 2010 – about three\ntenths of all VF funds were held as deposits in the banking system. Some of this money was thus available\n\nfor loans later in the year – for instance, for emergency purposes – and some of it remained idle, in the sense\n\nof not being lent to villagers, for the\n\nMention: \"2010 survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Village fund balance sheet positions are reported in this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005174:27:1", "occurrence_id": "12b39c0a8888ebb1e2332024", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: , we estimate the beneficiary mix as if the additional money did not affect\n\n\nthe income distribution among public school students. Using data from the government and the\n\n\nWorld Bank, we estimate that 36.3% of these students were from the lowest two income\n\n\n[5 See for instance http://www.ilo.org/public/english/region/asro/bangkok/events/sis/download/paper24.pdf](http://www.ilo.org/public/english/region/asro/bangkok/events/sis/download/paper24.pdf)\n6 <u>[http://go.worldbank.org/BEQZ2K3MR0](http://go.worldbank.org/BEQZ2K3MR0)</u>\n\n\n26\n\nMention: \"the government and the\n\n\nWorld Bank\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Combined records support the estimate of low-income student shares.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005175:55:5", "occurrence_id": "609e4aeb2eaa53379f66f4e1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  689.7 47.8\n<u><mark>Stationary and educational and sporting goods</mark></u> <u><mark>294.6</mark></u> <u><mark>451.8</mark></u> 157.1 53.3\n<u><mark>Garments, other fiber products, footwear</mark></u> <u><mark>1336.2</mark></u> <u><mark>2064.9</mark></u> 728.7 54.5\n<u><mark>Leather, Fur, Feather & related products</mark></u> <u><mark>578.6</mark></u> <u><mark>906.5</mark></u> 327.9 56.7\n<u><mark>Electrics and Telecommunications</mark></u> <u><mark>1623.8</mark></u> <u><mark>3250.8</mark></u> 1627.0 100.2\n<u><mark>Furniture Manufacturing</mark></u> <u><mark>180.5</mark></u> <u><mark>364.4</mark></u> <u>183.9</u> <u>101.9</u>\nSource: Banister and Cook, Monthly Labor Review March 2011, and NBS, China Labor Statistical Yearbook 2010\n(for 2009 data). Sectors marked in yellow are sunset sectors. The NBS stopped publishing statistics on rural\nenterprises in 2002, thus decomposition of employment by subsector for the entire economy is not available\n\nMention: \"China Labor Statistical Yearbook 2010\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Sectoral labour statistics for that year come from this yearbook.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005185:14:2", "occurrence_id": "4a3dbb07914f2869292156fb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  important during a crisis, as well as for consistency with the overall frequency of the dataset. However,\nQoQ data for inflation may result in extreme outliers, which we ameliorate by taking logarithms. Dropping these\ncases would potentially bias the results, since we lose valuable information from those observations. In any case,\nthe results reported here remain robust to either artificially limiting the inflation rate in these cases to 1,000\npercent, or dropping these observations entirely.\n\n16The measure of loan demand is from the Fed’s Senior Loan Officer Survey. The ECB collects comparable data\non loan demand faced by European banks, but we use the U.S. measure since U.S. banks lend less in developing\ncountries.\n\n\n13\n\nMention: \"Senior Loan Officer Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It supplies the measure of loan demand; the European equivalent is not used.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005193:35:1", "occurrence_id": "189d109b0d33f80070654957", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Report). Mokoro\nLtd., Oxford.\n\nMokoro Ltd., 2008b. Putting Aid on Budget (Literature Review). Mokoro Ltd., Oxford.\n\nMoore, Mick, 1998. Death Without Taxes: Democracy, State Capacity, and Aid Dependence in\n\nthe Fourth World. In: White, G., Robinson, M. (Eds.), Towards a Democratic\nDevelopmental State. Oxford University Press, Oxford.\n\nNORAD, 2006. Nordic Plus: Practical Guide to Delegated Cooperation. Norwegian Agency for\n\nDevelopment Cooperation, Oslo.\n\nOECD, 2011. Aid Effectiveness 2005-10: Progress in Implementing the Paris Declaration.\n\nOECD, Paris.\n\nOECD, 2009a. Aid Effectiveness: A Progress Report on Implementing the Paris Declaration.\n\nOECD, Paris.\n\nOECD, 2009b. Managing Development Resources: The Use of Country Systems in Public\n\nFinancial Management. OECD, Paris.\n\nOECD, 2008. 2008 Survey on Monitoring the Paris Declaration: Effective Aid by 2010? OECD,\n\nParis.\n\nOECD, 2006. Peer Review of the United States. OECD Development Assistance Committee,\n\nParis.\n\nOlson, Mancur, 1982. The Rise and Decline of Nations. Yale University Press, New Haven.\n\n\n34\n\nMention: \"2008 Survey on Monitoring the Paris Declaration\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"A reference-list entry for this aid effectiveness monitoring survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005195:14:1", "occurrence_id": "dd5e486f5a043a24a6a20eed", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Figure 4 our measure is reliable and consistent with official\nhistorical accounts of the geography of civil war in Sierra Leone (Truth and Reconciliation\nCommission 2004). As indicated, the effects of the war were concentrated in the eastern\nand southern districts, where a higher percentage of households were made refugees or had\nhouses that burned down during the conflict. Over time, however, the war spread across the\ncountry, forcing people to abandon their communities (the Western peninsula was the only\narea not affected by this massive internal displacement).\n\nDuring the war, the Revolutionary United Front (RUF) successfully targeted traditional\nauthorities (Paramount Chiefs) as well as members of their households (Bellows and Miguel,\n2009; Keen, 2005; Richards, 1996). Bellows and Miguel (2009) find that traditional authority\nhouseholds are statistically more likely to have experienced violence during the war. Thus,\nwe include a control for whether respondents are a member of a ‘ruling house’.\n\n\n**NGO** **Service** **Provision**\n\n\nAlthough NGOs provide a large portion of essential goods and services in Sierra Leone, there\nis no comprehensive data of the locations of NGO and non-state actor projects. The only\nmeasure of non-state actor service provision that is available is a question on the NPS that\nasks community leaders whether an “NGO (not government, Local Council, or MP) has\ndone any projects since the end of the war (Feb 2002) that this community benefits from?”\nWe aggregate the number of NGO projects that the community leaders believe have taken\nplace to the enumeration area. Albeit imperfect, the inclusion of this measure allows us to\n\n\n13\n\nMention: \"the NPS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its single question to community leaders is the only measure of non-state provision.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005203:1:0", "occurrence_id": "859a8cb05433d56287d23627", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Policy Research Working Paper 6030\n\n#### **Abstract**\n\nThis paper analyzes the impact of two distinct shocks\nstemming from the cross-border transmission of the\n2007-2009 crisis on credit availability for small firms.\nThe paper uses data from AccessBank Azerbaijan which\nwas affected in its liquidity position during the second\nand third quarters of 2008 by delays in its refinancing.\nThe Azeri real economy was hit by the global crisis from\nthe fourth quarter of 2008 onwards with a combined\ndecline in oil prices, exports, remittances, and domestic\ndemand. Therefore, a pure supply side shock con be\n\n\n\ncontrasted with a real economy shock that hit exactly\nwhen the bank’s funding position strengthened again.\nThe paper finds that during the _funding shock_ (potential)\nborrowers are discouraged from applying for loans.\nHowever, for those applications made, the likelihood of\nloan approval is not affected. The _real economy shock,_ in\ncontrast, reduces the approval likelihood for SME loans\nin particular, while agro and micro loans are considerably\nless affected. Finally, bank relationships increase credit\navailability in good as well as in bad times.\n\n\n\nThis paper is a product of the Finance and Private Sector Development Unit, Africa Region. It is part of a larger effort by\nthe World Bank to provide open access to its research and make a contribution to development policy discussions around\nthe world. Policy Research Working Papers are also posted on the Web at http://econ.worldbank.org. The author may be\ncontacted at gberg@worldbank.org.\n\n\n_The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development_\n_issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the_\n_names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those_\n_of the\n\nMention: \"AccessBank Azerbaijan\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its records show how the crisis affected small-firm credit.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005205:26:2", "occurrence_id": "9f97c88e1ab84227b7981b03", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  non-ag. wage 0.175 0.213 0.219 0.199 0.189 0.267\n…from non-ag. self 0.058 0.087 0.096 0.062 0.055 0.137\n…from remittances 0.309 0.348 0.387 0.299 0.412 0.256\n**Community char.**\nDist. to nearest bank (km) 4.009 3.753 3.914 4.070 5.201 2.784\nAccess to extension 0.203 0.241 0.206 0.313 0.216 0.200\nAccess to vet service 0.280 0.291 0.263 0.339 0.223 0.272\nMigration share in comm. 19.948 22.437 22.231 22.974 22.492 21.984\nShortage of labor in comm. 0.352 0.443 0.373 0.583 0.489 0.352\nShare of hhs with credit 0.319 0.478 0.397 0.617 0.511 0.424\n<u>No. of obs.</u> <u>1480</u> <u>316</u> <u>201</u> <u>115</u> <u>177</u> <u>139</u>\n_Source:_ Own computation from 2005 Albania LSMS\n\n\n24\n\nMention: \"2005 Albania LSMS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Household counts by group are computed from this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005217:38:0", "occurrence_id": "4dc773b168fef50cf50ea8e5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 37\n\n\nby 14.66 percent. But in G-Cubed, we get the capital data from the value added rows\nin Input-Output table which is the measure of capital payment. Therefore, we need to\nestimate the capital return to get the estimated capital stock both for private and\ninfrastructure. Kamps (2004)’s study got a well-estimated set of private and\ninfrastructure capital stock data so that we can back up the capital return for those 22\nOECD countries and fit in regions in G-Cubed (including USA, JPN, GBR, DEU,\nEUZ, CAN, AUS, NZL and OEC). Then we calculate the mean of capital return using\nthese estimates (except Japan <sup>17</sup> [^17: Japan is an exception in advanced countries who has a very high government capital stock and\ntherefore, very low government capital return rate.] ) which is approximately equal to 0.1704 and assume\nthat this return also applies to private capital. Finally, we can use this average capital\nreturn for the other regions in G-Cubed (including CHI, IND, OAS, LAM, LDC, EEB\nand OPC) to get the infrastructure capital stock by assuming that they all have the\nsame rate of return.\n\n\nHowever, we notice that the infrastructure capital return usually diverges to the\nprivate capital return, especially in the emerging countries (see Canning and\nBennathan, 2000). Therefore, we need to adjust the stock again. Here, we assume in\nadvanced countries there’s little divergence between the private and infrastructure\ncapital return; therefore, this is no need to adjust for regions such as USA, JPN, GBR,\nDEU, EUZ, CAN, AUS, NZL and OEC. But for emerging economies we assume\nthere is more divergence between the capital returns in private and public sectors and\ncertain adjustment is needed. To do this adjustment, we need the following\nassumptions.\n\n\n- For simplification, the marginal product of government capital is in the CobbDouglas production form, where is the output\n\nMention: \"Kamps (2004)\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its estimated private and infrastructure capital stocks are adopted here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005222:9:0", "occurrence_id": "5e2968c3f82e7365fed33e2e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: capacity of national (“Public”) generation and expenditures on maintenance (and energy\n\n\nconsumption) of inverters and generators represent a large share of the income of an average\nhousehold with these appliances.\n\n\n**Figure 3: Percentage of households that have an electrical inverter or a generator at home, by**\n**socioeconomic class**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nWhich strata of the population are opting for installing auto generation devices? Private\nsolutions to counter inefficiencies in public service provision are not always cheap, and, as\nshown in the graph, there is a large gap in terms of percentage of people that counts with these\ndevices in the lower classes <sup>5</sup> (averaging less than 3% coverage) and those belonging to the\n\nupper middle (28.9%) and high (60.8%) classes.\n\n\n5 “Lowest” and “low” groups according to the ONE “socioeconomic group” variable in ENIGH 2007.\n\n\n[8]\n\nMention: \"ENIGH 2007\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its socioeconomic group variable defines the income classes used.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005223:43:3", "occurrence_id": "b792e0932980bc4c073967ec", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: >-0.176<br>(-2.08)<br>(-0.90)<br>(-1.76)<br>(-1.06)<br>0.002<br>0.001<br>0.003<br>0.002<br>(0.75)<br>(0.27)<br>(0.93)<br>(0.51)<br>-0.029<br>0.261**<br>-0.050<br>-0.043<br>(-0.31)<br>(2.06)<br>(-0.53)<br>(-0.29)<br>0.031<br>0.018<br>(0.35)<br>(0.13)<br>-0.058<br>0.059<br>(-0.62)<br>(0.40)<br>0.095**<br>0.039<br>(2.09)<br>(0.58)<br>0.020<br>(0.19)|\n|**Pseudo R-Squared**|0.07<br>0.07<br>0.11|\n\n\n\n***,**, * Significant at 1,5, and 10 percent level. T-statistics in parentheses.\nSource: Authors‘ calculations based upon World Bank Enterprise Surveys\nNote: See Table 3 for additional notes\n\n\n42\n\nMention: \"World Bank Enterprise Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The regression estimates are calculated from these firm surveys.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005187:31:0", "occurrence_id": "503d4513e22c3a222f4654b7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: percent— reduced agricultural water use would be necessary (FAO, 2009). <sup>19</sup> The\n\n\nsimulations do not address the negative impact that this is likely to have on agricultural\n\n\nproduction.\n\n\nFigure 4.1: Government (current and capital) spending under alternative scenarios\n\n(Average annual percentage share of GDP, 2004-2015)\n\n\n\n\n\n\n\n\n\n\n\n\n\nSource: Authors’ estimates based on results from simulations with MAMS.\n\n\nIn addition, targeting all MDGs together reveals the importance of cross-MDG\n\n\ncomplementarities in assessing the costs of reaching multiple MDGs. In fact, progress in\n\n\nthe coverage of drinking water and sanitation (MDG 7) exerts a positive influence on\n\n\nhealth (MDG 4), and thus allows for savings in the production of health services; in turn,\n\n\na healthier student population more easily achieves completion of educational cycles\n\n\n(MDG 2). For example, improvements in water and sanitation alone have a positive\n\n\nimpact on the under-five mortality rate; under a foreign financing scenario (not shown),\n\n\nthe MDG 4 in 2015 is 5.1 per 1000 lower than in the base scenario. Overall, the savings\n\n\nin total spending due to cross-MDG synergies is equivalent 2.4 percent of GDP.\n\n\n19 See < _http://www.fao.org/nr/water/aquastat/countries/yemen/index.stm_ >.\n\n30\n\nMention: \"MAMS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Estimates of output effects across the goals come from simulations with this model.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005198:23:0", "occurrence_id": "77a73a5bca7c16df35ddcd2d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 44 percent of non-account-holders in that country cite cost as a reason for not\nhaving a formal account. Analysis finds a significant relationship between cost\nas a self-reported barrier and an objective measure of costs.\n\n\nBut fixed fees and high costs of opening and maintaining accounts also often\nreflect lack of competition and underdeveloped physical or institutional infrastructure. These issues seem to be particularly important in Sub-Saharan Africa\nand Latin America and the Caribbean, where improvements that reduce costs\ncould potentially increase the share of adults with a formal account by up to 24\npercentage points. <sup>17</sup>\n\n\nLack of trust in banks can be a difficult barrier to overcome. This distrust can stem\nfrom cultural norms, discrimination against certain population groups, past episodes of government expropriation of banks, or economic crises and uncertainty.\nIn Europe and Central Asia 31 percent of non-account-holders cite lack of trust\nin banks as a reason for not having an account—a share almost three times that\nin other regions on average. <sup>18</sup>\n\n\nReligious reasons for not having a formal account are most commonly cited in\nthe Middle East and North Africa and South Asia. In these regions, developing\nfinancial products compatible with religious beliefs (Islamic finance) could pay\noff—potentially increasing the share of adults with a formal account by up to 10\npercentage points in the Middle East and North Africa and by up to 5 percentage\npoints in South Asia.\n\n\nGlobal Findex data suggest that indirect use of an account is most common\nin South Asia: 34 percent of adults in the region without a formal account cite\nanother family member already having one as a reason, compared with a global\naverage of 23 percent. Women tend to be more likely to be indirect users as well:\nin South Asia and the Middle East and North Africa there is a gender gap of about\n10 percentage points in citing this reason. A recent study shows that lack of account ownership (and\n\nMention: \"Global Findex data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Indirect account use across regions is measured in these data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005227:13:0", "occurrence_id": "81f67468961b0aa680b354f4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: matching of children to parents allows us to estimate both sibling and intergenerational correlations\n\n\nfor the same sample of children. The second advantage of NFHS is that education data are more\n\n\ndetailed. Instead of reporting education level in discrete intervals as is done in the case of NSS,\n\n\nNFHS collected information on years of schooling for both parents and children, which facilitates\n\n\nmore precise estimates of sibling and intergenerational correlations.\n\n\nTo define the estimation sample, we follow the literature and restrict our sample to closely\n\n\nspaced young adult siblings between the age of 16 and 27 years. The argument for estimating sibling\n\n\ncorrelations from closely spaced siblings rests on the fact that there may be important changes in the\n\n\nfamily structure as well as shocks to family life over a longer time horizon diluting the already\n\n\nconservative estimate of family background on children‘s outcome. To check the sensitivity of our\n\n\nresults, we report the estimates of sibling and intergenerational correlations for other age groups also.\n\n\nWhile the sample for estimation of the sibling correlations should ideally focus on co\n\nresident children, it may bias the estimation of intergenerational link in education between parents\n\n\nand children. If, for example, among older children, the best educated ones tend to leave household\n\n\nearlier than less educated children, it may bias intergenerational regression coefficient β downward,\n\n\nbut may not necessarily bias the estimate of intergenerational correlation. Because such exit of better\n\n\neducated children from the household would also reduce the variance in children‘s education, thus\n\n\noffsetting the decline in the intergenerational regression coefficients.\n\n\nThe problem of not observing all of the children as co-residents in the household is more\n\n\nprevalent in the case of older age cohorts, particularly for women who usually leave their natal\n\n\nhousehold upon marriage. In the case of women, if educated women delay marriage and we have\n\n\nbetter probability of observing them as co-resident children, then estimate of intergenerational\n\n\ncorrelations from our sample may be biased upward. On the other hand, if marriage timing follows\n\n\nbirth order and there is a substantial birth order effect as reported in\n\nMention: \"NFHS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its more detailed education data aid the intergenerational analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005228:18:0", "occurrence_id": "356b168fdc962c7bfb46fb04", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The evolution of unpaid family work has been strongly gender biased. The increase in the\n\n\nshare of unpaid family work in employment was due to the increase of women’s\n\n\nemployment in this type of work. The share of working women in this type of\n\n\nemployment went from 13.8 percent to 22.2 percent of employment between 2007 and\n\n\n2009, while the share of men’s employment remained around 11.8 percent. Women\n\n\nshifted from own-account work into unpaid family work and to some extent into wage\n\nemployment and to a small extent into being employers (Table 14). <sup>4</sup> [^4: Nguyễn et al (2010) suggest that the shift between own-account work and unpaid family work may be due\nto misallocation of women between the two, very similar, categories.]\n\n\n**<u>Table 14: Structure of employment for working age population (15+) in 2007 and 2009 (%)</u>**\n**2007** **2009**\n**All** **Male** **Female** **All** **Male** **Female**\nEmployer 3.2 3.8 2.7 4.8 6.3 3.2\n\n\nSelf-account worker 53.5 48.55 58.6 44.7 42.6 46.9\n\n\nUnpaid family worker 12.9 11.8 13.95 16.8 11.8 22.2\n\n\nWage worker 30.0 35.4 25.45 33.4 38.9 27.5\n\n\nMember of cooperative 0.1 0.2 0.1 0.1 0.2 0.1\n\n\nOthers 0.2 0.25 0.2 0.2 0.3 0.1\n\n\nTotal 100.0 100.0 100.0 100.0 100.0 100.0\n\n\n_Source: (LFS, 2007, 2009)_\n\n\nThis gender bias is also translated\n\nMention: \"LFS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Distribution shares across these two rounds give the gender bias.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005229:3:0", "occurrence_id": "09d43bb4d0f5365e6fdec955", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 2\n\n\n**1** **Introduction**\n\n\nInfant and under-5 mortality has declined markedly in Sub-Saharan Africa during the last decade.\n\n\nAmong the 20 Sub-Saharan African countries that have conducted Demographic and Health\n\n\nSurveys (DHS) since 2005, 18 countries showed improvements in infant survival since the\n\n\ncountry‘s previous survey, and broad improvements have taken place in under-5 mortality as\n\n\nwell. The Millennium Development Goal for under-5 mortality is a reduction of two-thirds over\n\n\nthe 25 year period 1990-2015, which implies an annual rate of decline of 4.4 percent; twelve of\n\n\nthe 20 recent DHS countries have rates of decline that exceed this rate. For infant mortality,\n\n\nKenya stands out among the success stories, with an annual rate of decline of 7.6 percent, the\n\n\nhighest among the 20 countries.\n\n\nThe gains in infant mortality are just one element of a wide set of improvements in human\n\n\nwelfare in Kenya and many countries in Sub-Saharan Africa. In one review of these changes,\n\n\nRadelet (2010) argues that five fundamental changes have driven the broader turnaround in the\n\n\n―emerging African countries‖ that he profiles: 1) more democratic and accountable governments,\n\n\n2) more sensible economic policies, 3) the end of the debt crisis and major changes in\n\n\nrelationships with the international community, 4) new technologies that are creating new\n\n\nopportunities for business and political accountability, and 5) a new generation of policymakers,\n\n\nactivists, and business leaders. Accompanying these changes has been a new push by\n\n\ngovernments and the international community to implement public health initiatives, including\n\n\nthose advanced by the Gates Foundation, the Global Fund to Fight AIDS, Tuberculosis and\n\n\nMalaria, as well as other organizations.\n\n\nMillions of dollars have been poured into these initiatives, many of which have been shown to be\n\n\neffective on a small scale in randomized controlled trials. However, given the challenges of\n\n\ngovernance and large-scale implementation in many countries, the effectiveness of such\n\n\ninitiatives\n\nMention: \"Demographic and Health\n\n\nSurveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Eighteen of twenty countries show improved infant survival in these rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005231:19:0", "occurrence_id": "638f93eac95263797ab796e2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 8: Annual service employment growth as a function of initial service employment density\n\n(logs), Indian districts 1999-2004 and Chinese prefecture-level cities (2000-2007)\n\n\nFigure 9: Growth in services employment, predicted based on NSS data (left panel) and\n\ncounterfactual based on U.S. counties (right panel)\n\n\n18\n\nMention: \"NSS data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Predicted services employment growth is computed from these data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005233:19:1", "occurrence_id": "229f4bfaceb0bb9c28056ae2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  **_k_** indexes the 4-digit industries within the 3-digit industry _j_ and\n**Ek-1** **_Ykt_**\n_Y_ refers to employment.\n\nData on trade policy are obtained from Das **(2003).** The author computes the effective\n\nrate of protection _(ERP)_ for 3-digit Indian manufacturing industries in four sub-periods:\n\n**1980-81** to **1985-86,** **1986-87** to **1990-91,** **1991-92** to **1994-95,** and **1995-96** to **1999-00.23** We\n\nalso use tariff data from Topalova and Khandelwal (2011) which are available for every year\n\nbut only from **1987** to **1999** therefore reducing our sample size. However, these data have rich\n\ncross-sectional and time-series variation **-** which the ERP data lack. We estimate all our main\n\nspecifications with both measures of trade protection. We use the FDI deregulation variable\n\nfrom Sivadasan **(2009)** which is a dummy variable indicating which 4-digit industries were\n\nFDI-deregulated in **1992.24** We combine this dummy variable with the **ASI** data to calculate\n\nthe proportion of employment in a 3-digit industry that was exposed to FDI. Hence, our\n\n**_Nj_**\nmeasure of FDI deregulation is given by _FDIjt_ _FDIktyt_ where **_k_** indexes the 4-digit\nindustries within the 3-digit industry **_j_** and _Y_ refers to **Ek=1** employment.Ykt\n\n23Das\n\nMention: \"ASI\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Combined with a deregulation dummy it yields exposed employment shares.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005242:16:0", "occurrence_id": "d8277ab25c568c29f95a65db", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: such regimes the availability of substantial resources is more likely to be channeled in ways that\n\n\ndeter rebellion (the military and rents), plausibly countering the grievances generates by natural\n\n\nresources and rebels’ viewing of such resources as a prize for taking over the state.\n\n\nFollowing the theoretical discussion we derive two hypotheses:\n\n\n_H1: Presence of natural resources increases the risk of civil war for semi-democratic and_\n\n\n_unstable regimes and states where leaders appeal to ethnicity in order to govern._\n\n\n_H2: While coup risk increases the risk of rebellion, natural resource rents allow high coup d’état_\n\n\n_risk regimes to lower the risk of civil war._\n\n\n**3. Data and research design**\n\n\nWe use two alternative definitions for civil war onset, based on widely used measures in the\n\n\nliterature: Fearon and Laitin (2003) and the Prio/Uppsala data (Gleditsch et al. 2002) for years\n\n\n1946-2003. The Fearon and Laitin definition captures the onset of large scale civil wars, coded\n\n\nusing a threshold of 1000 battle deaths over the course of conflict with a yearly average of at\n\n\nleast 100 deaths. This data is updated to 2003 using Fearon (2011). The Prio/Uppsala measure\n\n\ncaptures the onset of smaller scale conflict, earlier onsets for the large scale conflicts and flare\n\nups of violence in a dormant conflict. This definition uses a lower threshold of 25 battle deaths\n\n\nper year and we use the Lujala (2010) as a data source. The dependent variable is coded 1 for\n\n\ncountry years with a civil war onset and 0 in all other cases, and we do not eliminate the\n\n\nobservations with an ongoing civil war.\n\n\nWe attempt to measure state weakness in ways that capture the theoretical account. To\n\n\ncapture mixed regimes we use the same indicator variable for anocracy as Fearon and Laitin\n\n\n(2003) based on Polity IV scores. An\n\nMention: \"Prio/Uppsala data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its conflict onset definition over this span is one of two standard measures.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005243:13:0", "occurrence_id": "244a3983435f015df58711a8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 1997-1998, 2001, and 2008. Due to sample restrictions, among bond funds we only observe the\n\n\ndrop in assets in 2008.\n\n\nFund assets may grow because of higher returns of their investments or because of\n\ninjections to the fund by the underlying investors. In fact, the growth rate of fund ’s total\n\nassets,, can be trivially written as\n\n(1)\n\n\nwhere is the (net) return to fund at time, and is the injection to the fund\n\nexpressed as a fraction of the fund’s initial assets . While injections are not directly\n\nobservable, we can estimate them. To do so, we compute individual fund returns on a given\n\n\nmonth using data from Bloomberg and Datastream and obtain injections from the difference\n\n\nbetween the change in total net assets and individual returns. More formally,\n\n, (2)\n\nwhere is the gross rate of returns to fund at time, computed as, with being\n\nthe fund price or net asset value (NAV), adjusted by dividend payments. <sup>19</sup> [^19: A fund’s net asset value (NAV) corresponds to the total net assets ( divided by the number of shares ( ).\nThus, the ratio of NAV in two consecutive periods correspond to the ratio of the total asset values times the\ninverse ratio of total shares . The flows into the fund can also be\nexpressed as the increase (decrease) in shares times the value of the share\nReplacing this in Equation (2), we obtain that the gross returns correspond to the ratio of net asset values. The\nonly caveat to our calculation is that total net assets discount the value of a fund’s liabilities, such as the fees paid\nto the managers. However, if these fees are proportional to the assets under management they would only bias the\nlevels of the variables but cancel out when computing the returns and flows relative to initial assets.]\n\n\nThe evolution of returns and injections for the median fund is shown in Figure 5, Panels\n\n\nB\n\nMention: \"Bloomberg and Datastream\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Monthly fund returns are computed from these vendor series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005232:15:2", "occurrence_id": "c9727e93baa9ee0595473e1f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  and 2006 Household Income and Expenditure\nSurveys and the 2002 and 2008 Labor Force Surveys.\n_Note:_ The grand mean effects of the industries are calculated; additional controls include age squared, industry,\nand occupation dummies. Hours worked were not available in the Sri Lankan surveys.\n\n##### _4.4. Working Conditions_\n\n\nWorking conditions directly impact a worker’s well-being through the number of hours that\nare worked overtime, a hazardous work environment, social benefits, or workplace\n\n\n14\n\nMention: \"Household Income and Expenditure\nSurveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Two rounds paired with labour surveys give the industry effects.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005246:11:0", "occurrence_id": "6348250dce681f6e3b1fc97f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 4. Average Proportion of Own-Source Revenue to GRDP 2008 and 2010:**\n**Provinces (Provinsi), Districts (Kabupaten), and Cities (Kota)**\n\n\n_Undermining agreements with special autonomy regions._ The Government of Indonesia (GOI)\nhas entered into special arrangements with Aceh, Papua and Papua West and allows them a\ngreater share of resource revenues through the tax sharing system. The DAU offsets a large part\nof those gains by including 95% of those gains as increases in fiscal capacity for the provinces\nand 63% for cities and districts. Thus the GOI taxes back most of the gains to resource rich\nregions through the operation of the DAU formula.\n\n_Incentives for local governments to serve as employment creation agencies to the neglect of_\n_their role as service providers._ The basic allocation provides financing for public sector wages.\nThis creates incentives for padding up local rolls. Such perverse behavior in Indonesia is\ncircumvented by central controls over local recruitment and staffing. But this takes away local\nautonomy for hiring and firing and setting terms of employment of local employees. It also ties\nlocal governments to the personnel policies of the central government taking away any\nincentives they might have to experiment with new public management paradigms to improve\naccountability for servicer delivery performance e.g. through contracting out or partnership\narrangements within and beyond government agencies. In short, wages compensation creates\nan incentives and accountability regime that works against good local governance.\n\n_Inappropriate indicators of fiscal need._ Beyond basic allocation, formula based expenditure\nneeds determination as done in Indonesia has important limitations. Regional per capita income\nis used twice as a need factor—real per capita GDP adjusted for purchasing power parity in the\nformation of the Human Development Index and nominal per capita GDP more directly.\nRegional per capita income is an imperfect measure of fiscal capacity but not a very useful\nmeasure of fiscal need. The\n\nMention: \"the Human Development Index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its purchasing-power-adjusted income measure serves as the need factor.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005247:21:0", "occurrence_id": "9dc462e793ae9434337d250e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: As yet we cannot predict the future with any degree of scientific validity; we evaluate the\n\n\ncurrent of intergenerational transfers from a recent comprehensive survey.\n\n\nWith detailed and high-quality data on intergenerational transfers as well as rich\n\n\ninformation on both parents and their children, the China Health and Retirement\n\n\nLongitudinal Study (CHARLS) 2008 pilot provides a fine opportunity to achieve this\n\n\ngoal. This paper developed empirical models to explore the patterns and correlates of\n\n\nintergenerational transfers between the elderly parents and adult children in Zhejiang and\n\n\nGansu Provinces.\n\n\nContrary to the situations in most developed countries, we find that transfers are\n\n\npredominantly from children to elderly parents, and still play an important role in the\n\n\nsupport of China’s elderly. Our results reveal that older people with a larger number of\n\n\noffspring more likely to receive transfers, a result indicating the potential challenges\n\n\nfaced with dwindling numbers of children. Parental income has a mixed prediction\n\n\ndepending on the magnitude of pre-transfer income of each parent. For those among the\n\n\nbottom income group the relationship is positive, but becomes negative for the top\n\n\nincome group. Within the family, there is evidence consistent with division of\n\n\nresponsibility among children, potentially based on children’s capabilities. For example,\n\n\nhighly educated children transfer more frequently, so do married children. Although there\n\n\nis no significant difference in amount of transfers, oldest sons appear less likely to\n\n\nprovide transfers, which seems to be in contradiction with conventional wisdom.\n\n\n20\n\nMention: \"China Health and Retirement\n\n\nLongitudinal Study\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its pilot carries intergenerational transfers and parent-child information.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005248:12:0", "occurrence_id": "d4fb939eab9fccec1f3b4be0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **4.1. Identification using department and year of birth: Age-cohorts from ENV-2008**\n\n\nAccording to the ENV-2008 survey data, for more than 90 percent of the individuals who earned\n\n\nthe CEPE (completed six years of primary education), it took between 6 to 10 years. This\n\n\nsuggests the majority of the students in the primary school are in the 6 to 16 age group with the\n\n\nplausible assumption that primary education normally starts at the age of six. To identify the\n\n\npotential victims of war, we construct a young cohort including all primary school goers who\n\n\nwere exposed to the conflict between 2002 and 2006. Based on this, the individuals aged\n\n\nbetween 10 and 22 years constitute the young cohort in the ENV-2008 survey. Using ENV-2008,\n\n\nwe compare average years of education for individuals in the young cohort against an older\n\n\ncohort, aged between 23 and 32. The individuals in the old cohort are likely to be over the age of\n\n\nprimary school goers between 2002 and 2006. We use the year of birth and the department of\n\n\nbirth to identify an individual‟s exposure to war. To begin with, a straight forward difference-in\n\ndifference of average years of education is calculated based on year and department of birth.\n\n\n<u>Table 4.1.1 Means of Years of Education by Cohort and War Prevalence</u>\n<u>Years of education</u>\n<u>No War</u> <u>War</u> <u>Difference</u>\n\n7.84 9.18 -1.34\nOld Cohort (Aged 23 to 32 in 2008)\n\n<u>(0.14)</u> <u>(0.08)</u> <u>(0.18)</u>\n\n6.32 6.46 -0.14\nYoung cohort (Aged 10 to 22\n\nMention: \"ENV-2008\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Age cohorts by department and birth year are identified from this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005248:12:0", "occurrence_id": "ce4cdf628a6b9fa3ae9d0d20", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **4.1. Identification using department and year of birth: Age-cohorts from ENV-2008**\n\n\nAccording to the ENV-2008 survey data, for more than 90 percent of the individuals who earned\n\n\nthe CEPE (completed six years of primary education), it took between 6 to 10 years. This\n\n\nsuggests the majority of the students in the primary school are in the 6 to 16 age group with the\n\n\nplausible assumption that primary education normally starts at the age of six. To identify the\n\n\npotential victims of war, we construct a young cohort including all primary school goers who\n\n\nwere exposed to the conflict between 2002 and 2006. Based on this, the individuals aged\n\n\nbetween 10 and 22 years constitute the young cohort in the ENV-2008 survey. Using ENV-2008,\n\n\nwe compare average years of education for individuals in the young cohort against an older\n\n\ncohort, aged between 23 and 32. The individuals in the old cohort are likely to be over the age of\n\n\nprimary school goers between 2002 and 2006. We use the year of birth and the department of\n\n\nbirth to identify an individual‟s exposure to war. To begin with, a straight forward difference-in\n\ndifference of average years of education is calculated based on year and department of birth.\n\n\n<u>Table 4.1.1 Means of Years of Education by Cohort and War Prevalence</u>\n<u>Years of education</u>\n<u>No War</u> <u>War</u> <u>Difference</u>\n\n7.84 9.18 -1.34\nOld Cohort (Aged 23 to 32 in 2008)\n\n<u>(0.14)</u> <u>(0.08)</u> <u>(0.18)</u>\n\n6.32 6.46 -0.14\nYoung cohort (Aged 10 to 22\n\nMention: \"ENV-2008 survey data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Shares completing primary education are computed from these data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005248:12:0", "occurrence_id": "860d7375c2cfe54fcb4f6004", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **4.1. Identification using department and year of birth: Age-cohorts from ENV-2008**\n\n\nAccording to the ENV-2008 survey data, for more than 90 percent of the individuals who earned\n\n\nthe CEPE (completed six years of primary education), it took between 6 to 10 years. This\n\n\nsuggests the majority of the students in the primary school are in the 6 to 16 age group with the\n\n\nplausible assumption that primary education normally starts at the age of six. To identify the\n\n\npotential victims of war, we construct a young cohort including all primary school goers who\n\n\nwere exposed to the conflict between 2002 and 2006. Based on this, the individuals aged\n\n\nbetween 10 and 22 years constitute the young cohort in the ENV-2008 survey. Using ENV-2008,\n\n\nwe compare average years of education for individuals in the young cohort against an older\n\n\ncohort, aged between 23 and 32. The individuals in the old cohort are likely to be over the age of\n\n\nprimary school goers between 2002 and 2006. We use the year of birth and the department of\n\n\nbirth to identify an individual‟s exposure to war. To begin with, a straight forward difference-in\n\ndifference of average years of education is calculated based on year and department of birth.\n\n\n<u>Table 4.1.1 Means of Years of Education by Cohort and War Prevalence</u>\n<u>Years of education</u>\n<u>No War</u> <u>War</u> <u>Difference</u>\n\n7.84 9.18 -1.34\nOld Cohort (Aged 23 to 32 in 2008)\n\n<u>(0.14)</u> <u>(0.08)</u> <u>(0.18)</u>\n\n6.32 6.46 -0.14\nYoung cohort (Aged 10 to 22\n\nMention: \"ENV-2008 survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Young and older cohort schooling is compared using this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005251:70:1", "occurrence_id": "a52f861e94b31d5a491b8bc4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ~|Primary CDM<br>~~Seconardy CDM~~||||||\n|Primary CDM<br>~~Seconardy CDM~~|Primary CDM<br>~~Seconardy CDM~~|Primary CDM<br>~~Seconardy CDM~~|Primary CDM<br>~~Seconardy CDM~~|Primary CDM<br>~~Seconardy CDM~~||||||\n|<br>Other Offsets|<br>Other Offsets|<br>Other Offsets|<br>Other Offsets|<br>Other Offsets||||||\n|||||||||||\n|||||||||||\n|||||||||||\n|||||||||||\n|||||||||||\n|||||||||||\n|||||||||||\n|||||||||||\n\n\n2005 2006 2007 2008 2009 2010\n\n\n\nSource State and Trends of the Carbon Market 2011 (World Bank, 2011)\n\n\n\n69\n\nMention: \"State and Trends of the Carbon Market 2011\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Annual carbon market values across these years are tabulated in this report.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005255:8:0", "occurrence_id": "cca740b20d711d689fed0162", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: , then the OLS estimate of β (namely, RIF-OLS estimator) provides a consistent\nestimator of the marginal effect on the unconditional quantile of a small location shift in the\ndistribution of X, holding else constant. So, one can decompose any two distributions at\nquantiles in a similar spirit as Oaxaca-Blinder approach. It is worth noting, however, that the\nassumption of linearity used in implementing this detailed decomposition implies that the\noutcome distribution depends on the marginal distribution of covariates only through their mean\n(Rothe 2010b).\n\nThe decomposition of test score change over-time can be rewritten as <sup>3</sup> [^3: Since and, taking expectations on both sides yields .\nSo, .] :\n\n\nwhere and are approximation errors that appear because of the first order approximations\nand the assumption of linear specification. They are calculated as differences between the\naggregate explained effect and the total composition effect:, and that between\nunexplained component and the total coefficients structure effect: .\n\nAn identification problem exists in this sort of detailed decompositions as pointed out by Oaxaca\nand Ransom (1999), as the coefficients effect is not invariant to the choice of reference group;\nthis problem has no easy solution. One approach is to derive results using more than one\nalternative reference group. The other is to use one of the solutions suggested to deal with this\nproblem. We used the averaging approach proposed by Yun (2005).\n\n**3.** **Data and Summary Statistics**\n\n_3.1._ _Data_\n\nWe used the Trends in International Mathematics and Science Study (TIMSS) data for 8 <sup>th</sup> grade\nGhanaian students. TIMSS is an international assessment conducted every four years in order to\ngenerate information about trends in mathematics and science achievement in each country and\nover time. The age of tested student population is 9 to10 years (4 <sup>th</sup> grade) and 13 to 14 years (8 <s\n\nMention: \"the Trends in International Mathematics and Science Study (TIMSS)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its eighth-grade assessment data for one country form the sample.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005257:5:0", "occurrence_id": "db90e2b39dc7ca796079b838", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 4\n\n\nadjustment, European bank leverage and reliance on wholesale funding remain relatively high\ncompared to other regions where deleveraging also took place but started from lower levels. For\nmany global European banks, the L/D ratios were above 100 percent and median asset values\nstood at almost 19 times equity, compared to around 10 times for their U.S. counterparts at end2010 (figure 2). <sup>2</sup> [^2: The limited decline in loan to deposit ratios may also be due to declining deposits as a result of flight to quality, as\nexperienced by Greece, Ireland, Italy, and Spain.]\n\n\n**Figure 3. Foreign Claims of European Banks on Emerging and Developing Countries**\n\n\n\n1,600\n\n\n1,400\n\n\n1,200\n\n\n1,000\n\n\n800\n\n\n600\n\n\n400\n\n\n200\n\n\n   \n\n\n\n\nSource: Bank for International Settlements (BIS), Consolidated Banking Statistics, table 9D.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n|Col1|Col2|Col3|Col4|Col5|Col6|Col7|Col8|\n|---|---|---|---|---|---|---|---|\n|||||||||\n|||||||||\n|||||||||\n|||||||||\n|||||||||\n\n\n\n\n\n\n\n\n|Figure 4. European Consolidated Banking System<br>Leverage, 2007–11|Figure 5. European Consolidated Banking<br>System Loan to Deposit Ratios, 2007–10|\n|---|---|\n|<br>19.0<br>25.0<br>19.8<br>19.1<br>18.7<br>20.9<br>27.9<br>21.1<br>20.2<br>19.7<br>9<br>10<br>\n\nMention: \"Bank for International Settlements (BIS), Consolidated Banking Statistics\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Cross-border banking claims on developing countries come from this table.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005259:15:0", "occurrence_id": "d86d5b99323d53b84d042614", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 7: Percent of Firms with a Bank Account/Loan or Line of Credit by Size**\n\n\n**_Panel A: Bank Account (%)_** **_Panel B: Loan or Line of Credit (%)_**\n\n\n\n\n\n\n\n\n\n\n\n_Source_ : World Bank Enterprise Surveys, various years. Available at: www.Enterprise Surveys.org.\n_Note_ : Sample size is 106 countries. Africa includes 38 countries all from Sub-Saharan Africa. All country point\nestimates are weighted. All regional averages are simple averages.\n\n\nNext, we examine one-year growth rates in employment and identify ―high-growth‖\n\n\nfirms, defined as 1-year growth in employment greater or equal to 20% (OECD, 2008). On\n\n\naverage, we categorize about 15% of SME’s in both Africa and other developing countries as\n\n\nhigh-growth firms. However, we find important differences in the sources of financing used to\n\n\nfinance new investments. In Africa, 84% of investments of SME’s are financed through internal\n\n\nfunds compared with 70% in other developing economies. The share of bank financing is 8%\n\n\n(compared to an average of 11% in other countries) and the share of equity financing in Africa is\n\n\nless than 2%, as compared to 8% in other developing economies (Figure 8). This data suggest\n\n\nthat firms in Africa, with similar growth opportunities as firms in other developing countries, are\n\n\nmore dependent on internal funds and more credit constrained in accessing external formal\n\n\nfinancing.\n\n\n14\n\nMention: \"World Bank Enterprise Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Account and credit access across 106 countries come from these rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005260:25:1", "occurrence_id": "24a36d1376060de1e292e3c7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  2010, to under\n3 months thereafter due to the ramp-up in imports related to the iron ore sector.\n\nThe international iron ore market, like many commodity markets, is subject to considerable volatility.\nThis volatility risk is compounded by the fact that a small number of countries— China, India and\nSouth Korea—account for a large share of world demand for the commodity. On the face of it, The\nMines and Minerals Act, (2009), does not provide for any stabilization mechanism, such as a\nStabilization Fund (SF).\n\nSierra Leone has built a fairly solid legal and regulatory framework for public financial management <sup>28</sup> ;\nbut weaknesses remain in many areas. This includes lack of budget credibility, as evidenced by large\ncompositional variances; weaknesses in expenditure controls, payroll in particular; uneven direction of\nresources to priority areas to support high quality expenditure outcomes; and imperfect revenue\nretention, as the Consolidated fund does not receive fees collected from oil exploration licenses.\nMoving forward, the government intends to fully roll-out the IFMIS to decentralized units and develop\nperformance based budgeting, systematically clean the civil service payroll, strengthen internal audits,\nand improve procurement practices across MDAs and local councils.\n\nIn 2012, the government started to draft a revenue management bill. The bill, still at an early stage of\ndevelopment and consultation, includes provisions regarding (i) accounting of revenues from mineral\nactivities to be collected and channeled into the Consolidated Fund, (ii) the establishment of a Natural\nResource Fund (NRF) for stabilization purposes, and (iii) reporting mechanisms on revenue collection\nand use.\n\n\n26 See World Bank (2009b).\n27 See Government of Sierra Leone (2010).\n28 The key legal framework for PFM is embedded in the Government Budgeting and Accountability Act, 2005 and its\nrelated regulations, the Financial Management Regulations, 2005.\n\n\n                             - 24\n\nMention: \"the IFMIS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"A planned roll-out to decentralized units alongside performance budgeting.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005261:22:2", "occurrence_id": "4869c0da694836c972174b82", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 35\nRegion dummies Yes Yes Yes Yes\nObservations 2,128 2,128 1,085 973\n<u>Pseudo R2</u> <u>0.176</u> <u>0.176</u> <u>0.0905</u> <u>0.0907</u>\nSource: MHYS 2009-2010. Probit estimations. Marginal effects, evaluated at sample means for continuous variables, are\nshown. Standard errors in parentheses: *** p<0.01, ** p<0.05, * p<0.1. The reference groups is rural unmarried youth aged 1519 without any education (for model 1 and 2 add female)\n\n\n\nWe would expect youth who wish to migrate abroad to be unsatisfied with domestic labor market\nconditions and suffer from stronger feelings of deprivation. Interestingly, employment status in itself is\nnot a significant predictor of the wish to migrate. However, relative labor deprivation among men is\n\n\n6 Specifying the above regression as a probit with the dependent variable being defined as satisfied/dissatisfied\n(1/0) yields results that are similar in interpretation.\n\n\n21\n\nMention: \"MHYS 2009-2010\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Probit marginal effects are estimated on this household round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005264:12:0", "occurrence_id": "ed4123466097a2c1964a2c2a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: however, considerable variation in these demographic variables despite being drawn from the\n\n\nsame population of tobacco farmers in central Malawi.\n\n\nThe summary statistics reveal important features of participants’ wealth and income. When\n\n\ncompared to typical households from low-income countries, the households in the sample are\n\n\npoor. In the central Malawi region we study, however, tobacco farmers have similar poverty and\n\nincome levels to those of non-tobacco-producing households. <sup>22</sup> [^22: Based on authors’ calculations from the 2004 Malawi Integrated Household Survey (IHS), individuals in tobacco\nfarming rural households in central Malawi live on PPP$1.48/day on average, while the average for central\nMalawian rural households overall is PPP$1.51/day.] At the time of the baseline\n\n\nsurvey, the median participant’s household has zero balances in a formal bank, and the 90th\n\n\npercentile of the bank balance distribution is just 700 Malawi Kwacha (henceforth MK), or\n\napproximately US$4.67. <sup>23</sup> [^23: At the time of the study, the relevant exchange rate was roughly 150 MK per US dollar.] The survey also inventoried the household’s non-cash assets\n\n\nincluding livestock and agricultural and household durables; and participants were asked to\n\n\nestimate how much they could earn by selling those assets. When we include the self-reported\n\n\nvalue of these assets, we find that the median household held just 4,446 MK of wealth and the\n\n\n90th percentile held 25,800 MK.\n\n\nSince the baseline was conducted during the rainy season, several months would elapse\n\n\nbefore the cash crop or primary staple (maize) would be ready to be harvested in mid-April or\n\n\nearly May. The summary statistics bear this out. The median household expects virtually no\n\n\nincome between the interview date and April 2010.\n\n\n**2.3** **Stage One: Choices Under Commitment**\n\n\nFigure 3 presents the timeline of the experiment. Upon initiating the interview, the\n\n\nhousehold\n\nMention: \"the 2004 Malawi Integrated Household Survey (IHS)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its calculations give daily consumption for tobacco-farming households.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005266:37:1", "occurrence_id": "4757d78a3232ae41fc7be08f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 5) = 18.232,\n\n\n\np = 0.000\n\n\n\nχ <sup>2</sup> (5) = 1.825,\n\n\n\np = 0.446\n\n\n\nχ <sup>2</sup> (5) = 0.001,\n\n\n\np = 0.854\n\n\n\nχ <sup>2</sup> (5) = 1.613,\n\n\n\nχ <sup>2</sup> (5) = 1.198,\n\n\n\np = 0.446 p = 0.549\n\nModerate poverty headcount χ <sup>2</sup> (5) = 9.139, χ <sup>2</sup> (5) = 14.240,\n\n\n\np = 0.446\n\n\n\nχ <sup>2</sup> (5) = 14.240,\n\n\n\n<u>p = 0.010</u>\n\n\n\n<u>p = 0.001</u>\n\n\n\n_Source_ : 2005 IHDS.\n\n\n**Table A3: Results from underidentification and weak instruments tests**\n\n\n\n**Household access to** **for electricity**\n\n**<u>Tests and sample type</u>** **<u>electricity</u>** **<u>(kWh/month)</u>**\n\n<u>Underidentification test (H0: Model is underidentified based on Kleibergen-Paap LM statistics)</u>\n\n\n\n**Log household demand**\n\n\n\n**<u>Tests and sample type</u>**\n\n\n\n**Household access to**\n\n\n\n**<u>electricity</u>**\n\n\n\nFor household level outcome equation χ <sup>2</sup> (6) = 629.037,\n\n\n\nχ <sup>2</s\n\nMention: \"2005 IHDS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Test statistics for household access are computed on this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005272:15:1", "occurrence_id": "f1b5b7cfd88e0cdc438faf47", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: u>Slovenia</u> <u>1</u> <u>1</u> <u>2</u> <u>23</u>\n<u>Total</u> <u>146</u> <u>122</u> <u>250</u> <u>531</u>\nSource: Authors‘ calculation from the Investment Climate Survey Databank, World Bank.\n\n\nThe previous table 2 comprises the sectors analyzed in this research paper respectively and\nincludes secondary activity (industry) sectors.\n\n4.2 Variable Selection\n\nIn order to estimate technical efficiency with DEA, there exists considerable disagreement in\nempirical literature on the definition of outputs (in the context of multiple-outputs firms) of an\nindustrial institution. Nevertheless, according to Coelli et al. (2005), a commonly-used\nclassification of inputs involves five categories: capital (K), labour (L), energy (E), material\ninputs (M), and purchase services (S). The use of data according to these categories in\nproductivity measurement is sometimes referred to as KLEMS approach <sup>19</sup> [^19: In our sample, Capital (K) and Purchased Services (S) are not available.] .\n\nTo ensure the validity of the DEA model specification, an isotonicity test (according to Golany\nand Roll, 1989) was conducted. This isotonicity test involves the calculation of all intercorrelations between inputs and outputs for identifying whether increasing amounts of inputs lead\nto greater outputs. Finally, according to Golany and Roll (1989), the number of DMUs should be\nat least twice of the total number of input and output factors considered when applying the DEA\nmodel. In this study the number of DMUs was 146 (for Beverages sector), 531 (for Metals and\nMachinery sector), 250 (Garments sector) and 122 (for Food sector), at least ten times of\n\nMention: \"Investment Climate Survey Databank\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Firm counts by sector are calculated from this databank.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005275:30:0", "occurrence_id": "3ae972a0b774ecc43627f51a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 20%\n\n\n10%\n\n\n0%\n\n\n-10%\n\n\n-20%\n\n\n-30%\n\n\n-40%\n\n\n-50%\n\n\n-60%\n\n\n-70%\n\n\n-80%\n\n\n0%\n\n\n-5%\n\n\n-10%\n\n\n-15%\n\n\n-20%\n\n\n-25%\n\n\n-30%\n\n\n-35%\n\n\n-40%\n\n\n-45%\n\n\n-50%\n\n\n10%\n\n\n5%\n\n\n0%\n\n\n-5%\n\n\n-10%\n\n\n-15%\n\n\n-20%\n\n\n-25%\n\n\n-30%\n\n\n-35%\n\n\n\n**Figure 4**\n**Net Foreign Equity and Debt Assets**\n\n_As a Percentage of GDP_\n\n\n**Asia** **China**\n\n\n60%\n\n\n50%\n\n\n40%\n\n\n30%\n\n\n20%\n\n\n10%\n\n\n0%\n\n\n-10%\n\n\n-20%\n\n\n-30%\n\n\n-40%\n\n\n**Eastern Europe** **G-7**\n\n\n20%\n\n\n15%\n\n\n10%\n\n\n5%\n\n\n0%\n\n\n-5%\n\n\n-10%\n\n\n-15%\n\n\n-20%\n\n\n**India** **Latin America**\n\n\n10%\n\n\n5%\n\n\n0%\n\n\n-5%\n\n\n-10%\n\n\n-15%\n\n\n-20%\n\n\n-25%\n\n\n-30%\n\n\n-35%\n\n\n\n**Other Developed Countries**\n\n\n0%\n\n\n-5%\n\n\n-10%\n\n\n-15%\n\n\n-20%\n\n\n-25%\n\n\n-30%\n\n\n-35%\n\n\n-40%\n\n\n-45%\n\n\n<mark>Net Equity Position</mark> <mark>Net Debt Position</mark> <mark>Net Debt Position (Excl. Int. Reserves)</mark>\n\n\nThis figure shows the evolution of the stock of net foreign equity assets and net foreign debt assets (with and without international reserves) as a percentage of\nGDP across regions between 1990 and 2007. The data sources are the World Bank's WDI and Lane and Milesi-Ferretti (2007).\n\nMention: \"WDI\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Regional shares of output over this span come from this compilation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005277:20:0", "occurrence_id": "cca0f4b228bcef69ab5d01d8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **_Financial literacy affects insurance take-up in lower-income countries_**\n\n\nSeveral studies on microinsurance suggest that knowledge of and attitudes towards insurance\n\n\nproducts play an important role in take-up. Giné et al. (2008) find that after insufficient funds,\n\n\nlack of understanding of the product is the second most commonly cited reason for not\n\n\npurchasing rainfall insurance policies in rural India. However, they find that many fewer\n\n\nhouseholds cite lack of understanding in a survey two years later after becoming more familiar\n\n\nwith the product.\n\n\nSimilarly, Cohen and Young (2007) note that the main reason for non-renewal of\n\n\ninsurance policies issued by the company CALUMNA in Guatemala was lack of understanding.\n\n\nThey also observe that Ugandans who understand health insurance are much more willing to buy\n\n\nit, while negative attitudes, such as thinking insurance will be bring ―bad luck‖ or that it’s ―only\n\n\nfor the rich‖, tend to reduce take-up. Tran and Yun (2004), an evaluation of a microinsurance\n\n\nproject in Vietnam, find that most clients had previously been unexposed to insurance products,\n\n\nand generally lacked a thorough understanding of the concept of insurance and its benefits once\n\n\nthey had enrolled in microinsurance. Some clients were subsequently unaware that they were\n\n\nable to make claims upon events such as death or illness.\n\n\nIndeed, the FinScope surveys indicate that basic awareness of insurance is low in many\n\n\nlow-income countries in Africa. Only four percent of Ghanaians have formal insurance, and an\n\n\neven smaller percentage has informal insurance. While two-thirds of those surveyed cited\n\n\naffordability as the main reason for not purchasing insurance, more than a quarter of individuals\n\n\nalso reported reasons such as not knowing what insurance is, how it works, or how to buy it. In\n\n\nMalawi, almost 50 percent of adults ―did not know that insurance products can protect you when\n\n\nyou have a problem.‖ Many people rely instead on support from their family and community\n\nMention: \"the FinScope surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"They show low basic insurance awareness across many low-income countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005278:2:0", "occurrence_id": "1c04ff8871189ca2745ab975", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ###### **Guide to the Services Trade Restrictions Database**\n\nIngo Borchert, Batshur Gootiiz and Aaditya Mattoo*\n\n\n_JEL Classification_ : F13, F14, L80\n\n\n_Keywords_ : Services trade barriers, quantification, trade policy patterns.\n\n\n[*World Bank, 1818 H Street NW, Washington DC 20433. Contact information: I.Borchert@sussex.ac.uk,](mailto:I.Borchert@sussex.ac.uk)\n<u>[bgootiiz@worldbank.org, and amattoo@worldbank.org. The authors would like to thank Rudolf Adlung, Nora Dihel,](mailto:bgootiiz@worldbank.org)</u>\nAna M. Fernandes, Shushanik Hakobyan, Bernard Hoekman, Aart Kraay, Hildegunn Nordås, Sebastian Saez and David\nTarr and the World Bank‟s Council of Chief Economists, for many helpful suggestions. We are grateful to Thelma\nChoi, Samantha De Martino, Yinuo Geng, Anna Rakhman, Nadia Rocha and, especially, Christine Lee, who\ncontributed significantly to the creation of the database, and to Qinghua Zhao, who designed the website. This paper is\npart of a World Bank research project on trade in services supported in part by the governments of Norway, Sweden,\nand the United Kingdom through the Multidonor Trust Fund for Trade and Development, and by the UK Department\nfor International Development (DFID). The findings, interpretations, and conclusions expressed in this paper are\nentirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction\nand Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or\nthe governments they represent, or\n\nMention: \"Services Trade Restrictions Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The paper is a guide to this policy restrictions database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005289:16:0", "occurrence_id": "2a809bdf6f4db5b755fb990b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: To get an initial indication of whether the data confirm these hypotheses we can calculate the two means\nbetween different groups (e.g. the proportion observation where feed-in tariffs have been used and the one\nwhere they were not) and compare them to see if one is greater than the other, and by how much. The\nsignificance of differences between two sample means can be assessed using the t-statistic calculated as part of\nthe t-test. The t-statistic may be thought of as a scaled difference between the two means, where the absolute\ndifference between means is rescaled using and estimate of the variability of the means. Such tests will be\nperformed for each of the following hypotheses.\n\n\nThe data reported in Fig 10 shows that countries that have introduced a feed-in tariff are almost four times\nmore likely to attract private investment in renewable-based energy and about seven times more investment\nthan countries where such support mechanisms have not been introduced. Both differences in the proportion\n(and total investment) of countries who have used (or not) feed-in tariffs are highly significant at 1%\nconfidence level.\n\n\n**Fig. 10**\n**Links between Investment in Renewable-based energy and Governance**\n\n\n_a) Likelihood of PPPs in renewable-based energy           b) Investment in PPP renewable-based energy_\n_(proportion)                         (average investment,US$ million)_\n\n\n_Source:_ World Bank/PPIAF PPI Database\n_Note:_ *, **, *** indicate significance of the t test of the difference\nbetween the two pair means respectively at 10%, 5% and 1% confidence level.\n\n\n**Hypothesis 2A** The likelihood of attracting private investors to provide electricity from renewable-based\nenergy is higher in countries characterized by higher standards of economy-wide governance and a higher\ndegree of political competition.\n\n\n**Hypothesis 2B** The amount of private investment in renewable\n\nMention: \"PPI Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Investment size and risk share differences are tested using it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005294:14:1", "occurrence_id": "cc67743f391a7a762bb340e7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: SU** **Former Soviet Union** _Recipient_\n\n\n**MDE** **Middle East** (Turkey, Rest of Middle East) _Recipient_\n\n\n**CAM** **Central America** (Mexico, Central America, Carribean) _Recipient_\n\n\n\n**SAM** **South America** (Colombia, Perù, Venezuela, Rest of Andean\nPact, Argentina, Brazil, Chile, Uruguay, Rest of South\nAmerica)\n\n\n\n_Recipient_\n\n\n\n**SAS** **South Asia** (Bangladesh, India, Sri Lanka, Rest of South Asia) _Recipient_\n\n\n\n**SEA** **Southeast Asia** (Taiwan, Indonesia, Malaysia, Philippines,\nSingapore, Thailand, Vietnam)\n\n\n\n_Recipient_\n\n\n\n**CHI** **China** (China, Hong Kong) _Recipient_\n\n\n**NAF** **North Africa** (Morocco, Rest of North Africa) _Recipient_\n\n\n\n**SSA** **Sub-Saharan Africa** (Botswana, Rest of SACU, Malawi,\nMozambique, Tanzania, Zambia, Zimbabwe, Other Southern\nAfrica, Uganda, Rest of Sub-Saharan Africa)\n\n\n\n_Recipient_\n\n\n\n**ROW** **Rest of the world** _Recipient_\n\n\nSource: Authors’ modelling aggregation based on GTAP data base\n\n\n**13**\n\nMention: \"GTAP data base\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The modelling aggregation of regions is built on this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005297:18:0", "occurrence_id": "7fd9a25e0b6d58a36ed238d5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **<u>Table 3</u>** indicate that NSP improves attitudes to government figures at almost all levels, including\n\n\ndistrict and provincial governors, central government officials, the President of Afghanistan, Members\n\nof Parliament, and government judges. Magnitudes of these effects vary from between 8 percentage\n\n\npoints for Members of Parliament to 4 percentage points for the national police. There is also a\n\npositive effect of NSP on the attitudes of villagers toward NGOs and ISAF soldiers. The results for\n\n\nthe summary measure indicate that NSP improves villagers’ attitudes by 13 percent of a standard\n\ndeviation. Results for the two eastern districts, however, are completely different. There is no positive\n\n\neffect of NSP on attitudes toward any government bodies, ISAF soldiers, or NGOs, and the effect on\n\nattitudes towards the president and the police is significantly negative.\n\n\nOverall, NSP improves attitudes to government and allied entities. This provides strong support for\n\nHypothesis 2. However, the positive effects on attitudes are not observed in areas with high levels of\n\n\ninitial violence, with villages receiving NSP in the eastern districts reporting significantly lower\n\napproval of the President and of police.\n\n\n**VI.3. Hypothesis 3**\n\n\nHypothesis 3 is tested using data from the follow-up survey on male and female villagers’ security\n\nperceptions and data on security incidents from villager surveys and from ISAF data. A summary\n\n\nindex is also used here to estimate the aggregate effect of NSP on perceptions of security.\n\n\nAs reported in **<u>Table 4</u>**, NSP improves villagers’ perceptions of security. The proportion of male\n\n\nrespondents in NSP villages who report an improvement in the security situation in the past two years\n\nis 6 percentage points higher, whereas the proportion of respondents who think that security has\n\ndeteriorated is 3 percentage points lower. <sup>19</sup> Among females, the proportion of female respondents\n\nwho think that women and girls feel safer compared to two years ago is\n\nMention: \"ISAF data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Security incident counts complement villager perceptions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005301:9:1", "occurrence_id": "6350e7a6d21b5b10cfe97497", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  household (See Table A1 for a complete list of variables and their descriptive statistics).\n\n\nWe use a household wealth proxy with land ownership and livestock indicator variables. Land ownership\ntakes the value 1 if the household owns more than the average land in the survey and 0 otherwise. The\naverage land is the mean of land owned by the households that own land. Livestock indicator takes the\nvalue 1 if the household owns any livestock and 0 otherwise. Access to electricity and cell or land-line\nphones are used as measures of a proxy for access to infrastructure and communication. The Upazila\npopulation is expected to proxy for economic development.\n\n\n**_Rainfall variability_**\nWe considered two sources of rainfall data. First, historical monthly rainfall data from 1948 to 2010 was\nfrom the Meteorological Department of Bangladesh for 35 weather stations around the country. The\nweather station data represents actual measured rainfall at the specific station and time and thus, are\nmore accurate. On the other hand, substantial manipulation is needed to estimate rainfall at each\nUpazila from the weather station data. We used inverse distance weighted average of three nearest\nstations to estimate the rainfall for each Upazila for each month. However, this method does not control\nfor altitude and other geographical characteristics and Upazilas near international boundaries do not\ntake weather stations in other countries that may be closer than those in Bangladesh into account.\n\n\nThe second source of rainfall data is from the Climate Research Unit of the University of East Anglia.\nEstimated monthly rainfall for most of the world is available by the half degree resolution from 1902 to\n2009. The CRU estimation combines weather station data with other information to arrive at the\nestimates. Previous versions of the CRU data were homogenized to reduce variability and provide more\n\n\n6\n\nMention: \"CRU data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Station readings are combined with other information; earlier versions were homogenised.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005309:27:0", "occurrence_id": "86010abe3bfc36a728eff682", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **3.4 Tangible versus intangible investment in Brazil**\n\n\nFigure 10 compares tangible and intangible investment in Brazil. <sup>21</sup> [^21: Note that for reasons of comparability with tangible investment, the chart shows estimates of intangible\n_investment_, and not expenditure. CHS and similar papers also present the same comparison of business tangible and\nintangible investment.] Compared to gross fixed capital\n\n\nformation (GFCF) in Brazil, which has ranged between 15-18 percent of GDP during 2000-2008,\n\n\nintangible investment is a small share of GDP. However, GFCF includes non-business tangible\n\n\ninvestment (such as government and household investment, and a part of infrastructural investment), so it\n\nis strictly speaking comparable to our estimate of intangible investment. <sup>22</sup> [^22: Source of GFCF Data: National Accounts] So we have also presented an\n\n\nestimate of business tangible investment, estimated from IBGE's surveys on Industry, Services,\n\n\nCommerce and Construction Industry using data on acquisitions and improvements to the firms‘\n\n\npermanent assets. Our estimates indicate that the ratio of tangible to intangible investment in Brazil‘s\n\n\nfirms is close to 1, which is lower than that reported for for the US and the UK, but comparable to that\n\n\nreported for some other advanced economies like France and Germany(CHS, 2004; Marrano and Haskel,\n\n\n2006). It is, however, markedly higher than China‘s ratio, which is estimated to be 0.2 (Hulten and Hao,\n\n\n2011).\n\n\nThere is some hint that in recent years, growth in intangibles has fallen behind that in tangibles. A caveat\n\n\nto this observation is that that the particularly sharp increase in business tangible investment after 2007\n\n\nappears to be associated with a revision of IBGE's methodology for their surveys on Industry, Services,\n\n\nCommerce and Construction Industry. However, even the GFCF series – which should not have been\n\n\naffected by this methodological revision –shows a markedly higher post-2005 rate of growth than\n\n\nintangible investment.\n\nMention: \"IBGE's surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Tangible business investment is estimated from surveys across four sectors.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005314:19:0", "occurrence_id": "54e9ff0953f595e5883f3476", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Although the total number of non-production workers is about one-fifth of production workers’\n\n\nin the data and the number is stable for the entire sample period, household data from _Susenas_\n\n\nshow that the estimated share of non-production workers (ages 15-65) in the manufacturing\n\nsector in 2006 is non-trivial, about 47%. Tables 3A, 3B, and 3C summarize the results using the\n\n\nlog of paid non-production worker employment as the dependent variable for all firms, small\n\n\nfirms, and large firms, respectively.\n\n\n**Table 3A: Employment Effects of Min. Wages on Non- Prod. Workers - All Firms**\n\n\n<u>Dependent variable: Log (Number of paid non-production workers)</u>\n<u>(1)</u> <u>(2)</u> <u>(3)</u> <u>(4)</u>\nLog (Min. Wage) 0.5997*** 0.3341*** -0.0585*** -0.0542**\n(0.0402) (0.0357) (0.0211) (0.0255)\nLog (Firm Age) 0.1969***\n(0.0101)\nForeign Share 0.0012***\n(0.0003)\nExport Share 0.0007***\n(0.0001)\nGovt. Share 0.0007***\n<u>(0.0003)</u>\nFixed Effects No Province Firm Firm\nYear Effects Yes Yes Yes Yes\nObservations 233,512 233,512 233,512 160,013\nNumber of firms 32,390 32,390 32,390 30,945\n<u>R-squared</u> <u>0.0093</u> <\n\nMention: \"Susenas\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its household data give the stable share of non-production manufacturing workers.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005332:9:0", "occurrence_id": "9daf318496a8279ede614f0b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Most migrants (40 percent), regardless of their origins and destinations, migrate to accompany\nsomeone. Marriage is the second most frequent reason for migration in any direction; with the\nexception of rural to urban migration where 26.6 percent migrants move to seek work.\n\n\nTable 7 – Direction of Migration and Reasons\n\n\nDirection of Migration\nReason of Migration Urban-Urban Urban-Rural Rural - Urban Rural - Rural\nFor work only 18.8 16.3 26.7 17.5\nEducation 1.3 0.3 0.7\nMarriage 33.3 33.9 23.4 28.1\nAccompany 40.9 40.2 42.1 45.2\nOthers 5.7 9.5 7.2 9.2\nTotal 100 100 100 100\nSource: Authors’ calculations using Labor Force Survey\n\n\nMost men migrate either to work or accompany another migrant (more than one-third each).\nEmployment is the reason behind migration from rural to urban localities (53 percent) as well\nas from one rural area to another. Women mostly migrate because of marriage or to\naccompany a migrant. The percentage that migrates for work is negligible (see Tables A3 and\nA4 in the appendix).\n\n\n**Internal Migration – Labor Mobility**\n\n\nThose who migrate to work are mostly males (97.8% of migrants to work are males while 2.2%\nare females), and the distribution of these migrants by level of education is multimodal (Table\n8): the biggest fraction is composed by illiterate workers (26%) and starts decreasing gradually\nwith the level of attainment until it reaches the technical secondary level (25%), and then the\nuniversity level at 17%.\n\n\nTable 8 – Educational Attainment of Migrants (for work only)\n\n\nEducation level Percent\nIlliterate 26.1\nRead & write 13.0\nLess than Intermediate level 10.2\nGeneral Secondary 2.7\nTechnical Secondary 25.3\nAbove Intermediate level 4.4\nUniversity 17.1\n\nMention: \"Labor Force Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Migration reasons and occupational shares are calculated from this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005334:7:0", "occurrence_id": "0732e0b35b03f7c92013a670", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The findings of these serious studies are all well and good, and they suggest that\n\nfurther investments in PMGSY may well be socially profitable, but they do not estab\n\nlish as much. For what must be demonstrated is that spending yet more billions on\n\nPMGSY will yield a greater improvement in welfare than like spending on the next\n\n\nbest alternative.\n\n\nThe plan of the paper is as follows. Section 2 deals with three key shadow prices\n\n\n(SPs), namely, the social discount rate, the premium on public income and the conver\nsion factor for road-building. Section 3 describes the planned extension of the network\n\nand estimates the cost, at SPs, of a representative feeder road over an assumed 30-year\n\n\nlifecycle. Estimating the benefits is altogether much harder. In keeping with the three\n\nbroad domains of commerce, education and health, Section 4 is split into three corre\n\nsponding subsections. Various methods and sources are used, notably a survey of 30\n\nvillages in Orissa’s poor uplands (hereinafter, the Orissa Survey). Those forms of ben\nefits that have defied my attempts to measure them are indicated and briefly discussed.\n\n\nThe costs and measured benefits are compared in Section 5. If the former exceed the\n\nlatter, the difference is the smallest contribution required of all non-measured benefits\n\n\nsuch that the project be socially profitable. In other words, the difference then mea\nsures the leap of faith involved in asserting such profitability. Section 6 takes up the\n\nprogram’s external effects on the wider economy, and I argue that, on balance, these\n\n\nwill be beneficial. Alternative approaches to the measurement of benefits are briefly\n\ndiscussed in Section 7. The main conclusions concerning social profitability and the\n\n\ncomplementary policies needed to ensure it are drawn together in Section 8.\n\n#### **2 Shadow Prices**\n\n\nMarket prices depart from shadow prices for various reasons: market failure, externali\nties and, especially important in\n\nMention: \"Orissa Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"A survey of thirty villages in the poor uplands is one of two sources.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005343:36:1", "occurrence_id": "1d6319cfc081f5b11cba66b8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: : Primary School 0.117 0.0476\n(0.104) (0.0724)\nHH head education: Junior or Senior Secondary School 0.156 0.0735\n(0.116) (0.0828)\nHH head education: Higher Education 0.0653 0.175\n(0.221) (0.200)\nNumber of infants in HH (ages 0-3) -0.108** -0.0638\n(0.0505) (0.0519)\nNumber of children in HH (ages 4-14) 0.00810 0.00319\n(0.0162) (0.0168)\nNumber of adults in HH (ages 15-59) 0.0184 0.0186\n(0.0198) (0.0173)\nNumber of elderly in HH (ages 60+) -0.121* 0.0110\n(0.0630) (0.0476)\n\nObservations 360 360 360 360\n\n\nTest for joint significance: H0: 3=0, 6=0\nchi2( 2) = 6.66 6.09 8.43 6.52\n\n<u>Prob > chi2</u> <u>0.0357</u> <u>0.0476</u> <u>0.0148</u> <u>0.0383</u>\nRobust standard errors (clustered at the PSU level) provided in parentheses *** p<0.01, ** p<0.05, * p<0.1.\nS _ource data:_ Turkey Welfare Monitoring Survey (May 2009)\n\n\n\n35\n\nMention: \"Turkey Welfare Monitoring Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Regression results clustered at the primary sampling unit use this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005348:20:1", "occurrence_id": "3d3e0fcdcc34803c9b41480f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: *Additional** **Percent** **Investment,**\n**Rooms (current,** **Rooms** **Increase in** **US$ ’000**\n\n\n\n**Additional**\n**Employment,**\n**Number of Jobs**\n\n\n\n**Additional**\n\n\n\n**Percent**\n**Increase in**\n**Number of**\n\n\n\n**Investment,**\n\n\n\n**Rooms**\n\n\n\n**US$ ’000**\n\n\n\n**10/2007)**\n\n\n\n**<u>Rooms</u>**\n\nBocas del Toro 539 597 111% 58 1,389\nColon 1,297 528 41% 35 1,017\nCocle 1,443 313 22% 35 460\nChiriqui 1,746 339 19% 91 560\nDarien 84 - 0% - Herrera 330 - 0% - Los Santos 272 6 2% 0 3\nPanama 10,017 7,074 71% 1,861 26,145\nVeraguas 561 - 0% - <u>Total</u> <u>16,289</u> <u>8,857</u> <u>54%</u> <u>2,080</u> <u>29,574</u>\n_Note:_ Data on 52 new investment projects from the National Tourism Register. Investments include\ntourist accommodation projects of at least US$300,000 in the metropolitan area and US$50,000 in the rest\nof the country, excluding land value. For restaurants and clubs, the data include investments of at least\nUS$120,000 in the metropolitan area and US$20,000 in the rest of the country, excluding land value.\n_Source:_ IPAT, Sustainable Tourism Masterplan (2008:25).\n\n\n19\n\nMention: \"National Tourism Register\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Fifty-two new accommodation projects above a threshold are listed here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005349:36:2", "occurrence_id": "3ec503872696732a213e231a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  0.32 0.53 0.327\nDummies for sector of industry Yes Yes Yes Yes\n<u>Dummies for region and district</u> <u>Yes</u> <u>Yes</u> <u>Yes</u> <u>Yes</u>\n_Source_ : Ghana (GLSS5 2005/06), Rwanda (EICV 2005/06), Tanzania (ILFS 2005/06).\n\n\n\n_Notes_ : earnings are net income from HE divided by hours worked by owner.\n\n\n**<u>Mean of Variables</u>**\nTanzania Rwanda Ghana Uganda\nLog earnings per hour in int $ ppp -0.56 5.64 2.50 4.12\nMale HE owner 51.8% 54.3% 26.7%\nAge of HE owner 35.7 34.1 38.6\nAge squared and divided by 100 14.4 13.1 16.4\nNo education 16.8% 17.2% 36.7% 13.2%\nIncomplete primary 13.6% 44.3% 12.2% 42.6%\nComplete primary 61.9% 24.2% 5.1% 17.0%\nComplete lower secondary 2.5% 11.9% 38.8% 16.6%\nComplete upper secondary 4.7% 1.9% 2.5% 5.8%\nAbove primary 0.6% 0.5% 4.6% 4.8%\nPast Apprentice 8.2% 26.2% 31.4% na\nUrban 49.4% 31.1% 45.5% 34.1%\nHours worked a week 52.4 32.2 42.1 na\nHours worked a week squared and divided by 100 31.6 15.9 22.3 na\n<u>Observations</u> <u>677\n\nMention: \"GLSS5\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Regional and district earnings from household enterprises come from this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005349:36:2", "occurrence_id": "3e657cc70e92e96a0077fc75", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  0.32 0.53 0.327\nDummies for sector of industry Yes Yes Yes Yes\n<u>Dummies for region and district</u> <u>Yes</u> <u>Yes</u> <u>Yes</u> <u>Yes</u>\n_Source_ : Ghana (GLSS5 2005/06), Rwanda (EICV 2005/06), Tanzania (ILFS 2005/06).\n\n\n\n_Notes_ : earnings are net income from HE divided by hours worked by owner.\n\n\n**<u>Mean of Variables</u>**\nTanzania Rwanda Ghana Uganda\nLog earnings per hour in int $ ppp -0.56 5.64 2.50 4.12\nMale HE owner 51.8% 54.3% 26.7%\nAge of HE owner 35.7 34.1 38.6\nAge squared and divided by 100 14.4 13.1 16.4\nNo education 16.8% 17.2% 36.7% 13.2%\nIncomplete primary 13.6% 44.3% 12.2% 42.6%\nComplete primary 61.9% 24.2% 5.1% 17.0%\nComplete lower secondary 2.5% 11.9% 38.8% 16.6%\nComplete upper secondary 4.7% 1.9% 2.5% 5.8%\nAbove primary 0.6% 0.5% 4.6% 4.8%\nPast Apprentice 8.2% 26.2% 31.4% na\nUrban 49.4% 31.1% 45.5% 34.1%\nHours worked a week 52.4 32.2 42.1 na\nHours worked a week squared and divided by 100 31.6 15.9 22.3 na\n<u>Observations</u> <u>677\n\nMention: \"EICV\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The companion round for the second country.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005349:36:2", "occurrence_id": "2f72d52320622682f7ff9c96", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  0.32 0.53 0.327\nDummies for sector of industry Yes Yes Yes Yes\n<u>Dummies for region and district</u> <u>Yes</u> <u>Yes</u> <u>Yes</u> <u>Yes</u>\n_Source_ : Ghana (GLSS5 2005/06), Rwanda (EICV 2005/06), Tanzania (ILFS 2005/06).\n\n\n\n_Notes_ : earnings are net income from HE divided by hours worked by owner.\n\n\n**<u>Mean of Variables</u>**\nTanzania Rwanda Ghana Uganda\nLog earnings per hour in int $ ppp -0.56 5.64 2.50 4.12\nMale HE owner 51.8% 54.3% 26.7%\nAge of HE owner 35.7 34.1 38.6\nAge squared and divided by 100 14.4 13.1 16.4\nNo education 16.8% 17.2% 36.7% 13.2%\nIncomplete primary 13.6% 44.3% 12.2% 42.6%\nComplete primary 61.9% 24.2% 5.1% 17.0%\nComplete lower secondary 2.5% 11.9% 38.8% 16.6%\nComplete upper secondary 4.7% 1.9% 2.5% 5.8%\nAbove primary 0.6% 0.5% 4.6% 4.8%\nPast Apprentice 8.2% 26.2% 31.4% na\nUrban 49.4% 31.1% 45.5% 34.1%\nHours worked a week 52.4 32.2 42.1 na\nHours worked a week squared and divided by 100 31.6 15.9 22.3 na\n<u>Observations</u> <u>677\n\nMention: \"ILFS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The companion round for the third country.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005350:12:0", "occurrence_id": "b6ab497c272d3aaa70865801", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **2010**\n\n\n\n**Better facilitation of trade in** **2010**\n**services** **could** **generate**\n**momentum for EU income and**\n**productivity** **growth** . With\nreforms that make adopting newer\ntechnologies easier, better\nregulations, and greater mobility of\nworkers, Europe’s trade in services\ncould triple in the next decade.\nMore importantly, productivity in\nthe general services sector—which\nis about 70 percent of GDP in\nEurope—would increase.\n\n**Integration of financial services**\n**in the last two decades was**\n**crucial in facilitating financial**\n**flows** **from** **the** **EU’s** **richer,**\n**slower-growth countries to less-** _Source:_ Eurostat, World Bank staff calculations\n**developed,** **fast-growing** **new**\n**member countries.** While financial integration worldwide progressed rapidly starting in\nthe late 1990s, Europe stands out for the regional deepening of financial integration,\nespecially in the EU11. As theory would predict, in Europe capital flowed downhill, from\nricher to poorer countries. <sup>11</sup> [^11: In other parts of the world, capital flows went uphill – from poorer countries like China to richer ones like the United\nStates. This pattern, though puzzling, is well-established.] Between 2004 and 2008, the average annual capital flow from\nadvanced to emerging economies was more than 10 percent of GDP, compared to about 4\npercent in Latin America and the Caribbean.\n\n**Furthermore, the composition of the capital flowing into EU11 was different from**\n**that of capital flowing into emerging countries in Asia or Latin America.** First, the\namount of foreign direct investment (FDI) in emerging Europe was higher than in other\nemerging\n\nMention: \"Eurostat\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Migration flows between richer and faster-growing member states come from this office.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005353:7:0", "occurrence_id": "ab34257768e254b6faaa5350", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: CLICROP was run at a 1°×1° resolution (i.e., 111 square kilometer grids in a country\nmeasuring 945,000 square kilometers). Separate models were developed for the 9 major crops of\nTanzania (i.e., cassava, groundnuts, maize, millet, potatoes, sorghum, soybeans, sweet potatoes\nand wheat). Predicted yields for each sub-national region are calculated as the sum of overlaid\ngridded results weighted by geographic area. CLICROP was calibrated to information on soil\nparameters from the FAO Soils Database (e.g., field capacity, wilting point and saturated\nhydraulic conductivity) (FAO-UNESCO, 2005). Regional information on crop growing seasons\nand planting dates were provided by Sacks et al. (forthcoming). Finally, crop locations were\nbased on You et al. (2006) for the year 2000, and crop parameters were drawn from Allen et al.\n(1998) and Doorenbos and Kassam (1979). Overall, the process is detailed with more than 13\nmillion potential simulation sets. <sup>3</sup> [^3: The product of approximately 8500 grids times nine crops times 43 crop years times four climate scenarios.]\n\n\n_Crop modeling results_\n\n\nAlthough CLICROP analysis was conducted for nine crops, we focus on the results for maize for\npurposes of exposition. Maize is the principal food crop in Tanzania, representing 35 and 45\npercent of calories consumed by poor urban and rural households, respectively (Pauw and\nThurlow, 2010). Tanzanian farmers allocate about one-third of their crop land to growing maize,\nmainly without the use of irrigation. Deviations in dry-land maize yields therefore provide a\nfirst-cut indicator of food availability in Tanzania (Thornton et al. 2009). In addition, the\nimplications of climate outcomes for maize are likely to be similar for sorghum and millet,\nmaking the coverage of caloric sources and land use, particularly for more vulnerable\npopulations, even broader.\n\n\nTable 2 summarizes the deviations in mean maize\n\nMention: \"the FAO Soils Database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Soil parameters calibrate the crop model.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005354:38:1", "occurrence_id": "db982897993b8899686b9984", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  almost 80,000 villages in Thailand\n\n\nto start village banks that would lend to local households. The total amount of funding to each\n\n\nvillage was the same regardless of the size of the village, so village size provides a plausibly\n\n\nexogenous source of variation in the amount of credit increase per household. The funds were\n\n\nnot directly targeted at women, but women were among the program beneficiaries. Kaboski and\n\n\nTownsend‟s (2007) analysis samples came from the Townsend Thai project, which gathered\n\n\npanel data on rural and semi-urban households and businesses from 64 villages in four Thai\n\n\nprovinces from 1997 to 2007.\n\n\nThe behavioral model is based on the standard buffer stock model of savings behavior under\n\n\nincome uncertainty (e.g., Aiyagiri (1994) and Deaton (1991)) with the additional feature that\n\n\nhouseholds have an investment option in each period. The household‟s problem is to maximize\n\n\n34\n\nMention: \"Townsend Thai project\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its panel across sixty-four villages underlies the earlier analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005355:13:0", "occurrence_id": "8a3ce85d1f9f0ee433846148", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |El Salvador|National Development<br>Plan 2010-2014|X|Col4|\n|---|---|---|---|\n|Mexico|Vision 2030|X|X|\n|Panama|National Concertation|X||\n|Peru|National<br>Competitiveness Plan<br>2003-2010<br>National<br>Competitiveness<br>Agenda<br>|<br> <br>      X<br> <br> <br>      X<br>||\n|Uruguay|Industrial<br>Development<br>Strategy (2008)<br> <br>Sectoral<br>Industrial<br>Plans (2011)|<br> <br>      X<br> <br> <br>      X|<br> <br>       X<br> <br> <br>       X|\n\n\n*Discontinued by the current government\n\n\n\nSource: The authors based on Official Data\n\n\nTable 2 also shows that some of the strategies combine with promotion of new activities, which\nwould fall into the Hausmann, Rodrik and Sabel industrial policies “in the large”. However,\nmany of these more ambitious initiatives are “in the large ‘ _light_ ’ ”, i.e., they are not grand bets\ndistant from existing comparative advantages or learning capabilities. As an illustration, Table 3\nlists the initiatives in Colombia for productive transformation that aim to build capacities\nleading to new world class sectors/activities in that country. In Chile, the 2007 innovation\nstrategy’s push for new activities was focused on cluster development of eight carefully\nselected sectors, but most are based on activities close to country’s abundant natural\nresources <sup>14</sup> .\n\n\n14 The current government, which entered power in 2010, suspended the strategy launched by the opposition.\nThe National Innovation\n\nMention: \"Official Data\n\n\nTable 2\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Government strategy records, some discontinued by the current administration.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005363:29:4", "occurrence_id": "58c1117242f4d290fe7d3ef5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .011</mark> <mark>0.000</mark>\n\n<mark>(0.006)</mark> <mark>(0.002)</mark> <mark>(0.001)</mark>\n\n<mark>Contributions to the decline in FGT(0)</mark>\n<mark>Growth</mark> <mark>90%</mark> <mark>84%</mark> <mark>103%</mark>\n\n<u><mark>Redistribution</mark></u> <mark>10%</mark> <mark>16%</mark> <mark>0%</mark>\n\n\n<mark>Source: Own estimations based on Peru's ENAHO 2005 -2009, Thailand's SES 2000 -2009, and</mark>\n<mark>Bangladesh's HIES 2000 -2010.</mark>\n\n<mark>Standard errors for a two-sided 95% confidence interval shown in parentheses.</mark>\n\n\n28\n\nMention: \"ENAHO\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The Peruvian household panel behind the estimated shares.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005363:29:4", "occurrence_id": "c55c47c21af058ab6e450715", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .011</mark> <mark>0.000</mark>\n\n<mark>(0.006)</mark> <mark>(0.002)</mark> <mark>(0.001)</mark>\n\n<mark>Contributions to the decline in FGT(0)</mark>\n<mark>Growth</mark> <mark>90%</mark> <mark>84%</mark> <mark>103%</mark>\n\n<u><mark>Redistribution</mark></u> <mark>10%</mark> <mark>16%</mark> <mark>0%</mark>\n\n\n<mark>Source: Own estimations based on Peru's ENAHO 2005 -2009, Thailand's SES 2000 -2009, and</mark>\n<mark>Bangladesh's HIES 2000 -2010.</mark>\n\n<mark>Standard errors for a two-sided 95% confidence interval shown in parentheses.</mark>\n\n\n28\n\nMention: \"SES\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The Thai household panel behind the estimated shares.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005363:29:4", "occurrence_id": "829b8a57cbdb051a5cb6bfaa", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .011</mark> <mark>0.000</mark>\n\n<mark>(0.006)</mark> <mark>(0.002)</mark> <mark>(0.001)</mark>\n\n<mark>Contributions to the decline in FGT(0)</mark>\n<mark>Growth</mark> <mark>90%</mark> <mark>84%</mark> <mark>103%</mark>\n\n<u><mark>Redistribution</mark></u> <mark>10%</mark> <mark>16%</mark> <mark>0%</mark>\n\n\n<mark>Source: Own estimations based on Peru's ENAHO 2005 -2009, Thailand's SES 2000 -2009, and</mark>\n<mark>Bangladesh's HIES 2000 -2010.</mark>\n\n<mark>Standard errors for a two-sided 95% confidence interval shown in parentheses.</mark>\n\n\n28\n\nMention: \"HIES\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The Bangladeshi household panel behind the estimated shares.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005371:5:1", "occurrence_id": "78fcf45902a5ce99fa362797", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: . <sup>4</sup> Finally, on key water, sanitation, and health indicators the study sample meets or exceeds the\n\n\nnational level statistics. Ninety-six percent of households have access to an improved water source, 68%\n\n\nimproved sanitation, 81% a place for handwashing with soap and water available, and just 14% of children\n\n\nunder two are stunted at baseline.\n\n\nVietnam was selected for the HWWS project due to its engagement in handwashing promotion under the\n\n\nPublic Private Partnership for Handwashing (PPPHW), a global handwashing initiative established in 2001.\n\n\nHandwashing promotion activities began in 2006 under the name of the Handwashing Initiative (HWI) with\n\n\nvarious partners including the Vietnam Ministry of Health, Women‟s Union and WSP. The HWWS\n\n\nproject sought to improve and scale-up these efforts through capacity building of local organizations, such as\n\n\nthe Vietnam Women‟s Union, training them in behavior change approaches, and providing technical support\n\n\nfor development of behavior change communication campaign materials. WSP-supported HWWS activities\n\n\ntargeted at caregivers and schoolchildren have been underway since 2007 in 7 provinces. These activities had\n\n\na target of reaching over 2 million mothers and children with the goal of improving the handwashing\n\n\n3 MDG 4 seeks to reduce infant and child mortality by two-thirds\n4 Vietnam Population and Housing Census, 2009\n\n4\n\nMention: \"Vietnam Population and Housing Census, 2009\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"A footnote listing this enumeration as a source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005372:37:0", "occurrence_id": "13d5afeae7ebe5bd5d7ed28c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: particles. In particular, PM10 has an impact on the respiratory diseases. The most\nwidely known adverse health impact of PM10 is premature mortality. Long-term\nexposure to PM10 can impact mortality and morbidity levels. Levels of particulate\nmatter are often much higher in developing countries as compared to those in developed\ncountries. Ostro (1994) coefficients are used to calculate PM10 health effects.\n\nSince PM10 causes premature death, one of the implications of high PM10 levels in the\ncountry would be a decrease in the available labor force. With a large population the\neffect on mortality rates and on the labor force are calculated using estimates from the\nJakarta study (Ostro, 1994). The health damages calculated outside of the CGE model\nproduced central, upper and lower estimates for the coefficients of change in mortality.\nAll three figures have been used to calculate a range of results in the CGE simulations.\n\nAs for the PM10 emissions, the projections from the CGE results have been used to\ncalculate the PM10 concentrations for India. The base year concentration level of\n97.58  g/m <sup>3</sup> is the average concentration level of the pollutant across all cities in India\n(calculated using Central Pollution Control Board of India data). The concept of\nuniform rollback was used to calculate the concentrations for the subsequent years.\nUniform rollback states that the percent change in pollutant emissions on an annual\nbasis will equal to the percent change in pollutant concentrations on an annual basis.\nTherefore, using the base year average for PM10 for the year 2010, projections can be\nmade for PM10 concentrations using the percent change in PM10 emissions from the\nCGE model.\n\nThe dose response coefficients are from the Ostro study on Jakarta (Ostro, 1994). Such\nan epidemiological study has not been carried out for India Jakarta is the next best study\nas its data provide more plausible health estimates than data from industrialized\nnations. For the\n\nMention: \"Central Pollution Control Board of India data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Average pollutant concentrations across cities are calculated from these records.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005376:6:0", "occurrence_id": "5123c959e88de31fa8021a10", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: neighboring communities (Chomitz et al. 2006; Kaimowitz and Angelsen 1998). This is a core\n\n\nissue for the Global Tiger Initiative (GTI), which must allocate its resources effectively while\n\n\nconfronting a short time horizon (to prevent tiger subspecies extinctions), a complex, constantly\n\nchanging spatial distribution of potential conservation benefits and costs, and the risk of rapid,\n\n\nirreversible losses in areas where conservation is weak. Locally-tailored GTI solutions will\n\n\nundoubtedly include REDD+ conservation payments; targeted monitoring and enforcement of\n\n\nconservation regulations where threats are urgent; and enlistment of critical local and national\n\n\ninstitutions.\n\n\nTo support cost-effective solutions, our research project is developing a spatially\n\nreferenced empirical model that can identify critical habitat areas for potential GTI programs.\n\n\nIdentification of such areas is not a static problem, because tropical forest clearing constantly\n\n\nevolves in response to changing economic and environmental conditions. This paper describes\n\n\nthe first stage of the project, which has developed two supporting resources:\n\n\n(1) A frequently-updated, high-resolution database for tracking threats to remaining\ntiger habitat in 11 tropical forest countries: Bangladesh, Bhutan, Cambodia, India,\nIndonesia, Lao PDR, Malaysia, Myanmar, Nepal, Thailand and Vietnam.\n\n\n(2) Country-specific spatial econometric models that identify critical drivers of\nforest habitat destruction.\n\n\nThe remainder of the paper is organized as follows. Section 2 develops a model of forest\n\n\nclearing that highlights economic determinants. Section 3 introduces the most critical input to\n\n\nour analysis: FORMA (Forest Monitoring for Action), a new high-resolution database that\n\n\npermits near-real-time assessment of forest habitat conditions in the tropical tiger range\n\n\ncountries. In Section 3, we develop a spatial formatting protocol for our database that is based\n\n\non critical minimum tiger habitat size. We use this protocol to integrate the FORMA data with\n\n\nspatially-referenced information on remaining forest habitat, currently-protected areas, and\n\n\n5\n\nMention: \"FORMA\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"A new high-resolution database permitting near-real-time deforestation monitoring.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005382:11:0", "occurrence_id": "edc5fa1f19dc8505008f8a62", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: likely may have a global component, as countries tend to “specialize” in the same form of\n\n\nsovereign theft at the same time. In the next section we explore a number of such potential\n\n\nfactors more systematically.\n\n\n**5. Correlates of Sovereign Theft Episodes**\n\n\nIn this section we investigate in more detail a set of potential correlates of sovereign theft\n\n\nepisodes that have been discussed in the literature on defaults and expropriations. Our starting\n\n\npoint is the core empirical specification in Kraay and Nehru (2006), who investigate the\n\n\ncorrelates of \"debt distress\", defined as episodes of debt servicing difficulties marked by\n\n\nexceptional financing in the form of recourse to the Paris Club or the IMF, as well as arrears\n\n\naccumulation. In a large sample of developing countries they find that debt distress is more\n\n\nfrequent in countries with high levels of external debt, with weak policy performance, and in\n\n\ncountries experiencing adverse macroeconomic shocks. One immediate difference however is\n\n\nthat Kraay and Nehru (2006) study debt servicing difficulties vis-a-vis both private and official\n\n\ncreditors, while the dataset we study in this paper covers only defaults against private\n\n\ncreditors. <sup>5</sup> [^5: Rescheduling of debts owed to official creditors (via the Paris Club in the case of bilateral creditors, and the\nHeavily-Indebted Poor Countries (HIPC) initiative in the case of multilateral creditors) may to some extent reflect a\nvoluntary decision on the part of creditors to provide a resource transfer to debtors, and so are different from\ndefaults against private creditors. Nevertheless, as we shall see the correlates of defaults against private creditors\nare broadly similar to those identified in Kraay and Nehru (2006).]\n\n\nWe begin by considering how expropriation and sovereign default events are related to the\n\n\nstock of external debt owed to private creditors and the stock of FDI in a country. Data on\n\n\nthese are taken from the Sovereign Wealth of Nations dataset by Lane and Milesi-Ferretti\n\n\n(2007). We also measure policy performance using the World Bank's Country Policy and\n\n\nInstitutional\n\nMention: \"the Sovereign Wealth of Nations dataset\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"External liabilities and inward investment stocks are taken from this dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005383:39:0", "occurrence_id": "492204e2c85fa3e790ca1a7c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Strukova, E. (2004). _Opportunity Cost Of Deforestation In The Brazilian Amazon: Aggregated_\n_Estimate Per Ton Of Avoided Carbon Emission._ Working Paper. Washington DC\n\n\nStrukova E. (2010) Cost-Benefit Analysis of Development and Conservation Alternatives in the\nSundarbans of India. Prepared for: Climate Change Adaptation, Biodiversity Conservation &\nSocio-Economic Development of the Sundarbans Area of West Bengal World Bank Technical\nAssistance\n\n\nSummers, L.and Zeckhauser, R. (2008). _Policy Making for Posterity_ . Journal of Risk and\nUncertainty 37(2), 115-140.\n\n\nTERI (2001). _Review of Past and On-Going Work on Urban Air Quality in India._ New Delhi.\nTata Energy Research Institute. TERI Project Report N 2001 EE41.\n\n\nThenkabail, P.S., Biradar, C.M., Noojipady, P., Dheeravath, V., Li, Y.J., Velpuri, M.,Reddy,\nG.P.O, Cai, X.L. and Krishna. M.G.(2008) A Global Irrigated Area Map (GIAM). Using Remote\nSensing at the End of the Last Millennium. International Journal of Remote Sensing.\n\n\n<mark>UN Economic Commission for Europe: Tajikistan Environmental Performance review, 2004</mark>\n<u><mark>[http://www.unece.org/env/epr/studies/Tajikistan/welcome.htm](http://www.unece.org/env/epr/studies/Tajikistan/welcome.htm)</mark></u>\n\n\nViscusi, W.K. and Aldi, J.E. (2002). _The Value of a Statistical Life: A critical\n\nMention: \"Global Irrigated Area Map\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Remote sensing at the millennium end maps irrigated area worldwide.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005385:7:1", "occurrence_id": "8cfeca130f5f8eff3e708a99", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .4 trillion|37million|21 million|1.5 trillion|15 million|37 million|1 trillion|4 trillion|\n\n\nSource: World Bank (2010), Development Research Center for the State Council (2010), and “NDRC\n\nreveals details of stimulus package”, China Daily, November 27, 2008.\n\n\nBy mid-2009, the macroeconomic stimulus and other interventions had succeeded in boosting\n\ndemand for migrant labor. By early 2010, scarcity reemerged in China’s labor market, and wages\n\nstarted to climb once again (Giles and others 2012). In 2010, China’s exports recovered to\n\nprecrisis levels, and real GDP grew 10.5 percent, which slowed down to 9.2 percent in 2011 and\n\nis now projected to slow down further to around 8 percent 2012 due to the slowdown in the\n\nglobal economy.\n\n\nThe success of China’s fiscal stimulus spurred domestic demand and supported the global\n\neconomic recovery. Strong economic growth in the world’s second-largest economy stimulated\n\n\n1 The latest IMF data shows a sharp increase in general government gross debt as percent of GDP to 33.5 percent in\n2010 before starting to decline to about 26 percent in 2011, though remaining well above its average level of about\n17.5 percent in 2008-9.\n\n\n6\n\nMention: \"IMF data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its figures show a sharp rise in general government gross debt.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005392:34:2", "occurrence_id": "c58066655d4b054ae29eb2f3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: >** **<u>82,334</u>** **<u>0.14</u>** **<u>0.11</u>** **<u>0.13</u>** **<u>0.10</u>**\n<u>Notes: Descriptive statistics taken from National Sample Statistics.</u>\n\nMention: \"National Sample Statistics\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Descriptive statistics are taken from this national sample source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005393:24:0", "occurrence_id": "da0e098574cd1e6a4bcb762c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 1 presents the scatter plot between each of a set of measures in the Database at the\n\n\ncountry level and the stage of development measured by GDP per capita along with the R\n\nsquared from the corresponding regression among the plotted pair of variables. Appendix 3\n\n\npresents analogous scatter plots between each of a set of measures in the Database and either the\n\n\neconomic size of the economy measured by GDP or the importance of the export sector\n\n\nmeasured by the ratio of exports to GDP. The Database measures as well as the GDP per capita,\n\n\nGDP, and the ratio of exports to GDP values are taken as averages over the period 2006-2008. <sup>37</sup>\n\n\n**Figure 1: Correlations of Selected Database Measures with GDP per Capita**\n\n**(averages over 2006-2008)**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n37 GDP per capita, GDP, and total exports in current USD are taken from the World Development Indicators (WDI) of the World\nBank.\n\n\n23\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Output and exports in current dollars come from this compilation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005406:19:0", "occurrence_id": "aba80df1c1ae08371532633f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: - _Factor-shift effect:_ Post-regulation production technologies may be more or less\n\nlabor intensive, as more /less labor is required per dollar of output.\n\n\nMorgenstern, Pizer and Shih (2002) empirically estimate a model incorporating these\ndifferent effects of regulation on employment for four highly polluting/regulated\nindustries using data for the period 1979-91. Importantly, they measure regulation by the\nplant’s expenditures on pollution controls, as self-reported in the Commerce\nDepartment’s survey on Pollution Abatement Control Expenditures (PACE).\n\n\nWhile there are some differences across the four industries, Morgenstern et al. find that\nincreased abatement expenditures generally do _not_ cause a significant change in\nemployment at the industry level. Since these estimates do not consider the effect that\nregulation in one industry may have on other industries that make complementary or\nsubstitute products, the results should not be interpreted as relevant to the whole\neconomy, but they do highlight the complex responses to the new regulations exhibited\nby plants in the four industries.\n\n\nInterestingly, other studies have focused on air pollution control induced changes in\nemployment, using measures of regulation tied to specific policy actions rather than to\nself-reported pollution control expenditures.  Examples include the stringent air quality\nstandards imposed in the South Coast Air Quality Management District in the 1990s, and\nthe nationwide attainment/nonattainment designations used for criteria air pollutants, as\nwell as the realignment of such designations following the 1990 Clean Air Act\nAmendments.\n\n\nIn an early study, Berman and Bui (1997) found small, statistically insignificant\nemployment _gains_ associated with the more stringent South Coast air quality rules,\nconsistent with the later findings of Morgenstern et al. In contrast, in the analyses of\nfederal attainment/nonattainment designations, studies by Greenstone (2002) and Walker\n(2011) found adverse employment impacts associated with the regulations.\n\n\nImportantly, all three of these studies use measures of regulation tied to specific policy\nactions rather\n\nMention: \"survey on Pollution Abatement Control Expenditures\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Firms self-report their pollution control spending in this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005412:25:0", "occurrence_id": "14931b63528426dc9087c352", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 1 In the mid-2000s, with a Gini coefficient of 0.530, Latin America was 18 percent more unequal than SubSaharan Africa, 36 percent more unequal than East Asia and the Pacific and 65 percent more unequal than\nadvanced countries. Ferreira and Ravallion (2008). Around 2010, the Gini coefficient for the distribution of\nhousehold per capita income ranged from 0.387, 0.442 and 0.453 in Venezuela, Argentina (urban) and Uruguay,\nrespectively, to 0.556 and 0.567 in Bolivia and Honduras, respectively.\n2 Colombia constitutes one of the important exceptions to the declining trend in the region. See Moller (2012).\n3 “Income” in these estimates is current (i.e., excludes capital gains/losses, gifts and inheritance) income. It\nincludes wages, salaries, rents, and other incomes from capital as well as government cash transfers,\nremittances, and other private transfers. Income is presumed to be net of taxes for wage earners and before\ntaxes for the self-employed.\n4 The data on inequality and poverty used here is the one published by SEDLAC. The Socio-economic\nDatabase for Latin America and the Caribbean or SEDLAC is a joint project of CEDLAS (at Universidad de\nLa Plata, Argentina) and The World Bank. The version of SEDLAC used in this paper is March 2012\n(http://sedlac.econo.unlp.edu.ar/eng/).\n5 Interestingly, the decline in inequality has been widespread. Inequality has fallen in high-inequality Brazil and\nlow-inequality (by Latin American standards) Argentina, Uruguay and Venezuela; in countries with a large\nindigenous population (Ecuador and Peru) and countries with a small indigenous population (Argentina);\ncountries governed by the left (Brazil and Chile) and countries governed by non-leftist regimes (Mexico and\nPeru); countries with\n\nMention: \"Socio-economic\nDatabase for Latin America and the Caribbean\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"A joint university and bank project publishing the inequality and poverty series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005412:25:0", "occurrence_id": "695145cd57923756d115910e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 1 In the mid-2000s, with a Gini coefficient of 0.530, Latin America was 18 percent more unequal than SubSaharan Africa, 36 percent more unequal than East Asia and the Pacific and 65 percent more unequal than\nadvanced countries. Ferreira and Ravallion (2008). Around 2010, the Gini coefficient for the distribution of\nhousehold per capita income ranged from 0.387, 0.442 and 0.453 in Venezuela, Argentina (urban) and Uruguay,\nrespectively, to 0.556 and 0.567 in Bolivia and Honduras, respectively.\n2 Colombia constitutes one of the important exceptions to the declining trend in the region. See Moller (2012).\n3 “Income” in these estimates is current (i.e., excludes capital gains/losses, gifts and inheritance) income. It\nincludes wages, salaries, rents, and other incomes from capital as well as government cash transfers,\nremittances, and other private transfers. Income is presumed to be net of taxes for wage earners and before\ntaxes for the self-employed.\n4 The data on inequality and poverty used here is the one published by SEDLAC. The Socio-economic\nDatabase for Latin America and the Caribbean or SEDLAC is a joint project of CEDLAS (at Universidad de\nLa Plata, Argentina) and The World Bank. The version of SEDLAC used in this paper is March 2012\n(http://sedlac.econo.unlp.edu.ar/eng/).\n5 Interestingly, the decline in inequality has been widespread. Inequality has fallen in high-inequality Brazil and\nlow-inequality (by Latin American standards) Argentina, Uruguay and Venezuela; in countries with a large\nindigenous population (Ecuador and Peru) and countries with a small indigenous population (Argentina);\ncountries governed by the left (Brazil and Chile) and countries governed by non-leftist regimes (Mexico and\nPeru); countries with\n\nMention: \"SEDLAC\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The specific version used for the paper's estimates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005413:10:3", "occurrence_id": "876920fef598d2bce65e0d69", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .6 0.1 4.0 3.9 -0.1 23.0 -9.1 5.1 -1.5 4.3 -0.3\n\nTafiela 30.5 17.8 -12.7 1.5 1.6 0.1 1.5 1.3 -0.2 26.5 -8.7 1.9 -0.3 1.4 0.1\n\nMa'an 45.4 26.6 -18.8 3.0 3.4 0.4 2.1 1.9 -0.2 31.0 -4.4 3.0 0.4 1.8 0.3\n\nAqaba 34.0 19.7 -14.3 2.2 3.0 0.8 2.0 2.2 0.2 12.8 6.9 1.4 1.5 2.2 -0.1\n\n\n<u>Kingdom</u> <u>32.0</u> <u>14.9</u> <u>-17.1</u> <u>100.0</u> <u>100.0</u> <u>0.0</u> <u>100.0</u> <u>100.0</u> <u>0.0</u> <u>19.3</u> <u>-4.4</u> <u>100.0</u> <u>0.0</u> <u>100.0</u> <u>0.0</u>\n\nSource: World Bank Staff Calculations using Official Household Income and Expenditure Surveys\n\n\nTable 2 reveals a regional pattern for poverty changes. All governorates witnessed a\n\nstrong and commendable decline in poverty in the past\n\nMention: \"Official Household Income and Expenditure Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Governorate-level poverty changes are computed from these national surveys.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005420:18:0", "occurrence_id": "9f9a1ce938531f0ee5edb870", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <u>3. Global inequality over the long-run: From proletarians to migrants</u>\n\n\nI will now look at global inequality over the long sweep of history. It is here that we can\n\n\nestablish an important finding, which goes, I think, into some core issues of political philosophy\n\n\nand economics.\n\n\nLet us try to do for the entire period since the Industrial Revolution the same type of global\n\n\ninequality calculations which we have just shown for the last 20 years. We ask, “what was global\n\n\ninequality then—say, around the mid-19 <sup>th</sup> century?” It is a question impossible to answer with any\n\n\nprecision, because we do not have household surveys or any other reliable sources of income data\n\n\nfor these times. Nonetheless, some important attempts to estimate it have been made before,\n\n\nnotably by François Bourguignon and Christian Morrisson (2002), who were the first to take such a\n\n\nlong-run view. They used income levels (GDP per capita) from Angus Maddison’s database (2004;\n\n\n2007), and some of their own (necessarily often shaky) estimates of income distributions for\n\n\ndifferent parts of the world, to create global income distributions for eleven benchmark years\n\n\nspanning the period 1820-1992.  They did the best one could do with the available data —and\n\n\ntheir results have been corroborated, to the extent that it is possible to corroborate something as\n\n\ntentative, by several later authors (Van Zanden, Baten, Foldvari and van Leeuwen 2010; Milanovic,\n\n\n2011). The basic story that emerges from these calculations of income inequality in far-away\n\n\ntimes is that since the Industrial Revolution, which launched a score of European countries and their\n\n\noverseas off-shoots onto a path of faster growth, global inequality kept on rising until the mid-20 <sup>th</sup>\n\n\ncentury. There was a period of more than a\n\nMention: \"Angus Maddison’s database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Output per head over the long run is taken from this historical database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005427:8:1", "occurrence_id": "e1ac47accf9622ecf4446ceb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  we know,\nthere are no applications to the gravity model of trade.\n\n### **3. Data**\n\nBilateral trade in current USD comes from the United Nations’ (2012) _COMTRADE_ database,\naccessed through WITS. <sup>9</sup> Exporter and importer GDPs are from the World Bank’s (2012) _World_\n_Development Indicators_ and, consistent with the underlying theory, are measured in nominal\nUSD. <sup>10</sup> Bilateral country characteristics come from the CEPII distance dataset (Mayer and\nZignago 2011).\n\nThe analysis focuses on a country’s total merchandise exports disaggregated into differentiated\nproduct exports and commodity exports. A commodity is defined according to the International\nMonetary Fund’s (2012a) _Primary Commodity Prices_ and is coded at the six-digit HS 1988/92\n\n\n9 Instead of combining exports and imports for each country pair to construct symmetric trade flows, we use the\nThis technique is commonly used to minimize the risk of underreporting due to the fact that customs agencies unidirectional value of trade from country 𝑖 to country 𝑗 and introduce both importing and exporting country fixed\nusually monitor imports more closely than exports. effects. The exports from country 𝑖 to country 𝑗 are inferred using the imports reported by country 𝑗 from country 𝑖.\n10 On the use of trade and GDP data in current USD, see Baldwin and Taglioni (2006).\n\n\n7\n\nMention: \"CEPII distance dataset\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Bilateral country characteristics come from this distance dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005432:9:1", "occurrence_id": "c98391b6f163bb7d44b0f50a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  total number of days and incomes for each activity (block 23). Thus we are\n\nable to estimate income gains (net of forgone earnings) from NREGS in Bihar, and impacts on\n\npoverty. Moreover, we know if household members had enough NREGS working days if they\n\nworked on NREGS, or if they did not want to work on the scheme, both during the year ended.\n\n\nAnother attraction of the Bihar Survey is that the questions asked allow us to plausibly\n\nidentify households who are not rationed in the casual rural labor market. These are identified as\n\nthe sub-sample who report that they do not want more work on the NREGS than what they have\n\nalready. Since NREGS pays a higher wage rate (about 10% higher in Bihar) than the casual labor\n\n\n9 One difference is that we do not have the split manual/non-manual work in the Bihar survey: we only know\nif the activity was or was not casual work. This implies higher participation rate in this type of work in the Bihar\nSurvey (21.1%% in the NSS data on rural Bihar versus 39.8% in the Bihar Survey).\n\n\n8\n\nMention: \"Bihar Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its questions identify households not rationed in the employment scheme.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005432:9:1", "occurrence_id": "2012f96985a8b30bb48ea569", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  total number of days and incomes for each activity (block 23). Thus we are\n\nable to estimate income gains (net of forgone earnings) from NREGS in Bihar, and impacts on\n\npoverty. Moreover, we know if household members had enough NREGS working days if they\n\nworked on NREGS, or if they did not want to work on the scheme, both during the year ended.\n\n\nAnother attraction of the Bihar Survey is that the questions asked allow us to plausibly\n\nidentify households who are not rationed in the casual rural labor market. These are identified as\n\nthe sub-sample who report that they do not want more work on the NREGS than what they have\n\nalready. Since NREGS pays a higher wage rate (about 10% higher in Bihar) than the casual labor\n\n\n9 One difference is that we do not have the split manual/non-manual work in the Bihar survey: we only know\nif the activity was or was not casual work. This implies higher participation rate in this type of work in the Bihar\nSurvey (21.1%% in the NSS data on rural Bihar versus 39.8% in the Bihar Survey).\n\n\n8\n\nMention: \"NSS data on rural Bihar\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its lower participation rate contrasts with the other survey's estimate.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005442:12:0", "occurrence_id": "13f4d2ecee8df116b091041c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: appear to be motivating many countries and voters to contribute more (given that some of the richest\n\n\nOECD donors do not contribute the most, as measured by share of GDP). <sup>5</sup> [^5: The share of GNI target may well have played a role in motivating an increase in ODA in some countries, such as\nthe Norway, Sweden, United Kingdom, and Ireland, in the period from 2000 to 2011. But it is not clear, yet, whether\npublic support for increases in ODA has consolidated. For instance, Hudson and VanHeerde-Hudson (2012) find\nthat OECD public support for development aid, as measured by Eurobarometer surveys, rapidly fell back to 1990s\nlevels after peaking in run up to the 2005 Gleneagles Summit.] Moving to the more nudging\n\n\nindicator may well be worth the loss in precision.\n\n\nA similar argument might be made for carbon dioxide emissions, which while not an MDG\n\n\ntarget, has been suggested as a target for MDGs 2.0. Thousand metric tons of carbon dioxide equivalent\n\n\nemissions, per capita or per dollar of national GDP, is a difficult-to-grasp indicator. Even measures of\n\n\nglobal average temperature may not be intuitive – an increase of two degrees centigrade in the average\n\n\ntemperature of the world does not seem like much to the untrained ear. Rather, the contributions of an\n\n\naverage person in a country to the rise in sea levels, in millimeters, over a defined time period, would be a\n\n\nmore palpable target. That indicator can be visualized. It also embeds within it the causal narrative\n\n\nbetween carbon emissions and some of its potentially catastrophic consequences. It is true that estimating\n\n\ncurrent and future sea levels as a function of carbon emissions is an ongoing area of research, and the\n\n\nestimates would invariably have a confidence interval around them. But again, if the goal is nudging, and\n\n\nif the confidence intervals are not too large, this would be an acceptable cost for a motivational device.\n\n\nFor fair trade, which is another element of the\n\nMention: \"Eurobarometer surveys\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Public support for development aid across donor countries is measured here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005444:9:0", "occurrence_id": "5a1d4510749d172402678519", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **4.** **Sources and likely persistence of foreign investor demand**\n\n\n\nThe growing popularity of EM local currency debt is\nno mystery when its strong performance versus other\nmajor asset classes in recent years is considered\n(Figure 5). <sup>6</sup> [^6: The finding that EM local currency debt has been a strong performer relative to other asset classes since at least the early 2000s\nis a robust one, but it should be noted that specific numbers such as those shown here are very sensitive to the time period\nchosen and to the inclusion of the extreme market price moves seen at the height of the financial crisis.] Indeed, the basic long-term investment\ncase for EM local currency debt—exposure to\nrelatively high real yields, prospects for currency\ngains (more on this immediately below), and capital\nallocation to sounder sovereign balance sheets than\nthose of many advanced economies—has not only\nweathered the crises of recent years, but arguably\nbeen strengthened by them.\n\n\nFrom a multi-asset perspective, some allocation to\nEM local currency debt of course also confers\nportfolio diversification benefits. This holds even\nthough, as a “risky asset”, the correlation of\nbenchmark GBI-EM index returns to those of other\nasset classes has tended to be quite high (Figure 6).\nInspecting historical correlations also reveals the\nclose link between GBI-EM returns and EM currency\nreturns (proxied using the JP Morgan ELMI+ index –\nan EM currency return index). Gaining exposure to\nEM currencies, which have shown secular\nappreciation trends, has likely been a major\nmotivation for investors’ increased allocations to\nthese countries’ domestic bonds.\n\n\nBut who exactly are these investors? Unfortunately,\nthere are no comprehensive data on the current\ncomposition of the foreign investor base for EM local\ncurrency debt, nor is it straightforward to gauge its\npotential future size. However, some recent estimates\nare available from market participants.\n\n\n\n**<u>Figure 5</u>** <u>:</u>\n\nMention: \"JP Morgan ELMI+ index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Emerging market currency returns are proxied by this index.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005446:41:2", "occurrence_id": "1a7613c60ba567a1b55989d0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |<br>91|\n|Is there a regulatory requirement for general provisions on loans and advances?|63|36|73|\n|Do you require banks to write off non-performing loans after a specific time period?|40|14|49|\n|Does accrued, though unpaid, interest/principal enter the bank's income statement<br>while the loan is classified as non-performing?|28<br>(14)|39<br>(31)|23<br>(8)|\n|If a customer has multiple loans and advances and one of them is classified as non-<br>performing, are all the other exposures classified as non-performing as well?<br>|53<br>(50)<br>|48<br>(35)<br>|54<br>(55)<br>|\n|**C. Some Additional Information**<br>|**Median**|**Median **|**Median**|\n|<br>What is the maximum time (in months) that a loan or advance can be non-performing<br>before it has to be written off?|<br>12|<br>18|<br>12|\n|<br>Up to what level (if any) is specific provisions tax deductible?|0|0|1|\n|Up to what level (if any) is general provisions tax deductible?<br>|0 <br>|0 <br>|0 <br>|\n\n\nparentheses refer to the 2007 Bank Regulation and Supervision Survey, which captures data as of 2005.\nADV: advanced economies, EMDEs: emerging and developing economies (Table 1 provides the complete list of countries).\n\n\n40\n\nMention: \"2007 Bank Regulation and Supervision Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its questions on tax deductibility capture conditions as of that earlier year.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005460:4:0", "occurrence_id": "ce99951247bd6cdfa52151a9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: with the recent aggregation by the World Bank of country level data from 45 countries\n(with different time coverage), which provides a real international comparison base for\nvarious elements of export dynamics (Cebeci, et al. 2012).\n\nThis paper complements the empirical literature by exploring a rich firm-level\ndataset of customs transactions from the Lao PDR to identify determinants of export\nsurvival. Such detailed analysis for a small, developing and landlocked economy, which,\nfurthermore, relies heavily on exports of raw materials, can be of considerable help to\npolicymakers and economists interested in export-led growth and diversification, and\nthere are currently very few other examples available (Volpe and Carballo 2008). It\ncontinues work already carried out with this dataset (Stirbat, Record and\nNghardsaysone 2011) which has provided novel descriptive statistics of the Lao\nexporter population, in particular the dominance of the intensive margin for export\ngrowth and the great deal of churning and experimentation happening in exports. That\narticle, looking at the chances of an export unit (a firm exporting an HS 6-figure\nproduct to a particular destination) to survive past the first year (in a cross-section\nsetup), has identified positive externalities from agglomeration and a positive impact of\nproduct- and destination-related experience.\n\nIn contrast to this previous work, the current analysis of the determinants of\nexporter survival is carried out in a dynamic context, using the full panel structure of\nthe data. It has also been slightly extended to cover five full years. The aggregation of\ninternational trade data by years as well as a general shortage of detailed firm-level\ndata means that yearly reporting is the general norm in the literature. In this case,\nhowever, the richness of the dataset allows for an analysis of month-level evolutions\ninstead of just yearly ones. There are, however, several exceptions (Tovar and Martinez\n2011), including one of the first instances of survival analysis of exporter data\n(Sab\n\nMention: \"firm-level\ndataset of customs transactions\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Export survival determinants are identified from these transaction records.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005461:14:0", "occurrence_id": "abee1d150d2583b12516ede0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Firm-level dynamics of exports**\n\n\nThe recent public availability of microdata allows crucial evaluation of export performance, that is, the\nentry, exit, and survival dynamics of firms in global markets. Export survival sustainability—first\naddressed by the seminal work of Besedes and Prusa (2006)—is a key dimension to understanding export\nperformance, and has become an important factor for policy makers to consider in the design of policy to\npromote exports.\n\n\nThe World Bank has recently assembled an Exporter Dynamics Database that includes exporter\ncharacteristics and exporter growth measures based on firm-level customs information from 38\ndeveloping and 7 developed countries, primarily for the period between 2003 and 2010. <sup>5</sup> [^5: See Cebeci and others (2012) for a full and detailed description of the database.] The use of this\ndatabase makes it possible to compare Brazilian exports with other middle-income economies in the\nworld (results are shown in figures 11–13).\n\n\nThe survival rate of exporters in Brazil stood at very high levels during 2003–9 (figure 11), only lower\nthan levels observed in Turkey. While this result can be considered positive, it reflects a small and\ndecreasing entry rate of firms into foreign trade. The entry rate of exporters was already low in Brazil, and\ndropped further to 22 percent, much lower than the values observed in countries chosen for comparison\n(around 35 percent). <sup>6</sup> [^6: Cebeci and others (2012) argued that Brazil presents the lowest entry rate among 44 countries in the above-mentioned database.]\n\n\nThat result—a matter of concern—also confirms other indicators recently released by the Ministry of\nDevelopment, Trade and Tourism for Brazil. <sup>7</sup> [^7: _Estado de São Paulo_, August 3, 2012 (editorial).] Several studies, such as Clerides, Lach, and Tybout (1998)\nfor Colombia, Mexico and Morocco, or Bernard and Jensen (\n\nMention: \"Exporter Dynamics Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Exporter characteristics and growth measures assembled from customs data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005462:16:2", "occurrence_id": "67e53398896afadfb412a7b1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: </sup> -,\n\nA final international comparison issue our data can shed light on is the association between the\ndissimilarity index (the measure of inequality of opportunity contained within the HOI) and the index of\ninequality of economic opportunity (IEO-R). The dissimilarity index can be interpreted as the proportion\n\n\n\n17 A possible caveat with viewing the dissimilarity index within the HOI as a measure of inequality of opportunity is\nthat the index is typically calculated “for children”. This justifies the use of certain variables - like geographic\nlocation or education of the adults in the household - as circumstances, which are clearly in the realm of choices\nfor the adults. The argument is that the index applies to children, and these are circumstances from their\nperspective. But this then raises the issue of age of responsibility, and whether or not all inequalities in access to\nservices for children below a certain age should not be considered inequality of opportunity. Under that view,\nunequal access to water or sanitation among five-year olds within the same type (i.e. sharing identical observed\ncircumstances) should also be counted as inequality of opportunity.\n18 The correlation with the inequality-adjusted Human Development Index introduced for the first time in 2011 is\nalmost the same: 0.95.\n\n15\n\nMention: \"inequality-adjusted Human Development Index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Introduced in that year, it correlates almost fully with the opportunity measure.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005469:4:0", "occurrence_id": "efc3739b7a9be18dff73c66a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ###### **Private Chinese Investment in Africa: Myths and Realities** By Xiaofang Shen\n\nThe recently increased Chinese investment in Africa has grabbed the world’s attention. Press\nviews and opinions triggered by this new development are starkly polarized. At one end of the\nspectrum, one hears praise for China’s role in filling financial and technological gaps for Africa,\na role which western companies have been reluctant to play especially since the 2008 financial\ncrises. At the other end, one hears considerable skepticism regarding the motivation behind\nChinese investment and its consequential impact on Africa, including charges that China is recolonializing the continent. <sup>2</sup>\n\n\nAt first glance, available statistics do not justify the degree of attention China’s investment in\nAfrica is receiving. Official data from the Ministry of Commerce (MOF) on Chinese overseas\ninvestment shows total Chinese outbound foreign direct investment (OFDI) in 2011 reached a\nrecord high of about US $74.65 billion, but only $1.7 billion, or a mere 2.2%, went to Africa. <sup>3</sup>\nCompared against the global foreign direct investment (FDI) flowing to the continent that year,\naround $42.65 billion according to the United Nations Conference of Trade and Development\n(UNCTAD, 2012), China’s contribution was a mere 4%. When measured in terms of FDI stocks,\nChinese OFDI in Africa appears even more trivial – reportedly standing at $14.7 billion by the\nend of 2011, or 2.6 percent of the total $570 billion FDI stock for the whole African continent. <sup>4</sup>\n\n\nSo why is China’s investment attracting so much attention?\n\n\nOne reason is the pace at which Chinese OFDI has risen in Africa, which, truly, has been\nbreathtaking. Until about 15 years ago, China’s\n\nMention: \"the Ministry of Commerce (MOF)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Official records of outbound investment are published by this ministry.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005471:7:0", "occurrence_id": "5297e2bf8bbf91872db3d24f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: In line with worldwide AD activities, the chemical and iron-steel industries dominate the filings\n\n\nin terms of the number of cases. <sup>5</sup>\n\n\nThe remainder of the paper is organized as follows. The second section describes the\n\n\ndata, the third section presents the empirical methodology and results, and the fourth section\n\n\nconcludes.\n\n\n**_<<A>>_** **_DATA_**\n\n\nWe employed detailed export data from the Brazilian customs office, SECEX _(Secretaria_\n\n\n_de Comércio Exterior_ ), which gathers export reports using product codes at the plant, month, and\n\n\nNCM ( _Nomenclatura Comum do Mercosul_ ) levels. The NCM codes are eight-digit numbers, the\n\n\nfirst six digits of which coincide with the first six digits in the HS. The destination information is\n\n\ntransformed from the Brazilian country codes into the international ISO system. The product\n\n\ncodes at the six-digit level in the Brazilian data, for which no corresponding HS entries exist,\n\n\nwere removed from the data. All export values in the SECEX data are reported in current U.S.\n\n\ndollars (USD), free on board (fob). We used observations on exporting plants, declared export\n\n\nvalues, export destinations and export quantities, and we aggregated monthly plant-level export\n\n\ninformation to years and firms. We focused solely on manufactured products. The main variable\n\n\nof interest in our analysis is the unit price of the exported good _p_, shipped by firm _f_ to destination\n\n\n\n_d,_ which is modeled as follows:\n\n\n\n𝑢𝑓𝑝𝑑 = <sup><u>𝐸𝑥𝑝𝑜𝑟𝑡 𝑣𝑎𝑙𝑢𝑒𝑓𝑝𝑑</u></sup>\n\n~~𝑄𝑢𝑎𝑛𝑡𝑖𝑡𝑦𝑓𝑝𝑑~~\n\n\n6\n\n\n\n.\n\nMention: \"SECEX data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Export values in current dollars, free on board, come from these records.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005483:22:0", "occurrence_id": "d2d7a8c225466465d7eb20ca", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <u>21</u>\n\n**Appendix 3 - Variable Definition and Sources**\n\n\n\n**Variables** **Description** **Data sources**\n\n\n\n**Variables** **Description** **Data sources**\n\n\n_Poverty incidence_\n\n~~The percentage of the population~~\n~~living below the $1.25/day~~\ninternational poverty line\n_Poverty gap_\nWorld Bank Global Poverty Index\n<u>The average shortfall of the poor</u> <u>Database</u>\n<u>with respect to the poverty line,</u>\n<u>multiplied by the headcount ratio</u>\n_<u>Trade openness</u>_\n~~Sum of exports and imports as a~~\n~~share of GDP~~\n\nAnnual percentage change in\n\n\n\n_Poverty incidence_\n\n\n\n~~share of GDP~~\n\n_Inflation_\nAnnual percentage change in\nconsumer prices\n\n\n\n_GDP per capita_\nNominal GDP divided by\npopulation size\n\n\n\n_Private Credit/GDP_\n<u>Domestic credit to private sector</u>\n<u>(% of GDP)</u>\n\n~~Primary completion rate: total~~\n\n\n\nInternational Financial Statistics and\nWorld Development Indicators\n\n\n\n_Education_\n\n\n_Law and order_\n\n\n\n<u>(% of GDP)</u>\n\n~~Primary completion rate: total~~\n~~number of new entrants in the last~~\ngrade of primary education,\n<u>regardless of age, expressed as</u>\n<u>percentage of the total population</u>\n<u>of the theoretical entrance age to</u>\n<u>the last grade of primary.</u>\n\n~~Strength and impartiality of the~~\n\n\n\n<u>of the theoretical entrance age to</u>\n<u>the last grade of primary.<\n\nMention: \"World Bank Global Poverty Index\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The average shortfall of the poor relative to the line comes from this index.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005483:22:0", "occurrence_id": "e4a493a5aa0fcb2f93fd8fbb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <u>21</u>\n\n**Appendix 3 - Variable Definition and Sources**\n\n\n\n**Variables** **Description** **Data sources**\n\n\n\n**Variables** **Description** **Data sources**\n\n\n_Poverty incidence_\n\n~~The percentage of the population~~\n~~living below the $1.25/day~~\ninternational poverty line\n_Poverty gap_\nWorld Bank Global Poverty Index\n<u>The average shortfall of the poor</u> <u>Database</u>\n<u>with respect to the poverty line,</u>\n<u>multiplied by the headcount ratio</u>\n_<u>Trade openness</u>_\n~~Sum of exports and imports as a~~\n~~share of GDP~~\n\nAnnual percentage change in\n\n\n\n_Poverty incidence_\n\n\n\n~~share of GDP~~\n\n_Inflation_\nAnnual percentage change in\nconsumer prices\n\n\n\n_GDP per capita_\nNominal GDP divided by\npopulation size\n\n\n\n_Private Credit/GDP_\n<u>Domestic credit to private sector</u>\n<u>(% of GDP)</u>\n\n~~Primary completion rate: total~~\n\n\n\nInternational Financial Statistics and\nWorld Development Indicators\n\n\n\n_Education_\n\n\n_Law and order_\n\n\n\n<u>(% of GDP)</u>\n\n~~Primary completion rate: total~~\n~~number of new entrants in the last~~\ngrade of primary education,\n<u>regardless of age, expressed as</u>\n<u>percentage of the total population</u>\n<u>of the theoretical entrance age to</u>\n<u>the last grade of primary.</u>\n\n~~Strength and impartiality of the~~\n\n\n\n<u>of the theoretical entrance age to</u>\n<u>the last grade of primary.<\n\nMention: \"International Financial Statistics and\nWorld Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Domestic credit and primary completion measures come from these compilations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005488:46:0", "occurrence_id": "a3541248d12b026cb4715137", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Aggregate of all activities 0.367** 0.361**\n\n<u>**denotes that the figures are statistically significant at less than 5% level. Note: Marginal return is calculated from the</u>\nfindings of Table 11 and is defined as the return (profit) on each additional taka invested in working capital and the value\nof the enterprise.\nSource: HIES 2000, 2005 and 2010\n\n\n**Table 13: Household distribution by constraints in the operation of microenterprise**\n\n\n<u>Constraint type</u> **<u>2000</u>** **<u>2005</u>** **<u>2010</u>**\n\n<u>No constraints</u> <u>33.0</u> <u>23.8</u> <u>49.7</u>\n\nInadequate capital or credit 27.1 25.5 22.1\n\nInadequate knowhow 3.8 3.9 2.2\n\nHigh operating cost 0.1 0 3.6\n\nUnreliable/inadequate power supply 0.5 1.1 2.9\n\nProblems with equipment/spare parts 1.7 0.6 0.4\n\nGovernment regulations 2.5 4.0 1.1\n\nLack of raw materials 10.5 18.2 2.9\n\nInadequate demand of products 8.3 6.2 7.1\n\nTransport problems 2.5 7.1 5.0\n\nOther miscellaneous problems 10.0 9.6 3.0\n\nN 1424 1418 1904\n\n<u>Source: HIES 2000, 2005, 2010</u>\n\n\n\n**<u>Table 14: Extent of credit and non-credit constraints by microenterprise activity</u>**\n**<u>Enterprise activity</u>** **<u>Enterprises are</\n\nMention: \"HIES 2000, 2005 and 2010\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Returns to working capital and enterprise value are computed across these rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005489:10:3", "occurrence_id": "a975d2d48d07d02bfef54312", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: >6.6<br> <br>1.7<br>7.3<br>13.8<br>7.5<br>5.1<br>7.2<br>15.8<br>9.3<br>2.4<br>7.0<br>11.9<br>7.0<br>-3.5<br>9.4<br>12.1<br>5.8<br>3.0<br>2.7<br>12.3<br>5.9<br> <br>8.3<br>3.9<br>18.7<br>10.1<br>-2.8<br>8.9<br>11.5<br>5.7|\n\n\n\nNote: The classification of manufacturing products in exports and imports is defined as SITC 5+6+7+8-68 in revision 3.\n\n\nSource: Based on UN COMTRADE Statistics.\n\nMention: \"UN COMTRADE Statistics\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Export and import composition is measured from these statistics.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005496:38:0", "occurrence_id": "58dbfa72aa964abdee3da42f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: highest growth firms and products (Melitz, 2003). Pakistan’s performance in this respect is\nconsiderably deteriorating over time. The rates of survival after 2004 increased. At the same\ntime, the number of new exporters each year steadily decreased. These two results together are\na potentially worrisome signal of a decrease in firm dynamism, whereby the greater persistence\nin export relationships is due to less competition in the domestic market. This may contribute to\nexplain why Pakistan’s export performance in the past decade was positive but more moderate\nthan the one of other Asian economies.\n\n\n**Figure 28**\n**Survival rates after 1 year of export activity**\n\n\n\n\n\n\n\nSource: Authors’ calculations based on data from the Exporter Dynamics Database (World Bank).\n\nTo investigate further, whether the lower dynamism of firms’ exports has had a detrimental\neffect on export growth, the next sub-section will decompose the year-on-year contribution of\nthe extensive and the intensive margin to aggregate export growth.\n\nExports growth is computed using the mid-point growth rate (Davis and Haltiwanger 1992 as\napplied to firm-level exports by Bricogne et al 2011). The main advantage of this approach is\nthat it allows the computation of the contribution of the extensive and the intensive margin to\nexport growth. Other methodologies only allow a static assessment of the extensive margin\n(Hummels and Klenow 2005).\n\nTable 11 that provides the decomposition reads as follows. In each of the columns the average\nannual growth relative to the indicated period is reported. The aggregate figure for all trade\nflows is reported in the last row. The remainder of the table reports the contribution to\naggregate growth by each of the margins. For example, looking at annual growth over the year\n2002-2003, the first row indicates that export growth due to entry of new firms into exporting\nequaled 9.3%. Since the corresponding loss in export growth due to the exit of firms equaled\n7.8%,\n\nMention: \"Exporter Dynamics Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Survival after one year of exporting is calculated from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005503:2:1", "occurrence_id": "3705cbf9147e7437730ca008", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: __________________\n*Hallsworth Professor of International Macroeconomics and Development Economics, University of\nManchester, and co-Director, Centre for Growth and Business Cycle Research; and **Vice President, World\nBank. We would like to thank Lim King Yoong and Baris Alpaslan for research assistance, Lilian Lopes Ribeiro\nfor helping us to identify the PNAD data used for calibration, and Pablo Fajnzylber for helpful comments. The\nAppendix is available upon request.\n\nMention: \"PNAD data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"These household records were identified for calibrating the model.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005504:38:0", "occurrence_id": "cb79309824aa1a50ba672586", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ##### **Appendix 1: List of restrictions in the services and manufacturing sectors** **included in the OECD Restrictiveness Index.**\n\n**Types of Restrictions:**\n\n\n_I - Equity restrictions_ _III - Key personnel_\n\n_II - Screening & Approval_ _IV - Other restrictions_\n\n\n**Sectoral**\n\n\n**_<u>Year</u>_** **_<u>Sectors</u>_** **_<u>Sub-sectors</u>_** **_<u>Measure</u>_** **_<u>Justifications / comments</u>_**\n\n**<u>1997</u>** <u>Manufacturing</u> <u>Transport equipment</u> <u>I</u> <u>Air transport equipmentt only in joint-venture with</u>\nITPN\n\n**1997** Electricity Electricity generation I Joint venture requirement existed in 1995. In 2000 it\nwas 95%, as it was in 2010. Assume that the same\nminimum applied before 2000.\n\n**1997** Electricity Electricity distribution I JV requirement has always existed. In 2000 it was\n95%, as it was in 2010. Assume that the same\nminimum applied before 2000. It also applies to\ndistribution\n**1997** Construction I Foreign construction firms are only permitted to be\nsubcontractors or advisors to local firms in areas\nwhere the governmentt believes that a local firm is\nunable to do the work.\n**1997** Distribution Wholesale I Foreign firms and foreign-controlled firms are not\nallowed to distribute in Indonesia unless they\nproduce there (and even then only at the wholesale\nlevel).\n**1997** Distribution Retail I All retail distribution must be handled by Indonesian\nfirms.\n*\n\nMention: \"the OECD Restrictiveness Index\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its appendix lists the equity, personnel and screening restrictions counted.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005506:15:0", "occurrence_id": "a68d5d744e1bb711f4bc576c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Indian height deficit by an increasing association between sanitation and height.\n\n###### **2.3 The gap between South Asia and sub-Saharan Africa**\n\n\nAlthough people in South Asia are, on average, richer than people in sub-Saharan Africa,\n\n\nchildren in South Asia are shorter, on average, and open defecation is much more common\n\n\nthere. How much of the South Asia-Africa gap can sanitation statistically explain, at the\n\n\nlevel of country averages?\n\n\nTable 5 estimates regressions in the form of equation 1, with the sample restricted to\n\n\ncountries in South Asia and sub-Saharan Africa and with an indicator variable added for\n\n\ndata from South Asia. Of the 140 DHS surveys in figure 1, 11 are from South Asia and 78\n\n\nare from sub-Saharan Africa. In these data, children in South Asia are, on average, about\n\n\none-third of a height-for-age standard deviation shorter. <sup>10</sup> [^10: Jayachandran and Pande (2012), using individual-level DHS data from Africa and South Asia, suggest\nthat first-born South Asian children are _taller_ than first-born African children. The country-level data\nstudied here, however, show no similar reversal. If country means are computed using only first-born children,\nI find that South Asian children are 0.22 standard deviations shorter ( _s.e._ = 0 _._ 05), a reduction but not an\nelimination of the 0.36 gap in table 5. In this sample of country means of first-borns, the gap falls to 0.15\nwith a control for open defecation and to 0.08 with a control for open defecation per square kilometer. In the\nfull sample of country-level means of first-borns, analogously to column 1 of table 1, moving from an open\ndefecation rate of 0 to 1 is linearly associated with a decline in height for age of 1.11 standard deviations.]\n\n\nHow do further controls change the estimate of this South Asia indicator? Merely lin\n\nearly controlling for open defecation reduces the gap by 30 percent from\n\nMention: \"DHS surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Child height outcomes across 140 surveys in two regions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005506:15:0", "occurrence_id": "6899a40860a97667db14b9f6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Indian height deficit by an increasing association between sanitation and height.\n\n###### **2.3 The gap between South Asia and sub-Saharan Africa**\n\n\nAlthough people in South Asia are, on average, richer than people in sub-Saharan Africa,\n\n\nchildren in South Asia are shorter, on average, and open defecation is much more common\n\n\nthere. How much of the South Asia-Africa gap can sanitation statistically explain, at the\n\n\nlevel of country averages?\n\n\nTable 5 estimates regressions in the form of equation 1, with the sample restricted to\n\n\ncountries in South Asia and sub-Saharan Africa and with an indicator variable added for\n\n\ndata from South Asia. Of the 140 DHS surveys in figure 1, 11 are from South Asia and 78\n\n\nare from sub-Saharan Africa. In these data, children in South Asia are, on average, about\n\n\none-third of a height-for-age standard deviation shorter. <sup>10</sup> [^10: Jayachandran and Pande (2012), using individual-level DHS data from Africa and South Asia, suggest\nthat first-born South Asian children are _taller_ than first-born African children. The country-level data\nstudied here, however, show no similar reversal. If country means are computed using only first-born children,\nI find that South Asian children are 0.22 standard deviations shorter ( _s.e._ = 0 _._ 05), a reduction but not an\nelimination of the 0.36 gap in table 5. In this sample of country means of first-borns, the gap falls to 0.15\nwith a control for open defecation and to 0.08 with a control for open defecation per square kilometer. In the\nfull sample of country-level means of first-borns, analogously to column 1 of table 1, moving from an open\ndefecation rate of 0 to 1 is linearly associated with a decline in height for age of 1.11 standard deviations.]\n\n\nHow do further controls change the estimate of this South Asia indicator? Merely lin\n\nearly controlling for open defecation reduces the gap by 30 percent from\n\nMention: \"individual-level DHS data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"First-born height and sex ratios are analysed at the individual level there.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005486:12:0", "occurrence_id": "d544f437957ba3978dd02ca7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: country size using GDP from the Penn World Tables and the real exchange rate. We use the bilateral\n\n\nreal exchange rate between France and other countries using producer prices of France and importer\n\n\ncountries from the International Financial Statistics (IFS) of the IMF. Finally, to compute the proxy for\n\n\nforeign competition at the extensive and intensive margin, we use the BACI dataset at the HS6 product\n\n\nlevel by country of origin. <sup>14</sup>\n\n\nIn order to keep a constant sample throughout the paper and to establish the stability of the point es\n\ntimates, we only consider firms that have information on all control variables. In the main specifications,\n\n\nthis leaves us with around 4,900 firm-country pairs for the 9 Asian countries in the period 1999-2005, a\n\n\ntotal of 34,327 observations.\n\n\nTo perform the robustness check exercises, we use two additional firm level datasets. First, to identify\n\n\nFrench multinational firms, we match our main dataset with firm level dataset on multinational groups\n\n\nlocated in France from the _Enquete Echanges Internationaux Intra-Groupe_ produced by the French Of\n\nfice of Industrial Studies and Statistics (SESSI). These data provide a good representation of the activity\n\n\nof international groups located in France. They account for around 82 percent of total trade flows by\n\n\nmultinationals, and for 55 and 61 percent of total French imports and exports respectively. Second, to\n\n\nadd information on firms’ characteristics, we merge our main dataset with the Annual French Business\n\n\nSurveys (EAE), available from INSEE. The EAE survey is conducted every year and provides detailed\n\n\nfirm-level information for all French firms with more than 20 employees whose main activity is in manu\n\nfacturing. This survey allows us to have information on firms’ employment and labor productivity (value\n\n\nadded per worker).\n\n##### _Market Access: firm level tariff measures_\n\n\nTo identify the impact of Asian’s\n\nMention: \"Penn World Tables\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Country size is measured with output taken from these tables.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005486:12:0", "occurrence_id": "bf5caaac6f6c8c924550607f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: country size using GDP from the Penn World Tables and the real exchange rate. We use the bilateral\n\n\nreal exchange rate between France and other countries using producer prices of France and importer\n\n\ncountries from the International Financial Statistics (IFS) of the IMF. Finally, to compute the proxy for\n\n\nforeign competition at the extensive and intensive margin, we use the BACI dataset at the HS6 product\n\n\nlevel by country of origin. <sup>14</sup>\n\n\nIn order to keep a constant sample throughout the paper and to establish the stability of the point es\n\ntimates, we only consider firms that have information on all control variables. In the main specifications,\n\n\nthis leaves us with around 4,900 firm-country pairs for the 9 Asian countries in the period 1999-2005, a\n\n\ntotal of 34,327 observations.\n\n\nTo perform the robustness check exercises, we use two additional firm level datasets. First, to identify\n\n\nFrench multinational firms, we match our main dataset with firm level dataset on multinational groups\n\n\nlocated in France from the _Enquete Echanges Internationaux Intra-Groupe_ produced by the French Of\n\nfice of Industrial Studies and Statistics (SESSI). These data provide a good representation of the activity\n\n\nof international groups located in France. They account for around 82 percent of total trade flows by\n\n\nmultinationals, and for 55 and 61 percent of total French imports and exports respectively. Second, to\n\n\nadd information on firms’ characteristics, we merge our main dataset with the Annual French Business\n\n\nSurveys (EAE), available from INSEE. The EAE survey is conducted every year and provides detailed\n\n\nfirm-level information for all French firms with more than 20 employees whose main activity is in manu\n\nfacturing. This survey allows us to have information on firms’ employment and labor productivity (value\n\n\nadded per worker).\n\n##### _Market Access: firm level tariff measures_\n\n\nTo identify the impact of Asian’s\n\nMention: \"International Financial Statistics\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Producer prices for the exchange rate construction come from this publication.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005486:12:0", "occurrence_id": "6407ee8cd6ebb3bbdaac3156", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: country size using GDP from the Penn World Tables and the real exchange rate. We use the bilateral\n\n\nreal exchange rate between France and other countries using producer prices of France and importer\n\n\ncountries from the International Financial Statistics (IFS) of the IMF. Finally, to compute the proxy for\n\n\nforeign competition at the extensive and intensive margin, we use the BACI dataset at the HS6 product\n\n\nlevel by country of origin. <sup>14</sup>\n\n\nIn order to keep a constant sample throughout the paper and to establish the stability of the point es\n\ntimates, we only consider firms that have information on all control variables. In the main specifications,\n\n\nthis leaves us with around 4,900 firm-country pairs for the 9 Asian countries in the period 1999-2005, a\n\n\ntotal of 34,327 observations.\n\n\nTo perform the robustness check exercises, we use two additional firm level datasets. First, to identify\n\n\nFrench multinational firms, we match our main dataset with firm level dataset on multinational groups\n\n\nlocated in France from the _Enquete Echanges Internationaux Intra-Groupe_ produced by the French Of\n\nfice of Industrial Studies and Statistics (SESSI). These data provide a good representation of the activity\n\n\nof international groups located in France. They account for around 82 percent of total trade flows by\n\n\nmultinationals, and for 55 and 61 percent of total French imports and exports respectively. Second, to\n\n\nadd information on firms’ characteristics, we merge our main dataset with the Annual French Business\n\n\nSurveys (EAE), available from INSEE. The EAE survey is conducted every year and provides detailed\n\n\nfirm-level information for all French firms with more than 20 employees whose main activity is in manu\n\nfacturing. This survey allows us to have information on firms’ employment and labor productivity (value\n\n\nadded per worker).\n\n##### _Market Access: firm level tariff measures_\n\n\nTo identify the impact of Asian’s\n\nMention: \"BACI dataset\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Product-level trade by origin feeds the competition proxies.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005486:12:0", "occurrence_id": "314c595bfa41847b358ff802", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: country size using GDP from the Penn World Tables and the real exchange rate. We use the bilateral\n\n\nreal exchange rate between France and other countries using producer prices of France and importer\n\n\ncountries from the International Financial Statistics (IFS) of the IMF. Finally, to compute the proxy for\n\n\nforeign competition at the extensive and intensive margin, we use the BACI dataset at the HS6 product\n\n\nlevel by country of origin. <sup>14</sup>\n\n\nIn order to keep a constant sample throughout the paper and to establish the stability of the point es\n\ntimates, we only consider firms that have information on all control variables. In the main specifications,\n\n\nthis leaves us with around 4,900 firm-country pairs for the 9 Asian countries in the period 1999-2005, a\n\n\ntotal of 34,327 observations.\n\n\nTo perform the robustness check exercises, we use two additional firm level datasets. First, to identify\n\n\nFrench multinational firms, we match our main dataset with firm level dataset on multinational groups\n\n\nlocated in France from the _Enquete Echanges Internationaux Intra-Groupe_ produced by the French Of\n\nfice of Industrial Studies and Statistics (SESSI). These data provide a good representation of the activity\n\n\nof international groups located in France. They account for around 82 percent of total trade flows by\n\n\nmultinationals, and for 55 and 61 percent of total French imports and exports respectively. Second, to\n\n\nadd information on firms’ characteristics, we merge our main dataset with the Annual French Business\n\n\nSurveys (EAE), available from INSEE. The EAE survey is conducted every year and provides detailed\n\n\nfirm-level information for all French firms with more than 20 employees whose main activity is in manu\n\nfacturing. This survey allows us to have information on firms’ employment and labor productivity (value\n\n\nadded per worker).\n\n##### _Market Access: firm level tariff measures_\n\n\nTo identify the impact of Asian’s\n\nMention: \"firm level dataset on multinational groups\n\n\nlocated in France\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Produced by the national statistics office, it identifies multinational affiliates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005486:12:0", "occurrence_id": "6f281d9a28810fc74f6e4ee2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: country size using GDP from the Penn World Tables and the real exchange rate. We use the bilateral\n\n\nreal exchange rate between France and other countries using producer prices of France and importer\n\n\ncountries from the International Financial Statistics (IFS) of the IMF. Finally, to compute the proxy for\n\n\nforeign competition at the extensive and intensive margin, we use the BACI dataset at the HS6 product\n\n\nlevel by country of origin. <sup>14</sup>\n\n\nIn order to keep a constant sample throughout the paper and to establish the stability of the point es\n\ntimates, we only consider firms that have information on all control variables. In the main specifications,\n\n\nthis leaves us with around 4,900 firm-country pairs for the 9 Asian countries in the period 1999-2005, a\n\n\ntotal of 34,327 observations.\n\n\nTo perform the robustness check exercises, we use two additional firm level datasets. First, to identify\n\n\nFrench multinational firms, we match our main dataset with firm level dataset on multinational groups\n\n\nlocated in France from the _Enquete Echanges Internationaux Intra-Groupe_ produced by the French Of\n\nfice of Industrial Studies and Statistics (SESSI). These data provide a good representation of the activity\n\n\nof international groups located in France. They account for around 82 percent of total trade flows by\n\n\nmultinationals, and for 55 and 61 percent of total French imports and exports respectively. Second, to\n\n\nadd information on firms’ characteristics, we merge our main dataset with the Annual French Business\n\n\nSurveys (EAE), available from INSEE. The EAE survey is conducted every year and provides detailed\n\n\nfirm-level information for all French firms with more than 20 employees whose main activity is in manu\n\nfacturing. This survey allows us to have information on firms’ employment and labor productivity (value\n\n\nadded per worker).\n\n##### _Market Access: firm level tariff measures_\n\n\nTo identify the impact of Asian’s\n\nMention: \"Annual French Business\n\n\nSurveys\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its annual firm detail supplements the main dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005486:12:0", "occurrence_id": "fdfedfe67da10cfdbaee8578", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: country size using GDP from the Penn World Tables and the real exchange rate. We use the bilateral\n\n\nreal exchange rate between France and other countries using producer prices of France and importer\n\n\ncountries from the International Financial Statistics (IFS) of the IMF. Finally, to compute the proxy for\n\n\nforeign competition at the extensive and intensive margin, we use the BACI dataset at the HS6 product\n\n\nlevel by country of origin. <sup>14</sup>\n\n\nIn order to keep a constant sample throughout the paper and to establish the stability of the point es\n\ntimates, we only consider firms that have information on all control variables. In the main specifications,\n\n\nthis leaves us with around 4,900 firm-country pairs for the 9 Asian countries in the period 1999-2005, a\n\n\ntotal of 34,327 observations.\n\n\nTo perform the robustness check exercises, we use two additional firm level datasets. First, to identify\n\n\nFrench multinational firms, we match our main dataset with firm level dataset on multinational groups\n\n\nlocated in France from the _Enquete Echanges Internationaux Intra-Groupe_ produced by the French Of\n\nfice of Industrial Studies and Statistics (SESSI). These data provide a good representation of the activity\n\n\nof international groups located in France. They account for around 82 percent of total trade flows by\n\n\nmultinationals, and for 55 and 61 percent of total French imports and exports respectively. Second, to\n\n\nadd information on firms’ characteristics, we merge our main dataset with the Annual French Business\n\n\nSurveys (EAE), available from INSEE. The EAE survey is conducted every year and provides detailed\n\n\nfirm-level information for all French firms with more than 20 employees whose main activity is in manu\n\nfacturing. This survey allows us to have information on firms’ employment and labor productivity (value\n\n\nadded per worker).\n\n##### _Market Access: firm level tariff measures_\n\n\nTo identify the impact of Asian’s\n\nMention: \"INSEE\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The agency from which the annual business survey is obtained.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005509:27:0", "occurrence_id": "13bc69fbe01e03f3028cc9d5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 2\nPanel A Panel B\n\n\n\n% Employment in Agriculture vs. GDP per capita, 2008\nSri Lanka + Region Trends & Comparator Countries\n\n\n\n% Employment in Manufacturing vs. GDP per capita, 2008\nSri Lanka + Region Trends & Comparator Countries\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n|Col1|Col2|Col3|Col4|Col5|Col6|Col7|Col8|Col9|\n|---|---|---|---|---|---|---|---|---|\n|<br>IN - 1983<br>NP - 2008|||||||||\n|BGD - 1984<br>BGD - 2008<br>IN - 2008<br>PK - 1980<br>PK - 2008<br>SL - 1981|Thailan|d|||||||\n|Philippin<br> <br>SL - 2008|Brazil<br>es<br>||||||||\n|||Mal|aysia||Kor|eaRep<br>Uni|t~~e~~dK<br>U|ngdom<br>nitedStates|\n|||||||Singapo<br>|re<br><br>||\n\n\n.5 5 10 15 20 25 30 40 50\nGDP per capita, 000s (in constant 2005 international $ at PPP)\n\nData source: ILO (KILM 3rd & 4th eds. & LABORSTA) & World Bank World Development Indicators, 2010. India data from National Sample Survey.\n\n2008 employment share regressed on log(2008 GDP per capita) & log(2008 GDP per capita) squared. Excludes ECA countries in transition.\n\n\nPanel C\n\n\n% Employment in Services vs. GDP per capita, 2008\nSri Lanka + Region Trends & Comparator Countries\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Employment shares across regions come from this compilation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005509:27:0", "occurrence_id": "41b9e83139ef68da1c6d30de", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 2\nPanel A Panel B\n\n\n\n% Employment in Agriculture vs. GDP per capita, 2008\nSri Lanka + Region Trends & Comparator Countries\n\n\n\n% Employment in Manufacturing vs. GDP per capita, 2008\nSri Lanka + Region Trends & Comparator Countries\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n|Col1|Col2|Col3|Col4|Col5|Col6|Col7|Col8|Col9|\n|---|---|---|---|---|---|---|---|---|\n|<br>IN - 1983<br>NP - 2008|||||||||\n|BGD - 1984<br>BGD - 2008<br>IN - 2008<br>PK - 1980<br>PK - 2008<br>SL - 1981|Thailan|d|||||||\n|Philippin<br> <br>SL - 2008|Brazil<br>es<br>||||||||\n|||Mal|aysia||Kor|eaRep<br>Uni|t~~e~~dK<br>U|ngdom<br>nitedStates|\n|||||||Singapo<br>|re<br><br>||\n\n\n.5 5 10 15 20 25 30 40 50\nGDP per capita, 000s (in constant 2005 international $ at PPP)\n\nData source: ILO (KILM 3rd & 4th eds. & LABORSTA) & World Bank World Development Indicators, 2010. India data from National Sample Survey.\n\n2008 employment share regressed on log(2008 GDP per capita) & log(2008 GDP per capita) squared. Excludes ECA countries in transition.\n\n\nPanel C\n\n\n% Employment in Services vs. GDP per capita, 2008\nSri Lanka + Region Trends & Comparator Countries\n\nMention: \"National Sample Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The country series for one large economy comes from this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005517:3:0", "occurrence_id": "ec801e944bdb139b3b262c37", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: There is a rapidly growing body of literature on the economics of the mass media. This\n\n\nliterature has led to a series of important new hypotheses and insights in an area that has\n\n\nbeen long neglected by economists. <sup>2</sup> An important strand of this literature concerns the\n\n\nrole of mass media in political markets and its effect on public policy making. Most of\n\n\nthis literature on the relationship between the mass media and public policy is theoretical.\n\n\nA few empirical studies have attempted to assess the effect of media on policy outcomes.\n\n\nSome key findings from this literature suggest that access to mass media empowers\n\n\npeople politically and, as such, increases the benefits they receive from government\n\n\nprograms (Strömberg and Snyder 2008). This influence has been found for different types\n\n\nof government programs and different countries, such as unemployment relief in the\n\n\nUnited States (Strömberg 2004b), public food provision and disaster relief in India\n\n\n(Besley and Burgess 2001, 2002), and educational spending in Uganda and Madagascar\n\n\n(Reinikka and Svensson 2005; Francken et al. 2009). All of these studies measure the\n\n\neffect within a single country, which has the benefit of holding many other factors fixed\n\n\nbut has the potential disadvantage of limited variation in policy and media.\n\n\nOur paper seeks to contribute to this empirical literature by analyzing the impact\n\n\nof mass media on policy making for a specific type of policy across a wide variety of\n\n\ncountries and years. We use a new dataset from the World Bank that includes measures\n\n\nof agricultural subsidization and taxation for a much wider set of countries and a longer\n\n\nperiod of time than has been previously available (Anderson and Valenzuela 2008). We\n\n\nuse these data as the dependent variable.\n\n\n2 See McCluskey and Swinnen (2010) and Pratt and Strömberg (2011) for\n\n\nreviews.\n\n\n2\n\nMention: \"a new dataset from the World Bank\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Subsidisation and taxation measures across many countries and years.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005520:8:0", "occurrence_id": "bfde7a670f96c7797406996f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: negative impact on the labor market as it still employs a third of the domestic labor force, and\nprovides at least partial subsistence to about half the population (Prohnitchi and Oprunenco, 2010).\n\n\nDrawing on the MAMS database, the economic structure of the Moldovan economy in 2008 is\nsummarized in Table 1. Primary exports represent only a moderate fraction of the total primary\nproduction while a quarter of the domestic demand for primary products is satisfied by imports.\nTherefore, the bulk of agricultural products are produced for the home market. This has important\nimplications for the impact of changes in the agro-industrial sphere on the broader economy.\n\n\nMoldova lacks own energy sources and imports all of its supplies of petroleum, coal, and natural gas,\nlargely from Russia. It has been estimated that, by 2020, Moldova will only be able to meet around\n10-12 percent of its energy needs from own (renewable) energy sources (Prohnitchi and Oprunenco,\n2010). The availability and price of energy is thus a major factor causing uncertainty about the future\neconomic development. In Table 1, fuel is part of an aggregate that also includes chemicals and fueland chemical-based products); for this broader sector, 80 percent of domestic demand is satisfied by\nimports.\n\n\nThe value added and employment shares indicate that private services have the highest level of value\nadded per employee, while the opposite is true for agriculture. This relatively low level of labor\nproductivity of is in part due to the fact that Moldova, despite its fertile soil and favorable climate,\nhas not been able to reach yields that are on par with levels reached in EU or Central and Eastern\nEuropean (CEE) countries (Prohnitchi and Oprunenco, 2010). Given this, there should be considerable\nscope for raising agricultural productivity.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n|Table 1. Structure of production|in 2008|Col3|Col4|Col5|Col6|Col\n\nMention: \"the MAMS database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The economic structure of one economy in that year is summarised from it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005533:6:0", "occurrence_id": "976a89e62f94913b38c3102b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ###### **2.2. Data and Sampling**\n\nData for variables in (1) come from different sources. _Budget Balance_ data are from the\n\nWorld Bank. Time series end in 2004 when change in fiscal accounting methods and reporting\n\nwas completed at the International Monetary Fund. Data for _Elections_ are from the International\n\nFederation of Electoral Systems and from Block and Vaaler (2004). Election years are adjusted\n\nfor government fiscal years consistent with Hanusch and Keefer (2011). _Ratings_ data are from\n\n_Bloomberg_, and are measured annually as an average of up to three ratings published by\n\nMoody’s, S&P and or Fitch (or CRAs merged into Fitch in the 1990s) on January 1. These raters\n\ndominate the industry, including sovereign rating segments, with the broadest country coverage,\n\nthe most analysts and the largest market shares of ratings for individual bond issuances (White,\n\n2010). Their ratings tend to move together with financial market players following them as a\n\ngroup. Data for other controls include the World Bank’s World Development Indicators\n\n( _Employment, Inflation, Reserves, Curr Crisis_ ) and Database of Political Institutions (DPI) (Beck\n\net al _._, 2004) ( _Checks_ ), and Freedom House ( _Rights_ ).\n\n\nWe sample from 18 prominent emerging democracies including: Argentina, Bolivia,\n\nBrazil, Bulgaria, Chile, Colombia, Ecuador, Indonesia, Mexico, Paraguay, Peru, Philippines,\n\nPoland, Russia, South Africa, South Korea, Uruguay, and Venezuela. They uniquely meet three\n\nsampling requirements. Starting in different years from 1989 and the fall of the Berlin Wall to\n\n1998 and the fall of the Suharto regime in Indonesia, these countries have: 1) competitive\n\npresidential executive electoral systems that are new (e.g., Russia) or recently reformed (e.g.,\n\nBrazil)\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Employment, inflation, reserves and crisis dummies come from this compilation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005533:6:0", "occurrence_id": "73024c275f4a3841b248f960", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ###### **2.2. Data and Sampling**\n\nData for variables in (1) come from different sources. _Budget Balance_ data are from the\n\nWorld Bank. Time series end in 2004 when change in fiscal accounting methods and reporting\n\nwas completed at the International Monetary Fund. Data for _Elections_ are from the International\n\nFederation of Electoral Systems and from Block and Vaaler (2004). Election years are adjusted\n\nfor government fiscal years consistent with Hanusch and Keefer (2011). _Ratings_ data are from\n\n_Bloomberg_, and are measured annually as an average of up to three ratings published by\n\nMoody’s, S&P and or Fitch (or CRAs merged into Fitch in the 1990s) on January 1. These raters\n\ndominate the industry, including sovereign rating segments, with the broadest country coverage,\n\nthe most analysts and the largest market shares of ratings for individual bond issuances (White,\n\n2010). Their ratings tend to move together with financial market players following them as a\n\ngroup. Data for other controls include the World Bank’s World Development Indicators\n\n( _Employment, Inflation, Reserves, Curr Crisis_ ) and Database of Political Institutions (DPI) (Beck\n\net al _._, 2004) ( _Checks_ ), and Freedom House ( _Rights_ ).\n\n\nWe sample from 18 prominent emerging democracies including: Argentina, Bolivia,\n\nBrazil, Bulgaria, Chile, Colombia, Ecuador, Indonesia, Mexico, Paraguay, Peru, Philippines,\n\nPoland, Russia, South Africa, South Korea, Uruguay, and Venezuela. They uniquely meet three\n\nsampling requirements. Starting in different years from 1989 and the fall of the Berlin Wall to\n\n1998 and the fall of the Suharto regime in Indonesia, these countries have: 1) competitive\n\npresidential executive electoral systems that are new (e.g., Russia) or recently reformed (e.g.,\n\nBrazil)\n\nMention: \"Database of Political Institutions\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its checks-and-balances variable is one of the institutional controls.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005537:44:8", "occurrence_id": "316c127fa02c81a371151902", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: -0.052</u>**\n\n<u>Uruguay</u> <u>-0.006</u> <u>0.015</u> <u>0.021</u>\n\n<u>-0.001</u> <u>-0.034</u> <u>-0.017</u>\n\n\n\n**0.031**\n\n**<u>LAC</u>** **<u>-0.052</u>**\n\n<u>Uruguay</u> <u>-0.006</u> <u>0.015</u> <u>0.021</u>\n\n<u>-0.001</u> <u>-0.034</u> <u>-0.017</u>\n~~Source: Author’s calculations with data from SEDLAC (CEDLAS and The World Bank) using labor population-weighted~~\n~~averages.~~\n\n\n\n43\n\nMention: \"SEDLAC (CEDLAS and The World Bank)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Labour population-weighted averages are computed from this distributional database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005538:16:1", "occurrence_id": "0e5fb38b050dbb203c65be5e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: > e Minas<br> i|G<br>Ipating<br>Juiz de Fora|\n\n\nLavras\n\n\n\nDivinópolis\n\n\n\n\n\nSource: Author’s calculations using data from Peixoto et al (2010)\n\n\nFigure 6 represents a simple way to present comparison for many police districts simultaneously. This\ntype of figure is useful when the number of districts is limited, but would (obviously) not be useful for\n\n\n15\n\nMention: \"Peixoto et al (2010)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Police district comparisons are calculated from these authors' data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005540:15:1", "occurrence_id": "12c831f26b496ef63c610cce", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 60<br>80<br>100<br>120<br>140<br>160<br>T<br>1<br>2<br>3<br>4<br>5<br>Finland (91)<br>Spain (08)<br>Sweden (91)<br>UK (08)<br>USA (07)<br>Euro Area (08)|\n|<br>**_III.3_ Many European banks will deleverage via run-**<br>**offs, divestments, and constraining asset growth…**<br>|<br>**_III.4_ … and will mainly target loan portfolios in their**<br>**divestment strategies**<br>|\n|<br>How European banks plan to deleverage<br>_% of respondents rating these items 4 or 5 on a 1-5 scale_|<br>Which assets European banks are likely to divest<br>_% of respondents rating these items 4 or 5 on a 1-5 scale_|\n|<br> <br>Source: Deloitte Bank Survey 2012 <br> <br>0%<br>20%<br>40%<br>60%<br>**Natural run-off**<br>**Divestments**<br>**Constraining asset growth**<br>**Write-offs/provisions**|<br>Source: Deloitte Bank Survey 2012<br>0%<br>20%<br>40%<br>60%<br>**Loan portfolios**<br>**Securities and trading**<br>**portfolios**<br>**Bank subsidiaries and**\n\nMention: \"Deloitte Bank Survey 2012\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Respondent ratings of banking pressures come from this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005485:6:1", "occurrence_id": "11aca465a1d7bfe6a22075f1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: NSECURITY**\n\n\n**INDICATOR**\n\n\nIn this section, we provide an overview of the GWP and the specific food security indicator\n\n\nused in this study. Our goal is limited to answering three questions. First, what is the general\n\n\nquality of the GWP surveys? Second, what limitations might the GWP indicator of self-assessed\n\n\n5\n\nMention: \"GWP surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"The study's object is the general quality of these surveys' indicator.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005525:2:0", "occurrence_id": "7b938eb4a7fd5518beea1591", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ### **Gender and Agriculture: Inefficiencies, Segregation, and** **Low Productivity Traps 1**\n\n##### **Andre Croppenstedt, Markus Goldstein and Nina Rosas †**\n\n\n\n1 This paper was prepared as a background paper for the World Development Report 2012. It has\n\nbenefitted from numerous conversations with the core team, especially Carolina Sanchez-Paramo, Ana\n\n\n\n1 This paper was prepared as a background paper for the World Development Report 2012. It has\n\nbenefitted from numerous conversations with the core team, especially Carolina Sanchez-Paramo, Ana\n\nRevenga, and Sudhir Shetty, as well as numerous colleagues at the World Bank and the FAO. In addition, this\n\n\n\nbenefitted from numerous conversations with the core team, especially Carolina Sanchez-Paramo, Ana\n\nRevenga, and Sudhir Shetty, as well as numerous colleagues at the World Bank and the FAO. In addition, this\n\npaper draws on work performed for the FAO’s State of Food and Agriculture 2010–2011. We are grateful to\n\n\n\nRevenga, and Sudhir Shetty, as well as numerous colleagues at the World Bank and the FAO. In addition, this\n\npaper draws on work performed for the FAO’s State of Food and Agriculture 2010–2011. We are grateful to\n\nKatia Covarrubias (FAO) for her support with the Rural Income Generating Activities data.\n\n\n\n1\n\n\n\npaper draws on work performed for the FAO’s State of Food and Agriculture 2010–2011. We are grateful to\n\nKatia Covarrubias (FAO) for her support with the Rural Income Generating Activities data.\n\nMention: \"the Rural Income Generating Activities data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Acknowledged support in handling this rural livelihood dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005547:33:0", "occurrence_id": "1da9cd17bfc8a98b2f294b09", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **DHS 2011**\n\nSurvey period in 2011\n\nJan    Feb    March    April    May   June   July    Aug     Sept    Oct\nNov     Dec\n\nMonsoon\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n|Cohort|Year of<br>Birth|Month of<br>Birth|Rainfall shock experienced in the….|Col5|Col6|Col7|Col8|\n|---|---|---|---|---|---|---|---|\n|Cohort<br>|Year of<br>Birth<br>|Month of<br>Birth<br>|<br>..1~~st~~ year<br>|<br>..2~~nd~~ year<br>|<br>..3~~rd~~ year<br>|<br>  ..4~~th~~ <br>year<br>|<br>..5~~th~~ year<br>|\n|5 <br>|2010<br>|Jan. –<br>Sep.<br>|Rain<br>2010<br>|<br>|<br>|<br> <br>|<br>|\n|5 <br>|2009<br>|<br>Sept. –<br>Dec.<br>|<br>Rain<br>2010<br>|<br>|<br>|<br>|<br>|\n|4 <br>|2009<br>|<br>Jan. –<br>Sep.<br>|<br>Rain<br>2009<br>|Rain 2010<br>|<br>|<br>|<br>|\n|4 <br>|2008<br>|<br>Sept.\n\nMention: \"DHS 2011\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"Fieldwork periods across the calendar year are scheduled for this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005549:46:44", "occurrence_id": "e5328d2f101ff97dedb9cdd6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Swaziland, and Zambia.|<br>If the table shows only one row for a country, it is equivalent to the bottom row for countries with two rows and means that no years were designated as<br>intensity 3 in the PRIO/Uppsala data.<br>African countries for which the data set reports no year of “internal armed conflict” or “internationalized internal armed conflict” of intensity 1, 2, or 3<br>between 1960 and 2004 are Benin, Malawi, Swaziland, and Zambia.|<br>If the table shows only one row for a country, it is equivalent to the bottom row for countries with two rows and means that no years were designated as<br>intensity 3 in the PRIO/Uppsala data.<br>African countries for which the data set reports no year of “internal armed conflict” or “internationalized internal armed conflict” of intensity 1, 2, or 3<br>between 1960 and 2004 are Benin, Malawi, Swaziland, and Zambia.|<br>If the table shows only one row for a country, it is equivalent to the bottom row for countries with two rows and means that no years were designated as<br>intensity 3 in the PRIO/Uppsala data.<br>African countries for which the data set reports no year of “internal armed conflict” or “internationalized internal armed conflict” of intensity 1, 2, or 3<br>between 1960 and 2004 are Benin, Malawi, Swaziland, and Zambia.|<br>If the table shows only one row for a country, it is equivalent to the bottom row for countries with two rows and means that no years were designated as<br>intensity 3 in the PRIO/Uppsala data.<br>African countries for which the data set reports no year of “internal armed conflict” or “internationalized\n\nMention: \"PRIO/Uppsala data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its conflict intensity coding leaves some countries without a year designated.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005550:7:0", "occurrence_id": "376949e3c9d09880b8f9dfe5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: #### **I Data and indicators**\n\n###### **Trade data sources**\n\nThe data used in this study come from Chinese Customs and provide export flows aggregated by\n\n\nprovince, year, product, and destination country over the 1997-2007 period. <sup>1</sup> We reaggregate\n\n\nthe original eight-digit-level data into HS4-level data (more than 1,200 product lines). An\n\n\ninteresting feature of this dataset is that it allows us to identify whether the export flows\n\n\nemanate from domestic or foreign firms <sup>2</sup> and whether they correspond to processing trade or\n\n\nto ordinary trade. <sup>3</sup> The processing trade includes all trade flows by firms operating in the\n\n\nassembly sector; that is, firms that import and process inputs in China and then re-export\n\n\nthe final products abroad. Firms engaged in this type of activity might be less embedded in\n\n\ntheir local environment and might consequently generate fewer (and possibly benefit less from)\n\n\nexternalities.\n\n###### **Explained variable: creation of new export linkages**\n\n\nThe creation of a new export transaction is measured by a dummy that takes the value 1 if\n\n\nthe domestic firms in province _i_ begin exporting product _k_ to country _j_ at time _t_ + 1 and 0\n\n\notherwise. A specific database is constructed that incorporates the set of alternatives faced by\n\n\n1We do not have firm-level data, but we believe that province/firm-type/trade-type/product/destination\ncountry data are suitable for the investigation of micro-phenomena such as export spillovers. The information\nthat we have is very detailed. Feenstra and Hanson (2005) argue, for example, that their city/firm-type/tradetype/product/destination country dataset approaches the precision of a firm-\n\nMention: \"Chinese Customs\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Export flows aggregated by province, product and destination come from this agency.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005551:14:2", "occurrence_id": "aae0e20970425c0b1a930252", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  sex ratio, and the literacy\nratio are all statistically similar across treatment assignment. Similarly, the sex ratio of workers\nand non-workers also does not vary between treated and untreated blocks. Thus, these t-tests also\nfail to yield evidence that would cause us to believe that the program was systematically targeted\nat villages where women were either less empowered or already more empowered, and hence more\nlikely to respond to the treatment.\n\n\n10Blocks are small groups of villages and are the administrative unit directly above the village council. Within each\ndistrict, the placement decision of _Mahila_ _Samakhya_ is made by block, rather than by village, hence these blocks are\nthe appropriate unit of analysis when examining whether the program is targeted in any way. For blocks without\nthe program, we use data from the 2001 census.\n\n\n13\n\nMention: \"the 2001 census\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"For areas without the programme, block characteristics come from this enumeration.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005553:46:1", "occurrence_id": "77a337ec2a5c588fe072fedb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: mark> <mark>5.24%</mark> <mark>***</mark>\n\n\n\n**<mark>No. of Firms (Orbis database)</mark>** 1,915\n\n\n**Investment**\n\n\n\n425 195\n\n\n\nCapital Expenditures 6,628 16,233 *** 52,840 ***\nCapital Expenditures/Sales 5.84% 8.13% *** 10.22% ***\n\n\n\n**No. of Firms (Worldscope database)** 2,235\n\n\n**Panel B. India**\n\n\n**<u>Firm Characteristics</u>** **<u>No Issues</u>**\n**Size**\n\n\n\n615 224\n\n\n**<u>Firms with</u>**\n\n**<u>Equity Issues</u>** **<u>Bond Issues</u>**\n\n\n\nTotal Assets 8,967 55,633 *** 596,587 ***\nSales 9,521 41,918 *** 272,083 ***\nEmployees 837 2,700 *** 4,600 ***\n\n\n**Growth**\n\nTotal Assets Growth 5.24% 15.77% *** 18.18% ***\nSales Growth 9.84% 17.33% *** 17.60% ***\nEmployee Growth 0.96% 3.62% *** 4.35% ***\n\n\n**Capital Structure and Financial Health**\n\nLong-Term Debt/Total Liabilities 49.49% 56.76% *** 62.95% ***\nLeverage 52.46% 62.17% *** 64.39% ***\nRetained Earnings/Total Assets 3.69% 4.97% *** 7.08% ***\n\n\n**Profit\n\nMention: \"Worldscope database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Firm counts and capital expenditure ratios come from this corporate database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005555:3:0", "occurrence_id": "388017c526346a3066f2b9a5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ###### **1. Introduction**\n\nThe economic integration of developing countries into world markets is an important stepping\n\nstone for economic transitions and growth. This integration can be quite challenging, however,\n\ndue to the many differences across countries in language, cultural understanding, legal\n\nregulations, and much more. As a consequence, business and social networks are valuable\n\nmechanisms for achieving this integration (Rauch 2001). Ethnicity-based interactions and\n\ndiaspora connections are a prominent form of these networks. The benefits typically cited for\n\ndiaspora networks include stronger access to information (especially very recent or tacit\n\nknowledge), matching and referral services that link firms together, language skills and cultural\n\nsensitivity that improve interactions, and repeated relationships that embed trust in uncertain\n\nenvironments and provide sanction mechanisms for misbehavior. Such traits are hard to\n\nconstruct yet crucial for business success in many developed and emerging economies. The\n\nhistory of these connections stretches back to the earliest of international exchanges (e.g., Aubet\n\n2001), and studies continue to find diasporas important for trade flows, foreign investments, and\n\nknowledge diffusion.\n\nOver the last two decades, the Internet has become a potent force for global economic\n\nexchanges. The Internet links customers and companies together worldwide, enables labor to be\n\nprovided at a distance, provides instant access to information about foreign locations, and much\n\nmore. How will the Internet affect the importance of diaspora networks? On one hand, the\n\nsubstantial improvements in connectivity and reduced frictions of the Internet may reduce the\n\nimportance of diaspora connections. On the other hand, the online capabilities may provide an\n\neffective tool that complements traditional diaspora connections (e.g., Saxenian 2006). To shed\n\nlight on these important questions, we investigate the role of the Indian diaspora in online\n\noutsourcing to India using data from oDesk. oDesk is the world’s largest online labor market,\n\nprocessing $30 million per month in contracts as of May 2012. It provides a platform for\n\nMention: \"data from oDesk\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The largest online labour market, processing thirty million monthly in contracts.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005563:44:3", "occurrence_id": "fd7b0778f39300ad87c34d1a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  of the five regions (north, north\nwest, center, east, and south) from individual employment data in NSS61-EUS. Comparable\n\ndata on hired labor expenses (for regular and casual farm workers), _Wℓh,_ and on total value\n\nof crop production are available at the farm-level by season from NSS59. Summing up _Wℓh_\n\nacross seasons and households within each region (using sampling weights) multiplying by\n\n(1 + _f_ ) and dividing by a similarly computed sum of production value gives the regional labor shares. I use the same approach for the intermediate input shares _αI_ = _PIIA/_ <sup>�</sup> _j_ <sup>_PjYj_</sup> <sup>,</sup>\n\n\n\nbor shares. I use the same approach for the intermediate input shares _αI_ = _PIIA/_ <sup>�</sup> _j_ <sup>_PjYj_</sup> <sup>,</sup>\n\nwhere the numerator is the total expenditures on non-labor variable inputs as reported in\n\nNSS59 (seed, fertilizer, pesticide, and irrigation). The following table reports the results of\n\n\n\n42\n\nMention: \"NSS61-EUS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Individual employment across five regions comes from this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005563:44:3", "occurrence_id": "2d0fc5c9f65bdf0b05cbe7e0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  of the five regions (north, north\nwest, center, east, and south) from individual employment data in NSS61-EUS. Comparable\n\ndata on hired labor expenses (for regular and casual farm workers), _Wℓh,_ and on total value\n\nof crop production are available at the farm-level by season from NSS59. Summing up _Wℓh_\n\nacross seasons and households within each region (using sampling weights) multiplying by\n\n(1 + _f_ ) and dividing by a similarly computed sum of production value gives the regional labor shares. I use the same approach for the intermediate input shares _αI_ = _PIIA/_ <sup>�</sup> _j_ <sup>_PjYj_</sup> <sup>,</sup>\n\n\n\nbor shares. I use the same approach for the intermediate input shares _αI_ = _PIIA/_ <sup>�</sup> _j_ <sup>_PjYj_</sup> <sup>,</sup>\n\nwhere the numerator is the total expenditures on non-labor variable inputs as reported in\n\nNSS59 (seed, fertilizer, pesticide, and irrigation). The following table reports the results of\n\n\n\n42\n\nMention: \"NSS59\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Farm-level hired labour expenses and crop output by season come from this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005572:23:1", "occurrence_id": "eb8a97f315bf9ba80e9ab0ae", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  in turn reflect the changes in both hourly wages and hours worked. The\n\n\naverage weekly wage growth of 23 percent is slightly higher than the average hourly wage\n\n\n40 For changes in ownership categories within the formal sectors, the changes may reflect either the change in\nnationality of the establishment for which an individual works or individual’s move from one sector to another.\n41 The VHLSS is likely to exclude migrants who hold a temporary or no registration status since the sampling of the\nsurveys is limited to the households who had permanent registration status. Based on the _Vietnam Migration Survey_\n(VMS) data in 2004, Fukase (2013b) finds that only 10 percent of migrants who work for foreign firms possess\npermanent registration status.\n42 The wages are converted to January 2002 constant prices using the deflators available in the VHLSS.\n\n22\n\nMention: \"VHLSS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Its sampling frame may exclude migrants without registration.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005574:20:1", "occurrence_id": "9f841378d21397845692e004", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  1.3 (dropping from 7.5 to\n\n\n6.6). However, PFM quality remains significant at the .001 level in equation 2.2.\n\n\nEquation 2.3 adds the Transparency indicator from the World Bank’s CPIA.\n\n\nTransparency and PFM Quality are correlated at .59 in our sample. The latter’s coefficient\n\n\ndeclines from 7.5 in equation 1.3 to 5.5 in equation 2.3, but is still significant at the .05 level.\n\n\n19\n\nMention: \"CPIA\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its transparency indicator enters the extended specification.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005575:11:3", "occurrence_id": "57403bd6b61c95740cf071b4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 5), Ghana (16), Kenya\n\n\n\nThe focus of this section is on foreign investors, but we also compare their characteristics with\n\ndomestic producers. The foreign investors’ surveys cover 87 firms in Chile (5), Ghana (16), Kenya\n\n(20), Lesotho (15), Mozambique (10), Swaziland (11) and Vietnam (10). Table 1 shows that the\n\n\n\ndomestic producers. The foreign investors’ surveys cover 87 firms in Chile (5), Ghana (16), Kenya\n\n(20), Lesotho (15), Mozambique (10), Swaziland (11) and Vietnam (10). Table 1 shows that the\n\nmajority of foreign investors are in apparel (43), followed by agribusiness (30) and mining (14).\n\n\n\n(20), Lesotho (15), Mozambique (10), Swaziland (11) and Vietnam (10). Table 1 shows that the\n\nmajority of foreign investors are in apparel (43), followed by agribusiness (30) and mining (14).\n\nDomestic producers’ surveys cover 64 firms in Chile (5), Ghana (10), Kenya (26), Mozambique (6)\n\n\n\nmajority of foreign investors are in apparel (43), followed by agribusiness (30) and mining (14).\n\nDomestic producers’ surveys cover 64 firms in Chile (5), Ghana (10), Kenya (26), Mozambique (6)\n\nand Vietnam (17). The majority of these firms are in agribusiness (46), followed by apparel (13)\n\n\n\nDomestic producers’ surveys cover 64 firms in Chile (5), Ghana (10), Kenya (26), Mozambique (6)\n\nand Vietnam (17). The majority of these firms are in agribusiness (46), followed by apparel (13)\n\nand mining (5).\n\n\n\nand mining (5).\n\n\n\n10\n\nMention: \"foreign investors’ surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Eighty-seven firms across eight countries are covered in these surveys.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005575:11:3", "occurrence_id": "df15a2b9550e327b07c41468", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 5), Ghana (16), Kenya\n\n\n\nThe focus of this section is on foreign investors, but we also compare their characteristics with\n\ndomestic producers. The foreign investors’ surveys cover 87 firms in Chile (5), Ghana (16), Kenya\n\n(20), Lesotho (15), Mozambique (10), Swaziland (11) and Vietnam (10). Table 1 shows that the\n\n\n\ndomestic producers. The foreign investors’ surveys cover 87 firms in Chile (5), Ghana (16), Kenya\n\n(20), Lesotho (15), Mozambique (10), Swaziland (11) and Vietnam (10). Table 1 shows that the\n\nmajority of foreign investors are in apparel (43), followed by agribusiness (30) and mining (14).\n\n\n\n(20), Lesotho (15), Mozambique (10), Swaziland (11) and Vietnam (10). Table 1 shows that the\n\nmajority of foreign investors are in apparel (43), followed by agribusiness (30) and mining (14).\n\nDomestic producers’ surveys cover 64 firms in Chile (5), Ghana (10), Kenya (26), Mozambique (6)\n\n\n\nmajority of foreign investors are in apparel (43), followed by agribusiness (30) and mining (14).\n\nDomestic producers’ surveys cover 64 firms in Chile (5), Ghana (10), Kenya (26), Mozambique (6)\n\nand Vietnam (17). The majority of these firms are in agribusiness (46), followed by apparel (13)\n\n\n\nDomestic producers’ surveys cover 64 firms in Chile (5), Ghana (10), Kenya (26), Mozambique (6)\n\nand Vietnam (17). The majority of these firms are in agribusiness (46), followed by apparel (13)\n\nand mining (5).\n\n\n\nand mining (5).\n\n\n\n10\n\nMention: \"Domestic producers’ surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Sixty-four domestic firms are covered for comparison.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005582:19:0", "occurrence_id": "b7f6eea8a7b0c7f45983e1ce", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Last but not least, formality is positively associated with better access to formal sources of financing at\nstart-up. Columns 3-4 of Table 2 show that, controlling for other factors, more formal firms are more\nlikely to have received some portion of their initial capital from formal sources, which are defined as\nprivate and public banks and micro-finance organizations. <sup>20</sup> [^20: Public banks include Banco Nacional de Fomento and Corporación Financiera Nacional. The micro-finance\ncategory also includes financing received through Credito de Desarrollo Humano (CDH).] The EMES data on access to credit shows\nthat formal sources of financing tend to offer credit on better terms, which in turn positively affects the\nprofitability of firms who are able to access these sources. Clearly, these results fall short of establishing\ncausality between formality and access to finance, because we do not observe the formality status of firms\nat start-up. <sup>21</sup> [^21: Knowing whether a firm received a loan from a formal source in the last year does not resolve the issue because,\nfor example, more formal firms could have a lower demand for loans due to their higher profitability which could\nallow them to use retained earnings instead of credit.] However, as shown in Figure 10, firms who complied with regulatory requirements from the\nstart do have a higher likelihood of having benefitted from formal sources of financing; on the other hand,\ndue to a relatively small number of firms who actually obtained credit from formal sources, the standard\nerrors are too large for the differences to be statistically significant. Therefore, the evidence on a positive\nlink between formality and improved access to credit in Table 2 and Figure 10 is suggestive, but not\nconclusive.\n\n\n**Figure 10: Formal financing and time of registration**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n_Source_ : Authors’ calculations with EMES data.\n\n##### **6. Conclusions**\n\nIn order to understand the links between formality and firm profitability, this paper relies on a new survey\n\nMention: \"EMES data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Credit access and terms across formal and informal sources are recorded here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005583:7:0", "occurrence_id": "3ae82c67c389ce0187ddf789", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: While national service flows are based on national input-output data, the data\n\n\non services trade flows come from a number of sources. The OECD, Eurostat\n\n\nand the UN all provide data in some form on bilateral services trade flows (both\n\n\nimports and exports), by partners and BOPs (balance of payments sectors) for\n\n\nup to 24 sectors and subsectors. The most comprehensive coverage of reporting\n\n\ncountries among the three sources is UN, which at present count provides data\n\n\non roughly 190 reporters. Eurostat and OECD provide data for a more limited\n\n\nnumber of reporters, though with more sector detail than the UN. Eurostat covers\n\n\n27 EU members plus Croatia, Iceland, Japan, Norway, Turkey, Switzerland, and\n\n\nUSA, while the OECD covers 28 countries (all the OECD members apart from\n\n\nChile, Iceland, Israel, Slovenia, and Switzerland). Time coverage is the deepest\n\n\nwith EUROSTA. IMF data cover almost as many countries as the UN, and for a\n\n\nlonger time span, but only for trade with world. <sup>3</sup> Within GTAP, the efforts to\n\n\nreconcile services balance of payments data with national input-ouput data are well\n\n\ndocumented, and reflect innovative approaches based on information theory and\n\n\nentropy methods to flesh out the pattern of trade enough to allow construction of a\n\n\nglobal dataset for estimating cross-border trade in services. (See MacDougall and\n\n\nHagemejer 2005, and Van Leeuwen and Lejour 2006). For the transport sector,\n\n\nreconciliation of trade data is explicitly linked to the parallel reconciliation of data\n\n\non trade in goods, as services play a critical role in bridging the gap between reported\n\n\nexport flows for goods, in one country’s national accounts data, and reported import\n\n\nflows in another country’s data (Gelhar 1996).\n\n\nThe structure of the data provides a comprehensive and consistent record of\n\n\nnational income accounting relationships between different sectors and regions. It\n\n\nis based on a fundamental, general equilibrium principle\n\nMention: \"IMF data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its coverage is nearly as wide as the UN's but over a longer span.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005583:7:0", "occurrence_id": "7b6278383fc53006677f3329", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: While national service flows are based on national input-output data, the data\n\n\non services trade flows come from a number of sources. The OECD, Eurostat\n\n\nand the UN all provide data in some form on bilateral services trade flows (both\n\n\nimports and exports), by partners and BOPs (balance of payments sectors) for\n\n\nup to 24 sectors and subsectors. The most comprehensive coverage of reporting\n\n\ncountries among the three sources is UN, which at present count provides data\n\n\non roughly 190 reporters. Eurostat and OECD provide data for a more limited\n\n\nnumber of reporters, though with more sector detail than the UN. Eurostat covers\n\n\n27 EU members plus Croatia, Iceland, Japan, Norway, Turkey, Switzerland, and\n\n\nUSA, while the OECD covers 28 countries (all the OECD members apart from\n\n\nChile, Iceland, Israel, Slovenia, and Switzerland). Time coverage is the deepest\n\n\nwith EUROSTA. IMF data cover almost as many countries as the UN, and for a\n\n\nlonger time span, but only for trade with world. <sup>3</sup> Within GTAP, the efforts to\n\n\nreconcile services balance of payments data with national input-ouput data are well\n\n\ndocumented, and reflect innovative approaches based on information theory and\n\n\nentropy methods to flesh out the pattern of trade enough to allow construction of a\n\n\nglobal dataset for estimating cross-border trade in services. (See MacDougall and\n\n\nHagemejer 2005, and Van Leeuwen and Lejour 2006). For the transport sector,\n\n\nreconciliation of trade data is explicitly linked to the parallel reconciliation of data\n\n\non trade in goods, as services play a critical role in bridging the gap between reported\n\n\nexport flows for goods, in one country’s national accounts data, and reported import\n\n\nflows in another country’s data (Gelhar 1996).\n\n\nThe structure of the data provides a comprehensive and consistent record of\n\n\nnational income accounting relationships between different sectors and regions. It\n\n\nis based on a fundamental, general equilibrium principle\n\nMention: \"services balance of payments data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Reconciling these with national accounts is a documented effort.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005584:5:1", "occurrence_id": "879872c96ce13136f6abaacd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  emphasis on the potential for agricultural-policy\nreforms.\n\n\n**2. Stylized Facts**\n\nMany agricultural- and trade-policy developments of the past half-century have happened\nquite suddenly and been transformational. Such events include decolonization in Africa and\nelsewhere around 1960; the creation of the Common Agricultural Policy (CAP) in Europe in\n1962; the introduction of flexible exchange rates from the 1970s; liberalization, deregulation,\nprivatization, and democratization in many countries from the mid-1980s; the opening of\nmarkets in China in 1979, in Vietnam in 1986, and in Eastern Europe (following the fall of\nthe Berlin Wall) in 1989; and the demise of the Soviet Union in 1991. More subtle are the\ninfluences of policies that change gradually in the course of economic development, as\nincomes grow and comparative advantages evolve. But what do the quantitative measures of\ndistortions and reforms reveal?\n\nEmpirical indicators of agricultural price distortions (called Producer Support and\nConsumer Subsidy Estimates, or PSEs and CSEs) have been provided consistently for 25\nyears by the Secretariat of the OECD (2011) for its thirty member countries. However, the\nOECD provides no comprehensive time-series rates of assistance to producers of nonagricultural goods to compare with the PSEs, nor of what took place in those advanced\neconomies during earlier decades. Data for these earlier decades from developed economies\nis needed in order to assess how various countries’ policies evolved during stages of\ndevelopment similar to those of today’s middle-income countries. As for developing\ncountries, almost no comparable time-series estimates were generated in the two decades\n\n\npolicies more thoroughly (while giving somewhat less attention to factor market distortions). And third, we\nplace particular emphasis on the substantial body of theoretical and empirical research that has been published\nduring the past fifteen years (that is, since the most recent citations in Binswanger and Deininger).\n\nMention: \"Producer Support and\nConsumer Subsidy Estimates\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its secretariat has provided price distortion indicators consistently for twenty-five years.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005592:15:0", "occurrence_id": "d2f9e5aa6239d9a40713d8b0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: the comprehensive wealth accounts published in World Bank (2011) – land values as a share of\ntotal wealth.\n\nThe physical indicators are clearly not the explanatory variables we are seeking – protected area\nper capita in developing countries is about 2/3 of the value in OECD countries, while there is\nlittle variation in the protected area share of total land area across the different income classes.\nThe third indicator tells the story. Land values (excluding urban land) make up only 1.4% of the\ntotal wealth of OECD countries, while it is 10 to 20 times higher as a proportion of wealth in\nmiddle and low income countries. Since the opportunity cost of creating a protected area is based\non the rents from agricultural land, this will necessarily be a much larger share of GDP in\ndeveloping countries compared with OECD countries.\n\n**<u>Table 2. Protected area extent and land values by income class, 2005</u>**\n<u>Low income</u> <u>Middle income</u> <u>OECD</u>\n\nProtected area per capita, ha. 0.22 0.21 0.34\n\n\n\nProtected area as a percent\nof total land area\n\n\n\nLand value (cropland,\npastureland, forested land,\nprotected areas) as a percent\n<u>of total wealth</u>\n\n\n\n10.8% 10.6% 11.8%\n\n\n29.5% 12.8% 1.4%\n\n\n\n<u>Source: derived from World Bank (2011) and</u> _<u>World Development Indicators</u>_\n\n_Policy implications, and some caveats_\n\nProtected areas and the biodiversity they harbor provide both local and global benefits. The\ninstitution that finances environmental global public goods is the Global Environment Facility\n(GEF), and it is worth assessing the level of finance that it disburses for biodiversity protection.\n\nAccording to the GEF’s _Annual Report 2010_\n\nMention: \"comprehensive wealth accounts\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Land values as a share of total wealth come from these accounts.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005592:15:0", "occurrence_id": "0d62b893e99c5c7718a628bc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: the comprehensive wealth accounts published in World Bank (2011) – land values as a share of\ntotal wealth.\n\nThe physical indicators are clearly not the explanatory variables we are seeking – protected area\nper capita in developing countries is about 2/3 of the value in OECD countries, while there is\nlittle variation in the protected area share of total land area across the different income classes.\nThe third indicator tells the story. Land values (excluding urban land) make up only 1.4% of the\ntotal wealth of OECD countries, while it is 10 to 20 times higher as a proportion of wealth in\nmiddle and low income countries. Since the opportunity cost of creating a protected area is based\non the rents from agricultural land, this will necessarily be a much larger share of GDP in\ndeveloping countries compared with OECD countries.\n\n**<u>Table 2. Protected area extent and land values by income class, 2005</u>**\n<u>Low income</u> <u>Middle income</u> <u>OECD</u>\n\nProtected area per capita, ha. 0.22 0.21 0.34\n\n\n\nProtected area as a percent\nof total land area\n\n\n\nLand value (cropland,\npastureland, forested land,\nprotected areas) as a percent\n<u>of total wealth</u>\n\n\n\n10.8% 10.6% 11.8%\n\n\n29.5% 12.8% 1.4%\n\n\n\n<u>Source: derived from World Bank (2011) and</u> _<u>World Development Indicators</u>_\n\n_Policy implications, and some caveats_\n\nProtected areas and the biodiversity they harbor provide both local and global benefits. The\ninstitution that finances environmental global public goods is the Global Environment Facility\n(GEF), and it is worth assessing the level of finance that it disburses for biodiversity protection.\n\nAccording to the GEF’s _Annual Report 2010_\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The companion compilation behind the tabulated ratios.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005598:35:0", "occurrence_id": "2cdb3b7c9e7cd8015ac50125", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **<u>Table 3. Job Status and political attitudes. World Value Survey</u>**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n|Col1|High income Higher middle income Low income Lower middle income|\n|---|---|\n|<br>Dependent<br>variable<br> <br>unemp<br> <br>Job qual<br> <br>Observations<br> <br>|<br> <br> <br> <br>Confidence in<br>government<br>Confidence in<br>government<br>Confidence in<br>government<br>Confidence in<br>government<br>Confidence in<br>government<br>Confidence in<br>government<br>Confidence in<br>government<br>Confidence in<br>government<br> <br> <br> <br> <br> <br> <br> <br> <br>-0.007<br> <br>0.018<br> <br>-0.025<br> <br>-0.005<br> <br>(0.018)<br> <br>(0.012)<br> <br>(0.024)<br> <br>(0.015)<br> <br> <br>-0.000<br> <br>0.001<br> <br>-0.002<br> <br>-0.001<br> <br>(0.003)<br> <br>(0.002)<br> <br>(0.007)<br> <br>(0.003)<br>11827<br>9614<br>17534<br>\n\nMention: \"World Value Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Job status and political attitudes are estimated across income groups from it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005600:11:0", "occurrence_id": "1d415aff7c99e55fe9c372d7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: we have inequality estimates) and 2010 increased by more than 16 times. <sup>9</sup> [^9: GDP per capita data, expressed in 1990 Geary-Khamis PPPs, are from Maddison (2007). The physiological\nsubsistence minimum in the same prices is assumed to be $PPP 300 (for discussion see MLW 2011, p. 262).] The figure\n\nshows the large difference between the IERs calculated with subsistence only, and under\n\nthe assumption that the social minimum increases as the mean income goes up. Under the\n\nfirst assumption, the IER in England/UK oscillated around 60 percent throughout 18 <sup>th</sup> and\n\n19 <sup>th</sup> century. But if we assume that the social minimum had risen in proportion with real\n\nincome (b=0.5), the IERs attained 80 percent. The numbers for the United States are very\n\nsimilar.\n\n\nThe political implications of b=0 and b=0.5 are very different. If the elites had to\n\nacquiesce to a rise of the minimum living standard for the masses, then an 80% inequality\n\nextraction ratio shows that they were able or willing to push overall inequality rather close\n\nto the maximum. If, on the other hand, we believe that there was no social pressure to\n\nincrease the social minimum, then the elites’ appetites may be thought to have been\n\nrelatively modest.\n\n\nIn both countries, it is only in the 20 <sup>th</sup> century, as illustrated more clearly in Figure 5,\n\nthat the IER, calculated under the more stringent conditions of a changing social minimum,\n\nbegan its downward slide, to be arrested and reversed in the last quarter of the 20 <sup>th</sup> century\n\nand early 21 <sup>st</sup> . But because real incomes have by then risen to very high levels (exceeding,\n\nin the United States, 100 times the subsistence minimum), the inequality possibility\n\nMention: \"1990 Geary-Khamis PPPs\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Output per head and the subsistence minimum are both expressed in these prices.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005602:3:1", "occurrence_id": "c421de04e0578f66befcd0b5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  look at the transformation of India’s\ncountryside currently underway. We provide detailed evidence covering the period between 1983\nand 2009/10, documenting a process of diversification out of agriculture that is slow but\naccelerating. <sup>3</sup> [^3: We are concerned in this paper with the rural non-farm sector defined in terms of all rural economic activities\noutside of agriculture. Non-farm activity may occur in the home or in factories or be performed by mobile traders.\nOur definition can be contrasted with a broader notion of “off-farm” activity which also considers economic\nactivities that occur “off of the owner’s own farm” but that could for example include wage-employment in\nagriculture earned on a neighbor’s farm, or the renting out of the services of a tractor to neighboring farmer,  along\nwith nonfarm earnings from the owner’s nonfarm enterprises or from nonfarm wage earnings. Chapter 1 in\nHaggblade, Hazell and Reardon (2007) provides a valuable discussion of definitions and concepts.] Section 2 also shows that with growth of the non-farm sector there is also\nevidence of declining “quality” of non-farm jobs, notably in the direction of increased\ncasualization of non-farm employment, away from regular, salaried, employment. The section\nfurther documents a persistently high share of the overall non-farm workforce engaged in selfemployment activities.  The section ends with the observation that since the 1980s, rural poverty\nin India has been falling and points to analysis in support of the contention that casualization of\nnon-farm employment opportunities has translated into improved access of the historically\ndisadvantaged segments of rural society to non-farm employment. As returns from casual nonfarm employment are higher than from agricultural labor (though markedly lower than from\nregular non-farm employment), the growing participation of disadvantaged groups in this (sub)\nsector is likely to have been a positive force for poverty reduction.\n\n\n2 The survey based analysis in this paper draws primarily on five “thick” rounds of the NSS—1983, 1993-94, 19992000, 2004-05, and 2009/10. We\n\nMention: \"the NSS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Five thick rounds across these years carry the survey analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005604:13:1", "occurrence_id": "ae933d8b4ac88929ec11b202", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 21 912.0\n\n110 3.26 3.93 1119.7 3.20 910.6\n\n115 3.41 4.07 1157.9 3.20 909.2\n\n120 3.55 4.20 1196.0 3.19 907.8\n\nSource: calculated\n\n\nThe very close co-movement of actual and predicted soybean oil prices depicted in\nFigure 2 empirically supports the theoretical link between biodiesel and soybean oil prices\nestablished by equation (6). <sup>19</sup> [^19: The underlying data come from the “Historical Biodiesel Operating Margins” spreadsheet of the Centre for\nAgricultural and Rural Development of the Iowa State University.\n[http://www.card.iastate.edu/research/bio/tools/hist_bio_gm.aspx](http://www.card.iastate.edu/research/bio/tools/hist_bio_gm.aspx)] This link appears to be very strong because prices of biodiesel and\nsoybean oil exhibit almost identical development as do actual and predicted soybean oil prices\n(not shown). <sup>20</sup> [^20: Biodiesel prices ran up in 2007-08 because of ‘splash and dash’ (see de Gorter et al. 2011) and then crashed as the\nEuropean Union invoked anti-dumping and countervailing duties. But the U.S. biodiesel mandate started to be\nenforced in mid-2010, causing prices to rise sharply again.] The three visible periods when our model over predicts (April – September 2007,\nFebruary 2008 – April 2009, and February 2011 – April 2012) correspond to the periods of\npositive profits earned in biodiesel production, likely due to capacity constraints.\n\n**5.** **Conclusions**\nThis paper has advanced an analytical framework to analyze the economics of biodiesel\npolicies (tax exemptions/tax credits and mandates) and the effects of crude oil price shocks under\neither of these policies. We show how the jointness in soybean (rapeseed) crushing – i.e\n\nMention: \"Historical Biodiesel Operating Margins\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its spreadsheet supplies the underlying margin data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005611:3:0", "occurrence_id": "041370a27db5e673d3fd841c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **1.** **Introduction**\n\nCentral banks intervene in foreign exchange (FX) markets when they target particular levels of\n\nthe exchange rate, through pegs, crawls, or bands with respect to other currencies or currency\n\nbaskets. Even when they do not target a particular rate, they may intervene when there is\n\nexcessive volatility in the exchange rate, which can be destabilizing for traders and ordinary\n\nconsumers. When faced with current account surpluses or capital inflows, central banks that\n\ntarget the exchange rate end up accumulating large amounts of FX reserves. Accumulation of\n\nreserves can also be a symptom of an undervalued exchange rate that some governments have\n\nused to promote exports—or to prevent an appreciation that could undermine competitiveness.\n\nThe extraordinary accumulation of reserves by the People’s Bank of China since the mid-1990s\n\nand by the Swiss National Bank in the aftermath of the 2008-09 global financial crisis—\n\nespecially following the Swiss franc’s peg to the euro in September 2011—have been the focus\n\nof a lot of discussion and analysis. <sup>1</sup> [^1: See, for example, Ross and Jones, (2013), who point to evidence that, in the face of excessive accumulation of FX reserves in a\nnear-zero interest rate environment for US dollar and euro denominated bonds, central banks have been increasingly searching\nfor opportunities to diversify reserve holding into higher-yielding assets, including from emerging market economies. According\nto IMF statistics, worldwide FX reserve holdings had more than doubled, from US$ 5 trillion in 2006 to US$ 10.8 trillion in\n2012.] By contrast, central banks that intervene to prevent exchange\n\nrate depreciation could see their FX reserves get depleted.\n\n\nCentral banks committed to a floating exchange rate intervene in foreign exchange markets for\n\nvarious reasons: When, for instance, a country’s currency appreciates because of some shock and\n\nthe central bank believes that this is a temporary fluctuation, it may want to prevent the\n\nappreciation of the exchange rate that may undermine growth\n\nMention: \"IMF statistics\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Worldwide foreign exchange reserve holdings doubled between these years.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005612:6:0", "occurrence_id": "558defbc18a51d35b96910ed", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: of the direction and possible magnitude of the poverty effects from climate change in Mexico,\n\n\nrather than actual forecasts.\n\n\nThe rest of this paper proceeds as follows. Section 2 introduces key definitions and data\n\n\nsources applied. Section 3 presents the empirical methodology followed by a discussion of the\n\n\nresults in Section 4. Section 5 concludes.\n\n##### **2 Data**\n\n\nThis study is carried out in 2,069 municipalities from all states in Mexico, representing 84%\n\n\nof the total number of municipalities in the country. Missing municipalities are largely from\n\n\nOaxaca and Puebla where climate model projections could not be fitted into smaller munic\n\nipalities. Both states display pockets of high poverty (see Figure 3). However, their highly\n\n\nfragmented political geography, especially Oaxaca, it made unfeasible to analyze climate data\n\n\n(available at a resolution of 50 x 50 km approx) in some small municipalities.\n\n\nThe analysis uses five types of information: (i) income and geographic data, (ii) climate and\n\n\nweather data, (iii) poverty rates, (iv) climate change scenarios, and (v) population and output\n\n\n(GDP) projections. Per capita GDP and geographic controls come from the National Insti\n\ntute of Statistics and Geography (INEGI). Daily precipitation in millimeters and temperature\n\n\ncome from meteorological stations and the National Weather Service (Servicio MeteorolÃşgico\n\n\nNacional - SMN). Historical Climate data were aggregated at the municipality level from a\n\n\ngridded historical dataset derived from observational data produced by the Climatic Research\n\n\nUnit (CRU) of University of East Anglia (UEA). These datasets were accessed through the\n\n\nWorld Bank Climate Change Knowledge Portal (CCKP) <sup>5</sup> .The CRU TS 2.1 Global Climate\n\n\nDataset is comprised of 1,224 monthly time series of climate variables, including temperature\n\n\nand precipitation, for the period 1901-2009, and covering the global land surface, excluding\n\n\nAntarctica,\n\nMention: \"INEGI\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Output and geographic controls come from this statistics institute.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005612:6:0", "occurrence_id": "b508655c2afb0c601dae973f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: of the direction and possible magnitude of the poverty effects from climate change in Mexico,\n\n\nrather than actual forecasts.\n\n\nThe rest of this paper proceeds as follows. Section 2 introduces key definitions and data\n\n\nsources applied. Section 3 presents the empirical methodology followed by a discussion of the\n\n\nresults in Section 4. Section 5 concludes.\n\n##### **2 Data**\n\n\nThis study is carried out in 2,069 municipalities from all states in Mexico, representing 84%\n\n\nof the total number of municipalities in the country. Missing municipalities are largely from\n\n\nOaxaca and Puebla where climate model projections could not be fitted into smaller munic\n\nipalities. Both states display pockets of high poverty (see Figure 3). However, their highly\n\n\nfragmented political geography, especially Oaxaca, it made unfeasible to analyze climate data\n\n\n(available at a resolution of 50 x 50 km approx) in some small municipalities.\n\n\nThe analysis uses five types of information: (i) income and geographic data, (ii) climate and\n\n\nweather data, (iii) poverty rates, (iv) climate change scenarios, and (v) population and output\n\n\n(GDP) projections. Per capita GDP and geographic controls come from the National Insti\n\ntute of Statistics and Geography (INEGI). Daily precipitation in millimeters and temperature\n\n\ncome from meteorological stations and the National Weather Service (Servicio MeteorolÃşgico\n\n\nNacional - SMN). Historical Climate data were aggregated at the municipality level from a\n\n\ngridded historical dataset derived from observational data produced by the Climatic Research\n\n\nUnit (CRU) of University of East Anglia (UEA). These datasets were accessed through the\n\n\nWorld Bank Climate Change Knowledge Portal (CCKP) <sup>5</sup> .The CRU TS 2.1 Global Climate\n\n\nDataset is comprised of 1,224 monthly time series of climate variables, including temperature\n\n\nand precipitation, for the period 1901-2009, and covering the global land surface, excluding\n\n\nAntarctica,\n\nMention: \"National Weather Service\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Daily precipitation and temperature come from its stations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005612:6:0", "occurrence_id": "25063d65654703a20e3e61a6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: of the direction and possible magnitude of the poverty effects from climate change in Mexico,\n\n\nrather than actual forecasts.\n\n\nThe rest of this paper proceeds as follows. Section 2 introduces key definitions and data\n\n\nsources applied. Section 3 presents the empirical methodology followed by a discussion of the\n\n\nresults in Section 4. Section 5 concludes.\n\n##### **2 Data**\n\n\nThis study is carried out in 2,069 municipalities from all states in Mexico, representing 84%\n\n\nof the total number of municipalities in the country. Missing municipalities are largely from\n\n\nOaxaca and Puebla where climate model projections could not be fitted into smaller munic\n\nipalities. Both states display pockets of high poverty (see Figure 3). However, their highly\n\n\nfragmented political geography, especially Oaxaca, it made unfeasible to analyze climate data\n\n\n(available at a resolution of 50 x 50 km approx) in some small municipalities.\n\n\nThe analysis uses five types of information: (i) income and geographic data, (ii) climate and\n\n\nweather data, (iii) poverty rates, (iv) climate change scenarios, and (v) population and output\n\n\n(GDP) projections. Per capita GDP and geographic controls come from the National Insti\n\ntute of Statistics and Geography (INEGI). Daily precipitation in millimeters and temperature\n\n\ncome from meteorological stations and the National Weather Service (Servicio MeteorolÃşgico\n\n\nNacional - SMN). Historical Climate data were aggregated at the municipality level from a\n\n\ngridded historical dataset derived from observational data produced by the Climatic Research\n\n\nUnit (CRU) of University of East Anglia (UEA). These datasets were accessed through the\n\n\nWorld Bank Climate Change Knowledge Portal (CCKP) <sup>5</sup> .The CRU TS 2.1 Global Climate\n\n\nDataset is comprised of 1,224 monthly time series of climate variables, including temperature\n\n\nand precipitation, for the period 1901-2009, and covering the global land surface, excluding\n\n\nAntarctica,\n\nMention: \"World Bank Climate Change Knowledge Portal\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The access point through which the climate datasets were obtained.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005612:6:0", "occurrence_id": "df71d4ffde81d9161d30f91d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: of the direction and possible magnitude of the poverty effects from climate change in Mexico,\n\n\nrather than actual forecasts.\n\n\nThe rest of this paper proceeds as follows. Section 2 introduces key definitions and data\n\n\nsources applied. Section 3 presents the empirical methodology followed by a discussion of the\n\n\nresults in Section 4. Section 5 concludes.\n\n##### **2 Data**\n\n\nThis study is carried out in 2,069 municipalities from all states in Mexico, representing 84%\n\n\nof the total number of municipalities in the country. Missing municipalities are largely from\n\n\nOaxaca and Puebla where climate model projections could not be fitted into smaller munic\n\nipalities. Both states display pockets of high poverty (see Figure 3). However, their highly\n\n\nfragmented political geography, especially Oaxaca, it made unfeasible to analyze climate data\n\n\n(available at a resolution of 50 x 50 km approx) in some small municipalities.\n\n\nThe analysis uses five types of information: (i) income and geographic data, (ii) climate and\n\n\nweather data, (iii) poverty rates, (iv) climate change scenarios, and (v) population and output\n\n\n(GDP) projections. Per capita GDP and geographic controls come from the National Insti\n\ntute of Statistics and Geography (INEGI). Daily precipitation in millimeters and temperature\n\n\ncome from meteorological stations and the National Weather Service (Servicio MeteorolÃşgico\n\n\nNacional - SMN). Historical Climate data were aggregated at the municipality level from a\n\n\ngridded historical dataset derived from observational data produced by the Climatic Research\n\n\nUnit (CRU) of University of East Anglia (UEA). These datasets were accessed through the\n\n\nWorld Bank Climate Change Knowledge Portal (CCKP) <sup>5</sup> .The CRU TS 2.1 Global Climate\n\n\nDataset is comprised of 1,224 monthly time series of climate variables, including temperature\n\n\nand precipitation, for the period 1901-2009, and covering the global land surface, excluding\n\n\nAntarctica,\n\nMention: \"CRU TS 2.1 Global Climate\n\n\nDataset\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Twelve hundred twenty-four monthly series of temperature and precipitation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005617:21:0", "occurrence_id": "86c872bd4d0a2b5291e261d0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: over time. The difference between the 2001 and 2007 models is now larger in both urban and rural\nareas as it should be expected given the smaller samples, but the difference is still very small and the\ntrends depicted by the two models are the same for all periods considered. In fact, by separating urban\nand rural areas, the 2001 and 2007 models are even closer in depicting seasonal variations.\n\n\n**Figure 6 – Quarterly Poverty Estimates 2001-2010, Urban and Rural Areas**\n\n\nSource: LFSs (2001-2010)\n\n\nThe figures below (Figures 7 and 8) further disaggregate the urban and rural poverty trends into subregions. Note that the sub-regions consist of groupings of Morocco’s original 16 regions. <sup>8</sup> [^8: These groupings were determined by the High Commission for the Plan of Morocco based on population density.] What is\napparent from these estimates is that by this sub-division all sub-regions of Morocco show the same\ndeclining trend in poverty. (Results not reported here show that when we disaggregate further, some of\nMorocco’s original 16 regions are found to be more stagnant.) Note that, at this level of disaggregation,\nwe also start to see that the reduction in poverty has been less steep in certain regions as compared to\nothers. We also observe a larger difference between the 2001 and 2007 model estimates, as is to be\nexpected, although the differences remain modest.\n\n\nDespite these encouraging results, some caution should be applied when considering the imputationbased estimates. Note that while the impact of the recent global financial crisis may have been offset at\nleast party by exceptional agricultural output as a result of good harvests shortly after the crisis, it is\nconceivable that not all changes brought on by the crisis are well captured by the labor force surveys.\nFor example, while households might have been able to hold on to employment, they might have readjusted consumption patterns by shifting to lower quality goods and/or reducing consumption\naltogether. Also, not all variables that may signal\n\nMention: \"LFSs\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Quarterly poverty estimates for urban and rural areas come from these rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005620:13:0", "occurrence_id": "b6537c5a03504debb415665b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: empirical evidence on the precise magnitude of such indirect impacts for the long-term income growth\nof the beneficiaries (reviewed in Alderman and Yemtsov 2013). However, all of these studies typically\ncover the effect of one program in one country, and it is hard to generalize across the world.\n\n\nTo approach the issue of cross-country comparability and comprehensiveness, we limit our\nattention to just the first, direct, channel of transfer impact on the current wellbeing of beneficiaries.\nFor this we use household survey information and compute what the household’s income would be\nwithout the transfers it receives from social protection programs. Thus, generally the transfer value is\nsubtracted from individual or household welfare <sup>30</sup> [^30: Welfare is a measure of well-being, and is defined either as per capita income or consumption], providing a synthetic counterfactual. Of course these\nresults are more indicative than authoritative since behavioral effects, for example, are not simulated. <sup>31</sup> [^31: For example, it is assumed that if a person stops receiving a pension, he or she will try to survive on the\nremaining income after the pension is taken away. In practice that person can seek assistance from relatives and\ncommunity that will partly compensate this loss of income, so will not be as poor as assuming no replacement.]\nIn the short run, if households reduced their efforts to earn market income in response to income\ntransfers through social protection, then this method would give an overestimate of the impact on\npoverty. Evidently _any_ medium term improvements in household market income discussed above, imply\nthat the direct assessment would be an underestimate of the poverty and inequality impact. It is difficult\nto know where exactly the balance would come out, but our estimates can provide an anchor for\ndiscussion and sensitivity analysis.\n\n\nUsing the ASPIRE dataset and an international poverty line of $1.25 per day per capita in PPP, Figure\n2 below uses this simulation to estimate how many people are moved out\n\nMention: \"ASPIRE dataset\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Estimates anchored to the international line are computed from this dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005634:9:0", "occurrence_id": "c1b0ad5c1f768cefa040b71c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: leveraged buy-outs (LBOs) that resulted in complete transfer of control to the PE investor.\n\n\nHowever, in recent years PE firms have also started to take minority equity stakes in already\n\n\npublicly-listed firms via PIPE transactions.\n\n\nPE firms are usually credited with three potential value creating strategies. Kaplan and\n\n\nStromberg (2009) define them as financial, governance and operational improvements. Financial\n\n\nengineering entails the use of higher levels of debt to create tax shields as well as to sharpen the\n\n\nincentives of the management team. Governance improvement is largely driven by providing a\n\n\nsignificantly larger share of equity to the operating management and through mitigation of\n\n\nagency problems via concentrated ownership. Operational improvement entails deploying\n\n\nsignificant industry-specific knowledge to create better-run organizations. Thus, unlike other\n\n\ninstitutional investors, PE firms are more likely to be involved in the day to day operations of the\n\n\nfirms that they invest in. Successful PE investors are credited with the ability to source and\n\n\nscreen investment ideas as well as the ability to monitor firms, especially those with poor\n\n\ncorporate governance, using mechanisms such as board representation and active involvement in\n\n\nthe strategic decisions of the firm.\n\n###### **3. Data and Summary Statistics**\n\n\n**3.1** **PIPE and SEO Sample**\n\n\nOur data on PE financed PIPE issues come from the Asian Venture Capital Journal (AVCJ)\n\n\nDatabase, a new electronic database that maintains information on private equity (PE), venture\n\n\ncapital (VC), and mergers and acquisitions (M&A) transactions in Asia. We consider PIPEs from\n\n\na set of nine countries, including Australia, China, Hong Kong, Special Administrative Region of\n\n\nChina, India, Japan, Singapore, South Korea, Taiwan, China, and Thailand, all of which report at\n\n\n8\n\nMention: \"Asian Venture Capital Journal (AVCJ)\n\n\nDatabase\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"A new electronic record of private equity and venture capital activity.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005636:9:2", "occurrence_id": "df02439a59404e4dc5989f78", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  <sup>7</sup> electricity surcharge on electricity consumption.\nTotal surcharge revenues that accumulated from 2006 to 2010 amounted to roughly RMB 16\nbillion (US$2.54 billion), of which 70 percent were used for wind power subsidies.\n\n18. According to the most recent wind power industry development report, <sup>8</sup> gridconnected wind capacity reached 44 GW, and annual wind electricity generation reached 50.1\nTWh by the end of 2010. The 12 <sup>th</sup> FYP increased the wind capacity target to 100 GW by 2015.\nThe target for 2020 might be increased to 150 GW, 5 times the target announced in 2007 (see\nFigure 2).\n\n\n7 The surcharge increased to 0.008 RMB/kWh in 2012.\n\n8 “Wind Turbine Installed Capacity in 2010,” Chinese Wind Energy Association.\n\n5\n\nMention: \"Wind Turbine Installed Capacity in 2010\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"A source note for the installed capacity figure.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005641:5:0", "occurrence_id": "cfe743d33b68ce0280017157", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 3\n\n\n\nChina\nIndustry Share of CO2 Emissions\nand Four Energy Intensive Industries Share\n\n\n\nof Industry CO2 Emissions\n\n\n\n1.0\n\n\n0.8\n\n\n0.6\n\n\n0.4\n\n\n0.2\n\n\n0.0\n\n1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010\n\n\nSource: IEA Online Data Services, 2012\n\n\n\nCO2 Emissions from 4 Energy Intensive Industries as a Share of Industry Emissions\nCO2 Emissions from Industry as a Share of Total Emissions\n\n\n\nAlong the way, we argue that China’s success in reducing the CO2 intensity of industrial\nproduction has not been easy or without cost. Following adoption of a new industrial policy to\n“grasp the large, let go the small” (Sutherland 2003: 2 and 10), attempts to create a small number\nof technologically sophisticated national champion enterprises that can save energy and\ndominate in each of these industries have not been entirely successful. <sup>8</sup> [^8: See discussion in the appendix.] Even where successful,\nsome (World Bank 2012) have worried about the economic costs associated with the anticompetitive effects of this strategy. Others (Allen et al 2008) argue that bank loans to China’s\nstate-owned enterprises, including those to national champions, undermine an efficient allocation\nof capital thereby imposing additional significant economic losses on the Chinese economy.\nFinally, given China’s highly fragmented industrial structure (Oh 2012: 5) and its highly\ndecentralized industrial governance system, it should not be surprising that the central\ngovernment has not been entirely successful in “letting go the small” by closing nearly all of the\nsmall scale, energy inefficient, and polluting facilities in these industries. Yet despite these\nproblems, as we show, China has been quite successful in reducing the CO2 intensity of\nproduction in the industries we study.\n\nOur argument proceeds in four steps. Because improving energy efficiency in particular\nenterprises and industries depends on the ability of enterprises in particular industries to\nsuccessfully adopt the minor process changes and the major technological changes that enable\nthem to reap environmental technique effects\n\nMention: \"IEA Online Data Services\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Emissions shares across energy-intensive industries over these years come from this service.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005642:11:1", "occurrence_id": "b397c5607b02c88cd1537137", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  data is used. The first (All DB) is a comprehensive index including all DB\nindicators in all topics. The second (Business Start-up_DB) measures the complexity of a number of\nprocedural aspects related to the entry and exit process; it encompasses the indicators for _starting a_\n_business, registering property_ and _closing a business_ . The third (Business Operations_DB) captures the\nburden of regulations faced by an enterprise in managing regular operations; it covers the topics of\n_dealing with construction permits, paying taxes, trading across borders_ and _employing workers_ . The\nfourth and last index (Institutional Environment_DB) measures the quality of the legal and institutional\nframework and aggregates the indicators for _protecting investors, getting credit_ and _enforcing contracts_ .\nAll PCA indices based on DB data are coded such that higher values indicate simpler regulation. <sup>24</sup> It is\nworth mentioning that none of the regulatory proxies are available at the sector level, only at the country\n\nfor all proxies of regulatory environment used for the estimation at sector level while Figure 4 shows the\nlevel. The same happens with 𝑂𝑝𝑒𝑛_𝑡𝑟𝑎𝑑𝑒 and ℎ𝑢𝑚𝑎𝑛𝐾. Table VI (Annex) provides summary statistics\nevolution of the comprehensive PCA indicator on DB data (Table VII, Annex, shows the values for all\nPCA indicators for all comparator countries).\n\n\n\n22 See Table IV in the annex.\n23 The productivity leader is allowed to change over time and across sectors.\n24 In this sense, the PCA indices – based on DB data - are consistent with the regulatory quality indicator (based on\nWGI data); even though the PCA indices are scaled differently, they are also defined in such a way that higher\nvalues mean better regulatory environment.\n\n\n10\n\nMention: \"DB data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"Its investor, credit and contract indices are recoded so higher means simpler.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005642:11:1", "occurrence_id": "3ae9445158764dbc79d33f9b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  data is used. The first (All DB) is a comprehensive index including all DB\nindicators in all topics. The second (Business Start-up_DB) measures the complexity of a number of\nprocedural aspects related to the entry and exit process; it encompasses the indicators for _starting a_\n_business, registering property_ and _closing a business_ . The third (Business Operations_DB) captures the\nburden of regulations faced by an enterprise in managing regular operations; it covers the topics of\n_dealing with construction permits, paying taxes, trading across borders_ and _employing workers_ . The\nfourth and last index (Institutional Environment_DB) measures the quality of the legal and institutional\nframework and aggregates the indicators for _protecting investors, getting credit_ and _enforcing contracts_ .\nAll PCA indices based on DB data are coded such that higher values indicate simpler regulation. <sup>24</sup> It is\nworth mentioning that none of the regulatory proxies are available at the sector level, only at the country\n\nfor all proxies of regulatory environment used for the estimation at sector level while Figure 4 shows the\nlevel. The same happens with 𝑂𝑝𝑒𝑛_𝑡𝑟𝑎𝑑𝑒 and ℎ𝑢𝑚𝑎𝑛𝐾. Table VI (Annex) provides summary statistics\nevolution of the comprehensive PCA indicator on DB data (Table VII, Annex, shows the values for all\nPCA indicators for all comparator countries).\n\n\n\n22 See Table IV in the annex.\n23 The productivity leader is allowed to change over time and across sectors.\n24 In this sense, the PCA indices – based on DB data - are consistent with the regulatory quality indicator (based on\nWGI data); even though the PCA indices are scaled differently, they are also defined in such a way that higher\nvalues mean better regulatory environment.\n\n\n10\n\nMention: \"WGI data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its regulatory quality indicator checks the consistency of the other indices.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005613:33:1", "occurrence_id": "316b34b6c47a9a37e91fccdd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  with the decline in\n\nthe treasury rate following the 2007-2008 crisis.\n\n\nSecond, the model generates a decline in investment in the developed\n\neconomy, and an increase in investment in the emerging economy. The decline\n\nin investment in the developed economy reflects the lower private return to\n\ncapital. The increase in investment in the emerging economy is a result of the\n\ndeveloped economy running down its capital stock: in part, this is achieved\n\nthrough saving in emerging economies, to smooth the declining consumption\n\n\n14Source: World Development Indicators.\n\n\n32\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Consumption smoothing across emerging economies is traced with this source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005621:49:1", "occurrence_id": "19e8e495281b3ea63b18803d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  president is impeached and the vice-president succeeds the presidency in a legal and proper way,\nEIEC remains as was. If EIEC was 7 under the old president, it remains 7 under the new president.\n\nFor “Electoral Rules” variables: all get an NA if the LIEC is 1. If LIEC is 2, then legislature is unelected and we\ninfer that district magnitude is NA. If LIEC is less than or equal to 4, then PR is also NA irrespective of district\nmagnitude. If LIEC is less than or equal to 3.5, then both PR and Plurality are NA.\nIn order to assess electoral rules we use the IPU website as well as the Europa Yearbook (and to a lesser extent\nBanks). IPU has the most recent information whereas Europa has information up to 1984, and from 1990 to\n1994. If there are discrepancies between Europa (to 1984) and IPU (1998), we assume that changes have\noccurred, and only input the IPU information for 1995, leaving blanks from 1985 to 1994. If the IPU matched\nthe Europa exactly, we assumed no changes took place, and filled in the intervening years. In the event that a\nsystem changed and then switched back, this introduces errors. Since this assumption was made only when\ninstitutions from 1984 matched those in 1998, these cases are limited to very stable democracies.\n\n\n48\n\nMention: \"Europa Yearbook\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Used with the parliamentary union site to assess electoral rules.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005623:7:0", "occurrence_id": "17ab4ee458b22084692864a4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: by the models. Resource-poor countries, such as Moldova, Ukraine, or Kyrgyz Republic, appear below\nthe 45-degree line; their export structures are more diversified than predicted by the model.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nFigure 6: (Log) Count of Export Products Against\nPrediction\n\n\n\nThe same pattern emerges from Figure 6 that\ncompared the predicted and actual number of\n\ndiversification indicators, which suggests that\nresource abundance, development and size explain\n\nmatches the actual count of export products.\n\nSource: Authors’ calculations based on WDI and Comtrade data\nfrom WITS.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nStill, most of the resource rich countries of the region (Turkmenistan, Azerbaijan, and Kazakhstan) are\nbelow the 45-degree line, which means that they export fewer products than is predicted by the model.\nRussia and Uzbekistan’s numbers of export products match those predicted by the model, while\nTajikistan exports more products than predicted.\n\n\n**Diversification and the Commodity Boom**\nWe assess whether the diversification performance of ECA countries during the commodity boom period\ndiffered between resource-rich and resource-poor countries, also taking into consideration\ndiversification of exports into services.\n\n\n6\n\nMention: \"WDI and Comtrade data\nfrom WITS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The count of export products is calculated from these combined sources.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005646:13:0", "occurrence_id": "9a073167d68bc80694cc0566", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: percent in 2010 (see detailed calculation in Appendix Table A.3), this apparent improvement comes with\n\ncaveats and should be interpreted with caution appropriate to use of administrative data voluntarily\n\nprovided by firms. The shortfall in social insurance contributions due to evasion is calculated as the\n\ndifference between what registered firms should contribute, given their reported wage bill, less the total of\n\ntheir actual contributions divided by required contribution levels.\n\n\nThe decline in outright evasion among formal sector firms is consistent with micro-evidence suggesting\n\nrising levels of the social insurance coverage, but the decline in evasion calculated from official yearbook\n\ndata suggests higher participation rates than observed in micro data. There are two sources of exclusion:\n\nfirms may hire some workers informally without contracts and the NBS yearbook data may exclude many\n\nsmaller firms that are less likely to participate.\n\n\nAs suggested above, one disincentive for participation by both employers and employees lies with\n\nChina’s high social insurance contribution rate, which leads to a “labor tax wedge” that is among the\n\nhighest found in both developing and developed countries (Figure 5). <sup>26</sup> Moreover the design of the social\n\ninsurance system includes minimum payments that imply significantly higher contribution rates for low\n\nincome workers, primarily part-time workers, migrants and lower income self-employed. More\n\nspecifically, the urban social insurance contribution, based on average wages, requires a minimum\n\npayment for workers equal to that which would be paid by a worker earning 60 percent of the local\n\naverage wage and a maximum payment of 300 percent. If a worker earns less than 60 percent of the local\n\naverage annual wage, for example, the worker and employer must make the same contribution as a\n\nworker earning exactly 60 percent of the average wage. This non-linearity at low income levels introduces\n\na significant disincentive for low wage workers to participate in social insurance (DRC and World Bank,\n\n2012).\n\n\n**3.3 Consequences of a Fragmented System: Benefit Disparity and Limited Portability\n\nMention: \"administrative data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Voluntary firm returns; the improvement should be read with caution.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005646:13:0", "occurrence_id": "c514a1a7b7255e5bd0487c33", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: percent in 2010 (see detailed calculation in Appendix Table A.3), this apparent improvement comes with\n\ncaveats and should be interpreted with caution appropriate to use of administrative data voluntarily\n\nprovided by firms. The shortfall in social insurance contributions due to evasion is calculated as the\n\ndifference between what registered firms should contribute, given their reported wage bill, less the total of\n\ntheir actual contributions divided by required contribution levels.\n\n\nThe decline in outright evasion among formal sector firms is consistent with micro-evidence suggesting\n\nrising levels of the social insurance coverage, but the decline in evasion calculated from official yearbook\n\ndata suggests higher participation rates than observed in micro data. There are two sources of exclusion:\n\nfirms may hire some workers informally without contracts and the NBS yearbook data may exclude many\n\nsmaller firms that are less likely to participate.\n\n\nAs suggested above, one disincentive for participation by both employers and employees lies with\n\nChina’s high social insurance contribution rate, which leads to a “labor tax wedge” that is among the\n\nhighest found in both developing and developed countries (Figure 5). <sup>26</sup> Moreover the design of the social\n\ninsurance system includes minimum payments that imply significantly higher contribution rates for low\n\nincome workers, primarily part-time workers, migrants and lower income self-employed. More\n\nspecifically, the urban social insurance contribution, based on average wages, requires a minimum\n\npayment for workers equal to that which would be paid by a worker earning 60 percent of the local\n\naverage wage and a maximum payment of 300 percent. If a worker earns less than 60 percent of the local\n\naverage annual wage, for example, the worker and employer must make the same contribution as a\n\nworker earning exactly 60 percent of the average wage. This non-linearity at low income levels introduces\n\na significant disincentive for low wage workers to participate in social insurance (DRC and World Bank,\n\n2012).\n\n\n**3.3 Consequences of a Fragmented System: Benefit Disparity and Limited Portability\n\nMention: \"NBS yearbook data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its coverage may omit smaller firms less likely to participate.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005647:12:0", "occurrence_id": "3e179c2473b19db546a7c811", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: calibrate our constant price elasticity demand and supply curves, we use elasticities found in the\n\nliterature as well as the elasticities used for similar developing countries in the FAPRI model.\n\nWe use Hamilton’s (2009) estimate of fuel demand elasticity of -0.26. <sup>15</sup> [^15: This value is close to the average long-run elasticity obtained reported in a meta-analysis by Havránek et al.\n(2012).] The demand for soybean\n\noil for human consumption, demand for soybean meal for feed, and supply of soybeans, has price\n\nelasticities -0.15, -0.35, and 0.34, respectively. <sup>16</sup> [^16: The FAPRI elasticities database does not provide elasticities specifically for Zambia. We therefore use the\nelasticities reported for Brazil as both countries are developing countries.]\n\n\n_The Persistence of the Observed Production and Trade Patterns_\n\nThe objective of this section is to provide rough bounds on the key market prices and\n\nparameters for which one might expect the observed production, consumption, and trade patterns\n\nto be preserved. The observed patterns suggest that the price regime 4 in Table 1 prevailed in\n\nZambia in 2010. This price regime implies that in order for domestic biodiesel production to\n\nbegin, Zambia’s biodiesel producers need to be compensated for the difference between the\n\nmarket price of soybean oil and their willingness to pay for the feedstock to break even. The\n\ndetermination of this compensatory payment is crucial for modeling the effects of alternative\n\nbiodiesel policies.\n\nSinkala (2011) reports a processing cost of soybean-based biodiesel in Zambia of\n\n$0.22/liter. Because the local market price of biodiesel in Zambia is $1.31/liter, using equation\n\n(5) we estimate the Zambian biodiesel producers’ willingness to pay for soybean oil to be\n\n$1,180/mt. The difference between the market price of soybean, $1,284/mt, and the willingness\n\nto pay then determines the compensatory payment which\n\nMention: \"FAPRI elasticities database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It lacks one country, so reported elasticities are borrowed instead.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005657:15:0", "occurrence_id": "b4455b5c27109b8c49b74cc1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: A caveat here is that when we compare number of full-time employees from the IMSS data to\n\nour follow-up survey data, the IMSS numbers are lower, suggesting that the enterprises in our sample\n\ndid not register all their employees with IMSS (the follow-up survey suggests that our enterprises had\n\nabout 15 full-time paid employees in 2009, and 2010 IMSS data shows about 10 employees). The\n\nincrease in number of employees in the IMSS data could thus reflect more employees being registered\n\ninstead of more employees being hired (this still is a desirable outcome though from a societal\n\nperspective). However, given that the increase in employees recorded in the IMSS data is long term, not\n\nmerely a short-term increase after the training, it appears to be a genuine increase in actual\n\nemployment at these firms.\n\nNot all of the firms in our sample were found in IMSS records, so that the results in Table 4 are\n\nbased on about 57 percent of our sample. To assess whether there are important compositional changes\n\nthat influence our key results, Appendix Table 9 reports the main specifications for impact as measured\n\nby the follow-up survey, but restricted to the sample of firms for which we have IMSS data. The results\n\nare similar.\n\n\n**4.2 Process Variables**\n\nIn order to investigate the channels that drive the observed treatment effects, we now study\n\nhow the consulting program changed processes within the enterprise. We measure these processes as\n\nfollows. First, the surveys asked enterprise owners whether or not they implemented certain changes\n\nduring the past year, such as developing new products, attracting new investors, and launching a new\n\nmarketing campaign. Note that if treatment enterprises believed they should please the program by\n\nreporting process changes that did not actually occur, these estimates will be upwardly biased.\n\nTable 5 displays the treatment effects on business process variables. We start with an all\nencompassing standardized index, calculated as per Kling et al. (2007),\n\nMention: \"IMSS data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Its employee counts fall below the follow-up survey, indicating undercount.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005662:13:10", "occurrence_id": "ea1c2f6f1e195ef5302becf8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  important aspect of child labor that is not commonly analyzed in the literature is_\n_seasonality._ Seasonal and temporary labor can follow seasonal and temporary school\nattendance patterns, with Figure 3 illustrating school-labor tradeoffs and a higher prevalence\nof child labor during the major planting and harvesting seasons (see appendix). <sup>24</sup> Simply\nengaging children seasonally on the farm is easier for many households than trying to search\nfor and hire adult workers willing to work periodically during the peak seasons when\n\n24 Given the survey design of GLSS and that data on each child is only collected at one point in time, there is\nlikely some degree of measurement error here as the survey does not include questions on children’s main\noccupation (school or work) throughout the year or whether some may likely only work seasonally. Rather,\nchildren are classified as laborers if they worked during the last 7 days, as defined in section 2. Furthermore,\nnationally representative panel data on child labor, which could help tackle this measurement issue, is not\navailable in Ghana.\n\n\n12\n\nMention: \"GLSS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Its design collects child data at one point only, limiting the panel.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005663:16:1", "occurrence_id": "74bcb7dfe4a01a600fc2522a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  + 𝛾𝑋𝑖𝑣𝑘 + 𝜇𝑘 + 𝛥𝜀𝑖𝑣𝑘 (4)\n\n\n\nwhere 𝛥𝑙𝑛(𝑦𝑖𝑣𝑘) is real income growth for household _i_, in village _v_ and kecamatan _k_ . It is the log\n\n<u>𝐼09,𝑖𝑣𝑘</u> <u>𝐼07,𝑖𝑣𝑘</u>\n\n\n\ndifference of real per capita income: ∆𝑙𝑛(𝑦𝑖𝑣𝑘) = 𝑙𝑛 ~~�~~\n\n\n\n~~𝑛~~ <u>𝐼09,𝑖𝑣𝑘</u> ~~09,𝑖𝑣𝑘~~ <sup>�</sup> <sup>−𝑙𝑛</sup> <sup>�</sup>\n\n\n\n~~𝑛~~ <u>𝐼07,𝑖𝑣𝑘</u> ~~07,𝑖𝑣𝑘~~ <sup>~~�~~</sup> <sup>where</sup> <sup>𝐼𝑡</sup> <sup>is weekly</sup>\n\n\n\nearning (in real term, deflated by the consumer price index in Aceh province from BPS ( _Badan Pusat_\n\n\n\nearning (in real term, deflated by the consumer price index in Aceh province from BPS ( _Badan Pusat_\n\n_Statistik_ ) website) and 𝑛𝑡 is household size in year _t_ . Right hand side includes a dummy for fishing\n\n\n\n15\n\nMention: \"BPS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its provincial consumer price index deflates real earnings.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005671:19:1", "occurrence_id": "6406c3ffcaa290c01e6dc4bd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: -stage designs and are nationally\n\nrepresentative. The EICVM is conducted in four stages, but only the first stage is used by\n\nthe present study. After data clearance, 8,421 households are retained out of the 9,075\n\nhouseholds initially included in the sample. Regarding the CWIQ, 6,048 households are\n\nincluded in the initial sample.\n\n\nIndividuals aged 15 to 64 years are the unit of analysis. In the case of Burkina Faso, the\n\nfinal sample includes all individuals, which represents 26,124 people from 8,258\n\nhouseholds, including 11,698 men (45%) and 14,426 women (55%). Also, 27% of people\n\nare urban while 73% live in rural areas. By contrast, in Togo, only the household head\n\nand his spouses are considered for the question involving financial resources. This leads\n\nto retaining 8,229 individuals from 4,980 households, for 3,716 men (45%) and 4,513\n\nwomen (55%). This large sample drop may introduce a bias in the short sample. In fact,\n\nsome socio-demographic features (e.g. average household size and average age) appear\n\nsignificantly different from one sample to another. Moreover, this country appears more\n\nurbanized than the previous one since urban represents 42% against 58% for rural.\n\n\n18\n\nMention: \"EICVM\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"A nationally representative four-stage design whose first stage is used here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005671:19:1", "occurrence_id": "0fc1216b6ff70719dd779350", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: -stage designs and are nationally\n\nrepresentative. The EICVM is conducted in four stages, but only the first stage is used by\n\nthe present study. After data clearance, 8,421 households are retained out of the 9,075\n\nhouseholds initially included in the sample. Regarding the CWIQ, 6,048 households are\n\nincluded in the initial sample.\n\n\nIndividuals aged 15 to 64 years are the unit of analysis. In the case of Burkina Faso, the\n\nfinal sample includes all individuals, which represents 26,124 people from 8,258\n\nhouseholds, including 11,698 men (45%) and 14,426 women (55%). Also, 27% of people\n\nare urban while 73% live in rural areas. By contrast, in Togo, only the household head\n\nand his spouses are considered for the question involving financial resources. This leads\n\nto retaining 8,229 individuals from 4,980 households, for 3,716 men (45%) and 4,513\n\nwomen (55%). This large sample drop may introduce a bias in the short sample. In fact,\n\nsome socio-demographic features (e.g. average household size and average age) appear\n\nsignificantly different from one sample to another. Moreover, this country appears more\n\nurbanized than the previous one since urban represents 42% against 58% for rural.\n\n\n18\n\nMention: \"CWIQ\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its initial sample of just over six thousand households is retained here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005675:21:0", "occurrence_id": "a3aa02fe68a526621c6bf50a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: possible the criteria in question 16. <sup>13</sup> [^13: The index includes 20 variables from five sources: the World Economic Forum’s (WEF) “Executive Opinion\nSurvey,” the International Country Risk Guide (ICRG), Economist Intelligence Unit (EIU), Freedom House, and\nReporters Without Borders. Data are all from 2012. The ICRG and EIU ratings are updated monthly, and the others\nannually.] Again, the IDA dummy is not significant, but the ECA\ndummy is positive and significant at the .05 level in both regressions. The average ECA country\nproposed rating (equation 6.7) is about one-fifth of a point higher than predicted. Following the\nnetwork review process, the magnitude of the bias is reduced slightly, from .192 to .182. <sup>14</sup> [^14: Results for questions 12 and 16 in the table are very similar if the 2011 Worldwide Governance Indicators (WGI)\n“Rule of Law,” “Control of Corruption” and “Voice and Accountability” indexes are substituted for the guidepost\nindexes. The WGI indexes include more sources than the guidepost indexes, but are not designed to match the\ncontent of the CPIA criteria and contain less up to date information.]\n\nThe network review has the effect of strengthening (albeit modestly) the relationship between\nCPIA ratings and comparator indicators. The coefficient on the comparator indicator is slightly\ngreater in the second regression within each pair, for questions 12, 15 and 16. This result is not\nsurprising, as the network reviewers make use of these indicators (as well as other quantitative\nand qualitative information) in formulating their ratings recommendations.\n\nTable 7 reports regressions similar to those for the even-numbered equations in Table 6, i.e.\nusing the final rather than proposed ratings as dependent variables. They differ however by\nspecifying dummy variables for the other 5 regions, with ECA as the base category. The\nregional dummy coefficients in this table thus show\n\nMention: \"the 2011 Worldwide Governance Indicators (WGI)\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Substituting its three rule-of-law style indexes leaves results similar.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005676:20:1", "occurrence_id": "286bb6f5b3a72180630f1ba2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: . The data include the date at which each power plant became\n\nactive. Using the GPS locations of the REDS survey villages, we calculated the straight-line\n\ndistance from each village to the nearest power plant for 1965 (as a benchmark before the REDS\n\nsurvey, during the time when rural electricity generation capacity was beginning to expand), and\n\n1975.\n\n##### **5. Specification and identification**\n\n\nWe postulate two distinct ways that a rural household can benefit from electrification.\n\nFirst, it can gain from village electrification, even if it does not have electricity itself, though we\n\nallow for the possibility that the extent of that gain may depend on whether the household itself\n\n\n19\n\nMention: \"REDS\n\nsurvey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Village distances to the nearest plant are benchmarked to its date.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005682:4:0", "occurrence_id": "bdc3a71bdda8df66d0c2f1c9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Accordingly, this paper builds on the work of the forthcoming World Bank report on jobs and\nzooms into the patterns of job creation across surviving firms in the EU11 economies. The analysis\nutilizes the Amadeus database (provided by Bureau van Dijk) which contains comparable and\ncomprehensive balance sheet and profit and loss account data for the EU11 countries for a time\nperiod spanning from 2002 to 2009. <sup>4</sup> [^4: A detailed data description for the Amadeus database is provided in Udomsaph (2013). One should note the large variation in the\ncoverage and representativeness across the EU11 countries in the Amadeus data base, and hence the quality of data across the EU11\ncountries. The results presented here pertain solely to the surveyed firms in the data base.] This data set is augmented with data from the World Bank's\nBusiness Environment and Enterprise Performance Survey (BEEPS) which collects information on\nthe business environment in which the firms operate. In the EU11 economies, these surveys have\nbeen conducted in the years 2002, 2005, 2007 and 2009, allowing the analysis of dynamics of job\ncreation at the firm level in the boom years prior to the global financial crisis as well as during the\n“bust”. The next section presents stylized facts on job creation in the EU11 economies, while\nSection 3 offers a structural analysis of job creation patterns across and within-industries. The\ncomparison of employment growth dynamics across industries allows for examining whether the\nEU11 economies have been successful in shifting employment to more productive sectors, while the\nwithin-industry analysis provides evidence on which firms have been most crucial for overall job\ncreation. Section 4 re-examines this latter issue by investigating the key-industry and firm-specific\ncharacteristics of high-growth firms. Section 5 offers policy conclusions.\n\n# Stylized Facts\n\n\n**Many jobs created in EU11 during 2002-2008 were lost during the 2008 global financial**\n**crisis.** Among the surveyed firms, the number of employees in firms that were active in 2002\nincreased by approximately 76 percent (Figure 1).\n\nMention: \"Amadeus database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Comparable balance sheet data across surviving firms come from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005682:4:0", "occurrence_id": "d750a4faeec1f1a5cc3f5f24", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Accordingly, this paper builds on the work of the forthcoming World Bank report on jobs and\nzooms into the patterns of job creation across surviving firms in the EU11 economies. The analysis\nutilizes the Amadeus database (provided by Bureau van Dijk) which contains comparable and\ncomprehensive balance sheet and profit and loss account data for the EU11 countries for a time\nperiod spanning from 2002 to 2009. <sup>4</sup> [^4: A detailed data description for the Amadeus database is provided in Udomsaph (2013). One should note the large variation in the\ncoverage and representativeness across the EU11 countries in the Amadeus data base, and hence the quality of data across the EU11\ncountries. The results presented here pertain solely to the surveyed firms in the data base.] This data set is augmented with data from the World Bank's\nBusiness Environment and Enterprise Performance Survey (BEEPS) which collects information on\nthe business environment in which the firms operate. In the EU11 economies, these surveys have\nbeen conducted in the years 2002, 2005, 2007 and 2009, allowing the analysis of dynamics of job\ncreation at the firm level in the boom years prior to the global financial crisis as well as during the\n“bust”. The next section presents stylized facts on job creation in the EU11 economies, while\nSection 3 offers a structural analysis of job creation patterns across and within-industries. The\ncomparison of employment growth dynamics across industries allows for examining whether the\nEU11 economies have been successful in shifting employment to more productive sectors, while the\nwithin-industry analysis provides evidence on which firms have been most crucial for overall job\ncreation. Section 4 re-examines this latter issue by investigating the key-industry and firm-specific\ncharacteristics of high-growth firms. Section 5 offers policy conclusions.\n\n# Stylized Facts\n\n\n**Many jobs created in EU11 during 2002-2008 were lost during the 2008 global financial**\n**crisis.** Among the surveyed firms, the number of employees in firms that were active in 2002\nincreased by approximately 76 percent (Figure 1).\n\nMention: \"Business Environment and Enterprise Performance Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It augments the financial data with business environment detail.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005686:7:0", "occurrence_id": "c435c3f27ea6d54e04a937e2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The price and the income effect is the main channel through which crisis impact upon\n\nhouseholds. Evidence from the Asian economic crisis in the late 1990s as well as from Africa\n\nconsistently shows that as prices increase, households first reduce consumption of more\n\nexpensive food items, typically animal source foods (meat, poultry, eggs, fish, and milk), and fruit\n\nand vegetables that are good sources of high-quality nutrients. This is followed by a reduction of\n\nthe size and frequency of meals (Fuere et al, 2000, Bloem et al, 2005, Thorne-Lyman et al, 2010,\n\nKoumou, 2008, Block et al, 2004). World Food Programme’s household-level food security\n\nassessments conducted in 2008 in a number of countries around the world found similar\n\nevidence as well as reductions in health care visits or health expenditures, increased school drop\nouts, and sale of assets (Sanogo, 2009).\n\n\nIndividual country case studies have looked at the impact of crises on nutrition in: Indonesia,\n\nJamaica, South Africa and Cameroon. Over the last twenty years, another strand of country\n\nstudies has emerged, which has been solely focused on Russia. Using official Goskomstat for\n\n1991 and 1992, Cornia (1994) concludes that there is only a small correlation between household\n\nresources (measured by income or expenditure) and caloric intake. Zohoori et al (2001), Popkin\n\net al (1996), Vella (1997) and Dore et al (2003) use RLMS data to examine caloric intake and\n\nother measures of nutrition. These studies report little variation in aggregate levels of nutrition\n\nintake and nutritional status over time.\n\n\nHowever, the absence of a robust link between economic crises and nutrition in the papers\n\nabove might be a result of the methodological and data shortcomings. Indeed, as some of the\n\nauthors point out, the nexus between the two variables is a complex one, thus involving various\n\ncoping strategies that are usually\n\nMention: \"Goskomstat\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Official statistics for these two years underpin the inequality conclusion.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005686:7:0", "occurrence_id": "6125e8e0d081ff481e487a55", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The price and the income effect is the main channel through which crisis impact upon\n\nhouseholds. Evidence from the Asian economic crisis in the late 1990s as well as from Africa\n\nconsistently shows that as prices increase, households first reduce consumption of more\n\nexpensive food items, typically animal source foods (meat, poultry, eggs, fish, and milk), and fruit\n\nand vegetables that are good sources of high-quality nutrients. This is followed by a reduction of\n\nthe size and frequency of meals (Fuere et al, 2000, Bloem et al, 2005, Thorne-Lyman et al, 2010,\n\nKoumou, 2008, Block et al, 2004). World Food Programme’s household-level food security\n\nassessments conducted in 2008 in a number of countries around the world found similar\n\nevidence as well as reductions in health care visits or health expenditures, increased school drop\nouts, and sale of assets (Sanogo, 2009).\n\n\nIndividual country case studies have looked at the impact of crises on nutrition in: Indonesia,\n\nJamaica, South Africa and Cameroon. Over the last twenty years, another strand of country\n\nstudies has emerged, which has been solely focused on Russia. Using official Goskomstat for\n\n1991 and 1992, Cornia (1994) concludes that there is only a small correlation between household\n\nresources (measured by income or expenditure) and caloric intake. Zohoori et al (2001), Popkin\n\net al (1996), Vella (1997) and Dore et al (2003) use RLMS data to examine caloric intake and\n\nother measures of nutrition. These studies report little variation in aggregate levels of nutrition\n\nintake and nutritional status over time.\n\n\nHowever, the absence of a robust link between economic crises and nutrition in the papers\n\nabove might be a result of the methodological and data shortcomings. Indeed, as some of the\n\nauthors point out, the nexus between the two variables is a complex one, thus involving various\n\ncoping strategies that are usually\n\nMention: \"RLMS data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Caloric intake and nutrition measures are examined in these data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005687:6:0", "occurrence_id": "60927cc447f5a189a5ee7c54", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Federal Reserve, suggesting that median wealth has declined. <sup>4</sup> Hence, the prospects of current\nand future retirees are clearly much diminished.\n\n\nWe know that one dollar will buy three times as much in Thailand and twice as much in Costa\nRica as it does in the United States. <sup>5</sup> Retirees could maintain their standard of living by relocating to places with a lower cost of living. In 2011, there were about 567,000 beneficiaries\nliving abroad, of whom 350,000 were retired American workers or about 1 percent of American\nretirees (US Social Security Administration, 2012, Table 5.J2). <sup>6</sup> A quarter of retirees live in\ndeveloping countries (Table 5.J11). Between 2008 and 2010, 35 percent of new overseas\nAmerican retirees headed for the developing world –double the flows over the preceding 7 years.\nAccording to US Census Bureau projections, by 2020, the number of people above age 60 will\nincrease by about 22 million and by a further 14 million by 2030. If the fraction of those\nchoosing to live overseas increases to say 2 percent by 2020 and 3 percent by 2030, about 2.5-3\nmillion US retirees could be living abroad within two decades.\n\n\n_Infirm_\n\n\nThe rising costs of healthcare in the United States imply that if current levels of access to health\ncare are maintained, spending on the major federal health care programs alone would grow from\nmore than 5 percent of GDP today to almost 10 percent in 2037. <sup>7</sup>\n\n\nSince healthcare of comparable quality is much cheaper in other countries, treatment abroad\ncould yield big savings. For example, inpatient knee surgery, roughly 400,000 of which were\nperformed in 2010 in the United States, costs more than $15,000 in a typical community hospital\n(Centers for Disease Control and Prevention,\n\nMention: \"US Census Bureau\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its projections give the increase in older migrants.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005692:19:0", "occurrence_id": "c0c0ae3cc537afaceb93fd3d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: person marketing campaign. The personal marketing visit may have swamped the\n\n\ntreatment effects of the messaging such that without the personal marketing visits, the\n\n\nmessaging would have changed behavior. In other words, people who do not respond to a\n\n\npersonalized visit may be more committed nonrecyclers and may be unresponsive to\n\n\nmessaging treatments, but those who do respond to a personalized visit may have\n\n\nresponded to a messaging treatment even if they had received one without a personalized\n\n\nvisit. This issue is not one of internal validity, but it does make the results less widely\n\n\napplicable. Because the messaging treatments were tested in an environment in which\n\n\neveryone received a personalized visit to promote recycling, we can only conclude that\n\n\nmessaging is ineffective under those specific constraints. Burn and Oskamp (1986) also\n\n\nused a mix of in-person visits and written messaging, but their control group received\n\n\nneither. Their results showed a large, significant increase in recycling behavior for all\n\n\ntreatments relative to their control but no significant differences between the three\n\n\ntreatment arms. Although not conclusive, this finding provides further evidence that in\n\nperson visits may be the most important factor in incentivizing good recycling behavior.\n\n\nWe propose several other explanations for the failure of the treatment to generate\n\n\nan increase in participation or participation intensity. One possible explanation is that the\n\n\nmessages may have been inappropriate for this context because they were motivated by\n\n\nstudies in the United States, where there are different norms, attitudes, and knowledge\n\n\nabout recycling. A related explanation is that the strength of opinions may differ. It is\n\n\ndifficult to find information on attitudes specific to recycling, but we were able to look\n\n\nmore closely at the environmental questions from the World Values Survey (2005) in\n\n\nPeru and the United States. As noted previously, Peruvians were just as likely as\n\n\nAmericans to report that they felt that protecting the environment was important and were\n\n\n18\n\nMention: \"the World Values Survey (2005)\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its environmental questions are examined for two countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005695:8:0", "occurrence_id": "b379200c2baa561381551016", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 5: Changes in the Gini coefficient and in the return to education; 2000-2010**\n_Annual % change_\n\nChange in Gini Change in returns on primary schooling\n\n\n4.0\n\n\n2.0\n\n\n\n0.0\n\n\n-2.0\n\n\n-4.0\n\n\n-6.0\n\n\n-8.0\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n_Source_ : Authors' calculations, based on data from SEDLAC (CEDLAS and The World Bank), May 2013. _Note_ :\nThe returns to different levels of education are calculated with respect to no schooling or incomplete primary\nschool. Skill categories are determined by level of formal education. Educational levels correspond to\ncompleted primary school, lower- and upper-secondary school, and tertiary education. The time periods used\nto estimate the percentage changes are the same as those used in Figure 1.\n\n\nThe determinants of the decline in non-labor income inequality include: returns to\n\ncapital —interests, profits and rents—, private transfers —for example, remittances— and\n\npublic transfers —for example, CCTs and noncontributory pensions. The contribution of\n\nchanges in returns to capital in Argentina, Brazil and Mexico, for example, tended to be\n\nsmall and unequalizing (Lustig et al. 2013). However, a well-known fact is that household\n\nsurveys under-estimate income from capital so the unequalizing effect may have been\n\nlarger than current estimates indicate. Esquivel et al. (2010) show that, in Mexico,\n\nremittances proved to be equalizing and became even more so in the 2000s, because they\n\nclosed the gap between rural and urban household per capita incomes. Cornia (2013) also\n\nshows that the increase in migrant remittances in total household income appears to have\n\nhad an equalizing effect in El Salvador and Mexico; however, in Honduras their effect was\n\nunequalizing.\n\nIn terms of public transfers, as mentioned above, Azevedo et al. (2012) find that,\n\non average\n\nMention: \"data from SEDLAC\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Returns to education levels are calculated from this version.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005702:4:0", "occurrence_id": "b35c38f11a2a8e77254a8b9c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: expansion of social program provisions. <sup>3</sup> The cost of health insurance through employment in\n\n\nColombia is often cited as anecdotal evidence of a powerful disincentive for formal employment.\n\n\nHowever, prior to this paper, no systematic empirical evidence has been presented to confirm or\n\n\nreject this hypothesis. <sup>4</sup>\n\n\nTo determine whether the SR expansion generated incentives for informal employment, we\n\n\nuse two datasets and different identification strategies. The first comes from repeated cross\n\nsections of the Colombian Household Survey. Using individual-level data and controlling for\n\n\nmunicipality and year effects, we estimate whether there was an increase in informal employment\n\n\nafter the municipality started conducting the interviews needed to determine eligibility for the SR\n\n\n(known as SISBEN interviews). These interviews were implemented at different points in time\n\n\nacross municipalities, which allows us to disentangle the effects of the expansion of the health\n\n\ninsurance provision from other labor market reforms and macroeconomic shocks occurring at\n\n\nthe national level during the same period of time. In section III, we show that pre-existing\n\n\ninformality rates are uncorrelated with the timing of the SISBEN, and we find robust and\n\n\nconsistent effects of an increase in informal employment of approximately 4 percentage points.\n\n\nThis finding corresponds to approximately 8% of the labor force not entering the formal sector\n\n\nand an increase in the government’s health expenditures of approximately 11%.\n\n\nOur second dataset is a panel of people interviewed in the first and second SISBEN. Using\n\n\ninformation about an individual’s eligibility status (as determined by her Poverty Index Score\n\n\nfrom the first SISBEN) <sup>5</sup> and individual fixed effects, we corroborate the increase in informal\n\n\nemployment found with the household survey dataset. Additional estimations that account\n\n\n3See Grosh (1994) for a review of 30 social programs in 11 Latin American countries.\n4For example, in a news article reported by _Presidencia_ _de_ _la\n\nMention: \"Colombian Household Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Repeated cross-sections with municipality and year effects.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005702:4:0", "occurrence_id": "7d9ae80374f908b207d52229", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: expansion of social program provisions. <sup>3</sup> The cost of health insurance through employment in\n\n\nColombia is often cited as anecdotal evidence of a powerful disincentive for formal employment.\n\n\nHowever, prior to this paper, no systematic empirical evidence has been presented to confirm or\n\n\nreject this hypothesis. <sup>4</sup>\n\n\nTo determine whether the SR expansion generated incentives for informal employment, we\n\n\nuse two datasets and different identification strategies. The first comes from repeated cross\n\nsections of the Colombian Household Survey. Using individual-level data and controlling for\n\n\nmunicipality and year effects, we estimate whether there was an increase in informal employment\n\n\nafter the municipality started conducting the interviews needed to determine eligibility for the SR\n\n\n(known as SISBEN interviews). These interviews were implemented at different points in time\n\n\nacross municipalities, which allows us to disentangle the effects of the expansion of the health\n\n\ninsurance provision from other labor market reforms and macroeconomic shocks occurring at\n\n\nthe national level during the same period of time. In section III, we show that pre-existing\n\n\ninformality rates are uncorrelated with the timing of the SISBEN, and we find robust and\n\n\nconsistent effects of an increase in informal employment of approximately 4 percentage points.\n\n\nThis finding corresponds to approximately 8% of the labor force not entering the formal sector\n\n\nand an increase in the government’s health expenditures of approximately 11%.\n\n\nOur second dataset is a panel of people interviewed in the first and second SISBEN. Using\n\n\ninformation about an individual’s eligibility status (as determined by her Poverty Index Score\n\n\nfrom the first SISBEN) <sup>5</sup> and individual fixed effects, we corroborate the increase in informal\n\n\nemployment found with the household survey dataset. Additional estimations that account\n\n\n3See Grosh (1994) for a review of 30 social programs in 11 Latin American countries.\n4For example, in a news article reported by _Presidencia_ _de_ _la\n\nMention: \"SISBEN\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"A panel of people scored for eligibility across two rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005704:21:0", "occurrence_id": "8dcfe9100cfa9f846797be4d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 1. Budget share of wheat consumption and food consumption\n\n\n\n\n\nData source: PSLM 2007-08\n\n\n20\n\nMention: \"PSLM\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Budget shares of wheat and food consumption come from this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005684:42:0", "occurrence_id": "f5986c370a92b53dd75aaa90", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **_Wheat_**\n\n\nAccording to the World Bank’s Global Economic Monitor database, the price of wheat traded\n\n\ninternationally increased substantially between 2006 and 2008, with US soft red winter wheat\n\n\nrising by 71 percent and Canadian Western Red Spring wheat rising by 110 percent. Using the\n\n\ndata from the FAO database, however, we observe much greater variation in border price\n\n\nchanges between countries (Appendix table 1). Some countries experienced negligible price\n\n\nchanges while other countries experienced price increases of over 200 percent. For the majority\n\n\nof countries, the wheat border price rise was between 80 and 120 percent.\n\n\nDomestic wheat prices rose much less than the large increases in border prices for most\n\n\ncountries. The median domestic price change was 70 percent, and most countries experienced a\n\n\nprice increase between 35 and 100 percent (Appendix figure 1). Because domestic prices\n\n\nchanged at very different rates from the border prices, there is a sharp reduction in protection.\n\n\nThe observed median change in the protection index is only -0.1 percent, but most of the\n\n\ncountries’ protection fell between zero and 25 percent.\n\n\n**_Maize_**\n\n\nThe international price of maize rose by 83 percent between 2006 and 2008 (Appendix table 1).\n\n\nThe median change in border prices was 94 percent, with half the observations between 60 and\n\n\n110 percent (Appendix figure 1). Domestic prices changed much less with a median price change\n\n\nof 49 percent and most countries experiencing increases in domestic prices between 30 and 85\n\n\npercent. Corresponding to these differences in domestic and international price changes we\n\n\nobserve a reduction in protection with a median fall of 17 percent and the majority of the cuts\n\n\nbetween 10 and 30 percent.\n\n\n41\n\nMention: \"Global Economic Monitor database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"International wheat price increases across these years come from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005684:42:0", "occurrence_id": "2f5c07a80ac8a96d259fdec4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **_Wheat_**\n\n\nAccording to the World Bank’s Global Economic Monitor database, the price of wheat traded\n\n\ninternationally increased substantially between 2006 and 2008, with US soft red winter wheat\n\n\nrising by 71 percent and Canadian Western Red Spring wheat rising by 110 percent. Using the\n\n\ndata from the FAO database, however, we observe much greater variation in border price\n\n\nchanges between countries (Appendix table 1). Some countries experienced negligible price\n\n\nchanges while other countries experienced price increases of over 200 percent. For the majority\n\n\nof countries, the wheat border price rise was between 80 and 120 percent.\n\n\nDomestic wheat prices rose much less than the large increases in border prices for most\n\n\ncountries. The median domestic price change was 70 percent, and most countries experienced a\n\n\nprice increase between 35 and 100 percent (Appendix figure 1). Because domestic prices\n\n\nchanged at very different rates from the border prices, there is a sharp reduction in protection.\n\n\nThe observed median change in the protection index is only -0.1 percent, but most of the\n\n\ncountries’ protection fell between zero and 25 percent.\n\n\n**_Maize_**\n\n\nThe international price of maize rose by 83 percent between 2006 and 2008 (Appendix table 1).\n\n\nThe median change in border prices was 94 percent, with half the observations between 60 and\n\n\n110 percent (Appendix figure 1). Domestic prices changed much less with a median price change\n\n\nof 49 percent and most countries experiencing increases in domestic prices between 30 and 85\n\n\npercent. Corresponding to these differences in domestic and international price changes we\n\n\nobserve a reduction in protection with a median fall of 17 percent and the majority of the cuts\n\n\nbetween 10 and 30 percent.\n\n\n41\n\nMention: \"FAO database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its records show much greater variation in border price changes.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005706:14:0", "occurrence_id": "7713b508cf8d3ba33609da40", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: When we estimate the effect of the country risk on the level of the investment, we see that the\nenergy sector remains the most sensitive to changes in country risk ratings not only in the\nlikelihood of undertaking PPI investments but also when predicting the level of the investments.\n\nA change of one point in country risk is linked on average to an increase of 3.8 percent in\ninvestments in energy. After energy comes transport with a 2.5 percent and then\ntelecommunication with 1.6 percent, while “Water and Sewerage” does not have a statistically\nsignificant effect. Hence, improving an SD (13.8 points) in the country risk, implies a 53.4, 34.6,\nand 22.4 percent increase in the probability of undertaking PPI in energy, transport, and\ntelecom sectors, respectively. In the “Water and Sewerage” sector the country risk explains the\nlikelihood to commit but not the level of commitment as in other sectors, mainly explained by\nthe fact that almost every water and sewerage investments in PPI database are greenfield\nprojects (around 98%). The nature of the projects is analyzed in the following section.\n\n\n**Table 6: The Effect of Country Risk on PPI Investment Levels by Sector**\n\n\n**VARIABLES** **Log PPI Energy** **Log PPI Telecom** **Log PPI Transport** **Log PPI Water**\n**<u>(1)</u>** **<u>(2)</u>** **<u>(3)</u>** **<u>(4)</u>**\nCountry Risk 0.0387*** 0.0162*** 0.0251* 0.0151\n(0.0110) (0.00552) (0.0128) (0.0184)\nLog GDP\n\nMention: \"PPI database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Nearly all water and sewerage entries here are greenfield projects.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005709:19:0", "occurrence_id": "c5cbb27a815cd1c3d1c093d5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Table 4: Length and percentage of federal road network not meeting standards (2009)**\n\n|Col1|Length (km)|Percentage|\n|---|---|---|\n|Failing to meet minimum riding quality requirements<br>Failing to meet minimum grip requirements<br>Failing to meet minimum strength requirements<br>Failing to meet minimum defects requirements<br>|28,500<br>12,200<br>24,900<br>35,100<br>|57.1<br>24.4<br>49.9<br>70.3<br>|\n\n\n\nDecember 31, 2009.\n\n\n**Given these findings based on descriptive statistics, the question arises whether they**\n**can be corroborated by econometric analysis and to what extent infrastructure**\n**deficiencies affect firm productivity.** This is the question we take up by explicitly\nmodeling changes in TFP productivity growth using a vector of infrastructure variables\nfrom the BEEPs surveys, combined with firm characteristics.\n\n**Since BEEPs surveys are infrequent, and in this case were conducted in 2004 and**\n**2008 only, results in this section use data for 2004 and 2008 only.** Following\nEscribano and Gausch (2004) and Alam et al (2008), we postulate the general regression\nmodel, with changes in productivity being the dependent variable, and include controls\nfor firm characteristics.\n\n\n∆lnTFPijt=α + β1∆ **infrastructure** t-1 + **_∑ⱷn X_** <sup>**_n_**</sup> **_i.t_** <sup>+  ε</sup> i,t\n\n\nFor each of Russia’s regions, and for each sector (2-\n\nMention: \"BEEPs surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Infrastructure variables from these infrequent rounds feed the productivity model.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005714:6:0", "occurrence_id": "8f345f97d2a0c7007c8a3f8b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: At the country level, financial inclusion can boost efficiency of financial intermediation,\nthrough intermediation of greater amounts of domestic savings, and thus improve soundness of\ninvestment financing and investment cycles. Provided that quality financial infrastructure and\ncompetent supervision are implemented, such improvement will, in turn, produce both greater\neconomic and financial stability (Prasad, 2010; Cull et al., 2012).\n\n\nThe paper proceeds as follows. Section two discusses some stylized facts. Section three\nexplains the employed regression model and estimation approaches. Section four describes the\ndata used for estimation of the regression model. Section five discusses the baseline regression\nresults for the entire sample of 95 countries, along with a robustness analysis using the robust\nregression approach, and an alternative access to deposit proxy together with a GMM estimator.\nSection six presents and discusses differences in the model’s estimates across HICs, MICs, and\nLICs, and reports related robustness tests. Section seven concludes and offers some policy\nimplications.\n\n\n**2.** **Some Stylized Facts**\n\n\nIn this section, we preview some stylized facts about the main variables of interest, i.e. bank\ndeposit growth around the 2008 global financial crisis and access to bank deposits before the\ncrisis. The preview motivates some choices that we make for our regression analysis, and further\nillustrates the extent of heterogeneity in deposit growth and access to deposits across country\nincome groups.\n\n\n**Figure 2: Global Growth of Bank Deposits During 2005-2010**\n\n\n\n35\n\n\n30\n\n\n25\n\n\n20\n\n\n15\n\n\n10\n\n\n5\n\n\n0\n\n\n\n2005 2006 2007 2008 2009 2010\n\n\n\nSource: Authors’ calculation based on deposits data from IMF’s FAS database.\nNote: This figure displays the year on year percentage growth rate of outstanding deposits of commercial banks in national\ncurrency calculated for 173 countries.\n\n\nConsider the global growth in bank deposits first. Figure 2 shows that the median growth\nin bank deposit across countries dropped significantly from 2007 to 2009 by about 9 percentage\n\n\n5\n\nMention: \"FAS database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Outstanding deposit growth across these years is calculated from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005716:9:1", "occurrence_id": "581e986ebdeca3f835da6394", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  a population\ndensity of at least 1,000 persons per square kilometer; (ii) _poblaciones_ or central districts\nof a municipality or city with a population density of at least 500 persons per square\nkilometer; (iii) there is presence of the following: (a) street pattern or network of streets in\neither parallel or right angle orientation; (b) at least six economic establishments ; or (c)\n(commercial banks, manufacturing, recreational, and/or personal services); or (d) at least\nthree social/public establishments (town hall, church, public plaza, park, cemetery, school,\nhospital, health center, library, market place, or building where trading activities are\ncarried on at least once a week); and (iv) having at least 1,000 residents with occupations\npredominantly not in farming or fishing. <sup>10</sup> [^10: The “URB” variable in all three rounds of the FIES (2003, 2006 and 2009) is based on the 2000 urban\nclassification. This classification is updated every time a new barangay census is made available. For our\npurpose, we held the classification constant for a more consistent comparison across time periods.]\n**Sampling.** The total sample is 118,832 households. Focusing on rural areas, we have a\nsample of 65,956 households. In each round the sample is stratified by PSUs that have a\n<u>probability of selection very close to 1.</u>\n\n~~8 In spite of requests, we were una~~ ble to obtain the data collected by visit in each year of the FIES survey. This\n<u>would have allowed us to have a better match between the timing of the rainfall during the year and the date</u>\nof collection of the household consumption data.\n9 We take average and not total because in 60% of the municipalities, there is at least one month with missing\ndata\n\nMention: \"FIES survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"The visit-level data could not be obtained, blocking a rainfall timing match.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005711:40:0", "occurrence_id": "117c4573401bbb9f70b7786a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <u>Number of HHs (groups)</u> <u>1,509</u> <u>1,509</u> <u>1,509</u> <u>1,509</u>\n<u>Number of observations</u> <u>5,525</u> <u>5,525</u> <u>4,037</u> <u>4,037</u>\n<u>*=statistically significant at 10% level; **=statistically significant at 5% level or better</u>\nNote: Figures in parentheses are t-statistics based on robust standard errors clustered at the village level.\nModel 1 is based on all 3 survey periods while model 2 includes 2 periods: 1998/99 and 2010/11. Estimates\ncontrol for initial conditions, and the regressions additionally include village level control variables such as\ncommunity prices of consumer goods, daily wage rates of men and women, and infrastructure variables such\nas presence of schools, government and NGO food programs, electricity, paved roads, and so on.\nSource: WB-BIDS surveys 1991/92 and 1998/99, and WB-InM survey 2010/11.\n\n\n39\n\nMention: \"WB-BIDS surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Two rounds record school, food programme, electricity and road presence.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005711:40:0", "occurrence_id": "0cbf429f672ed0ca47f2b1a4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <u>Number of HHs (groups)</u> <u>1,509</u> <u>1,509</u> <u>1,509</u> <u>1,509</u>\n<u>Number of observations</u> <u>5,525</u> <u>5,525</u> <u>4,037</u> <u>4,037</u>\n<u>*=statistically significant at 10% level; **=statistically significant at 5% level or better</u>\nNote: Figures in parentheses are t-statistics based on robust standard errors clustered at the village level.\nModel 1 is based on all 3 survey periods while model 2 includes 2 periods: 1998/99 and 2010/11. Estimates\ncontrol for initial conditions, and the regressions additionally include village level control variables such as\ncommunity prices of consumer goods, daily wage rates of men and women, and infrastructure variables such\nas presence of schools, government and NGO food programs, electricity, paved roads, and so on.\nSource: WB-BIDS surveys 1991/92 and 1998/99, and WB-InM survey 2010/11.\n\n\n39\n\nMention: \"WB-InM survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The later round recording the same community assets.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005715:13:0", "occurrence_id": "398a8ef967ce676e01544208", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **(a) Growth truncated**\n\n\n25 30 35 40 45 50\n\n\nTime\n\n\n\n0.6\n\n\n0.4\n\n\n0.2\n\n\n0.0\n\n\n−0.2\n\n\n\n**(b) Acceleration truncated**\n\n\n25 30 35 40 45 50\n\n\nTime\n\n\n\n6\n\n\n4\n\n\n2\n\n\n0\n\n\n−2\n\n\n\nFigure 5: MODWT scalogram of the truncated synthetic GDP: (a) growth and (b) acceleration.\n\n\nand acceleration series by using a truncated version of the data set introduced above. In\n\n\nparticular, the new starting point is moved to 25th period of the synthetic data. The\n\n\noriginal synthetic series have 30 years of growth in the begining (with the exception of 5\n\n\nyears of stagnant GDP in the middle). Therefore, with the truncated data, there is only\n\n\n5 years of growth in the beginning of the new projection horizon. The figures show that\n\n\nthe decomposition by using the truncated data series does not differ from the one with\n\n\nlonger series (compare to Figure 4b). This illustrates that the choice of starting period\n\n\ndoes not affect the decomposition at later periods.\n\n\nNext, we consider actual data. We use annual series of income, consumption,\n\n\nand investment data from Penn World Tables 7.0, all in 2005 international dollars per\n\n\ncapita terms. The income series are provided by the Purchasing Power Parity (PPP)\n\n\nconverted GDP per capita (chain series) data, whereas consumption and investment series\n\n\nare computed by using the share of actual consumption and investment in the PPP\n\n\nconverted GDP series at 2005 prices.\n\n\nDecomposition for demonstration purposes are limited to the following five coun\n\ntries: Brazil (BRA), Fiji (FJI), Turkey (TUR), United States (USA), and Zimbabwe\n\n\n(ZWE). The selection of the countries roughly reflects a cross-section of the geographic\n\n\nspread, as well as per capita income levels as defined by the World Bank classification\n\n\nusing 2011 Gross National Incomes.\n\n\n12\n\nMention: \"Penn World Tables 7.0\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Annual income, consumption and investment in constant per capita dollars.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005720:18:0", "occurrence_id": "486840c94ef77ab8982d4057", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 17\n\n\nestablished by the leader. This is key, because in a setting where the programmatic promises\nof rulers include infrastructure development, better schools or improved environmental\nquality, the ability of party members to act collectively to enforce those promises has\nspillover benefits for citizens more generally.\n\n\n**The development effects of political party organization**\n\nThe foregoing discussion indicates that political parties in both democracies and\nnon-democracies differ in the degree to which their organization supports collective action\nby party members. In particular, parties seeking to mobilize support on the basis of\nprogrammatic commitments, either in democratic or non-democratic settings, must allow\nparty members to act collectively. Otherwise, leader incentives to shirk in their pursuit of\nthe party program render it non-credible and the program ineffective as a mobilizing device.\nSuccessful machine parties also allow members to act collectively to restrain leader shirking,\nthough they confront significant obstacles to using this organizational endowment to make\nprogrammatic appeals. In contrast, patron-client parties have no capacity at all to make\nprogrammatic appeals.\n\nIf the analysis is correct and patron-client parties are the only parties that are\nsystematically incapable of make credible commitments to pursue policies in the broader\npublic interest, then we should observe an association between the presence of such parties\nand the lower quality of public policies and development outcomes, in both democratic and\nnon-democratic settings.\n\nEvidence on these implications is difficult to assemble, since data on whether parties\nexhibit leader monitoring of member contributions and member collective action to restrain\nleader shirking are not available. However, Keefer (2011) exploits several characteristics of\nparties, for which time-series data are available, that plausibly capture the degree to which\nparties are organized to better promote the collective interests of members. One of these,\nfrom the Database of Political Institutions (Beck, et al. 2001), is whether parties can be\ncharacterized as espousing economic policies that are left (more redistributionist), right, or\ncentrist; or whether instead the party\n\nMention: \"Database of Political Institutions\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its party position variable classifies the economic policies parties espouse.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005722:5:0", "occurrence_id": "19f0eb006b7460c7a285b6c9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The second section presents an overview of the road sector in Zambia and the latest developments,\nsection 3 describes the project selection analysis, section 4 presents the main findings of procurement\nanalysis and the final section concludes.\n\n#### **2. Overview of the road sector in Zambia**\n\n\nThe major audit by the Office of the Auditor General in 2009 brought out weaknesses relating to public\nfinance management, budgetary control, procurement efficacy and quality control (Republic of Zambia\n2010). Zambian contractors, engineers, and government officials surveyed in 2008 reported that\nproviding materials of lower quality than the contract called for was the single most “unethical” practice\nin the industry (Sichombo et al. 2009). A 2010 audit of 18 Zambian road projects jointly financed by the\ngovernment and donors, shown in Table 1, confirms the extent of the problem (Government of Zambia\n2010 quoted in Messick 2011).\n\n\n**Table 1: Types of defect found in surveyed projects**\n\n**<u>Defect found in project</u>** **<u>Percentage of contracts affected</u>**\n<u>Improperly sized aggregate particles</u> <u>44%</u>\n<u>Too much clay</u> <u>75%</u>\n<u>Aggregates did not meet crushing strength</u> <u>67%</u>\n<u>Base thinner than required</u> <u>81%</u>\n<u>Surface dressing layers thinner than required</u> <u>82%</u>\n<u>Cement content less than specified</u> <u>100%</u>\n<u>Concrete samples weaker than required</u> <u>50%</u>\n_Source:_ Government of Zambia (2010).\n\n\nContract\n\nMention: \"2010 audit of 18 Zambian road projects\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Jointly financed road projects audited to confirm the extent of practice problems.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005728:16:1", "occurrence_id": "0d1053c97e46dfdc0b7e4fff", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  to assess whether the levels of foreign direct investment received by\nChina are abnormally high, or whether they can be explained by fundamentals (Fan, Morck, Xu, and Yeung, 2008).\n13 These results are similar to those in Beck _et al_ ., (2008). Aside from our proxy for natural resource intensity, our\nregressions differ from theirs in two ways. First, we use the more conservative cross-sectional approach, whereas they\nuse the full panel. We choose to do this because errors from observations from the same country are likely to be\ncorrelated with one another. This could artificially deflate standard errors, thus increasing the significance level of\ncoefficients. For robustness, we ran our models also on the full panel and qualitative results are similar. Second, they\ninclude a poverty gap variable in their regressions, which is the proportion of the population under the poverty line,\ntimes the average distance from the poverty line (Source: Povcal Net, World Bank _)._ Since the poverty gap is so tightly\nlinked to income levels, they use the residuals from a regression of the poverty gap on income in their regressions. We\ndid the same, but the variable provided little explanatory power and reduced our sample. Also, unlike the other variables\nin our analysis, the poverty counts are based on surveys that do not occur at regular intervals, and thus the panel is\nhighly unbalanced. For these reasons, we present models without the poverty gap variable.\n\n\n15\n\nMention: \"Povcal Net\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The population below the line times its average distance is computed here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005729:8:1", "occurrence_id": "33fda3198e27857a072b9014", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  household were randomly selected\n\nbased on a listing of all household members aged 16 years and above, using the Kish Grid-diagram. <sup>4</sup>\n\n\nWe use the FinAccess surveys in our analysis for a number of reasons. _First_, it captures important\n\ndimensions of financial inclusion as it was specifically designed to measure access to financial\n\nservices in Kenya. _Second,_ it also provides information on individual and household characteristics,\n\nsuch as education level, gender, asset ownership, income source, age group, language, location and\n\nhousehold size. _Third,_ it is a nationally representative household survey that covers the universe of\n\ndistricts in Kenya, alleviating potential sample selection bias problems. _Fourth,_ its implementation\n\nover different years allows us to explore the within-district time variation in financial inclusion.\n\n\n4 The Kish Grid-diagram is a method that uses a pre-assigned table of random numbers for selecting individuals within a\nhousehold to be interviewed.\n\n7\n\nMention: \"FinAccess surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Selected for the financial dimensions it captures among adults.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005730:6:0", "occurrence_id": "ea26bdea130d12580ef08ab4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: for taking a relativist approach. <sup>5</sup> [^5: For a new “weakly relative” measure that makes an explicit accounting for social cohesion, see Ravallion and\nChen (2011). For the implications for global poverty measurement, see Chen and Ravallion (2012).] It may well be that the average Colombian or Pole has a higher\neconomic standard of living than the average Western European or American in 1930; however,\nthey still aspire be as well off as the average Western European or American today.\n\n\n**Figure 2 (a) and (b): Two Views on a “Trap”—Relative and Absolute MIC Thresholds**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nSource: Authors’ elaboration based on data from Maddison (2010).\n\n\nThe relative or absolute approach also has an impact on the type of analysis that one\nmight undertake to understand the dynamics of a presumed MIC trap. The relative approach\nimmediately takes one down the path of two relevant threads of the economics literature: (a)\nabsolute convergence (or the lack thereof); and (b) inter-country distribution of income. On the\nformer, there is a vast theoretical literature on neo-classical growth models, <sup>6</sup> [^6: Parente and Prescott (2005), and Banerjee and Duflo (2005) present surveys from different perspectives.] and a vast empirical\nliterature on testing convergence or lack thereof (Pritchett, 1997). On the second thread, the\nliterature on the world distribution of income has noted that, in recent decades, widening global\ngaps between rich and poor individuals have been driven more by increases in the difference\nbetween countries’ incomes per capita than by differences in income across households within\ncountries (Bourguignon and Morrison, 2002; Milanovic, 2005).\n\nThe absolute definition might lead one to focus more on why some countries enter actual\nstagnation rather than failed “catching up.” Here issues of macroeconomic stability may have\nplayed a leading role—as noted, in particular\n\nMention: \"Maddison (2010)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Threshold elaborations rest on this historical database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005735:29:0", "occurrence_id": "97facd09490004c2af0dd110", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Domic´ılios (PNAD), a representative survey of the Brazilian population, are used.\n\n\nThe unit of observation is the household, and the individual outcome is measured as the monthly\n\n\nhousehold equivalent income, expressed in 2008 Brazilian real. <sup>32</sup> Household income is computed\n\n\nas the sum of all household members’ individual incomes, including earnings from all jobs, and all\n\n\nother reported income, including income from assets, pensions, and transfers.\n\n\nThe population is partitioned into 15 types using the information on two circumstances: region\n\n\nof birth and race. Region of birth is coded in five categories (North, Northeast, Southeast, South,\n\n\nCenter-west), and race is coded in three categories (white/east Asian, black/mixed race, and in\n\ndigenous). Individuals who were born abroad and those classified as “other” for the variable race\n\n\nare excluded because the number of observations is too low to make appropriate inference. Hence,\n\n\nthe sample sizes of each wave considered in this analysis are as follows: 366,388 households in 2002,\n\n\n390,046 in 2005, and 372,581 in 2008. <sup>33</sup> .\n\n\nThe full opportunity profiles for the three waves are reported in Table 4 in the appendix. <sup>34</sup> In\n\n\nthis table, it is clear that race is the main determinant of the disparity in opportunities. Consistent\n\n\nwith a number of contributions on socio-economic inequality in Brazil, racial relationships appear\n\n\nto be the major source of outcome and opportunity inequality in Brazil (Telles 2004; Bourguignon\n\n\net al. 2007; among others).\n\n\nTo analyze the distributional impact of growth in Brazil according to the EOp perspective,\n\n\ntwo three-year period growth processes are considered: 2002–05 and 2005–08. The choice of these\n\n\nparticular periods is driven by the observation that during these years, Brazil experienced\n\nMention: \"Domic´ılios (PNAD)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"A representative national household survey forming the observation unit.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005740:6:1", "occurrence_id": "018606696b82fbf760515d13", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \n|Rest of ECA|Rest of ECA|\n|||\n|||\n|||\n\n\n\n5 10 15 20 25 30\nDistribution of observations\n\n\nRussia Rest of ECA\n\n\n\n|Col1|Col2|\n|---|---|\n|Russia<br>Rest of the World|Russia<br>Rest of the World|\n|||\n|||\n\n\n0 5 10\nDistribution of observations\n\n\nRussia Rest of the World\n\n\nSource: Enterprise Surveys comprehensive dataset (May 2012)\n\n\nSize distribution of firms based on sales revenue (log)\nRussia vs. Rest of the World\n\n|Col1|Col2|\n|---|---|\n|Russia|Russia|\n|||\n|Rest of the World||\n|||\n|||\n\n\n\n0 10 20 30 40\nDistribution of observations\n\n\nRussia Rest of the World\n\n\n\nSource: Enterprise Surveys comprehensive dataset (May 2012)\n\n\n\nSource: Enteprise Surveys comprehensive dateset (May 2012)\n\n\n\nOf equal concern is the indication that the right-hand side of the size distribution of manufacturing Source: Enterprise Surveys comprehensive dataset (May 2012) Source: Enteprise Surveys comprehensive dateset (May 2012)\nfirms in Russia may also be shorter than that of other economies. In other words, the biggest firms\ndo not grow to be as big in Russia as in other parts of the world. Examining Figure 2 (above) once\nagain, the reader can see that the right-hand side tail of the distribution is also shorter for Russia\nthan in other economies. This finding calls into question whether even efficient firms get the\n<u>resources they require to grow in the Russian economy. In well-functioning economies, markets</u>\n<u>efficiently allocate resources to the most productive firms irrespective of their size and age (Hsieh</u>\n\n10 The data are taken from the World Bank’s Enterprise Surveys on May 2012. For each country, only the latest survey is\nused.\n\nMention: \"Enterprise Surveys comprehensive dataset\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The observation and size distribution across regions comes from this dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005740:6:1", "occurrence_id": "2244f5eb3de7aab1cfc9529e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \n|Rest of ECA|Rest of ECA|\n|||\n|||\n|||\n\n\n\n5 10 15 20 25 30\nDistribution of observations\n\n\nRussia Rest of ECA\n\n\n\n|Col1|Col2|\n|---|---|\n|Russia<br>Rest of the World|Russia<br>Rest of the World|\n|||\n|||\n\n\n0 5 10\nDistribution of observations\n\n\nRussia Rest of the World\n\n\nSource: Enterprise Surveys comprehensive dataset (May 2012)\n\n\nSize distribution of firms based on sales revenue (log)\nRussia vs. Rest of the World\n\n|Col1|Col2|\n|---|---|\n|Russia|Russia|\n|||\n|Rest of the World||\n|||\n|||\n\n\n\n0 10 20 30 40\nDistribution of observations\n\n\nRussia Rest of the World\n\n\n\nSource: Enterprise Surveys comprehensive dataset (May 2012)\n\n\n\nSource: Enteprise Surveys comprehensive dateset (May 2012)\n\n\n\nOf equal concern is the indication that the right-hand side of the size distribution of manufacturing Source: Enterprise Surveys comprehensive dataset (May 2012) Source: Enteprise Surveys comprehensive dateset (May 2012)\nfirms in Russia may also be shorter than that of other economies. In other words, the biggest firms\ndo not grow to be as big in Russia as in other parts of the world. Examining Figure 2 (above) once\nagain, the reader can see that the right-hand side tail of the distribution is also shorter for Russia\nthan in other economies. This finding calls into question whether even efficient firms get the\n<u>resources they require to grow in the Russian economy. In well-functioning economies, markets</u>\n<u>efficiently allocate resources to the most productive firms irrespective of their size and age (Hsieh</u>\n\n10 The data are taken from the World Bank’s Enterprise Surveys on May 2012. For each country, only the latest survey is\nused.\n\nMention: \"Enteprise Surveys comprehensive dateset\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"A second source note for the same edition behind the size distribution.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005740:6:1", "occurrence_id": "f5e2a8d40b4224c98029bc82", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \n|Rest of ECA|Rest of ECA|\n|||\n|||\n|||\n\n\n\n5 10 15 20 25 30\nDistribution of observations\n\n\nRussia Rest of ECA\n\n\n\n|Col1|Col2|\n|---|---|\n|Russia<br>Rest of the World|Russia<br>Rest of the World|\n|||\n|||\n\n\n0 5 10\nDistribution of observations\n\n\nRussia Rest of the World\n\n\nSource: Enterprise Surveys comprehensive dataset (May 2012)\n\n\nSize distribution of firms based on sales revenue (log)\nRussia vs. Rest of the World\n\n|Col1|Col2|\n|---|---|\n|Russia|Russia|\n|||\n|Rest of the World||\n|||\n|||\n\n\n\n0 10 20 30 40\nDistribution of observations\n\n\nRussia Rest of the World\n\n\n\nSource: Enterprise Surveys comprehensive dataset (May 2012)\n\n\n\nSource: Enteprise Surveys comprehensive dateset (May 2012)\n\n\n\nOf equal concern is the indication that the right-hand side of the size distribution of manufacturing Source: Enterprise Surveys comprehensive dataset (May 2012) Source: Enteprise Surveys comprehensive dateset (May 2012)\nfirms in Russia may also be shorter than that of other economies. In other words, the biggest firms\ndo not grow to be as big in Russia as in other parts of the world. Examining Figure 2 (above) once\nagain, the reader can see that the right-hand side tail of the distribution is also shorter for Russia\nthan in other economies. This finding calls into question whether even efficient firms get the\n<u>resources they require to grow in the Russian economy. In well-functioning economies, markets</u>\n<u>efficiently allocate resources to the most productive firms irrespective of their size and age (Hsieh</u>\n\n10 The data are taken from the World Bank’s Enterprise Surveys on May 2012. For each country, only the latest survey is\nused.\n\nMention: \"Enterprise Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The latest round per country is retained for each economy.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005745:18:0", "occurrence_id": "7269d0f3577b4093d007ec86", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: We observe that the level of the NIM is consistently lower than the SELIC rate (Figure 5b). This means\nthat the returns of banks are lower than by just lending to the government, which is theoretically risk\nfree. This is counter-intuitive because theory suggests that by taking risk, lending to the individuals\nand corporations, banks should theoretically earn a higher return than by lending to the government. A\npossible explanation for this is that banks underestimate default risk and the realized default rates are\nhigher than the anticipated default rates. However, the fact that the SELIC rate is consistently higher\nthan NIM is of concern. A risk-averse bank would choose to lend risk free to the government and get\nhigher returns. There is an arbitrage opportunity that is not being exploited by the banking sector. <sup>11</sup> [^11: Another possibility would be that the banks believe that the government would not repay its debts. Nevertheless,\nthis rationale is improbable given that the Brazilian government’s debt ratings have improved significantly and the\ndebt to GDP ratio has decreased.]\n\nEven when we take out the competition factors from the pure spread, in our Step 2 regression, we see\nthat macro-factors are still a very small part of the NIM in 2009 (Figure 6a and 6b). This supports our\nargument that macro-factors only play a small part in the determination of spreads in Brazil and microfactors are the most significant (Figure 6b).\n\n**Figure 6a. Breakdown of NIM Spread in Brazil** **Figure 6b. Breakdown of NIM Spread as % in**\n**<u>(2009)</u>** **<u>Brazil (2009)</u>**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n**_5.2 Brazil's Bank Spreads in an International Context_**\n\nMany studies have concluded that spreads in Brazil are the highest in the world (generally as measured\nby the data provided by IMF’s IFS). We find that Brazil’s IFS spreads\n\nMention: \"data provided by IMF’s IFS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Lending spreads across countries are measured with these series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005746:25:0", "occurrence_id": "95bf3d9b77243d0f2e9592d9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ###### Table 1a: Manufacturing employment distribution by gender of owner and sector\n\nValue relative to 1994\n\n<u>1994</u> <u>2000</u> <u>2005</u> <u>2000</u> <u>2005</u>\n\nA. Employment weighted distribution in shares\n\nTotal organized share 0.197 0.165 0.185 0.839 0.941\nTotal unorganized share 0.803 0.835 0.815 1.039 1.014\nFemale proprietorships 0.092 0.136 0.187 1.477 2.027\nMale proprietorships 0.663 0.653 0.577 0.985 0.870\nOther categories 0.048 0.045 0.051 0.944 1.054\n\n\nB. Employment weighted distribution in raw counts (in 000s)\n\nTotal sector size 34,424 40,702 40,336 1.182 1.172\nTotal organized value 6,775 6,723 7,470 0.992 1.103\nTotal unorganized value 27,649 33,979 32,866 1.229 1.189\nFemale proprietorships 3,180 5,554 7,555 1.746 2.376\nMale proprietorships 22,813 26,576 23,265 1.165 1.020\nOther categories 1,656 1,849 2,046 1.116 1.235\n\n\nC. Household distribution for female proprietorships\n\nHousehold value of female businesses 2,882 4,934 6,800 1.712 2.359\nHousehold share of female businesses 0.906 0.888 0.900 0.980 0.993\nFemale household share of total sector 0.084 0.121 0.169 1.448 2.013\n\nNotes: Indian descriptive statistics taken from Annual Survey of Industries and National Sample Statistics.\n\n\n24\n\nMention: \"Annual Survey of Industries\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Sector shares in the descriptive statistics come from this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005746:25:0", "occurrence_id": "faa474792668d1b90b930624", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ###### Table 1a: Manufacturing employment distribution by gender of owner and sector\n\nValue relative to 1994\n\n<u>1994</u> <u>2000</u> <u>2005</u> <u>2000</u> <u>2005</u>\n\nA. Employment weighted distribution in shares\n\nTotal organized share 0.197 0.165 0.185 0.839 0.941\nTotal unorganized share 0.803 0.835 0.815 1.039 1.014\nFemale proprietorships 0.092 0.136 0.187 1.477 2.027\nMale proprietorships 0.663 0.653 0.577 0.985 0.870\nOther categories 0.048 0.045 0.051 0.944 1.054\n\n\nB. Employment weighted distribution in raw counts (in 000s)\n\nTotal sector size 34,424 40,702 40,336 1.182 1.172\nTotal organized value 6,775 6,723 7,470 0.992 1.103\nTotal unorganized value 27,649 33,979 32,866 1.229 1.189\nFemale proprietorships 3,180 5,554 7,555 1.746 2.376\nMale proprietorships 22,813 26,576 23,265 1.165 1.020\nOther categories 1,656 1,849 2,046 1.116 1.235\n\n\nC. Household distribution for female proprietorships\n\nHousehold value of female businesses 2,882 4,934 6,800 1.712 2.359\nHousehold share of female businesses 0.906 0.888 0.900 0.980 0.993\nFemale household share of total sector 0.084 0.121 0.169 1.448 2.013\n\nNotes: Indian descriptive statistics taken from Annual Survey of Industries and National Sample Statistics.\n\n\n24\n\nMention: \"National Sample Statistics\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The companion household source for the same statistics.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005726:11:1", "occurrence_id": "8bce39c196047e277b09729d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  gap between the ERSS livestock survey and our RGCS study for which we\ncompensate in the analysis. Though each survey has its shortcomings, our expectation was that the\nsurveys should agree in the aggregate.\n\nTo construct comparable measures, we use retrospective questions about livestock in the RGCS\nto derive the number of livestock the household owned at the time of the ERSS survey. These questions\naccount for slaughter, loss, death, purchase, and birth over the six months prior to the survey. We\ncalculate two weighted estimates, one using the base weights and the other using the weight which\nadjusts for the proportion of the selected circle actually observed, as discussed above.\n\n\nTable 3: Weighted Estimates of Total Livestock in Study Area and Average Livestock Held by Household\n\n(Conditional on Ownership), by Animal Type\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n|Col1|Mean (SE)|Col3|Col4|Total (SE)|Col6|Col7|\n|---|---|---|---|---|---|---|\n||RGCS<br>(unadjusted<br>weights)|RGCS<br>(adjusted<br>weights)|ERSS|RGCS<br>(unadjusted<br>weights)|RGCS (adjusted<br>weights)|ERSS|\n|Cattle|10.4<br>(1.5)|10.8<br>(1.8)|15.3<br>(3.3)|153,505<br>(34,384)|186,164<br>(51,283)|1,092,752<br>(367,307)|\n|Camels|8.1<br>(1.4)|7.7<br>(1.4)|6.2<br\n\nMention: \"ERSS livestock survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its livestock counts differ from the other study, a gap compensated in analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005726:11:1", "occurrence_id": "2cf55bff0d696c51f5ddac0e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  gap between the ERSS livestock survey and our RGCS study for which we\ncompensate in the analysis. Though each survey has its shortcomings, our expectation was that the\nsurveys should agree in the aggregate.\n\nTo construct comparable measures, we use retrospective questions about livestock in the RGCS\nto derive the number of livestock the household owned at the time of the ERSS survey. These questions\naccount for slaughter, loss, death, purchase, and birth over the six months prior to the survey. We\ncalculate two weighted estimates, one using the base weights and the other using the weight which\nadjusts for the proportion of the selected circle actually observed, as discussed above.\n\n\nTable 3: Weighted Estimates of Total Livestock in Study Area and Average Livestock Held by Household\n\n(Conditional on Ownership), by Animal Type\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n|Col1|Mean (SE)|Col3|Col4|Total (SE)|Col6|Col7|\n|---|---|---|---|---|---|---|\n||RGCS<br>(unadjusted<br>weights)|RGCS<br>(adjusted<br>weights)|ERSS|RGCS<br>(unadjusted<br>weights)|RGCS (adjusted<br>weights)|ERSS|\n|Cattle|10.4<br>(1.5)|10.8<br>(1.8)|15.3<br>(3.3)|153,505<br>(34,384)|186,164<br>(51,283)|1,092,752<br>(367,307)|\n|Camels|8.1<br>(1.4)|7.7<br>(1.4)|6.2<br\n\nMention: \"RGCS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its retrospective questions reconstruct holdings at the earlier date.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005726:11:1", "occurrence_id": "070357b51a0bcd1e4e16bb9e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  gap between the ERSS livestock survey and our RGCS study for which we\ncompensate in the analysis. Though each survey has its shortcomings, our expectation was that the\nsurveys should agree in the aggregate.\n\nTo construct comparable measures, we use retrospective questions about livestock in the RGCS\nto derive the number of livestock the household owned at the time of the ERSS survey. These questions\naccount for slaughter, loss, death, purchase, and birth over the six months prior to the survey. We\ncalculate two weighted estimates, one using the base weights and the other using the weight which\nadjusts for the proportion of the selected circle actually observed, as discussed above.\n\n\nTable 3: Weighted Estimates of Total Livestock in Study Area and Average Livestock Held by Household\n\n(Conditional on Ownership), by Animal Type\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n|Col1|Mean (SE)|Col3|Col4|Total (SE)|Col6|Col7|\n|---|---|---|---|---|---|---|\n||RGCS<br>(unadjusted<br>weights)|RGCS<br>(adjusted<br>weights)|ERSS|RGCS<br>(unadjusted<br>weights)|RGCS (adjusted<br>weights)|ERSS|\n|Cattle|10.4<br>(1.5)|10.8<br>(1.8)|15.3<br>(3.3)|153,505<br>(34,384)|186,164<br>(51,283)|1,092,752<br>(367,307)|\n|Camels|8.1<br>(1.4)|7.7<br>(1.4)|6.2<br\n\nMention: \"ERSS survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The reference date to which the retrospective reconstruction refers.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005748:14:0", "occurrence_id": "a52cac0e0e88ff9232c9c939", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: the time of their fifth interview. Similar to the other two countries, we derive four year-on-year\n\n\ntransitions matrices that cover the period between 2007 and 2011.\n\n\nAs mentioned in the introduction, the two countries analyzed herein cover a large range\n\n\nof informality rates observed in Latin America. Existing estimates suggest that the informal\n\n\nsectors in Mexico are large relative to other countries in the region, whereas Brazil tends to\n\n\nhave lower informality rates than other Latin American economies. According to existing calcu\n\nlations for the case of Mexico, the size of the informal sector represents about 60% of employed\n\n\nsalaried male workers and 55% of females. Brazil’s informality appears to be below the average\n\n\nof Latin America, with a rate of informality of about 32% of employed males and 33% for\n\n\nfemales <sup>4</sup> .\n\n\nTable 1 presents summary statistics of the data used for the estimation outlined in the\n\n\nprevious section. In Mexico, the size of the informal sector (59.9% of the employed workers)\n\n\nand the employment shares (60.45%, which corresponds to a residual sector of 39.55%) are\n\n\nalso consistent with estimates found in the literature. For example, Bosch and Maloney (2010)\n\n\nreport an informal sector of 38% and a residual sector of 20%. However, their sample includes\n\n\nonly men in urban areas (using the “Encuesta Nacional de Empleo Urbano,” ENEU, from\n\n\n1984-2004). Although we use the ENOE, which is representative of both urban and rural areas,\n\n\nwith our sample of men in urban areas we obtain similar rates as Bosch and Maloney (2010).\n\n\nNonetheless, it is noteworthy that the residual sector for both urban and rural male workers\n\n\nin our data is almost exactly 20%, whereas the informality rate of 49.0% is unsurprisingly\n\n\nmuch higher than the 38% for urban males, because self-employment\n\nMention: \"Encuesta Nacional de Empleo Urbano\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its urban male-only sample limits the earlier estimate.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005748:14:0", "occurrence_id": "f71d183edfdf0fb94e671ffd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: the time of their fifth interview. Similar to the other two countries, we derive four year-on-year\n\n\ntransitions matrices that cover the period between 2007 and 2011.\n\n\nAs mentioned in the introduction, the two countries analyzed herein cover a large range\n\n\nof informality rates observed in Latin America. Existing estimates suggest that the informal\n\n\nsectors in Mexico are large relative to other countries in the region, whereas Brazil tends to\n\n\nhave lower informality rates than other Latin American economies. According to existing calcu\n\nlations for the case of Mexico, the size of the informal sector represents about 60% of employed\n\n\nsalaried male workers and 55% of females. Brazil’s informality appears to be below the average\n\n\nof Latin America, with a rate of informality of about 32% of employed males and 33% for\n\n\nfemales <sup>4</sup> .\n\n\nTable 1 presents summary statistics of the data used for the estimation outlined in the\n\n\nprevious section. In Mexico, the size of the informal sector (59.9% of the employed workers)\n\n\nand the employment shares (60.45%, which corresponds to a residual sector of 39.55%) are\n\n\nalso consistent with estimates found in the literature. For example, Bosch and Maloney (2010)\n\n\nreport an informal sector of 38% and a residual sector of 20%. However, their sample includes\n\n\nonly men in urban areas (using the “Encuesta Nacional de Empleo Urbano,” ENEU, from\n\n\n1984-2004). Although we use the ENOE, which is representative of both urban and rural areas,\n\n\nwith our sample of men in urban areas we obtain similar rates as Bosch and Maloney (2010).\n\n\nNonetheless, it is noteworthy that the residual sector for both urban and rural male workers\n\n\nin our data is almost exactly 20%, whereas the informality rate of 49.0% is unsurprisingly\n\n\nmuch higher than the 38% for urban males, because self-employment\n\nMention: \"ENOE\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Representative of both urban and rural areas, unlike its predecessor.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005749:40:0", "occurrence_id": "eeeeb5a7d18ace7efd33b6f8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 39\n\n\nilliteracy. Alternatively, there may be a lack of profitable investment projects in the economy\n\n\nthat deserve financing based on their expected return. This problem can actually not be\n\n\nillustrated in our figure as it focuses on loan applicants. A second type of access problem can\n\n\narise from regulatory distortions or insufficient contestability that cause lenders not to fully\n\n\nexploit all the outreach opportunities and thus settle at a point below the access possibilities\n\n\nfrontier with a higher marginal interest rate (Figure 10, Point II, B). A third and very\n\n\ndifferent access problem is associated with “excess access,” that is, an equilibrium above the\n\n\naccess possibilities frontier with loans being granted to a larger share of loan applicants than\n\n\nis prudently warranted or SMEs achieve too high a leverage, given the lending interest rate\n\n\nand the institutional framework (Figure 10, Point III, C). A final access problem consists of\n\n\ntoo low a prudent access possibilities frontier, caused by deficiencies in an economy’s\n\n\ninstitutional framework compared with that of countries with similar levels of economic\n\n\ndevelopment. An improvement along these lines would lead to an expansion of the frontier\n\n\nfrom S* to S*’’ in Figure 11. Similarly, lower opportunity costs of funding (imc), e.g. due to\n\n\nbetter macroeconomic conditions, will increase the universe of potential loan applicants\n\n\nreceiving finance (Figure 11, Point II, B).\n\n\n**6.2.** **Matching the concept with data**\n\n\nAs in the case of aggregate indicators, the frontier concept can be matched with\n\n\nempirical data. First, using the benchmarking model described above, one can predict the\n\n\nshare of (small) firms with access to a credit line or loan by a formal financial institution and\n\n\ncompare them to the actual share, as gauged by the Enterprise Surveys, undertaken by the\n\n\nWorld Bank/IFC. Figure 12 graphs the gap between the predicted and the actual share of\n\n\nsmall enterprises that use a credit line or loan from a formal financial\n\nMention: \"the Enterprise Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Actual loan uptake by firms is gauged in these surveys.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005750:45:0", "occurrence_id": "748a685bfb1536b769d62aa4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: - **Armed Conflict Database**      - <mark>This dataset is a collaborative project between the Department of</mark>\n<mark>Peace and Conflict Research at Uppsala University and the Centre for the Study of Civil War at</mark>\n<mark>PRIO. The database includes armed conflicts, both internal and external, in the period 1946 to the</mark>\n<mark>present.</mark> [(http://www.prio.no/Data/Armed-Conflict/UCDP-PRIO/)](http://www.prio.no/Data/Armed-Conflict/UCDP-PRIO/)\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n|Variables sources and definitions|Col2|Col3|\n|---|---|---|\n|<br>|**Source**<br>|**Definition**|\n|**Structure of the Economy**<br> <br> <br>|**Structure of the Economy**<br> <br> <br>|**Structure of the Economy**<br> <br> <br>|\n|GDP per capita<br>|WDI<br>|GDP per capita is gross domestic product divided by midyear population.<br>|\n|Industry, value added to<br>GDP (%)<br>|WDI<br>|Industry corresponds to ISIC divisions 10-45 and includes manufacturing (ISIC<br>divisions 15-37). It comprises value added in mining, manufacturing (also<br>reported as a separate subgroup), construction, electricity, water, and gas. Value <br>added is the net output of a sector after adding up all outputs and subtracting<br>intermediate inputs. It is calculated without making deductions for depreciation of\n\nMention: \"Armed Conflict Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"A collaborative departmental project listing conflict episodes.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005752:30:0", "occurrence_id": "bc9b1a68e0b2f027664cec76", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |Table 1: Reasons for attending suppleme (percent), Vietnam 2006|entary education clas|sses for students age|\n|---|---|---|\n||**Tutoring organized**<br>**by school**|**Tutoring not**<br>**organized by school**|\n|Prepare for examinations|47.2|41.7|\n|<br>Do not catch up with the class|12.9|14.4|\n|<br>Acquire skills for future employment|12.2|12.7|\n|<br>Like this subject|6.4|11.3|\n|<br>Parents too busy to take care|2.7|1.6|\n|<br>Poor quality lessons in school|2.7|6.0|\n|<br>Subjects not taught in mainstream classes|0.5|1.5|\n|<br>Others|15.4|10.9|\n||||\n|Total|100|100|\n|N|376|301|\n|Source: Household and Tutoring Survey 2007- 2008.|Source: Household and Tutoring Survey 2007- 2008.||\n\n\n29\n\nMention: \"Household and Tutoring Survey 2007- 2008\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Sample sizes of 376 and 301 households underlie the class participation shares.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005753:14:1", "occurrence_id": "d0ea4416fc1921350ef2029a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  80 percent of MFIs in Bangladesh charge 11\n15 percent interest rate, and about 1 percent charges more than 20 percent (Rahman (2003)). Starting\n\nfrom July 2004, the wholesale microcredit fund provider PKSF capped the interest rate at 12.5 percent\n\nflat.\n\nThe average MFI coverage rate is about 42 percent in our sample of villages (Table A.1) which is\n\ncomparable to coverage rate from our panel data (38 percent). According to the Household Income and\n\nExpenditure Survey (HIES) 2010, about 45 percent of households with less than an acre of land in the\n\nRangpur division covering areas included in our sample are active borrowers from MFIs. The summary\n\nstatistics for all other variables used in the regression are also reported in Table A.1.\n\n\n**2.4. OLS, MATCHING AND IPW ESTIMATES**\n\n\nWe start with the simplest specification where the moneylender interest rate is regressed on the\n\nMFI coverage dummy (D=1 if coverage in a village is more than the mean coverage rate) without any\n\n23 According to one estimate reported in late eighties, the average interest rate charged by moneylenders was 51.86\npercent in rural India (Dasgupta, 1989); Aleem (1993) reports an average lending rate of 78.5 percent in Pakistan.\nFor a summary of the evidence on informal interest rates in developing countries see Banerjee (2003).\n24 It is not uncommon to have 25-50 percent interest rate for a consumption loan for a month, which becomes\nextremely high interest rates when annualized. Most of the moneylender interest rates reported in the literature are\nannualized rates on short term consumption loans.\n25 Note, however, that it assumes that the repayment schedule is enforced strictly, which is unrealistic, for both the\nmoneylender loans and MFI loans.\n\n\n13\n\nMention: \"Household Income and\n\nExpenditure Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its coverage rate for smallholders benchmarks the panel figure.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005759:60:2", "occurrence_id": "41048ed97573ef247dee3533", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  taxes and minus any subsidies not included in the value<br>of the products. It is calculated without making deductions for depreciation of fabricated assets or for<br>depletion and degradation of natural resources taking the average from 2007 to the latest of World<br>Development Indicators in the World Bank database.|41|4.61|4.55|0.51|8.17|1.95|\n|Rule of law|Rule of law|Country|Rule of law captures perceptions of the extent to which agents have confidence in and abide by the<br>rules of society, and in particular the quality of contract enforcement, property rights, the police, and<br>the courts, as well as the likelihood of crime and violence derived from the Worldwide Governance<br>Indicators (WGI) in the World Bank database.|41|-0.59|-0.63|-1.58|1.27|0.53|\n|Bank overhead costs / Total assets|Bank overhead costs / Total assets|Country|Accounting value of a bank's overhead costs as a share of its total assets, taking the average from<br>2007 to the latest derived from Financial Development and Structure Dataset in the World Bank<br>database.|41|0.05|0.05|0.01|0.17|0.04|\n|Ease of doing business index<br>(1=most business-friendly<br>regulations)<br>|Ease of doing business index<br>(1=most business-friendly<br>regulations)<br>|Country|Ease of doing business ranks economies from 1 to 185, with first place being the best. A high ranking<br>(a low numerical rank) means that the regulatory environment is conducive to business operation.<br>The index averages\n\nMention: \"World Bank database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Averages from that year to the latest carry the perception indicators.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005759:60:2", "occurrence_id": "30c3fe286e712abff7a9c637", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  taxes and minus any subsidies not included in the value<br>of the products. It is calculated without making deductions for depreciation of fabricated assets or for<br>depletion and degradation of natural resources taking the average from 2007 to the latest of World<br>Development Indicators in the World Bank database.|41|4.61|4.55|0.51|8.17|1.95|\n|Rule of law|Rule of law|Country|Rule of law captures perceptions of the extent to which agents have confidence in and abide by the<br>rules of society, and in particular the quality of contract enforcement, property rights, the police, and<br>the courts, as well as the likelihood of crime and violence derived from the Worldwide Governance<br>Indicators (WGI) in the World Bank database.|41|-0.59|-0.63|-1.58|1.27|0.53|\n|Bank overhead costs / Total assets|Bank overhead costs / Total assets|Country|Accounting value of a bank's overhead costs as a share of its total assets, taking the average from<br>2007 to the latest derived from Financial Development and Structure Dataset in the World Bank<br>database.|41|0.05|0.05|0.01|0.17|0.04|\n|Ease of doing business index<br>(1=most business-friendly<br>regulations)<br>|Ease of doing business index<br>(1=most business-friendly<br>regulations)<br>|Country|Ease of doing business ranks economies from 1 to 185, with first place being the best. A high ranking<br>(a low numerical rank) means that the regulatory environment is conducive to business operation.<br>The index averages\n\nMention: \"Financial Development and Structure Dataset\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"External claims as a share of assets come from this dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005764:6:0", "occurrence_id": "2b9493d12b07b6e7632e3477", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: account at a post office. <sup>4</sup> [^4: For seven countries (Algeria, Congo, Rep., Egypt, Arab Rep., Greece, Morocco, Tunisia, and Yemen, Rep.) the\ndata is for 2012 instead of 2011 because of data quality issues in the 2011 round.] In our sample, we focus on the over 65,000 individuals who were\n\n\ninterviewed in 60 of these countries. <sup>5</sup> [^5: We drop Madagascar from our sample because more than 20% of the population is not sampled and Taiwan, China\nbecause no country-level post office information is available.] Appendix 1 lists all the countries included in our sample.\n\n\nThe Global Findex data allow us to distinguish between three types of account\n\n\nownership: an account at a financial institution only, accounts at both a financial institution and\n\n\nthe post office, and an account at the post office only. Within our sample of countries that offer\n\n\npostal accounts, we find that 49.9 percent of adults have an account at either or both institutions,\n\n\n11.6 percent of adults have an account at the post office, and 2.8 percent of adults have an\n\n\naccount at the post office only (5.6 percent of account holders).\n\n\nFigure 1 shows a wide variation in overall account ownership between high income and\n\n\ndeveloping countries. While account penetration is close to universal in the developed countries\n\n\nincluded in our sample (90 percent), on average, only 44 percent of adults in developing\n\n\ncountries offering postal financial services report having an account. <sup>6</sup> [^6: This compares to 89 percent account penetration in high income countries and 41 percent account penetration in\ndeveloping countries for the full sample of countries in the Global Findex database. The data are population\nweighted.] However, beyond account\n\n\npenetration, we are interested in the types of accounts owned by individuals. In most of countries\n\n\nin our sample\n\nMention: \"Global Findex data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"These data distinguish three types of account ownership.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005772:14:2", "occurrence_id": "7a8901831ff84559441721be", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  _RWi,c,r_ ) are calculated as\n\nthe average expected real wage of the entire labor force, regardless of employment status;\n\n\n\nemployment once the migrant arrives in the host region. Real wages ( _RWi,c,r_ ) are calculated as\n\nthe average expected real wage of the entire labor force, regardless of employment status;\n\nhence the expected real wage that a migrant faces once the migrant arrives in the host region\n\n\n\nthe average expected real wage of the entire labor force, regardless of employment status;\n\nhence the expected real wage that a migrant faces once the migrant arrives in the host region\n\nincludes unemployment. If unemployment in the host economy rises then the expected real\n\n\n\nhence the expected real wage that a migrant faces once the migrant arrives in the host region\n\nincludes unemployment. If unemployment in the host economy rises then the expected real\n\nwage falls, due to a higher probability of unemployment, and hence migration falls. This reflects\n\n\n\nincludes unemployment. If unemployment in the host economy rises then the expected real\n\nwage falls, due to a higher probability of unemployment, and hence migration falls. This reflects\n\na main assumption of the Harris-Todaro model that the migration decision between rural and\n\n\n\nwage falls, due to a higher probability of unemployment, and hence migration falls. This reflects\n\na main assumption of the Harris-Todaro model that the migration decision between rural and\n\nurban areas is based on expected income differentials, rather than just wage differentials.\n\n\n\na main assumption of the Harris-Todaro model that the migration decision between rural and\n\nurban areas is based on expected income differentials, rather than just wage differentials.\n\nUnderlying the _GMig2Dyn_ model is a database that captures both bilateral labor (GMig2\n\n\n\nurban areas is based on expected income differentials, rather than just wage differentials.\n\nUnderlying the _GMig2Dyn_ model is a database that captures both bilateral labor (GMig2\n\nDatabase) and\n\nMention: \"GMig2\n\nDatabase\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Bilateral labour flows underlying the migration model come from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005761:3:1", "occurrence_id": "f5a1be97674140d229b319a1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  well as for their neighbors who do not.\n\nWhile the transition to cleaner sources of domestic energy will ultimately reduce the\n\ndisease burden associated with this risk factor, during the transition good ventilation, improved\n\nstoves, and public information leading to behavior change can significantly reduce exposure to\n\nhousehold smoke. While reduction of HAP has been recognized as an important development\n\nobjective, the design of HAP reduction strategies has been hindered by lack of information about\n\nactual air quality in households and health benefits of potential mitigation measures. Data on\n\nHAP have been scarce because monitoring in households has been difficult and costly. While\n\nmany studies have measured gaseous and sometimes particulate emissions from standardized or\n\nlaboratory tests, it is not clear whether these emissions are representative of cook-stoves in use. <sup>9</sup>\n\n\n4 WHO, 2009\n5 The Global Burden of Disease Study 2010 is a collaborative project of nearly 500 researchers in 50 countries led\nby the Institute for Health Metrics and Evaluation (IHME) at the University of Washington.\n6 Lim et al., 2012\n7 Disability-adjusted life years (DALYs) quantify both premature mortality and disability within a population.\n8 Ibid.\n9 Roden et al., 2009\n\n\n2\n\nMention: \"Global Burden of Disease Study 2010\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"A collaborative effort of nearly five hundred researchers in fifty countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005776:20:0", "occurrence_id": "47e4c574f2a03f8b7b50e953", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: and service sectors. Most of the firms in the country are in the trade industry\n\n\n\nand service sectors. Most of the firms in the country are in the trade industry\n\n(around 57% of all the firms). However, firms in all three sectors are similarly likely\n\n\n\nand service sectors. Most of the firms in the country are in the trade industry\n\n(around 57% of all the firms). However, firms in all three sectors are similarly likely\n\nto sign up for e-filing, which is around 15% in trade and 18% in service sectors. Lack\n\n\n\n(around 57% of all the firms). However, firms in all three sectors are similarly likely\n\nto sign up for e-filing, which is around 15% in trade and 18% in service sectors. Lack\n\nof detailed data on sectoral allocation of firms limits our ability to say more on this.\n\n\n\nto sign up for e-filing, which is around 15% in trade and 18% in service sectors. Lack\n\nof detailed data on sectoral allocation of firms limits our ability to say more on this.\n\n\n\n**2.1.3 Ukraine**\n\nAccording to the TCC survey carried out in Ukraine as of 2007 fiscal year\n\n(excluding proprietorships, which were covered in a separate survey), 21% of\n\n\n\n**2.1.3 Ukraine**\n\n\n\nAccording to the TCC survey carried out in Ukraine as of 2007 fiscal year\n\n(excluding proprietorships, which were covered in a separate survey), 21% of\n\ncompanies had signed up for e-filing. <sup>17</sup> On the other hand, floppy disk usage in filing\n\n\n\n(excluding proprietorships, which were covered in a separate survey), 21% of\n\ncompanies had signed up for e-filing. <sup>17</sup> On the other hand, floppy disk usage in filing\n\nwas 26%, which was slightly higher than electronic filing. One reason may be\n\nMention: \"TCC survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Conducted for that fiscal year, excluding proprietorships.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005779:0:0", "occurrence_id": "a4204fc79663241e60f83173", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ## WPS6651 Policy Research Working Paper 6651\n\n# Demand Collapse or Credit Crunch to Firms?\n\n## Evidence from the World Bank’s Financial Crisis Survey in Eastern Europe\n\n#### _Ha Nguyen_ _Rong Qian_ The World Bank Development Research Group Macroeconomics and Growth Team October 2013\n\nMention: \"the World Bank’s Financial Crisis Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Firm demand and credit conditions across Eastern Europe are examined in this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005782:15:0", "occurrence_id": "9b8422619879d45d1072de58", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 14\n\n\nin the late 1990s. Since neither series overlaps sufficiently with the WVS data, we turn instead to the\n\n\nonly expenditure variable that is available throughout the 1990s and 2000s and that has broad country\n\n\ncoverage, general government final consumption expenditure from the WDI.\n\n\nThis variable differs from the IFS series in two ways. First, it includes general rather than central\n\n\ngovernment spending and so includes government spending at subnational levels. However, coverage\n\n\nof subnational spending is sporadic and unlikely to be a significant issue. Second, and more importantly,\n\n\nfinal government consumption expenditure excludes the capital budget. Since previous research has\n\n\nidentified capital spending as one source of funds for pre-electoral mobilization, this is problematic.\n\n\nHowever, Block (2002) finds for a sample of African countries that fiscal expansions around elections\n\n\nare particularly pronounced in government consumption expenditure and the correlation between this\n\n\nvariable and the IFS data is 0.7. Assuming that cycles in government consumption and capital\n\n\nexpenditure are related, this measure should be a good approximation of total government expenditure.\n\n\nMoreover, any bias generated by using the general government measure is likely to be against the\n\n\nhypothesized relationship between budget cycles and party characteristics. Subnational spending\n\n\nintroduces noise into central government spending, making it more difficult to detect cycles in central\n\n\ngovernment spending. Similarly, to the extent that capital spending is more important for pre-electoral\n\n\nexpenditures, it should once again be more difficult to detect cycles in spending using the general\n\n\ngovernment consumption data.\n\n\nPrevious research has not addressed the noise that hyperinflation can introduce into analyses of\n\n\npolitical budget cycles. Hyperinflation generates very large within-fiscal-year differences in the real value\n\n\nof nominal government spending. As a consequence, small errors in assumptions regarding the timing\n\n\nof election spending can have large effects on the estimation. For example, if the election falls in the\n\n\nbeginning of the fiscal year, even a large increase in government spending might go unnoticed, since\n\n\nnominal spending could be one-half of the\n\nMention: \"WVS data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its series does not overlap sufficiently with the consumption measures.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005782:15:0", "occurrence_id": "0553d057f448c50926839553", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 14\n\n\nin the late 1990s. Since neither series overlaps sufficiently with the WVS data, we turn instead to the\n\n\nonly expenditure variable that is available throughout the 1990s and 2000s and that has broad country\n\n\ncoverage, general government final consumption expenditure from the WDI.\n\n\nThis variable differs from the IFS series in two ways. First, it includes general rather than central\n\n\ngovernment spending and so includes government spending at subnational levels. However, coverage\n\n\nof subnational spending is sporadic and unlikely to be a significant issue. Second, and more importantly,\n\n\nfinal government consumption expenditure excludes the capital budget. Since previous research has\n\n\nidentified capital spending as one source of funds for pre-electoral mobilization, this is problematic.\n\n\nHowever, Block (2002) finds for a sample of African countries that fiscal expansions around elections\n\n\nare particularly pronounced in government consumption expenditure and the correlation between this\n\n\nvariable and the IFS data is 0.7. Assuming that cycles in government consumption and capital\n\n\nexpenditure are related, this measure should be a good approximation of total government expenditure.\n\n\nMoreover, any bias generated by using the general government measure is likely to be against the\n\n\nhypothesized relationship between budget cycles and party characteristics. Subnational spending\n\n\nintroduces noise into central government spending, making it more difficult to detect cycles in central\n\n\ngovernment spending. Similarly, to the extent that capital spending is more important for pre-electoral\n\n\nexpenditures, it should once again be more difficult to detect cycles in spending using the general\n\n\ngovernment consumption data.\n\n\nPrevious research has not addressed the noise that hyperinflation can introduce into analyses of\n\n\npolitical budget cycles. Hyperinflation generates very large within-fiscal-year differences in the real value\n\n\nof nominal government spending. As a consequence, small errors in assumptions regarding the timing\n\n\nof election spending can have large effects on the estimation. For example, if the election falls in the\n\n\nbeginning of the fiscal year, even a large increase in government spending might go unnoticed, since\n\n\nnominal spending could be one-half of the\n\nMention: \"WDI\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"General government final consumption with broad coverage is taken from here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005782:15:0", "occurrence_id": "7f3cccf723ca8be6d9942551", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 14\n\n\nin the late 1990s. Since neither series overlaps sufficiently with the WVS data, we turn instead to the\n\n\nonly expenditure variable that is available throughout the 1990s and 2000s and that has broad country\n\n\ncoverage, general government final consumption expenditure from the WDI.\n\n\nThis variable differs from the IFS series in two ways. First, it includes general rather than central\n\n\ngovernment spending and so includes government spending at subnational levels. However, coverage\n\n\nof subnational spending is sporadic and unlikely to be a significant issue. Second, and more importantly,\n\n\nfinal government consumption expenditure excludes the capital budget. Since previous research has\n\n\nidentified capital spending as one source of funds for pre-electoral mobilization, this is problematic.\n\n\nHowever, Block (2002) finds for a sample of African countries that fiscal expansions around elections\n\n\nare particularly pronounced in government consumption expenditure and the correlation between this\n\n\nvariable and the IFS data is 0.7. Assuming that cycles in government consumption and capital\n\n\nexpenditure are related, this measure should be a good approximation of total government expenditure.\n\n\nMoreover, any bias generated by using the general government measure is likely to be against the\n\n\nhypothesized relationship between budget cycles and party characteristics. Subnational spending\n\n\nintroduces noise into central government spending, making it more difficult to detect cycles in central\n\n\ngovernment spending. Similarly, to the extent that capital spending is more important for pre-electoral\n\n\nexpenditures, it should once again be more difficult to detect cycles in spending using the general\n\n\ngovernment consumption data.\n\n\nPrevious research has not addressed the noise that hyperinflation can introduce into analyses of\n\n\npolitical budget cycles. Hyperinflation generates very large within-fiscal-year differences in the real value\n\n\nof nominal government spending. As a consequence, small errors in assumptions regarding the timing\n\n\nof election spending can have large effects on the estimation. For example, if the election falls in the\n\n\nbeginning of the fiscal year, even a large increase in government spending might go unnoticed, since\n\n\nnominal spending could be one-half of the\n\nMention: \"IFS data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Correlated at 0.7 with the government consumption measure.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005783:59:1", "occurrence_id": "a46b3417fe474f880d96a4b4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .6 standard<br>deviation drop in the net interest margin. For instance, giving Mexico the same level<br>of activity restriction as Korea would lead to a 1 percentage point decrease in net<br>margins in Mexico. However, once institutional framework is controlled for,<br>regulations cease to have an impact on efficiency. Better institutional environment<br>(according to the World Governance Indicators) and economic freedom are strongly<br>associated with lower net interest margins. Specifically, a one standard deviation<br>improvement in economic freedoms (0.54) would lower interest margins by 1.1<br>percentage points.<br>• If Romania had the same banking concentration as Poland, its net interest margin<br>would drop from 8.45 to 7.97. Also, if Burundi had the same banking concentration<br>as Kenya, its net interest margin would drop from 10.74 to 9.95.<br>|\n|Pasiouras,<br>Tanna, and<br>Zopounidis<br>(2009)<br>|<br>• A one-unit increase in capital requirements reduces bank’s cost inefficiency by 9<br>percent. Under the same scenario, market discipline, activity restrictions, and<br>official supervisory power lead to 10 percent, 9 percent, and 8 percent reductions in<br>cost inefficiency. The effect of official supervision on profit inefficiency is about<br>6.6 percent, whereas market discipline leads to a 25 percent reduction in profit<br>inefficiency.<br>|\n|Bayraktar and<br>Wang (2004)|<br>• A one standard deviation increase in the asset share of foreign banks in the<br>economy (i.e., 15.9 percent) leads to a 0.03 percentage point decrease\n\nMention: \"the World Governance Indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Institutional environment and economic freedom are measured with these indicators.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005784:7:0", "occurrence_id": "3163845e5fadde4f26ff0bc9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: We use the two-digit level of the NIC system for industry variation. This level of\n\naggregation contains 22 manufacturing industries. Discussions of India’s industrial landscape\n\nand its long-run transformation often use the terms “traditional” and “modern”, although there\n\nare no established or precise definitions of these groups. Following Ghani, Kerr, and O’Connell\n\n(2013a), we classify an industry as being modern if its unorganized sector share is less than the\n\nunweighted average unorganized share across manufacturing industries. We report several traits\n\nbelow that utilize this separation, and the notes to Table 3a document the industries included in\n\nthe two groupings explicitly. <sup>4</sup> [^4: For clarity, we use term “sector” when discussing organized and unorganized sectors, “industries” when\ndiscussing the 22 manufacturing industries within manufacturing, and “groups” when describing “traditional” and\n“modern” industry groups.]\n\nThe NSS surveys the ownership type of each establishment. Establishments can be listed\n\nas male proprietary, female proprietary, other owned, cooperative, household partnership, multi\nhousehold partnership, private LLC, and unknown. In some of our analyses of the unorganized\n\nsector, we separately consider establishments listed as either male proprietary or female\n\nproprietary. These two groups constitute 98% of establishments in the informal manufacturing\n\nsector in 2000. Ghani, Kerr, and O’Connell (2012) in particular emphasize the partitioned\n\nbusiness networks in India for female-owned businesses, and this work extends these earlier\n\nstudies to understand the implications for input usage.\n\nFinally, and most important, we collect raw material input counts information from the\n\nsurvey questionnaires (ASI ‘00 - Block H, NSS ‘00 (56 <sup>th</sup> Round) - Block 3). In each survey,\n\nrespondents are asked to list the first five major raw material inputs into their production,\n\nproviding product codes, quantities and values. From this block of information we extract the\n\ntotal number of reported inputs (with a maximum\n\nMention: \"NSS surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"It surveys the ownership type of each establishment.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005784:7:0", "occurrence_id": "309de51452bbf272a819ca35", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: We use the two-digit level of the NIC system for industry variation. This level of\n\naggregation contains 22 manufacturing industries. Discussions of India’s industrial landscape\n\nand its long-run transformation often use the terms “traditional” and “modern”, although there\n\nare no established or precise definitions of these groups. Following Ghani, Kerr, and O’Connell\n\n(2013a), we classify an industry as being modern if its unorganized sector share is less than the\n\nunweighted average unorganized share across manufacturing industries. We report several traits\n\nbelow that utilize this separation, and the notes to Table 3a document the industries included in\n\nthe two groupings explicitly. <sup>4</sup> [^4: For clarity, we use term “sector” when discussing organized and unorganized sectors, “industries” when\ndiscussing the 22 manufacturing industries within manufacturing, and “groups” when describing “traditional” and\n“modern” industry groups.]\n\nThe NSS surveys the ownership type of each establishment. Establishments can be listed\n\nas male proprietary, female proprietary, other owned, cooperative, household partnership, multi\nhousehold partnership, private LLC, and unknown. In some of our analyses of the unorganized\n\nsector, we separately consider establishments listed as either male proprietary or female\n\nproprietary. These two groups constitute 98% of establishments in the informal manufacturing\n\nsector in 2000. Ghani, Kerr, and O’Connell (2012) in particular emphasize the partitioned\n\nbusiness networks in India for female-owned businesses, and this work extends these earlier\n\nstudies to understand the implications for input usage.\n\nFinally, and most important, we collect raw material input counts information from the\n\nsurvey questionnaires (ASI ‘00 - Block H, NSS ‘00 (56 <sup>th</sup> Round) - Block 3). In each survey,\n\nrespondents are asked to list the first five major raw material inputs into their production,\n\nproviding product codes, quantities and values. From this block of information we extract the\n\ntotal number of reported inputs (with a maximum\n\nMention: \"NSS ‘00\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its block three questionnaire collects raw material input listings.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005787:2:0", "occurrence_id": "0a1483e97173c9e58fcb0c35", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ### Effects of Early-Life Exposure to Sanitation on Childhood Cognitive Skills: Evidence from India’s Total Sanitation Campaign\n\nDean Spears and Sneha Lamba <sup>_∗_</sup>\n\n\nOctober 11, 2013\n\n\nKeywords: sanitation, cognitive achievement, human capital, early-life health, India\nJEL codes: I25, I15, O15\n\n\n_∗_ Spears is corresponding author: dean@riceinstitute.org. Affiliations: Spears: Centre for Development\nEconomics, Delhi School of Economics, Delhi, India and r.i.c.e. (www.riceinstitute.org); Lamba: Barcelona\nGraduate School of Economics. We deeply appreciate the outstanding research assistance of Aashish Gupta.\nWe appreciate helpful comments from Anne Case, Diane Coffey, Juan Costain, Janet Currie, Aashish Gupta,\nJeff Hammer, Payal Hathi, Avinash Kishore, Manisha Shah, and Sangita Vyas. We are grateful to Pratham\nand the ASER Centre for making the ASER data available for this project; we particularly thank Natasha\nArora, Ranajit Bhattacharyya, and Sushmita Das. Dean Spears acknowledges with thanks support for the\nresearch program of which this paper is a part from the World Bank Water and Sanitation Program (WSP)\nand the Bill and Melinda Gates Foundation. All conclusions and any errors are our own.\n\n\n1\n\nMention: \"ASER data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Acknowledged provision of these learning outcome data for the project.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005788:17:3", "occurrence_id": "c63644243f74e75257f5b7fb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: >166.7</mark> <mark>200.0</mark> <mark>100</mark>\n<mark>Portugal</mark> <mark>0.0</mark> <mark>−20.0</mark> <mark>0.0</mark> <mark>120.0</mark> <mark>100</mark>\n<mark>Spain</mark> <mark>−66.7</mark> <mark>−100.0</mark> <mark>66.7</mark> <mark>200.0</mark> <mark>100</mark>\n<u><mark>Sweden</mark></u> <u><mark>0.0</mark></u> <u><mark>0.0</mark></u> <u><mark>0.0</mark></u> <u><mark>100.0</mark></u> <u><mark>100</mark></u>\n_<u><mark>Source:</mark></u>_ <mark>World Bank staff calculations using EU-SILC data.</mark>\n\n##### **6. Results**\n\n\nThis section presents the simulated poverty and social exclusion indicators (at-risk-of-poverty rates, low\nwork intensity rates, and anchored poverty rates) that would result if each country achieved the Europe\n2020 national targets in the following areas by 2020: (i) four separate employment outcomes; (ii) early\nschool leaving rates; and (iii) tertiary school completion rates. <sup>14</sup> The three employment outcomes are\ndescribed below and depicted in Figure 2.\n\nMention: \"EU-SILC data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Simulated poverty and exclusion outcomes are calculated from these data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005780:12:1", "occurrence_id": "c2bf078cefc67ed54f275517", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  constructed.\n\n\nThe vector of individual controls for individual i in country j includes education level,\n\ngender, access to formal banking, age and income quintile. We set up the first three of these\n\nvariables as 0/1 dummy variables - whether the individual has completed secondary school,\n\nwhether the individual is female, and finally whether the individual had access to a formal bank\n\naccount. Income quintile refers to the within country relative income provided by the Global\n\nFindex dataset, and we include a set of four dummy variables in our analysis to capture the\n\nvariable. Age is measured both by the age and age squared to account for non-linearity in how\n\nage affects usage as young adults could be more likely to use m-banking.\n\n\nThe country controls vary depending of the specification we use, but for our baseline\n\nregression it includes log of GDP per capita, percent of unbanked population, percent of urban\n\npopulation, percentage of population owning a mobile phone, concentration of banks (top three),\n\n\n11\n\nMention: \"the Global\n\nFindex dataset\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Within-country relative income quintiles come from this dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005790:9:0", "occurrence_id": "641a9c2c53591f89ed5c2a6b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Italy, Netherlands, Poland and Spain. Table 1 lists the top 15 countries in recorded emissions in\n\nCITL. The ten countries covered in the analysis are among the biggest polluters. Table 2 shows\n\nthe proportion of CITL installations that are matched – the ratio of matched firms and total firms\n\nin each CITL sector. Though the power sector has a low match ratio, most of the non-matched\n\nfirms are identified as being in non-power sectors, therefore we believe that our control group\n\nwill not contain ‘treated’ firms.\n\n\nTable 1: Top 15 Polluting Countries\n\n\n\n<u>Rank</u> <u>Country</u>\n\n\n\n<u>Verified Emissions</u>\n\n\n\n<u>Country</u> <u>(million tonnes)</u>\n\n1 Germany* 466\n2 Great Britain* 247\n3 Italy* 213\n4 Poland* 203\n5 Spain* 162\n6 France* 122\n7 Netherlands* 81\n8 Czech Republic* 80.6\n9 Greece 67.9\n10 Belgium* 52.5\n11 Finland 38.7\n12 Romania 38.3\n13 Austria* 31.3\n14 Portugal 30.5\n<u>15</u> <u>Denmark</u> <u>27.9</u>\n\n\n\nNote: The ‘verified emissions’ is a simple time average of the total verified emissions of all\ninstallations on the CITL record. Starred countries indicate the ones in our sample. The names and\naddresses of Greek firms in the AMADEUS database did not transcript properly so we did not try to\nmatch these installations and their respective firms. * indicates countries that are covered in the study.\n\n\n8\n\nMention: \"CITL record\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Total verified emissions of all installations are held in this record.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005790:9:0", "occurrence_id": "71d434b5819fd27e384d5ab4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Italy, Netherlands, Poland and Spain. Table 1 lists the top 15 countries in recorded emissions in\n\nCITL. The ten countries covered in the analysis are among the biggest polluters. Table 2 shows\n\nthe proportion of CITL installations that are matched – the ratio of matched firms and total firms\n\nin each CITL sector. Though the power sector has a low match ratio, most of the non-matched\n\nfirms are identified as being in non-power sectors, therefore we believe that our control group\n\nwill not contain ‘treated’ firms.\n\n\nTable 1: Top 15 Polluting Countries\n\n\n\n<u>Rank</u> <u>Country</u>\n\n\n\n<u>Verified Emissions</u>\n\n\n\n<u>Country</u> <u>(million tonnes)</u>\n\n1 Germany* 466\n2 Great Britain* 247\n3 Italy* 213\n4 Poland* 203\n5 Spain* 162\n6 France* 122\n7 Netherlands* 81\n8 Czech Republic* 80.6\n9 Greece 67.9\n10 Belgium* 52.5\n11 Finland 38.7\n12 Romania 38.3\n13 Austria* 31.3\n14 Portugal 30.5\n<u>15</u> <u>Denmark</u> <u>27.9</u>\n\n\n\nNote: The ‘verified emissions’ is a simple time average of the total verified emissions of all\ninstallations on the CITL record. Starred countries indicate the ones in our sample. The names and\naddresses of Greek firms in the AMADEUS database did not transcript properly so we did not try to\nmatch these installations and their respective firms. * indicates countries that are covered in the study.\n\n\n8\n\nMention: \"AMADEUS database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Greek firm names and addresses failed to transcribe, blocking a match.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005797:38:3", "occurrence_id": "8165cca01b465c05a488a708", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: </u> <u>(0.400)</u> <u>(0.366)</u> <u>(0.391)</u>\n<u>Observations</u> <u>28,627</u> <u>28,627</u> <u>29,527</u> <u>29,527</u> <u>28,556</u> <u>28,556</u> <u>29,551</u> <u>29,551</u>\n\n\n\n<u>Has Access to Finance</u> <u>Has a Line of Credit or Loan</u>\n\n\n\nHas a Checking or\n\n\n\nSource: Enterprise Surveys Database\n\nNotes: A firm has access to finance if they are of type NCC or MCC. Standard errors in parentheses, *** p<0.01, ** p<0.05, * p<0.1. A full set of country and\nstratification sectors dummy variables are included.\n\n\n37\n\nMention: \"Enterprise Surveys Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Access to finance and credit holding are classified from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005798:35:0", "occurrence_id": "1b8865a185b5e4281d737354", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The third speci…cation deals with the possibility that the potential negative e¤ects of heavy\n\n\nrainfall (and ‡ood) on wages may induce substitution e¤ects. We report IV estimates that control\n\n\nfor agricultural wages and spatial cost of living index (using our main IV, the heavy rainfall dummy\n\n\nbased on the period 2000-2009). The results are reported in columns (5) and (6) in Table 6. They\n\n\ncon…rm the argument that the potential negative e¤ects on wages and prices are not relevant for\n\n\nour results.\n\n\nAlternative Time Period for the Rainfall Instrument\n\n\nThe heavy rainfall dummy we so far used for identi…cation is de…ned on the basis of 10\n\n\nyears data on average rainfall for the period 2000-2009. If our estimates are in fact the e¤ects of\n\n\n‘economic status’(i.e, permanent income) of a household on bribing in schools, then the estimates\n\n\nshould not change signi…cantly if we use average rainfall data from a somewhat di¤erent period.\n\n\nWe checked the sensitivity of the estimates for alternative periods, and it is reassuring that the\n\n\nestimates remain robust. For example, consider the estimates using rainfall data for 1999-2005\n\n\npresented in columns (7) and (8) in Table 6, they are almost identical to the estimates we found\n\n\nearlier using the 2000-2009 rainfall data.\n\n\n(9) Discussion and Interpretation of the IV Results\n\n\nIt is now widely appreciated that, when the monotonicity condition is satis…ed, the IV es\n\ntimates provide us with Local Average Treatment E¤ect (LATE), i.e., they provide estimates\n\n\nof the average causal e¤ect for those households which are a¤ected by the instrument (i.e., the\n\n\n‘compliers’). Since we exploit variations in the rainfall for our identi…cation, our estimates are\n\n\nmost relevant for those households whose income primarily\n\nMention: \"10\n\n\nyears data on average rainfall\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The heavy rainfall instrument is defined from this decade of averages.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005801:7:0", "occurrence_id": "244d8ed1e31f81b825ed180a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The rest of the paper is organized as follows. Section II discusses the data. Section III\n\n\nlays out the empirical methodology. Section IV presents the empirical results. Section V\n\n\nconcludes.\n\n\n**II-** **Data**\n\n\nOur database combines annual bank-level financial information from Bankscope, a\n\n\ncommercial dataset provided by Bureau van Dijk, with bank ownership information collected\n\n\nfrom various sources including Fitch Research, The Bankers’ Almanac, bank websites, Central\n\n\nBank publications, parent company’s reports and bank regulation authorities. Overall, our dataset\n\n\ncovers 1,194 banks <sup>5</sup> operating in 51 countries from 2005 to 2009. Our sample spans three\n\n\nregions, including 7 countries in East Asia and the Pacific (EAP), 25 in Europe and Central Asia\n\n\n(ECA), and 19 in Latin America and the Caribbean (LAC). Because ECA is the largest and most\n\n\ndiverse region in terms of the number of countries, we further divide the region into three sub\n\ngroups: 5 countries that participated in the so-called “Vienna Initiative” (ECA VI) <sup>6</sup> [^6: The Vienna Initiative was an action plan in which multinational banks, international financial institutions (EBRD,\nIMF, World Bank), and European governments agreed to cooperate to support the local banking systems in Bosnia,\nHungary, Latvia, Serbia, and Romania. The international financial institutions provided support in exchange for\ncountries’ commitments to keep their economic programs on track. In turn, various multinational banks (Alpha Bank,\nBayerische Landesbank, Erste Group, Eurobank EFG, Hype Alpe-Adria, ING, Intesa San Paolo, KBC Group,\nNational Bank of Greece, Nordea Bank, OTP, Piraeus Bank, Raiffeisen International, Skandinaviska Enskilda\nBanker, Societe Generale, Swedbank, UniCredit, and Volksbank) signed commitment letters in which they pledged\nto maintain exposures and keep their affiliates in these countries adequately capitalized.], 8 countries\n\n\nthat were part of the former Soviet Union (ECA FSU\n\nMention: \"Bankscope\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Annual bank-level financial information comes from this commercial dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005809:18:1", "occurrence_id": "8fd73918f7460c16a81b2c89", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  lack connections that give them access to these intermediaries. Of\ncourse it is not only regulatory interpretations that may cause problems, but also the regulations\nthemselves. Regulatory frameworks in the region range from the very good to the very poor but\nin general smaller firms will be less well equipped to deal with complex and costly regulations.\n\nSMEs also face greater hurdles when attempting to enter export markets. According to World\nBank enterprise data, the percentage of large firms that export is over three times that of small\nfirms and almost double that of medium firms, <sup>18</sup> implying difficulties for SMEs to enter export\nmarkets. Another indicator on ease of exporting is average number of days to clear customs. On\naverage, small firms spend an extra day and a half to clear customs compared to medium and\nlarge firms. <sup>19.</sup> Small firms are at a disadvantage if there are fixed costs relating to obtaining\nmarket information, product standards and quality certification and other procedures necessary\nto enter export markets. Newer and smaller firms may not have the contacts of established firms\nin export markets and large firms may control key distribution channels. Also, due to their\ndisadvantaged access to finance, SMEs may have profitable opportunities to export for which\nthey cannot obtain funding.\n\nOn the regulatory environment, SMEs overall face a bigger challenge, as SMEs place a high\npriority on rule of law—they feel they could operate and grow better with better governance and\n\n\n18Around 35 percent for large firms vs. 21 percent for medium and 9 percent for small enterprises\n197.3 days vs. 5.8 days for medium firms\n\n17\n\nMention: \"World\nBank enterprise data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Export participation by firm size is measured in these data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005819:4:0", "occurrence_id": "fd6a30c7f10aea069b3e3d83", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The motivation for focusing on trade facilitation measures comes from existing studies that\n\n\nshow that with the decline of tariff and non-tariff barriers around the globe, it is the trade facilitation\n\n\nmeasures that are becoming increasingly more important for the overall expansion of international\n\n\ntrade (see, for example, Wilson et al. 2003). There is no standard definition of trade facilitation in public\n\n\npolicy discourse. In a narrow sense, trade facilitation efforts simply address the logistics of moving goods\n\n\nthrough ports and the documentation associated with cross-border trade. Some of the factors included\n\n\nunder trade facilitation are Internet availability (Freund and Weinhold 2002), time to clear shipments at\n\n\nports (Djankov et al. 2010) and standards harmonization and automated customs procedures (Herter et\n\n\nal. 2001). The present paper focuses on a regulatory aspect of trade facilitation, the number of\n\n\ndocuments required to export and import as measured by the World Bank Doing Business project. There\n\n\nis a lot of variation in the measure across countries. For example, as of May 2010, while two documents\n\n\nare required to export a container from France, it requires 11 documents to do the same in Namibia. Is\n\n\nsome of this cross-country variation in the number of required documents due to differences in country\n\n\nsize? The present paper attempts to answer this question.\n\n\nMuch like the studies mentioned above, our focus is on a single element of trade facilitation. We\n\n\nwould like to caution that trade facilitation is a multi-faceted phenomenon and there is no guarantee\n\n\nthat the results presented in this paper for the number of documents required to export and import\n\n\nwould necessarily extend to other aspects of trade facilitation. A careful analysis of the various trade\n\n\nfacilitation measures one by one may be required to understand the determinants and effects of these\n\n\ndifferent measures.\n\n\nThe motivation for exploring the possibility of non-linearity in the relationship between country\n\n\nsize and trade facilitation is largely empirical in nature. That is, there is no strong theoretical reason\n\nMention: \"the World Bank Doing Business project\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Document counts required to export and import are measured by this project.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005821:13:0", "occurrence_id": "27e092afeed88f40c2a4bea0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: whether women were more likely to work outside the home, in full-time occupations and in those\n\n\nwith higher requirements for participation.\n\n\n**III. Ethiopia**\n\n\nEthiopia is one of the poorest countries in the world, although it has made progress in\n\n\nreducing poverty and achieving other Millennium Development Goals in recent years. It had an\n\n\nestimated GDP per capita of $400 in 2012 (IMF 2012). Ethiopia ranked 171 of 182 countries in\n\n\nthe UNDP’s Human Development Index in 2009 (UNDP 2009) and 89 of 102 on its Social\n\n\nInstitutions and Gender Index (SIGI 2009). Ethiopia is the second most populous country in Sub\n\nSaharan Africa, and also among the most ethnically diverse. The country is a federal republic,\n\n\nwith nine regions and two city councils. <sup>8</sup> [^8: In our analysis, we treat the two city councils (Addis Ababa and Dire Dawa) as their own geographic areas since\nthey are administered separately from the other regions. As a result, we end up 11 geographic areas in our analysis.]\n\n\nTable 1 provides some information on differences across geographic areas in the country.\n\n\nWhile there are similarities between areas in recent rates of population growth, there are\n\n\nsignificant differences in urbanization rates as well as population density. Given the correlation\n\n\nbetween urbanization and education, there is also high geographic variation in education. In our\n\n\nanalysis, it is important to capture the effects of these differences across geographic areas. We\n\n\ndo so by including dummies for the various regions and chartered cities. We also repeat our\n\n\nresults excluding the large cities to be sure they are not driving the results and the findings hold.\n\n\nEthiopia’s labor market is characterized by significant gender gaps in participation and in\n\n\nremuneration. Gender differences in the labor market are also present. In a 2004 survey report on\n\n\nurban employment, there is a significant gender difference in the employment-to-population ratio\n\n\nin the urban areas of all regions (CSA 2004). For women, the employment-to-population ratio is\n\n\n8\n\nMention: \"UNDP’s Human Development Index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its ranking of one country among 182 sets the development context.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005821:13:0", "occurrence_id": "d936b8b1ae8257c8083bda77", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: whether women were more likely to work outside the home, in full-time occupations and in those\n\n\nwith higher requirements for participation.\n\n\n**III. Ethiopia**\n\n\nEthiopia is one of the poorest countries in the world, although it has made progress in\n\n\nreducing poverty and achieving other Millennium Development Goals in recent years. It had an\n\n\nestimated GDP per capita of $400 in 2012 (IMF 2012). Ethiopia ranked 171 of 182 countries in\n\n\nthe UNDP’s Human Development Index in 2009 (UNDP 2009) and 89 of 102 on its Social\n\n\nInstitutions and Gender Index (SIGI 2009). Ethiopia is the second most populous country in Sub\n\nSaharan Africa, and also among the most ethnically diverse. The country is a federal republic,\n\n\nwith nine regions and two city councils. <sup>8</sup> [^8: In our analysis, we treat the two city councils (Addis Ababa and Dire Dawa) as their own geographic areas since\nthey are administered separately from the other regions. As a result, we end up 11 geographic areas in our analysis.]\n\n\nTable 1 provides some information on differences across geographic areas in the country.\n\n\nWhile there are similarities between areas in recent rates of population growth, there are\n\n\nsignificant differences in urbanization rates as well as population density. Given the correlation\n\n\nbetween urbanization and education, there is also high geographic variation in education. In our\n\n\nanalysis, it is important to capture the effects of these differences across geographic areas. We\n\n\ndo so by including dummies for the various regions and chartered cities. We also repeat our\n\n\nresults excluding the large cities to be sure they are not driving the results and the findings hold.\n\n\nEthiopia’s labor market is characterized by significant gender gaps in participation and in\n\n\nremuneration. Gender differences in the labor market are also present. In a 2004 survey report on\n\n\nurban employment, there is a significant gender difference in the employment-to-population ratio\n\n\nin the urban areas of all regions (CSA 2004). For women, the employment-to-population ratio is\n\n\n8\n\nMention: \"Social\n\n\nInstitutions and Gender Index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its ranking of the same country across 102 economies complements the first.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005822:14:1", "occurrence_id": "8cc3ee43d6638630e39ab4fe", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .11\n\n\n**South Asia** **1.77** **0.07** **0.39** **1.27** **0.03**\n\n\nSub-Saharan Africa 4.72 1.51 1.05 1.63 0.00\n\n\nSource: International Trade Centre and World Bank Staff Calculations based on 2009 statistics\n\n##### **_Source Countries for South Asia’s FDI Inflows_**\n\n\n**The number of developing countries investing in South Asia has increased in recent years, which**\n\n**reflects a global trend of more south-south FDI flows** . Growth in the number of countries investing in\n\nSouth Asia may suggest greater trade linkages and knowledge spillovers between developing countries\n\nand industries, rather than the expansion of investments from a single country with only marginal\n\nadditional knowledge spillovers. As Table 4 shows, the number of developing countries with FDI\n\npositions in South Asia increased from 38 to 45 between the early and late 2000s, while the number of\n\ndeveloped countries making investments in South Asia grew by only one (Greece).\n\n\n**Table 4: FDI into South Asia: Source Country Counts**\n\n\n**Developed Countries** **Developing Countries** **Total**\n\n\n2003-2006 23 38 61\n\n\n<u>2007-2010</u> <u>24</u> <u>45</u> <u>69</u>\n\n\nSource: World Bank Staff estimates from UNCTAD statistics and fDimarkets\n\n\n13\n\nMention: \"2009 statistics\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Trade performance calculations rest on these statistics.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005822:14:1", "occurrence_id": "0e0ae2dca917aa06fe801928", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .11\n\n\n**South Asia** **1.77** **0.07** **0.39** **1.27** **0.03**\n\n\nSub-Saharan Africa 4.72 1.51 1.05 1.63 0.00\n\n\nSource: International Trade Centre and World Bank Staff Calculations based on 2009 statistics\n\n##### **_Source Countries for South Asia’s FDI Inflows_**\n\n\n**The number of developing countries investing in South Asia has increased in recent years, which**\n\n**reflects a global trend of more south-south FDI flows** . Growth in the number of countries investing in\n\nSouth Asia may suggest greater trade linkages and knowledge spillovers between developing countries\n\nand industries, rather than the expansion of investments from a single country with only marginal\n\nadditional knowledge spillovers. As Table 4 shows, the number of developing countries with FDI\n\npositions in South Asia increased from 38 to 45 between the early and late 2000s, while the number of\n\ndeveloped countries making investments in South Asia grew by only one (Greece).\n\n\n**Table 4: FDI into South Asia: Source Country Counts**\n\n\n**Developed Countries** **Developing Countries** **Total**\n\n\n2003-2006 23 38 61\n\n\n<u>2007-2010</u> <u>24</u> <u>45</u> <u>69</u>\n\n\nSource: World Bank Staff estimates from UNCTAD statistics and fDimarkets\n\n\n13\n\nMention: \"UNCTAD statistics\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Investment flow estimates are derived from these statistics.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005822:14:1", "occurrence_id": "62fafb81ed7fbe8c81d7d96b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .11\n\n\n**South Asia** **1.77** **0.07** **0.39** **1.27** **0.03**\n\n\nSub-Saharan Africa 4.72 1.51 1.05 1.63 0.00\n\n\nSource: International Trade Centre and World Bank Staff Calculations based on 2009 statistics\n\n##### **_Source Countries for South Asia’s FDI Inflows_**\n\n\n**The number of developing countries investing in South Asia has increased in recent years, which**\n\n**reflects a global trend of more south-south FDI flows** . Growth in the number of countries investing in\n\nSouth Asia may suggest greater trade linkages and knowledge spillovers between developing countries\n\nand industries, rather than the expansion of investments from a single country with only marginal\n\nadditional knowledge spillovers. As Table 4 shows, the number of developing countries with FDI\n\npositions in South Asia increased from 38 to 45 between the early and late 2000s, while the number of\n\ndeveloped countries making investments in South Asia grew by only one (Greece).\n\n\n**Table 4: FDI into South Asia: Source Country Counts**\n\n\n**Developed Countries** **Developing Countries** **Total**\n\n\n2003-2006 23 38 61\n\n\n<u>2007-2010</u> <u>24</u> <u>45</u> <u>69</u>\n\n\nSource: World Bank Staff estimates from UNCTAD statistics and fDimarkets\n\n\n13\n\nMention: \"fDimarkets\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"A companion commercial source for the investment estimates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005823:1:0", "occurrence_id": "95898832dcc98fe7048f1b1d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Policy Research Working Paper 6697\n\n### **Abstract**\n\nLittle is known about which of the skills that make up\nworkers’ human capital contribute to higher earnings.\nPast empirical evidence suggest that most of the return to\nschooling is generated by effects or correlates unrelated\nto the skills measured by the available tests. This paper\nuses the International Adult Literacy and the Adult\nLiteracy and Life Skills surveys to obtain multi-country\nestimates of the components of the return to schooling.\nThe results reveal considerable heterogeneity and a\n\n\n\ndichotomy between two groups of countries. For a\nsubgroup of educationally advanced countries, nearly\nhalf of the return to schooling can be attributed to labor\nmarker-relevant functional literacy skills associated with\nschooling, while for a subgroup of less educationally\nadvanced countries, such skills account for just over 20\npercent of the return to schooling, while the return to\nschooling mostly reflects the signaling value of schooling.\n\n\n\nThis paper is a product of the Education Unit, Human Development Network. It is part of a larger effort by the World\nBank to provide open access to its research and make a contribution to development policy discussions around the world.\nPolicy Research Working Papers are also posted on the Web at http://econ.worldbank.org. The authors may be contacted\nat hpatrinos@worldbank.org.\n\n\n_The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development_\n_issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the_\n_names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those_\n_of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and_\n_its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent._\n\nMention: \"International Adult Literacy\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Multi-country estimates of schooling returns come from this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005823:1:0", "occurrence_id": "70087bc1ac43556617f79b2f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Policy Research Working Paper 6697\n\n### **Abstract**\n\nLittle is known about which of the skills that make up\nworkers’ human capital contribute to higher earnings.\nPast empirical evidence suggest that most of the return to\nschooling is generated by effects or correlates unrelated\nto the skills measured by the available tests. This paper\nuses the International Adult Literacy and the Adult\nLiteracy and Life Skills surveys to obtain multi-country\nestimates of the components of the return to schooling.\nThe results reveal considerable heterogeneity and a\n\n\n\ndichotomy between two groups of countries. For a\nsubgroup of educationally advanced countries, nearly\nhalf of the return to schooling can be attributed to labor\nmarker-relevant functional literacy skills associated with\nschooling, while for a subgroup of less educationally\nadvanced countries, such skills account for just over 20\npercent of the return to schooling, while the return to\nschooling mostly reflects the signaling value of schooling.\n\n\n\nThis paper is a product of the Education Unit, Human Development Network. It is part of a larger effort by the World\nBank to provide open access to its research and make a contribution to development policy discussions around the world.\nPolicy Research Working Papers are also posted on the Web at http://econ.worldbank.org. The authors may be contacted\nat hpatrinos@worldbank.org.\n\n\n_The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development_\n_issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the_\n_names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those_\n_of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and_\n_its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent._\n\nMention: \"Adult\nLiteracy and Life Skills surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The companion survey supplying the skill components.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005824:10:0", "occurrence_id": "d22959935c8cf19a7f1d218e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 4: State control of economic activity in EU countries**\n\n\nSource: own elaborations based on OECD data\n\n\nDespite the decline in state control over the past decade, Romania’s performance still lags behind that of\nmost EU countries (Figure 4). Important energy generation companies, whose governance requires further\nimprovement, continue to be managed by\n\n**Figure 5: Decomposition of PMR: State Control**\n\nthe state, while some energy tariffs have\nnot yet reached import price parity.\n\n\nThe two sub-domains of state control in\nthe PMR system are _public ownership_ and\n_state involvement in business operations_ .\nAs the decomposition exercise below\nreveals (Figure 5), public ownership poses\na significant challenge to competition.\nWhereas in Romania the government’s\ndirect involvement in business operations\nhas been reduced, mainly through price\nliberalization for many product categories,\npublic ownership remains high, both because the state still holds on to at least one firm in a large number\nof industries and due to its ownership of large network companies.\n\n\n**3.1.1** **PUBLIC OWNERSHIP**\n\n\nThe ‘public ownership’ sub-domain covers three policy areas: _scope of the public enterprise sector_, _direct_\n_control over business enterprises_, and _government involvement in the infrastructure sector_ . Due to the\nlarge role played by the level of public ownership in many of the countries in the sample, we present\nbelow the detailed score distribution of these three components. For Romania, results indicate that while\nthe level of direct control over business enterprises is not particularly high, both the pervasiveness of state\nownership across industries and the degree of government involvement in infrastructure are above the EU\naverage.\n\n\n9\n\nMention: \"OECD data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Elaborations on state control of economic activity rest on this source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005829:19:0", "occurrence_id": "8841d3ad20c51e43ca9edb66", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 18\n\n\n2011—80,000 more than in 2010, with many families cutting down basic necessities such as\nfood and reducing the quality and the amount of food intake. The well-being of children in the\nCzech Republic, Poland, Greece, and the United States also seems to have deteriorated.\nSimilarly, the UN composite Human Development Index, computed as a function of measures of\nlife expectancy at birth, access to knowledge, and a decent standard of living, declined between\n2007 and 2012 for a number of middle and high income countries in the Middle East and Europe\n(the United Arab Emirates, Bahrain, Greece, Luxembourg, Ireland, and Iceland) as well as for\nsmall island states. Advanced countries and regions closely linked to them registered the smallest\nimprovement in the index (figure 12).\n\n\n**Figure 9. Development indicators before and after the crisis by income group**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n_Source_ : Authors’ computations, based on World Bank’s World Development Indicators database.\n\nMention: \"UN composite Human Development Index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Life expectancy, knowledge access and a decent standard of living enter this composite.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005829:19:0", "occurrence_id": "8ecada99d90514077c9a9166", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 18\n\n\n2011—80,000 more than in 2010, with many families cutting down basic necessities such as\nfood and reducing the quality and the amount of food intake. The well-being of children in the\nCzech Republic, Poland, Greece, and the United States also seems to have deteriorated.\nSimilarly, the UN composite Human Development Index, computed as a function of measures of\nlife expectancy at birth, access to knowledge, and a decent standard of living, declined between\n2007 and 2012 for a number of middle and high income countries in the Middle East and Europe\n(the United Arab Emirates, Bahrain, Greece, Luxembourg, Ireland, and Iceland) as well as for\nsmall island states. Advanced countries and regions closely linked to them registered the smallest\nimprovement in the index (figure 12).\n\n\n**Figure 9. Development indicators before and after the crisis by income group**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n_Source_ : Authors’ computations, based on World Bank’s World Development Indicators database.\n\nMention: \"World Development Indicators database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The computations behind the group comparison come from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005833:3:1", "occurrence_id": "0736fb78c10d7e6b1e683420", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _ legal questionnaires. Their input was the fundamental source of data for the analysis.\nGrateful acknowledgment of these contributors is made in the Contributors section at the end of this\nreport. Additional details about the contributors across all topics of the FDI Regulation database are\n[provided at http://iab.worldbank.org/Local-Partners/2012.](http://iab.worldbank.org/Local-Partners/2012)\n\nThe second is the expert consultative group formed to provide technical input to the development of\nboth topics. This group of global experts provided input on the focus of the legal questionnaires and on\nthe methodology to develop the indicators as well as on the analysis of the data. Members of the expert\nconsultative group include:\n\n   - Keith Clifford Bell, World Bank Group, Washington DC\n\n   - Kim Bettcher, Center for International Private Enterprise(CIPE), Washington DC\n\n   - Thomas Farole, World Bank Group, Washington DC\n\n   - Jean-Paul Gauthier, World Economic Processing Zones Association, Delaware\n\n   - Susan Karamanian, George Washington University Law School, Washington DC\n\n   - Gerald Korngold, New York Law School, New York\n\n   - Terry Selzer, Husen Advokater, Denmark\n\n\n2\n\nMention: \"FDI Regulation database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Contributor detail across its topics is published alongside the report.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005836:21:1", "occurrence_id": "06609711e0d7daaaf078f65e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: , 1999, and 2006 (Table 1). The number of EIAs expanded from four in\n\n\n[6 The database on EIAs maintained by Baier and Bergstrand (http://www.nd.edu/~jbergstr/) also provides](http://www.nd.edu/~jbergstr/)\n\n\ndetailed and useful information on EIAs and links to the legal text of the Agreements. Similarly, Horn et al.\n\n\n(2009) study the legal text to investigate whether the areas (technical regulations as well as customs\n\n\nadministration, export taxes, public procurement, and labour market regulations) covered by the Agreements fall\n\n\nunder or outside the current mandate of the WTO. We thank A. Sapir for providing data. We tried a cross-check\n\n\nwith our data. Unfortunately, their database does not provide a detailed analysis of the content of each area.\n\n\n20\n\nMention: \"database on EIAs\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It also supplies the count of preferential agreements over time.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005839:12:0", "occurrence_id": "a93c11d46e19f03525832e92", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Nevertheless, if the national poverty line is used to identify the poor, could we still\ndefend the Tendulkar line as adequate? We argue in the affirmative.\n\n\nSource: Calculations from the NSSO expenditure survey, 2004-05, by Rahul Ahluwalia\nof International School of Business, Hyderabad\n\n\nGoing by the urban and rural population weights of 0.298 and 0.702 implicit in\nthe population projections for January 1, 2010, the average countrywide per-capita MRP\nexpenditure during 2009-10 works out to 40.2 rupees per person per day. Therefore,\n\n\n9\n\nMention: \"the NSSO expenditure survey, 2004-05\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Calculations testing the adequacy of the poverty line use this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005841:15:0", "occurrence_id": "f189245f71c0f5d0f85df4d9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: instrument political instability, but unfortunately finding credible instruments is hard. Instead, we\n\n\nrecognize this problem and caution that our results should be interpreted as conditional\n\n\nassociations, rather than reflecting causal relationships. <sup>10</sup> [^10: We also do not control explicitly for potential spillovers across countries, although the ICRG political instability\nindex has a component capturing external conflict and the risk of foreign interference and therefore does reflect the\nimpact of unrest in the region on the political stability in the host economy.]\n\n\nA related concern is omitted variable bias, which we try to minimize by including as many\n\n\nrelevant economic variables as we could obtain quarterly data for, notably inflation, changes in\n\n\nthe nominal exchange rate, industrial production and changes therein. <sup>11,12</sup> [^11: In the cases of countries without industrial production data (Bahrain, Lebanon, and Yemen), we use quarterly\nexports data. Since we are only looking at changes within countries over time this does not yield any major problems.] [^12: Ideally, we should have used quarterly Gross Domestic Product (GDP) data, but such series were not available for\nthe majority of countries in the region] While we would ideally\n\n\nhave liked to control for a richer array of macroeconomic factors, high frequency economic data\n\n\nare unfortunately not widely available.\n\n\nLast but not least, the presence of a lagged endogenous variable creates a potential\n\n\nupward endogeneity bias when estimating equation (1) using Ordinary Least Squares, due to a\n\n\ncorrelation between the time invariant unobserved fixed effects and explanatory variables. By\n\n\ncontrast, fixed effects estimates tend to be biased downwards because of the so-called Nickell\n\n\nbias (Nickell, 1981; Kiviet, 1995) – a non-negligible correlation between the transformed residuals\n\n\nand the transformed error term. However, Monte Carlo evidence suggests that bias-reduction\n\n\nmethods, developed by Bun and Kiviet (2003) and Bruno (2005), work well when T is relatively\n\n\nlarge and the serial correlation relatively modest as appears\n\nMention: \"ICRG political instability\nindex\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its component captures external conflict and foreign interference risk.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005843:52:1", "occurrence_id": "1711f0c2413c00d830450edf", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: -19<br>20-24<br>25-29<br>30-34<br>35-39<br>40+<br>Age<br>Small-firm, Capital Intensive|50<br>100 150 200 250<br>Employment (Linear Prediction)<br><5<br>5-9<br>10-14<br>15-19<br>20-24<br>25-29<br>30-34<br>35-39<br>40+<br>Age<br>Low Dependence on External Finance|<br>100 200 300 400 500<br>Employment (Linear Prediction)<br><5<br>5-9<br>10-14<br>15-19<br>20-24<br>25-29<br>30-34<br>35-39<br>40+<br>Age<br>High Dependence on External Finance|\n\n\n\n\n\nSource: ASI Data pooled from the following census years: 1983/84, 1989/90, 1994/95, 2000/01, 2004/05\n\n\n\n51\n\nMention: \"ASI Data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Pooled across five benchmark years for the age and finance analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005844:21:1", "occurrence_id": "2026091a1b862125e23534d9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: up>44</sup>\n\n\n39 Currency conversions include changes in the currency being used, such as the formation of the Eurozone, and\ncurrency redenominations as often observed in high-inflation environments.\n\n40 As a first step, we use domestic CPI figures from the WDI. Where these are unavailable, we resort to the IMF’s\nWorld Economic Outlook or the country’s statistical offices directly.\n\n41 The WDI does not provide PPPs for Kosovo, so we used the implied PPP GDP conversion factors reported in the\nIMF’s World Economic Outlook.\n\n42 Data for China and Indonesia are provided by PovcalNet so they already incorporate adjustments for differential\ncosts of living in rural and urban areas. For India, where we added one survey from the WYD, we use urban and\nrural PPP exchange rates of Rs 17.24 and Rs 11.40 respectively which are given in Ravallion (2008).\n\n43 These countries include Argentina, Colombia, Ecuador, Honduras, Micronesia and Uruguay, and are all taken from\nPovcalNet.\n\n\n20\n\nMention: \"domestic CPI figures from the WDI\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The first step in deflating uses these consumer price series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005844:21:1", "occurrence_id": "7fc99cc6423e4068f206df3a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: up>44</sup>\n\n\n39 Currency conversions include changes in the currency being used, such as the formation of the Eurozone, and\ncurrency redenominations as often observed in high-inflation environments.\n\n40 As a first step, we use domestic CPI figures from the WDI. Where these are unavailable, we resort to the IMF’s\nWorld Economic Outlook or the country’s statistical offices directly.\n\n41 The WDI does not provide PPPs for Kosovo, so we used the implied PPP GDP conversion factors reported in the\nIMF’s World Economic Outlook.\n\n42 Data for China and Indonesia are provided by PovcalNet so they already incorporate adjustments for differential\ncosts of living in rural and urban areas. For India, where we added one survey from the WYD, we use urban and\nrural PPP exchange rates of Rs 17.24 and Rs 11.40 respectively which are given in Ravallion (2008).\n\n43 These countries include Argentina, Colombia, Ecuador, Honduras, Micronesia and Uruguay, and are all taken from\nPovcalNet.\n\n\n20\n\nMention: \"Data for China and Indonesia\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"These two economies' data already embed rural-urban cost adjustments.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005844:21:1", "occurrence_id": "78c8e5f9d8c3ff0bcbc57cfa", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: up>44</sup>\n\n\n39 Currency conversions include changes in the currency being used, such as the formation of the Eurozone, and\ncurrency redenominations as often observed in high-inflation environments.\n\n40 As a first step, we use domestic CPI figures from the WDI. Where these are unavailable, we resort to the IMF’s\nWorld Economic Outlook or the country’s statistical offices directly.\n\n41 The WDI does not provide PPPs for Kosovo, so we used the implied PPP GDP conversion factors reported in the\nIMF’s World Economic Outlook.\n\n42 Data for China and Indonesia are provided by PovcalNet so they already incorporate adjustments for differential\ncosts of living in rural and urban areas. For India, where we added one survey from the WYD, we use urban and\nrural PPP exchange rates of Rs 17.24 and Rs 11.40 respectively which are given in Ravallion (2008).\n\n43 These countries include Argentina, Colombia, Ecuador, Honduras, Micronesia and Uruguay, and are all taken from\nPovcalNet.\n\n\n20\n\nMention: \"PovcalNet\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"It supplies survey data with differential cost adjustments.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005844:21:1", "occurrence_id": "864a79780b7d025bad1c4dca", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: up>44</sup>\n\n\n39 Currency conversions include changes in the currency being used, such as the formation of the Eurozone, and\ncurrency redenominations as often observed in high-inflation environments.\n\n40 As a first step, we use domestic CPI figures from the WDI. Where these are unavailable, we resort to the IMF’s\nWorld Economic Outlook or the country’s statistical offices directly.\n\n41 The WDI does not provide PPPs for Kosovo, so we used the implied PPP GDP conversion factors reported in the\nIMF’s World Economic Outlook.\n\n42 Data for China and Indonesia are provided by PovcalNet so they already incorporate adjustments for differential\ncosts of living in rural and urban areas. For India, where we added one survey from the WYD, we use urban and\nrural PPP exchange rates of Rs 17.24 and Rs 11.40 respectively which are given in Ravallion (2008).\n\n43 These countries include Argentina, Colombia, Ecuador, Honduras, Micronesia and Uruguay, and are all taken from\nPovcalNet.\n\n\n20\n\nMention: \"WYD\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"One additional survey is added from this world dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005853:4:0", "occurrence_id": "f3d9f6fbd8f43e87a4d935f3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: We show that inequality-of-opportunity aversion and outcome-inequality aversion\nneed not be inconsistent; the tension between them can be resolved, as long as one\nis given priority over the other. Our axioms allow us to characterize two classes of\nopportunity-sensitive poverty measures (OSPM) - one a strict subset of the other.\nDominance conditions are established for each class. The first condition, based\non the the broad class of OSPM, turns out to be a re-interpretation of an existing\nresult in the literature on poverty dominance for heterogeneous populations (Atkinson and Bourguignon 1987, Jenkins and Lambert 1993). This condition involves\na type-sequential comparison of the cumulative distribution functions truncated at\nthe poverty line. The second condition, for the narrower class of OSPM, is a new\nsequential dominance procedure, which involves comparisons of the group speficic\nheadcount ratios and of average incomes among the poor in each type. Moving\nfrom partial to complete orderings, we also propose a specific parametric family of\nOSPM which combines elements from the Sen and Foster et al. frameworks. We\ncall it the opportunity-sensitive Foster, Greer and Thorbecke index, or OS-FGT.\n\nWe then apply both the partial-ordering analysis and the OS-FGT index to\neighteen countries in the European Union, using the EU-SILC 2005 data. When\naversion to inequality of opportunity is incorporated into the assessment of poverty,\nwe are able to uncover aspects not captured when using standard income-poverty\nmethods. The broad OSPM dominance condition yields a large number of unambiguous rankings: 128 of 153 possible pairwise comparisons. The more demanding\nsufficient conditions we establish for dominance in the narrow OSPM class are\nsatisfied much more seldom. In addition, income poverty rankings and OS-FGT\nrankings, although positively correlated, frequently differ: whereas the Netherlands has less income poverty than Germany, for example, the reverse is true for\nopportunity-sensitive poverty\n\nMention: \"the EU-SILC 2005 data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Poverty ordering across eighteen member states is computed from this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005855:5:1", "occurrence_id": "687bfa0726b745ba5358b0e6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  municipalities, thus distinguishing if the\nimpact of crime rates over growth is the result of common crime, organized crime, or both.\n\n\n**3.** **Data and methodology**\n\nData on income per capita by municipality for 2005 and 2010 are obtained by using the poverty\nmaps methodology (Elbers et al. 2003). Simply put, the method consists of imputing income to\nhouseholds in the Population Census (2010) and Population Count data (2005) using a model that\npredicts income from a household survey (the National Survey on Household Income and\nExpenditures, ENIGH, 2005 and 2010). The advantage of this methodology is that it benefits from\nthe strengths of both the household survey and the censuses while avoiding their weaknesses.\nEmpirical evidence based on this method has proven its precision when applied to a large sample of\ncountries, and particularly for aggregated data at the municipality level in Mexico (see Elbers et al.\n2001; Lopez-Calva, et al. 2005).\n\nAs mentioned previously, the data on the total number of homicides at the municipal level is\navailable from the official figures made public by Mexico's Technical Secretary for the National\nSecurity Council ( _Sistema Nacional de Seguridad Publica_, SNSP), a federal institution that is part of the\nSecretary of the Interior. The SNSP compiles information through an extensive collaborative\n\n\nother expenses: for example, it is less than half of the social security budget (i.e. pensions and healthcare), and 68 percent\nof what is spent in public education.\n4 Average losses for middle-high income countries like Mexico (Argentina, Botswana, Chile, Colombia, Mexico, Panama,\nPeru, Uruguay, Venezuela) are 2.7 percent (IFC and WB, 2012).\n\n\n4\n\nMention: \"Population Census\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Income is imputed to its households using a model fitted on survey data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005855:5:1", "occurrence_id": "8e3c6d63de7e1a97e87c9e59", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  municipalities, thus distinguishing if the\nimpact of crime rates over growth is the result of common crime, organized crime, or both.\n\n\n**3.** **Data and methodology**\n\nData on income per capita by municipality for 2005 and 2010 are obtained by using the poverty\nmaps methodology (Elbers et al. 2003). Simply put, the method consists of imputing income to\nhouseholds in the Population Census (2010) and Population Count data (2005) using a model that\npredicts income from a household survey (the National Survey on Household Income and\nExpenditures, ENIGH, 2005 and 2010). The advantage of this methodology is that it benefits from\nthe strengths of both the household survey and the censuses while avoiding their weaknesses.\nEmpirical evidence based on this method has proven its precision when applied to a large sample of\ncountries, and particularly for aggregated data at the municipality level in Mexico (see Elbers et al.\n2001; Lopez-Calva, et al. 2005).\n\nAs mentioned previously, the data on the total number of homicides at the municipal level is\navailable from the official figures made public by Mexico's Technical Secretary for the National\nSecurity Council ( _Sistema Nacional de Seguridad Publica_, SNSP), a federal institution that is part of the\nSecretary of the Interior. The SNSP compiles information through an extensive collaborative\n\n\nother expenses: for example, it is less than half of the social security budget (i.e. pensions and healthcare), and 68 percent\nof what is spent in public education.\n4 Average losses for middle-high income countries like Mexico (Argentina, Botswana, Chile, Colombia, Mexico, Panama,\nPeru, Uruguay, Venezuela) are 2.7 percent (IFC and WB, 2012).\n\n\n4\n\nMention: \"Population Count data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The earlier inter-censal count receiving the same imputation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005855:5:1", "occurrence_id": "68b578642f0041ebf0147712", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  municipalities, thus distinguishing if the\nimpact of crime rates over growth is the result of common crime, organized crime, or both.\n\n\n**3.** **Data and methodology**\n\nData on income per capita by municipality for 2005 and 2010 are obtained by using the poverty\nmaps methodology (Elbers et al. 2003). Simply put, the method consists of imputing income to\nhouseholds in the Population Census (2010) and Population Count data (2005) using a model that\npredicts income from a household survey (the National Survey on Household Income and\nExpenditures, ENIGH, 2005 and 2010). The advantage of this methodology is that it benefits from\nthe strengths of both the household survey and the censuses while avoiding their weaknesses.\nEmpirical evidence based on this method has proven its precision when applied to a large sample of\ncountries, and particularly for aggregated data at the municipality level in Mexico (see Elbers et al.\n2001; Lopez-Calva, et al. 2005).\n\nAs mentioned previously, the data on the total number of homicides at the municipal level is\navailable from the official figures made public by Mexico's Technical Secretary for the National\nSecurity Council ( _Sistema Nacional de Seguridad Publica_, SNSP), a federal institution that is part of the\nSecretary of the Interior. The SNSP compiles information through an extensive collaborative\n\n\nother expenses: for example, it is less than half of the social security budget (i.e. pensions and healthcare), and 68 percent\nof what is spent in public education.\n4 Average losses for middle-high income countries like Mexico (Argentina, Botswana, Chile, Colombia, Mexico, Panama,\nPeru, Uruguay, Venezuela) are 2.7 percent (IFC and WB, 2012).\n\n\n4\n\nMention: \"National Survey on Household Income and\nExpenditures\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its two rounds supply the income predictors for the imputation model.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005857:17:2", "occurrence_id": "b7f9981d7f7117e270ff76d3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: REER_ is replaced by _ULC_ ) remain\nunchanged.\n\n\nThe finding of limited contribution of real exchange rate (or unit labor cost) changes to the\nexternal imbalances of Eurozone countries may, to some extent, reflect the merging in a single\ngroup of periphery and core countries with large and opposite external imbalances especially in\nrecent years. Dynamics of the current account balance in core and periphery countries have been\nmostly asymmetric in recent years while patterns of real exchange movements have been less\nmarkedly divergent (see Figures 1 and 5b). It is then suitable to split the sample in two groups\n(core and periphery) to explore if any significant differences exist in the dynamics of the external\nbalance between these two groups.\n\n\n15 The _ULC_ series are taken from the OECD database.\n\n\n16\n\nMention: \"OECD database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Unit labour cost series behind the external balance come from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005859:11:1", "occurrence_id": "804cc324463230b23014332d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  information as an indicator of\n\nindigenous mother tongue and use it as a proxy for ethnic background. <sup>21</sup> In all three of these\n\n\n17 Although several other country censuses in the IPUMS-I database included linguistic variables, these countries\nwere excluded from the final sample because country specialists indicated that language is not a good proxy for\nethnicity in their specific contexts.\n\n18 The identification of the reference and minority languages as proxies for ethnicity was supplemented through\nconsultations with specialists in each country.\n\n19 In Bolivia, for example, the official classification of minorities cuts across multiple languages (including Spanish\nand indigenous languages) as well as ancestral roots.\n\n20 In Peru (2007), we limit the analysis to individuals who speak Spanish and Quechua as their mother tongue, since\nQuechua is by far the largest indigenous group in the country. However, using a broader “indigenous mother\ntongue” category, which encompasses all native languages including Quechua, does not change the results.\n\n21 Admittedly, this distinction may mistakenly classify those members of the reference group who learned an\nindigenous language, but who otherwise have no ethnic relation to indigenous groups in Mexico, as a member of the\nminority group (e.g., those who speak an indigenous language for professional reasons or through bilingual\neducation). Although our data is not conducive to identifying these cases, we expect these situations to comprise a\n\n\n10\n\nMention: \"IPUMS-I database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Other censuses with language variables were excluded over sample consistency.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005862:7:1", "occurrence_id": "4252c251215dc6436dfc27f2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  environment resulting in continued contamination of soil and water, and poten\ntially infecting others in the community who practice safe sanitation. In other words, even\n\nif my household has chosen to practice safe sanitation, if others in my community continue\n\nto defecate in the open, what are the consequences for the welfare of my household?\n\n\nObservational and quasi-experimental studies have looked at the impact of improved sanita\ntion on child health outcomes including diarrhea, infant mortality, and stunting. However,\n\nmost of these studies have focused on the effects at the household level and have not consid\nered the importance, or additional effects, that may be present when an entire community\n\nhas access to sanitation. The most relevant for the current study is a recent paper that\n\nused propensity score matching to estimate the effect of access to improved sanitation on\n\ndiarrhea in children under five using the District Level Household Survey (DLHS-3) (Ku\n\n3Indeed, some have suggested there is too little evidence on the complementary effects of WSH in\nterventions and that too much attention has been focused on understanding the epidemiology of single\n\ntransmission pathways (see Eisenberg (2007 and 2012)).\n\n4This effect also goes the other way, with poor nutritional status associated with increased risk of diarrhea\n\n(see for example Chen et al. (1981)).\n\n\n6\n\nMention: \"District Level Household Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Child diarrhoea and sanitation access are estimated from this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005862:7:1", "occurrence_id": "8a96505132d9bf9ea6b2b552", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  environment resulting in continued contamination of soil and water, and poten\ntially infecting others in the community who practice safe sanitation. In other words, even\n\nif my household has chosen to practice safe sanitation, if others in my community continue\n\nto defecate in the open, what are the consequences for the welfare of my household?\n\n\nObservational and quasi-experimental studies have looked at the impact of improved sanita\ntion on child health outcomes including diarrhea, infant mortality, and stunting. However,\n\nmost of these studies have focused on the effects at the household level and have not consid\nered the importance, or additional effects, that may be present when an entire community\n\nhas access to sanitation. The most relevant for the current study is a recent paper that\n\nused propensity score matching to estimate the effect of access to improved sanitation on\n\ndiarrhea in children under five using the District Level Household Survey (DLHS-3) (Ku\n\n3Indeed, some have suggested there is too little evidence on the complementary effects of WSH in\nterventions and that too much attention has been focused on understanding the epidemiology of single\n\ntransmission pathways (see Eisenberg (2007 and 2012)).\n\n4This effect also goes the other way, with poor nutritional status associated with increased risk of diarrhea\n\n(see for example Chen et al. (1981)).\n\n\n6\n\nMention: \"DLHS-3\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The same round as the short-name reference.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005863:26:0", "occurrence_id": "b0501c3106b83eb36546b9e3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Table 1: Summary Statistics (Project-level)**\n\n\n**<u>Source</u>** **<u>N</u>** **<u>Mean</u>** **<u>SD</u>** **<u>P50</u>** **<u>Min</u>** **<u>Max</u>**\nDisputes PPI and UNCTAD 5237 0.080 0.272 0 0 1\n\n\n**<u>Micro variables</u>**\nDuration (Contract age) PPI 5237 9.569 5.999 9 0 28\nIFI involvement PPI 5156 0.131 0.338 0 0 1\nPrivate share > 50% PPI 4989 0.905 0.293 1 0 1\nPrivate share > 80% PPI 4989 0.755 0.430 1 0 1\nCompetitive bidding PPI 1987 0.724 0.447 1 0 1\n\n\n**<u>Macro variables</u>**\nChange in real per capita GDP (during contract) WDI 5199 0.009 0.056 0.002 -0.360 0.563\nChange in term of trade (TOT) (during contract) WDI 4991 0.114 0.380 0 -0.684 5.139\nIncrease in input cost (average) World Bank 4722 0.064 0.060 0.057 -0.272 0.446\nCommodity windfall revenue (average) World Bank, UN Comtrade 5191 0.081 0.053 0.091 -0.230 0.345\nDemocracy Polity IV 5187 0.741 0.438 1 0 1\nRight wing government (last year) Database of political institution 3846 0.197 0.398 0 0 1\nLog income per capita in 2000 Besley and\n\nMention: \"Database of political institution\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its variable records whether a right-wing government held office the previous year.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005865:12:1", "occurrence_id": "c6bd15caca73197c1776c146", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  inequality over time_\n\nAs mentioned before, the distribution of household expenditure per capita is changing over time for many\ncountries. In the following analysis, we propose what may be termed “inequality-adjusted growth rate” to\ncapture the impact of change in inequality on poverty. Although we will still have to rely on the\ndistribution-neutral growth assumption using the 2010 income distribution in PovcalNet, we do so, for all\navailable countries, by translating changes in inequality between 2002 and 2010 into the projected growth\nrate such that countries that have reduced inequality during this period would have a higher projected\ngrowth rate than the actual growth rate (so that projected poverty rate in 2030 will be lower) and vice\nversa. An underlying assumption is that the trend in inequality between 2002 and 2010 would continue\nbetween 2010 and 2030.\n\nThe inequality adjustment is carried out in the following manner. In PovcalNet, for each country, we\nchoose the 2010 distribution, and keep shifting the mean of the distribution until we obtain the 2002\npoverty rate. <sup>10</sup> [^10: Practically speaking, the user of PovcalNet must keep reducing poverty line (as opposed to increasing the mean)\nfrom the default value of US$38 (the monthly value of US$1.25 per day poverty line) until we obtain the 2002\npoverty rate, because poverty line is the only variable in PovcalNet.] The ratio of the original mean to the resulting mean will give us the total growth rate\nrequired to reduce the poverty rate measured at US$1.25 per day between 2002 and 2010 if the\ndistribution is constant at the 2010 level. Finally, we annualize the required growth rate.\n\nFigure 4 illustrates this process. The upper panel shows the 2002 and the 2010 distributions of household\nexpenditure per capita. In this example, it appears that the 2010 distribution is more unequal than the\n2002 distribution. The raw growth rate of mean household expenditure per capita is calculated by\ncomparing mean expenditures. In the lower panel, the 2010 distribution\n\nMention: \"PovcalNet\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Its income distribution supports the distribution-neutral growth projection.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005867:6:0", "occurrence_id": "d4a13c6f60e3112e191c8e51", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 2\nExport growth and time to export vs. lagged AFT, by quintile of the export per capita distribution\n\n\n\n120.0%\n\n100.0%\n\n\n\n120.0%\n\n100.0%\n\n80.0%\n\n\n\n80.0%\n\n60.0%\n\n\n\n80.0%\n\n60.0%\n\n40.0%\n\n\n\n60.0%\n\n40.0%\n\n20.0%\n\n\n\n40.0%\n\n20.0%\n\n0.0%\n\n\n\n-20.0%\n\n\n\n0.0%\n\n\n\n(a) Export growth (five-year cumulative) (b) Time to export\n\n\nQ1 Q2 Q3 Q4 Q5\n\nLOW HIGH LOW HIGH LOW HIGH LOW HIGH LOW HIGH\n0.0%\n\n\n-5.0%\n\n\n-10.0%\n\n\n-15.0%\n\n\n\n\n\n\n|Col1|Col2|Col3|Col4|Col5|\n|---|---|---|---|---|\n||||||\n||||||\n||||||\n||||||\n||||||\n|~~Low~~<br>~~High~~<br>Q1|~~Low~~<br>~~High~~<br>Q2|~~Low~~<br>~~High~~<br>Q3|~~Low~~<br>~~High~~<br>Q4|~~Low~~<br>~~High~~<br>Q5|\n\n\n\nSource: Authors’ calculations using OECD CRS database and WDI\n\n###### _b. Evidence from Econometric Studies_\n\n\nUsing the OECD's CRS, <sup>6</sup> Cali and te Velde (2011) regress trading costs and aggregate export value\n\n\non lagged AFT disbursements and control variables, on a panel of developing\n\nMention: \"OECD CRS database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Aid flows by recipient are drawn from this creditor system database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005867:6:0", "occurrence_id": "2c9b04cd57842209b7f84e8d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 2\nExport growth and time to export vs. lagged AFT, by quintile of the export per capita distribution\n\n\n\n120.0%\n\n100.0%\n\n\n\n120.0%\n\n100.0%\n\n80.0%\n\n\n\n80.0%\n\n60.0%\n\n\n\n80.0%\n\n60.0%\n\n40.0%\n\n\n\n60.0%\n\n40.0%\n\n20.0%\n\n\n\n40.0%\n\n20.0%\n\n0.0%\n\n\n\n-20.0%\n\n\n\n0.0%\n\n\n\n(a) Export growth (five-year cumulative) (b) Time to export\n\n\nQ1 Q2 Q3 Q4 Q5\n\nLOW HIGH LOW HIGH LOW HIGH LOW HIGH LOW HIGH\n0.0%\n\n\n-5.0%\n\n\n-10.0%\n\n\n-15.0%\n\n\n\n\n\n\n|Col1|Col2|Col3|Col4|Col5|\n|---|---|---|---|---|\n||||||\n||||||\n||||||\n||||||\n||||||\n|~~Low~~<br>~~High~~<br>Q1|~~Low~~<br>~~High~~<br>Q2|~~Low~~<br>~~High~~<br>Q3|~~Low~~<br>~~High~~<br>Q4|~~Low~~<br>~~High~~<br>Q5|\n\n\n\nSource: Authors’ calculations using OECD CRS database and WDI\n\n###### _b. Evidence from Econometric Studies_\n\n\nUsing the OECD's CRS, <sup>6</sup> Cali and te Velde (2011) regress trading costs and aggregate export value\n\n\non lagged AFT disbursements and control variables, on a panel of developing\n\nMention: \"WDI\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The companion compilation of recipient characteristics.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005879:19:0", "occurrence_id": "9c146207df38a1d35b25074f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Appendix B:**\n\n\n**<u>Variable</u>** **<u>Description</u>**\n\n\n\nPercent Change in Nominal\nExchange Rate\n\n\n\nPercent Change in Nominal Percent change in official exchange rate over the period indicated in the\nExchange Rate text. Data expressed as local currency per USD. Original data are\n\nrecorded as monthly averages. Source: World Bank GEM database.\nPercent Change in Stock Percent change in stock price index over the period indicated in the text.\nMarket Indices Original data are indices of local currency, with the index = 100 in the\n\n\n\nPercent Change in Stock Percent change in stock price index over the period indicated in the text.\nMarket Indices Original data are indices of local currency, with the index = 100 in the\n\nyear 2000. Exceptions: Kazakhstan and Morocco have data indexed to\nthe year 2010 in local currency, Israel has data indexed to the year 2000\nin USD, and Hong Kong SAR, China, and Serbia have data indexed to the\nyear 2005 in local currency. Source: World Bank GEM database; IMF\nIFS database (for Hong Kong SAR, China, and Serbia only).\nPercent Change in External Percent change in total reserves over the period indicated in the text.\nReserves Original data expressed in million USD. Source: World Bank GEM\n\n\n\nPercent Change in External Percent change in total reserves over the period indicated in the text.\nReserves Original data expressed in million USD. Source: World Bank GEM\n\ndatabase.\nBasis Points Change in Bond Absolute change in bond interest rate spreads over the period indicated in\nYields the text. Original data expressed in basis points over US Treasuries.\n\n\n\nBasis Points Change in Bond Absolute change in bond interest rate spreads over the period indicated in\nYields the text. Original data expressed in basis points over US Treasuries.\n\nSource: World Bank GEM database.\nBasis Points Change in Absolute change in 5 year credit default swap spread over the period\nCountry Default Swaps indicated in the text. Original data expressed\n\nMention: \"World Bank GEM database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Monthly currency and stock averages come from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005879:19:0", "occurrence_id": "af5f185ef86b55df58382a7f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Appendix B:**\n\n\n**<u>Variable</u>** **<u>Description</u>**\n\n\n\nPercent Change in Nominal\nExchange Rate\n\n\n\nPercent Change in Nominal Percent change in official exchange rate over the period indicated in the\nExchange Rate text. Data expressed as local currency per USD. Original data are\n\nrecorded as monthly averages. Source: World Bank GEM database.\nPercent Change in Stock Percent change in stock price index over the period indicated in the text.\nMarket Indices Original data are indices of local currency, with the index = 100 in the\n\n\n\nPercent Change in Stock Percent change in stock price index over the period indicated in the text.\nMarket Indices Original data are indices of local currency, with the index = 100 in the\n\nyear 2000. Exceptions: Kazakhstan and Morocco have data indexed to\nthe year 2010 in local currency, Israel has data indexed to the year 2000\nin USD, and Hong Kong SAR, China, and Serbia have data indexed to the\nyear 2005 in local currency. Source: World Bank GEM database; IMF\nIFS database (for Hong Kong SAR, China, and Serbia only).\nPercent Change in External Percent change in total reserves over the period indicated in the text.\nReserves Original data expressed in million USD. Source: World Bank GEM\n\n\n\nPercent Change in External Percent change in total reserves over the period indicated in the text.\nReserves Original data expressed in million USD. Source: World Bank GEM\n\ndatabase.\nBasis Points Change in Bond Absolute change in bond interest rate spreads over the period indicated in\nYields the text. Original data expressed in basis points over US Treasuries.\n\n\n\nBasis Points Change in Bond Absolute change in bond interest rate spreads over the period indicated in\nYields the text. Original data expressed in basis points over US Treasuries.\n\nSource: World Bank GEM database.\nBasis Points Change in Absolute change in 5 year credit default swap spread over the period\nCountry Default Swaps indicated in the text. Original data expressed\n\nMention: \"IMF\nIFS database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Two economies' series are indexed using this source instead.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005886:31:5", "occurrence_id": "18d9392593ee62f2b2845ae8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 9<br>Latin America<br>(7)|1990-9<br>2000-9<br>G7 (6)|1990-9<br>2000-9<br>G7 (6)|1990-9<br>2000-9<br>G7 (6)|1990-9<br>2000-9<br>G7 (6)|1990-9<br>2000-9<br>Oth. Adv.<br>Economies (7)|1990-9<br>2000-9<br>Oth. Adv.<br>Economies (7)|1990-9<br>2000-9<br>Oth. Adv.<br>Economies (7)|1990-9<br>2000-9<br>Oth. Adv.<br>Economies (7)|1990-9<br>2000-9<br>Oth. Adv.<br>Economies (7)|\n\n\n\nBanks Bonds Equities\n\n\nThis figure shows the average size and structure of domestic financial systems between 1990 and 2009. Panel A shows total banking claims,\noutstanding bonds and equity market capitalization as a percentage of GDP. Panel B shows the same figures expressed as percentage of total\ndomestic financial system. Numbers in parentheses show the number of countries in each region. The data sources are the IMF's\nInternational Financial Statistics (IFS), the Bank for International Settlements (BIS), and World Bank's World Development Indicators\n(WDI).\n\nMention: \"International Financial Statistics\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"One of three sources behind the regional counts.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005886:31:5", "occurrence_id": "6b16d953f9d2c6b456e3c296", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 9<br>Latin America<br>(7)|1990-9<br>2000-9<br>G7 (6)|1990-9<br>2000-9<br>G7 (6)|1990-9<br>2000-9<br>G7 (6)|1990-9<br>2000-9<br>G7 (6)|1990-9<br>2000-9<br>Oth. Adv.<br>Economies (7)|1990-9<br>2000-9<br>Oth. Adv.<br>Economies (7)|1990-9<br>2000-9<br>Oth. Adv.<br>Economies (7)|1990-9<br>2000-9<br>Oth. Adv.<br>Economies (7)|1990-9<br>2000-9<br>Oth. Adv.<br>Economies (7)|\n\n\n\nBanks Bonds Equities\n\n\nThis figure shows the average size and structure of domestic financial systems between 1990 and 2009. Panel A shows total banking claims,\noutstanding bonds and equity market capitalization as a percentage of GDP. Panel B shows the same figures expressed as percentage of total\ndomestic financial system. Numbers in parentheses show the number of countries in each region. The data sources are the IMF's\nInternational Financial Statistics (IFS), the Bank for International Settlements (BIS), and World Bank's World Development Indicators\n(WDI).\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The third source behind the regional figures.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005890:17:0", "occurrence_id": "e1cefee00db4b9e6049fc9fa", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: cycle: project evaluation, implementation and facility operations, selection and budgeting,\nscreening and appraisal (vertical axis). In doing so, it provides a benchmark against which one\ncan assess the potential impact of public investment on growth given connections between costeffective asset creation _ex post_ and the conditional quality of project proposal screening,\nappraising and budgeting _ex ante_ . <sup>26</sup> [^26: See also Dabla-Norris et al. (2011) for a different approach to creating cross-country PIM indexes.] The link between capital investment – asset creation – and\ngrowth cannot be taken for granted, however, especially in the context of poor governance and\nrudimentary public investment systems, as reflected in multiple gaps in all stages of the PIMa\ncycle. This fact underscores why it is also important to examine each country’s results beyond\nthe headline number in the index and consider a country’s performance on various individual\ncomponents of the index. While the PIMa indicators cover a broad range of components, what\ndistinguishes the outcomes across a range of country studies is the degree of centralization of the\nproject screening and appraisal responsibilities and the connection between capital and recurrent\nexpenditures beyond the project construction period.\n\n\n**Figure 1** : Public Investment Management (PIMa) - Selected Countries\n\n\n\n\n\n\n\n\n\n\n\n\n\n_Sources_ : Rajaram et al, et al. (2011) and World Bank country case studies.\n\n\n**A review of in-depth country studies across different country-settings and different levels**\n**of development** (ranging from high income countries with more advanced public investment\nmanagement systems to donor-dependent and resource rich countries) suggests that these\ncountries lie between two extremes: one set of countries with full or near fully effective\nfunctioning PIMa’s at all stages and another set of countries where core PIMa functions are\neither missing or existing only de jure but not in practice. <sup>27</sup> It is for the second group of\ncountries with weak PIMas that an upstream ‘gate keeping’ stage, linked with\n\nMention: \"the PIMa indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Its broad component range distinguishes country performance.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005899:18:0", "occurrence_id": "4c08bf3334fd8ad0295de0ef", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 9: Firms with Bank Accounts and Loans**\n\n\nSource: Author’s calculations based on data from Enterprise Surveys and Informal Enterprise Surveys.\n\n\nLooking at various sources firms use for financing their day-to-day operations, Figure 10 shows\n\nthat majority of firms - 70 percent informal and 63 percent formal, rely on their retained earnings, their\n\nfamilies and friends or on moneylenders for financing. Trade credit is the second most frequently used\n\nsource with 23 percent of informal and 24 percent of formal firms relying on it. Bank financing is the\n\nnext option firms opt for, however only 7 percent of informal firms use this option. This is not surprising\n\ngiven that informal firms lack the required documentation to qualify for a bank loan. In the formal sector\n\n24 percent of firms rely on banks to finance their working capital. Lastly, 7 percent of informal firms use\n\nmicrofinance institutions (MFIs) and 6 percent of formal firms use non-bank financial institutions (NBFIs)\n\nfor their day-to-day financing. Figure 11, which highlights sources of financing for investments, reveals a\n\nvery similar picture. Around 76 percent of informal firms and 65 percent of formal firms identify using\n\ntheir retained earnings, their families and friends or moneylenders for financing investments. The next\n\noption used by informal firms is trade credit (15% of firms) and by formal firms is banks (24% of firms).\n\nMFIs by informal firms (5% of firms) and NBFIs by formal firms (5% of firms) are the least used means of\n\nfinancing investments.\n\n\n17\n\nMention: \"Enterprise Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Bank account and loan holding are calculated from these surveys.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005899:18:0", "occurrence_id": "a9902050fe85af2354c8a7e3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 9: Firms with Bank Accounts and Loans**\n\n\nSource: Author’s calculations based on data from Enterprise Surveys and Informal Enterprise Surveys.\n\n\nLooking at various sources firms use for financing their day-to-day operations, Figure 10 shows\n\nthat majority of firms - 70 percent informal and 63 percent formal, rely on their retained earnings, their\n\nfamilies and friends or on moneylenders for financing. Trade credit is the second most frequently used\n\nsource with 23 percent of informal and 24 percent of formal firms relying on it. Bank financing is the\n\nnext option firms opt for, however only 7 percent of informal firms use this option. This is not surprising\n\ngiven that informal firms lack the required documentation to qualify for a bank loan. In the formal sector\n\n24 percent of firms rely on banks to finance their working capital. Lastly, 7 percent of informal firms use\n\nmicrofinance institutions (MFIs) and 6 percent of formal firms use non-bank financial institutions (NBFIs)\n\nfor their day-to-day financing. Figure 11, which highlights sources of financing for investments, reveals a\n\nvery similar picture. Around 76 percent of informal firms and 65 percent of formal firms identify using\n\ntheir retained earnings, their families and friends or moneylenders for financing investments. The next\n\noption used by informal firms is trade credit (15% of firms) and by formal firms is banks (24% of firms).\n\nMFIs by informal firms (5% of firms) and NBFIs by formal firms (5% of firms) are the least used means of\n\nfinancing investments.\n\n\n17\n\nMention: \"Informal Enterprise Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The companion round covering unregistered firms.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005909:11:0", "occurrence_id": "65280734fe3015201134066f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **_Car Imports_**\n\nBecause trade classification does not distinguish between new and secondhand cars, our study\n\nfocused on informal traders and secondhand cars. The analysis then proceeded in several\n\ndifferentiation phases to assess the importance of informal trade.\n\n\nDuring the first phase, we excluded buses from tariff line 8703 because interviewees rejected the\n\nhypothesis that buses were being imported by informal traders.\n\n\nDuring the second phase, we attempted to identify cars sold on the informal market. In 2012\n\napproximately 32,000 vehicles were imported into Douala port for consumption <sup>6</sup> [^6: The cars were identified in ASYCUDA as “supplementary units.”] purposes. The\n\nbreakdown of car imports by value appears in table 2.\n\n\n**Table 2. Breakdown of Imported Cars by Value, Douala port, Cameroon**\n\n|Category of vehicles by declared value|Percentage of total number<br>of secondhand vehicles|\n|---|---|\n|≤1 million CFAF (€1,500) <br>|49.0<br>|\n|<br>>1 million and≤2 million CFAF (€3,000) <br>|28.0 <br>|\n|<br>>2 million and≤3 million CFAF (€4,500) <br>|<br>10.0 <br>|\n|<br>>3 million and≤4 million CFAF (€6,000) <br>|<br>4.0 <br>|\n|<br>>4 million and≤5 million CFAF (€7,500) <br>|<br>2.0 <br>|\n|<br>>5 million and≤10 million CFAF (€3,000) <br>|<br>3.0 <br>|\n|<br>>10 million and≤15 million CFAF (€3,000) <br\n\nMention: \"ASYCUDA\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Vehicle imports are identified in this customs system as supplementary units.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005880:10:2", "occurrence_id": "7864d5050e449e325ad0a380", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  total gross value added divided by total employment|* Usage of power is ignored and only the criteria of 10 or more employees is considered for registration under the Factories Act,1948.<br>  Excludes firms with missng or zero employment.<br>**Includes unpaid family membes/helpers working in the firm.<br>***Labor productivity is total gross value added divided by total employment|* Usage of power is ignored and only the criteria of 10 or more employees is considered for registration under the Factories Act,1948.<br>  Excludes firms with missng or zero employment.<br>**Includes unpaid family membes/helpers working in the firm.<br>***Labor productivity is total gross value added divided by total employment|* Usage of power is ignored and only the criteria of 10 or more employees is considered for registration under the Factories Act,1948.<br>  Excludes firms with missng or zero employment.<br>**Includes unpaid family membes/helpers working in the firm.<br>***Labor productivity is total gross value added divided by total employment|\n\n\nSome idea of the trends is given by comparing the 2010-11 NSSO survey and 2009-10 ASI with the 200506 NSSO survey and ASI in (Table 2). Share of formal firms went from 0.7% of all firms to 0.8% over this five year\nperiod, and their employment share increased from 19.5% to 24.8%. Thus, the employment share of total\ninformality (B, C and D) declined correspondingly. The employment share of noncompliant firms (B) went from\n8.9% to 9.8%, a clear upward trend. The share of employment of the avoiders (C) does not seem to have\nchanged at all, while the share of the outsiders (D\n\nMention: \"NSSO survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Two rounds bracket the change in the formal firm share.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005880:10:2", "occurrence_id": "6a58ecca2ecfccdc45b42d90", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  total gross value added divided by total employment|* Usage of power is ignored and only the criteria of 10 or more employees is considered for registration under the Factories Act,1948.<br>  Excludes firms with missng or zero employment.<br>**Includes unpaid family membes/helpers working in the firm.<br>***Labor productivity is total gross value added divided by total employment|* Usage of power is ignored and only the criteria of 10 or more employees is considered for registration under the Factories Act,1948.<br>  Excludes firms with missng or zero employment.<br>**Includes unpaid family membes/helpers working in the firm.<br>***Labor productivity is total gross value added divided by total employment|* Usage of power is ignored and only the criteria of 10 or more employees is considered for registration under the Factories Act,1948.<br>  Excludes firms with missng or zero employment.<br>**Includes unpaid family membes/helpers working in the firm.<br>***Labor productivity is total gross value added divided by total employment|\n\n\nSome idea of the trends is given by comparing the 2010-11 NSSO survey and 2009-10 ASI with the 200506 NSSO survey and ASI in (Table 2). Share of formal firms went from 0.7% of all firms to 0.8% over this five year\nperiod, and their employment share increased from 19.5% to 24.8%. Thus, the employment share of total\ninformality (B, C and D) declined correspondingly. The employment share of noncompliant firms (B) went from\n8.9% to 9.8%, a clear upward trend. The share of employment of the avoiders (C) does not seem to have\nchanged at all, while the share of the outsiders (D\n\nMention: \"ASI\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The companion industry series for the same comparison.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005905:10:1", "occurrence_id": "7c05659bf7e612982c0e2f4b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  not complain” (Chigwenembe, 2011). Practices that fail to respect patients’\ndignity remain widespread even though they violate established professional standards.\n\n\nThus, negative attitudes toward certain groups appear to be widespread and are likely to find expression\nin behavior, sometimes leading to exclusion. In order to better understand what might drive intolerant\nattitudes, we next turn to our statistical analysis.\n\n\n**3. Statistical analysis of intolerant attitudes**\n\n\nThis section presents an empirical analysis of unwelcoming attitudes and their determinants, using LiTS\nsurvey responses. We find three distinct clusters or components of unwelcoming attitudes. This means\nthat the degree to which people are welcoming or rejecting of other groups can be traced back to three\ndistinct sets of values. This allows us to propose a novel distinction of social exclusion attitudes not\npreviously identified in the empirical or theoretical literature: (1) intolerance for the poor and for\ndifferent lifecycle stages; (2) intolerance toward stigmatized attributes and behaviors; and (3)\nintolerance to specific identity groups. We infer that that these three distinct value clusters\nsystematically shape attitudes to other groups. Further, our empirical analysis demonstrates that\n\n\n‘non-institutional behavior’, that is behavior not guided by any norms, and that of Kabeer (2000) who uses the\nterm ‘unruly practices’ in referring to corruption, for example, or discriminatory behavior by teachers. We\nunderstand ‘off-protocol’ behavior to be that which is either non-institutional, or that which relies on norms from\nother institutions. The existence of these behaviors means that even where the norms of a given institution are not\ninherently discriminatory, or even where institutions attempt to address inequalities, the day to day practices of\nindividuals may nevertheless be discriminatory in nature.\n5 The survey question cannot discern what might have caused the disrespectful behavior, however, negative\nattitudes on the part of the service provider are likely to be an important factor. In comparison, among various\ncomplaints about health services\n\nMention: \"LiTS\nsurvey responses\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Attitudes toward outsiders and their components are estimated from these responses.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005911:36:1", "occurrence_id": "7ba1aff15671eafad3097e9b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: sup>**</sup>\n\n(1.70) (2.34)\n\nChange in number of members aged 18– 0.0836 <sup>*</sup> 0.0264\n60\n\n(1.82) (1.17)\n\n<u>Province dummies</u> <u>Yes</u> <u>Yes</u>\n\n<u>Number of observations</u> <u>2,210</u> <u>2,169</u>\n\nSource: Estimation from household samples from IFLS 2000 and 2007.\nNotes:  t-statistics in parentheses; <sup>*</sup> _p_ < 0.10, <sup>**</sup> _p_ < 0.05, <sup>***</sup> _p_ < 0.01.\n\nMention: \"IFLS 2000 and 2007\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Household samples across these two rounds carry the estimation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005917:9:1", "occurrence_id": "a0d35c9ba7fb189595e3f83d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  drawn from Bown (2012). This dataset contains information on the case numbers, countries\n\n\nfiled against, HS codes and the date a petition is filed. The presence of the HS codes allows us to link\n\n\nthis information to detailed U.S. trade data. Our trade data are the HS product level information\n\n\non U.S. imports available in the monthly reports of the _U.S._ _Imports_ _of_ _Merchandise_ _Data_ . <sup>12</sup> These\n\n\ndata contain substantial information on the value and quantity of imports at the product level.\n\n\nFrom this value and quantity information we derive unit values, providing us with a very detailed\n\n\ndataset for analysis. We use _Imports_ _for_ _Consumption_, the classification of U.S. imports used in\n\n\n9The focus on supply and demand shocks is similar to that in Grossman (1986), Pindyck and Rotemberg (1987)\nand Kelly (1988). However, their methods and data differ from ours along with the specific questions they address.\n\n10We employ product-specific fixed effects, which should sweep out average differences in the underlying propensity\nto file.\n\n11Unfortunately the relevant trade data contained in the petition is not made available to the public.\n12Despite the public availability of these data they are relatively underutilized. Exceptions include Hummels and\nSchaur (2010), Chor and Manova (2012), and within the AD literature, Besedeˇs and Prusa (2013).\n\n\n7\n\nMention: \"U.S. trade data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Harmonized-system level import information is linked through these codes.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005919:52:3", "occurrence_id": "5ce8fbc9d5df9d889728de53", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: u>\n\n\n\n<u>Marginal effects;</u> _<u>t</u>_ <u>statistics in parentheses; (d) for discrete change of dummy variable from 0 to 1</u>\n\n<u>*</u> _<u>p</u>_ <u>< 0.10, **</u> _<u>p</u>_ <u>< 0.05, ***</u> _<u>p</u>_ <u>< 0.01</u>\n_<u>Note:</u>_ <u>Estimates employ the same set of covariates as model 3 in annex table A.7. DPI = database of political institutions.</u>\n\n\n\n**Table A.10 Ordered Probit Regression Estimates with Ordinal IEG Outcome Ratings for Project Management Variables**\n\n**Observable during the First Half of Project Implementation: Marginal Effects Reported for PSM Projects**\n\n\n\nModerately\nUnsatisfactory Rating\nUnsatisfactory Rating\n\n\n\n<u>Marginal Probability of…</u>\n\n\n\nSatisfactory\n\nRating\n\n\n\n<u>Highly</u>\nSatisfactory\n\n\n\n<u>Rating</u>\n\n\n\n<u>Moderately</u>\nSatisfactory\n\n\n\n<u>Rating</u>\n\n\n\nVariable\n\n\n\n<u>Highly</u>\nUnsatisfactory\n\n\n\n<u>Rating</u>\n\n\n\n<u>Committed Amount</u> <u>-0.0000123</u> <u>-0.0000646</u> <u>-0.0000362</u> <u>-0.0000146</u> <u>0.000108</u> <u>0.0000197</u>\n(-0.61) (-0.60)\n\nMention: \"database of political institutions\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its abbreviation defines the covariates in the annex model.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005947:19:1", "occurrence_id": "60d5b54df4d06991f11da3ce", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  student body of 1,450 undergraduate students and 200\n\npostgraduate and doctoral students - the total of 1,650 students. In the Faculty for Business Studies\n\nthere were 450 students in all programs, with 100 new students (1st year of study) and 138 new\n\nstudents in the previous year. The data are from Monstat – the statistical office of Montenegro.\n\n\n18\n\nMention: \"Monstat\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The national statistical office supplying the student records.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005874:12:0", "occurrence_id": "c74181949939ad80a4a8a6c2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **3.2.** **Employment**\nIn this paper, we predominantly use data from two labor force surveys (LFS) and two household\nincome consumption and expenditure surveys (HICES), namely: HICES 1995/96, LFS 1999, LFS\n2005 and HICES 2010/11. The HICES 1999/00 and HICES 2004/05 are not included because the\nlabor force surveys cover similar time periods and provide more comprehensive information on\nemployment issues. <sup>7</sup> Moreover, the HICES 1999/00 does not provide data on the sector of\nemployment, which is a crucial variable for our analysis on structural transformation. In addition,\nwe do not use Census data. The 2007 Census also does not provide data on the sector of\nemployment, <sup>8</sup> [^8: The questions relating to occupation and industry were dropped from the questionnaire when it was observed that\n[some trainers had difficulty in understanding concepts and definitions relating to the occupational and industrial](http://unstats.un.org/unsd/censuskb20/Attachment489.aspx)\nclassifications (CSA, 2012: 49). <u>[http://unstats.un.org/unsd/censuskb20/Attachment489.aspx](http://unstats.un.org/unsd/censuskb20/Attachment489.aspx)</u>] while the 1984 Census includes data for Eritrea. The 1994 Census is fairly close to\nthe first HICES and is not included to minimize the use of multiple data sources. The use of the\nHICES 2010/11 survey is particularly crucial, since it enables us to investigate recent sectoral\nemployment dynamics (and thus structural change).\nBefore analyzing employment dynamics, it is important to investigate the underlying demographic\ntrends. However, as we will note below, there can be significant discrepancies across data sources.\nFor instance, while Census data is the primary source of population statistics, the methodology we\nutilize later\n\nMention: \"labor force surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Two rounds supply the employment series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005874:12:0", "occurrence_id": "9372e8d26827af360976e8af", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **3.2.** **Employment**\nIn this paper, we predominantly use data from two labor force surveys (LFS) and two household\nincome consumption and expenditure surveys (HICES), namely: HICES 1995/96, LFS 1999, LFS\n2005 and HICES 2010/11. The HICES 1999/00 and HICES 2004/05 are not included because the\nlabor force surveys cover similar time periods and provide more comprehensive information on\nemployment issues. <sup>7</sup> Moreover, the HICES 1999/00 does not provide data on the sector of\nemployment, which is a crucial variable for our analysis on structural transformation. In addition,\nwe do not use Census data. The 2007 Census also does not provide data on the sector of\nemployment, <sup>8</sup> [^8: The questions relating to occupation and industry were dropped from the questionnaire when it was observed that\n[some trainers had difficulty in understanding concepts and definitions relating to the occupational and industrial](http://unstats.un.org/unsd/censuskb20/Attachment489.aspx)\nclassifications (CSA, 2012: 49). <u>[http://unstats.un.org/unsd/censuskb20/Attachment489.aspx](http://unstats.un.org/unsd/censuskb20/Attachment489.aspx)</u>] while the 1984 Census includes data for Eritrea. The 1994 Census is fairly close to\nthe first HICES and is not included to minimize the use of multiple data sources. The use of the\nHICES 2010/11 survey is particularly crucial, since it enables us to investigate recent sectoral\nemployment dynamics (and thus structural change).\nBefore analyzing employment dynamics, it is important to investigate the underlying demographic\ntrends. However, as we will note below, there can be significant discrepancies across data sources.\nFor instance, while Census data is the primary source of population statistics, the methodology we\nutilize later\n\nMention: \"household\nincome consumption and expenditure surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Two rounds supply income and consumption measures.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005874:12:0", "occurrence_id": "c96699492cf7ee08d293c790", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **3.2.** **Employment**\nIn this paper, we predominantly use data from two labor force surveys (LFS) and two household\nincome consumption and expenditure surveys (HICES), namely: HICES 1995/96, LFS 1999, LFS\n2005 and HICES 2010/11. The HICES 1999/00 and HICES 2004/05 are not included because the\nlabor force surveys cover similar time periods and provide more comprehensive information on\nemployment issues. <sup>7</sup> Moreover, the HICES 1999/00 does not provide data on the sector of\nemployment, which is a crucial variable for our analysis on structural transformation. In addition,\nwe do not use Census data. The 2007 Census also does not provide data on the sector of\nemployment, <sup>8</sup> [^8: The questions relating to occupation and industry were dropped from the questionnaire when it was observed that\n[some trainers had difficulty in understanding concepts and definitions relating to the occupational and industrial](http://unstats.un.org/unsd/censuskb20/Attachment489.aspx)\nclassifications (CSA, 2012: 49). <u>[http://unstats.un.org/unsd/censuskb20/Attachment489.aspx](http://unstats.un.org/unsd/censuskb20/Attachment489.aspx)</u>] while the 1984 Census includes data for Eritrea. The 1994 Census is fairly close to\nthe first HICES and is not included to minimize the use of multiple data sources. The use of the\nHICES 2010/11 survey is particularly crucial, since it enables us to investigate recent sectoral\nemployment dynamics (and thus structural change).\nBefore analyzing employment dynamics, it is important to investigate the underlying demographic\ntrends. However, as we will note below, there can be significant discrepancies across data sources.\nFor instance, while Census data is the primary source of population statistics, the methodology we\nutilize later\n\nMention: \"HICES 1995/96\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The earliest retained round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005874:12:0", "occurrence_id": "76f91d4ca7e0b68b02c06a2b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **3.2.** **Employment**\nIn this paper, we predominantly use data from two labor force surveys (LFS) and two household\nincome consumption and expenditure surveys (HICES), namely: HICES 1995/96, LFS 1999, LFS\n2005 and HICES 2010/11. The HICES 1999/00 and HICES 2004/05 are not included because the\nlabor force surveys cover similar time periods and provide more comprehensive information on\nemployment issues. <sup>7</sup> Moreover, the HICES 1999/00 does not provide data on the sector of\nemployment, which is a crucial variable for our analysis on structural transformation. In addition,\nwe do not use Census data. The 2007 Census also does not provide data on the sector of\nemployment, <sup>8</sup> [^8: The questions relating to occupation and industry were dropped from the questionnaire when it was observed that\n[some trainers had difficulty in understanding concepts and definitions relating to the occupational and industrial](http://unstats.un.org/unsd/censuskb20/Attachment489.aspx)\nclassifications (CSA, 2012: 49). <u>[http://unstats.un.org/unsd/censuskb20/Attachment489.aspx](http://unstats.un.org/unsd/censuskb20/Attachment489.aspx)</u>] while the 1984 Census includes data for Eritrea. The 1994 Census is fairly close to\nthe first HICES and is not included to minimize the use of multiple data sources. The use of the\nHICES 2010/11 survey is particularly crucial, since it enables us to investigate recent sectoral\nemployment dynamics (and thus structural change).\nBefore analyzing employment dynamics, it is important to investigate the underlying demographic\ntrends. However, as we will note below, there can be significant discrepancies across data sources.\nFor instance, while Census data is the primary source of population statistics, the methodology we\nutilize later\n\nMention: \"LFS 1999\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The first retained labour round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005874:12:0", "occurrence_id": "9cba236f8ab7b7fef9ba712f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **3.2.** **Employment**\nIn this paper, we predominantly use data from two labor force surveys (LFS) and two household\nincome consumption and expenditure surveys (HICES), namely: HICES 1995/96, LFS 1999, LFS\n2005 and HICES 2010/11. The HICES 1999/00 and HICES 2004/05 are not included because the\nlabor force surveys cover similar time periods and provide more comprehensive information on\nemployment issues. <sup>7</sup> Moreover, the HICES 1999/00 does not provide data on the sector of\nemployment, which is a crucial variable for our analysis on structural transformation. In addition,\nwe do not use Census data. The 2007 Census also does not provide data on the sector of\nemployment, <sup>8</sup> [^8: The questions relating to occupation and industry were dropped from the questionnaire when it was observed that\n[some trainers had difficulty in understanding concepts and definitions relating to the occupational and industrial](http://unstats.un.org/unsd/censuskb20/Attachment489.aspx)\nclassifications (CSA, 2012: 49). <u>[http://unstats.un.org/unsd/censuskb20/Attachment489.aspx](http://unstats.un.org/unsd/censuskb20/Attachment489.aspx)</u>] while the 1984 Census includes data for Eritrea. The 1994 Census is fairly close to\nthe first HICES and is not included to minimize the use of multiple data sources. The use of the\nHICES 2010/11 survey is particularly crucial, since it enables us to investigate recent sectoral\nemployment dynamics (and thus structural change).\nBefore analyzing employment dynamics, it is important to investigate the underlying demographic\ntrends. However, as we will note below, there can be significant discrepancies across data sources.\nFor instance, while Census data is the primary source of population statistics, the methodology we\nutilize later\n\nMention: \"LFS\n2005\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The second retained labour round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005874:12:0", "occurrence_id": "7c256f8ed797ae9409dcfd54", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **3.2.** **Employment**\nIn this paper, we predominantly use data from two labor force surveys (LFS) and two household\nincome consumption and expenditure surveys (HICES), namely: HICES 1995/96, LFS 1999, LFS\n2005 and HICES 2010/11. The HICES 1999/00 and HICES 2004/05 are not included because the\nlabor force surveys cover similar time periods and provide more comprehensive information on\nemployment issues. <sup>7</sup> Moreover, the HICES 1999/00 does not provide data on the sector of\nemployment, which is a crucial variable for our analysis on structural transformation. In addition,\nwe do not use Census data. The 2007 Census also does not provide data on the sector of\nemployment, <sup>8</sup> [^8: The questions relating to occupation and industry were dropped from the questionnaire when it was observed that\n[some trainers had difficulty in understanding concepts and definitions relating to the occupational and industrial](http://unstats.un.org/unsd/censuskb20/Attachment489.aspx)\nclassifications (CSA, 2012: 49). <u>[http://unstats.un.org/unsd/censuskb20/Attachment489.aspx](http://unstats.un.org/unsd/censuskb20/Attachment489.aspx)</u>] while the 1984 Census includes data for Eritrea. The 1994 Census is fairly close to\nthe first HICES and is not included to minimize the use of multiple data sources. The use of the\nHICES 2010/11 survey is particularly crucial, since it enables us to investigate recent sectoral\nemployment dynamics (and thus structural change).\nBefore analyzing employment dynamics, it is important to investigate the underlying demographic\ntrends. However, as we will note below, there can be significant discrepancies across data sources.\nFor instance, while Census data is the primary source of population statistics, the methodology we\nutilize later\n\nMention: \"HICES 2010/11\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The latest round retained for the analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005874:12:0", "occurrence_id": "9a185071cfe02d1dd3efaadc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **3.2.** **Employment**\nIn this paper, we predominantly use data from two labor force surveys (LFS) and two household\nincome consumption and expenditure surveys (HICES), namely: HICES 1995/96, LFS 1999, LFS\n2005 and HICES 2010/11. The HICES 1999/00 and HICES 2004/05 are not included because the\nlabor force surveys cover similar time periods and provide more comprehensive information on\nemployment issues. <sup>7</sup> Moreover, the HICES 1999/00 does not provide data on the sector of\nemployment, which is a crucial variable for our analysis on structural transformation. In addition,\nwe do not use Census data. The 2007 Census also does not provide data on the sector of\nemployment, <sup>8</sup> [^8: The questions relating to occupation and industry were dropped from the questionnaire when it was observed that\n[some trainers had difficulty in understanding concepts and definitions relating to the occupational and industrial](http://unstats.un.org/unsd/censuskb20/Attachment489.aspx)\nclassifications (CSA, 2012: 49). <u>[http://unstats.un.org/unsd/censuskb20/Attachment489.aspx](http://unstats.un.org/unsd/censuskb20/Attachment489.aspx)</u>] while the 1984 Census includes data for Eritrea. The 1994 Census is fairly close to\nthe first HICES and is not included to minimize the use of multiple data sources. The use of the\nHICES 2010/11 survey is particularly crucial, since it enables us to investigate recent sectoral\nemployment dynamics (and thus structural change).\nBefore analyzing employment dynamics, it is important to investigate the underlying demographic\ntrends. However, as we will note below, there can be significant discrepancies across data sources.\nFor instance, while Census data is the primary source of population statistics, the methodology we\nutilize later\n\nMention: \"HICES 2004/05\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Excluded because the labour surveys cover similar periods.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005874:12:0", "occurrence_id": "2b75b5f5cc291f277c3e64bb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **3.2.** **Employment**\nIn this paper, we predominantly use data from two labor force surveys (LFS) and two household\nincome consumption and expenditure surveys (HICES), namely: HICES 1995/96, LFS 1999, LFS\n2005 and HICES 2010/11. The HICES 1999/00 and HICES 2004/05 are not included because the\nlabor force surveys cover similar time periods and provide more comprehensive information on\nemployment issues. <sup>7</sup> Moreover, the HICES 1999/00 does not provide data on the sector of\nemployment, which is a crucial variable for our analysis on structural transformation. In addition,\nwe do not use Census data. The 2007 Census also does not provide data on the sector of\nemployment, <sup>8</sup> [^8: The questions relating to occupation and industry were dropped from the questionnaire when it was observed that\n[some trainers had difficulty in understanding concepts and definitions relating to the occupational and industrial](http://unstats.un.org/unsd/censuskb20/Attachment489.aspx)\nclassifications (CSA, 2012: 49). <u>[http://unstats.un.org/unsd/censuskb20/Attachment489.aspx](http://unstats.un.org/unsd/censuskb20/Attachment489.aspx)</u>] while the 1984 Census includes data for Eritrea. The 1994 Census is fairly close to\nthe first HICES and is not included to minimize the use of multiple data sources. The use of the\nHICES 2010/11 survey is particularly crucial, since it enables us to investigate recent sectoral\nemployment dynamics (and thus structural change).\nBefore analyzing employment dynamics, it is important to investigate the underlying demographic\ntrends. However, as we will note below, there can be significant discrepancies across data sources.\nFor instance, while Census data is the primary source of population statistics, the methodology we\nutilize later\n\nMention: \"HICES 1999/00\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Excluded for lacking the sector of employment variable.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005874:12:0", "occurrence_id": "6abd82d7b7e979035917a731", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **3.2.** **Employment**\nIn this paper, we predominantly use data from two labor force surveys (LFS) and two household\nincome consumption and expenditure surveys (HICES), namely: HICES 1995/96, LFS 1999, LFS\n2005 and HICES 2010/11. The HICES 1999/00 and HICES 2004/05 are not included because the\nlabor force surveys cover similar time periods and provide more comprehensive information on\nemployment issues. <sup>7</sup> Moreover, the HICES 1999/00 does not provide data on the sector of\nemployment, which is a crucial variable for our analysis on structural transformation. In addition,\nwe do not use Census data. The 2007 Census also does not provide data on the sector of\nemployment, <sup>8</sup> [^8: The questions relating to occupation and industry were dropped from the questionnaire when it was observed that\n[some trainers had difficulty in understanding concepts and definitions relating to the occupational and industrial](http://unstats.un.org/unsd/censuskb20/Attachment489.aspx)\nclassifications (CSA, 2012: 49). <u>[http://unstats.un.org/unsd/censuskb20/Attachment489.aspx](http://unstats.un.org/unsd/censuskb20/Attachment489.aspx)</u>] while the 1984 Census includes data for Eritrea. The 1994 Census is fairly close to\nthe first HICES and is not included to minimize the use of multiple data sources. The use of the\nHICES 2010/11 survey is particularly crucial, since it enables us to investigate recent sectoral\nemployment dynamics (and thus structural change).\nBefore analyzing employment dynamics, it is important to investigate the underlying demographic\ntrends. However, as we will note below, there can be significant discrepancies across data sources.\nFor instance, while Census data is the primary source of population statistics, the methodology we\nutilize later\n\nMention: \"2007 Census\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Not used because it lacks the employment sector question.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005874:12:0", "occurrence_id": "316c8d88b91ac7fd0bba1a65", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **3.2.** **Employment**\nIn this paper, we predominantly use data from two labor force surveys (LFS) and two household\nincome consumption and expenditure surveys (HICES), namely: HICES 1995/96, LFS 1999, LFS\n2005 and HICES 2010/11. The HICES 1999/00 and HICES 2004/05 are not included because the\nlabor force surveys cover similar time periods and provide more comprehensive information on\nemployment issues. <sup>7</sup> Moreover, the HICES 1999/00 does not provide data on the sector of\nemployment, which is a crucial variable for our analysis on structural transformation. In addition,\nwe do not use Census data. The 2007 Census also does not provide data on the sector of\nemployment, <sup>8</sup> [^8: The questions relating to occupation and industry were dropped from the questionnaire when it was observed that\n[some trainers had difficulty in understanding concepts and definitions relating to the occupational and industrial](http://unstats.un.org/unsd/censuskb20/Attachment489.aspx)\nclassifications (CSA, 2012: 49). <u>[http://unstats.un.org/unsd/censuskb20/Attachment489.aspx](http://unstats.un.org/unsd/censuskb20/Attachment489.aspx)</u>] while the 1984 Census includes data for Eritrea. The 1994 Census is fairly close to\nthe first HICES and is not included to minimize the use of multiple data sources. The use of the\nHICES 2010/11 survey is particularly crucial, since it enables us to investigate recent sectoral\nemployment dynamics (and thus structural change).\nBefore analyzing employment dynamics, it is important to investigate the underlying demographic\ntrends. However, as we will note below, there can be significant discrepancies across data sources.\nFor instance, while Census data is the primary source of population statistics, the methodology we\nutilize later\n\nMention: \"1984 Census\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Mentioned because it includes data for one neighbouring country.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005874:12:0", "occurrence_id": "eef430b555610bfd414951f0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **3.2.** **Employment**\nIn this paper, we predominantly use data from two labor force surveys (LFS) and two household\nincome consumption and expenditure surveys (HICES), namely: HICES 1995/96, LFS 1999, LFS\n2005 and HICES 2010/11. The HICES 1999/00 and HICES 2004/05 are not included because the\nlabor force surveys cover similar time periods and provide more comprehensive information on\nemployment issues. <sup>7</sup> Moreover, the HICES 1999/00 does not provide data on the sector of\nemployment, which is a crucial variable for our analysis on structural transformation. In addition,\nwe do not use Census data. The 2007 Census also does not provide data on the sector of\nemployment, <sup>8</sup> [^8: The questions relating to occupation and industry were dropped from the questionnaire when it was observed that\n[some trainers had difficulty in understanding concepts and definitions relating to the occupational and industrial](http://unstats.un.org/unsd/censuskb20/Attachment489.aspx)\nclassifications (CSA, 2012: 49). <u>[http://unstats.un.org/unsd/censuskb20/Attachment489.aspx](http://unstats.un.org/unsd/censuskb20/Attachment489.aspx)</u>] while the 1984 Census includes data for Eritrea. The 1994 Census is fairly close to\nthe first HICES and is not included to minimize the use of multiple data sources. The use of the\nHICES 2010/11 survey is particularly crucial, since it enables us to investigate recent sectoral\nemployment dynamics (and thus structural change).\nBefore analyzing employment dynamics, it is important to investigate the underlying demographic\ntrends. However, as we will note below, there can be significant discrepancies across data sources.\nFor instance, while Census data is the primary source of population statistics, the methodology we\nutilize later\n\nMention: \"1994 Census\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Excluded as too close in time to the first household survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005874:12:0", "occurrence_id": "79ac0dd1e2d8b8c46cdb447d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **3.2.** **Employment**\nIn this paper, we predominantly use data from two labor force surveys (LFS) and two household\nincome consumption and expenditure surveys (HICES), namely: HICES 1995/96, LFS 1999, LFS\n2005 and HICES 2010/11. The HICES 1999/00 and HICES 2004/05 are not included because the\nlabor force surveys cover similar time periods and provide more comprehensive information on\nemployment issues. <sup>7</sup> Moreover, the HICES 1999/00 does not provide data on the sector of\nemployment, which is a crucial variable for our analysis on structural transformation. In addition,\nwe do not use Census data. The 2007 Census also does not provide data on the sector of\nemployment, <sup>8</sup> [^8: The questions relating to occupation and industry were dropped from the questionnaire when it was observed that\n[some trainers had difficulty in understanding concepts and definitions relating to the occupational and industrial](http://unstats.un.org/unsd/censuskb20/Attachment489.aspx)\nclassifications (CSA, 2012: 49). <u>[http://unstats.un.org/unsd/censuskb20/Attachment489.aspx](http://unstats.un.org/unsd/censuskb20/Attachment489.aspx)</u>] while the 1984 Census includes data for Eritrea. The 1994 Census is fairly close to\nthe first HICES and is not included to minimize the use of multiple data sources. The use of the\nHICES 2010/11 survey is particularly crucial, since it enables us to investigate recent sectoral\nemployment dynamics (and thus structural change).\nBefore analyzing employment dynamics, it is important to investigate the underlying demographic\ntrends. However, as we will note below, there can be significant discrepancies across data sources.\nFor instance, while Census data is the primary source of population statistics, the methodology we\nutilize later\n\nMention: \"HICES\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The household survey family used across the analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005874:12:0", "occurrence_id": "a8eb165967c8b2c994e943ca", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **3.2.** **Employment**\nIn this paper, we predominantly use data from two labor force surveys (LFS) and two household\nincome consumption and expenditure surveys (HICES), namely: HICES 1995/96, LFS 1999, LFS\n2005 and HICES 2010/11. The HICES 1999/00 and HICES 2004/05 are not included because the\nlabor force surveys cover similar time periods and provide more comprehensive information on\nemployment issues. <sup>7</sup> Moreover, the HICES 1999/00 does not provide data on the sector of\nemployment, which is a crucial variable for our analysis on structural transformation. In addition,\nwe do not use Census data. The 2007 Census also does not provide data on the sector of\nemployment, <sup>8</sup> [^8: The questions relating to occupation and industry were dropped from the questionnaire when it was observed that\n[some trainers had difficulty in understanding concepts and definitions relating to the occupational and industrial](http://unstats.un.org/unsd/censuskb20/Attachment489.aspx)\nclassifications (CSA, 2012: 49). <u>[http://unstats.un.org/unsd/censuskb20/Attachment489.aspx](http://unstats.un.org/unsd/censuskb20/Attachment489.aspx)</u>] while the 1984 Census includes data for Eritrea. The 1994 Census is fairly close to\nthe first HICES and is not included to minimize the use of multiple data sources. The use of the\nHICES 2010/11 survey is particularly crucial, since it enables us to investigate recent sectoral\nemployment dynamics (and thus structural change).\nBefore analyzing employment dynamics, it is important to investigate the underlying demographic\ntrends. However, as we will note below, there can be significant discrepancies across data sources.\nFor instance, while Census data is the primary source of population statistics, the methodology we\nutilize later\n\nMention: \"HICES 2010/11 survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"It enables investigation of recent sectoral employment dynamics.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005875:7:1", "occurrence_id": "b475fa11eb0b422de584e708", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: , 0 otherwise (i.e. interaction of the oil rents variable with a\ndummy variable indicating the first quartile)\nOil rents, 2nd quartile Oil rents if oil rents belong to the 2nd quartile in corresponding period,\n0 otherwise\nOil rents, 3rd quartile Oil rents if oil rents belong to the 3rd quartile in corresponding period,\n0 otherwise\nOil rents, 4th quartile Oil rents if oil rents belong to the highest quartile in corresponding\nperiod, 0 otherwise\nNote: The Variables oilrent ~~0~~ 025, oilrent ~~2~~ 550, oilrent 5075 and\noilrent ~~7~~ 500 add up to the variable oilrent\n\n\nPWT corresponds to the Penn World Tables (Heston, Summers, and Aten, 2012)\n\nIIASA corresponds to the IIASA-VID dateset on educational attainment (Lutz, Goujon, K.C., and Sanderson, 2007)\n\nPolity IV corresponds to the Polity IV dataset obtained from Teorell, Charron, Samanni, Holmberg, and Rothstein (2011)\n\nWDI corresponds to the World Development Indicators (2013)\n\n\n6\n\nMention: \"Penn World Tables\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"A variable key in the data appendix.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005875:7:1", "occurrence_id": "453f4fff027d9103e2ab0482", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: , 0 otherwise (i.e. interaction of the oil rents variable with a\ndummy variable indicating the first quartile)\nOil rents, 2nd quartile Oil rents if oil rents belong to the 2nd quartile in corresponding period,\n0 otherwise\nOil rents, 3rd quartile Oil rents if oil rents belong to the 3rd quartile in corresponding period,\n0 otherwise\nOil rents, 4th quartile Oil rents if oil rents belong to the highest quartile in corresponding\nperiod, 0 otherwise\nNote: The Variables oilrent ~~0~~ 025, oilrent ~~2~~ 550, oilrent 5075 and\noilrent ~~7~~ 500 add up to the variable oilrent\n\n\nPWT corresponds to the Penn World Tables (Heston, Summers, and Aten, 2012)\n\nIIASA corresponds to the IIASA-VID dateset on educational attainment (Lutz, Goujon, K.C., and Sanderson, 2007)\n\nPolity IV corresponds to the Polity IV dataset obtained from Teorell, Charron, Samanni, Holmberg, and Rothstein (2011)\n\nWDI corresponds to the World Development Indicators (2013)\n\n\n6\n\nMention: \"IIASA-VID dateset\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Educational attainment projections come from this dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005875:7:1", "occurrence_id": "4d04926c93984da4f1e022a2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: , 0 otherwise (i.e. interaction of the oil rents variable with a\ndummy variable indicating the first quartile)\nOil rents, 2nd quartile Oil rents if oil rents belong to the 2nd quartile in corresponding period,\n0 otherwise\nOil rents, 3rd quartile Oil rents if oil rents belong to the 3rd quartile in corresponding period,\n0 otherwise\nOil rents, 4th quartile Oil rents if oil rents belong to the highest quartile in corresponding\nperiod, 0 otherwise\nNote: The Variables oilrent ~~0~~ 025, oilrent ~~2~~ 550, oilrent 5075 and\noilrent ~~7~~ 500 add up to the variable oilrent\n\n\nPWT corresponds to the Penn World Tables (Heston, Summers, and Aten, 2012)\n\nIIASA corresponds to the IIASA-VID dateset on educational attainment (Lutz, Goujon, K.C., and Sanderson, 2007)\n\nPolity IV corresponds to the Polity IV dataset obtained from Teorell, Charron, Samanni, Holmberg, and Rothstein (2011)\n\nWDI corresponds to the World Development Indicators (2013)\n\n\n6\n\nMention: \"Polity IV dataset\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Regime measures come from this dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005875:7:1", "occurrence_id": "a2b24b3ea23a2cb71d9e09d4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: , 0 otherwise (i.e. interaction of the oil rents variable with a\ndummy variable indicating the first quartile)\nOil rents, 2nd quartile Oil rents if oil rents belong to the 2nd quartile in corresponding period,\n0 otherwise\nOil rents, 3rd quartile Oil rents if oil rents belong to the 3rd quartile in corresponding period,\n0 otherwise\nOil rents, 4th quartile Oil rents if oil rents belong to the highest quartile in corresponding\nperiod, 0 otherwise\nNote: The Variables oilrent ~~0~~ 025, oilrent ~~2~~ 550, oilrent 5075 and\noilrent ~~7~~ 500 add up to the variable oilrent\n\n\nPWT corresponds to the Penn World Tables (Heston, Summers, and Aten, 2012)\n\nIIASA corresponds to the IIASA-VID dateset on educational attainment (Lutz, Goujon, K.C., and Sanderson, 2007)\n\nPolity IV corresponds to the Polity IV dataset obtained from Teorell, Charron, Samanni, Holmberg, and Rothstein (2011)\n\nWDI corresponds to the World Development Indicators (2013)\n\n\n6\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Development indicators come from this edition.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005949:36:2", "occurrence_id": "2e7191e737c1c1ee0c4766a8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: , IMF International Financial Statistics, World Development Indicators</u>\n\n<u>External debt/GDP</u> <u>World Development Indicators, Datastream, BIS</u>\n\n\n\n<u>Trade/GDP</u>\n\n<u>External debt/GDP</u> <u>World Development Indicators, Datastream, BIS</u>\n\n<u>Private sector credit/GDP</u> <u>IMF International Financial Statistics</u>\n\n\n\n<u>External debt/GDP</u> <u>World Development Indicators, Datastream, BIS</u>\n\n<u>Private sector credit/GDPNote: All variables are at quarterly frequency, unless indicated otherwiseIMF International Financial Statistics</u>\n\n\n\nNote: All variables are at quarterly frequency, unless indicated otherwise\n\n\n\nobtain a complementary fund inflow measure. The main explanatory and control variables were\nobtained from IFS, World Development Indicators (WDI), and central banks, supplemented with\nDatastream and Haver where gaps exist (see specific sources in Table A2. 2). Both capital flows and\nexplanatory variables in the model are measured in real terms, in constant 2010 exchange rates and\nprices.\n\n\n_Model_\n\n\nThe main dependent variable of interest, financial inflows (GFIit), and its component parts\n(portfolio investment flows, foreign direct investment, and cross-border bank lending) are each\nmodeled as a function of variables meant to proxy for various factors associated with the movement\nof cross-border flows. The model with both global and local determinants of capital flows is\nconsistent with the recent policy and academic literature (see, for instance, Ahmed and Zlate 2013;\nFratscher 2011; Bruno and Shin 2013; Forbes and Warnock 2012). This approach is also consistent\nwith an earlier literature on capital flows (Chuhan, Claessens and Mamingi 1998; Sarno and Taylor\n1997; Calvo Leiderman and Reinhart 1996\n\nMention: \"IMF International Financial Statistics\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Financial and monetary series in the variable table come from this source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005949:36:2", "occurrence_id": "7965a7d5a5a24bc44e1de95b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: , IMF International Financial Statistics, World Development Indicators</u>\n\n<u>External debt/GDP</u> <u>World Development Indicators, Datastream, BIS</u>\n\n\n\n<u>Trade/GDP</u>\n\n<u>External debt/GDP</u> <u>World Development Indicators, Datastream, BIS</u>\n\n<u>Private sector credit/GDP</u> <u>IMF International Financial Statistics</u>\n\n\n\n<u>External debt/GDP</u> <u>World Development Indicators, Datastream, BIS</u>\n\n<u>Private sector credit/GDPNote: All variables are at quarterly frequency, unless indicated otherwiseIMF International Financial Statistics</u>\n\n\n\nNote: All variables are at quarterly frequency, unless indicated otherwise\n\n\n\nobtain a complementary fund inflow measure. The main explanatory and control variables were\nobtained from IFS, World Development Indicators (WDI), and central banks, supplemented with\nDatastream and Haver where gaps exist (see specific sources in Table A2. 2). Both capital flows and\nexplanatory variables in the model are measured in real terms, in constant 2010 exchange rates and\nprices.\n\n\n_Model_\n\n\nThe main dependent variable of interest, financial inflows (GFIit), and its component parts\n(portfolio investment flows, foreign direct investment, and cross-border bank lending) are each\nmodeled as a function of variables meant to proxy for various factors associated with the movement\nof cross-border flows. The model with both global and local determinants of capital flows is\nconsistent with the recent policy and academic literature (see, for instance, Ahmed and Zlate 2013;\nFratscher 2011; Bruno and Shin 2013; Forbes and Warnock 2012). This approach is also consistent\nwith an earlier literature on capital flows (Chuhan, Claessens and Mamingi 1998; Sarno and Taylor\n1997; Calvo Leiderman and Reinhart 1996\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Output, trade and external debt ratios come from this compilation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005949:36:2", "occurrence_id": "24d6ac76ebf2e51160c57933", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: , IMF International Financial Statistics, World Development Indicators</u>\n\n<u>External debt/GDP</u> <u>World Development Indicators, Datastream, BIS</u>\n\n\n\n<u>Trade/GDP</u>\n\n<u>External debt/GDP</u> <u>World Development Indicators, Datastream, BIS</u>\n\n<u>Private sector credit/GDP</u> <u>IMF International Financial Statistics</u>\n\n\n\n<u>External debt/GDP</u> <u>World Development Indicators, Datastream, BIS</u>\n\n<u>Private sector credit/GDPNote: All variables are at quarterly frequency, unless indicated otherwiseIMF International Financial Statistics</u>\n\n\n\nNote: All variables are at quarterly frequency, unless indicated otherwise\n\n\n\nobtain a complementary fund inflow measure. The main explanatory and control variables were\nobtained from IFS, World Development Indicators (WDI), and central banks, supplemented with\nDatastream and Haver where gaps exist (see specific sources in Table A2. 2). Both capital flows and\nexplanatory variables in the model are measured in real terms, in constant 2010 exchange rates and\nprices.\n\n\n_Model_\n\n\nThe main dependent variable of interest, financial inflows (GFIit), and its component parts\n(portfolio investment flows, foreign direct investment, and cross-border bank lending) are each\nmodeled as a function of variables meant to proxy for various factors associated with the movement\nof cross-border flows. The model with both global and local determinants of capital flows is\nconsistent with the recent policy and academic literature (see, for instance, Ahmed and Zlate 2013;\nFratscher 2011; Bruno and Shin 2013; Forbes and Warnock 2012). This approach is also consistent\nwith an earlier literature on capital flows (Chuhan, Claessens and Mamingi 1998; Sarno and Taylor\n1997; Calvo Leiderman and Reinhart 1996\n\nMention: \"Datastream\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"A commercial companion source for the same ratios.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005951:15:1", "occurrence_id": "e07b9afddb4803f3700959a7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  3.2.\nAdditional details regarding data construction and basic data sources can be found in Timmer (ed.,\n2012).\n\n\n**3.1 World input-output tables**\n\n\nThe WIOD provides a time-series of world input-output tables (WIOTs) from 1995 onwards. It covers\n40 countries, including all EU 27 countries and 13 other major advanced and emerging economies\nnamely Australia; Brazil; Canada; China; India; Indonesia; Japan; Mexico; Russia; the Republic of\nKorea; Taiwan, China; Turkey; and the United States. In total it covers more than 85 percent of world\nGDP in 2008. In addition a model for the remaining non-covered part of the world economy is made\nsuch that the decomposition of final output as given in equation (3) is complete.\n\n\nThe WIOTs have been constructed on the basis of national Supply and Use Tables (SUTs) which\nprovide information on the intra-industry flows within a country. A Supply table indicates for each\nproduct its source (domestic industries and imports), while the Use table indicates for each product its\ndestination (intermediate use by domestic industries, domestic final demand or exports). National SUTs\nhave dimensions of 35 industries and 59 product groups. The 35 industries cover the overall economy\nand are mostly at the 2-digit NACE rev. 1 level or groups there from. They include agriculture, mining,\nconstruction, utilities, fourteen manufacturing industries, eight trade and transport services, telecom,\nfinance, business services, personal services, and three public services. The product groups are more\nfinely defined and are all two-digits in the 2002 Classification of Products by Activity (CPA), including\ntwenty-three manufacturing products. SUTs provide a more natural starting point than input-output\n\n\n14\n\nMention: \"World input-output tables\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"A time series across forty countries beginning in this year.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005951:15:1", "occurrence_id": "19130df4b27b092c4013c26a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  3.2.\nAdditional details regarding data construction and basic data sources can be found in Timmer (ed.,\n2012).\n\n\n**3.1 World input-output tables**\n\n\nThe WIOD provides a time-series of world input-output tables (WIOTs) from 1995 onwards. It covers\n40 countries, including all EU 27 countries and 13 other major advanced and emerging economies\nnamely Australia; Brazil; Canada; China; India; Indonesia; Japan; Mexico; Russia; the Republic of\nKorea; Taiwan, China; Turkey; and the United States. In total it covers more than 85 percent of world\nGDP in 2008. In addition a model for the remaining non-covered part of the world economy is made\nsuch that the decomposition of final output as given in equation (3) is complete.\n\n\nThe WIOTs have been constructed on the basis of national Supply and Use Tables (SUTs) which\nprovide information on the intra-industry flows within a country. A Supply table indicates for each\nproduct its source (domestic industries and imports), while the Use table indicates for each product its\ndestination (intermediate use by domestic industries, domestic final demand or exports). National SUTs\nhave dimensions of 35 industries and 59 product groups. The 35 industries cover the overall economy\nand are mostly at the 2-digit NACE rev. 1 level or groups there from. They include agriculture, mining,\nconstruction, utilities, fourteen manufacturing industries, eight trade and transport services, telecom,\nfinance, business services, personal services, and three public services. The product groups are more\nfinely defined and are all two-digits in the 2002 Classification of Products by Activity (CPA), including\ntwenty-three manufacturing products. SUTs provide a more natural starting point than input-output\n\n\n14\n\nMention: \"world input-output tables\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Covering 40 countries including all member states of one union.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005952:18:0", "occurrence_id": "68d999a2b0bb1492c2d35026", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: poor could not benefit as much from the increase in formal jobs (structural changes are discussed further\nin section 3.2) because of entry barriers such as good education, networks and so forth. <sup>8</sup> [^8: This hypothesis is supported by the findings from Grun et al. (2013) showing that in the post-crisis period most of the jobs were\nformal jobs requiring university education.] Besides wages,\npensions were the second most important equalizing factor during the considered period. The share of\nadults in the household and social assistance also contributed to the reduction of the Gini coefficient.\n\n**Figure 3.4a: Sources of welfare improvement** **Figure 3.4b: Sources of welfare improvement**\n**across classes in 2002-2011, %** **for the bottom 40 percent of the population in**\n\n**<u>2002-2011, %</u>**\n\n\nSource: Authors’ calculation based on ECAPOV data.\nNotes: Welfare aggregate is income per capita. Negative share means negative contribution to welfare. Thresholds for poor,\nvulnerable and middle class are 5 and 10 USD/PPP 2005 per day. Methodological details are presented in Azevedo and Nguyen\n<u>(2014).</u>\n\n26. **The contribution of women to poverty and inequality reduction was very limited.** Income\npoverty decomposition across gender demonstrates that female contribution to poverty reduction through\nthe labor market, share of adults and pensions was much lower than that of males (figure 3.5). Moreover,\nboth female earnings and added worker effects were inequality increasing (figure 3.6). This may be\nrelated to the fact that women with better education benefit from entering the labor market and earning\nhigher wages. One possible explanation of lower contribution of the share of female adults to poverty\nreduction is that women tend to get married earlier than men and leave the household. In addition, there\nwas a tendency for men to\n\nMention: \"ECAPOV data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Welfare contribution shares over this decade are calculated from these data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005954:5:1", "occurrence_id": "bf684ef9a1e27a6fbdd92aac", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  among those near the center of the political\n\nspectrum as they are less attached to extreme political ideologies. In a similar study, Pop-Eleches\n\nand Pop-Eleches (2009) analyze the case of a Romanian program that awarded low income\n\nfamilies with school age children vouchers for purchasing new personal computers and find that\n\nit had a significant impact on political attitudes and electoral behavior. In particular, voucher\n\nrecipients were more likely to report vote intention in upcoming elections, and governing parties\n\nreaped most of the benefits of increased participation. They also find some evidence of vote\n\nswitching from the main opposition party to the current incumbents and this effect was\n\nsubstantially stronger in towns where the governing parties controlled the local government.\n\n\nThis paper is structured as follows: Section 2 briefly describes the electoral context in Mexico\n\nand the Weather Index Insurance program. Section 3 presents the data. Section 4 discusses the\n\nstatistical methodology and presents the main results. Finally, section 5 discusses the\n\nimplications of the results found from the perspective of the study of political behavior and voter\n\nresponsiveness to relief aid after a natural disaster.\n\n\nPage **4** of **29**\n\nMention: \"the Weather Index Insurance program\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The programme's design is described before the data are presented.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005956:26:1", "occurrence_id": "8b664b62c54e1c1f43e2e2a0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  in the Reinhart & Rogoff\n\n\n(2009) dataset that are represented in our aggregate crisis experience variable (as described in\n\n\nSubsection 4.1). The most striking result is that every single crisis type has a negative coefficient—\n\n\nin line with our aggregate—but only experiences in currency and domestic debt crises enter with a\n\n\nsignificant coefficient. We also note that experience in banking crises enters with a relatively high\n\n\ncoefficient, and is statistically insignificant mainly because it is accompanied by correspondingly\n\n\nlarge standard errors. This latter result may be understood by recognizing the fact that banking\n\n\ncrises are generally rather heterogeneous in terms of both context and causes. As a result, banks\n\n\nmay acquire mixed experiences as a result of past banking crises, and accordingly respond to them\n\n\nin ways that go beyond contractions in their loan portfolio.\n\n\nThe relative importance of currency crises and domestic debt crises in terms of crisis experience\n\n\nsuggests that these sorts of crises offer important learning opportunities, insofar as lending activity is\n\n\nconcerned. Currency and domestic debt crises often impose real costs (Caballero & Krishnamurthy\n\n\n2001), which may or may not always occur with stock market and banking crises. However, there\n\n\nis also the possibility that these crises are especially important because they directly impact bank\n\n\nbalance sheets (as opposed to stock market crashes or external debt crises, which would have either\n\n\na small or indirect impact on bank balance sheets, or banking crises, which affect bank operations\n\n\nacross their entire spectrum of operations, not just their lending).\n\n\nAnother approach to decomposing the crisis experience variable is to capture the independent\n\n\ncomponents embedded in the crisis experience measure after stripping out correlated elements\n\n\nwithin each separate type of crisis experience. We perform this decomposition with principal\n\n\ncomponents analysis (PCA). More specifically, we calculate the proportion-weighted average of the\n\n\n25\n\nMention: \"the Reinhart & Rogoff\n\n\n(2009) dataset\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its crisis episodes feed the aggregate crisis experience variable.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005961:6:0", "occurrence_id": "86b9bfa92b90fa3584e6fed1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: main regression results are provided in Section 3. In section 4, we discuss some extension of the\n\n\nmain results. The concluding section summarizes the main findings of the paper.\n\n\n**2. Data and Main Variables**\n\n\nThe main data source we use consists of firm-level surveys for 87 developing countries\n\n\nconducted by the World Bank’s Enterprise Surveys between 2007 and 2009. Table 1 contains the\n\n\nlist of countries in our sample along with the sample size (number of firms) for each survey.\n\n\nThese surveys were conducted in some cases across the whole region (such as Latin America and\n\n\nEastern Europe and Central Asia) and sometimes in individual countries. However, a common\n\n\nsampling methodology – stratified random sampling – was followed in all the surveys along with\n\n\na common questionnaire. The sample for each country was stratified by firm-size, sector of\n\n\nactivity and location within the country. Weights are provided in the survey to ensure that the\n\n\nsample is representative of the non-agricultural private sector of the economy. We note that\n\n\nEnterprise Surveys cover manufacturing as well as services sectors but certain services such as\n\n\neducation and health are not covered. Also, the primary sector, which encompasses agriculture,\n\n\nmining, forestry, etc., is also excluded from the survey. Finally, the sample we use is a pure cross\n\n\nsection of firms with one observation per firm. We also use the latest round of the survey\n\n\nconducted in each country. Enterprise Surveys does provide panel data with more than one\n\n\nobservation per firm but currently, these data are limited in size and are available for only a\n\n\nhandful of countries.\n\n\nThe regression results discussed below are obtained from logit estimation with Huber\n\nWhite robust standard errors and clustered on the country. Marginal effects from the logit\n\n\nestimation discussed below are calculated at the mean value of the all the explanatory variables.\n\n\n5\n\nMention: \"Enterprise Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Firm-level surveys across 87 developing countries over these years.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005964:7:0", "occurrence_id": "8d6f5325a930cc46fa52b371", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Regardless, the increase in private investment and non-state-owned firms provided the basis for\n\nthe deepening of the enterprise reforms.\n\n\nWith the development of the market economy, the volatility of firm level productivity becomes a\n\ncrucial aspect of the micro performance of an economy. It reflects responses of firms to\n\nidiosyncratic shocks and frictions in product, factor and credit markets. Firm level volatility\n\noften links to their investment patterns, their access to external finance, as well as the regulation\n\nreforms. The volatility of the entire private sector reflects the joint forces of firm entry and exit\n\nand the volatility of the individual firms in the market.\n\n\n**3 Evolution of productivity volatility**\n\n\nFirm level productivity volatility is an important aspect of development of the whole economy. In\n\nprevious studies, volatility is measured in different ways. In Loayza and Servén (2010),\n\nmacroeconomic volatility is represented by the standard deviation of the output gap, obtained as\n\nthe difference between the actual and trend real GDP per capita. Trend output is estimated using\n\nthe band-pass filter of Baxter and King (1999). In Hausmann and Gavin (1996), macroeconomic\n\nvolatility is defined as the standard deviation of the level of GDP per capita. In Breen and García\nPeñalosa (2005), output volatility is defined as the standard deviation of the annual growth rate of\n\nreal per capita GDP.\n\n\nIn this study, we follow the methodology developed in Comin and Philippon (2005). They used\n\naggregate data from the National Income and Product Accounts (NIPAs), and firm level data\n\nfrom COMPUSTAT (a database of financial, statistical and market information on active and\n\ninactive global companies throughout the world) and the Center for Research in Security Prices\n\n(CRSP), which is composed of time series that vary from 1955 to 2000. For each firm _i_ at time _t_,\n\nthey computed the volatility as the standard deviation of the annual growth rate of sales during\n\neach ten\n\nMention: \"National Income and Product Accounts\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Aggregate national accounts behind the earlier methodology.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005964:7:0", "occurrence_id": "7a46daa869eb633f99cdf8a2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Regardless, the increase in private investment and non-state-owned firms provided the basis for\n\nthe deepening of the enterprise reforms.\n\n\nWith the development of the market economy, the volatility of firm level productivity becomes a\n\ncrucial aspect of the micro performance of an economy. It reflects responses of firms to\n\nidiosyncratic shocks and frictions in product, factor and credit markets. Firm level volatility\n\noften links to their investment patterns, their access to external finance, as well as the regulation\n\nreforms. The volatility of the entire private sector reflects the joint forces of firm entry and exit\n\nand the volatility of the individual firms in the market.\n\n\n**3 Evolution of productivity volatility**\n\n\nFirm level productivity volatility is an important aspect of development of the whole economy. In\n\nprevious studies, volatility is measured in different ways. In Loayza and Servén (2010),\n\nmacroeconomic volatility is represented by the standard deviation of the output gap, obtained as\n\nthe difference between the actual and trend real GDP per capita. Trend output is estimated using\n\nthe band-pass filter of Baxter and King (1999). In Hausmann and Gavin (1996), macroeconomic\n\nvolatility is defined as the standard deviation of the level of GDP per capita. In Breen and García\nPeñalosa (2005), output volatility is defined as the standard deviation of the annual growth rate of\n\nreal per capita GDP.\n\n\nIn this study, we follow the methodology developed in Comin and Philippon (2005). They used\n\naggregate data from the National Income and Product Accounts (NIPAs), and firm level data\n\nfrom COMPUSTAT (a database of financial, statistical and market information on active and\n\ninactive global companies throughout the world) and the Center for Research in Security Prices\n\n(CRSP), which is composed of time series that vary from 1955 to 2000. For each firm _i_ at time _t_,\n\nthey computed the volatility as the standard deviation of the annual growth rate of sales during\n\neach ten\n\nMention: \"COMPUSTAT\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"A database of financial and market information on active and inactive firms.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005964:7:0", "occurrence_id": "b85142b2d58dd048115892f2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Regardless, the increase in private investment and non-state-owned firms provided the basis for\n\nthe deepening of the enterprise reforms.\n\n\nWith the development of the market economy, the volatility of firm level productivity becomes a\n\ncrucial aspect of the micro performance of an economy. It reflects responses of firms to\n\nidiosyncratic shocks and frictions in product, factor and credit markets. Firm level volatility\n\noften links to their investment patterns, their access to external finance, as well as the regulation\n\nreforms. The volatility of the entire private sector reflects the joint forces of firm entry and exit\n\nand the volatility of the individual firms in the market.\n\n\n**3 Evolution of productivity volatility**\n\n\nFirm level productivity volatility is an important aspect of development of the whole economy. In\n\nprevious studies, volatility is measured in different ways. In Loayza and Servén (2010),\n\nmacroeconomic volatility is represented by the standard deviation of the output gap, obtained as\n\nthe difference between the actual and trend real GDP per capita. Trend output is estimated using\n\nthe band-pass filter of Baxter and King (1999). In Hausmann and Gavin (1996), macroeconomic\n\nvolatility is defined as the standard deviation of the level of GDP per capita. In Breen and García\nPeñalosa (2005), output volatility is defined as the standard deviation of the annual growth rate of\n\nreal per capita GDP.\n\n\nIn this study, we follow the methodology developed in Comin and Philippon (2005). They used\n\naggregate data from the National Income and Product Accounts (NIPAs), and firm level data\n\nfrom COMPUSTAT (a database of financial, statistical and market information on active and\n\ninactive global companies throughout the world) and the Center for Research in Security Prices\n\n(CRSP), which is composed of time series that vary from 1955 to 2000. For each firm _i_ at time _t_,\n\nthey computed the volatility as the standard deviation of the annual growth rate of sales during\n\neach ten\n\nMention: \"Center for Research in Security Prices\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Long time series of security returns behind the earlier study.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005966:6:0", "occurrence_id": "35df6226b0ea5c34a7d74831", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: our estimation strategy and results are presented in section 3.4. We conclude with a\ndiscussion of our findings in section 3.6.\n\n\n**2 Data**\n\nIn this section, we examine youth employment outcomes using the most recent year of\navailable data, before documenting trends between 1990 and 2010. We then describe the\nevolution of cohort size for the same time period.\n\n\n2.1 Labor Market Data\nThe labor market outcome data comes from the International Income Distribution Survey\n(I2D2). The I2D2 is a harmonized collection of household and labor surveys from 19902010 constructed and maintained by the research group of the World Bank. <sup>9</sup> [^9: An earlier version of the data set is described in detail in Montenegro and Hirn (2009).] Our data set\nincludes repeated cross-section data from 450 surveys at the individual level in 83 lower,\nmiddle and upper middle income countries. The available years and frequency of waves\ndiffer among countries as shown in Table 5 in the Appendix.\n\n\nWhen combining different micro surveys in one data set, it is important to ensure that\nquestions are asked consistently over time. Although the I2D2 has been harmonized,\nthere can be large changes in variables when the underlying survey changes, due to\nchanges in survey design driven by differences in the formulation of questions or the\ntarget group. This will introduce measurement error into our measures of labor market\noutcomes, which could potentially be correlated with changes in cohort size and bias the\nestimates. We therefore identify potential structural breaks, using the outlier detection\ntool “bacon” in Stata. We first identified annualized changes in labor market outcomes\namounting to more than four percentage points, which we considered to be outliers. We\nthen used bacon on the full set of observations to identify outliers conditional on changes\nin GDP. These outliers were excluded from the main analysis. We include all outliers as a\nrobustness check and find little difference to our main results.\n\n\nWe use four labor market outcome variables: (1)\n\nMention: \"International Income Distribution Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"A harmonised collection of household and labour surveys over these years.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005966:6:0", "occurrence_id": "e68bf9f234b7001917c5cc7e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: our estimation strategy and results are presented in section 3.4. We conclude with a\ndiscussion of our findings in section 3.6.\n\n\n**2 Data**\n\nIn this section, we examine youth employment outcomes using the most recent year of\navailable data, before documenting trends between 1990 and 2010. We then describe the\nevolution of cohort size for the same time period.\n\n\n2.1 Labor Market Data\nThe labor market outcome data comes from the International Income Distribution Survey\n(I2D2). The I2D2 is a harmonized collection of household and labor surveys from 19902010 constructed and maintained by the research group of the World Bank. <sup>9</sup> [^9: An earlier version of the data set is described in detail in Montenegro and Hirn (2009).] Our data set\nincludes repeated cross-section data from 450 surveys at the individual level in 83 lower,\nmiddle and upper middle income countries. The available years and frequency of waves\ndiffer among countries as shown in Table 5 in the Appendix.\n\n\nWhen combining different micro surveys in one data set, it is important to ensure that\nquestions are asked consistently over time. Although the I2D2 has been harmonized,\nthere can be large changes in variables when the underlying survey changes, due to\nchanges in survey design driven by differences in the formulation of questions or the\ntarget group. This will introduce measurement error into our measures of labor market\noutcomes, which could potentially be correlated with changes in cohort size and bias the\nestimates. We therefore identify potential structural breaks, using the outlier detection\ntool “bacon” in Stata. We first identified annualized changes in labor market outcomes\namounting to more than four percentage points, which we considered to be outliers. We\nthen used bacon on the full set of observations to identify outliers conditional on changes\nin GDP. These outliers were excluded from the main analysis. We include all outliers as a\nrobustness check and find little difference to our main results.\n\n\nWe use four labor market outcome variables: (1)\n\nMention: \"I2D2\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The same harmonised collection under its short name.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005970:42:2", "occurrence_id": "645345f6777e294c86ad92d6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 06<br>|\n\n\n\n_Figure B1: Roofer wages in an area where losses have been significant after the 2004 hurricane season in_\n_Florida. Data from the Bureau of Labor Statistics, Occupational Employment Surveys in May 03, Nov 03,_\n_May 04, Nov 04, May 05, May 06, May 07._\n\nFigure B2 is a classical quantity-price plot, showing the long-term demand and supply curves for a goods\nor service aggregated at the macroeconomic level. The green line is the demand curve: it shows how the\n\n33 The only case where this result does not hold is when the production function is in the form: _Y=f(K1+K2+…+Kn)._\n\n\n41\n\nMention: \"Bureau of Labor Statistics, Occupational Employment Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Employment across these survey dates traces hurricane employment losses.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005974:11:0", "occurrence_id": "0627778f3ec18feef0ec5cbc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _migrants’ transfers_, which are a capital account transfer reflecting the movement of assets by a\n\nmigrant from one country to another when he or she migrates (IMF 1993). The World Bank\n\ncombines workers’ remittances and compensation of employees together to form the remittance\n\nmeasure reported in its World Development Indicators, and this has been the definition of\n\nremittances used in a number of studies in the literature (e.g. Catrinescu et al. 2009). <sup>3</sup> [^3: There has been some debate over this definition with Barajas et al. (2009) just using workers’ remittances, and Guiliano and\nRuiz-Arranz (2009) using the sum of all three components, but then making adjustments on a country-by-country basis.]\n\n\nHowever, in practice it often proved difficult to separate transfers made by migrant workers from\n\ntheir employment income from a number of other transfers. The Balance of Payments Manual 6\n\n(BPM6) therefore replaced the category of _workers’ remittances_ with _personal transfers_ which\n\nconsist of “all current transfers in cash or in kind made or received by resident households to or\n\nfrom nonresident households” (IMF 2009, p. 20). Personal remittances in the World\n\nDevelopment Indicators (and Figure 1) are then the sum of these personal transfers and\n\ncompensation of employees.\n\n\nThe underlying data that feeds into these aggregates comes from central banks in each individual\n\ncountry. They typically require banks, money transfer operators and other institutions to provide\n\nreports on the transactions they process (Orozco 2006). However, in many cases the coverage of\n\nthis reporting has been partial. De Luna Martínez (2005) reports on a survey of central banks in\n\n40 developing countries that took place in 2004. He finds that 90 percent of the countries\n\ncollected remittance data from commercial banks, but such data were collected in only 65\n\npercent of the countries in which credit unions and exchange houses processed remittances, in\n\nonly 38 percent of the countries in which money transfer operators processed remittances, and in\n\nonly 35 percent\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its combined transfers and compensation measure defines the remittance series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005975:18:0", "occurrence_id": "c4c3a4ad995ad573b4cc2cb2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ##### **3. Price Rising Effects**\n\n###### **a. Price Data: The Trade Unit Value Database**\n\nTrade unit values from the French Research Center in International Economics (CEPII) are used to\napproximate trade prices. <sup>3</sup> [^3: This database provides a world-level dataset reporting reliable unit values at a high level of disaggregation. One\nshortcoming is important to note: despite the treatment of missing unit values in the database, there is still a\nrelatively large share of missing data. Berthou and Emlinger (2011) estimate that the database covered around 80%\nof world total trade values in 2008.] These data provide reliable and comparable unit values across countries.\nFollowing Cadot and Gourdon (2012), we use Cost of Insurance and Freight (CIF) trade unit values\n(reported by the importer country) to estimate any potential price-raising effects of NTMs. The CIF price\nwas deemed most appropriate for the analysis as it includes the Freight on Board (FOB) price of the\nproduct along with the cost of compliance with SPS and TBT regulations, and any other expenses from\npre-shipment inspection, licensing, or any other sort of NTMs.\n\n###### **b. Methodology**\n\n\nThe estimation below can be thought of as a simple treatment-effect approach where prices are “treated”\nby NTMs. We assume that prices are determined by Equation 3. Here, i indexes countries, k indexes\nproducts, p∗k is the world price of product k, pik is its price in country i, tik the tariff imposed by country i\n\na number of other characteristics such as landlocked status, remoteness, the quality of the infrastructure,\non product k, and λi a cost-of-living adjustment factor for country i depending on its level of income and\n\n\n\nThe estimation below can be thought of as a simple treatment-effect approach where prices are “treated”\nby NTMs. We assume that prices are determined by Equation\n\nMention: \"Trade Unit Value Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Product-level price series behind the price effects analysis come from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005982:29:0", "occurrence_id": "e75212d88ee1191426b79097", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **5. Conclusions**\n\n\n\nThis paper has analyzed the relation between energy consumption and income levels,\n\nand other characteristics, in Mexican households. To the extent that subsidies are\n\n\n\nThis paper has analyzed the relation between energy consumption and income levels,\n\nand other characteristics, in Mexican households. To the extent that subsidies are\n\ncurrently applied to energy sources for general consumption, or that a social tariff has\n\n\n\nand other characteristics, in Mexican households. To the extent that subsidies are\n\ncurrently applied to energy sources for general consumption, or that a social tariff has\n\nto be designed, this issue becomes relevant for better targeting of public mechanisms.\n\n\n\ncurrently applied to energy sources for general consumption, or that a social tariff has\n\nto be designed, this issue becomes relevant for better targeting of public mechanisms.\n\nPrevious studies linking energy consumption and income have only focused on the\n\n\n\nto be designed, this issue becomes relevant for better targeting of public mechanisms.\n\nPrevious studies linking energy consumption and income have only focused on the\n\nextreme poor who receive conditional cash transfer schemes (i.e. Gertler et al., 2009).\n\n\n\nPrevious studies linking energy consumption and income have only focused on the\n\nextreme poor who receive conditional cash transfer schemes (i.e. Gertler et al., 2009).\n\n\n\nWe used microdata at the household level from the Income-Expenditure Surveys\n\n(ENIGH) 2010 for Mexico. The surveys record energy expenditure, income, and socio\n\n\nWe used microdata at the household level from the Income-Expenditure Surveys\n\n(ENIGH) 2010 for Mexico. The surveys record energy expenditure, income, and socio\neconomic characteristics, among other things. We first calculated energy consumption\n\n\n\n(ENIGH) 2010 for Mexico. The surveys record energy expenditure, income, and socio\neconomic characteristics, among other things. We first calculated energy consumption\n\nstandardized in KgOE, so that all results can be directly compared to each other. We\n\n\n\neconomic characteristics, among other things. We\n\nMention: \"microdata at the household level\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Household energy expenditure and income records are drawn at this level.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005982:29:0", "occurrence_id": "bcb471f6127eac522ec70ce5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **5. Conclusions**\n\n\n\nThis paper has analyzed the relation between energy consumption and income levels,\n\nand other characteristics, in Mexican households. To the extent that subsidies are\n\n\n\nThis paper has analyzed the relation between energy consumption and income levels,\n\nand other characteristics, in Mexican households. To the extent that subsidies are\n\ncurrently applied to energy sources for general consumption, or that a social tariff has\n\n\n\nand other characteristics, in Mexican households. To the extent that subsidies are\n\ncurrently applied to energy sources for general consumption, or that a social tariff has\n\nto be designed, this issue becomes relevant for better targeting of public mechanisms.\n\n\n\ncurrently applied to energy sources for general consumption, or that a social tariff has\n\nto be designed, this issue becomes relevant for better targeting of public mechanisms.\n\nPrevious studies linking energy consumption and income have only focused on the\n\n\n\nto be designed, this issue becomes relevant for better targeting of public mechanisms.\n\nPrevious studies linking energy consumption and income have only focused on the\n\nextreme poor who receive conditional cash transfer schemes (i.e. Gertler et al., 2009).\n\n\n\nPrevious studies linking energy consumption and income have only focused on the\n\nextreme poor who receive conditional cash transfer schemes (i.e. Gertler et al., 2009).\n\n\n\nWe used microdata at the household level from the Income-Expenditure Surveys\n\n(ENIGH) 2010 for Mexico. The surveys record energy expenditure, income, and socio\n\n\nWe used microdata at the household level from the Income-Expenditure Surveys\n\n(ENIGH) 2010 for Mexico. The surveys record energy expenditure, income, and socio\neconomic characteristics, among other things. We first calculated energy consumption\n\n\n\n(ENIGH) 2010 for Mexico. The surveys record energy expenditure, income, and socio\neconomic characteristics, among other things. We first calculated energy consumption\n\nstandardized in KgOE, so that all results can be directly compared to each other. We\n\n\n\neconomic characteristics, among other things. We\n\nMention: \"Income-Expenditure Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"They record energy expenditure, income and household characteristics.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005984:8:0", "occurrence_id": "627ba00001d45549df4fde72", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **2.** **Data**\n\nOur database consists of: (i) country weights or weights,, which are the country\n\nportfolio allocations of international mutual funds (those investing in several countries);\n\n(ii) benchmark weights,, which are the country allocations in the relevant\n\nbenchmarks; (iii) mutual fund-specific information, such as its assets, returns, and\n\nrelevant benchmarks; (iv) country-specific information, such as stock and bond market\n\nindex returns. <sup>6</sup> [^6: Benchmark weights are fund-specific because each fund chooses its benchmark. We thus denote it\nwith sub-index _i._ The same applies to other benchmark characteristics such as benchmark returns.] Throughout the paper, the sub-index _i_ refers to funds, _c_ to countries, and\n\n\nthe supra-index _B_ to benchmarks. For the final database, we cleaned the raw data and\n\n\nmerged data from several sources, some of which had not been previously used or\n\nmatched in the literature. The final data cover the period from January 1996 to July 2012\n\n\nand constitute an unbalanced panel.\n\n\nOur two main sources for country portfolio allocations of international mutual\n\n\nfunds are EPFR (Emerging Portfolio Fund Research) and Morningstar Direct (MS).\n\nBoth sources include dead and live mutual funds. The data from EPFR are at a monthly\n\n\nfrequency, and include open-end equity and bond funds classified according to their\n\n\ngeographical investment scope. Global funds invest anywhere in the world, global\n\n\nemerging funds only in emerging countries, and regional funds in groups of countries\n\nwithin a specific geographical region (e.g., developed Asia). <sup>7</sup> [^7: While global funds theoretically can invest anywhere in the world, a large proportion of them track the\nMSCI World Index, which only has developed countries as constituents. A minor proportion of these\nfunds gauge their performance relative to the MSCI All Country World Index that contains both\ndeveloped and emerging countries.] Frontier market funds are\n\n\nusually classified as regional\n\nMention: \"data from EPFR\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Monthly open-end equity and bond fund records come from this source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005984:8:0", "occurrence_id": "74d966ee50a3b1f9c1cdf94b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **2.** **Data**\n\nOur database consists of: (i) country weights or weights,, which are the country\n\nportfolio allocations of international mutual funds (those investing in several countries);\n\n(ii) benchmark weights,, which are the country allocations in the relevant\n\nbenchmarks; (iii) mutual fund-specific information, such as its assets, returns, and\n\nrelevant benchmarks; (iv) country-specific information, such as stock and bond market\n\nindex returns. <sup>6</sup> [^6: Benchmark weights are fund-specific because each fund chooses its benchmark. We thus denote it\nwith sub-index _i._ The same applies to other benchmark characteristics such as benchmark returns.] Throughout the paper, the sub-index _i_ refers to funds, _c_ to countries, and\n\n\nthe supra-index _B_ to benchmarks. For the final database, we cleaned the raw data and\n\n\nmerged data from several sources, some of which had not been previously used or\n\nmatched in the literature. The final data cover the period from January 1996 to July 2012\n\n\nand constitute an unbalanced panel.\n\n\nOur two main sources for country portfolio allocations of international mutual\n\n\nfunds are EPFR (Emerging Portfolio Fund Research) and Morningstar Direct (MS).\n\nBoth sources include dead and live mutual funds. The data from EPFR are at a monthly\n\n\nfrequency, and include open-end equity and bond funds classified according to their\n\n\ngeographical investment scope. Global funds invest anywhere in the world, global\n\n\nemerging funds only in emerging countries, and regional funds in groups of countries\n\nwithin a specific geographical region (e.g., developed Asia). <sup>7</sup> [^7: While global funds theoretically can invest anywhere in the world, a large proportion of them track the\nMSCI World Index, which only has developed countries as constituents. A minor proportion of these\nfunds gauge their performance relative to the MSCI All Country World Index that contains both\ndeveloped and emerging countries.] Frontier market funds are\n\n\nusually classified as regional\n\nMention: \"MSCI All Country World Index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Some funds gauge performance against this developed and emerging index.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005987:44:2", "occurrence_id": "dd4ee9f3af27e1b5d89bb0cc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  0 6 6572\nland (at hh level) 0.334 0.472 0 1 6559\nwoman working in agric 0.047 0.211 0 1 6446\n<u>husband</u> <u>working</u> <u>in</u> <u>agric</u> <u>0.095</u> <u>0.294</u> <u>0</u> <u>1</u> <u>6446</u>\n\n\n_Notes._ 2008 NDHS. Using sample weights. Statistics are at the household level. Excludes households in which there is no woman eligible for the\nwoman questionnaire (age 15–49) for which the number and sex of children is not available.\n\n\n43\n\nMention: \"NDHS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Household-level statistics with sample weights come from this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005998:17:1", "occurrence_id": "1668fa87261cb56da652142a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: br>6.77|\n\n\n\n_Source: Authors calculations based on medical records_\n\n\nThe sample sizes for tetanus toxoid vaccine are smaller than for the other\nindicators. This is primarily because the vaccine is prescribed only to a subset of\npregnant women. <sup>11</sup> Based on these criteria 44 percent of women in the sample\nrequired a tetanus vaccine.\n\n\nThe sample size for neonatal mortality is also lower. We exclude one\nprovince and a number of hospitals that coded neonatal mortality aggregated with\nother outcomes such as abortion. In these cases we could not separate neonatal\nmortality from abortion. Overall, we have data from 40 hospitals representing\naround 45 percent of the births. While this should not affect the internal validity of\nour estimates of the causal impact, it could affect the external validity. However, the\nneonatal mortality rate from our sample (8.3 per thousand) is close to the neonatal\nmortality rate based on death certificate data (10.1 per thousand). We would expect\nthe death rate from death certificates to be slightly higher than the in-hospital\nmortality rate as the former includes deaths that did not occur in hospitals.\n\n\n11 Women requiring a tetanus vaccine include those who had not previously been vaccinated or had\nreceived only one dose, those whose previous immunization status is not known, those who have had\ntwo doses but whose last vaccination is more than three years ago, and those that have had three\ndoses but the last dose or booster was more than five years ago (the maximum length recommended\nby clinical guidelines).\n\n\n16\n\nMention: \"medical records\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Vaccine indicator calculations rest on these clinical records.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005998:17:1", "occurrence_id": "4a48bce9574dbb4aad9a644a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: br>6.77|\n\n\n\n_Source: Authors calculations based on medical records_\n\n\nThe sample sizes for tetanus toxoid vaccine are smaller than for the other\nindicators. This is primarily because the vaccine is prescribed only to a subset of\npregnant women. <sup>11</sup> Based on these criteria 44 percent of women in the sample\nrequired a tetanus vaccine.\n\n\nThe sample size for neonatal mortality is also lower. We exclude one\nprovince and a number of hospitals that coded neonatal mortality aggregated with\nother outcomes such as abortion. In these cases we could not separate neonatal\nmortality from abortion. Overall, we have data from 40 hospitals representing\naround 45 percent of the births. While this should not affect the internal validity of\nour estimates of the causal impact, it could affect the external validity. However, the\nneonatal mortality rate from our sample (8.3 per thousand) is close to the neonatal\nmortality rate based on death certificate data (10.1 per thousand). We would expect\nthe death rate from death certificates to be slightly higher than the in-hospital\nmortality rate as the former includes deaths that did not occur in hospitals.\n\n\n11 Women requiring a tetanus vaccine include those who had not previously been vaccinated or had\nreceived only one dose, those whose previous immunization status is not known, those who have had\ntwo doses but whose last vaccination is more than three years ago, and those that have had three\ndoses but the last dose or booster was more than five years ago (the maximum length recommended\nby clinical guidelines).\n\n\n16\n\nMention: \"death certificate data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its neonatal rate provides an external check on the sample estimate.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006001:28:1", "occurrence_id": "aab5032fd7bcc8621431de0f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  surveying and meetings with:\n\nSociété Mauritanienne de Commercialisation de Poissons (Mauritanian Company for Fish Exports, SMCP),\n\nFederation National des Pêches (National Federation of Fishery, FNP), the Institut Mauritanien de Recherches\n\nOcéanographiques et des Pêches (Mauritanian Institute for Oceanographic Research and Fishery, IMROP),\n\nDirection de la Marine Marchande du Ministere des Peches (Directorate of Merchant Shipping at the Ministry of\n\nFishery). In addition, World Bank staff visited the artisanal port of Nouakchott and the artisanal and commercial\n\nport of Nouadhibou, conducting direct interviews to local fishermen and insiders to obtain empirical data. This\n\nresearch utilized data on the number of boats, on the average catch (in tons) per year, on the average number of\n\ntrips (per boat) per year, on the average number of sailors per boat, on labor costs (docking costs, crew costs, fuel\n\ncosts and equipment costs), and on the price of various type of catches. The team provided assumptions for a\n\n\n48 Online database available at: data.worldbank.org/data-catalog/world-development-indicators\n49 Data available at http://go.worldbank.org/QNYZRM5540\n\n22\n\nMention: \"world-development-indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"A footnote giving the web address of this online database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:005979:27:0", "occurrence_id": "1ed49339c6c4337822ed96d3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 26\n\n\npocket payments than immiserized families. Nonetheless, as a count-based measure\n\n\nit is not bad, capturing as it does the idea that from a poverty-reduction perspective\n\n\nimpoverishing and immiserizing payments are worse than catastrophic payments\n\n\n(at least if defined in terms of whether they leave a family close to the poverty line),\n\n\nwhich are in turn worse than non-catastrophic payments, which are in turn worse\n\n\nthan no payments (especially if this group is restricted to users). Moreover, this\n\n\nindex is easy to explain to a non-technical audience.\n\n\nWe have illustrated our ideas using data from 51 countries for which out-of\n\npocket spending data are available in the World Health Survey. We use $1.25-a\n\nday, $2-a-day and $5-a-day international poverty lines, and a catastrophic payment\n\n\nthreshold of 40 percent. While intended to be only illustrative, the results are not\n\n\nwithout interest. For the 51 countries as a whole, we find very small shares of\n\n\nhouseholds falling into the impoverishment and catastrophic spending categories,\n\n\nand a much larger fraction falling into the immiserizing category. We also find a\n\n\nlarger fraction of households in the catastrophic payment category when\n\n\ncatastrophic payments are defined as payments that leave a family’s nonmedical\n\n\nconsumption below _Z_ times the poverty line. Unsurprisingly, as we raise the\n\n\npoverty line, the share of households falling into the immiserizing spending\n\n\ncategory increase, and the shares falling into the impoverishing and catastrophic\n\n\nspending categories decrease. In our more detailed analysis of the BRICS countries\n\n\n(Brazil, China, India, Russia and South Africa), we see appreciable fractions of the\n\n\npopulation in some countries falling into the impoverishing and catastrophic\n\nMention: \"World Health Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Out-of-pocket spending for 51 countries is available in this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006003:13:1", "occurrence_id": "90e1086f15b79d1449f0399f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  headcount estimate.\n\n\nIn light of this, the estimated further increase in poverty in 2011 seems plausible, particularly since the\nstructural parameters of the underlying model are unlikely to have changed dramatically over time. At the\nsame time, modeling assumptions vis-à-vis the dynamics of consumption and income, as well as the effects of\nthe social protection payments, are important, and the 2011 prediction should be regarded with the\nappropriate caveats in mind.\n\n\nNotably, the microsimulation-based forecast for 2011 diverges from the prediction based on the elasticity\napproach, which suggests a reduction in poverty in 2011, based on positive growth in GDP during 20102011. Furthermore, estimates for 2010 suggest that the elasticity approach forecasts can fall rather far from\nthe true poverty estimates, and can even imply poverty dynamics of opposite sign from the true poverty\nchanges between two points in time. Using HFCE data instead of GDP data appear to offer an improvement,\nespecially when the poverty measure is based on expenditure data, which need not follow GDP dynamics.\n\n\n12\n\nMention: \"HFCE data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Using consumption instead of output improves the measured poverty changes.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006004:27:3", "occurrence_id": "5511d899c38e382dab1a0dde", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  2.5\nmillion has of national park land). In turn, this means that the maximum carrying capacity of about 1.65 million,\n\n**Appendix 2 Description of SAM and CGE Multipliers** which is close to the estimate of the carrying capacity in Stratton ( _op cit_ ).\n\n\n\nπ _N_\n\n\n\n**Appendix 2 Description of SAM and CGE Multipliers**\n\n\n\nThe model is based on a SAM estimated on the basis of the GTAP data set supplemented by official and other\nstatistics. The basic model represents the Tanzanian economy with 10 economic sectors, 4 production factors, 2\ninstitutions, Capital Formation and The Rest of the World, or, in more detail:\nOur model represents the economy with 29 sectors. In the GTAP model the Natural Resources sector only\nincludes the value of the resources used by mining and extraction. The factor Land includes only cultivated land.\n\n26\n\nThe GTAP SAM is augmented in to take better account of the Serengeti economy:\n\nMention: \"the GTAP data set\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The social accounting matrix for one economy is estimated on this basis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006006:22:0", "occurrence_id": "43285f56f16262a789f15834", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Table 2: Previous studies of differential progress by wealth toward the health MDGs**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n|Col1|Moser, Leon and Gwatkin<br>(2005)|Suzuki, Sharan and Bos (2012)|Victora et al. (2012)|\n|---|---|---|---|\n|Time period<br>|<br>1990-2001<br>|1995-2007<br>|1990-2007<br>|\n|<br>No. countries<br>|<br>22<br>|<br>40<br>|<br>35<br>|\n|<br>Data source<br>|<br>Demographic and Health<br>Survey (DHS)<br>|<br>Demographic and Health Survey<br>(DHS), Multiple Indicator Cluster<br>Survey (MICS)<br>|<br>Demographic and<br>Health Survey (DHS)<br>|\n|No. surveys per<br>country<br>|2 <br>|<br>2 <br>|2 <br>|\n|<br>Official indicators<br>|Under-five mortality<br>|Prevalence of underweight children<br>under five years of age<br>Under-five mortality rate<br>Infant mortality rate<br>Measles immunization among children<br>under one<br>Proportion of births attended by<br>skilled health personnel<br>|Skilled birth attendant<br>Measles immunization<br>|\n|Additional core<br>intermediate<br>monitoring indicators<br>|None<br>|<br>Use of insecticide-treated bednets by<br>under-five children<br>\n\nMention: \"Demographic and Health Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Several indicator columns list this survey as their data source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006009:8:0", "occurrence_id": "0f8096ffa8020cc4f412c024", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: selected subgroup (females vs. males, urban vs. rural). However we do not use these additional\n\ndata to examine the nature of the evolution of private school participation at the same depth as\n\nwith the 1998/99 PIHS and 2004/05 and 2010/11 PSLM survey data.\n\nAll the surveys cover the four provinces and the Islamabad Capital Territory (ICT). ICT\n\naccounted for less than 1 percent of the population of Pakistan in 2012. <sup>7</sup> [^7: The denominator for the percentage is the population in the four provinces and ICT. The percentage is calculated\nfrom population projection data from the National Institute of Population Studies (NIPS).] Given its relatively\n\nsmall size, we exclude ICT from our analysis and only examine private school participation in\n\nthe four provinces. <sup>8</sup> [^8: Excluding ICT from the analysis does not influence the patterns we observe in private school participation in\nPakistan. In addition, given its size, and albeit territories and provinces are at the same administrative tier, it seems\ninappropriate to include ICT as a separate unit in any analysis where we compare patterns in private school\nparticipation across provinces. Using the 2010/11 PSLM survey, we estimate total school participation and private\nschool participation rates for the six to 10 age group of 89 percent and 33 percent respectively for ICT, which are\nhigher than in each of the provinces. In any analysis of schooling, what might be more appropriate is to compare\nICT to the provincial capitals or other major urban centers in the country.] For ease of exposition, we refer to the four provinces taken together as the\n\ncountry.\n\nThe household sample surveys ask a small set of basic questions about education.\n\nIndividuals age four and above are asked their current schooling status. Those who report being\n\ncurrently enrolled in school are asked in which grade or level they are and in which type of\n\nschool. Regarding school type, the response options include government, private,\n\nMention: \"PIHS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The earliest round allowing depth on private school participation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006009:8:0", "occurrence_id": "fccf9a82ccae3b413937d0ce", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: selected subgroup (females vs. males, urban vs. rural). However we do not use these additional\n\ndata to examine the nature of the evolution of private school participation at the same depth as\n\nwith the 1998/99 PIHS and 2004/05 and 2010/11 PSLM survey data.\n\nAll the surveys cover the four provinces and the Islamabad Capital Territory (ICT). ICT\n\naccounted for less than 1 percent of the population of Pakistan in 2012. <sup>7</sup> [^7: The denominator for the percentage is the population in the four provinces and ICT. The percentage is calculated\nfrom population projection data from the National Institute of Population Studies (NIPS).] Given its relatively\n\nsmall size, we exclude ICT from our analysis and only examine private school participation in\n\nthe four provinces. <sup>8</sup> [^8: Excluding ICT from the analysis does not influence the patterns we observe in private school participation in\nPakistan. In addition, given its size, and albeit territories and provinces are at the same administrative tier, it seems\ninappropriate to include ICT as a separate unit in any analysis where we compare patterns in private school\nparticipation across provinces. Using the 2010/11 PSLM survey, we estimate total school participation and private\nschool participation rates for the six to 10 age group of 89 percent and 33 percent respectively for ICT, which are\nhigher than in each of the provinces. In any analysis of schooling, what might be more appropriate is to compare\nICT to the provincial capitals or other major urban centers in the country.] For ease of exposition, we refer to the four provinces taken together as the\n\ncountry.\n\nThe household sample surveys ask a small set of basic questions about education.\n\nIndividuals age four and above are asked their current schooling status. Those who report being\n\ncurrently enrolled in school are asked in which grade or level they are and in which type of\n\nschool. Regarding school type, the response options include government, private,\n\nMention: \"PSLM survey data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Two later rounds covering the four provinces and the capital territory.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006014:16:1", "occurrence_id": "f80e87a202d6a78172023be6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  of different savings products offered by the staff. Commercial banks and SOFIPOs\n\noffered the least number of products. Similarly, auditors seeking an investment account are offered fewer\n\nproducts, simply because the institutions have less distinct investment products. According to the\n\nCONDUSEF dataset, of the 125 savings products, only 35 percent are investment products. Experienced\n\nauditors are not provided with more choices.\n\n[TABLE 2A HERE]\n\nWe note that the number of observations drops from 112 to 107 because all the auditors that were\n\nassigned to visit SOFOMs were turned away empty-handed, suggesting that SOFOMs do comply with the\n\nlaw. In addition, 92 percent of the visits the auditor was offered a product that aligned with his or her\n\npreferences (column 4). In sum, experienced auditors spent more time with the staff but were not offered\n\nmore products.\n\n[TABLE 2B HERE]\n\nTable 2B reports similar results to those of Table 2A for credit audits. Since an observation is now\n\nan auditor-branch pair that may involve more than one visit, columns 1 and 2 report the sum of each visit’s\n\nlength in minutes. For control auditors, now defined as informally dressed, with low experience and low\n\nindebtedness, the average wait time was around 6 minutes and the total time in face to face interactions\n\n15\n\nMention: \"CONDUSEF dataset\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Of 125 savings products, only 35 percent are investment products here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006024:35:1", "occurrence_id": "1eadbeea399c5e9b212c5e19", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  of observations (Total and per year)** **118** **4** **7** **29** **41** **37**\n\n**<u>Number of countries</u>** **<u>43</u>**\n\n_Notes: Income refers to per capita household income (net). Expenditure refers to per capita household expenditure. The circumstances birth region and region of residence may_\n_contain more than one variable (e.g. administrative region and rural/urban). The number of types is the average across the years for a given country, rounded to the nearest_\n_integer. The number of types may differ across years for a given country if some categories are unobserved in a particular year._\n_LIS: Luxemburg Income Study_\n_SEDLAC: Socioeconomic Database for Latin America and the Caribbean (CEDLAS-World Bank)_\n_I2D2: International Income Distribution Database_\n\nMention: \"Luxemburg Income Study\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"A note defining this abbreviation for the harmonised income dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006024:35:1", "occurrence_id": "ff3a7526957625b1ffc98e23", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  of observations (Total and per year)** **118** **4** **7** **29** **41** **37**\n\n**<u>Number of countries</u>** **<u>43</u>**\n\n_Notes: Income refers to per capita household income (net). Expenditure refers to per capita household expenditure. The circumstances birth region and region of residence may_\n_contain more than one variable (e.g. administrative region and rural/urban). The number of types is the average across the years for a given country, rounded to the nearest_\n_integer. The number of types may differ across years for a given country if some categories are unobserved in a particular year._\n_LIS: Luxemburg Income Study_\n_SEDLAC: Socioeconomic Database for Latin America and the Caribbean (CEDLAS-World Bank)_\n_I2D2: International Income Distribution Database_\n\nMention: \"Socioeconomic Database for Latin America and the Caribbean\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"A note defining this regional distributional database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006024:35:1", "occurrence_id": "e7d689ba8b7faf9976156f35", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  of observations (Total and per year)** **118** **4** **7** **29** **41** **37**\n\n**<u>Number of countries</u>** **<u>43</u>**\n\n_Notes: Income refers to per capita household income (net). Expenditure refers to per capita household expenditure. The circumstances birth region and region of residence may_\n_contain more than one variable (e.g. administrative region and rural/urban). The number of types is the average across the years for a given country, rounded to the nearest_\n_integer. The number of types may differ across years for a given country if some categories are unobserved in a particular year._\n_LIS: Luxemburg Income Study_\n_SEDLAC: Socioeconomic Database for Latin America and the Caribbean (CEDLAS-World Bank)_\n_I2D2: International Income Distribution Database_\n\nMention: \"International Income Distribution Database_\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"A note defining this harmonised labour collection.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006026:4:0", "occurrence_id": "c843f4142847c44375cb99e0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: more efficiently lobby for rents. Thus rent seeking activities may attract labor from different sectors of the\n\n\neconomy, potentially depriving the economy of potential entrepreneurs. (iii) Increases in private subsidies\n\n\ncan come at the cost of spending on public goods such as education and health. The positive correlation\n\n\nbetween education attainment and entrepreneurship has well been established (Parker, 2004). Thus private\n\n\nsubsidies may crowd out the government budget at the cost of other types of spending that may encourage\n\n\nentrepreneurship.\n\n\nOur empirical strategy is to exploit individual level variation in entrepreneurship activity and cross\n\n\ncountry variation in government spending. To achieve this we combine cross-country micro-economic\n\n\npanels from the Global Entrepreneurship Monitor with country-specific government spending data. Our\n\n\ndata cover 43 countries, spanning 2001-2009 for about 360,000 observations <sup>1</sup> [^1: List of countries include: Algeria, Argentina, Australia, Bolivia, Chile, China, Croatia, Czech Republic, Denmark,\nEgypt, Arab Rep., Finland, Greece, Guatemala, Hungary, India, Iran, Islamic Rep., Italy, Jordan, Kazakhstan,\nLatvia, Lebanon, Malaysia, New Zealand, Norway, Philippines, Poland, Portugal, Romania, Russian Federation,\nSerbia, Singapore, Slovenia, South Africa, Spain, Sweden, Switzerland, Syrian Arab Republic, Thailand, Uganda,\nUnited Kingdom, United States, Uruguay, Venezuela RB.] . Concerns about\n\n\nsimultaneity bias are ameliorated in our use of aggregate country-level explanatory variables because the\n\n\nindividual decision of a potential entrepreneur should not affect country-level factors. However,\n\n\nendogeneity may arise because the mean country-level individual entrepreneurship outcome may affect\n\n\nsome of the country-level variables, so we lag all our macroeconomic and institutional variables by one\n\n\nyear.\n\n\n2. Empirical Methods\n\n\nWe obtain individual level data from the Global Entrepreneurship Monitor (GEM). The data are\n\n\ngenerated through stratified samples of 2,000 individuals surveyed per country. We use two definitions of\n\n\nentrepreneurship. The first definition is the standard definition by Reynolds et al. (2005) and\n\nMention: \"Global Entrepreneurship Monitor\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Cross-country micro panels across 43 countries combined with spending data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006015:26:1", "occurrence_id": "8da92ec177be74fef788d5c1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  of ICS were able to reduce the cost of fuels (by\n\nusing less fuelwood) and time for cooking. The study finds that commercial users of ICS were\n\nable to save Euro 1.1 - Euro 6.6 per day. Based on Expenditure and Consumption Survey in Lao\n\nPeople’s Democratic Republic, national average time spent by women collecting fuelwood have\n\nfallen from 18 minutes in 2003 to about 12 minutes in 2008 mainly due to shift from fuelwood to\n\ncharcoal for cooking in urban areas (ASTAE, 2013b). However, time spent by women collecting\n\nfuelwood in the villages is significantly higher, in the range of 1-3 hours per day. Although time\n\nsavings and the opportunity cost of time are important, Jeuland and Pattanayak (2012), however,\n\nsuggest that the private net benefits of ICS are more likely negative because the ways in which\n\nusers change behaviors lead to no change or net increases in time spent cooking or preparing\n\nfuels resulting reduced health benefits.\n\n\nIn the case of costs associated ICS and modern fuels, WHO (2006b) finds that fuel-,\n\nstove- and program- costs are some of the main cooking system intervention costs. For instance,\n\n\n25\n\nMention: \"Expenditure and Consumption Survey in Lao\n\nPeople’s Democratic Republic\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"National average fuelwood collection time for women falls across these years.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006023:6:1", "occurrence_id": "7d5771d5b6524714804387ae", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 2012), Harrigan, Ma,\n\n\nand Shlychkov (2012), Kneller and Yu (2008), and Gervais (2013).\n\n\nAlthough we focus on Portugal, a middle-income country, we believe that our findings have\n\n\nimplications for our understanding of the upgrading process in developing countries as well. In\n\n\nparticular, the results reinforce the idea that raising the quality of outputs requires raising the\n\n\nquality of inputs (Verhoogen, 2008; Kugler and Verhoogen, 2012). This in turn suggests that\n\n\nincreasing exports to high income destinations may require the upgrading of entire complexes\n\n\nof suppliers and downstream producers, not just of exporters. The particular empirical setting\n\n\nhas the advantage that it allows us to identify cleanly a causal relationship between destination\n\n\nincome and material input prices, but the basic findings seem likely to apply more broadly.\n\n#### **2 Data**\n\n\nThe analysis in this paper draws on two main datasets, both collected by the _Instituto_ _Nacional_\n\n\n_de_ _Estat´ıstica_ _(INE)_, the Portuguese national statistical agency:\n\n\n1. Customs data on firm-level international trade transactions, which are collected separately\n\n\nfor European Union (EU) partner countries ( _Estat´ısticas Correntes do Com´ercio Intracomu-_\n\n\n_nit´ario_ [Current Statistics on Intra-community Trade]) and non-EU partners ( _Estat´ısticas_\n\n\n4\n\nMention: \"Customs data on firm-level international trade transactions\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Collected separately for union and non-union partner countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006023:6:1", "occurrence_id": "b5692011a3fed3979dc93d3c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 2012), Harrigan, Ma,\n\n\nand Shlychkov (2012), Kneller and Yu (2008), and Gervais (2013).\n\n\nAlthough we focus on Portugal, a middle-income country, we believe that our findings have\n\n\nimplications for our understanding of the upgrading process in developing countries as well. In\n\n\nparticular, the results reinforce the idea that raising the quality of outputs requires raising the\n\n\nquality of inputs (Verhoogen, 2008; Kugler and Verhoogen, 2012). This in turn suggests that\n\n\nincreasing exports to high income destinations may require the upgrading of entire complexes\n\n\nof suppliers and downstream producers, not just of exporters. The particular empirical setting\n\n\nhas the advantage that it allows us to identify cleanly a causal relationship between destination\n\n\nincome and material input prices, but the basic findings seem likely to apply more broadly.\n\n#### **2 Data**\n\n\nThe analysis in this paper draws on two main datasets, both collected by the _Instituto_ _Nacional_\n\n\n_de_ _Estat´ıstica_ _(INE)_, the Portuguese national statistical agency:\n\n\n1. Customs data on firm-level international trade transactions, which are collected separately\n\n\nfor European Union (EU) partner countries ( _Estat´ısticas Correntes do Com´ercio Intracomu-_\n\n\n_nit´ario_ [Current Statistics on Intra-community Trade]) and non-EU partners ( _Estat´ısticas_\n\n\n4\n\nMention: \"Current Statistics on Intra-community Trade\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The series covering partner countries inside the union.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006027:6:0", "occurrence_id": "9f9ca13a45aaf0d8bceb062c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: In Eastern Europe and Central Asia (ECA), we use the following two questions in the survey to get\n\ndata for innovation:\n\n\n_Q1: In the last three years, has this establishment introduced new products or services (Yes/No/_\n\n_Don’t answer)_\n\n\n_Q2: In the last three years, has this establishment upgraded an existing product line or service_\n\n_(Yes/No/Don’t answer)._\n\n\nWe define that a firm innovates when it answer Yes to either of the questions. By doing so, we can\n\nharmonize the innovation variables between LAC and ECA and hence can increase the sample size.\n\nThe downside of this is that in ECA, we will mix the return of an upgraded product and that of a\n\ncompletely new product <sup>4</sup> .\n\n\nOf 3,798 observations that answer, 1,855 answer Yes to one of the innovation questions. Figure 1\n\nsummarizes the profile of innovating firms by size and by regions. Large firms are more likely to\n\ninnovate than small firms. Europe and Central Asia firms are more likely to innovate than Latin\n\nAmerican firms.\n\n\n**Figure 1. Firms’ Innovation in Emerging Markets Economies**\n\n\n(number of observations)\n\n\nSource: Authors’ calculation based on The World Bank’s Enterprise Survey.\n\n\n4 See Akcigit and Kerr (2010) for a discussion about the innovation implications of completely new products and\nimprove products.\n\n\n5\n\nMention: \"The World Bank’s Enterprise Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Observation counts for emerging markets are calculated from this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006036:18:4", "occurrence_id": "11d8e65c2579e914adea282f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .288 20\nQatar 50-54 0.873** -0.110 12,838.886 -575.424 -71,604.843 11\nSaudi Arabia 50-54 0.804*** -0.024 -127.275 6.369 637.421 21\nSyria 50-54 1.695*** -0.767*** 12.749 -0.779 -51.207 26\nTunisia 50-54 1.679*** -0.794*** -1.344 0.121 4.200 31\n<u>Yemen</u> <u>50-54</u> <u>1.500***</u> <u>-0.524**</u> <u>36.434</u> <u>-2.365</u> <u>-139.518</u> <u>21</u>\nSource: ILO-EAPEP database (Female LFP) and World Bank, World Development Indicators (GDP per\ncapita in PPP values). Significance level: ***=.01; **= .05; *=.1 - *; Standard errors omitted.\n\n\n17\n\nMention: \"ILO-EAPEP database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Female labour force participation across economies comes from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006036:18:4", "occurrence_id": "73c41a23ce70aac2627d49ce", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .288 20\nQatar 50-54 0.873** -0.110 12,838.886 -575.424 -71,604.843 11\nSaudi Arabia 50-54 0.804*** -0.024 -127.275 6.369 637.421 21\nSyria 50-54 1.695*** -0.767*** 12.749 -0.779 -51.207 26\nTunisia 50-54 1.679*** -0.794*** -1.344 0.121 4.200 31\n<u>Yemen</u> <u>50-54</u> <u>1.500***</u> <u>-0.524**</u> <u>36.434</u> <u>-2.365</u> <u>-139.518</u> <u>21</u>\nSource: ILO-EAPEP database (Female LFP) and World Bank, World Development Indicators (GDP per\ncapita in PPP values). Significance level: ***=.01; **= .05; *=.1 - *; Standard errors omitted.\n\n\n17\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Output per head in purchasing power terms comes from this compilation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006037:61:3", "occurrence_id": "6ec0b164864a4ea06511d1d5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  (yield is proportional to the amount of\nthe most limiting nutrient, whichever nutrient it may be) the production of other two types\nof fertilizers will follow the path of nitrogen fertilizers.\n\n2Thorough description of the FAOSTAT database is available from the following website:\nhttp://faostat.fao.org/.\n\n3Thorough description of the dataset is available from the following website: http://www.\ners.usda.gov/Data/FertilizerUse/.\n\n\n15\n\nMention: \"the FAOSTAT database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Fertilizer use projections rest on this agricultural statistics database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006045:24:0", "occurrence_id": "1daec309c40c56443d2a7ea9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: distributed somewhat differently across countries. Egypt sees the most new city-to-city routes (6) with\n\n\nTurkey under this specification.\n\n\n**VI.** **Conclusions**\n\n\nIn this study, we estimate a set of empirical models of air passenger traffic, in order to better\n\n\nunderstand the relationship between air transport policy and international traffic. We use a WTO index\n\n\nmeasure of policy commitments in both bilateral and plurilateral air services agreements, and relate these\n\n\nmeasures to ICAO data on air passenger traffic. We find that country pairs with more liberal policy\n\n\ncommitments see greater air passenger traffic. Subsequent work at the city level indicates that, to a large\n\n\ndegree, larger passenger numbers in more liberal agreements occur because more city pairs are served by\n\n\na flight.\n\n\nWe use our empirical models to investigate the likely impact of two policy changes. First, we\n\n\nconsider the impact of deepening the ACAC agreement, awarding it the same policy score as the UK\n\nSingapore bilateral agreement. Because policy within the ACAC is already fairly liberal, the deepening\n\n\nof the agreement has modest, but significant effects. Our estimates suggest a 30% increase in traffic\n\n\nbetween ACAC countries as a result of the policy change.\n\n\nIn a second scenario we add Turkey to the deeper ACAC agreement, assigning the same liberal\n\n\nmeasures to Turkey-ACAC passenger flows. Due to the restrictive nature of policy commitments\n\n\nbetween Turkey and many Arab countries, this scenario generates much more growth in passenger traffic.\n\n\nOur estimates suggest that passenger traffic would more than double. Much of this would arise through\n\n\nincreasing numbers of direct flights between Turkish cities and cities in the ACAC member states.\n\n\nWhile these results may seem optimistic, there are three reasons to suggest they are not\n\n\nunrealistic. First, current air passenger traffic levels in the region are low, and fast growth is plausible off\n\n\na small base. Second, much faster growth rates have been observed in recent years in selected regional\n\n\nmarkets. Finally, the policy changes\n\nMention: \"ICAO data on air passenger traffic\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Traffic outcomes are related to the liberalisation of air services agreements.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006052:8:1", "occurrence_id": "58417299e739758f89ef0af3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  payments are also subtracted, the difference equals the primary fiscal balance. <sup>4</sup> We also\nused the Treasury database to obtain information on “social” public expenditures both at the\nstate and municipal level, which comprises expenditure on education and culture, health and\nsanitation as well as social security and social assistance (similar to what was used in\nFerreira, Leite and Ravaillon, 2010).\n\nMoreover, we used Regional National Accounts Statistics from IBGE to obtain a series for\nstate-level GDP and the respective deflators. Finally, we also use information on federal\nsocial transfers at the state level from a dataset constructed by the Institute for Applied\nEconomic Research (IPEA). This comprises information on three main federal social\nprograms: _Bolsa Familia_, _Beneficio de Prestacao Continuada_, and _Renda Mensal Vitalicia_ .\nThese federal social transfers are direct cash transfers to households.\n\n3 We linearly interpolate values for these years based on information from adjacent years. We also exclude\nhouseholds that reported zero income so that different measures of inequality could be computed in a\ncomparable manner. Sensitivity analysis shows that our results remain broadly the same when households with\nzero income are included in the calculation of inequality indicators.\n\n\n4 The National Treasury dataset used is publicly available at https://www.tesouro.fazenda.gov.br/pt/prefeituras<u>governos-estaduais/sobre.</u>\n\nMention: \"Treasury database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Social public expenditure at state and municipal levels comes from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006052:8:1", "occurrence_id": "1fe521db519617d815908719", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  payments are also subtracted, the difference equals the primary fiscal balance. <sup>4</sup> We also\nused the Treasury database to obtain information on “social” public expenditures both at the\nstate and municipal level, which comprises expenditure on education and culture, health and\nsanitation as well as social security and social assistance (similar to what was used in\nFerreira, Leite and Ravaillon, 2010).\n\nMoreover, we used Regional National Accounts Statistics from IBGE to obtain a series for\nstate-level GDP and the respective deflators. Finally, we also use information on federal\nsocial transfers at the state level from a dataset constructed by the Institute for Applied\nEconomic Research (IPEA). This comprises information on three main federal social\nprograms: _Bolsa Familia_, _Beneficio de Prestacao Continuada_, and _Renda Mensal Vitalicia_ .\nThese federal social transfers are direct cash transfers to households.\n\n3 We linearly interpolate values for these years based on information from adjacent years. We also exclude\nhouseholds that reported zero income so that different measures of inequality could be computed in a\ncomparable manner. Sensitivity analysis shows that our results remain broadly the same when households with\nzero income are included in the calculation of inequality indicators.\n\n\n4 The National Treasury dataset used is publicly available at https://www.tesouro.fazenda.gov.br/pt/prefeituras<u>governos-estaduais/sobre.</u>\n\nMention: \"Regional National Accounts Statistics\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"State-level output and deflators come from this source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006052:8:1", "occurrence_id": "1b0620326b8679c060c0ebfb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  payments are also subtracted, the difference equals the primary fiscal balance. <sup>4</sup> We also\nused the Treasury database to obtain information on “social” public expenditures both at the\nstate and municipal level, which comprises expenditure on education and culture, health and\nsanitation as well as social security and social assistance (similar to what was used in\nFerreira, Leite and Ravaillon, 2010).\n\nMoreover, we used Regional National Accounts Statistics from IBGE to obtain a series for\nstate-level GDP and the respective deflators. Finally, we also use information on federal\nsocial transfers at the state level from a dataset constructed by the Institute for Applied\nEconomic Research (IPEA). This comprises information on three main federal social\nprograms: _Bolsa Familia_, _Beneficio de Prestacao Continuada_, and _Renda Mensal Vitalicia_ .\nThese federal social transfers are direct cash transfers to households.\n\n3 We linearly interpolate values for these years based on information from adjacent years. We also exclude\nhouseholds that reported zero income so that different measures of inequality could be computed in a\ncomparable manner. Sensitivity analysis shows that our results remain broadly the same when households with\nzero income are included in the calculation of inequality indicators.\n\n\n4 The National Treasury dataset used is publicly available at https://www.tesouro.fazenda.gov.br/pt/prefeituras<u>governos-estaduais/sobre.</u>\n\nMention: \"National Treasury dataset\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"A publicly available dataset behind the inequality indicators.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006057:10:1", "occurrence_id": "7d7c433bd80d5adfb5257653", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  quality of\n\npolitical institutions. <sup>9</sup> [^9: Van de Ploeg and Poelhekke 2010 show that adverse growth effects of natural resources results mainly from the\nvolatility of commodity prices, especially for point-based resources.  Indeed, the indirect effects of resource exports\non growth via the volatility channel outweigh a direct positive effect of resource endowments on growth. Arezki,\nHamilton and Kazimov 2011 find that overall government spending in resource-exporting countries has been procyclical relative to commodity prices and that in the long run resource windfalls have negative effects on the growth\nof non-resource GDP. Both the effects of windfalls on macroeconomic stability and on growth are moderated by the\nquality of political institutions.] Improved capacity to manage shocks, a more robust financial sector, and efforts to\n\nencourage a more diversified economy are some of the ways in which resource-rich countries can work to\n\nminimize the curse.  A few, like Chile, have succeeded in sustaining counter-cyclical policies using fiscal\n\nrules and transparent and independent projections of long-run resource prices to guide policy (de Gregorio\n\nand Labbe 2011), but even for these it has not been easy to resist the pressures to increase spending at\n\n\n7 Oil prices have very high long-term volatility and hard mineral prices are not far behind. Hamilton 2008 estimates\nthe likely price band for oil prices up to four years into the future, assuming that the logarithm of prices follows a\nGaussian random walk. Starting from an initial price of $115 per barrel, the 95% confidence range for the four year\ndistribution was between $34 and $391. At the time, in the middle of the super-boom with prices spiking up to\n$140, no one could have imagined prices at the low end of the scale, but they collapsed to that level shortly after the\npublication of the study with the onset of the global crisis. In addition, many of the new producers face high base\nproduction costs, further increasing the sensitivity of rent flows to market prices.\n8 The IMF commodities outlook for December 17, 2013, putsmod\n\nMention: \"IMF commodities outlook\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its outlook puts forward the price path behind rent sensitivity.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006065:21:2", "occurrence_id": "85d61ba4dba7a4b9af89158c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .017 0.026\nSewing Machine 0.0396 0.0089 0.447 0.451\nPesticider 0.0816 0.0201 0.043 0.048\nThree wheeler 0.2041 0.0212 0.048 0.074\nHas own or shared toilet 0.0561 0.0242 0.976 0.983\nTractor 2 wheel 0.1571 0.024 0.031 0.038\nTractor 4 wheel 0.1487 0.0709 0.007 0.009\nTV 0.067 0.011 0.777 0.818\nUnemployed ratio to household size -0.0784 0.0315 0.048 0.052\nVCD 0.0691 0.0099 0.259 0.346\nWall: pressed soil blocks -0.09 0.0162 0.051 0.041\n<u>Washing machine</u> <u>0.0907</u> <u>0.018</u> <u>0.088</u> <u>0.121</u>\n\n<u>Source: Authors’ calculations using HIES 2006/07 and 2009/10</u>\n\n\n20\n\nMention: \"HIES\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Regression results are computed across these two household rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006093:29:0", "occurrence_id": "c3329aa355b8db4fbd37dda3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: granted in Croatia in 2011 amounted to 2.47% of GDP <sup>22</sup> ; which is much higher than the\naverage for EU 27 in the same period (0.51% of GDP). <sup>23</sup> In addition, the perception of\nintensity of local competition in the Croatian economy is well below the regional peers (see\nWEF data for 2011-12).\n\n\n22 Annual State Aid report Croatian Competition Agency\n23 Croatian data draws on the “Annual Report on State Aid for 20122”- Croatian Competition Agency. EU-27\naverage is from European Commission- DG Competition.\n\n\n20\n\nMention: \"WEF data for 2011-12\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Competition intensity relative to regional peers is measured in these data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006099:13:1", "occurrence_id": "9c068248d90e4f1ce92a9e60", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  10-29 years.</u>\n\n\n**3.** **Inequities in Learning Achievements**\n\n\nIn Section 2, we showed that while there has been democratization in access to basic\n\n\neducation in Egypt, there remain persistent disparities in access to general secondary and college\n\n\neducation. Given that educational trajectories depend to a large extent on tracking into general and\n\n\nvocational secondary schools at completion of basic education, the quality of the education at lower\n\n\nlevels of schooling is likely to be a major determinant of educational inequalities in Egypt. <sup>4</sup> We now\n\n\nturn to inequalities in learning achievements, as measured by TIMSS evaluation of pupils’\n\n\n4 The tracking is determined by scores at the preparatory completion exam organized at the level of governorate.\n\n\n12\n\nMention: \"TIMSS evaluation of pupils\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Inequalities in learning achievement are measured by this assessment.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006107:9:0", "occurrence_id": "bf6a97399529a6eaeda7d70c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: impact of the treatment.\n\n\nIn addition to the counterfactual that is missing, <sup>16</sup> another issue is that export and import\n\n\nactivities can be carried by traders. Indeed, promoting the relationship between producers and\n\n\ntraders might be one of the targets of EPAs, <sup>17</sup> but traders are accounted as services that are\n\n\nnot directly related to the firms that produce the goods that are exported, so results could\n\n\nbe misleading. I concentrate on the manufacturing sector where I assume that trade (export\n\n\nand import) is more likely to be accounted for producers, though manufacturing firms also\n\n\nmay trade goods they do not produce (see Bernard et al. (2012) and Crozet et al. (2013)). I\n\n\nuse the sectoral definition of CNAE 2.0 <sup>18</sup> (Portuguese acronym for National Economic Activity\n\n\nClassification) that is equivalent to the International Standard Industrial Classification of All\n\n\nEconomic Activities (ISIC, Rev.4.).\n\n\nRAIS provides yearly information for approximately 250,000 manufacturing firms per year,\n\n\nwhich can work as an annual census for the formal manufacturing sector in Brazil. This is critical\n\n\nbecause we can observe all formal firms over time, disregarding their export status. Since my\n\n\nidentification uses lagged variables in the first stage, I first impose a restriction in the dataset\n\n\naccording to which firms should appear in RAIS at periods _t −_ 1 and _t_ . <sup>19</sup> For example, there are\n\n\n257,473 manufacturing firms in RAIS in 2007. Among these firms 224,232 also appear in 2006\n\n\nand so on. <sup>20</sup>\n\n\nThe basic services provided by Apex-Brasil are related to matching domestic sellers and\n\n\nforeign buyers (e.g. through participation in trade fairs) and providing standard information on\n\n\nforeign markets, including standard qualification for exporting\n\nMention: \"RAIS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"It provides yearly records for around 250,000 manufacturing firms.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006110:10:1", "occurrence_id": "983e50bf8dc5f73809649b20", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  amount had there\n\n\n4 See Jensen and Anderson (2014). This is a revision and update of the effort by Valenzuela and Anderson (2008), who\nmade a similar adjustment to the Version 7 GTAP protection database for 2004. Specifically, the DAItoGTAP program\nhas four steps. _Firstly,_ it aggregates the World Bank’s NRA data, consisting of border market support (NRA_bms),\ndomestic price support (NRA_dms) and intermediate input support (NRA_I), to GTAP’s product groups. _Secondly,_ the\nNRA_bms are allocated as price wedges at national borders, as either import or export taxes depending on each country’s\ntrade status as a net importer or exporter. _Thirdly_, GTAP output and intermediate input subsidies are brought into line\nwith the NRA_dms and NRA_I estimates so that the GTAP database mirrors the DAI database in which NRA =\nNRA_bms + NRA_dms + NRA_I. _Fourthly_, the DAItoGTAP program generates a shock file used in the ALTERTAX\nprogram to incorporate the DAI price wedge estimates into our aggregation of the GTAP database.\n\nMention: \"GTAP protection database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"Its version seven protection data for that year received a similar adjustment.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006110:10:1", "occurrence_id": "86a76251eeaad02fb240a9cb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  amount had there\n\n\n4 See Jensen and Anderson (2014). This is a revision and update of the effort by Valenzuela and Anderson (2008), who\nmade a similar adjustment to the Version 7 GTAP protection database for 2004. Specifically, the DAItoGTAP program\nhas four steps. _Firstly,_ it aggregates the World Bank’s NRA data, consisting of border market support (NRA_bms),\ndomestic price support (NRA_dms) and intermediate input support (NRA_I), to GTAP’s product groups. _Secondly,_ the\nNRA_bms are allocated as price wedges at national borders, as either import or export taxes depending on each country’s\ntrade status as a net importer or exporter. _Thirdly_, GTAP output and intermediate input subsidies are brought into line\nwith the NRA_dms and NRA_I estimates so that the GTAP database mirrors the DAI database in which NRA =\nNRA_bms + NRA_dms + NRA_I. _Fourthly_, the DAItoGTAP program generates a shock file used in the ALTERTAX\nprogram to incorporate the DAI price wedge estimates into our aggregation of the GTAP database.\n\nMention: \"NRA data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"Border, domestic and input support components are aggregated from this source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006110:10:1", "occurrence_id": "593123210c555cbe6394f332", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  amount had there\n\n\n4 See Jensen and Anderson (2014). This is a revision and update of the effort by Valenzuela and Anderson (2008), who\nmade a similar adjustment to the Version 7 GTAP protection database for 2004. Specifically, the DAItoGTAP program\nhas four steps. _Firstly,_ it aggregates the World Bank’s NRA data, consisting of border market support (NRA_bms),\ndomestic price support (NRA_dms) and intermediate input support (NRA_I), to GTAP’s product groups. _Secondly,_ the\nNRA_bms are allocated as price wedges at national borders, as either import or export taxes depending on each country’s\ntrade status as a net importer or exporter. _Thirdly_, GTAP output and intermediate input subsidies are brought into line\nwith the NRA_dms and NRA_I estimates so that the GTAP database mirrors the DAI database in which NRA =\nNRA_bms + NRA_dms + NRA_I. _Fourthly_, the DAItoGTAP program generates a shock file used in the ALTERTAX\nprogram to incorporate the DAI price wedge estimates into our aggregation of the GTAP database.\n\nMention: \"GTAP database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"It is adjusted so that its assistance totals mirror the other database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006110:10:1", "occurrence_id": "a30adaca3100fde44b0d5fe6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  amount had there\n\n\n4 See Jensen and Anderson (2014). This is a revision and update of the effort by Valenzuela and Anderson (2008), who\nmade a similar adjustment to the Version 7 GTAP protection database for 2004. Specifically, the DAItoGTAP program\nhas four steps. _Firstly,_ it aggregates the World Bank’s NRA data, consisting of border market support (NRA_bms),\ndomestic price support (NRA_dms) and intermediate input support (NRA_I), to GTAP’s product groups. _Secondly,_ the\nNRA_bms are allocated as price wedges at national borders, as either import or export taxes depending on each country’s\ntrade status as a net importer or exporter. _Thirdly_, GTAP output and intermediate input subsidies are brought into line\nwith the NRA_dms and NRA_I estimates so that the GTAP database mirrors the DAI database in which NRA =\nNRA_bms + NRA_dms + NRA_I. _Fourthly_, the DAItoGTAP program generates a shock file used in the ALTERTAX\nprogram to incorporate the DAI price wedge estimates into our aggregation of the GTAP database.\n\nMention: \"DAI database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Its decomposition of assistance defines the target structure.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006111:43:1", "occurrence_id": "9517eefb1e5cbed595a1756c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: -19<br>20-24<br>25-29<br>30-34<br>35-39<br>40+<br>Age<br>Small-firm, Capital Intensive|50<br>100 150 200 250<br>Employment (Linear Prediction)<br><5<br>5-9<br>10-14<br>15-19<br>20-24<br>25-29<br>30-34<br>35-39<br>40+<br>Age<br>Low Dependence on External Finance|<br>100 200 300 400 500<br>Employment (Linear Prediction)<br><5<br>5-9<br>10-14<br>15-19<br>20-24<br>25-29<br>30-34<br>35-39<br>40+<br>Age<br>High Dependence on External Finance|\n\n\n\n\n\nSource: ASI Data pooled from the following census years: 1983/84, 1989/90, 1994/95, 2000/01, 2004/05\n\n\n\n42\n\nMention: \"ASI Data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Pooled across five benchmark years for the age and finance analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006112:46:0", "occurrence_id": "7f8a0dd7ded3f7cfdc2c378f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: where the parameter _ξ_ 1 <sup>_P_</sup> <sup>captures</sup> <sup>the</sup> <sup>curvature</sup> <sup>of</sup> <sup>the</sup> <sup>liquid</sup> <sup>fossil</sup> <sup>fuel</sup> <sup>cost</sup>\n\nfunction. We calibrate the values of the parameter _ξ_ 1 <sup>_P_</sup> <sup>to</sup> <sup>match</sup> <sup>observed</sup> <sup>ex-</sup>\n\ntraction rates of oil and natural gas in the rst decade of the 21st century. We\n\n\nobtain the extraction quantities of oil and natural gas from the BP Statistical\n\nReview of World Energy database (2013).\n\n\nB.1.1.3 Other Primary Resources\n\n\nThe initial values for the production of other primary inputs come from GTAP\n\n\nv.7 database and are de\u001cned as the value of output from labor (skilled and\n\n\nunskilled), capital, and natural resources. We set the growth rate of other\n\n\nprimary resources as the di\u001berence between global GDP growth and total factor\n\n\nproductivity growth rates from Jorgenson and Vu (2010) projections based on\n\n\neconometric estimates for 122 economies over the 1990 - 2008 period (see Figure\n\n\nB.1, panel b).\n\n\nB.1.2 Intermediate Inputs\n\n\nB.1.2.1 Petroleum Products\n\n\nThe re\u001cning of petroleum, _x_ <sup>_p_</sup> _t_ <sup>_,_</sup> <\n\nMention: \"BP Statistical\n\nReview of World Energy database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Oil and gas extraction quantities come from this edition.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006112:46:0", "occurrence_id": "8b00030dda15ee50124b8179", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: where the parameter _ξ_ 1 <sup>_P_</sup> <sup>captures</sup> <sup>the</sup> <sup>curvature</sup> <sup>of</sup> <sup>the</sup> <sup>liquid</sup> <sup>fossil</sup> <sup>fuel</sup> <sup>cost</sup>\n\nfunction. We calibrate the values of the parameter _ξ_ 1 <sup>_P_</sup> <sup>to</sup> <sup>match</sup> <sup>observed</sup> <sup>ex-</sup>\n\ntraction rates of oil and natural gas in the rst decade of the 21st century. We\n\n\nobtain the extraction quantities of oil and natural gas from the BP Statistical\n\nReview of World Energy database (2013).\n\n\nB.1.1.3 Other Primary Resources\n\n\nThe initial values for the production of other primary inputs come from GTAP\n\n\nv.7 database and are de\u001cned as the value of output from labor (skilled and\n\n\nunskilled), capital, and natural resources. We set the growth rate of other\n\n\nprimary resources as the di\u001berence between global GDP growth and total factor\n\n\nproductivity growth rates from Jorgenson and Vu (2010) projections based on\n\n\neconometric estimates for 122 economies over the 1990 - 2008 period (see Figure\n\n\nB.1, panel b).\n\n\nB.1.2 Intermediate Inputs\n\n\nB.1.2.1 Petroleum Products\n\n\nThe re\u001cning of petroleum, _x_ <sup>_p_</sup> _t_ <sup>_,_</sup> <\n\nMention: \"GTAP\n\n\nv.7 database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Initial values for other primary inputs come from this version.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006120:11:0", "occurrence_id": "669253ae7d4bd1970c0d0c04", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 4: Financial Depth vs. Outreach across the World**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nThis graph shows the correlation between the Bank Credit to GDP and the share of small firms with a bank\n[credit across countries. Source: authors’ calculations based on Enterprise Surveys (www.enterprisesurveys.org)](http://www.enterprisesurveys.org/)\nand the World Bank��s Global Financial Development Indicators.\n\n\n**3.** **What explains access to finance by SMEs in Africa and elsewhere?**\n\n\nThis section explores the covariates of the likelihood of having a loan from a formal\n\n\nfinancial institution. While the graphs in the previous section have shown that larger and\n\n\nolder firms are more likely to have a formal loan and have provided evidence for an African\n\n\ngap in the use of finance by SMEs, firm size and age might be correlated with other firm\n\n\ncharacteristics and the “Africa gap” in corporate finance might reflect the variation in the\n\n\ncomposition of the firm population. The following regression analysis will allow us to isolate\n\n\nthe firm characteristics that can explain the use of formal bank loans, gauge the existence of\n\n\n10\n\nMention: \"Enterprise Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Small-firm bank credit across countries is calculated from these surveys.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006120:11:0", "occurrence_id": "6e1683ba98db401e9cd25268", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 4: Financial Depth vs. Outreach across the World**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nThis graph shows the correlation between the Bank Credit to GDP and the share of small firms with a bank\n[credit across countries. Source: authors’ calculations based on Enterprise Surveys (www.enterprisesurveys.org)](http://www.enterprisesurveys.org/)\nand the World Bank’s Global Financial Development Indicators.\n\n\n**3.** **What explains access to finance by SMEs in Africa and elsewhere?**\n\n\nThis section explores the covariates of the likelihood of having a loan from a formal\n\n\nfinancial institution. While the graphs in the previous section have shown that larger and\n\n\nolder firms are more likely to have a formal loan and have provided evidence for an African\n\n\ngap in the use of finance by SMEs, firm size and age might be correlated with other firm\n\n\ncharacteristics and the “Africa gap” in corporate finance might reflect the variation in the\n\n\ncomposition of the firm population. The following regression analysis will allow us to isolate\n\n\nthe firm characteristics that can explain the use of formal bank loans, gauge the existence of\n\n\n10\n\nMention: \"Global Financial Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The companion indicators on financial access.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006124:9:0", "occurrence_id": "e0a78e679bd7cc9f3b2b9deb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: empirical measure is the stock of education (years of schooling per person available from Barro and Lee,\n2013) averaged over 1970-2010.\n\n\n_Food sufficiency_ :  Adequacy of food among mothers is likely to be related positively to higher baby\nweight, better health at birth, and higher levels of breastfeeding and thus to better chances for baby\nand child survival. Similarly, adequacy of food among children is likely to lead to stronger and healthier\nchildren with a reduced susceptibility to disease and death. In the regression analysis below we have\nincluded a variable called “undernourishment” which measures food intake status.  This measure is\navailable from the WDI database but only since 1992 and so the average over 1992-2010 is used in the\nempirical work. <sup>5</sup> [^5: The measure is called “Prevalence of Undernourishment” in the WDI database and refers to the proportion of the\npopulation whose food intake is insufficient to meet minimum dietary energy requirements continuously. The\nstatistical methodology note in the database states: “Data on undernourishment are from the Food and\nAgriculture Organization (FAO) of the United Nations and measure food deprivation based on average food\navailable for human consumption per person, the level of inequality in access to food, and the minimum calories\nrequired for an average person.” This measure is among those used to track the Millennium Development Goal\ntargets related to poverty and hunger (MDG1C).]\n\n\n_Urbanization_ : The link between urbanization and child mortality could go either way. Greater\nurbanization usually reflects higher population densities. On the positive side, this means that a given\nhealth facility is likely to be able to cover more people in more urban settings. Thus, all other things\nbeing equal, urbanization should result in greater economies of scale from health spending and better\nhealth outcomes, including for child mortality. More urban settings are also likely to be associated with\neasier access to health facilities through better transport than more rural settings.  On the negative\nside, greater population densities could also be associated with higher probabilities of disease incidence\n\nMention: \"WDI database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its food intake measure begins only in this year, so an average is used.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006127:23:1", "occurrence_id": "02f4f2da4bfec22945ddda89", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  ages 6-22)|Educational Attainment by Age Level<br>(indicates lower total educational attainment)<br>15<br>Attainment<br>10<br>Educational<br>5<br>0<br>6 8 10 12 14 16 18 20 22<br>Age<br>Least advantaged group Most advantaged group<br>Source data: Egypt DHS 2000 (sample of children ages 6-22)|Educational Attainment by Age Level<br>(indicates lower total educational attainment)<br>15<br>Attainment<br>10<br>Educational<br>5<br>0<br>6 8 10 12 14 16 18 20 22<br>Age<br>Least advantaged group Most advantaged group<br>Source data: Egypt DHS 2000 (sample of children ages 6-22)|\n|---|---|---|---|\n|<br>0<br>.2<br>.4<br>.6<br>.8<br>1<br>Enrollment Rate<br>6<br>8<br>10<br>12<br>14<br>16<br>18<br>20<br>22<br>Age<br>Least advantaged group<br>Most advantaged group<br>Source data: Egypt DHS 2008 (sample of children ages 6-22)<br>(indicates lower enrollment and early drop-out)<br>Enrollment by Age Level|0<br>.2<br>.4<br>.6<br>.8<br>1<br>Enrollment Rate<br>6<br>8<br>10<br>12<br\n\nMention: \"Egypt DHS 2000\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Enrollment by age for the two rounds shows early drop-out.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006127:23:1", "occurrence_id": "f522c3aee615ca1193aa722a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  ages 6-22)|Educational Attainment by Age Level<br>(indicates lower total educational attainment)<br>15<br>Attainment<br>10<br>Educational<br>5<br>0<br>6 8 10 12 14 16 18 20 22<br>Age<br>Least advantaged group Most advantaged group<br>Source data: Egypt DHS 2000 (sample of children ages 6-22)|Educational Attainment by Age Level<br>(indicates lower total educational attainment)<br>15<br>Attainment<br>10<br>Educational<br>5<br>0<br>6 8 10 12 14 16 18 20 22<br>Age<br>Least advantaged group Most advantaged group<br>Source data: Egypt DHS 2000 (sample of children ages 6-22)|\n|---|---|---|---|\n|<br>0<br>.2<br>.4<br>.6<br>.8<br>1<br>Enrollment Rate<br>6<br>8<br>10<br>12<br>14<br>16<br>18<br>20<br>22<br>Age<br>Least advantaged group<br>Most advantaged group<br>Source data: Egypt DHS 2008 (sample of children ages 6-22)<br>(indicates lower enrollment and early drop-out)<br>Enrollment by Age Level|0<br>.2<br>.4<br>.6<br>.8<br>1<br>Enrollment Rate<br>6<br>8<br>10<br>12<br\n\nMention: \"Egypt DHS 2008\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The later round showing lower enrollment and early drop-out.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006129:0:0", "occurrence_id": "496e1bee71b462cde184f7fa", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Policy Research Working Paper\n7028\nGLS Estimation and Empirical Bayes \nPrediction for Linear Mixed Models  \nwith Heteroskedasticity and Sampling Weights\nA Background Study for the POVMAP Project\nRoy van der Weide\nDevelopment Research Group\nPoverty and Inequality Team\nSeptember 2014\nWPS7028\nPublic Disclosure Authorized\nPublic Disclosure Authorized\nPublic Disclosure Authorized\nPublic Disclosure Authorized\nPublic Disclosure Authorized\nPublic Disclosure Authorized\nPublic Disclosure Authorized\nPublic Disclosure Authorized\n\nMention: \"the POVMAP Project\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"inform\", \"summary\": \"This methodological background study was prepared for that project.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006119:8:0", "occurrence_id": "8d82734a5ad9c9551ccd18d1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: this may actually induce banks to take on more risk with potentially negative\n\n\nrepercussions for the stability of the financial system.\n\n\nIn the remainder, section 2 discusses the data, and section 3 presents the empirical\n\n\nresults. Section 3 starts with an analysis of the relationships between corporate\n\n\ngovernance and overall bank insolvency risk and bank returns. Then it considers the\n\n\n‘channels’ on the assets and liabilities sides of the bank’s balance sheet through which\n\n\ncorporate governance affects overall bank risk. Finally, it considers the relationship\n\n\nbetween corporate governance and bank risk taking over the business cycle. Section 4\n\n\nconcludes with policy implications.\n\n\n**2. The data**\n\n\nIn this study, we relate measures of bank risk to summary indicators of corporate\n\n\ngovernance for an international sample of banks over the 2004-2008 period. Accounting\n\n\nand market data necessary to construct our bank risk measures are taken from Bankscope\n\n\nof Bureau Van Dijk, CRSP and Compustat Global. Data on bank corporate governance\n\n\nare from the Corporate Governance Quotient data base assembled by Institutional\n\n\nShareholder Services (ISS).\n\n\nWe examine two main measures of bank insolvency risk. First, the Z-score\n\n\nrepresents the number of standard deviations that a bank’s rate of return on assets has to\n\n\nfall for the bank to become insolvent. The Z-score is constructed as the sum of the rate of\n\n\nreturn on assets and the equity to assets ratio divided by the standard deviation of the\n\n\nreturn on assets. Table A1 in the Appendix describes variable definitions and data sources.\n\n\nA higher Z-score signals that a bank has lower insolvency risk. We calculate a Z-score\n\n\n7\n\nMention: \"Bankscope\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Accounting and market data behind the bank risk measures come from here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006119:8:0", "occurrence_id": "0a1228c5d124054398f926a8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: this may actually induce banks to take on more risk with potentially negative\n\n\nrepercussions for the stability of the financial system.\n\n\nIn the remainder, section 2 discusses the data, and section 3 presents the empirical\n\n\nresults. Section 3 starts with an analysis of the relationships between corporate\n\n\ngovernance and overall bank insolvency risk and bank returns. Then it considers the\n\n\n‘channels’ on the assets and liabilities sides of the bank’s balance sheet through which\n\n\ncorporate governance affects overall bank risk. Finally, it considers the relationship\n\n\nbetween corporate governance and bank risk taking over the business cycle. Section 4\n\n\nconcludes with policy implications.\n\n\n**2. The data**\n\n\nIn this study, we relate measures of bank risk to summary indicators of corporate\n\n\ngovernance for an international sample of banks over the 2004-2008 period. Accounting\n\n\nand market data necessary to construct our bank risk measures are taken from Bankscope\n\n\nof Bureau Van Dijk, CRSP and Compustat Global. Data on bank corporate governance\n\n\nare from the Corporate Governance Quotient data base assembled by Institutional\n\n\nShareholder Services (ISS).\n\n\nWe examine two main measures of bank insolvency risk. First, the Z-score\n\n\nrepresents the number of standard deviations that a bank’s rate of return on assets has to\n\n\nfall for the bank to become insolvent. The Z-score is constructed as the sum of the rate of\n\n\nreturn on assets and the equity to assets ratio divided by the standard deviation of the\n\n\nreturn on assets. Table A1 in the Appendix describes variable definitions and data sources.\n\n\nA higher Z-score signals that a bank has lower insolvency risk. We calculate a Z-score\n\n\n7\n\nMention: \"Corporate Governance Quotient data base\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Bank corporate governance measures come from this assembled database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006132:9:0", "occurrence_id": "7f5128dac5426b68c36a9f96", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: aged 6-17 years old and out of school and inactive persons above the age of 60). This selection of\nvariables was largely based on findings from the literature on emerging economies and from our\nown screening of possible predictors of FLP.\n\n\nThe HBS data used for comparing results with LFS data is the 2007 National Living Standards\nSurvey covering 7,200 households. The survey focused on household expenditure and revenues\nand was principally administered to measure poverty, inequality and other dimensions of living\nstandards but contains modules on employment and education and uses the same employment and\neducational categories used by the LFS. It was therefore possible to reproduce with precision the\nset of variables used with LFS data.\n\n\n**Results**\n\n\n**FLP trends**\n\n\nThe period covered by this study (1999-2012) was a period characterized by sustained growth,\naveraging four percent a year, and by a very rich set of macroeconomic and social reforms that\ntransformed the economy. For most of the period considered, macroeconomic stability reigned in\nMorocco and GDP growth trickled down effectively to the population so that the country was\nable to deliver in terms of households income and poverty reduction. The overall economic\nperformance of Morocco during the period considered was good by macroeconomic and\nmicroeconomic standards (see Douidich et al, 2013 and Verme et al., 2014 for a full description\nof reforms and economic performance).\n\nBy contrast, the labor market performance has been disappointing. The unemployment rate\ndeclined marginally in recent years but only for men and the Labor Force Participation Rate\n(LFPR) and the Employment Rate experienced little changes, despite evident employment and\njob mobility, and females experienced overall declines in both rates (Verme et al, 2014).\nQuarterly data (Figure 1) show that: 1) the FPR in both urban and rural areas declines over time;\n2) the FPR rate is higher in rural areas as compared to urban areas; 3) The urban-rural gap\nincreases over time and 4\n\nMention: \"2007 National Living Standards\nSurvey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its 7,200 households allow comparison of results with the labour survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006132:9:0", "occurrence_id": "e0c97bf850149db41d3daf31", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: aged 6-17 years old and out of school and inactive persons above the age of 60). This selection of\nvariables was largely based on findings from the literature on emerging economies and from our\nown screening of possible predictors of FLP.\n\n\nThe HBS data used for comparing results with LFS data is the 2007 National Living Standards\nSurvey covering 7,200 households. The survey focused on household expenditure and revenues\nand was principally administered to measure poverty, inequality and other dimensions of living\nstandards but contains modules on employment and education and uses the same employment and\neducational categories used by the LFS. It was therefore possible to reproduce with precision the\nset of variables used with LFS data.\n\n\n**Results**\n\n\n**FLP trends**\n\n\nThe period covered by this study (1999-2012) was a period characterized by sustained growth,\naveraging four percent a year, and by a very rich set of macroeconomic and social reforms that\ntransformed the economy. For most of the period considered, macroeconomic stability reigned in\nMorocco and GDP growth trickled down effectively to the population so that the country was\nable to deliver in terms of households income and poverty reduction. The overall economic\nperformance of Morocco during the period considered was good by macroeconomic and\nmicroeconomic standards (see Douidich et al, 2013 and Verme et al., 2014 for a full description\nof reforms and economic performance).\n\nBy contrast, the labor market performance has been disappointing. The unemployment rate\ndeclined marginally in recent years but only for men and the Labor Force Participation Rate\n(LFPR) and the Employment Rate experienced little changes, despite evident employment and\njob mobility, and females experienced overall declines in both rates (Verme et al, 2014).\nQuarterly data (Figure 1) show that: 1) the FPR in both urban and rural areas declines over time;\n2) the FPR rate is higher in rural areas as compared to urban areas; 3) The urban-rural gap\nincreases over time and 4\n\nMention: \"LFS data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its variable set was reproduced precisely in the other survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006135:14:0", "occurrence_id": "e9cb4f8932b77561b22cf830", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: village that had experienced payouts would have also experienced uninsured rainfall shocks during the\n\n\nsame season. Therefore it may be possible that there are effects of trust and learning, but they are\n\n\noutweighed by opposite effects of the weather. As we saw in the previous section, rainfall shocks tend\n\n\nto have a negative effect on insurance demand, so the (lack of) evidence of spillovers may be a result\n\nof a more complex interaction between trust/learning and direct effects of weather.\n\n\nOverall, these results do not support the hypothesis that trust, learning, or any other effects of\n\n\nsimply witnessing insurance payouts are driving increased purchasing. While it is possible that our\n\n\nmeasurements of spillovers are too crude and miss more subtle effects, the data simply does not provide\n\n\nevidence that here are strong spillover effects. Since we do not see these spillover effects, this casts\n\n\ndoubt on the theory that repurchases are being driven by increased trust in the insurance company or\n\n\nlearning about insurance payouts.\n\n\n**4.3** **Direct** **Effects** **of** **Payouts** **on** **Wealth** **and** **Liquidity**\n\n\nThe previous two sections discount the possibility that trust, learning, or weather effects are driving\n\n\nthe result that receiving an insurance payout is correlated with purchasing insurance the following\n\n\nyear. This points to the actual reception of money from the insurance company as being the driving\n\n\nforce behind greater takeup. The most natural explanation for this phenomenon would be that recving\n\n\nan insurance payout could directly affect choices the next year due to its effects on wealth and liquidity.\n\n\nFor instance, if insurance is a normal good then increased wealth would result in greater insurance\n\ndemand. <sup>16</sup>\n\n\nWhile the BASIX data set does not offer the opportunity to test the direct effects of a cash payment\n\n\nseparately from an insurance payout, there are a number of reasons why it\n\nMention: \"BASIX data set\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"It does not allow testing the direct effect of a cash payment separately.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006138:3:3", "occurrence_id": "0e690c7da4497d8c30cf909d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  results of the most recent 2011 Business Environment\n\n\n\nin Russia’s regions vary considerably (Kisunko & Knack 2013; Plekhanov & Isakova 2011;\n\nYakovlev 2006; CEFIR 2006). The results of the most recent 2011 Business Environment\n\nand Enterprise Performance Survey (BEEPS) conducted jointly by the EBRD and the World\n\n\n\nYakovlev 2006; CEFIR 2006). The results of the most recent 2011 Business Environment\n\nand Enterprise Performance Survey (BEEPS) conducted jointly by the EBRD and the World\n\nBank once again highlight the regional diversity in business climate: regional location was\n\n\n\nand Enterprise Performance Survey (BEEPS) conducted jointly by the EBRD and the World\n\nBank once again highlight the regional diversity in business climate: regional location was\n\nfound to be significantly correlated with firms’ perceptions of administrative burden,\n\n\n\nBank once again highlight the regional diversity in business climate: regional location was\n\nfound to be significantly correlated with firms’ perceptions of administrative burden,\n\ncorruption, and state capture indicators. Further complicating the picture is the fact that\n\n\n\nfound to be significantly correlated with firms’ perceptions of administrative burden,\n\ncorruption, and state capture indicators. Further complicating the picture is the fact that\n\n<u>regions that rank high on some business climate indicators rank low on others. Thus, a</u>\n\n\n\ncorruption, and state capture indicators. Further complicating the picture is the fact that\n\n<u>regions that rank high on some business climate indicators rank low on others. Thus, a</u>\n\n<u>region that is low on perceptions of administrative corruption might at the same time be</u>\n\n\n\n<u>regions that rank high on some business climate indicators rank low on others. Thus, a</u>\n\n<u>region that is low on perceptions of administrative corruption might at the same time be</u>\n\n\n\n~~1~~ ~~The Russian Federation is divide~~ d into 85 federal subjects (\n\nMention: \"2011 Business Environment\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its regional results show considerable variation in business climate.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006138:3:3", "occurrence_id": "442c4d6cded12208b049f946", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  results of the most recent 2011 Business Environment\n\n\n\nin Russia’s regions vary considerably (Kisunko & Knack 2013; Plekhanov & Isakova 2011;\n\nYakovlev 2006; CEFIR 2006). The results of the most recent 2011 Business Environment\n\nand Enterprise Performance Survey (BEEPS) conducted jointly by the EBRD and the World\n\n\n\nYakovlev 2006; CEFIR 2006). The results of the most recent 2011 Business Environment\n\nand Enterprise Performance Survey (BEEPS) conducted jointly by the EBRD and the World\n\nBank once again highlight the regional diversity in business climate: regional location was\n\n\n\nand Enterprise Performance Survey (BEEPS) conducted jointly by the EBRD and the World\n\nBank once again highlight the regional diversity in business climate: regional location was\n\nfound to be significantly correlated with firms’ perceptions of administrative burden,\n\n\n\nBank once again highlight the regional diversity in business climate: regional location was\n\nfound to be significantly correlated with firms’ perceptions of administrative burden,\n\ncorruption, and state capture indicators. Further complicating the picture is the fact that\n\n\n\nfound to be significantly correlated with firms’ perceptions of administrative burden,\n\ncorruption, and state capture indicators. Further complicating the picture is the fact that\n\n<u>regions that rank high on some business climate indicators rank low on others. Thus, a</u>\n\n\n\ncorruption, and state capture indicators. Further complicating the picture is the fact that\n\n<u>regions that rank high on some business climate indicators rank low on others. Thus, a</u>\n\n<u>region that is low on perceptions of administrative corruption might at the same time be</u>\n\n\n\n<u>regions that rank high on some business climate indicators rank low on others. Thus, a</u>\n\n<u>region that is low on perceptions of administrative corruption might at the same time be</u>\n\n\n\n~~1~~ ~~The Russian Federation is divide~~ d into 85 federal subjects (\n\nMention: \"2011 Business Environment\n\nand Enterprise Performance Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Conducted jointly by two institutions across regions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006138:3:3", "occurrence_id": "5f410368a7e358a5a0bd55fd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  results of the most recent 2011 Business Environment\n\n\n\nin Russia’s regions vary considerably (Kisunko & Knack 2013; Plekhanov & Isakova 2011;\n\nYakovlev 2006; CEFIR 2006). The results of the most recent 2011 Business Environment\n\nand Enterprise Performance Survey (BEEPS) conducted jointly by the EBRD and the World\n\n\n\nYakovlev 2006; CEFIR 2006). The results of the most recent 2011 Business Environment\n\nand Enterprise Performance Survey (BEEPS) conducted jointly by the EBRD and the World\n\nBank once again highlight the regional diversity in business climate: regional location was\n\n\n\nand Enterprise Performance Survey (BEEPS) conducted jointly by the EBRD and the World\n\nBank once again highlight the regional diversity in business climate: regional location was\n\nfound to be significantly correlated with firms’ perceptions of administrative burden,\n\n\n\nBank once again highlight the regional diversity in business climate: regional location was\n\nfound to be significantly correlated with firms’ perceptions of administrative burden,\n\ncorruption, and state capture indicators. Further complicating the picture is the fact that\n\n\n\nfound to be significantly correlated with firms’ perceptions of administrative burden,\n\ncorruption, and state capture indicators. Further complicating the picture is the fact that\n\n<u>regions that rank high on some business climate indicators rank low on others. Thus, a</u>\n\n\n\ncorruption, and state capture indicators. Further complicating the picture is the fact that\n\n<u>regions that rank high on some business climate indicators rank low on others. Thus, a</u>\n\n<u>region that is low on perceptions of administrative corruption might at the same time be</u>\n\n\n\n<u>regions that rank high on some business climate indicators rank low on others. Thus, a</u>\n\n<u>region that is low on perceptions of administrative corruption might at the same time be</u>\n\n\n\n~~1~~ ~~The Russian Federation is divide~~ d into 85 federal subjects (\n\nMention: \"Enterprise Performance Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its results again highlight regional diversity in the business climate.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006143:21:0", "occurrence_id": "9aa8426adb02cb64ed2ad390", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: premium. We also report estimates from two regressions: a full model where all variables are introduced\n\n\nsimultaneously; and the estimates from the second stage regression where …rst stage regression did not\n\n\ninclude any of the three variables of our interest. The full model thus corresponds to speci…cations whose\n\n\nresults are reported in column 3 of Table 2 and columns 2, 4, and 6 of Table 3. The conditional estimates\n\n\nfrom second stage correspond to results reported in the even numbered columns of Table 4. We report the\n\n\nresults for all migrants in Table 5.\n\n\nThe regression results reported so far come from speci…cations where income and distance variables are\n\n\nmeasured in levels. In most migration studies, these variables are often introduced in the logarithmic form.\n\n\nThe logarithmic form would imply a log-linear utility function which –according to Grogger and Hanson\n\n\n(2011) –could be mis-speci…ed leading to omitted variable bias. To avoid mis-speci…cation, we estimated\n\n\nthe speci…cation consistent with the linear utility function where income and distance enter the regression\n\n\nequation linearly. In the …rst robustness check, we estimate the regression in log-linear form with both\n\n\nincome and distance variables in logarithms. The estimates of parameters of our main interest from the\n\n\nfull and conditional (second stage) regressions are reported in columns 1 and 2 of Table 5. The estimates\n\n\nare similar in sign and magnitudes to those reported in Tables 2 and 4.\n\n\nIn the next couple of robustness exercises, we address the issue of potential measurement errors in income\n\n\nestimation. The district level income estimates come from the NLSS 2002/3 data. Income estimates from\n\n\nhousehold surveys typically involve measurement error though NLSS 2002/3 is a nationally representative\n\n\nsurvey. To check whether our estimates are sensitive to alternative indicators of income, we conduct three\n\n\nrobustness checks: (i) Instead of average income adjusted for household size and composition, we use\n\n\nmedian per\n\nMention: \"NLSS 2002/3 data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"District-level income estimates are drawn from this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006157:7:0", "occurrence_id": "79c13980ebf3ac189accf4c4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: agreements require that imported products receive no less favorable treatment than ‘like\nproducts’ of national origin, such as an energy-efficient washing machine or the use of lowemission technology in aluminum production. <sup>5</sup> Along the same vein, criteria are lacking to\njudge what is “environmentally friendly”. Apart from the divergence in preferences\n(conceptions of the “environment”), this is also due to the hurdles facing the completion of a\nlife-cycle assessment as a same good may be used and disposed of in different ways. Again,\nusing “ex” headings for identification at the 8-digit or 10-digit level would be part of the\nsolution. Among the 411 products on a WTO combined list, more than two-thirds only\nappeared on one submission list, an indication that the perception of EGs differs widely\nacross countries (Balineau and Melo, 2011).\n\n\nAs examples, Steenblik (2007) and Hufbauer et al. (2009), stress that the use of bio-fuels to\nsave on energy and reduce CO2 emissions is, at best, doubtful, not to mention that today’s\ncleanest available technology will change as technological progress occurs, calling for regular\nupdates of the list of goods that would benefit from tariff exemptions. Moreover, the\nidentification of EPPs requires an efficient disclosure mechanism, which can be very costly\nwhen attributes are not observable in the final product (e.g. efficient third-party certification\nfor “credence goods”, calling for an international standard and certification process). <sup>6</sup>\n\n\nSeveral lists of EGs have been drawn. After many rounds of submissions, in 2010, 13\ndeveloped countries proposed lists resulting in a combined list (the ‘WTO list’ of 411\nproducts that included many of the products in the earlier OECD and APEC lists). Another\nnegotiated list is the “Friends of\n\nMention: \"WTO combined list\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Of 411 products, over two-thirds appeared on only one submission list.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006159:20:2", "occurrence_id": "3d6731bb6b7998348f990fba", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  and assets to GDP in 2007 or the outstanding amount of international loans from non-resident banks to GDP ratio (Data from Lane and Milesi-Ferretti (2007) or Beck et al. (2009)).\nWe failed to find any statistically significant additional effect. A banking crisis appears thus to give a superior signal about\nthe degree of financial distress in the banking sector of a given country during the period 2008-2010.\n\n35Data come from UNIDO Industrial Statistics Database (INDSTAT4), http://www.unido.org/en/resources/statistics/statistical\n18\n\nMention: \"UNIDO Industrial Statistics Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Industrial data behind the banking sector analysis come from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006159:20:2", "occurrence_id": "d64e4d445f244da4123e63a3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  and assets to GDP in 2007 or the outstanding amount of international loans from non-resident banks to GDP ratio (Data from Lane and Milesi-Ferretti (2007) or Beck et al. (2009)).\nWe failed to find any statistically significant additional effect. A banking crisis appears thus to give a superior signal about\nthe degree of financial distress in the banking sector of a given country during the period 2008-2010.\n\n35Data come from UNIDO Industrial Statistics Database (INDSTAT4), http://www.unido.org/en/resources/statistics/statistical\n18\n\nMention: \"INDSTAT4\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The same source under its short name.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006140:18:0", "occurrence_id": "4e2381fe0c1211d4382bd702", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: curves accessed from the AIDSinfo Online Database from the 2012 and 2013 UNAIDS\nreports, only differing by one more years of data, several countries showed considerable\ndifference between the two incidence curves. In many cases estimated peak incidence\nwas higher from the 2013 report (12 out of 19 countries had higher incidence peaks\nusing the 2013 report data; median difference between highest incidence from 2013 to\n2012 report among these 12 countries, 0.33, but as high as 1.37) sometimes occurring in\na different year than the 2012 report, and these changes were evident as far back as\n1990 (Annex 2).\n\nStill, changes in incidence, estimated by Spectrum, have been important for measuring\nprogress in the global fight against HIV/AIDS at regional and country levels. The 2012\nGlobal AIDS report grouped countries by overall changes in incidence between 20012011, with those falling into categories of increasing incidence over 25%, stable\nincidence (less than 25% change in incidence, up or down), and those with reductions in\nincidence greater than 25% and 50%. Although the majority of countries in the 2012\nreport have estimated incidence reductions over 25%, a third of them are estimated to\nhave experienced stable incidence, or worse, incidence increases over 25% between\n2001 and 2011 (including Congo, Lesotho, Nigeria, Tanzania, and Lesotho, which have all\nexperienced stable incidence over that period).\n\n**<u>2.2. Effect of ART on prevalence data, mortality, and infectiousness</u>**\n\nAchieving universal access to ART (reaching 80% of those in need of treatment) is a\nglobal priority, and the provision of ARVs is a key component of combination HIV\nprevention strategies, as ART reduces the likelihood of both sexual and mother-to-child\ntransmission of HIV. Following the hallmark HPTN052 study showing that early HIV\ntreatment of HIV positive partners among sero-discordant couples significantly reduced\ntransmission to their partners (Cohen\n\nMention: \"AIDSinfo Online Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Prevalence curves are accessed from these two report editions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006140:18:0", "occurrence_id": "3329c1b0b37941c65d484d7c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: curves accessed from the AIDSinfo Online Database from the 2012 and 2013 UNAIDS\nreports, only differing by one more years of data, several countries showed considerable\ndifference between the two incidence curves. In many cases estimated peak incidence\nwas higher from the 2013 report (12 out of 19 countries had higher incidence peaks\nusing the 2013 report data; median difference between highest incidence from 2013 to\n2012 report among these 12 countries, 0.33, but as high as 1.37) sometimes occurring in\na different year than the 2012 report, and these changes were evident as far back as\n1990 (Annex 2).\n\nStill, changes in incidence, estimated by Spectrum, have been important for measuring\nprogress in the global fight against HIV/AIDS at regional and country levels. The 2012\nGlobal AIDS report grouped countries by overall changes in incidence between 20012011, with those falling into categories of increasing incidence over 25%, stable\nincidence (less than 25% change in incidence, up or down), and those with reductions in\nincidence greater than 25% and 50%. Although the majority of countries in the 2012\nreport have estimated incidence reductions over 25%, a third of them are estimated to\nhave experienced stable incidence, or worse, incidence increases over 25% between\n2001 and 2011 (including Congo, Lesotho, Nigeria, Tanzania, and Lesotho, which have all\nexperienced stable incidence over that period).\n\n**<u>2.2. Effect of ART on prevalence data, mortality, and infectiousness</u>**\n\nAchieving universal access to ART (reaching 80% of those in need of treatment) is a\nglobal priority, and the provision of ARVs is a key component of combination HIV\nprevention strategies, as ART reduces the likelihood of both sexual and mother-to-child\ntransmission of HIV. Following the hallmark HPTN052 study showing that early HIV\ntreatment of HIV positive partners among sero-discordant couples significantly reduced\ntransmission to their partners (Cohen\n\nMention: \"2013 report data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its incidence peaks are higher in most countries relative to the prior edition.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006149:7:0", "occurrence_id": "d7a2c21435c9a1d256e961d6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: fulfilling the necessary data requirements for our analysis do not seem to bias the representation\n\n\nof the final sample, which covers 36 host countries. The data set used in the analysis also\n\n\nincludes stock market prices for all other banks and firms in the home and host countries that are\n\n\nused to construct default risk control variables in the regressions.\n\n\nWe use daily stock market information from Compustat Global for international banks\n\n\nand firms and stock market information from CRSP for U.S. banks and firms. Bank level\n\n\nvariables are constructed from Bankscope, a comprehensive commercial database of banks'\n\n\nfinancial statements produced by Bureau Van Dijk. Since we are interested in how bank\n\n\ncharacteristics affect the correlation between parents’ and their foreign subsidiaries’ default risks,\n\n\nand since there have been significant changes to bank balance sheets during the crisis, we\n\n\nconstruct and use bank-level variables measured prior to the crisis (as of December 2006). For\n\n\neach bank, we calculate relative bank _size_ (bank assets to total system assets), _capital_ _ratio_\n\n\n(regulatory capital to risk weighted assets), _equity ratio_ (equity to total assets), _provisions_ (loan\n\n\nloss provisions divided by total loans), _deposit_ _funding_ (deposits divided by total funding),\n\n\n_profitability_ (net income divided by total assets), and _liquidity_ (liquid assets divided by total\n\n\nassets). <sup>5</sup> [^5: Note that in the ratios mentioned here assets and equity are measured as book values. We also tried running\nregressions using the market value of assets and equity computed from the Merton model discussed above and found\nsimilar results.] We winsorize all financial variables at the 1 <sup>st</sup> and 99 <sup>th</sup> percentile level of their\n\n\ndistributions to reduce the influence of outliers and potential data errors.\n\n\nHost country-level variables are collected from a number of sources. We use data from\n\n\nthe World Bank 2007 Bank Regulation and Supervision survey to construct different indexes of\n\n\nbank regulation and supervision, following the methodology proposed\n\nMention: \"Compustat Global\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Daily market information for international banks and firms.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006149:7:0", "occurrence_id": "24cf2462e30cb0bd869ce78f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: fulfilling the necessary data requirements for our analysis do not seem to bias the representation\n\n\nof the final sample, which covers 36 host countries. The data set used in the analysis also\n\n\nincludes stock market prices for all other banks and firms in the home and host countries that are\n\n\nused to construct default risk control variables in the regressions.\n\n\nWe use daily stock market information from Compustat Global for international banks\n\n\nand firms and stock market information from CRSP for U.S. banks and firms. Bank level\n\n\nvariables are constructed from Bankscope, a comprehensive commercial database of banks'\n\n\nfinancial statements produced by Bureau Van Dijk. Since we are interested in how bank\n\n\ncharacteristics affect the correlation between parents’ and their foreign subsidiaries’ default risks,\n\n\nand since there have been significant changes to bank balance sheets during the crisis, we\n\n\nconstruct and use bank-level variables measured prior to the crisis (as of December 2006). For\n\n\neach bank, we calculate relative bank _size_ (bank assets to total system assets), _capital_ _ratio_\n\n\n(regulatory capital to risk weighted assets), _equity ratio_ (equity to total assets), _provisions_ (loan\n\n\nloss provisions divided by total loans), _deposit_ _funding_ (deposits divided by total funding),\n\n\n_profitability_ (net income divided by total assets), and _liquidity_ (liquid assets divided by total\n\n\nassets). <sup>5</sup> [^5: Note that in the ratios mentioned here assets and equity are measured as book values. We also tried running\nregressions using the market value of assets and equity computed from the Merton model discussed above and found\nsimilar results.] We winsorize all financial variables at the 1 <sup>st</sup> and 99 <sup>th</sup> percentile level of their\n\n\ndistributions to reduce the influence of outliers and potential data errors.\n\n\nHost country-level variables are collected from a number of sources. We use data from\n\n\nthe World Bank 2007 Bank Regulation and Supervision survey to construct different indexes of\n\n\nbank regulation and supervision, following the methodology proposed\n\nMention: \"CRSP\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The domestic counterpart for banks and firms.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006149:7:0", "occurrence_id": "0e639aec5a80e3103e518731", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: fulfilling the necessary data requirements for our analysis do not seem to bias the representation\n\n\nof the final sample, which covers 36 host countries. The data set used in the analysis also\n\n\nincludes stock market prices for all other banks and firms in the home and host countries that are\n\n\nused to construct default risk control variables in the regressions.\n\n\nWe use daily stock market information from Compustat Global for international banks\n\n\nand firms and stock market information from CRSP for U.S. banks and firms. Bank level\n\n\nvariables are constructed from Bankscope, a comprehensive commercial database of banks'\n\n\nfinancial statements produced by Bureau Van Dijk. Since we are interested in how bank\n\n\ncharacteristics affect the correlation between parents’ and their foreign subsidiaries’ default risks,\n\n\nand since there have been significant changes to bank balance sheets during the crisis, we\n\n\nconstruct and use bank-level variables measured prior to the crisis (as of December 2006). For\n\n\neach bank, we calculate relative bank _size_ (bank assets to total system assets), _capital_ _ratio_\n\n\n(regulatory capital to risk weighted assets), _equity ratio_ (equity to total assets), _provisions_ (loan\n\n\nloss provisions divided by total loans), _deposit_ _funding_ (deposits divided by total funding),\n\n\n_profitability_ (net income divided by total assets), and _liquidity_ (liquid assets divided by total\n\n\nassets). <sup>5</sup> [^5: Note that in the ratios mentioned here assets and equity are measured as book values. We also tried running\nregressions using the market value of assets and equity computed from the Merton model discussed above and found\nsimilar results.] We winsorize all financial variables at the 1 <sup>st</sup> and 99 <sup>th</sup> percentile level of their\n\n\ndistributions to reduce the influence of outliers and potential data errors.\n\n\nHost country-level variables are collected from a number of sources. We use data from\n\n\nthe World Bank 2007 Bank Regulation and Supervision survey to construct different indexes of\n\n\nbank regulation and supervision, following the methodology proposed\n\nMention: \"Bankscope\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"A comprehensive commercial database of bank financial statements.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006161:40:0", "occurrence_id": "7782b1259cdd164a0face4c9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: the number of FDI projects. The positive FDI-specific effects of financial development are the strongest at\n\n\nthe intensive margin of FDI. On the other hand, the positive general effects of financial development are the\n\n\nstrongest at the extensive margin of FDI.\n\n\nThese observations explain why we have found in column (3) that the contribution of the direct effects to\n\n\nthe overall positive effects of SFD and DFD on the relative volume of bilateral FDI in financially vulnerable\n\n\nsectors is much larger than that of the indirect effects. They also underline that a fall in output created by a\n\n\nnegative financial shock is of little importance relative to tighter credit conditions. Finally they indicate that the\n\n\nrelative response of FDI to changes in financial development is likely to be much larger at the intensive margin\n\n\nof FDI than at the extensive margin of FDI. In the context of international trade, Manova (2013) reaches similar\n\n\nconclusions.\n\n\n**5.2.5** **Financial development, FDI, and international trade**\n\n\nWe end this section by investigating whether SFD and DFD influence exports and FDI, two alternative modes\n\n\nof internationalization, in the same way. We use as dependent variable the sector-specific value of bilateral trade\n\n\nin columns (10)-(12) of Table 4. Data come from the CEPII _Trade and Production Database_ . In columns (10)\n\n\nand (11), we use trade data for the 23 ISIC sectors listed in Table 5 (Appendix A). In column (12), we aggregate\n\n\nthe data into 13 broad sectors, in the same way as we have done for greenfield FDI. To ensure that the trade\n\n\nresults can be compared with our FDI results, we restrict the sample to those country-pair-year observations for\n\n\nwhich we also observed FDI.\n\n\nOn the source side, our results mirror those of Manova (2013). Higher SFD increases the relative volume\n\n\nof bilateral exports in financially vulnerable sectors and the trade-specific\n\nMention: \"CEPII _Trade and Production Database_\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Bilateral trade values across 23 sectors come from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006166:28:28", "occurrence_id": "e84905a42fa9d16b8cad4c86", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 28.54<br>62.41<br>90.95<br>38.43<br>33.10<br>71.53<br>32.30<br>44.17<br>76.47<br>36.85<br>36.10<br>72.95|100<br>100<br>100<br>100<br>**100**|100<br>100<br>100<br>100<br>**100**|\n\n\nNote: For the conversions of occupations to labor categories see note in Table 1. “Labor intensity” is the percentage weight of the number of workers of each category in the total number of\nworkers. The percentages in the columns “Young workers” and “Adult (females+males)” add up to 100 as reflected in the column “Total number of workers”, since they represent all the\nlabor available for production in the model, excluding the labor allocated to subsistence. In turn, the values in the column “Adult (females+males)” are the sum of the values in columns\n“All females” and “All males”.\n\nSource: Author's calculations based on NBS (2002a).\n\n\n27\n\nMention: \"NBS (2002a)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Calculations by sex are computed from this statistical yearbook.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006172:26:1", "occurrence_id": "331374cf16f1c4395b13a93f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  suggests that the relationship\n\n\nbetween our instruments and future land reallocation risk is not contaminated by other decisions of\n\n\nvillage leaders that may affect individual decisions over productive investments or migration.\n\n\n24The history of specific public goods investments is enumerated Table F4 of the Supplemental 2004 Village\nGovernance Survey.\n\n\n_Land Tenure Insecurity and Migration, Page 25_\n\nMention: \"2004 Village\nGovernance Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its supplemental module enumerates the history of public goods investments.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006176:21:0", "occurrence_id": "d965b00e1ca2bf152dc23e32", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |Col1|expected negative effects of higher levels of<br>aid dependency (Brautigam and Knack 2004)|Col3|Col4|\n|---|---|---|---|\n|<br>**Political/institutional variables **|<br>**Political/institutional variables **|<br>**Political/institutional variables **|<br>**Political/institutional variables **|\n|**Political**<br>**stability**|Greater political stability may facilitate<br>sustained improvements in institutions,<br>including PFM systems|Worldwide<br>Governance<br>Indicators (WGI)|Political stability indicator<br>(rescaled: 1=least stable,<br>10=most stable)|\n|**Regime**|<br>In more democratic regimes, citizens are<br>expected to have more opportunities to hold<br>governments to account, creating an<br>incentive for politicians to seek PFM<br>improvements (de Renzio 2009)|<br>Freedom House<br>|<br>Sum of Civil Liberties and<br>Political Rights Index (converted<br>to: 1=least democratic,<br>10=most democratic)|\n|**Programmat**<br>**ic Parties**<br>|<br>Programmatic parties are expected to be able<br>to make credible commitments about<br>improving public service delivery; and as part<br>of this should have an incentive and<br>capabilities to improve PFM systems.|DPI, Beck et. al<br>2001  and Cruz &<br>Keefer, 2010|Fraction of parties in a country<br>\n\nMention: \"Political Rights Index\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Summed with civil liberties and rescaled so ten is most democratic.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006188:12:2", "occurrence_id": "69ca85faf41ab414f7dc8feb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: . In Tables 1 and 2, we report models using the non-logged oil variable without excluding\noutliers. In the Appendix, we show results using the logged oil variable, including replications of the results of Ross\n(2012).\n17 We prefer scaling military spending by total population because the GDP data from sources other than the COW\nare not in the same metric. For comparison, in our summary Table 1 we scale military spending with GDP data from\nthe Penn World Table 7.1.\n18 Education spending, for instance, has significantly more missing values than health and social security spending. 11\n\nMention: \"COW\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its output metric is the only one consistent with the spending series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006188:12:2", "occurrence_id": "a6b571a90b5d0191d0a33c79", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: . In Tables 1 and 2, we report models using the non-logged oil variable without excluding\noutliers. In the Appendix, we show results using the logged oil variable, including replications of the results of Ross\n(2012).\n17 We prefer scaling military spending by total population because the GDP data from sources other than the COW\nare not in the same metric. For comparison, in our summary Table 1 we scale military spending with GDP data from\nthe Penn World Table 7.1.\n18 Education spending, for instance, has significantly more missing values than health and social security spending. 11\n\nMention: \"Penn World Table\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Version 7.1 output is used to scale military spending in the summary table.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006195:25:0", "occurrence_id": "21eac065a516f1bb53f647d1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: housing expenditures for three African cities (Table 4) suggest that housing expenditure is less responsive\nin these cities than in other developing countries (see Malpezzi and Mayo 1987).\n\nMalpezzi and Mayo (1987) review previous housing demand studies in over a dozen developing country\ncities and find that income elasticities are fairly consistent across countries, and in most cases fall\nbetween 0.5 and 1, with renters generally have slightly lower income elasticities than owners. <sup>31</sup> [^31: Likewise, Struyk, et al. (1990) find a similar difference between renters and owners in Indonesia and also note that income\nelasticities decrease with city size.] These\nestimates however, were mostly for Latin American and East Asian countries/cities, with little to no\nrepresentation of low-income or SSA countries. Using the SHIP data to estimate income elasticity of\nhousing expenditures in Kigali, Abidjan, and Maputo yields coefficients on log income ranging from 0.15\nand 0.37 (Table 4). Interestingly, early estimates of the income elasticity for low income housing in the\nUS are in a similar range: 0.19 for renters and 0.45 for owners (Mulford 1979). The estimated elasticities\nsuggest that a 10 percent increase in income will only lead to increases in housing expenditures ranging\nbetween 1.5 and 3.6 percent. Cash transfers to households would not lead to increased expenditures in\nhousing unless strongly earmarked.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n|Col1|Kigali|Col3|Abidjan|Col5|Maputo|Col7|\n|---|---|---|---|---|---|---|\n|Log Income<br>|0.162<br>(.016)***<br>|0.159<br>(.016)***<br>|0.150<br>(.015)***<br>|0.150<br>\n\nMention: \"SHIP data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Housing expenditure elasticities are estimated for three cities from these data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006196:5:1", "occurrence_id": "ec74b7aaf046b49578515251", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  procyclical if the CCI is positive and significantly different from zero at the\n10 percent level; countercyclical if it is negative and significantly different from zero at the 10\npercent level;  acyclical if it belongs to the residual category of currencies with insignificant CCI. To\ncheck the robustness of the significance threshold, we use a 15 percent confidence interval, as well\nas a one tailed test, and find the classifications according to the different significance criteria to be\nvery similar (Appendix Table C.1). We also calculate the CCI using the real effective exchange rate\nand find a large overlap with the one calculated using the nominal effective exchange rate. <sup>8</sup> The CCI\ncalculated using NEER is presented in Figure 1a, and the one using REER is in Figure 1b.\n\n\n6 We run the HP filter (with lambda equal to 1600) after seasonally adjusting the series using Census X11.\n\n7 Since nominal exchange rates are relative prices we have a _numéraire_ problem and we lose a degree of freedom, which in\nprinciple could affect the interpretation of the results. However the problem is not very severe in our analysis, which is\nbased on nominal effective exchange rate with weights that differ across countries. We would like to thank Andrew\nPowell for raising this important point.\n\n8 The Spearman rank correlation between NEER and REER procyclicality measures is 0.86.\n\n\n4\n\nMention: \"Census X11\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"The procedure used to seasonally adjust the series before filtering.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006181:48:0", "occurrence_id": "747274ab0bccf452d3bed884", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: [29] Itskhoki, Oleg and Elhanan Helpman (2014). “Firms, Trade and Labor Market Dynam\n\nics.” Mimeo, Princeton University.\n\n\n[30] Jones, Ronald I. (1971). “A Three-factor Model in Theory, Trade and History.” In\n\n\n_Trade,_ _Balance_ _of_ _Payments_ _and_ _Growth:_ _Essays_ _in_ _Honor_ _of_ _C.P._ _Kindleberger_, ed.\n\n\nby Bhagwati et. al. Amsterdam: North Holland.\n\n\n[31] Jovanovic, Boyan and Robert Moffitt (1991). “An Estimate of a Sectoral Model of Labor\n\n\nMobility,” _Journal_ _of_ _Political_ _Economy_, 98:4, pp. 827-852.\n\n\n[32] Karp, Larry and Thierry Paul (1994). “Phasing In and Phasing Out Protectionism with\n\n\nCostly Adjustment of Labour,” _Economic_ _Journal_ 104:427 (November), pp. 1379-92.\n\n\n[33] King, Miriam, Steven Ruggles, J. Trent Alexander, Sarah Flood, Katie Genadek,\n\n\nMatthew B. Schroeder, Brandon Trampe, and Rebecca Vick (2010). “Integrated Public\n\n\nUse Microdata Series, Current Population Survey: Version 3.0.” Minneapolis: Univer\n\nsity of Minnesota.\n\n\n[34] Kovak, Brian (2013). “Local Labor Market Effects of Trade Policy: Evidence from\n\n\nBrazilian Liberalization.” _American_ _Economic_ _Review_ 103:5, pp. 1960-1976.\n\n\n[35] Magee, Stephen P. (1989). “Three Simple Tests of the Stolper-Samuelson Theorem.”\n\n\nCh. 7 in Stephen P. Magee, William A. Brock, and Leslie\n\nMention: \"Integrated Public\n\n\nUse Microdata Series\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"A harmonised microdata series published by a university population centre.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006181:48:0", "occurrence_id": "a17055c5806d52b401b31da8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: [29] Itskhoki, Oleg and Elhanan Helpman (2014). “Firms, Trade and Labor Market Dynam\n\nics.” Mimeo, Princeton University.\n\n\n[30] Jones, Ronald I. (1971). “A Three-factor Model in Theory, Trade and History.” In\n\n\n_Trade,_ _Balance_ _of_ _Payments_ _and_ _Growth:_ _Essays_ _in_ _Honor_ _of_ _C.P._ _Kindleberger_, ed.\n\n\nby Bhagwati et. al. Amsterdam: North Holland.\n\n\n[31] Jovanovic, Boyan and Robert Moffitt (1991). “An Estimate of a Sectoral Model of Labor\n\n\nMobility,” _Journal_ _of_ _Political_ _Economy_, 98:4, pp. 827-852.\n\n\n[32] Karp, Larry and Thierry Paul (1994). “Phasing In and Phasing Out Protectionism with\n\n\nCostly Adjustment of Labour,” _Economic_ _Journal_ 104:427 (November), pp. 1379-92.\n\n\n[33] King, Miriam, Steven Ruggles, J. Trent Alexander, Sarah Flood, Katie Genadek,\n\n\nMatthew B. Schroeder, Brandon Trampe, and Rebecca Vick (2010). “Integrated Public\n\n\nUse Microdata Series, Current Population Survey: Version 3.0.” Minneapolis: Univer\n\nsity of Minnesota.\n\n\n[34] Kovak, Brian (2013). “Local Labor Market Effects of Trade Policy: Evidence from\n\n\nBrazilian Liberalization.” _American_ _Economic_ _Review_ 103:5, pp. 1960-1976.\n\n\n[35] Magee, Stephen P. (1989). “Three Simple Tests of the Stolper-Samuelson Theorem.”\n\n\nCh. 7 in Stephen P. Magee, William A. Brock, and Leslie\n\nMention: \"Current Population Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The underlying survey behind this harmonised version.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006197:3:1", "occurrence_id": "993ec90a011befb2baece777", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  and investment in irrigation (World Bank 2008).\n\nThe literature on agricultural intensification in Africa developed significantly in the 1980s\n\n\n\nand so have fertilizer use, mechanization and investment in irrigation (World Bank 2008).\n\nThe literature on agricultural intensification in Africa developed significantly in the 1980s\nand has resumed over the past decade. It generally finds that in most areas studied intensification\nhas progressed along the lines predicted by Boserup and Ruthenberg, and that agricultural\ninvolution is confined to few areas. These studies have been done using samples of locations for\nintensive studies. Under the Intensive Agricultural Surveys (ISA) that have been imbedded in\nbroader Living Standard Measurement Studies (World Bank, 2009), national household data for six\nAfrican countries is becoming available that contains the intensification and technology variables as\nwell as income. These panel data will over time present a unique opportunity to test the BR model\nin different countries and different agroclimate zones, and to see whether there are signs of\ninvolution. At the present cross section data are available for six countries, while panel data are or\nwill become available for all countries over the next few years. As we will show, only panel data\nover a sufficient period of time will be able to rigorously test the BR theory.\n\nIn this paper we are setting the stage for such later studies by describing the status of\n\n\n\nover a sufficient period of time will be able to rigorously test the BR theory.\n\nIn this paper we are setting the stage for such later studies by describing the status of\nintensification across the six countries (Ethiopia, Malawi, Niger, Nigeria Uganda and Tanzania).\nFollowing Binswanger et al. (1993) we also study the impact of the agro-ecological endowment and\naccess to markets of the environments in which the farmers live on population density,\ninfrastructure and agricultural variables including output, and use of modern inputs. We construct\na single variable for each of these, the agro-ecological potential derived from the potential yield\nestimates of To\n\nMention: \"Intensive Agricultural Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Location samples embedded in broader household studies.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006197:3:1", "occurrence_id": "47aa5d452e4096ebf2845666", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  and investment in irrigation (World Bank 2008).\n\nThe literature on agricultural intensification in Africa developed significantly in the 1980s\n\n\n\nand so have fertilizer use, mechanization and investment in irrigation (World Bank 2008).\n\nThe literature on agricultural intensification in Africa developed significantly in the 1980s\nand has resumed over the past decade. It generally finds that in most areas studied intensification\nhas progressed along the lines predicted by Boserup and Ruthenberg, and that agricultural\ninvolution is confined to few areas. These studies have been done using samples of locations for\nintensive studies. Under the Intensive Agricultural Surveys (ISA) that have been imbedded in\nbroader Living Standard Measurement Studies (World Bank, 2009), national household data for six\nAfrican countries is becoming available that contains the intensification and technology variables as\nwell as income. These panel data will over time present a unique opportunity to test the BR model\nin different countries and different agroclimate zones, and to see whether there are signs of\ninvolution. At the present cross section data are available for six countries, while panel data are or\nwill become available for all countries over the next few years. As we will show, only panel data\nover a sufficient period of time will be able to rigorously test the BR theory.\n\nIn this paper we are setting the stage for such later studies by describing the status of\n\n\n\nover a sufficient period of time will be able to rigorously test the BR theory.\n\nIn this paper we are setting the stage for such later studies by describing the status of\nintensification across the six countries (Ethiopia, Malawi, Niger, Nigeria Uganda and Tanzania).\nFollowing Binswanger et al. (1993) we also study the impact of the agro-ecological endowment and\naccess to markets of the environments in which the farmers live on population density,\ninfrastructure and agricultural variables including output, and use of modern inputs. We construct\na single variable for each of these, the agro-ecological potential derived from the potential yield\nestimates of To\n\nMention: \"Living Standard Measurement Studies\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"National household data for six African countries are becoming available.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006203:17:1", "occurrence_id": "a2e4aaad547a37fca3fcbb39", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: </sup> <sup>_,jlocal_</sup> <sup>_ODipsc ×_</sup> <sup>ln (</sup> <sup>_densitypsc_</sup> <sup>)</sup> <sup>_j_</sup> <sup>+</sup>\n\n\n_β_ 4 _household_ _ODipsc_ + _Xipscθ_ + _αpsc_ + _εipsc._\n\n\n\n(4)\n\n\n\nAs before, we are estimating health outcomes for child _i_, in region _p_, in DHS survey _s_, and\n\n\nin country _c_ . As described above, we introduce fixed effects _α_ at country, survey, and region\n\n\nlevels, in stages. We also include the same vector of extended controls _X_, as well as the\n\n\nhousehold’s own open defecation, as described above.\n\n\nThis functional form implies that the change in health associated with a change from 0%\n\n\nto 100% local open defecation is:\n\n\n\n_<u>∂health</u>_ <sup>�</sup>\n_∂local_ _OD_ <sup>=</sup> <sup>_β_</sup> <sup>ˆ1 +</sup>\n\n\n\n5\n\n\n\n_j_ =1\n\n\n\n_β_ ˆ3 _,j_ ln ( _densitypsc_ ) <sup>_i_</sup> (5)\n\n\n\nPanel A of figure 1 plots the dependence of the infant mortality-open defecation gradient\n\n\non population density; panel B plots the height-open defecation gradient as a function of\n\n\npopulation density. In both cases, the same six specifications that were used in section 3.3\n\nMention: \"DHS survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Health outcomes are estimated within survey and country fixed effects.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006205:16:0", "occurrence_id": "0c46f6f50f18f9a2afc348f1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The following four sections examine each channel in turn. For each channel, we discuss the literature that\ndescribes the channel, and also review what we know of the exposure, vulnerability, and ability to adapt of\nthe poor and vulnerable. Some determinants are very specific to a given channel; for instance, there is no\nreason for the exposure of the poor and vulnerable to food price changes to be similar to the exposure to\nasset losses from floods.\n\n\nBut some determinants are the same for all channels. For instance limited financial inclusion means that\nmany poor people cannot save in a diversified portfolio and are more vulnerable to a single localized shock,\nregardless of its origin. According to the World Bank’s FINDEX database (FINDEX, 2011), more than 20% of\nIndonesian households in the top 60% of the income distribution saved at a financial institution in 2011 vs.\nless than 8% for the bottom 40%. <sup>20</sup> [^20: This pattern is consistent across countries. In 159 out of 164 countries with data in 2011, the top 40% had higher\nproportion of households saving at a financial institution compared to the bottom 60%, with the average difference\n10%.] This difference suggests poor households are likely to have more of\ntheir assets and savings localized in or around their dwelling, and thus lose a larger fraction of their income\nin case of an asset-damaging disaster such as floods.\n\n\nIn addition, poorer households often have more limited access than the non-poor to social protection and\nsafety nets after disasters, which makes them more vulnerable to any sort of shock. According to the World\nBank’s ASPIRE database (ASPIRE, 2014), the average per capita transfer received by the extreme poor from\nsocial protection is much lower than the transfer received by the richest quintile. In Malawi, those in the\npoorest quintile receive on average 0.5 cents per day, while the richest 20% receive more than 17 cents. In\nVietnam, transfers are\n\nMention: \"FINDEX database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Household account ownership by income group comes from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006205:16:0", "occurrence_id": "ba339ea729fab84d08791295", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The following four sections examine each channel in turn. For each channel, we discuss the literature that\ndescribes the channel, and also review what we know of the exposure, vulnerability, and ability to adapt of\nthe poor and vulnerable. Some determinants are very specific to a given channel; for instance, there is no\nreason for the exposure of the poor and vulnerable to food price changes to be similar to the exposure to\nasset losses from floods.\n\n\nBut some determinants are the same for all channels. For instance limited financial inclusion means that\nmany poor people cannot save in a diversified portfolio and are more vulnerable to a single localized shock,\nregardless of its origin. According to the World Bank’s FINDEX database (FINDEX, 2011), more than 20% of\nIndonesian households in the top 60% of the income distribution saved at a financial institution in 2011 vs.\nless than 8% for the bottom 40%. <sup>20</sup> [^20: This pattern is consistent across countries. In 159 out of 164 countries with data in 2011, the top 40% had higher\nproportion of households saving at a financial institution compared to the bottom 60%, with the average difference\n10%.] This difference suggests poor households are likely to have more of\ntheir assets and savings localized in or around their dwelling, and thus lose a larger fraction of their income\nin case of an asset-damaging disaster such as floods.\n\n\nIn addition, poorer households often have more limited access than the non-poor to social protection and\nsafety nets after disasters, which makes them more vulnerable to any sort of shock. According to the World\nBank’s ASPIRE database (ASPIRE, 2014), the average per capita transfer received by the extreme poor from\nsocial protection is much lower than the transfer received by the richest quintile. In Malawi, those in the\npoorest quintile receive on average 0.5 cents per day, while the richest 20% receive more than 17 cents. In\nVietnam, transfers are\n\nMention: \"ASPIRE database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Average transfers received by the extreme poor come from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006210:17:0", "occurrence_id": "4e429e9953b8628b57a67908", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: range from Basic Facts, Territorial Organization, Local Democracy, Central–Local Government\nRelations and Local Responsibilities to Local Finance. To access a country profile, one has to\nclick on the region and choose a country. After selecting a region, click on the map to select a\nspecific country profile or access the profile of the entire region.\n\n\nOur Selection and Best Practices contains a list of websites on local democracy and\ngovernance which can be country specific or global in scope. This section lists some useful\nwebsites including the CLGF website and other important sites, some of which are listed here.\n\n\nKey Indicators on Local Governments is a really handy database or “pocketbook” with\ninformation on 82 countries in a wide range of topics such as population data, macroeconomics,\nterritorial organization and public finance. To access data on a specific country, click on the\nregion to select the country. Data for the selected country is then displayed in PDF files.\n\n\nAnother great source of information on local democracy is the **Commonwealth Local**\n**Government Forum (CLGF)** [: http://www.clgf.org.uk/index.cfm. CLGF seeks to promote and](http://www.clgf.org.uk/index.cfm)\nstrengthen democratic local government across the Commonwealth countries and to encourage\nthe development of democratic values and good governance. CLGF has been actively involved in\nencouraging and developing local elections, election monitoring, and capacity building support\nfor councilors and councils.\n\n\nCLGF contains a repository of information on local government in different countries\nwhich can be accessed through the Country Profiles tab at the top. This section includes a brief\nsummary of each country, key facts such as percentage of local government contribution to GDP,\nstructure of the local government, dates of most recent elections and links to other useful sites.\nThe first section entitled Summary contains a concise summary of the form, scope,\nresponsibilities and structure of local\n\nMention: \"Key Indicators on Local Governments\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"A pocketbook database covering 82 countries, listed among useful sites.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006211:2:0", "occurrence_id": "df91dd7ca38c87373bac17d4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Transparent Government and Business Regulation: “Open for Business?”**\n\n\nCarolin Geginat and Valentina Saltane <sup>1</sup>\n\n\n**JEL:** O Economic Development, Technological Change, and Growth\n\n\nKeywords: Transparency, Governance, Access to Information, Regulatory Transparency Index\n\n\n1 [2121 Pennsylvania Avenue NW, Washington, DC 20433, cgeginat@worldbank.org and vsaltane@worldbank.org.](mailto:cgeginat@worldbank.org)\nThis paper would not have been possible without the extensive research of the team who collects the annual Doing\nBusiness data at the World Bank Group and the 9,600 private contributors who answered to the surveys of the Doing\nBusiness Project in 2012. We would like to thank Sarmad Shaikh for his excellent research assistance. The full data\nset for each country can be found at www.doingbusiness.org. We would also like to thank -- without implication –\nAdrian Gonzalez, Aart Kraay, Alejandro Ponce, Rita Ramalho, Sylvia Solf and Hulya Ulku for helpful discussions\nin the preparation of this paper. The authors thank the United States Agency for International Development for its\nfinancing contribution to the data collection efforts.\n\nMention: \"annual Doing\nBusiness data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Collected annually with contributions from 9,600 private respondents.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006214:36:4", "occurrence_id": "5b5468ff2cb32280ac3bf8ed", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: MFN\n<u>rates) 100 % coverage</u>\n\nWITS&WTO (non-MFN\n<u>rates) 100 % coverage</u>\n\n\nWITS&WTO (non-MFN\n<u>rates) 99.98 % coverage</u>\n\n\nWITS&WTO (non-MFN\n<u>rates) 100 % coverage</u>\n\n\n\nWITS&WTO (non-MFN\n<u>rates) 100 % coverage</u> <u>WITS&WTO (MFN rates) 100 % coverage</u>\n\n\n\nWITS&WTO (non-MFN\n<u>rates) 100 % coverage</u>\n\n\nWITS&WTO (non-MFN\n<u>rates) 100 % coverage</u>\n\n\n\nWITS&WTO (non-MFN\n<u>rates) 100 % coverage</u> <u>WITS&WTO (MFN rates) 99.92 % coverage</u>\n\n\n\nWITS&WTO (non-MFN\n<u>rates) 100 % coverage</u>\n\n\nWITS&WTO (non-MFN\n<u>rates) 100 % coverage</u>\n\n\n\nWITS&EUROSTAT 90.64 % coverage;\nWITS&WTO (MFN rates) 7.27 % coverage;\n<u>WITS&GTAP 2.08 % coverage</u>\n\n\nWITS&EUROSTAT 99.42 % coverage;\n<u>WITS&WTO (MFN rates) 0.56 % coverage</u>\n\n\n\nNote: Unless specified otherwise, all information from WITS refers to imports for 2007.\n\n\n35\n\nMention: \"WITS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Unless otherwise noted, its figures refer to imports in this year.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006224:17:0", "occurrence_id": "a2207db78eca0b1b799ea8e1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 9. Growth of hourly wages in Brazil, Mexico and Peru, 2002-2012\n\n\na. Brazil b. Mexico\n\n\n<u>c.</u> <u>Peru</u>\n\n\nSource: Authors’ calculation from SEDLAC (CEDLAS and the World Bank). Note: The figures report the growth of\nhourly wage for the bottom, fifth and top decile of labor income, indexed as a proportion.\n\nThe growth rates seen across the income distributions of Brazil and Peru are consistent with the\nfalling skill premiums in the region, one of the main drivers of the reduction in inequality during the\nfirst decade of the 2000s (Gasparini et al. 2008; World Bank 2011; Lustig and Lopez-Calva 2010; and\nLustig et al. 2013). Since 2010, however, the skill premium’s rate of reduction has fallen, and in some\ncountries, reversed trends and increased (Table 1 and Table A3 in the annex). <sup>24</sup> Between 2003 and\n2010, most countries in Latin America saw a narrowing of the skill premium (measured using the\nhourly wage gap) between high skill and low skill workers. However, between 2010 and 2012, six out\nof the 15 countries showed increases in the skill gap, including four of six countries in the Central\nAmerica and Mexico region. Moreover, the gap between tertiary and secondary education increased in\neight countries, including in all of the countries of Central America, Mexico and Chile. This suggests a\nreduction in the relative returns to finishing secondary school and a widening gap between those with\nonly a secondary education and those with tertiary education in Mexico and Central America.\n\n\n24 Skill premium estimates are obtained from Mincer equations of log hourly wages on education levels, controlling for age\nand age squared.\n\n\n15\n\nMention: \"SEDLAC\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Hourly wage growth for the bottom decile is calculated from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006229:22:2", "occurrence_id": "bc5b593baee4bc5cc1810980", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: >486965<br>37720<br>43875<br>66981<br>56755<br>11024<br>56380<br>608502<br>497482<br>92464<br>86161<br>3290<br>2437379<br>51519<br>469584<br>1077928<br>94787<br>201710<br>578591<br>14664<br>286<br>31613<br>4776<br>407953<br>1225403<br>6931<br>163573<br>41845<br>53186<br>381252<br>357424<br>89568<br>9485<br>134905<br>1125383<br>58046<br>47108<br>52061<br>16553<br>350353<br>86574<br>71955<br>645028<br>108560<br>815154<br>34521<br>578633<br>3268006<br>35580<br>919326|\n\n\n\nNote: Albania and Israel excluded from the sample. For Israel we do not\ncount with data on pre-school attendance. For Albania we do not count\nwith data on economic, social and cultural status. Hong-Kong, Chinese\nTaipei, Macao and Shanghai are excluded.\nSource: Authors’ calculations based on data from the OECD, PISA 2012.\n\n\n\n21\n\nMention: \"data from the OECD\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Achievement calculations rest on this assessment cycle.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006230:16:0", "occurrence_id": "934de89140ce6cc36deee6c8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: (e.g., graduate studies). Therefore, I also introduce a lag of the dependent variable as a regressor to\ncontrol for serial correlation when analyzing the level of educational attainment. <sup>13</sup>\n\n\nFigure 3: Educational attainment vs. labor income Gini\n\n\n\n(a)\n\n\n\n(b)\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nSource: Author’s calculations based on data from Latin American household surveys, 1995-2009, biannual\nand Polity IV Project.\n\n\nSimilar to Table 1, in Table 2 each duplet of columns corresponds to computing Equation (2)\n\nusing 0.01, 0.03, 0.06 and 0.10 as values for _r_ respectively. Our results are consistent with our prior:\n\ndemocracy is related to the level of educational attainment. (Note that _β_ is biased downwards.)\n\nFurthermore, in columns (2), (4), (6) and (8), democracy and educational attainment show a slight\n\nnon-monotonic relationship.\n\n\nNow, let us consider the results of columns (2), (4), (6) and (8). A cohort with no existing\n\nstock of democratic capital experiences the following democratic educational effect: for each full\n\ndecade of high-quality democracy (Polity2 = 10), democracy stock increases by approximately 100\n\npoints. Considering panels A and B, I find that the predicted impact on educational attainment of\n\na decade of high-quality democracy is approximately between 0.09 and 0.19 years of education.\n\nConsidering that the average educational attainment is 5.6 years, having 20 years of high quality\n\ndemocratic institutions would increase educational attainment (on average) by 5%.\n\n\n13Verbeek and Vella (2005) show that if grouping variables are exogenous and relevant and cohorts sizes are large\nenough, it is possible to obtain and appropriate measure of _Y c,g,p,e−_ 1, equivalent to the lagged value of _\n\nMention: \"Latin American household surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Biannual distributional calculations across the region come from these surveys.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006230:16:0", "occurrence_id": "08cbe844dc6964b17ae48502", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: (e.g., graduate studies). Therefore, I also introduce a lag of the dependent variable as a regressor to\ncontrol for serial correlation when analyzing the level of educational attainment. <sup>13</sup>\n\n\nFigure 3: Educational attainment vs. labor income Gini\n\n\n\n(a)\n\n\n\n(b)\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nSource: Author’s calculations based on data from Latin American household surveys, 1995-2009, biannual\nand Polity IV Project.\n\n\nSimilar to Table 1, in Table 2 each duplet of columns corresponds to computing Equation (2)\n\nusing 0.01, 0.03, 0.06 and 0.10 as values for _r_ respectively. Our results are consistent with our prior:\n\ndemocracy is related to the level of educational attainment. (Note that _β_ is biased downwards.)\n\nFurthermore, in columns (2), (4), (6) and (8), democracy and educational attainment show a slight\n\nnon-monotonic relationship.\n\n\nNow, let us consider the results of columns (2), (4), (6) and (8). A cohort with no existing\n\nstock of democratic capital experiences the following democratic educational effect: for each full\n\ndecade of high-quality democracy (Polity2 = 10), democracy stock increases by approximately 100\n\npoints. Considering panels A and B, I find that the predicted impact on educational attainment of\n\na decade of high-quality democracy is approximately between 0.09 and 0.19 years of education.\n\nConsidering that the average educational attainment is 5.6 years, having 20 years of high quality\n\ndemocratic institutions would increase educational attainment (on average) by 5%.\n\n\n13Verbeek and Vella (2005) show that if grouping variables are exogenous and relevant and cohorts sizes are large\nenough, it is possible to obtain and appropriate measure of _Y c,g,p,e−_ 1, equivalent to the lagged value of _\n\nMention: \"Polity IV Project\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its regime measures accompany the distributional data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006231:24:0", "occurrence_id": "2b4e86cb4937f965789460c7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **8.** **APPENDIX 1:**\n\n\n**Table A.1: Local Budget module of the Life in Kyrgyzstan survey for 2012**\n\n**8.D. LOCAL BUDGET**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n|Now I would like to ask your opinion and involvem government authorities|ment in the process of budget decision of local|\n|---|---|\n|<br>How do you think, how informed are you about the<br>I828    budget of your local government body (Aiyl Aimak<br>or municipal government)?|Very well                                                                                  1|\n|<br>How do you think, how informed are you about the<br>I828    budget of your local government body (Aiyl Aimak<br>or municipal government)?|Well                                                                                           2|\n|<br>How do you think, how informed are you about the<br>I828    budget of your local government body (Aiyl Aimak<br>or municipal government)?|Badly                                                                                        3|\n|<br>How do you think, how informed are you about the<br>I828    budget of your local government body (Aiyl Aimak<br>or municipal government)?|Absolutely not informed                                                            4|\n|Please estimate, how much it is interesting to you,<br>how your local government body (Aiyl Aimak or<br>municipal<br>I829<br>government) spends your money from taxes and<br>payments?<br>|Very interesting                                                                        1|\n|Please estimate, how much it is interesting to you,<br>how your local government body (Aiyl Aimak or<br>municipal<br>I829<br>government) spends your money from taxes and<br>payments?<br>|Interesting                                                                                 2|\n|Please estimate, how much it is interesting to you,\n\nMention: \"Life in Kyrgyzstan survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its local budget module is reproduced in the appendix.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006233:3:0", "occurrence_id": "c0adfda1093cef5a38264f97", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: #### 1 Introduction\n\nWhen the US granted duty-free and quota-free access to Madagascar, under the African\n\n\nGrowth and Opportunity Act 2000, exports from Madagascar exploded, from $170 million\n\n\nin 2000 to $500 million in 2004. Over the same period, Madagascar’s exports to the rest of\n\n\nthe world also increased, from $750 million to $875 million. Similarly, based on Comtrade\n\n\nexport data, when the EU granted duty-free and quota-free access to Bangladesh under the\n\n\nEverything But Arms Initiative in 2001, knitwear exports from Bangladesh to the EU more\n\n\nthan doubled, rising from $823 million to $2,351 million between 2000 to 2004. During\n\n\nthe same period of time, knitwear exports from Bangladesh to the US increased from $316\n\n\nmillion to $465 million. <sup>1</sup> Exports to countries other than the US and EU rose from $60\n\nmillion to $190 million. <sup>2</sup> To the surprise of many, such generous trade preferences resulted\n\n\nnot in trade diversion from the rest of the world to the preference granting markets, but in\n\n\ntrade creation - tempered by the presence of quotas.\n\n\nThe model we develop and estimate in this paper predicts exactly these changes. Trade\n\n\npreferences given by one country have positive spillovers on exports to others in the presence\n\n\nof free entry. For example, preferences given by the EU make the industry more attractive,\n\n\ninduce entry, and some of these entering …rms export to the US so that exports to both\n\n\ncountries rise. We use customs data from Bangladesh to estimate a heterogeneous …rm\n\n\nmodel based on the ‡agship model of Melitz (2003), but structured to be suitable for trade\n\n\npolicy applications. Our work takes a heterogeneous …rm model literally, and confronts\n\n\nit with micro data and actual trade policies, to estimate all of its structural parameters,\n\n\nincluding the various levels of …\n\nMention: \"Comtrade\n\n\nexport data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Export values under duty-free access are traced in these data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006233:3:0", "occurrence_id": "69c2db988125e48ff80d4702", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: #### 1 Introduction\n\nWhen the US granted duty-free and quota-free access to Madagascar, under the African\n\n\nGrowth and Opportunity Act 2000, exports from Madagascar exploded, from $170 million\n\n\nin 2000 to $500 million in 2004. Over the same period, Madagascar’s exports to the rest of\n\n\nthe world also increased, from $750 million to $875 million. Similarly, based on Comtrade\n\n\nexport data, when the EU granted duty-free and quota-free access to Bangladesh under the\n\n\nEverything But Arms Initiative in 2001, knitwear exports from Bangladesh to the EU more\n\n\nthan doubled, rising from $823 million to $2,351 million between 2000 to 2004. During\n\n\nthe same period of time, knitwear exports from Bangladesh to the US increased from $316\n\n\nmillion to $465 million. <sup>1</sup> Exports to countries other than the US and EU rose from $60\n\nmillion to $190 million. <sup>2</sup> To the surprise of many, such generous trade preferences resulted\n\n\nnot in trade diversion from the rest of the world to the preference granting markets, but in\n\n\ntrade creation - tempered by the presence of quotas.\n\n\nThe model we develop and estimate in this paper predicts exactly these changes. Trade\n\n\npreferences given by one country have positive spillovers on exports to others in the presence\n\n\nof free entry. For example, preferences given by the EU make the industry more attractive,\n\n\ninduce entry, and some of these entering …rms export to the US so that exports to both\n\n\ncountries rise. We use customs data from Bangladesh to estimate a heterogeneous …rm\n\n\nmodel based on the ‡agship model of Melitz (2003), but structured to be suitable for trade\n\n\npolicy applications. Our work takes a heterogeneous …rm model literally, and confronts\n\n\nit with micro data and actual trade policies, to estimate all of its structural parameters,\n\n\nincluding the various levels of …\n\nMention: \"customs data from Bangladesh\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"A heterogeneous firm model is estimated on these transaction records.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006238:5:1", "occurrence_id": "51c8ac294180e386eebffabe", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  an empirical application for the Peruvian case, and\nfind that market and product diversification increases the survival rate of Peruvian firms in export\nmarkets. They explain the results with a portfolio argument for which firms exporting different\nproducts or in different destinations can take advantage of the non-perfect covariance of sales of\ndifferent products/markets to reduce the overall variability of sales and thus increase the\nlikelihood of survival in export markets.\n\n\nThe literature has also explored the role of network effects in the success of export flows under\nthe hypothesis that other exporters could facilitate entering and surviving into foreign markets for\nother firms of the same country. Cadot et al. (2013) find evidence of ‘network effects’ in\ndetermining export survival chances. The authors find that survival probabilities increase with the\nnumber of same-country firms exporting the same product to the same destination, suggesting\nthe existence of cross-firm informational externalities.\n\n###### **3. Data**\n\nThe analysis is based on a unique data set from Georgian Customs’ export transaction data\nmerged with firm level data obtained from Georgia’s National Statistics Office, Geostat. The firm\nlevel data consists of a panel of Georgian firms spanning the period 2006-2012. This is matched\nwith Customs data recording all export transactions occurred in the same period using a common\nfirm identifier. Large firms (those with more than 100 employees) are all included in the data set\nwhile small and medium enterprises have been randomly sampled. There are 13,816 firm-year\n\n4\n\nMention: \"Georgian Customs’ export transaction data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Merged with firm data from the national statistics office.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006238:5:1", "occurrence_id": "2ff87acb7c62a3afcde5688e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  an empirical application for the Peruvian case, and\nfind that market and product diversification increases the survival rate of Peruvian firms in export\nmarkets. They explain the results with a portfolio argument for which firms exporting different\nproducts or in different destinations can take advantage of the non-perfect covariance of sales of\ndifferent products/markets to reduce the overall variability of sales and thus increase the\nlikelihood of survival in export markets.\n\n\nThe literature has also explored the role of network effects in the success of export flows under\nthe hypothesis that other exporters could facilitate entering and surviving into foreign markets for\nother firms of the same country. Cadot et al. (2013) find evidence of ‘network effects’ in\ndetermining export survival chances. The authors find that survival probabilities increase with the\nnumber of same-country firms exporting the same product to the same destination, suggesting\nthe existence of cross-firm informational externalities.\n\n###### **3. Data**\n\nThe analysis is based on a unique data set from Georgian Customs’ export transaction data\nmerged with firm level data obtained from Georgia’s National Statistics Office, Geostat. The firm\nlevel data consists of a panel of Georgian firms spanning the period 2006-2012. This is matched\nwith Customs data recording all export transactions occurred in the same period using a common\nfirm identifier. Large firms (those with more than 100 employees) are all included in the data set\nwhile small and medium enterprises have been randomly sampled. There are 13,816 firm-year\n\n4\n\nMention: \"Customs data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Recording every export transaction across the panel period.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006242:11:0", "occurrence_id": "16516a98e01c79b5f1b61050", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **3.1. Who are the vulnerable to poverty in Mexico?**\n\n\nIn part because of modest economic growth and the expansion of more progressive social spending,\nMexico faced a reconfiguration of its social groups (mainly) over the last decade. As shown in Figure\n2, the reduction of poverty meant that between 2000 and 2012 the Mexican middle class rose by\nalmost 12 percentage points (from 26 to 38 percent). At the same time though, there has been an\nexpansion of the population in vulnerability in 5.4 percentage points. In fact, in 2012 roughly twothirds of the population in Mexico remained in a situation of economic insecurity: 22.2 percent in\npoverty, and 43 percent in vulnerability—this is the situation of nearly half of the Mexican states\nwhere the sum of poverty and vulnerability ranges from 64.5 percent in the State of Mexico to 84.2\npercent in Chiapas (Figure 3).\n\n\n**Figure 3: Distribution of socioeconomic groups by states**\n\n_Percentage of population; Mexico, 2012_\n\n\n\nNuevo Leon\n\nDistrito Federal\n\nSonora\n\nBaja California\n\nBaja California Sur\n\nColima\n\nCoahuila\n\nQuintana Roo\n\nChihuahua\n\nTamaulipas\n\nSinaloa\n\nJalisco\n\nQueretaro\n\nAguascalientes\n\nCampeche\n\nMorelos\n\nState of Mexico\n\nGuanajuato\n\nNayarit\n\nTabasco\n\nYucatan\n\nSan Luis Potosi\n\nVeracruz\n\nZacatecas\n\nMichoacan\n\nHidalgo\n\nDurango\n\nTlaxcala\n\nOaxaca\n\nPuebla\n\nGuerrero\n\nChiapas\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n0.0 20.0 40.0 60.0 80.0 100.0\n\n< $4.0 $4 - 10 $10 - 50          - $50.0\n\n_Source_ : Author’s estimations based on MPS-ENIGH 2012\n\n\n\n10\n\nMention: \"MPS-ENIGH 2012\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Income band shares are estimated from this module of the household survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006248:32:1", "occurrence_id": "ecd51b0a68e6e7f58b4e3abe", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \nto affect greenness positively, however, since cold also becomes a serious constraint for plant\n\n\ngrowth when moving away from the equator.\n\n\nWhile greenness is best seen from above, rainfall is harder to measure using satellites.\n\n\nThis study uses data from the Tropical Rainfall Measuring Mission (TRMM) which to our\n\n\n31\n\nMention: \"Tropical Rainfall Measuring Mission\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Satellite rainfall estimates, harder to measure than other variables.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006252:6:0", "occurrence_id": "5ede56144d7a37c26770a925", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: was taken due to the public sector’s limited size, and in order to exploit what was perceived to\n\n\nbe existing excess capacity in non-public schools, where student-teacher ratios were believed\n\n\nto be relatively low because parents could not afford to consistently send children to school.\n\n\nIn addition, in areas where schools were already full, it was expected that the tuition waiver\n\n\nincentive would trigger a supply response by private actors, encouraging them to build or\n\n\nexpand non-public schools.\n\n\nParticipation in the TWP is subject to several conditions, aimed at improving the learning\n\n\nenvironment and grade-for-age accordance. First, in order to qualify, schools need to have a\n\n\npermit to operate from the Ministry of Education. Second, only children entering grade 1 for\n\nthe first time, between the ages of 6-8, are eligible for the subsidy. <sup>4</sup> [^4: This is a potentially binding constraint in many cases, as the average age of a new 1st grader is nearly 8 years\nold (based on data from the 2012 DHS).] Each entering grade 1\n\n\ncohort is then supported for the subsequent years through grade 6, subject to compliance with\n\n\nthe program rules. Third, in addition to not charging parents any fees, schools are required to\n\n\nprovide students with at least three school textbooks. Finally, there is a limit of 45 children per\n\n\nclass, and a maximum of two classes per grade that can benefit from the program per school.\n\n\nCompliance with these conditions has been verified annually by independent organizations\n\n\nhired by the Ministry to do so. While the majority of schools have largely complied, the\n\n\nverification reports find that most fail to fully comply with at least one of the conditions.\n\n\nHowever, the Ministry has taken little to no action to enforce compliance (either by\n\n\nencouraging it or by sanctioning non-compliance).\n\n\nThe amount of the subsidy was set at $90 per student in 2007, above the estimated average\n\n\nschool tuition to account for the extra costs associated with providing textbooks and limiting\n\n\nclass size. The subsidy\n\nMention: \"the 2012 DHS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its data give the average age of a first-grade entrant.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006241:37:1", "occurrence_id": "c88e5e5b3f1a2efe7c875ebc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: % 5.1%\n<u>Car or Truck</u> <u>DHS</u> <u>1.1%</u> <u>0.6%</u>\n_<u>Net School Enrollment Rate</u>_\nPre-Primary National Panel 21.0% 20.4%\nPrimary National Panel 79.0% 67.9%\nSecondary National Panel 20.0% 16.2%\n<u>Higher</u> <u>National Panel</u> <u>1.0%</u> <u>0.0%</u>\n_<u>Health</u>_\n_Anemia_\n% of children under 5 with Hemoglobin\nDHS 5.4% 2.1%\nless than 8 grams per deciliter\n\n_Anthropometrics_\nHeight for Age 2SD Below DHS 44.5% 47.5%\nWeight for Height 2SD Below DHS 4.8% 2.5%\nWeight for Age 2SD Below DHS 16.9% 15.4%\nHeight for Age 3SD Below DHS 17.7% 17.1%\nWeight for Height 3SD Below DHS 1.3% 0.4%\nWeight for Age 3SD Below DHS 4.1% 3.2%\nHeight for Age Mean DHS -1.80 -1.95\nWeight for Height Mean DHS 0.00 0.02\n<u>Weight for Age Mean</u> <u>DHS</u> <u>-1.00</u> <u>-1.03</u>\nNotes: DHS = 2010 Demographic and Health Survey; National Panel = Tanzania\nNational Panel Survey (LSMS) 2010 - 2011\n\n\n\n36\n\nMention: \"2010 Demographic and Health Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its sample weights are compared against the panel survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006241:37:1", "occurrence_id": "3ec2ac0eef125353766d7299", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: % 5.1%\n<u>Car or Truck</u> <u>DHS</u> <u>1.1%</u> <u>0.6%</u>\n_<u>Net School Enrollment Rate</u>_\nPre-Primary National Panel 21.0% 20.4%\nPrimary National Panel 79.0% 67.9%\nSecondary National Panel 20.0% 16.2%\n<u>Higher</u> <u>National Panel</u> <u>1.0%</u> <u>0.0%</u>\n_<u>Health</u>_\n_Anemia_\n% of children under 5 with Hemoglobin\nDHS 5.4% 2.1%\nless than 8 grams per deciliter\n\n_Anthropometrics_\nHeight for Age 2SD Below DHS 44.5% 47.5%\nWeight for Height 2SD Below DHS 4.8% 2.5%\nWeight for Age 2SD Below DHS 16.9% 15.4%\nHeight for Age 3SD Below DHS 17.7% 17.1%\nWeight for Height 3SD Below DHS 1.3% 0.4%\nWeight for Age 3SD Below DHS 4.1% 3.2%\nHeight for Age Mean DHS -1.80 -1.95\nWeight for Height Mean DHS 0.00 0.02\n<u>Weight for Age Mean</u> <u>DHS</u> <u>-1.00</u> <u>-1.03</u>\nNotes: DHS = 2010 Demographic and Health Survey; National Panel = Tanzania\nNational Panel Survey (LSMS) 2010 - 2011\n\n\n\n36\n\nMention: \"Tanzania\nNational Panel Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The companion panel whose weights are contrasted with the other survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006244:1:0", "occurrence_id": "77a74f7355894b88e053e640", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Policy Research Working Paper 7167\n\n### **Abstract**\n\nDuring the early 2000s, Moldova experienced strong economic growth and poverty and inequality reductions. This\npaper aims at uncovering the patterns behind these poverty\ntrends by looking at economic mobility and its associated\nfactors in Moldova. The findings build on the synthetic\npanel approach and allow for a non-anonymous view of the\nprocess of poverty reduction. The data used for this country\nstudy on Moldova come from the Household Budget Survey,\nwhich is conducted on a yearly basis. The paper presents\n\n\n\ntransition matrices to look at the patterns of economic\nmobility across selected consumption thresholds, as well\nas descriptive statistics and a linear probability model aimed\nat identifying correlates of economic mobility. The findings\nshow that the observed poverty reductions happened with\nlittle churning, and highlight the importance of education\nand employment on upward economic mobility in Moldova\nin the 2000s, as well as the role of public and private transfers.\n\n\n\nThis paper is a product of the Poverty Global Practice Group. It is part of a larger effort by the World Bank to provide\nopen access to its research and make a contribution to development policy discussions around the world. Policy Research\nWorking Papers are also posted on the Web at http://econ.worldbank.org. The authors may be contacted at mmeyer3@\nworldbank.org.\n\n\n_The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development_\n_issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the_\n_names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those_\n_of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and_\n_its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they\n\nMention: \"the Household Budget Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Conducted yearly, it carries the transition matrices for the country study.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006254:19:0", "occurrence_id": "2ecea696f8c5f53f689e1639", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: To further rule out concerns over sample selection produced by endogenous migration, we\n\n\nrun econometric models of a binary migration variable that captures households that moved to\n\nanother Guatemalan municipality after Agatha occurred against the shock measures and a\n\n\nsubset of covariates. <sup>15</sup> [^15: It is worth noting that the _Encovi_ surveys only track domestic migration.] The results from these models –and also from others that measure\n\n\nmigration according to the location of both the household head and the spouse– do not provide\n\nevidence that Agatha pushed systematically more (or less) households to migrate away from\n\n\nthe villages where they resided before the shock happened (Table 8- Panel C).\n\n\nAnother potential concern is the possibility of a systematic –as opposed to random–\n\nmeasurement error in the shock variable. More specifically, the empirical models may be\n\n\npicking up the cumulative and persistent effect of multiple rainfall shocks (floods and droughts)\n\nthat occurred in the past if the location of those overlap closely with the geographic coordinates\n\n\nof Agatha’s path. To investigate this bias in measurement, we checked if rainfall variability–\n\nincluding the occurrence and frequency of extreme events– in the pre-Agatha period (1970\n\n2009) differ systematically between “affected” and “less- and non-affected” weather stations. To\n\naccount for the different probability distribution of precipitation between the two groups of\n\n\nstations, we compute the coefficient of variation. The calculation of this normalized measure of\n\ndispersion shows that the coefficient of variation is comparable between treated (1.17) and\n\n\ncontrol stations (1.12). Further, the geographical location of the precipitation anomalies of\n\n\nthree large events that hit Guatemala in the 2000s (Hurricane Stan 2005, Tropical Storm\n\nBarbara 2008 and Hurricane Mitch 2008) does not overlap with the path of Agatha.\n\n\nAnother possible source of measurement error is the misallocation of historical rainfall and\n\n\nprecipitation anomalies to households across municipalities. As noted before, municipalities are\n\nmatched to their closest weather station. However, due to the low density of stations\n\nMention: \"_Encovi_ surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Its design tracks only domestic migration.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006257:36:0", "occurrence_id": "5a421cd6ed789c0357d7468b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: spouse by cohort of entry (see table 8). <sup>45</sup> The only group for which we find a statistically impor\ntant impact on employment (after accounting for multiple hypotheses testing) are spouses (who are\n\nfemales in 98% of the cases), who were not employed before 2002, and who (potentially) entered\n\nthe program after 2004. We reject the null that the effects on early and later cohorts are the same\n\n(p-value is 0.02). This suggests that CS promotes the employment of married women. With an\n\naverage participation of 30%, female labor force participation in Chile is low by Latin American\n\nstandards, overall and among the poorest sections of the population. In this context, CS coupled\n\nwith the support of employment programs could constitute an important avenue to improve fe\nmale labor market participation. The expansion of access to child care might have enhanced these\n\nimprovements, although there are no reasons to expect that the access to public day care centers\n\nexhibits a discontinuity at the effective cutoffs we estimate (Medrano, 2009).\n\n\n**Other Outcomes** We also study program impacts on four other sets of outcomes: housing, ed\nucation, health and variables associated with behaviors. The effects we find in most of these\n\noutcomes are not statistically significant (even without adjusting inference for multiple hypothesis\n\ntesting).\n\n\nThere are seven variables related to housing conditions (collected from the CAS), related to\n\nconnections to the water and sewage networks, ownership of the house, and quality of the walls\n\nand ceilings (see table A.9 in Appendix A). We also analyze nine variables (collected from the FPS)\n\nconcerning school enrolment of children, health coverage of children and adults in the household,\n\nand participation in employment centers. The variables where we find a positive impact of par\nticipation in CS indicates that all adults aged 65 or above in the family have completed health\n\ncheck-ups (see\n\nMention: \"CAS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Seven housing condition variables are collected in this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006257:36:0", "occurrence_id": "602a68d646c2510a6f36dd37", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: spouse by cohort of entry (see table 8). <sup>45</sup> The only group for which we find a statistically impor\ntant impact on employment (after accounting for multiple hypotheses testing) are spouses (who are\n\nfemales in 98% of the cases), who were not employed before 2002, and who (potentially) entered\n\nthe program after 2004. We reject the null that the effects on early and later cohorts are the same\n\n(p-value is 0.02). This suggests that CS promotes the employment of married women. With an\n\naverage participation of 30%, female labor force participation in Chile is low by Latin American\n\nstandards, overall and among the poorest sections of the population. In this context, CS coupled\n\nwith the support of employment programs could constitute an important avenue to improve fe\nmale labor market participation. The expansion of access to child care might have enhanced these\n\nimprovements, although there are no reasons to expect that the access to public day care centers\n\nexhibits a discontinuity at the effective cutoffs we estimate (Medrano, 2009).\n\n\n**Other Outcomes** We also study program impacts on four other sets of outcomes: housing, ed\nucation, health and variables associated with behaviors. The effects we find in most of these\n\noutcomes are not statistically significant (even without adjusting inference for multiple hypothesis\n\ntesting).\n\n\nThere are seven variables related to housing conditions (collected from the CAS), related to\n\nconnections to the water and sewage networks, ownership of the house, and quality of the walls\n\nand ceilings (see table A.9 in Appendix A). We also analyze nine variables (collected from the FPS)\n\nconcerning school enrolment of children, health coverage of children and adults in the household,\n\nand participation in employment centers. The variables where we find a positive impact of par\nticipation in CS indicates that all adults aged 65 or above in the family have completed health\n\ncheck-ups (see\n\nMention: \"FPS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Nine enrolment and coverage variables are collected in this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006261:25:0", "occurrence_id": "9a4bd7152a78eab914620883", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Sánchez-Martín, de Piniés, and Antoine, 2014. Measuring the channels of FDI externalities in the Caribbean\n\n##### **4 Conclusions**\n\n\nThis paper has explored the determinants of backward linkages between foreign owned firms and domestic\nsuppliers in the host country. Using the World Bank-IFC Enterprise Survey, the intensity of these linkages has\nbeen preliminarily established in more than one hundred developing economies for the first time. It is worth\nmentioning that these surveys have not been designed to specifically ask local companies about their linkages\nwith foreign MNCs, and for that reason we have defined proxies for backward linkages.\n\n\nResults of the econometric regressions presented in this paper confirm that market oriented (horizontal) FDI\nis more predisposed to establishing backward linkages than export oriented (vertical) FDI as posited by\nAltenburg (2000). A negative and significant relationship between the percentage of inputs of domestic origin\n(our proxy for backward linkages) and the percentage of sales going to exports in foreign owned companies\n(the proxy for forward linkages) has also been found. In addition, subsidiaries with a lower percentage of\nforeign participation are shown to rely more on local inputs in line with Chen and Chang (2011). The reason\nis that it is easier for a joint venture to find reliable suppliers due to market specific knowledge of local\npartners. Holding foreign technology licenses is also shown to contribute significantly to developing fewer\nbackward linkages. This could be the result of the multinational having already identified international\nsuppliers to serve the needs of the specific production process using a given technology. Skilled labor or\nquality certificates, however, are not found to be significantly correlated with backward linkages.\n\n\nAnother important finding both in terms of foreign investment policy and understanding the development\noutcomes of FDI, is that the sector of destination matters. Drawing on our econometric exercise, we have\nshown that the food, wood and furniture, and automobile and auto parts sectors tend to develop many\nbackward linkages. In contrast,\n\nMention: \"the World Bank-IFC Enterprise Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Foreign-domestic supplier linkages established across 100+ economies.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006268:6:1", "occurrence_id": "7cbb5565686bb807c000a631", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  these results in Table 2 and an online appendix. The robustness checks include:\n\n\n1. _Including country dummies_ . Now country-level variables, including the country-level\n\n\ntelecom variables, are dropped due to collinearity.\n\n\n2. _Including additional controls_ (industry-country tariffs and local averages for markup\n\n\nand corruption). Such controls are potentially important if openness, competition and\n\n\n5 Tariffs are from the UN TRAINS database. All other variables are from the WBES.\n\n5\n\nMention: \"UN TRAINS database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The tariff controls in the specification come from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006268:6:1", "occurrence_id": "c9679607a8996073cdf68d0c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  these results in Table 2 and an online appendix. The robustness checks include:\n\n\n1. _Including country dummies_ . Now country-level variables, including the country-level\n\n\ntelecom variables, are dropped due to collinearity.\n\n\n2. _Including additional controls_ (industry-country tariffs and local averages for markup\n\n\nand corruption). Such controls are potentially important if openness, competition and\n\n\n5 Tariffs are from the UN TRAINS database. All other variables are from the WBES.\n\n5\n\nMention: \"WBES\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"All remaining firm variables come from this survey programme.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006269:5:1", "occurrence_id": "06bd9af87961aa3a74436ec0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  99 counties that were\nupgraded from counties to county-level cities between 1994 and 1997.\n\nThe firm-level data set spans from 1998 to 2009 and is compiled from the Annual Survey of\nIndustrial Firms (ASIF) collected by the National Bureau of Statistics of China. This data set is\noften referred to as the 5 Million RMB data set, since it contains non-SOE firms with main\noperating revenue (i.e., sales) of more than five million RMB, and all state-owned enterprises\n(SOEs) regardless of sales volume. Thus, the ASIF survey data set tracks the performance of all\nSOEs and all big non-SOEs. The ASIF data set details all operational, financial, and managerial\nfacts of firms in three broad categories: (1) mining, (2) manufacturing, and (3) production and\ndistribution of electricity, gas, and water, and classifies each firm to 6-digit sub-industries.\nCompared with the 2004 China Economic Census, the firms covered by this data set represented\n89.5% of the total main operating revenue (sales) from all enterprises in China (Nie, Jiang, & Yang,\n2012).\n\nThe 1998-2009 ASIF data set covers a period when some major structural reforms happened in\nChina, such as the SOE reform in the mid-to-late 1990s. Following the end of the county-to-city\nupgrading policy in 1997, China carried out a major reform of SOEs between 1998 and 2000. The\naim of the reform was to privatize or close down small and unprofitable SOEs (Song and Hsieh,\n2013). In 2004, an economic census covering all firms regardless of size was carried out and more\ndetailed information on firms was collected. <sup>1</sup> To mitigate the effects of the above SOE changes,\nand since we are only interested in understanding the effect of county-\n\nMention: \"5 Million RMB data set\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its threshold captures non-state firms above five million in revenue.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006269:5:1", "occurrence_id": "0d04bc2f626b4a4cb63baa24", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  99 counties that were\nupgraded from counties to county-level cities between 1994 and 1997.\n\nThe firm-level data set spans from 1998 to 2009 and is compiled from the Annual Survey of\nIndustrial Firms (ASIF) collected by the National Bureau of Statistics of China. This data set is\noften referred to as the 5 Million RMB data set, since it contains non-SOE firms with main\noperating revenue (i.e., sales) of more than five million RMB, and all state-owned enterprises\n(SOEs) regardless of sales volume. Thus, the ASIF survey data set tracks the performance of all\nSOEs and all big non-SOEs. The ASIF data set details all operational, financial, and managerial\nfacts of firms in three broad categories: (1) mining, (2) manufacturing, and (3) production and\ndistribution of electricity, gas, and water, and classifies each firm to 6-digit sub-industries.\nCompared with the 2004 China Economic Census, the firms covered by this data set represented\n89.5% of the total main operating revenue (sales) from all enterprises in China (Nie, Jiang, & Yang,\n2012).\n\nThe 1998-2009 ASIF data set covers a period when some major structural reforms happened in\nChina, such as the SOE reform in the mid-to-late 1990s. Following the end of the county-to-city\nupgrading policy in 1997, China carried out a major reform of SOEs between 1998 and 2000. The\naim of the reform was to privatize or close down small and unprofitable SOEs (Song and Hsieh,\n2013). In 2004, an economic census covering all firms regardless of size was carried out and more\ndetailed information on firms was collected. <sup>1</sup> To mitigate the effects of the above SOE changes,\nand since we are only interested in understanding the effect of county-\n\nMention: \"ASIF survey data set\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"It tracks all state enterprises and all large non-state enterprises.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006269:5:1", "occurrence_id": "12bc43c816427e4eaad3c5f4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  99 counties that were\nupgraded from counties to county-level cities between 1994 and 1997.\n\nThe firm-level data set spans from 1998 to 2009 and is compiled from the Annual Survey of\nIndustrial Firms (ASIF) collected by the National Bureau of Statistics of China. This data set is\noften referred to as the 5 Million RMB data set, since it contains non-SOE firms with main\noperating revenue (i.e., sales) of more than five million RMB, and all state-owned enterprises\n(SOEs) regardless of sales volume. Thus, the ASIF survey data set tracks the performance of all\nSOEs and all big non-SOEs. The ASIF data set details all operational, financial, and managerial\nfacts of firms in three broad categories: (1) mining, (2) manufacturing, and (3) production and\ndistribution of electricity, gas, and water, and classifies each firm to 6-digit sub-industries.\nCompared with the 2004 China Economic Census, the firms covered by this data set represented\n89.5% of the total main operating revenue (sales) from all enterprises in China (Nie, Jiang, & Yang,\n2012).\n\nThe 1998-2009 ASIF data set covers a period when some major structural reforms happened in\nChina, such as the SOE reform in the mid-to-late 1990s. Following the end of the county-to-city\nupgrading policy in 1997, China carried out a major reform of SOEs between 1998 and 2000. The\naim of the reform was to privatize or close down small and unprofitable SOEs (Song and Hsieh,\n2013). In 2004, an economic census covering all firms regardless of size was carried out and more\ndetailed information on firms was collected. <sup>1</sup> To mitigate the effects of the above SOE changes,\nand since we are only interested in understanding the effect of county-\n\nMention: \"ASIF data set\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its detail spans operational, financial and managerial facts.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006269:5:1", "occurrence_id": "329ebcd991a70a28b4325a75", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  99 counties that were\nupgraded from counties to county-level cities between 1994 and 1997.\n\nThe firm-level data set spans from 1998 to 2009 and is compiled from the Annual Survey of\nIndustrial Firms (ASIF) collected by the National Bureau of Statistics of China. This data set is\noften referred to as the 5 Million RMB data set, since it contains non-SOE firms with main\noperating revenue (i.e., sales) of more than five million RMB, and all state-owned enterprises\n(SOEs) regardless of sales volume. Thus, the ASIF survey data set tracks the performance of all\nSOEs and all big non-SOEs. The ASIF data set details all operational, financial, and managerial\nfacts of firms in three broad categories: (1) mining, (2) manufacturing, and (3) production and\ndistribution of electricity, gas, and water, and classifies each firm to 6-digit sub-industries.\nCompared with the 2004 China Economic Census, the firms covered by this data set represented\n89.5% of the total main operating revenue (sales) from all enterprises in China (Nie, Jiang, & Yang,\n2012).\n\nThe 1998-2009 ASIF data set covers a period when some major structural reforms happened in\nChina, such as the SOE reform in the mid-to-late 1990s. Following the end of the county-to-city\nupgrading policy in 1997, China carried out a major reform of SOEs between 1998 and 2000. The\naim of the reform was to privatize or close down small and unprofitable SOEs (Song and Hsieh,\n2013). In 2004, an economic census covering all firms regardless of size was carried out and more\ndetailed information on firms was collected. <sup>1</sup> To mitigate the effects of the above SOE changes,\nand since we are only interested in understanding the effect of county-\n\nMention: \"2004 China Economic Census\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its coverage benchmark shows the earlier dataset represents 89.5 percent of revenue.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006270:17:3", "occurrence_id": "b2100773cf5cd1b1f8d2d049", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: mark>0.5%</mark> <mark>8.1%</mark>\n<mark>8</mark> <mark>8.3%</mark> <mark>2.8%</mark> <mark>2.0%</mark> <mark>10.2%</mark> <mark>18.6%</mark> <mark>4.0%</mark> <mark>2.5%</mark> <mark>18.3%</mark>\n<mark>9</mark> <mark>17.7% 11.7%</mark> <mark>10.6%</mark> <mark>23.3%</mark> <mark>36.3%</mark> <mark>15.7%</mark> <mark>13.1%</mark> <mark>41.6%</mark>\n<u><mark>10</mark></u> <u><mark>63.7% 84.3%</mark></u> <u><mark>86.9%</mark></u> <u><mark>58.4%</mark></u> <u><mark>100.0% 100.0%</mark></u> <u><mark>100.0%</mark></u> <u><mark>100.0%</mark></u>\nSource: Own calculations based on IES 2010/11.\n1/ This is the distribution of market income for the population ordered by market income. All the others are\nconcentration shares for the population ordered by market income.\n2/ This is\n\nMention: \"IES 2010/11\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Market income distributions are calculated from this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006272:14:0", "occurrence_id": "fdbd3e634d0849c6c440cdcc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ###### **2. Estimate deaths among women aged 15-49**\n\nThe consequent distribution of deaths by gender and age distribution is presented in Table A2.\n\n\n**<u>Table A2: Estimated EVD Deaths by Age & Gender</u>**\n\n\n<u>Guinea</u> <u>Liberia</u> <u>Sierra Leone</u>\n\nTotal Deaths **<mark>11</mark>** 1,995 3,826 3,341\n\nFemale deaths <15 years old 167 360 380\n\n\n15-44 years old 540 1,028 895\n\n\n≥45 years old 322 490 444\n\n\nMale deaths <15 years old 159 373 359\n\n\n15-44 years old 514 1,064 847\n\n\n<u>≥45 years old</u> <u>306</u> <u>507</u> <u>420</u>\nSource: Total mortality by country is from WHO (2014). Distribution by age and gender is inferred by the author based on WHO Ebola Response\nTeam (2014).\n\n###### **3. Calculate surviving children**\n\nThe age distribution of Ebola deaths among women age 15-44 is assumed to reflect the age distribution of\nwomen age 15-44, in five-year age groups. For example, in Sierra Leone, 23% of women age 15-44 are\nbetween the ages of 15 and 19, so 23% of the 895 female deaths are applied to women age 15-19. <sup>b</sup> The\nsame process is applied for women age 45-49, using the distribution of deaths for women over age 45.\nThese five-year age group mortality counts are then multiplied by the mean number of living children\n(MNLC) for women in the same age groups, from the most recent DHS for the\n\nMention: \"the most recent DHS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its fertility measures for women in age groups feed the calculation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006275:24:0", "occurrence_id": "56e83bee928685167b0a5bb4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Table A2: Toilet facilities (Bangladesh DHS 2004, 2007)**\n\n|Facility Type|DHS Description|Freq.|%|\n|---|---|---|---|\n|**Dummy variables included**||||\n|**Flush toilet**|Flush to piped sewer system|2,496|6.38|\n||Flush to septic tank|2,646|6.76|\n||Flush to pit latrine|870|2.22|\n||Flush to somewhere else|690|1.76|\n||Flush, don't know where|36|0.09|\n|**Pit latrine**|Ventilated improved pit latrine|3,192|8.16|\n||Pit latrine with slab|9,846|25.16|\n||Pit latrine without slab/open pit|11,466|29.31|\n||Pit latrine, other|465|1.19|\n|||||\n|Dummy variables excluded<br>(for comparison)||||\n|No toilet facility|No facility/bush/field|3,453|8.83|\n|Excluded<br>(ambiguous definition or<br>insufficient data)||||\n||Bucket toilet|12|0.03|\n||Hanging toilet/latrine|2,160|5.52|\n||Not a de jure resident|1,794|4.59|\n|||||\n||Total|39,126|100|\n\n\n\n23\n\nMention: \"Bangladesh DHS 2004, 2007\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Sanitation facility types and their frequencies come from these two rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006279:37:1", "occurrence_id": "97f518f99ad54b5feb35f63c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  8,854 25,757 31,343 55,199 138,310 1462% 341%\n191% 22% 76% 151%\nFormal non-tradables 6,734 19,833 23,970 37,896 112,859 1576% 371%\n195% 21% 58% 198%\n\n\nNotes: Indian descriptive statistics taken from Annual Survey of Industries and National Sample Statistics. Tradables definition comes from Kothari (2014) based upon abovemedian values of exports plus imports to gross production at the three-digit industry level. Output is measured as annual gross sales. Average wage is measured as total annual\nwages divided by the number of employees in the firm. Output and average wages across all years are converted into 2005 rupees.\n\nMention: \"Annual Survey of Industries\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Sector shares in the descriptive statistics come from this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006279:37:1", "occurrence_id": "2c8fa3e0e63a546f4b4f9a30", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  8,854 25,757 31,343 55,199 138,310 1462% 341%\n191% 22% 76% 151%\nFormal non-tradables 6,734 19,833 23,970 37,896 112,859 1576% 371%\n195% 21% 58% 198%\n\n\nNotes: Indian descriptive statistics taken from Annual Survey of Industries and National Sample Statistics. Tradables definition comes from Kothari (2014) based upon abovemedian values of exports plus imports to gross production at the three-digit industry level. Output is measured as annual gross sales. Average wage is measured as total annual\nwages divided by the number of employees in the firm. Output and average wages across all years are converted into 2005 rupees.\n\nMention: \"National Sample Statistics\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The companion household source for the same statistics.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006283:10:0", "occurrence_id": "39e72266b3066e7128a958c4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 1: Reported Days in Customs - Trading Across Borders\n\n\n\n\n\n\n\nNote: Distributions of days in import customs and days in export customs, respectively.\n\nData from _Doing_ _Business_ _-_ _Trading_ _Across_ _Borders_ _2012_ for 182 countries.\n\n\n**3.2** **Quantitative** **measures** **of** **border** **policy**\n\n\nThe primary goal of this paper is to relate border management policies and procedures to border\n\n\noutcomes. Heretofore this has been a difficult issue to address empirically because quantitative\n\n\nrepresentations of border policies/procedures have not been available. <sup>15</sup> In anticipation of the Bali\n\n\nnegotiations the OECD set out to correct this problem, identifying 12 categories of trade facilitation\n\n\npolicies and procedures. After documenting national policies and procedures, the OECD scored\n\n\ncountries on their proximity to the world’s best practice in each of the 12 policy categories.\n\n\nThe raw material for the OECD’s 12 trade facilitation indicators was a detailed catalogue of\n\n\nthe policies and procedures used in border management agencies around the world. The OECD’s\n\n\ninitial catalogue, based in 2010, attempted to document 95 different indicators of border manage\n\nment procedures in OECD countries. A subsequent catalogue, representing the situation in 2012,\n\n\nattempted to track 78 procedures in non-OECD countries. <sup>16</sup> The OECD dropped 7 questions be\n\n15Measures such as the World Bank’s Logistics Performance Index, or information from TAB on the number of\ndocuments required to trade provide some high-level information. But these do not represent the breadth of the policy\nreforms considered in the TFA, nor do they provide sufficient policy detail for our purposes.\n\n16Our empirical specifications pool the two data series, but includes an OECD dummy variable. Among other things,\nthis should control for any systematic differences between the situations in 2010\n\nMention: \"Logistics Performance Index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It and trade document counts give only high-level trade facilitation measures.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006289:26:0", "occurrence_id": "aea19c09c768aa6a488719e7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: We provide estimates of the patterns and predictors of firm exit using a unified and\n\n\nharmonized data methodology for 47 economies. By relying on such survey data, we leverage a\n\n\nsingular methodology, with a well-defined universe of inference and are able to provide cross\neconomy comparisons, including estimates of the effects of macro-level variables, including the\n\n\nregulatory environment. We find evidence consistent with results well-established in the literature,\n\n\nbut for a broader range of economies, suggesting that the use of comparatively small, representative\n\n\nsurveys may provide a viable opportunity for further research. This approach appears to be widely\n\napplicable, particularly when compared to the feasibility of coordinating and scaling up census-based\n\n\ndata collection efforts. However, while the use of ES data enables such comparisons, we adopt a\n\n\nnecessarily conservative approach, due to the considerations of the difficulty of data collection by the\n\n\ncollection of survey samples, in periodic waves. As such, we find estimates for firm exit and job loss\n\nthat tend to fall below comparative estimates from census- and registry-based studies; this likely\n\n\nindicates that survey-based approaches to firm exit can be considered as lower bounds of firm exit.\n\n\nOne particular strength from our approach is the possibility of systematic cross-country\n\n\nresearch. We exploit measures for barriers to entry and exit as a logical first-effort: further cross\n\ncountry analysis utilizing additional topic areas will surely be needed. This agenda should also include\n\nfurther refinement of data measurement of firm exit, including building on ongoing, continuous data\n\n\ncollection efforts such as the ES and other globally comparable data sets. Additional research will\n\n\nbenefit from multi-wave panels, including the observations of several economies over more-frequent\n\n\nand heterogeneous external conditions, as well as further integration of exit measures with data entry.\n\n\n25\n\nMention: \"ES data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Its use enables comparisons, though a conservative approach is taken.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006289:26:0", "occurrence_id": "143ca5e14a9c26cc62b5bffb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: We provide estimates of the patterns and predictors of firm exit using a unified and\n\n\nharmonized data methodology for 47 economies. By relying on such survey data, we leverage a\n\n\nsingular methodology, with a well-defined universe of inference and are able to provide cross\neconomy comparisons, including estimates of the effects of macro-level variables, including the\n\n\nregulatory environment. We find evidence consistent with results well-established in the literature,\n\n\nbut for a broader range of economies, suggesting that the use of comparatively small, representative\n\n\nsurveys may provide a viable opportunity for further research. This approach appears to be widely\n\napplicable, particularly when compared to the feasibility of coordinating and scaling up census-based\n\n\ndata collection efforts. However, while the use of ES data enables such comparisons, we adopt a\n\n\nnecessarily conservative approach, due to the considerations of the difficulty of data collection by the\n\n\ncollection of survey samples, in periodic waves. As such, we find estimates for firm exit and job loss\n\nthat tend to fall below comparative estimates from census- and registry-based studies; this likely\n\n\nindicates that survey-based approaches to firm exit can be considered as lower bounds of firm exit.\n\n\nOne particular strength from our approach is the possibility of systematic cross-country\n\n\nresearch. We exploit measures for barriers to entry and exit as a logical first-effort: further cross\n\ncountry analysis utilizing additional topic areas will surely be needed. This agenda should also include\n\nfurther refinement of data measurement of firm exit, including building on ongoing, continuous data\n\n\ncollection efforts such as the ES and other globally comparable data sets. Additional research will\n\n\nbenefit from multi-wave panels, including the observations of several economies over more-frequent\n\n\nand heterogeneous external conditions, as well as further integration of exit measures with data entry.\n\n\n25\n\nMention: \"ES\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"A continuous collection effort that future work could build on.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006291:30:1", "occurrence_id": "671e8616a7cf05119c10aed0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  represents very much a\nlower bound.\n\n\nThe IMF produces two sets of “subsidy” estimations, which it refers to as “pre-tax” and “post-tax.” The\nreference point for the IMF’s pre-tax subsidies excludes taxes on consumers, such as fuel excise tax,\nVAT, and GST. In tabulating its pre-tax subsidies, the IMF combines estimates obtained by price-gap\nmeasurements; budgetary and tax expenditure support for producers in OECD countries for petroleum\nproducts and coal (mostly using the OECD’s data); and hidden costs in the power sector in 40 countries,\ninclusive of hydropower and nuclear power where they exist (based on a mix of IMF and World Bank\nsources). As such, the IMF’s pre-tax subsidy calculations are not strictly comparable across countries, nor\nare they confined solely to fossil fuels, although the contribution of energy sources not derived from\nfossil fuels should be small.\n\n\nIn terms of country coverage, the IMF reports petroleum product subsidies for 176 countries, some of\nwhich have been found not to have pre-tax subsidies. The most recent IEA analysis identified 40\ncountries that subsidized fossil fuel consumption in 2013 based on the IEA’s longstanding program of\nwork that monitors energy pricing across the world. The IEA estimates cover subsidies to fossil fuels\nconsumed by end-users (households and businesses) and subsidies to fossil-fuel inputs to electric power\ngeneration. The IEA notes that it believes that more than 40 countries had fossil fuel subsidies in 2013,\nbut there was insufficient information available to perform the necessary calculations to confirm their\nexistence with sufficient confidence for inclusion in its publication. The two numbers—176 versus 40—\nare not directly comparable, in part because types of subsidies covered are different. The numbers of\ncountries covered by the IMF for other types of energy are smaller than 176, and as few as 39 for coal.\n\n\n29\n\nMention: \"OECD’s data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Support for petroleum and coal producers across these countries is measured here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006294:9:0", "occurrence_id": "f5372656172cf522345dbaac", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: pricing mechanism at the border as an instrument of protection of domestic production. However, due\nto the surge in the price of wheat on the international market in 2007 and the need to meet the increasing\ndemand for bread, imports of soft wheat for the production of liberalized flours benefited from import\nsubsidies when prices exceeded the target price.\n\n\nThe high burden of subsidies stemming from the widening gap between domestic and international\nprices of wheat urged the government to take measures to reduce, albeit marginally, subsidies benefiting\nthe national flour. To this end, it reduced the quota of subsidized flour to 9 million tons annually, while\nstrengthening the control of production and delivery of subsidized flour, and redeploying its distribution\nto targeted populations, mainly using poverty maps. The government further limited the subsidizedflour’s annual quota to 8.5 million tons starting from the second half of 2013. The reduction of the quota\nwas limited to urban areas with a poverty rate below 10%.\n\n\n**3.** **Baseline data**\n\n\nThe distributional and simulation analyses that follow largely rely on household budget survey data.\nThese data are collected occasionally in Morocco and the last available survey is the 2007 Living\nStandards Survey (LSS). The first exercise before undertaking the distributional and simulation analyses\nthat follow is therefore to update the information available in the 2007 data to 2014, the year we consider\nfor the analysis. The 2007-2014 extrapolations are based on demographic and economic estimates. Table\n4 shows the reference statistics used for the extrapolation.\n\n\n**Table 4 - Reference statistics 2007-2013**\n\n\n\n\n\n\n\n\n\n|Source|Indicator|2007|2008|2009|2010|2011|2012|2013|\n|---|---|---|---|---|---|---|---|---|\n|1|GDP<br>(current<br>prices, bn\n\nMention: \"2007 Living\nStandards Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The last available round before the distributional simulation exercise.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006295:34:1", "occurrence_id": "45bbb5a2b3393bf0792bb155", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: *\n\n\n**Scenario**\n\n\n\n**Pre-**\n**<u>reform</u>**\n\n\n\n**1 (30%**\n**reduction**\n\n**in**\n**<u>subsidies)</u>**\n\n\n\n**Scenario 2**\n**(Elimination**\n\n\n\n**<u>reform</u>** **<u>subsidies)</u>** **<u>of subsidies)</u>**\n\nTotal real household expenditure (bn LYD) 12.53 11.79 9.29\nHousehold expenditure loss in real terms (% of pre-reform) n.a. -6 -19.9\nTotal subsidies (bn LYD) 2.49 1.74 0\nIncidence of subsidies (% tot exp.) 19.9 14.8 0\nChange in government spending following reform (bn LYD) n.a. -1.22 -2.49\nSavings to the government following reform (% govt. exp.)* n.a. 1.9 3.9\nPoverty headcount (%, international poverty line) 8.5 11.2 21.7\nPoverty headcount (%, national poverty line) 14.4 18.2 30.4\n<u>Inequality (%, Gini)</u> <u>30.2</u> <u>30.8</u> <u>33.4</u>\nSource: Libyan Household Consumption Survey 2007/8; Libyan authorities; and Bank staff calculations.\n\n\n\nClearly, energy subsidy reform would have a huge impact on the Libyan economy, which calls for\ngradualism. Full liberalization would imply price increases of between 7 and 10 times the existing prices,\nin a context where alternatives (such as more efficient production processes for electricity\n\nMention: \"Libyan Household Consumption Survey 2007/8\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Consumption levels and inequality behind the subsidy reform analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006297:42:2", "occurrence_id": "1e1a4733b39b6fcdfe742b8c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  24276.68 23763.79 23250.9 22738.01 22225.13 21712.24 21199.35 20686.46 20173.58 19660.69 19147.8\n\n\nbased on original Table 5.1. World Bank Cairo Pollution report\n\n\n41\n\nMention: \"World Bank Cairo Pollution report\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its original table supplies the pollution cost series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006278:57:0", "occurrence_id": "30d8bdba3be93b5cc1ca12e6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Trade Reform and Regional Dynamics <u>Dix-Carneiro</u> <u>and</u> <u>Kovak</u>\n\n\nFigure 17: Individual Share of Formally Employed Months Away from Initial Region - Tradable\nSector Worker Sample - 1990-2009\n\n\n0.6\n\n\n0.5\n\n\n0.4\n\n\n0.3\n\n\n0.2\n\n\n0.1\n\n\n0\n\n\n-0.1\n\n\n-0.2\n\n|Col1|Col2|\n|---|---|\n|90<br>1991<br>1992<br>1993<br>1994<br>19|95<br>1996<br>1997<br>1998<br>1999<br>2000<br>2001<br>2002<br>2003<br>2004<br>2005<br>2006<br>2007<br>2008<br>20|\n\n\n\nEach point reflects an individual regression coefficient where the dependent variable is the share of formally employed\nmonths in the year listed on the x-axis that the individual spent working in a region other than their initial region,\ncalculated using RAIS. For all years, the independent variable is the regional tariff change reflecting tariff changes\nfrom 1990-1995 (described in the text), with various controls (described in the text). Positive estimates imply a\nlower share of formally employed months away from the initial region for workers initially employed in regions facing\nlarger tariff declines. Vertical bar indicates that liberalization was complete in 1995. Dashed lines show 95 percent\nconfidence intervals. Standard errors adjusted for 112 mesoregion clusters.\n\n\n56\n\nMention: \"RAIS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Work days spent outside the initial region are calculated from these records.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006299:20:2", "occurrence_id": "8376ef1656e46f7c0760da68", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  States.|\n|What is the notice period for a<br>redundancy dismissal (in weeks of<br>salary)?a|Luxembourg, 17.3;**Sweden**, 14.4;**Finland** and**Iceland**, 10.1;<br>Germany, 10; Netherlands and**Norway**, 8.7; Portugal, 7.9;<br>Belgium, France and Italy, 7.2; United Kingdom, 5.3; Canada, 5;<br>Japan, 4.3; Ireland, 4; Spain, 2.1; Austria, 2; Denmark and United<br>States, 0.<br>Averages: EU15, 6.9; G7, 5.6; Nordic, 7.2.|\n|What is the severance payment for a<br>redundancy dismissal (in weeks of<br>salary)?a|Greece, 15.9; Portugal and Spain, 15.2; Germany, 11.6; Ireland,<br>8.2; Canada, 5; France, 4.6: Luxembourg, 4.3; United Kingdom,<br>3.1; Austria, Belgium,**Denmark**, **Finland**, **Iceland**, Italy, Japan,<br>Netherlands,**Norway**, **Sweden** and United States, 0.<br>Averages: EU15, 5.2; G7, 3.5; Nordic, 0.|\n\n\nSource: World Bank Group, Doing Business 2014. Note: The sample consists of all 20 countries in the EU15, G7 and\nNordic region. The Nordic average excludes Sweden. a. Average for workers with 1, 5 and 10 years of tenure.\n\n\n\n19\n\nMention: \"Doing Business 2014\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Regulatory timings across EU15, G7 and Nordic economies come from this edition.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006300:10:0", "occurrence_id": "bc8d48444bf89d12fc2d3d82", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Secretariat of the National Public Security Council (SNSP), a federal entity dependent of the Ministry\nof Interior. This variable has the advantage that aims to be more closely related to “war-on-drugs”\nspikes of violence that is the motivation of this study. However, this variable was created by individual\nofficials’ assessments of the relatedness to drug trafficking of the given homicide, and thus it is likely\nto have substantially more measurement error. Thus, although we include it in some models, we use\nthe standard homicide rate variable for our main specification.\n\n\n**Demographic Characteristics at the Municipal Level**\n\n\nDemographic characteristics at the municipality level were obtained from the Censos de Población y\nVivienda for 1990, 2000, 2005, and 2010, which include data on total population, share of rural\npopulation, share of population between 15 and 29 years of age, share of population over 60, median\nage, number of households, male to female ratio and fertility rate. For the years between censuses, we\ninterpolated values assuming linear growth.\n\n\nWe have also gathered data on aggregate figures of public expenditures, government transfers,\nand other public finance variables at the municipal level in Mexico. The data on public expenditures\nwas obtained from the State and Municipal System of Databases (SIMBAD) produced by the National\nInstitute of Statistics, Geography, and Information (INEGI).\n\n\n**Data on Education (schooling, enrollment rates)**\n\n\nThe yearly education data from the first data set includes student enrollment (total students), passing\nrates, and retention rates at the municipality level. These data were obtained from INEGI and are\nconsistently reported from 1994 to 2010. The variables are decomposed by the different levels of\nschooling, namely for _primaria_ (primary school, grades 1-6), _secundaria_ (lower secondary school, grades\n7-9), and, _preparatoria_ or _bachillerato\n\nMention: \"SNSP\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"A federal entity within the interior ministry supplying the violence measure.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006300:10:0", "occurrence_id": "23a158decc20794ff2473a8e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Secretariat of the National Public Security Council (SNSP), a federal entity dependent of the Ministry\nof Interior. This variable has the advantage that aims to be more closely related to “war-on-drugs”\nspikes of violence that is the motivation of this study. However, this variable was created by individual\nofficials’ assessments of the relatedness to drug trafficking of the given homicide, and thus it is likely\nto have substantially more measurement error. Thus, although we include it in some models, we use\nthe standard homicide rate variable for our main specification.\n\n\n**Demographic Characteristics at the Municipal Level**\n\n\nDemographic characteristics at the municipality level were obtained from the Censos de Población y\nVivienda for 1990, 2000, 2005, and 2010, which include data on total population, share of rural\npopulation, share of population between 15 and 29 years of age, share of population over 60, median\nage, number of households, male to female ratio and fertility rate. For the years between censuses, we\ninterpolated values assuming linear growth.\n\n\nWe have also gathered data on aggregate figures of public expenditures, government transfers,\nand other public finance variables at the municipal level in Mexico. The data on public expenditures\nwas obtained from the State and Municipal System of Databases (SIMBAD) produced by the National\nInstitute of Statistics, Geography, and Information (INEGI).\n\n\n**Data on Education (schooling, enrollment rates)**\n\n\nThe yearly education data from the first data set includes student enrollment (total students), passing\nrates, and retention rates at the municipality level. These data were obtained from INEGI and are\nconsistently reported from 1994 to 2010. The variables are decomposed by the different levels of\nschooling, namely for _primaria_ (primary school, grades 1-6), _secundaria_ (lower secondary school, grades\n7-9), and, _preparatoria_ or _bachillerato\n\nMention: \"Censos de Población y\nVivienda\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Municipal demographic characteristics including rural poverty share.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006300:10:0", "occurrence_id": "e4c4a4c98074098c213ba8ce", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Secretariat of the National Public Security Council (SNSP), a federal entity dependent of the Ministry\nof Interior. This variable has the advantage that aims to be more closely related to “war-on-drugs”\nspikes of violence that is the motivation of this study. However, this variable was created by individual\nofficials’ assessments of the relatedness to drug trafficking of the given homicide, and thus it is likely\nto have substantially more measurement error. Thus, although we include it in some models, we use\nthe standard homicide rate variable for our main specification.\n\n\n**Demographic Characteristics at the Municipal Level**\n\n\nDemographic characteristics at the municipality level were obtained from the Censos de Población y\nVivienda for 1990, 2000, 2005, and 2010, which include data on total population, share of rural\npopulation, share of population between 15 and 29 years of age, share of population over 60, median\nage, number of households, male to female ratio and fertility rate. For the years between censuses, we\ninterpolated values assuming linear growth.\n\n\nWe have also gathered data on aggregate figures of public expenditures, government transfers,\nand other public finance variables at the municipal level in Mexico. The data on public expenditures\nwas obtained from the State and Municipal System of Databases (SIMBAD) produced by the National\nInstitute of Statistics, Geography, and Information (INEGI).\n\n\n**Data on Education (schooling, enrollment rates)**\n\n\nThe yearly education data from the first data set includes student enrollment (total students), passing\nrates, and retention rates at the municipality level. These data were obtained from INEGI and are\nconsistently reported from 1994 to 2010. The variables are decomposed by the different levels of\nschooling, namely for _primaria_ (primary school, grades 1-6), _secundaria_ (lower secondary school, grades\n7-9), and, _preparatoria_ or _bachillerato\n\nMention: \"State and Municipal System of Databases\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Public expenditure at the municipal level comes from this system.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006300:10:0", "occurrence_id": "22d3f3ed34b287254174d588", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Secretariat of the National Public Security Council (SNSP), a federal entity dependent of the Ministry\nof Interior. This variable has the advantage that aims to be more closely related to “war-on-drugs”\nspikes of violence that is the motivation of this study. However, this variable was created by individual\nofficials’ assessments of the relatedness to drug trafficking of the given homicide, and thus it is likely\nto have substantially more measurement error. Thus, although we include it in some models, we use\nthe standard homicide rate variable for our main specification.\n\n\n**Demographic Characteristics at the Municipal Level**\n\n\nDemographic characteristics at the municipality level were obtained from the Censos de Población y\nVivienda for 1990, 2000, 2005, and 2010, which include data on total population, share of rural\npopulation, share of population between 15 and 29 years of age, share of population over 60, median\nage, number of households, male to female ratio and fertility rate. For the years between censuses, we\ninterpolated values assuming linear growth.\n\n\nWe have also gathered data on aggregate figures of public expenditures, government transfers,\nand other public finance variables at the municipal level in Mexico. The data on public expenditures\nwas obtained from the State and Municipal System of Databases (SIMBAD) produced by the National\nInstitute of Statistics, Geography, and Information (INEGI).\n\n\n**Data on Education (schooling, enrollment rates)**\n\n\nThe yearly education data from the first data set includes student enrollment (total students), passing\nrates, and retention rates at the municipality level. These data were obtained from INEGI and are\nconsistently reported from 1994 to 2010. The variables are decomposed by the different levels of\nschooling, namely for _primaria_ (primary school, grades 1-6), _secundaria_ (lower secondary school, grades\n7-9), and, _preparatoria_ or _bachillerato\n\nMention: \"yearly education data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Enrolment, passing rates and schooling within the first dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006302:27:1", "occurrence_id": "ea4480995dffdee119deb634", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 36</sup>, we find that the average unrefund rate for\n\n\nadvanced-technology products in our sample is higher than that of other products (5.84% vs.\n\n\n4.45%), but the average indirect export share of advanced-technology products is only\n\n\nslightly higher than that of other products (32.5% vs. 31.6%).\n\n\nAlthough including HS4 product fixed effects can help to alleviate the above concerns,\n\n\nwe also take them as problems of omitted variable bias by checking the robustness of our\n\n\n35 These data are based on the sample used in the first regression in Table 1. The three groups of\n\nproducts (Primary Products, Resource-based Products, and others) are mutually exclusive, accounting\n\nfor 840, 1317 and 4,852 observations respectively.\n\n36 The original primary product and resource-based product data in Lall (2000) are in SITC 3-digit level.\n\nWe concord them to HS 6-digit. The original ATP data are at 4-digit Chinese Industrial Classification\n\n(CIC). We first concord CIC to 4-digit ISIC, and then to HS 6-digit. Although we have access to other\n\nclassifications of high-tech products such as those in Lall (2000) and Ferrantino et al. (2007), we\n\nchoose to use China’s own official definition because that is what the rebate rates are actually based\n\non.\n\n\n26\n\nMention: \"ATP data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"Originally at four-digit national classification, concorded to six-digit trade codes.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006304:3:0", "occurrence_id": "86e438fe5ab5698345e6079c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ##### **_Introduction_**\n\nThe ratio of boys to girls rose in the South Caucasus countries in the 1990s to levels second only\nto China. Many parents have clearly practiced sex-selection to ensure that they have a boy. The\nphenomenon was not manifest in the data from preceding decades, and has thus generated a good\ndeal of debate about what may underlie it and above all what policies might help reduce it.\n\n\nThis paper begins by describing the levels and trends in sex-selection in these countries, based on\nthe country census and vital registration data, and survey data. <sup>2</sup> After sharp rises in child sex\nratios, the trend seems to be turning.\n\n\nSection 2 discusses whether sex-selection would be expected to be found in this region. On the\none hand, traditional kinship systems in the region are similar to those of other regions with\nstrong son preference. On the other hand, the region has been over 50 percent urbanized for\ndecades, and has a high level of gender equity along many dimensions: characteristics that have\nhelped reduce sex-selection elsewhere in Asia.\n\n\nWhat then might have caused the sudden rise in sex-selection in this region? As discussed in\nSection 3, the timing indicates it is related to the severe economic and other shocks experienced\nafter the dissolution of the USSR. Similar responses have been seen in China in response to\nsevere stresses. In the 1990s, an economic collapse in this region led to heavy out-migration in\nsearch of work, and sharp fertility decline. There was much forced migration as the newlyindependent countries re-aligned themselves along ethnic lines. Territorial conflicts added to the\nupheavals. Systems of service delivery, pensions, and safety nets, that had been highly\ndependent on transfers from other parts of the USSR, were badly affected. Levels of poverty\nwere very high in the early 2000s. Under these circumstances, one coping mechanism is to fall\nback on\n\nMention: \"country census and vital registration data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Levels and trends in child sex selection are described from these records.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006312:11:0", "occurrence_id": "cacd97392360efb9322e25e7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: We used the _Pesquisa Nacional por Amostra Domiciliar_ database – PNAD – (National\n\n\nHousehold Sample Survey) published by _Instituto Brasileiro de Geografia e Estatística_ - IBGE \n\n(Brazilian Institute of Geography and Statistics) annually in Brazil. The PNAD collects socioeconomic\n\n\ndata on individuals in a sample of households in Brazil. In addition to the collection of permanent\n\n\nvariables (work characteristics, education, income, and housing), the PNAD periodically conducts\n\n\nadditional research to include other topics such as healthcare characteristics and characteristics of\n\n\nother education types (technical-vocational education, for example). The PNAD is a sample survey\n\n\nbut is representative of Brazil. In 2007, the PNAD conducted a supplementary survey that sought to\n\n\nverify the technical-vocational education of the Brazilian population. Hence, we can examine the\n\n\nportion of the population who attends or has attended any technical-vocational course, as well as\n\n\nthe course type and other features. A significant portion (23.8%) of the Brazilian population <sup>6</sup> [^6: Population age 14 or older.] was\n\n\nattending or had already attended technical-vocational education in 2007.\n\n\nWe consider three types of technical-vocational education that we can observe in the data: i)\n\n\ntechnical education at the upper secondary level; ii) technical educational at the tertiary level; and iii)\n\n\nshort term training courses. Technical education courses at the upper secondary level (which will be\n\n\ncalled TEC for the remainder of the paper) can be attended by students who are pursuing or have\n\n\ncompleted a general education course and are further divided into subsequent, concurrent and\n\n\nintegrated courses. Subsequent courses are technical courses that require the completion of general\n\n\nupper secondary education for attendance. Concurrent courses are technical courses that take place\n\n\nduring the same period in which the individual pursues general upper secondary education, but they\n\n\nusually take place at different schools. Integrated courses include general upper secondary education\n\n\nand technical education courses in the same curriculum and\n\nMention: \"National\n\n\nHousehold Sample Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Published annually by the national statistics institute.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006312:11:0", "occurrence_id": "532abd2431f19819f1466c78", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: We used the _Pesquisa Nacional por Amostra Domiciliar_ database – PNAD – (National\n\n\nHousehold Sample Survey) published by _Instituto Brasileiro de Geografia e Estatística_ - IBGE \n\n(Brazilian Institute of Geography and Statistics) annually in Brazil. The PNAD collects socioeconomic\n\n\ndata on individuals in a sample of households in Brazil. In addition to the collection of permanent\n\n\nvariables (work characteristics, education, income, and housing), the PNAD periodically conducts\n\n\nadditional research to include other topics such as healthcare characteristics and characteristics of\n\n\nother education types (technical-vocational education, for example). The PNAD is a sample survey\n\n\nbut is representative of Brazil. In 2007, the PNAD conducted a supplementary survey that sought to\n\n\nverify the technical-vocational education of the Brazilian population. Hence, we can examine the\n\n\nportion of the population who attends or has attended any technical-vocational course, as well as\n\n\nthe course type and other features. A significant portion (23.8%) of the Brazilian population <sup>6</sup> [^6: Population age 14 or older.] was\n\n\nattending or had already attended technical-vocational education in 2007.\n\n\nWe consider three types of technical-vocational education that we can observe in the data: i)\n\n\ntechnical education at the upper secondary level; ii) technical educational at the tertiary level; and iii)\n\n\nshort term training courses. Technical education courses at the upper secondary level (which will be\n\n\ncalled TEC for the remainder of the paper) can be attended by students who are pursuing or have\n\n\ncompleted a general education course and are further divided into subsequent, concurrent and\n\n\nintegrated courses. Subsequent courses are technical courses that require the completion of general\n\n\nupper secondary education for attendance. Concurrent courses are technical courses that take place\n\n\nduring the same period in which the individual pursues general upper secondary education, but they\n\n\nusually take place at different schools. Integrated courses include general upper secondary education\n\n\nand technical education courses in the same curriculum and\n\nMention: \"PNAD\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"It collects socioeconomic data on individuals in sampled households.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006318:21:2", "occurrence_id": "83a4c0775f0ee42fd04f7ee3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |9|10|10|\n\n\n\nVery insecure Insecure\n\n\nSecure Very secure\n\n\n\n**How secure do you feel alone**\n\n**on the street at night?**\n\n\n\n|Col1|Col2|Col3|2|Col5|5|Col7|6|Col9|\n|---|---|---|---|---|---|---|---|---|\n||24||14<br>||43<br>5||8<br>6||\n||24||58|58|58|58|52|52|\n||24|24|24|24|24|24|24|24|\n||32|32|32|32|30|30|30|30|\n||20|20|26|26|23|23|34|34|\n\n\nVery insecure Insecure\n\n\nSecure Very secure\n\n\n\n\n\n\n\n\n\n\n\n\n\nSource: Listening to Displaced People Survey, 2014.\n\n\n15 Victims of the crisis refers to death as well as physical violence.\n16 Respondents in our survey have been exposed to less violence than the average population in northern Mali.\nComparing our results to Afrobarometer findings, the latter finds that 15% of the population in Timbuktu and 8% in\nGao reported death in their family as a result of the conflict. In our survey, the 5% of those originating from Timbuktu\ndo so and 7% for those from Gao.\n\n\n20\n\nMention: \"Listening to Displaced People Survey, 2014\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Perceptions of security and crisis victimisation come from this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006319:7:0", "occurrence_id": "8af336f63c9cd0ab5eb4a784", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: market and the legal system. The survey also contains information on firm ownership, sales,\n\n\nemployees, top manager experience, whether a firm is an exporter, and firm age.\n\n\nWe start with all observations in WBES, and we proceed to delete firms that do not answer\n\n\nthe audit question: “In the last fiscal year, did this establishment have its annual financial statement\n\n\nchecked and certified by an external auditor?” We also delete (4,330) publicly listed firms because they\n\n\nare required to have their annual financial statements audited. Thus, auditing is not a choice that can\n\n\nbe made by these firms. Our final sample consists of 71,677 firms in 121 countries from 2006 to 2014. <sup>7</sup> [^7: The dataset includes firms from 15 different industries according to the international industry classification codes\n(ISIC).]\n\n\nAmong these, 47% of firms had their annual financial statements checked and certified by external\n\n\nauditors.\n\n\nA common criticism of using survey data to conduct research related to firm performance and\n\n\ngrowth is that survey data are self-reported and therefore the findings may suffer from self-reporting\n\n\nbias. However, accounting data are more likely to be biased than survey data as the incentives to distort\n\n\ndata are likely to be higher in financial statements because many firm-level decisions, such as tax,\n\n\nfinancing, and managerial compensations, are in part based on variables from financial statements\n\n\n(Beck Demirgüç‐Kunt and Maksimovic, 2005). Furthermore, the self-reporting nature of the WBES\n\n\ndata is not likely to be a significant source of bias. The survey aims to evaluate the business\n\n\nenvironment instead of firm performance and growth. Even though some firm performance and\n\n\ngrowth related questions were asked, the survey was specifically designed to ask those questions at the\n\n\nend of the interview. This reduces the respondents’ need to justify their own performance when\n\n\nanswering the earlier questions about the business environment.\n\n\nThroughout our analysis, we control for country level governance and macroeconomic\n\n\nvariables. Country level financial development data are\n\nMention: \"WBES\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The starting observation set before dropping non-respondents to the audit question.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006319:7:0", "occurrence_id": "4d15ca948a50b2397c594c46", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: market and the legal system. The survey also contains information on firm ownership, sales,\n\n\nemployees, top manager experience, whether a firm is an exporter, and firm age.\n\n\nWe start with all observations in WBES, and we proceed to delete firms that do not answer\n\n\nthe audit question: “In the last fiscal year, did this establishment have its annual financial statement\n\n\nchecked and certified by an external auditor?” We also delete (4,330) publicly listed firms because they\n\n\nare required to have their annual financial statements audited. Thus, auditing is not a choice that can\n\n\nbe made by these firms. Our final sample consists of 71,677 firms in 121 countries from 2006 to 2014. <sup>7</sup> [^7: The dataset includes firms from 15 different industries according to the international industry classification codes\n(ISIC).]\n\n\nAmong these, 47% of firms had their annual financial statements checked and certified by external\n\n\nauditors.\n\n\nA common criticism of using survey data to conduct research related to firm performance and\n\n\ngrowth is that survey data are self-reported and therefore the findings may suffer from self-reporting\n\n\nbias. However, accounting data are more likely to be biased than survey data as the incentives to distort\n\n\ndata are likely to be higher in financial statements because many firm-level decisions, such as tax,\n\n\nfinancing, and managerial compensations, are in part based on variables from financial statements\n\n\n(Beck Demirgüç‐Kunt and Maksimovic, 2005). Furthermore, the self-reporting nature of the WBES\n\n\ndata is not likely to be a significant source of bias. The survey aims to evaluate the business\n\n\nenvironment instead of firm performance and growth. Even though some firm performance and\n\n\ngrowth related questions were asked, the survey was specifically designed to ask those questions at the\n\n\nend of the interview. This reduces the respondents’ need to justify their own performance when\n\n\nanswering the earlier questions about the business environment.\n\n\nThroughout our analysis, we control for country level governance and macroeconomic\n\n\nvariables. Country level financial development data are\n\nMention: \"WBES\n\n\ndata\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Its self-reporting nature is unlikely to be a significant bias source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006292:31:1", "occurrence_id": "2e5d39a8341acaccb227f50a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: don (2013) study gender bias in\neducation expenditure and show that in the period 1993–2005, when girls’ education caught up to boys’, households\nstill spent less on girls, primarily by sending girls to public schools and boys to private schools. Girls are thus likely to\nreceive lower quality schooling than boys, even if they attain the same level of education. Unfortunately we do not\nobserve the type of institution a woman attended or any other indicator of the quality of education in the NSS data.\n24 As discussed in Section 2 some recent studies show that (girls’) enrollment responds to the growth of jobs in IT and\nIT enabled services (Jensen 2012; Shastry 2012; Oster and Millett Steinberg 2013), but this is a very recent\ndevelopment and this sector still accounts for a very small share of employment. We are not aware of any research on\n\n30\n\nMention: \"NSS data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"It carries no indicator of institution quality.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006324:42:14", "occurrence_id": "17776b47c5f32863750f435f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: /u> <u>0.143</u> <u>0.294</u>\n\n\n\n_(0.292)_ _(0.273)_ _(0.426)_ _(0.195)_ _(0.116)_ _(0.232)_ _(0.176)_ _(0.141)_ _(0.276)_\n\n\n\n_(0.160)_ _(0.151)_ _(0.234)_ _(0.105)_ _(0.063)_ _(0.170)_ _(0.067)_ _(0.054)_ _(0.107)_\n\n\n\n_<u>Source: Household Budget Surveys (HBS) for 2001, 2007 and 2011/12. Numbers in parentheses are Standard deviations.</u>_\n\n\n\n41\n\nMention: \"Household Budget Surveys (HBS) for 2001, 2007 and 2011/12\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Estimates with their standard deviations are computed across these rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006332:17:1", "occurrence_id": "77cd1aadb83ce550453ff46c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  capita, at constant 2000 US dollars ( _Income_ ). We use\n\n\naverage values of the variable over the period 2007 and 2009 taken from World Development\n\n\nIndicators (WDI), World Bank. The variable serves as a useful proxy measure for various aspects\n\n\nof economic development that could be correlated with both women managers and women’s access\n\n\nto education.\n\n\nIt is natural to expect that ceteris paribus, countries with proportionately more women in\n\n\nthe population to have proportionately more women in top management. At the same time,\n\n\npopulations with more women could also reflect more favorable attitudes towards female children\n\n16\n\nMention: \"World Development\n\n\nIndicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Averages over these two years serve as a proxy measure.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006335:21:0", "occurrence_id": "04d5ef94b212553069951f57", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: we estimate average road speed by road condition from estimates for 12,322 road segments in\n\n\nthe DRC produced by Ali, et al. (2015).\n\n\n**Figure 5: Kasese, Maniema Province, DRC**\n\n\nTable 4 presents our results, along with derived travel time multipliers for a road segment\n\n\nwhose initial condition is very poor. In this condition, average travel speed on the segment is 24\n\n\nkm/hr. Progressive improvements in the segment’s condition increase average speed to 36, 55\n\n\nand 73 km/hr, respectively. With increased speed, travel time to the nearest urban center falls to\n\n\n75%, 49% and 33% of the initial time. <sup>5</sup> [^5: A reduction in travel time should also produce an increase in the opportunity value of land in the road corridor.\nThe opportunity value measure employed for this exercise is based on physical measures (e.g., soil quality) and does\nnot incorporate the effect of distance to market. We are therefore unable to estimate the effect of road quality\nimprovement on this variable. By implication, our results should be interpreted as conservative estimates of road\nimprovement impacts on forest clearing.]\n\n\n**Table 4: Road condition and travel time to nearest urban center**\n\n\n\n\n\n\n\n\n\n\n|Road<br>Condition|Average<br>Speed<br>(km/hr)|Travel<br>Time<br>Multiplier|\n|---|---|---|\n|<br>**1 **|<br>**24**|<br>**1.00**|\n|<br>**2 **|<br>**36**|<br>**0.75**|\n|<br>**3 **|<br>**55**|<br>**0.49**|\n|\n\nMention: \"Ali, et al. (2015)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its estimates across 12,322 road segments give speed responses to condition.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006337:40:1", "occurrence_id": "a8ecfffb152d21ad40c8a092", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  further details.|<br>_Note:_ The figure plots the predicted probability of private school existence against the number of years that different types<br>of schools have existed in the village. The predicted probability is based on a probit regression, with all schools included in a<br>single regression. The data area based on the census of private schools (Federal Bureau of Statistics) matched to public<br>schools (Educational Management Information System for Punjab). See Andrabi, Das and Khwaja 2013 for further details.|\n\n\n\n39\n\nMention: \"census of private schools\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Enumerated private schools matched to public school records.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006337:40:1", "occurrence_id": "506f7b8597bf69daeb7c8d40", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  further details.|<br>_Note:_ The figure plots the predicted probability of private school existence against the number of years that different types<br>of schools have existed in the village. The predicted probability is based on a probit regression, with all schools included in a<br>single regression. The data area based on the census of private schools (Federal Bureau of Statistics) matched to public<br>schools (Educational Management Information System for Punjab). See Andrabi, Das and Khwaja 2013 for further details.|\n\n\n\n39\n\nMention: \"Educational Management Information System for Punjab\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The administrative record of public schools used in the match.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006338:10:3", "occurrence_id": "707b70df5302ecee4b27dfa2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  fees)\n\ndivided by the asset replacement cost.\n\n\nOutput is measured by total sales in U.S. dollars. Ideally, this should be measured by some\n\nvariable in real terms, but this is only the usable common proxy for output in our BEEPS\n\ndata. <sup>4</sup> To control for sector heterogeneity, the empirical model incorporates the sector\n\n4 No physical output variable is available in the database, because firms in the sample engage in various\nbusinesses.\n\nMention: \"BEEPS\n\ndata\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its real-terms proxy is the only usable output measure available.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006346:16:0", "occurrence_id": "8e6fc2dabf005209c15a9907", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Business indicators apply to a standardized case and firm, and emphasize formal, required procedures.\n\nActual implementation may differ from formal procedures, however; e.g. the indicators assume no\n\nbribes are paid or personal connections exploited to bypass or expedite any procedures. Moreover,\n\nthe standardized case and firm (necessary for cross-country comparability) cannot possibly reflect\n\nreality for firms of all sizes, locations, and industries. For example, the standardized firm is assumed\n\nto be located in a country’s largest city. In a small number of countries, the Doing Business indicators\n\nhave been measured for multiple cities, and often show large differences between them. Other\n\nindicators based on expert perceptions may emphasize priorities and knowledge of Western-based\n\nadvocacy NGOs or commercial risk assessment firms, which may incorporate certain biases or neglect\n\nconcerns or views of domestic investors (Arndt and Oman, 2006).\n\n\nWhen the explanatory variable is measured with error (say _e_ ), the error, _e_, is embedded in the\n\nresidual. But the observed explanatory variable also contains _e_, causing a correlation of the regression\n\nerror term and the explanatory variable, and results in general in attenuation bias. The bias is especially\n\nsevere when the variability of _e_ is large—a case of particular concern for us if institutional variables\n\ntend to have large measurement errors. We attempt to correct for measurement error using an\n\ninstrumental variable approach (Greene, 1997, p.437). Specifically, we identify a second institutional\n\nvariable that is strongly related to the mis-measured variable, but with measurement error that is\n\nuncorrelated to the measurement error in the first variable. In our context, we have natural candidates\n\nfor such variables: our multiple indicators of contracting and property rights institutions are produced\n\nby different sources that reflect different methods and information sets. In implementing this\n\napproach for any particular institutional measure, we choose an instrument that has similar theoretical\n\ncontent, a relatively small incidence of missing values,\n\nMention: \"Doing Business indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"In some economies they are measured for several cities, showing large gaps.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006348:7:0", "occurrence_id": "4027f1ad5901ac5876737eae", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |Canada|Reform leader|Westminster<br>Common law / federalist|\n|---|---|---|\n|**Netherlands**|Recent MTEF reform|Codified<br>Germanic|\n\n\nFollowing an overview in Part 2 that surveys the design and implementation of MTEFs, the\n\npaper is structured around the key areas of MTEF practice: institutional arrangements and\n\nprocedures (Part 3), program costing frameworks and practices (Part 4), and budget and\n\nforward years cost estimates (Part 5). Part 6 is a concluding section that provides observations\n\non the current status of MTEF costing practices as reported by finance ministries. Readers who\n\nare familiar with MTEF concepts and practices can skip Part 2 and go straight to Part 3.\n\n#### **2. Defining and implementing MTEF**\n\n\n**2.1** **Defining a medium term expenditure framework**\n\nMTEF is one of the key budget planning and management practices that define contemporary\n\npublic financial management reform in advanced countries. In its 2012 budget practices survey,\n\nfor example, the OECD reported that nearly 90% of advanced economies had implemented\n\nsome form of MTEF as part of their national budgeting framework (see OECD International\n\nDatabase of Budget Practices and Procedures 2012).\n\n\nIn conjunction with other PFM developments, the MTEF is an important way of addressing\n\nsome of the key shortcomings in annual budgeting. An annual budget for revenues and\n\nexpenditures often does not take account of the consequences of current year budget decisions\n\nin subsequent years. To better understand the fiscal impact of policies, to more effectively set\n\ncurrent and future policy priorities, and to exercise control over budget aggregates, an MTEF\n\nintegrates the annual budget formulation cycle with a medium-term planning process.\n\n\nAn MTEF is therefore not simply the documentation of multiyear estimates of revenue and\n\nexpenditures alongside the annual budget, but rather ‘all the systems,\n\nMention: \"OECD International\n\nDatabase of Budget Practices and Procedures 2012\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It records which economies had implemented a medium-term expenditure framework.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006350:6:1", "occurrence_id": "87a9d773cf79ca7c37e9d589", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  inequality measure given the several desired properties. <sup>5</sup> [^5: Mean log deviation is defined as: Io =\n<u>1</u>] We\nclosely follow Nieheus and Peichl (2012) and use a parametric approach to estimate the IOL and IOR of\ninterest. Furthermore, we take advantage of the longitudinal nature of the RLMS and estimate the lower\nand upper bounds of the share of wage inequality explained by inequality of opportunity.\n\n\n_Lower bound._ Based on the parametric approach of Ferreira and Gignoux (2011), we estimate the impact\nof our circumstance variables. Log linearization of an equation such as (4) would yield the following\nspecifications:\n\n\n\n(7)\n\n(8)\n\n𝑦𝑦𝑖𝑖 = 𝛼𝛼𝐶𝐶𝑖𝑖 + 𝛽𝛽𝐸𝐸𝑖𝑖 + 𝜗𝜗𝑖𝑖\n\n𝐸𝐸𝑖𝑖 = 𝐻𝐻𝐶𝐶𝑖𝑖 + 𝜔𝜔𝑖𝑖\n\n\n\n(7)\n\n(8)\n\n= 𝛼𝛼𝐶𝐶𝑖𝑖 + 𝛽𝛽𝐸𝐸𝑖𝑖 + 𝜗𝜗𝑖𝑖\n\n𝐸𝐸𝑖𝑖 = 𝐻𝐻𝐶𝐶𝑖𝑖 + 𝜔𝜔𝑖𝑖\n\n\n\nwhere equation (8) represents the indirect effect of circumstances on the outcome through effort. To\ncompute IOL and IOR as described above, we can estimate by OLS the reduced-form equation:\n\n\n\ncompute IOL and IOR as described above, we can estimate by OLS the reduced-form equation:\n\n\n(9)\n𝑦𝑦𝑖𝑖 = 𝜑𝜑𝐶𝐶𝑖𝑖 + 𝜀𝜀𝑖𝑖\n\n**𝛽**\n\n\n\noutcomes, combining the direct and indirect effects. To construct the parametric estimate of the smoothed\nwhere 𝜑𝜑= 𝛼𝛼+ 𝛽 **𝛽** and 𝜀𝜀𝑖𝑖 = 𝛽𝛽𝜗𝜗𝑖𝑖 + 𝜔𝜔𝑖𝑖. The estimates 𝜑𝜑� measure the overall effect of circumstances on\ndistribution, we then calculate the predicted outcome:\n\n\n\n**𝛽**\n\n\n(10)\n𝑦𝑦�𝑖𝑖,𝐿 = 𝜑𝜑�𝐶𝐶𝑖𝑖\n\n\n\n4 An ex-post approach is based on identifying individuals with the same level of effort and then estimating the\ninequality in outcomes variables and then measure how outcomes vary by types. See (Ferreira and Gignoux, 2009)\n\n𝑁 𝑦 𝑖\n\n\n\n<u>1<\n\nMention: \"the RLMS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Its longitudinal design allows bounding the explained share of wage inequality.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006359:7:0", "occurrence_id": "883412a5b5541a620e527cea", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: This study uses the World Health Organization <mark>Child Growth Standards</mark> for the classification of\nstunting, wasting and underweight status. Children are classified as stunted, wasted and\nunderweight if their height-to-age Z-score, weight-to-height Z-score and weight-to-age Z-score\nare below -2 respectively. Children are classified as severe stunted, severe wasted and severe\nunderweight if their height-to-age Z-score, weight-to-height Z-score and weight-to-age Z-score\nare below -3 respectively (WHO Multicentre Growth Reference Study Group 2006).\n\n\nData from two parts of the survey were combined to estimate nutritional intake: the household\nstocks data and the personal diary data. The household stocks of food data, involved an interviewer\nasking the household head or representative about food currently in the house on day 1 and day 14\nof the survey. The stocks were weighed by the interviewer. Nineteen food items, identified as the\nnineteen most common items in the previous HIES, were specified on the questionnaire. Other\nitems were found, weighed and listed. The second source of data was derived from the participant’s\nrecorded transactions in a personal diary for 14 days. A total number of 295,804 transactions\ninvolving food and alcohol were extracted from this diary.\n\n\nThe quantity of food was converted to calories using food composition information from the New\nZealand FOODfiles 2012 database (2013). Energy and nutrient composition for each item was\nestimated by taking an average of relevant items from the FOODfiles database. The difference\nbetween stock levels on day1 and day14, plus additional purchased foods were converted to the\ntotal calories consumed by the households.\n\n\nTotal calories consumed by households during the two weeks were then converted to calories per\ncapita (adult equivalent) per day. Per capita is defined by adult equivalent using children 0-14 as\n0.5 adult. Guests who regularly have meals in the\n\nMention: \"HIES\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its nineteen most common food items defined the questionnaire.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006359:7:0", "occurrence_id": "6c528e29007afc99afb24f6d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: This study uses the World Health Organization <mark>Child Growth Standards</mark> for the classification of\nstunting, wasting and underweight status. Children are classified as stunted, wasted and\nunderweight if their height-to-age Z-score, weight-to-height Z-score and weight-to-age Z-score\nare below -2 respectively. Children are classified as severe stunted, severe wasted and severe\nunderweight if their height-to-age Z-score, weight-to-height Z-score and weight-to-age Z-score\nare below -3 respectively (WHO Multicentre Growth Reference Study Group 2006).\n\n\nData from two parts of the survey were combined to estimate nutritional intake: the household\nstocks data and the personal diary data. The household stocks of food data, involved an interviewer\nasking the household head or representative about food currently in the house on day 1 and day 14\nof the survey. The stocks were weighed by the interviewer. Nineteen food items, identified as the\nnineteen most common items in the previous HIES, were specified on the questionnaire. Other\nitems were found, weighed and listed. The second source of data was derived from the participant’s\nrecorded transactions in a personal diary for 14 days. A total number of 295,804 transactions\ninvolving food and alcohol were extracted from this diary.\n\n\nThe quantity of food was converted to calories using food composition information from the New\nZealand FOODfiles 2012 database (2013). Energy and nutrient composition for each item was\nestimated by taking an average of relevant items from the FOODfiles database. The difference\nbetween stock levels on day1 and day14, plus additional purchased foods were converted to the\ntotal calories consumed by the households.\n\n\nTotal calories consumed by households during the two weeks were then converted to calories per\ncapita (adult equivalent) per day. Per capita is defined by adult equivalent using children 0-14 as\n0.5 adult. Guests who regularly have meals in the\n\nMention: \"New\nZealand FOODfiles 2012 database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Food composition information converts quantities to calories.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006359:7:0", "occurrence_id": "d31455a026602b9d7ce4e344", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: This study uses the World Health Organization <mark>Child Growth Standards</mark> for the classification of\nstunting, wasting and underweight status. Children are classified as stunted, wasted and\nunderweight if their height-to-age Z-score, weight-to-height Z-score and weight-to-age Z-score\nare below -2 respectively. Children are classified as severe stunted, severe wasted and severe\nunderweight if their height-to-age Z-score, weight-to-height Z-score and weight-to-age Z-score\nare below -3 respectively (WHO Multicentre Growth Reference Study Group 2006).\n\n\nData from two parts of the survey were combined to estimate nutritional intake: the household\nstocks data and the personal diary data. The household stocks of food data, involved an interviewer\nasking the household head or representative about food currently in the house on day 1 and day 14\nof the survey. The stocks were weighed by the interviewer. Nineteen food items, identified as the\nnineteen most common items in the previous HIES, were specified on the questionnaire. Other\nitems were found, weighed and listed. The second source of data was derived from the participant’s\nrecorded transactions in a personal diary for 14 days. A total number of 295,804 transactions\ninvolving food and alcohol were extracted from this diary.\n\n\nThe quantity of food was converted to calories using food composition information from the New\nZealand FOODfiles 2012 database (2013). Energy and nutrient composition for each item was\nestimated by taking an average of relevant items from the FOODfiles database. The difference\nbetween stock levels on day1 and day14, plus additional purchased foods were converted to the\ntotal calories consumed by the households.\n\n\nTotal calories consumed by households during the two weeks were then converted to calories per\ncapita (adult equivalent) per day. Per capita is defined by adult equivalent using children 0-14 as\n0.5 adult. Guests who regularly have meals in the\n\nMention: \"FOODfiles database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Averages across relevant items give each item's energy and nutrient content.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006360:4:0", "occurrence_id": "9142f7a9a555a96bea7ea143", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 2013 Report (World health Organization 2013) stated that the current tobacco smoking rate for\nmale adults was 60.3% and 27.0% for female adults, 55.4% for male youth and 40.3% for female\nyouth, based on Papua New Guinea STEPwise Surveillance (STEPS) for Noncommunicable\nDiseases conducted in 2007.\n\n\nThis paper uses the national representative sample to look into the current smoking patterns in\nPapua New Guinea. In summary, the results show that tobacco consumption imposes a significant\nfinancial burden to households, particularly to poor households. Tobacco consumption accounts\nfor about 7% of total household expenditure and about 23% of total household food expenditure\nfor the poorest quartile. Adults who cannot read are more likely to smoke. Household wealth status\nis significantly correlated with tobacco choices but not smoking prevalence. Smokers from the\npoorest quartile are more likely to consume non-processed tobacco, while smokers from the richest\nquartile are more likely to consume cigarettes.\n\n\n**2. DATA AND METHODS**\n\n**2.1 Data**\n\nThis paper uses the data from the 2009-2010 Papua New Guinea Household Income and\nExpenditure Survey (2009-2010 PNG HIES). The 2009-2010 HIES is the first comprehensive and\nnationally representative survey of the socioeconomic status of PNG households since the 1996\nHousehold Survey of PNG conducted by Unisearch from the University of Papua New Guinea\n(UPNG) and the Institute of National Affairs (INA).\n\n\nThe 2009-2010 PNG HIES contains the final cross-section sample of 4,191 households. The\nsample of households was selected from a nationally representative sample frame. A\ncomprehensive set of multi-topic questionnaires was designed to elicit information on key topics\nsuch as family demography, education, health, employment, and consumption. In addition to the\nstandard questions on current smoking and ever smoking status, the survey also collects more indepth information on cigarette type and the number of cigarettes consumed daily for individuals\nwho reported as\n\nMention: \"Papua New Guinea STEPwise Surveillance\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its surveillance gives adult and youth prevalence rates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006360:4:0", "occurrence_id": "1262568ba4f6bcca4e22703d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 2013 Report (World health Organization 2013) stated that the current tobacco smoking rate for\nmale adults was 60.3% and 27.0% for female adults, 55.4% for male youth and 40.3% for female\nyouth, based on Papua New Guinea STEPwise Surveillance (STEPS) for Noncommunicable\nDiseases conducted in 2007.\n\n\nThis paper uses the national representative sample to look into the current smoking patterns in\nPapua New Guinea. In summary, the results show that tobacco consumption imposes a significant\nfinancial burden to households, particularly to poor households. Tobacco consumption accounts\nfor about 7% of total household expenditure and about 23% of total household food expenditure\nfor the poorest quartile. Adults who cannot read are more likely to smoke. Household wealth status\nis significantly correlated with tobacco choices but not smoking prevalence. Smokers from the\npoorest quartile are more likely to consume non-processed tobacco, while smokers from the richest\nquartile are more likely to consume cigarettes.\n\n\n**2. DATA AND METHODS**\n\n**2.1 Data**\n\nThis paper uses the data from the 2009-2010 Papua New Guinea Household Income and\nExpenditure Survey (2009-2010 PNG HIES). The 2009-2010 HIES is the first comprehensive and\nnationally representative survey of the socioeconomic status of PNG households since the 1996\nHousehold Survey of PNG conducted by Unisearch from the University of Papua New Guinea\n(UPNG) and the Institute of National Affairs (INA).\n\n\nThe 2009-2010 PNG HIES contains the final cross-section sample of 4,191 households. The\nsample of households was selected from a nationally representative sample frame. A\ncomprehensive set of multi-topic questionnaires was designed to elicit information on key topics\nsuch as family demography, education, health, employment, and consumption. In addition to the\nstandard questions on current smoking and ever smoking status, the survey also collects more indepth information on cigarette type and the number of cigarettes consumed daily for individuals\nwho reported as\n\nMention: \"2009-2010 Papua New Guinea Household Income and\nExpenditure Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The first comprehensive nationally representative round since the mid-1990s.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006360:4:0", "occurrence_id": "1f28495c5af648bc1e301bd5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 2013 Report (World health Organization 2013) stated that the current tobacco smoking rate for\nmale adults was 60.3% and 27.0% for female adults, 55.4% for male youth and 40.3% for female\nyouth, based on Papua New Guinea STEPwise Surveillance (STEPS) for Noncommunicable\nDiseases conducted in 2007.\n\n\nThis paper uses the national representative sample to look into the current smoking patterns in\nPapua New Guinea. In summary, the results show that tobacco consumption imposes a significant\nfinancial burden to households, particularly to poor households. Tobacco consumption accounts\nfor about 7% of total household expenditure and about 23% of total household food expenditure\nfor the poorest quartile. Adults who cannot read are more likely to smoke. Household wealth status\nis significantly correlated with tobacco choices but not smoking prevalence. Smokers from the\npoorest quartile are more likely to consume non-processed tobacco, while smokers from the richest\nquartile are more likely to consume cigarettes.\n\n\n**2. DATA AND METHODS**\n\n**2.1 Data**\n\nThis paper uses the data from the 2009-2010 Papua New Guinea Household Income and\nExpenditure Survey (2009-2010 PNG HIES). The 2009-2010 HIES is the first comprehensive and\nnationally representative survey of the socioeconomic status of PNG households since the 1996\nHousehold Survey of PNG conducted by Unisearch from the University of Papua New Guinea\n(UPNG) and the Institute of National Affairs (INA).\n\n\nThe 2009-2010 PNG HIES contains the final cross-section sample of 4,191 households. The\nsample of households was selected from a nationally representative sample frame. A\ncomprehensive set of multi-topic questionnaires was designed to elicit information on key topics\nsuch as family demography, education, health, employment, and consumption. In addition to the\nstandard questions on current smoking and ever smoking status, the survey also collects more indepth information on cigarette type and the number of cigarettes consumed daily for individuals\nwho reported as\n\nMention: \"2009-2010 PNG HIES\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The short-name reference to the same round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006360:4:0", "occurrence_id": "e6e6cdf7de38d8ca2cb6c910", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 2013 Report (World health Organization 2013) stated that the current tobacco smoking rate for\nmale adults was 60.3% and 27.0% for female adults, 55.4% for male youth and 40.3% for female\nyouth, based on Papua New Guinea STEPwise Surveillance (STEPS) for Noncommunicable\nDiseases conducted in 2007.\n\n\nThis paper uses the national representative sample to look into the current smoking patterns in\nPapua New Guinea. In summary, the results show that tobacco consumption imposes a significant\nfinancial burden to households, particularly to poor households. Tobacco consumption accounts\nfor about 7% of total household expenditure and about 23% of total household food expenditure\nfor the poorest quartile. Adults who cannot read are more likely to smoke. Household wealth status\nis significantly correlated with tobacco choices but not smoking prevalence. Smokers from the\npoorest quartile are more likely to consume non-processed tobacco, while smokers from the richest\nquartile are more likely to consume cigarettes.\n\n\n**2. DATA AND METHODS**\n\n**2.1 Data**\n\nThis paper uses the data from the 2009-2010 Papua New Guinea Household Income and\nExpenditure Survey (2009-2010 PNG HIES). The 2009-2010 HIES is the first comprehensive and\nnationally representative survey of the socioeconomic status of PNG households since the 1996\nHousehold Survey of PNG conducted by Unisearch from the University of Papua New Guinea\n(UPNG) and the Institute of National Affairs (INA).\n\n\nThe 2009-2010 PNG HIES contains the final cross-section sample of 4,191 households. The\nsample of households was selected from a nationally representative sample frame. A\ncomprehensive set of multi-topic questionnaires was designed to elicit information on key topics\nsuch as family demography, education, health, employment, and consumption. In addition to the\nstandard questions on current smoking and ever smoking status, the survey also collects more indepth information on cigarette type and the number of cigarettes consumed daily for individuals\nwho reported as\n\nMention: \"1996\nHousehold Survey of PNG\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"The prior national survey of household socioeconomic status.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006367:9:0", "occurrence_id": "3b459cd82c6d1b34e2c5322b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **figure 2.** **Evolution of parity fertility by cohort in Turkey**\n\n\nData Source: Census (i-pums) 2000\n\n\nFigure 3 confirms a low and stable proportion of childless women for the 1969 to 1974 cohort (SILC\n2011) but suggests an increase in the proportion of women having one and two children in comparison\nto older cohorts illustrated in figure 2. The proportion of women havingr three children stagnates, while\nthe proportion of women having at least four children has decreased (note however that the\nunderestimation bias in the SILC data leads to an overemphasis of this decline). Most importantly, figure\n3 illustrates that the majority of Turkish women of cohorts 1969 to 1974 either have two or three\nchildren. Just like on average in Europe, having two children is the norm for cohorts born around 1970\nin Turkey, which was not the case for younger cohorts born before 1960. At the same time, the\nproportion of women having three or more children is larger in Turkey than on average in other European\ncountries.\n\n\nWhen calculating parity progression ratios <sup>5</sup> [^5: Parity progression ratio (idem “transition probability”) from zero to one child: 89% Turkey and 85% EU; from three to four children: 40% in\nTurkey and 6% EU.] based on the proportions illustrated in figure 3, we find that\n80% of women having one child pass on to have a second child in Turkey, while the EU(26) average is\n72%. Among women having two children, 47% are likely to have a third child in Turkey, against only\n23% on average in the EU (26) as reported by SILC.\n\n\nFurther calculations based on the proportions of figure 3 reveal that almost half of the gap in approximate\ncompleted fertility between the EU(26) and Turkey (1.7 vs. 2.3 children per women) can be explained\nby an overrepresentation of women having at least two children in Turkey\n\nMention: \"SILC data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Its underestimation bias overemphasises the decline in larger families.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006368:20:0", "occurrence_id": "d68f1175aaec267185c92b2e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **_The main causes for unexpected clearance delays are inadequate documentation, congestion_**\n\n\n**_at the border, and trading in goods that require lengthy inspections_**\n\n\nRegarding the main causes of unexpected delays in the clearance process, a large\n\n\nmajority of freight forwarders indicate that inadequate documentation (29%), or congestion at\n\n\nthe border (25%), or trading in goods that require lengthy inspections (21%) are the first main\n\n\ncause, as seen in Figure 9. Only a tenth of the freight forwarders indicate the performance of the\n\n\ncustoms agency as the first main cause of delays and close to another tenth indicate the\n\nperformance of other agencies operating at the border being the first main cause of delays. <sup>14</sup>\n\n\nThese proportions increase slightly when we combine the responses of freight forwarders to what\n\n\nwere the first main cause, the second main cause, and the third main cause of clearance delays, as\n\n\nseen in the right set of bars in Figure 9. These patterns are quite similar across freight forwarders\n\n\nof any size.\n\n\n**Figure 9. One of Three Main Causes of Unexpected Clearance Delays**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nSource: Authors’ calculations based on data from the Survey of Serbian Freight Forwarders.\nNote: the statistics shown are based on responses by all 153 freight forwarders.\n\n\nFurther evidence on inspections as a cause of clearance delays is shown in Figure 10 that\n\n\nshows 63% of freight forwarders indicating that intensive physical inspections frequently lead to\n\n\n14 Many freight forwarders indicate that they do not trade in goods that require them to interact with the non-customs agency.\n\n\n19\n\nMention: \"the Survey of Serbian Freight Forwarders\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"All 153 forwarders' responses underlie the clearance delay statistics.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006369:5:0", "occurrence_id": "1511a8b5c870afd639ae719a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: However, during the final years of the Ali regime and the recent post-revolution period, the\nspending and composition of Tunisia’s subsidies have notably changed (Figure 1). During the\nlast 10 years, the combined spending on energy, food, and transportation has more than tripled,\nrising from 2% of GDP in 2005 to 7% in 2013. Energy subsidies, in particular, increased even\nmore, fourfold, during that period. Energy subsidies reached 4.7% of GDP in 2013, with\nsustained increases since 2010, reflecting the partial (rather than the full) pass-through of\ninternational oil prices to domestic prices sought by the government (IMF 2014). Regarding the\ncomposition of subsidies, during the post-revolution period, energy subsidies increased both in\nabsolute and relative terms. As a result, energy subsidies went from one-third of total public\nsubsidies prior to the revolution to about two-thirds in 2013. In contrast, food and other basic\nneeds’ subsidies have lost relative weight in the total subsidy bill despite having notably\nincreased in absolute terms. With respect to other public expenditures, energy subsidies in\nTunisia account for one-fifth of all public spending, or 7% of the GDP in 2013, the latest\navailable figure. Because they are considered the backbone of the social protection strategy, it is\nnot surprising that public spending on subsidies exceeds that of social assistance, health and\neducation, and individual programs on youth, children, or women (Figure 2).\n\n\n\n\n\n\n\n\n\n\n\n_Source:_ WB staff calculations using data from Ministère des Finances (October 2013).\n_Notes:_ Basic products refer to food products such as cereals, bread, sugar, and vegetable oil. * = forecast.\n\n\n4\n\nMention: \"Ministère des Finances (October 2013)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Basic product price data behind the incidence figures come from this ministry.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006370:15:0", "occurrence_id": "5f104a92c4588f567908c8a7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Applying 2015 tariffs has little negative impact on the per capita wellbeing of households. Given small\nincreases in tariffs that are focused mostly on rich consumers, expenditures per capita are expected to\ndecline on average by JD0.6 or about 0.03 percent (table 13). This would bring no changes in poverty and\npoverty gap measures.\n\n\n**<u>Table 13. The impact of 2015 tariffs on economic wellbeing</u>**\n<u>Quintiles</u> <u>Impact, JD</u> <u>Impact, %</u>\nQuintile 1 -0.02 0.00\nQuintile 2 -0.05 0.00\nQuintile 3 -0.15 -0.01\nQuintile 4 -0.31 -0.01\nQuintile 5 -2.48 -0.06\n<u>Total</u> <u>-0.60</u> <u>-0.03</u>\nSource: Authors’ calculations based on extrapolated HEIS 2010 data.\n<u>Notes: Population quintiles based on consumption per capita before the reform.</u>\n\nFull removal of subsidies, in contrast, will have a considerable impact on economic wellbeing. Fully\nremoving subsidies with a flat tariff rate is expected to reduce consumption per capita on average by JD\n72.5 or by 3.6 percent. The negative impact is expected to be the strongest for the poorest households, with\nthe bottom quintile experiencing on average a 5.7 percent reduction in per capita consumption. The negative\nburden on the poorest households can be reduced if a progressive increase of tariffs is applied. In this case,\nthe negative impact would be less pronounced for the poor even though the average household consumption\nwould drop by the same amount. Nevertheless, both sub-scenarios are quite severe and would be difficult\nto implement. A semi-progressive increase in tariffs leading to\n\nMention: \"HEIS 2010 data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Extrapolated consumption quintiles before the reform rest on this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006373:5:0", "occurrence_id": "a994a4bec76b04ac1b0042df", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: period of 14 years). To the best of our knowledge, there is no rigorous analysis of the evolution of\n\n\nintergenerational educational and occupational mobility in rural China in the post reform period\n\n\nin the existing economics literature. <sup>7</sup> [^7: The only paper we are aware of is Sato and Li (2007) which educational mobility among three\ngenerations of men in rural China using 2002 CHIP data set. In contrast, we focus on the _changes_, if any, in\nthe _average_ economic mobility in rural China from 1988 to 2002. Among the recent papers, Gong et al.\n(2012) and Deng et al. (2013) analyze income mobility in _urban_ China; Yuan and Chen (2013) focus on\nderiving intergenerational income elasticity, while we focus on causal effect.] We use two rounds of Chinese Household Income Project\n\n\n(CHIP) survey data for the years 1988 and 2002 for the analysis of intergenerational persistence\n\n\nin educational attainment and occupational choices over a span of almost a decade and a half.\n\n\nThe data in both 1988 and 2002 are based on almost identical questionnaires (2002 round has\n\n\nsome added information), and are comparable. This allows us to trace out the _changes_ in\n\n\nintergenerational occupational and educational persistence from 1988 to 2002.\n\n\nThe economics literature on the intergenerational economic mobility in developed countries\n\n\nfocuses on income mobility, especially between father and son. <sup>8</sup> [^8: The handful of studies focusing on the intergenerational correlations for daughters include Chadwick\nand Solon (2002) for USA and Dearden, Machin and Reed (1997) for UK.] There is also a relatively small\n\n\nliterature on occupational mobility in labor economics, again mostly in the context of developed\n\n\ncountries. <sup>9</sup> [^9: The intergenerational occupational mobility has, however, been the focus of a large, mostly descriptive,\nliterature in sociology.] Similar to income mobility, the existing literature on occupational mobility focuses\n\n\nprimarily on\n\nMention: \"2002 CHIP data set\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"An earlier study used it for educational mobility across three generations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006373:5:0", "occurrence_id": "90cfaea3c4ae8b3f531ee2d8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: period of 14 years). To the best of our knowledge, there is no rigorous analysis of the evolution of\n\n\nintergenerational educational and occupational mobility in rural China in the post reform period\n\n\nin the existing economics literature. <sup>7</sup> [^7: The only paper we are aware of is Sato and Li (2007) which educational mobility among three\ngenerations of men in rural China using 2002 CHIP data set. In contrast, we focus on the _changes_, if any, in\nthe _average_ economic mobility in rural China from 1988 to 2002. Among the recent papers, Gong et al.\n(2012) and Deng et al. (2013) analyze income mobility in _urban_ China; Yuan and Chen (2013) focus on\nderiving intergenerational income elasticity, while we focus on causal effect.] We use two rounds of Chinese Household Income Project\n\n\n(CHIP) survey data for the years 1988 and 2002 for the analysis of intergenerational persistence\n\n\nin educational attainment and occupational choices over a span of almost a decade and a half.\n\n\nThe data in both 1988 and 2002 are based on almost identical questionnaires (2002 round has\n\n\nsome added information), and are comparable. This allows us to trace out the _changes_ in\n\n\nintergenerational occupational and educational persistence from 1988 to 2002.\n\n\nThe economics literature on the intergenerational economic mobility in developed countries\n\n\nfocuses on income mobility, especially between father and son. <sup>8</sup> [^8: The handful of studies focusing on the intergenerational correlations for daughters include Chadwick\nand Solon (2002) for USA and Dearden, Machin and Reed (1997) for UK.] There is also a relatively small\n\n\nliterature on occupational mobility in labor economics, again mostly in the context of developed\n\n\ncountries. <sup>9</sup> [^9: The intergenerational occupational mobility has, however, been the focus of a large, mostly descriptive,\nliterature in sociology.] Similar to income mobility, the existing literature on occupational mobility focuses\n\n\nprimarily on\n\nMention: \"Chinese Household Income Project\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Two rounds carry the intergenerational persistence analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006353:7:0", "occurrence_id": "5a37135e014740c485bf464b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Living Standards Survey (VHLSSs), while data from India are synthetic panels that are constructed\n\n\nfrom cross sections of the National Sample Surveys (NSSs). To maximize comparability across\n\n\ncountries, we focus on the same time period for all three countries: 2004-2008 for the US and\n\n\nVietnam; and 2004/05- 2009/10 for India. <sup>6</sup> [^6: We use the 2007 and 2009 rounds of the PSID, whose income data are for the previous tax year. Thus we refer to\nthese two rounds by the tax year for convenience.] Detailed calculations and estimation results are drawn\n\n\nfrom two companions to the present paper (Dang and Lanjouw, 2014, 2015).\n\n\nThis note consists of four sections. We briefly review the growth experience for India, the US,\n\n\nand Vietnam in the next section, before delving into the analysis of shared prosperity in Section\n\n\nIII. Section IV concludes with further thoughts on research directions.\n\n\n**II. Overview on Growth Experience of Three Countries**\n\nTable 1 provides two measures of economic growth, one is growth in GDP per capita, and the\n\n\nother is growth in household consumption per capita (total household income for the US). The\n\n\nlatter measure is shown both for the whole population and the bottom 40 percent. While these\n\n\nmeasures can be qualitatively similar, they can offer different estimates of the speed of growth\n\n\n(e.g., Deaton, 2005). Thus combining the two can provide a more comprehensive picture. Indeed,\n\n\nTable 1 shows that growth rates are different, depending on whether the first or the second measure\n\n\nis employed.\n\n\nA couple of observations are in order for this table. First, India had the largest annual growth\n\n\nrate for GDP per capita, which is then followed by Vietnam and the US for the considered periods.\n\n\nThis order is, however, different when growth is defined in terms of household consumption. By\n\n\nthis measure, overall growth rates in India and Vietnam were the same,\n\nMention: \"Living Standards Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its repeated rounds form one country's panel.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006353:7:0", "occurrence_id": "555f69ed676b27caf2ac8a75", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Living Standards Survey (VHLSSs), while data from India are synthetic panels that are constructed\n\n\nfrom cross sections of the National Sample Surveys (NSSs). To maximize comparability across\n\n\ncountries, we focus on the same time period for all three countries: 2004-2008 for the US and\n\n\nVietnam; and 2004/05- 2009/10 for India. <sup>6</sup> [^6: We use the 2007 and 2009 rounds of the PSID, whose income data are for the previous tax year. Thus we refer to\nthese two rounds by the tax year for convenience.] Detailed calculations and estimation results are drawn\n\n\nfrom two companions to the present paper (Dang and Lanjouw, 2014, 2015).\n\n\nThis note consists of four sections. We briefly review the growth experience for India, the US,\n\n\nand Vietnam in the next section, before delving into the analysis of shared prosperity in Section\n\n\nIII. Section IV concludes with further thoughts on research directions.\n\n\n**II. Overview on Growth Experience of Three Countries**\n\nTable 1 provides two measures of economic growth, one is growth in GDP per capita, and the\n\n\nother is growth in household consumption per capita (total household income for the US). The\n\n\nlatter measure is shown both for the whole population and the bottom 40 percent. While these\n\n\nmeasures can be qualitatively similar, they can offer different estimates of the speed of growth\n\n\n(e.g., Deaton, 2005). Thus combining the two can provide a more comprehensive picture. Indeed,\n\n\nTable 1 shows that growth rates are different, depending on whether the first or the second measure\n\n\nis employed.\n\n\nA couple of observations are in order for this table. First, India had the largest annual growth\n\n\nrate for GDP per capita, which is then followed by Vietnam and the US for the considered periods.\n\n\nThis order is, however, different when growth is defined in terms of household consumption. By\n\n\nthis measure, overall growth rates in India and Vietnam were the same,\n\nMention: \"National Sample Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its cross sections are combined into a synthetic panel.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006377:30:0", "occurrence_id": "a7abc9cab81fb44e65fb81b1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 3.** **Average Index of Supervisory Quality by Crisis Experience.**\n\n\n\n\n\n\n\n\n\n_Source_ : Authors’ calculations based on Anginer, Demirgüç-Kunt, and Zhu 2013; <mark>Laeven and</mark>\n<mark>Valencia 2013.</mark> We followed Anginer, et al.’s (2013) methodology and computed the index by aggregating\nthe answers to fourteen selected questions regarding supervisory powers that were collected in the 2003, 2007\nand 2011 surveys conducted by Barth, Caprio, and Levine (2008).\n_Note:_ “Crisis countries” are those that experienced a financial crisis in 2008; “no-crisis countries” are those\nthat did not. The sample contains 25 crisis countries and 122 no-crisis countries. The index is the\naggregated answers to 14 selected questions on supervisory powers collected in 2003, 2007, and\n2011 surveys (Barth, Caprio, and Levine, 2008).\n\n\n**Figure 4.** **Average Number of Years before 2008 of Publishing a Financial Stability Report in Crisis and No-**\n**Crisis (left panel) Countries; and the Share of Crisis and No-Crisis Countries that Published a Financial**\n**Stability Report in 2007(right panel).**\n\n\n\n\n\n\n\n\n\n\n\n_Source:_ Authors’ calculations based on Čihák et al. (2012) and <mark>Laeven and Valencia (2013).</mark>\n_Note:_ “Crisis countries” are those that experienced a financial crisis in 2008; “no-crisis countries” are those that did not. The\nsample contains 25 crisis countries and 122 no-crisis countries.\n\n\n29\n\nMention: \"Barth, Caprio, and Levine (2008)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Their surveys across these years collect supervisory power measures.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006378:45:0", "occurrence_id": "087149bcce68b563f295c1d7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: two-digit industry, which is reported by _China Statistics Yearbook_ . <sup>20</sup> [^20: Since China Statistics Yearbook report the growth rate of nominal capital stock in the two‐digit industry from 1986, we\nassume firms established before 1986 are established in 1986.] Then, we can calculate the\n\n\nnominal capital stock in 1998 from the following equation:\n\n������ ���� ∏������� �1 ���� (A2)\n\nWhere ������ is the nominal captial stock in 1998 reported in the ASIF data, _s_ indicate the\n\nfirm’s estalished year, while ��� is the nominal captial stock in the firm’s. And �� is the growth\n\nrate of nominal capital stock in the two-digit industry in year t, which is reported by _China_\n\n\n_Statistics Yearbook_ . From equation (A2), we can calculate the nominal stock in each year between\n\n\nfirm’s established year and 1998.\n\nSecond, annual nominal investment ��� is the change in nominal capital stock between\n\nyears, that is, ��� ���� ������.\n\nThird, we derive the real capital stock for each year between firm’s established year and 1998.\n\nWe deflate annual nominal investment in each year ��� into real values ��� with the investment\n\ndeflator. _China Statistics Yearbook_ reports the investment deflator from 1990. For years 1986-1989,\n\n\nwe use the investment deflator constructed by Perkins and Rawski (2008).\n\nFourth, we obtain the real capital stock in 1998 from the perpetual inventory method.\n\n\nSpecifically,\n\n��� ��1 �������� ����\n\nWhere ��� is the real capital stock in year _t_, and - is the depreciation rate, which is estimated\n\n\n\nby\n\n\n\n<u>����������� �����������\n\nMention: \"ASIF data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Nominal capital stock reported here anchors the depreciation calculation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006387:23:2", "occurrence_id": "d08a8608b33f2718a2c60c73", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .21 0.30 0.24 0.19 0.26\nMozambique 2011 0.36 0.36 0.35 0.31 0.33 0.30\nRwanda 2010 0.33 0.31 0.34 0.28 0.27 0.28\nTanzania 2010 0.24 0.23 0.25 0.20 0.21 0.20\nUganda 2011 0.30 0.31 0.29 0.26 0.30 0.26\nZambia 2007 0.24 0.22 0.25 0.20 0.20 0.20\nSub-Total 0.28 0.30 0.28 0.23 0.27 0.23\n\n**SOUTHERN**\nLesotho 2009 0.42 0.42 0.42 0.34 0.35 0.33\nNamibia 2006 0.44 0.40 0.48 0.45 0.40 0.49\nSouth Africa 1998 0.42 0.36 0.50 0.43 0.36 0.50\nSwaziland 2006 0.48 0.39 0.53 0.51 0.42 0.54\nZimbabwe 2010 0.45 0.45 0.44 0.43 0.46 0.42\nSub-Total 0.43 0.38 0.48 0.43 0.38 0.48\n\n**<u>AFRICA</u>** **<u>0.26</u>** **<u>0.28</u>** **<u>0.25</u>** **<u>0.21</u>** **<u>0.24</u>** **<u>0.20</u>**\nNote: Based on information from the household roster in the latest available DHSs.\n\n\n22\n\nMention: \"DHSs\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The latest available rounds supply the household roster information.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006394:42:0", "occurrence_id": "775e3babbb23c693c1061cac", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: #### **Appendix A : Data Description and Sources**\n\n**Domestic Prices**\nWholesale white maize prices in Tanzania are monthly averages, expressed in TZ shillings\nper kg, collected by Tanzania’s Ministry of Industry and Trade in 18 markets. These markets are grouped into five zones and cover the entire country. Data on domestic fuel prices\nand CPI are collected by Tanzania’s National Bureau of Statistics (NBS).\n\n\n**External Prices**\nWholesale white maize prices in Nairobi (Kenya), Randfontein (South Africa), and Nam[pula (Mozambique) are sourced from FAO GIEWS (http://www.fao.org/giews/english/](http://www.fao.org/giews/english/index.htm)\n[index.htm); U.S. Maize prices represent no.](http://www.fao.org/giews/english/index.htm) 2, yellow, f.o.b. Gulf ports. Crude oil price\nis the average of Brent (U.K. 38 <sup>0</sup> API), West Texas Intermediate (WTI 40 <sup>0</sup> API), and Dubai\n(Fateh 32 <sup>0</sup> API) equally weighted. World prices for maize and oil were taken from the\n[World Bank’s pink sheet (http://go.worldbank.org/4ROCCIEQ50).](http://go.worldbank.org/4ROCCIEQ50)\n\n\n**Export Ban**\nAn export ban is represented by a dummy variable taking the value of 1 when the ban was\neffective and zero otherwise. The ban was effective during the following months: Jan—\nDec 2005, Mar 2006\n\nMention: \"Data on domestic fuel prices\nand CPI\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Collected by the national statistics bureau across 18 markets.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006394:42:0", "occurrence_id": "f91f66cb3f175256407ce875", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: #### **Appendix A : Data Description and Sources**\n\n**Domestic Prices**\nWholesale white maize prices in Tanzania are monthly averages, expressed in TZ shillings\nper kg, collected by Tanzania’s Ministry of Industry and Trade in 18 markets. These markets are grouped into five zones and cover the entire country. Data on domestic fuel prices\nand CPI are collected by Tanzania’s National Bureau of Statistics (NBS).\n\n\n**External Prices**\nWholesale white maize prices in Nairobi (Kenya), Randfontein (South Africa), and Nam[pula (Mozambique) are sourced from FAO GIEWS (http://www.fao.org/giews/english/](http://www.fao.org/giews/english/index.htm)\n[index.htm); U.S. Maize prices represent no.](http://www.fao.org/giews/english/index.htm) 2, yellow, f.o.b. Gulf ports. Crude oil price\nis the average of Brent (U.K. 38 <sup>0</sup> API), West Texas Intermediate (WTI 40 <sup>0</sup> API), and Dubai\n(Fateh 32 <sup>0</sup> API) equally weighted. World prices for maize and oil were taken from the\n[World Bank’s pink sheet (http://go.worldbank.org/4ROCCIEQ50).](http://go.worldbank.org/4ROCCIEQ50)\n\n\n**Export Ban**\nAn export ban is represented by a dummy variable taking the value of 1 when the ban was\neffective and zero otherwise. The ban was effective during the following months: Jan—\nDec 2005, Mar 2006\n\nMention: \"FAO GIEWS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Regional market prices at cross-border points come from this system.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006394:42:0", "occurrence_id": "894f3f1eebf3609e120b4986", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: #### **Appendix A : Data Description and Sources**\n\n**Domestic Prices**\nWholesale white maize prices in Tanzania are monthly averages, expressed in TZ shillings\nper kg, collected by Tanzania’s Ministry of Industry and Trade in 18 markets. These markets are grouped into five zones and cover the entire country. Data on domestic fuel prices\nand CPI are collected by Tanzania’s National Bureau of Statistics (NBS).\n\n\n**External Prices**\nWholesale white maize prices in Nairobi (Kenya), Randfontein (South Africa), and Nam[pula (Mozambique) are sourced from FAO GIEWS (http://www.fao.org/giews/english/](http://www.fao.org/giews/english/index.htm)\n[index.htm); U.S. Maize prices represent no.](http://www.fao.org/giews/english/index.htm) 2, yellow, f.o.b. Gulf ports. Crude oil price\nis the average of Brent (U.K. 38 <sup>0</sup> API), West Texas Intermediate (WTI 40 <sup>0</sup> API), and Dubai\n(Fateh 32 <sup>0</sup> API) equally weighted. World prices for maize and oil were taken from the\n[World Bank’s pink sheet (http://go.worldbank.org/4ROCCIEQ50).](http://go.worldbank.org/4ROCCIEQ50)\n\n\n**Export Ban**\nAn export ban is represented by a dummy variable taking the value of 1 when the ban was\neffective and zero otherwise. The ban was effective during the following months: Jan—\nDec 2005, Mar 2006\n\nMention: \"World Bank’s pink sheet\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"World prices for maize and oil are taken from this price series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006357:9:0", "occurrence_id": "839a56fdaf302f16d239ec90", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Rainfall:_\n\n\nOur rainfall data come from the GPCP <sup>3</sup> [^3: GPCP rainfall data is publicly available from http://precip.gsfc.nasa.gov/] database, which provides rainfall estimates from 1979 to present\n\n\nday, published by the precipitation research group at NASA. This rainfall measure has been used in the\n\n\nliterature several times, notably in Miguel (2004) and Bruckner and Ciccone (2011). GPCP provides\n\n\nmonthly averages of rainfall in 2.5 degree latitude and longitude intervals. These monthly averages are\n\n\nthen multiplied by the number of days in each month to produce an estimate of monthly rainfall. These\n\n\nmonthly rainfall estimates are then summed to get an annual rainfall estimate. Since 2.5 degree latitude\n\n\nand longitude intervals do not represent countries accurately, we follow the approach of Miguel (2004) to\n\n\naggregate rainfall in these intervals to country levels in SSA. The GPCP data set utilizes weather station\n\n\nmeasures in conjunction with satellite observations of cold cloud cover.\n\n\n_Commodity index:_\n\n\nWe assemble a country-specific commodity export price index following the model:\n\n\n𝜃𝑖,𝑐\n\n𝐶𝑜𝑚𝑃𝐼𝑖,𝑡 = ∏𝐶𝑜𝑚𝑃𝑟𝑖𝑐𝑒 (𝑡)𝑐,𝑡\n\n𝑐∈𝐶\n\n\nwhere _ComPricec,t_ is the international price of commodity c in year t, and _θi,c_ is the average (time\n\n\ninvariant) value of exports of commodity _c_ in the GDP of country _i._ Our commodity data comes from\n\n\ntwo sources. The first of these is the NBER-UN trade data, which we use for export values. Real\n\n\ncommodity price index come from the UNCTAD commodity statistics. We included aluminum, beef,\n\n\ncoffee, cocoa, copper, cotton, gold, iron, maize, oil, rice, rubber, sugar, tea, tobacco, wheat, and wood in\n\n\nour index, following Arezki and Bruckner (2014).\n\n\n**III.2 Capital Flows to SSA**\n\n\nFigures 1 and 2 present an overview about total capital inflows\n\nMention: \"GPCP\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Rainfall data come from this publicly available database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006357:9:0", "occurrence_id": "690b4fbf9c2dc4e95523d2ef", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Rainfall:_\n\n\nOur rainfall data come from the GPCP <sup>3</sup> [^3: GPCP rainfall data is publicly available from http://precip.gsfc.nasa.gov/] database, which provides rainfall estimates from 1979 to present\n\n\nday, published by the precipitation research group at NASA. This rainfall measure has been used in the\n\n\nliterature several times, notably in Miguel (2004) and Bruckner and Ciccone (2011). GPCP provides\n\n\nmonthly averages of rainfall in 2.5 degree latitude and longitude intervals. These monthly averages are\n\n\nthen multiplied by the number of days in each month to produce an estimate of monthly rainfall. These\n\n\nmonthly rainfall estimates are then summed to get an annual rainfall estimate. Since 2.5 degree latitude\n\n\nand longitude intervals do not represent countries accurately, we follow the approach of Miguel (2004) to\n\n\naggregate rainfall in these intervals to country levels in SSA. The GPCP data set utilizes weather station\n\n\nmeasures in conjunction with satellite observations of cold cloud cover.\n\n\n_Commodity index:_\n\n\nWe assemble a country-specific commodity export price index following the model:\n\n\n𝜃𝑖,𝑐\n\n𝐶𝑜𝑚𝑃𝐼𝑖,𝑡 = ∏𝐶𝑜𝑚𝑃𝑟𝑖𝑐𝑒 (𝑡)𝑐,𝑡\n\n𝑐∈𝐶\n\n\nwhere _ComPricec,t_ is the international price of commodity c in year t, and _θi,c_ is the average (time\n\n\ninvariant) value of exports of commodity _c_ in the GDP of country _i._ Our commodity data comes from\n\n\ntwo sources. The first of these is the NBER-UN trade data, which we use for export values. Real\n\n\ncommodity price index come from the UNCTAD commodity statistics. We included aluminum, beef,\n\n\ncoffee, cocoa, copper, cotton, gold, iron, maize, oil, rice, rubber, sugar, tea, tobacco, wheat, and wood in\n\n\nour index, following Arezki and Bruckner (2014).\n\n\n**III.2 Capital Flows to SSA**\n\n\nFigures 1 and 2 present an overview about total capital inflows\n\nMention: \"GPCP rainfall data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Publicly available at the stated address.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006357:9:0", "occurrence_id": "911586c167cee147ade501b6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Rainfall:_\n\n\nOur rainfall data come from the GPCP <sup>3</sup> [^3: GPCP rainfall data is publicly available from http://precip.gsfc.nasa.gov/] database, which provides rainfall estimates from 1979 to present\n\n\nday, published by the precipitation research group at NASA. This rainfall measure has been used in the\n\n\nliterature several times, notably in Miguel (2004) and Bruckner and Ciccone (2011). GPCP provides\n\n\nmonthly averages of rainfall in 2.5 degree latitude and longitude intervals. These monthly averages are\n\n\nthen multiplied by the number of days in each month to produce an estimate of monthly rainfall. These\n\n\nmonthly rainfall estimates are then summed to get an annual rainfall estimate. Since 2.5 degree latitude\n\n\nand longitude intervals do not represent countries accurately, we follow the approach of Miguel (2004) to\n\n\naggregate rainfall in these intervals to country levels in SSA. The GPCP data set utilizes weather station\n\n\nmeasures in conjunction with satellite observations of cold cloud cover.\n\n\n_Commodity index:_\n\n\nWe assemble a country-specific commodity export price index following the model:\n\n\n𝜃𝑖,𝑐\n\n𝐶𝑜𝑚𝑃𝐼𝑖,𝑡 = ∏𝐶𝑜𝑚𝑃𝑟𝑖𝑐𝑒 (𝑡)𝑐,𝑡\n\n𝑐∈𝐶\n\n\nwhere _ComPricec,t_ is the international price of commodity c in year t, and _θi,c_ is the average (time\n\n\ninvariant) value of exports of commodity _c_ in the GDP of country _i._ Our commodity data comes from\n\n\ntwo sources. The first of these is the NBER-UN trade data, which we use for export values. Real\n\n\ncommodity price index come from the UNCTAD commodity statistics. We included aluminum, beef,\n\n\ncoffee, cocoa, copper, cotton, gold, iron, maize, oil, rice, rubber, sugar, tea, tobacco, wheat, and wood in\n\n\nour index, following Arezki and Bruckner (2014).\n\n\n**III.2 Capital Flows to SSA**\n\n\nFigures 1 and 2 present an overview about total capital inflows\n\nMention: \"GPCP data set\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Station measures combined with satellite observations of cloud cover.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006357:9:0", "occurrence_id": "a6f780ecef59ffdc22de6967", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Rainfall:_\n\n\nOur rainfall data come from the GPCP <sup>3</sup> [^3: GPCP rainfall data is publicly available from http://precip.gsfc.nasa.gov/] database, which provides rainfall estimates from 1979 to present\n\n\nday, published by the precipitation research group at NASA. This rainfall measure has been used in the\n\n\nliterature several times, notably in Miguel (2004) and Bruckner and Ciccone (2011). GPCP provides\n\n\nmonthly averages of rainfall in 2.5 degree latitude and longitude intervals. These monthly averages are\n\n\nthen multiplied by the number of days in each month to produce an estimate of monthly rainfall. These\n\n\nmonthly rainfall estimates are then summed to get an annual rainfall estimate. Since 2.5 degree latitude\n\n\nand longitude intervals do not represent countries accurately, we follow the approach of Miguel (2004) to\n\n\naggregate rainfall in these intervals to country levels in SSA. The GPCP data set utilizes weather station\n\n\nmeasures in conjunction with satellite observations of cold cloud cover.\n\n\n_Commodity index:_\n\n\nWe assemble a country-specific commodity export price index following the model:\n\n\n𝜃𝑖,𝑐\n\n𝐶𝑜𝑚𝑃𝐼𝑖,𝑡 = ∏𝐶𝑜𝑚𝑃𝑟𝑖𝑐𝑒 (𝑡)𝑐,𝑡\n\n𝑐∈𝐶\n\n\nwhere _ComPricec,t_ is the international price of commodity c in year t, and _θi,c_ is the average (time\n\n\ninvariant) value of exports of commodity _c_ in the GDP of country _i._ Our commodity data comes from\n\n\ntwo sources. The first of these is the NBER-UN trade data, which we use for export values. Real\n\n\ncommodity price index come from the UNCTAD commodity statistics. We included aluminum, beef,\n\n\ncoffee, cocoa, copper, cotton, gold, iron, maize, oil, rice, rubber, sugar, tea, tobacco, wheat, and wood in\n\n\nour index, following Arezki and Bruckner (2014).\n\n\n**III.2 Capital Flows to SSA**\n\n\nFigures 1 and 2 present an overview about total capital inflows\n\nMention: \"NBER-UN trade data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Export values across commodities come from this dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006357:9:0", "occurrence_id": "cb41d4c04d5c1d17a7fd0736", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Rainfall:_\n\n\nOur rainfall data come from the GPCP <sup>3</sup> [^3: GPCP rainfall data is publicly available from http://precip.gsfc.nasa.gov/] database, which provides rainfall estimates from 1979 to present\n\n\nday, published by the precipitation research group at NASA. This rainfall measure has been used in the\n\n\nliterature several times, notably in Miguel (2004) and Bruckner and Ciccone (2011). GPCP provides\n\n\nmonthly averages of rainfall in 2.5 degree latitude and longitude intervals. These monthly averages are\n\n\nthen multiplied by the number of days in each month to produce an estimate of monthly rainfall. These\n\n\nmonthly rainfall estimates are then summed to get an annual rainfall estimate. Since 2.5 degree latitude\n\n\nand longitude intervals do not represent countries accurately, we follow the approach of Miguel (2004) to\n\n\naggregate rainfall in these intervals to country levels in SSA. The GPCP data set utilizes weather station\n\n\nmeasures in conjunction with satellite observations of cold cloud cover.\n\n\n_Commodity index:_\n\n\nWe assemble a country-specific commodity export price index following the model:\n\n\n𝜃𝑖,𝑐\n\n𝐶𝑜𝑚𝑃𝐼𝑖,𝑡 = ∏𝐶𝑜𝑚𝑃𝑟𝑖𝑐𝑒 (𝑡)𝑐,𝑡\n\n𝑐∈𝐶\n\n\nwhere _ComPricec,t_ is the international price of commodity c in year t, and _θi,c_ is the average (time\n\n\ninvariant) value of exports of commodity _c_ in the GDP of country _i._ Our commodity data comes from\n\n\ntwo sources. The first of these is the NBER-UN trade data, which we use for export values. Real\n\n\ncommodity price index come from the UNCTAD commodity statistics. We included aluminum, beef,\n\n\ncoffee, cocoa, copper, cotton, gold, iron, maize, oil, rice, rubber, sugar, tea, tobacco, wheat, and wood in\n\n\nour index, following Arezki and Bruckner (2014).\n\n\n**III.2 Capital Flows to SSA**\n\n\nFigures 1 and 2 present an overview about total capital inflows\n\nMention: \"UNCTAD commodity statistics\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Real price indices across the listed commodities come from these statistics.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006391:10:0", "occurrence_id": "4951f08362d77926ce4a8090", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: effectively provide public goods - for instance, when on group seeks to actively obstruct the\n\n\nauthority of the other - we also expect inter-elite competition to harm local development\n\n\noutcomes.\n\n\nThere are also theoretically based reasons to expect that relationships defined by\n\n\ncooperation, rather than competition, will be beneficial to the provision of local public goods.\n\n\nFor instance, as Fearon et al. (2009) suggest, chieftaincy support of local council projects may\n\n\nincrease the chances of programmatic success. Further, Logan’s (2008, iii) analysis of Afro\n\nbarometer data from 15 countries on the continent, finds that respondents’ evaluations of newly\n\n\ndecentralized local authorities and hereditary chiefs move together, noting “local traditional\n\n\nleaders appear to draw their sustenance and legitimacy from the same well as elected officials.”\n\n\nThis suggests that rather than control over local power and resources being a zero-sum game,\n\n\ncooperative relationships between chiefs and councilors may be beneficial for public goods\n\n\noutcomes as well as public perceptions of the quality of local government. If these types of\n\n\ncooperative inter-elite relationships are at play, closer ties between councilors and chiefs should\n\n\nbe associated with greater levels of public goods.\n\n\nYet other research on the political economy of Africa suggests an opposite effect. Close\n\n\ninter-elite relationships may indicate that councilors are benefitting from chiefs’ clientelistic\n\n\nnetworks to buy votes and ensure future electoral support. In such instances, patronage-based\n\n\nredistribution may come at the expense of public goods provision (Bratton 2008; Collier &\n\n\nVicente 2012; Keefer 2005; van de Walle 2003, 2012; Wantchekon 2003). In addition, if\n\n\ncouncilors draw their legitimacy indirectly from the support of chiefs rather than from their\n\n\nperformance, the accountability of local governance and thus councilors’ incentives to invest in\n\n\npublic goods also decline. In such scenarios, we expect that closer inter-elite relationships\n\n\ndefined by frequent contact in the absence of disputes and a higher frequency of councilor/chief\n\n\nfamilial ties will lead to worse\n\nMention: \"Afro\n\nbarometer data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its 15-country analysis records evaluations of new institutions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006393:12:1", "occurrence_id": "012c677d1bfaf18c8dfa3d40", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  household. <sup>22</sup> For each enterprise we compute net income as total annual sales minus\n\ncosts. In the Malawi ISA, households report the average net income for a bad, standard, and\n\ngood month. Households are then asked how many of each type of month occurred in the past\n\nyear. In Uganda, households report both gross income and costs. In Tanzania households report\nnet income directly. <sup>23</sup>\n\n\n19In Malawi, the majority of households receive seed and fertilizer subsidies by the Farm Inputs Subsidy Program.\nWe use the subsidized (coupon) prices reported by each household.\n\n20This is innocuous for our purposes of measuring household income because the labor income generated by\nhousehold members in agricultural production is also part of household income. Under some assumptions, this\nlabor income can be separated from net profits, but this is beyond the scope of our paper.\n\n21Previous LSMS datasets provide information on labor supply with the reference period of “the past 7 days”.\nISAs complement that information with recalled hours worked per day, week, and month over the past 12 months,\nwhich greatly facilitates the determination of the annualized labor supply. In particular, this avoids potential\nmeasurement error from labor supply seasonality. See Rosenzweig and Udry (2014) for a discussion on how wages\nare affected by seasonal weather patterns.\n\n22Privately held businesses per household (potentially more than one) include owned nonagricultural businesses\nthat process/sell agricultural byproducts (e.g. flour, juice, beer, jam, oil, seed, and livestock by-products),\nsales of forest-based products, street or market trading businesses, taxi or pickup truck drivers, bar/restaurants,\nprofessional services (e.g., doctor, accountant, lawyer, and midwife) etc..\n\n23As was the case for agricultural net income, business net income includes the contribution of household labor\n\nMention: \"LSMS datasets\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Earlier rounds used a seven-day labour reference period.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006393:12:1", "occurrence_id": "daa30c1c4837bddc937868b6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  household. <sup>22</sup> For each enterprise we compute net income as total annual sales minus\n\ncosts. In the Malawi ISA, households report the average net income for a bad, standard, and\n\ngood month. Households are then asked how many of each type of month occurred in the past\n\nyear. In Uganda, households report both gross income and costs. In Tanzania households report\nnet income directly. <sup>23</sup>\n\n\n19In Malawi, the majority of households receive seed and fertilizer subsidies by the Farm Inputs Subsidy Program.\nWe use the subsidized (coupon) prices reported by each household.\n\n20This is innocuous for our purposes of measuring household income because the labor income generated by\nhousehold members in agricultural production is also part of household income. Under some assumptions, this\nlabor income can be separated from net profits, but this is beyond the scope of our paper.\n\n21Previous LSMS datasets provide information on labor supply with the reference period of “the past 7 days”.\nISAs complement that information with recalled hours worked per day, week, and month over the past 12 months,\nwhich greatly facilitates the determination of the annualized labor supply. In particular, this avoids potential\nmeasurement error from labor supply seasonality. See Rosenzweig and Udry (2014) for a discussion on how wages\nare affected by seasonal weather patterns.\n\n22Privately held businesses per household (potentially more than one) include owned nonagricultural businesses\nthat process/sell agricultural byproducts (e.g. flour, juice, beer, jam, oil, seed, and livestock by-products),\nsales of forest-based products, street or market trading businesses, taxi or pickup truck drivers, bar/restaurants,\nprofessional services (e.g., doctor, accountant, lawyer, and midwife) etc..\n\n23As was the case for agricultural net income, business net income includes the contribution of household labor\n\nMention: \"ISAs\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"They add recalled hours per day, week and month.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006395:9:0", "occurrence_id": "a461299e9e9bef8ce79360ca", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **2.2.** **The Participants**\n\n\nThe recruitment of a total number of 700 participants took place between March 2010 and August\n\n\n2011. Five individuals had to be excluded from the trial because of a missing consent form,\n\n\nbecoming pregnant, developing cancer, and internal documentation problems. <sup>9</sup> [^9: Results are robust with respect to treating these individuals as dropouts in sensitivity checks described\nin Section 3.] The last\n\n\nparticipant finished the follow‐up phase by the end of July 2013.\n\n\nTable 1 provides some descriptive statistics of the study population. The average body\n\n\nweight at the start of the experiment (at the end of the rehabilitation stay) was 113.0 kg (249.4\n\n\nlbs.) or 37.6 in terms of BMI (Figure A1 shows the distribution of the BMI over time). About 68\n\n\npercent of the participants are men and 21 percent have a migration background. <sup>10</sup> [^10: These shares are substantially lower than the corresponding averages for obese individuals in Germany,\nwhich we obtained using the German Socioeconomic Panel (SOEP) – a representative panel of German\nhouseholds. The mean age of the study population (48 years) lies about ten years below the average age of\nobese in Germany, while the share of employed participants (82 percent) is almost twice as large. Only the\nshare of married participants does not deviate considerably from the respective share of obese in Germany.] For most\n\n\npatients of the cooperating clinics, medical rehabilitation is paid by the German pension fund,\n\n\nwhose predominant goal is to avoid work disability and early retirement. Since there are many\n\n\nobese in the overall population that are already retired, our study population oversamples\n\n\npersons that are available for the labor market.\n\n\nAs already noted, four rehabilitation clinics, located in different towns, have been involved\n\n\nin the trial. About 42 percent of the participants were recruited by the clinic in Bad Mergentheim,\n\nMention: \"the German Socioeconomic Panel (SOEP)\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"A representative household panel supplying the comparison averages.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006397:8:0", "occurrence_id": "783466942c4231fb10967efa", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: characteristics are obtained from the official website of Bangladesh’s national power utility,\nBangladesh Power Development Board (BPDB, 2013), and the 2010 Power Sector Master Plan\nprepared by the Japan International Cooperation Agency and Tokyo Electric Power Company for\nthe Ministry of Power, Energy and Mineral Resources (MPEMR, 2011). For Bhutan, physical\ncharacteristics of power plants are from the National Statistical Yearbook of Bhutan (NSB (2013)\nand Annual Reports of Druk Green Power Company Limited, Bhutan’s national power company\n(DGPCL, 2013).\n\nFor Pakistan, physical characteristics of power plants are from an existing Power Sector\nMaster Plan prepared by the SNC-Lavalin International Inc. in association with National\nEngineering Services Pakistan Limited for the National Transmission and Despatch Company\nLimited, Pakistan (NTDCL, 2011). The data are supplemented with information from an integrated\nenergy plan 2009-2022 prepared by an energy expert group for the Economic Advisory Council,\nMinistry of Finance (EAC, 2009). Information on candidate hydropower projects are from a survey\nreport prepared by Pakistan’s Water and Power Development Authority (PWAPDA, 2011). Data\nfor Sri Lanka are from the Central Electricity Boards’ Long Term Generation Expansion Planning\n2013-2032 prepared by for Government of Sri Lanka (CEB 2012). We tried to use the data from\nthe existing electricity master plan, where available, to be consistent with countries national\nelectricity expansion plan in the baseline. However, we did not force all aspects of the baseline in\nthe master plan to apply in our study, as the underlying assumptions in the master plan might be\ndifferent from ours.\n\nIn India, the list of all existing plants and their installed capacities, year of commissioning\nand fuel type were taken from various documents (CEA 2012a, 2012b, 2012c and 2012d)\npublished by the Central Electricity Authority (CEA), a statutory organization under the Central\nGovernment that advises on matters relating to\n\nMention: \"Bangladesh Power Development Board\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The national power utility's website gives plant characteristics.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006397:8:0", "occurrence_id": "2ffe744e92f5a13f63d4ebe7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: characteristics are obtained from the official website of Bangladesh’s national power utility,\nBangladesh Power Development Board (BPDB, 2013), and the 2010 Power Sector Master Plan\nprepared by the Japan International Cooperation Agency and Tokyo Electric Power Company for\nthe Ministry of Power, Energy and Mineral Resources (MPEMR, 2011). For Bhutan, physical\ncharacteristics of power plants are from the National Statistical Yearbook of Bhutan (NSB (2013)\nand Annual Reports of Druk Green Power Company Limited, Bhutan’s national power company\n(DGPCL, 2013).\n\nFor Pakistan, physical characteristics of power plants are from an existing Power Sector\nMaster Plan prepared by the SNC-Lavalin International Inc. in association with National\nEngineering Services Pakistan Limited for the National Transmission and Despatch Company\nLimited, Pakistan (NTDCL, 2011). The data are supplemented with information from an integrated\nenergy plan 2009-2022 prepared by an energy expert group for the Economic Advisory Council,\nMinistry of Finance (EAC, 2009). Information on candidate hydropower projects are from a survey\nreport prepared by Pakistan’s Water and Power Development Authority (PWAPDA, 2011). Data\nfor Sri Lanka are from the Central Electricity Boards’ Long Term Generation Expansion Planning\n2013-2032 prepared by for Government of Sri Lanka (CEB 2012). We tried to use the data from\nthe existing electricity master plan, where available, to be consistent with countries national\nelectricity expansion plan in the baseline. However, we did not force all aspects of the baseline in\nthe master plan to apply in our study, as the underlying assumptions in the master plan might be\ndifferent from ours.\n\nIn India, the list of all existing plants and their installed capacities, year of commissioning\nand fuel type were taken from various documents (CEA 2012a, 2012b, 2012c and 2012d)\npublished by the Central Electricity Authority (CEA), a statutory organization under the Central\nGovernment that advises on matters relating to\n\nMention: \"2010 Power Sector Master Plan\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Prepared with a power company, it supplies the sector outlook.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006397:8:0", "occurrence_id": "f4e8531c1ca2962621f2a821", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: characteristics are obtained from the official website of Bangladesh’s national power utility,\nBangladesh Power Development Board (BPDB, 2013), and the 2010 Power Sector Master Plan\nprepared by the Japan International Cooperation Agency and Tokyo Electric Power Company for\nthe Ministry of Power, Energy and Mineral Resources (MPEMR, 2011). For Bhutan, physical\ncharacteristics of power plants are from the National Statistical Yearbook of Bhutan (NSB (2013)\nand Annual Reports of Druk Green Power Company Limited, Bhutan’s national power company\n(DGPCL, 2013).\n\nFor Pakistan, physical characteristics of power plants are from an existing Power Sector\nMaster Plan prepared by the SNC-Lavalin International Inc. in association with National\nEngineering Services Pakistan Limited for the National Transmission and Despatch Company\nLimited, Pakistan (NTDCL, 2011). The data are supplemented with information from an integrated\nenergy plan 2009-2022 prepared by an energy expert group for the Economic Advisory Council,\nMinistry of Finance (EAC, 2009). Information on candidate hydropower projects are from a survey\nreport prepared by Pakistan’s Water and Power Development Authority (PWAPDA, 2011). Data\nfor Sri Lanka are from the Central Electricity Boards’ Long Term Generation Expansion Planning\n2013-2032 prepared by for Government of Sri Lanka (CEB 2012). We tried to use the data from\nthe existing electricity master plan, where available, to be consistent with countries national\nelectricity expansion plan in the baseline. However, we did not force all aspects of the baseline in\nthe master plan to apply in our study, as the underlying assumptions in the master plan might be\ndifferent from ours.\n\nIn India, the list of all existing plants and their installed capacities, year of commissioning\nand fuel type were taken from various documents (CEA 2012a, 2012b, 2012c and 2012d)\npublished by the Central Electricity Authority (CEA), a statutory organization under the Central\nGovernment that advises on matters relating to\n\nMention: \"National Statistical Yearbook of Bhutan\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Physical characteristics of power plants come from this yearbook.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006397:8:0", "occurrence_id": "1a0e032ee93fbaaedf245874", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: characteristics are obtained from the official website of Bangladesh’s national power utility,\nBangladesh Power Development Board (BPDB, 2013), and the 2010 Power Sector Master Plan\nprepared by the Japan International Cooperation Agency and Tokyo Electric Power Company for\nthe Ministry of Power, Energy and Mineral Resources (MPEMR, 2011). For Bhutan, physical\ncharacteristics of power plants are from the National Statistical Yearbook of Bhutan (NSB (2013)\nand Annual Reports of Druk Green Power Company Limited, Bhutan’s national power company\n(DGPCL, 2013).\n\nFor Pakistan, physical characteristics of power plants are from an existing Power Sector\nMaster Plan prepared by the SNC-Lavalin International Inc. in association with National\nEngineering Services Pakistan Limited for the National Transmission and Despatch Company\nLimited, Pakistan (NTDCL, 2011). The data are supplemented with information from an integrated\nenergy plan 2009-2022 prepared by an energy expert group for the Economic Advisory Council,\nMinistry of Finance (EAC, 2009). Information on candidate hydropower projects are from a survey\nreport prepared by Pakistan’s Water and Power Development Authority (PWAPDA, 2011). Data\nfor Sri Lanka are from the Central Electricity Boards’ Long Term Generation Expansion Planning\n2013-2032 prepared by for Government of Sri Lanka (CEB 2012). We tried to use the data from\nthe existing electricity master plan, where available, to be consistent with countries national\nelectricity expansion plan in the baseline. However, we did not force all aspects of the baseline in\nthe master plan to apply in our study, as the underlying assumptions in the master plan might be\ndifferent from ours.\n\nIn India, the list of all existing plants and their installed capacities, year of commissioning\nand fuel type were taken from various documents (CEA 2012a, 2012b, 2012c and 2012d)\npublished by the Central Electricity Authority (CEA), a statutory organization under the Central\nGovernment that advises on matters relating to\n\nMention: \"integrated\nenergy plan 2009-2022\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Prepared by an expert group, it supplements the plant records.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006397:8:0", "occurrence_id": "e102b04b24a73c88db9170b3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: characteristics are obtained from the official website of Bangladesh’s national power utility,\nBangladesh Power Development Board (BPDB, 2013), and the 2010 Power Sector Master Plan\nprepared by the Japan International Cooperation Agency and Tokyo Electric Power Company for\nthe Ministry of Power, Energy and Mineral Resources (MPEMR, 2011). For Bhutan, physical\ncharacteristics of power plants are from the National Statistical Yearbook of Bhutan (NSB (2013)\nand Annual Reports of Druk Green Power Company Limited, Bhutan’s national power company\n(DGPCL, 2013).\n\nFor Pakistan, physical characteristics of power plants are from an existing Power Sector\nMaster Plan prepared by the SNC-Lavalin International Inc. in association with National\nEngineering Services Pakistan Limited for the National Transmission and Despatch Company\nLimited, Pakistan (NTDCL, 2011). The data are supplemented with information from an integrated\nenergy plan 2009-2022 prepared by an energy expert group for the Economic Advisory Council,\nMinistry of Finance (EAC, 2009). Information on candidate hydropower projects are from a survey\nreport prepared by Pakistan’s Water and Power Development Authority (PWAPDA, 2011). Data\nfor Sri Lanka are from the Central Electricity Boards’ Long Term Generation Expansion Planning\n2013-2032 prepared by for Government of Sri Lanka (CEB 2012). We tried to use the data from\nthe existing electricity master plan, where available, to be consistent with countries national\nelectricity expansion plan in the baseline. However, we did not force all aspects of the baseline in\nthe master plan to apply in our study, as the underlying assumptions in the master plan might be\ndifferent from ours.\n\nIn India, the list of all existing plants and their installed capacities, year of commissioning\nand fuel type were taken from various documents (CEA 2012a, 2012b, 2012c and 2012d)\npublished by the Central Electricity Authority (CEA), a statutory organization under the Central\nGovernment that advises on matters relating to\n\nMention: \"Long Term Generation Expansion Planning\n2013-2032\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Prepared for the government, it supplies the generation outlook.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006398:19:2", "occurrence_id": "0da2281387eaca4880167ca2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: >68.13</u> <u>34.16</u> <u>55.40</u> <u>77.84</u> <u>66.72</u>\n_Source:_ Own computation from 2013 CSA large farm survey\n\n\n18\n\nMention: \"2013 CSA large farm survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Farm size distribution statistics are computed from this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006409:8:0", "occurrence_id": "18491eaf0236bea249a05ff0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: the capital market activity. Section 4 presents the results on which firms use these markets. Section 5\n\n\nshows the results on firm performance and the evolution of the FSD. Section 6 explores additional\n\n\nheterogeneity in firm behavior according to the type of securities issued, market of issuance, and firms’\n\n\nhome market. Section 7 studies the role of firm age. Section 8 concludes.\n\n\n**2.** **Data**\n\n\nTo assess which firms issue securities, the comparative performance of issuing and non-issuing firms,\n\n\nand the evolution of the size distribution of firms as they issue securities, we assemble a\n\n\ncomprehensive data set covering corporate bond and equity issuances in capital markets around the\n\n\nworld as well as balance sheet information on publicly listed firms. The data on firm capital raising\n\n\nactivity cover the period 1991-2011 and come from the Thomson Reuters Security Data Corporation\n\n\n(SDC) Platinum database, which provides transaction-level information on new issuances of common\n\n\nand preferred equity and publicly and privately placed bonds with an original maturity of more than\n\n\none year. <sup>5</sup> [^5: SDC Platinum collects data on security issuance mostly from filings with local regulatory agencies and stock exchanges.\nThese data are augmented with data from other sources such as offering circulars, prospectuses, surveys of investment\nbanks, brokers, and other financial advisors, news sources, trade publications, and wires.] Given that the SDC Platinum database does not collect data on debt issuances with\n\n\nmaturities shorter than one year, the data set does not cover commercial paper. For offerings in more\n\n\nthan one market, we consider each market a separate issuance. The data set includes 532,423 security\n\n\nissuances: 138,968 equity issuances and 393,455 bond issuances by listed and unlisted firms. Security\n\n\nissuances are classified as domestic or international based on the location of the main exchange where\n\n\nthe issuances take place and compared with the issuing firm’s nationality. The data set includes 411,180\n\n\nissuances in domestic markets and 116,811 issuances in foreign\n\nMention: \"SDC Platinum database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"It does not collect debt issues with maturities shorter than one year.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006410:19:2", "occurrence_id": "998e92d7e74ac20021c71212", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  complex (including the media company, hotels, theatres, etc.) by a politically connected businessman. The\ncomplex was declared a special economic zone shortly after, allowing him to benefit from tax exemptions (Ahram Online,\nvarious issues).\n\n28 Moreover, while the WBES is a survey, it covers most large firms within a sector, which typically includes the connected\nfirms in sectors where these firms are present.\n\n29 Among the 91 manufacturing sectors, 59 have no politically connected CEOs, while 32 have at least one politically\nconnected CEO; 26 have no politically connected owners, while the remaining 65 sectors have between one (18 sectors)\nand 35 (one sector) politically connected owners; and 19 have no politically connected firms of any kind, while the\nremaining 72 sectors exhibit between one (16 sectors) and 35 (two sectors) connected firms. In sectors with multiple\n\n\n18\n\nMention: \"the WBES\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Though a survey, it covers most large firms within a sector.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006400:9:0", "occurrence_id": "beab511dd22c5652a00751d3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **<u>Table 2: Effects of Health Care Worker Deaths from Ebola on Maternal Mortality</u>**\n\n\nDoctors, nurses, & midwives Maternal mortality ratio (per 100,000 live births)\n\n\n\nUpper\nbound %\n\nchange\n\n(95%)\n\n\n\nStock\npre-Ebola\n\n\n\nStock\n\npostEbola\n\n\n\n% change\n\n\n\nPreEbola\n\nratio\n(2013)\n\n\n\nMay\n2015\nratio\n\n\n\n% change\n\n\n\nLower\nbound %\n\nchange\n\n(95%)\n\n\n\nGuinea 5395 5317 -1% 650 897 **38%** 26% 50%\n\n\nLiberia 1029 946 -8% 640 1347 **111%** 76% 145%\n\n\nSierra Leone 1153 1074 -7% 1100 1916 **74%** 51% 97%\n\n\nSource: Author calculations based on Ebola mortality data from the WHO, <sup><u>8</u></sup> population and maternal mortality data\nfrom World Development Indicators, <sup><u>9</u></sup> and health worker-mortality coefficients from Speybroeck et al. <sup><u>18</u></sup>\nNote: Data on pre-Ebola stock of health workers is for the most recent years available for each country: 2004 (nurses\nand midwives) and 2005 (doctors) for Guinea, 2008 for Liberia, and 2010 for Sierra Leone.\n\n\n**<u>Table 3: Effects of Health Care Worker Deaths from Ebola on Infant Mortality*</u>**\n\n\nDoctors, nurses, & midwives Infant mortality rate (per 1,000 live births)\n\n\n\nUpper\nbound %\n\nchange\n\n(95%)\n\n\n\nStock\npre-Ebola\n\n\n\nStock\n\npostEbola\n\n\n\nLower\nbound %\n\nchange\n\n(95%)\n\n\n\n% change\n\n\n\nPreEbola\n\nrate\n(2013)\n\n\n\nMay\n2015 rate <sup>% change</sup>\n\n\n\nGuinea 5395 5317 -1% 64.9\n\nMention: \"population and maternal mortality data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Denominators and maternal mortality rates behind the projection come from this source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006400:9:0", "occurrence_id": "a867347df4c855ade39cb20c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **<u>Table 2: Effects of Health Care Worker Deaths from Ebola on Maternal Mortality</u>**\n\n\nDoctors, nurses, & midwives Maternal mortality ratio (per 100,000 live births)\n\n\n\nUpper\nbound %\n\nchange\n\n(95%)\n\n\n\nStock\npre-Ebola\n\n\n\nStock\n\npostEbola\n\n\n\n% change\n\n\n\nPreEbola\n\nratio\n(2013)\n\n\n\nMay\n2015\nratio\n\n\n\n% change\n\n\n\nLower\nbound %\n\nchange\n\n(95%)\n\n\n\nGuinea 5395 5317 -1% 650 897 **38%** 26% 50%\n\n\nLiberia 1029 946 -8% 640 1347 **111%** 76% 145%\n\n\nSierra Leone 1153 1074 -7% 1100 1916 **74%** 51% 97%\n\n\nSource: Author calculations based on Ebola mortality data from the WHO, <sup><u>8</u></sup> population and maternal mortality data\nfrom World Development Indicators, <sup><u>9</u></sup> and health worker-mortality coefficients from Speybroeck et al. <sup><u>18</u></sup>\nNote: Data on pre-Ebola stock of health workers is for the most recent years available for each country: 2004 (nurses\nand midwives) and 2005 (doctors) for Guinea, 2008 for Liberia, and 2010 for Sierra Leone.\n\n\n**<u>Table 3: Effects of Health Care Worker Deaths from Ebola on Infant Mortality*</u>**\n\n\nDoctors, nurses, & midwives Infant mortality rate (per 1,000 live births)\n\n\n\nUpper\nbound %\n\nchange\n\n(95%)\n\n\n\nStock\npre-Ebola\n\n\n\nStock\n\npostEbola\n\n\n\nLower\nbound %\n\nchange\n\n(95%)\n\n\n\n% change\n\n\n\nPreEbola\n\nrate\n(2013)\n\n\n\nMay\n2015 rate <sup>% change</sup>\n\n\n\nGuinea 5395 5317 -1% 64.9\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The compilation supplying the demographic and health series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006411:21:0", "occurrence_id": "48e9eaab935e5f325f589e88", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _A.3 A few examples of post disaster expenditures_\n\n**Seychelles**\nFollowing the 2013 floods the immediate fiscal impact of the disaster was estimated to be SR\n4.3 million (0.03 percent of GDP or 0.08 of the recurrent budget). According to information\nrecorded in the PDNA utilization of the government’s budget framework accommodated the\nincreased expenditures; no allocation for natural disasters was made to the contingency fund\nof SR 50 million. The SR4.3 million was spent on those areas identified with the greatest need.\nThese being the provision of relief goods by the Agency for Social Protection SR3.3 million;\nemergency assistance provided by the Ministry of Land and Housing SR480,000 and the\nDistrict Administrators have incurred expensed SR 0.6 million for similar activities. **Source:**\n**Government of Seychelles, 2013.**\n\n**Pakistan**\nFollowing the 2011 floods in Pakistan the additional expenditure needed to fund the necessary\nrelief, rehabilitation and reconstruction was estimated to cause equivalent to 0.3 percent of\nGDP. Consequently, the already stretched target of running a budget deficit equivalent to 4\npercent of GDP was forecast to reach 6 percent of GDP as a result of the increased expenditures.\n**Source: Government of Pakistan, 2011.**\n\n**Fiji**\nPost disaster expenditures will first and foremost be driven by the amount of cash readily\navailable to the government. This makes the time that a disaster occurs a key issue; different\namounts of funding will be available depending on whether the disaster occurs at the beginning\nor towards the end of the budget cycle. In December 2012 Fiji was struck by tropical cyclone\nEvan causing damage on the northern island of Vanua Levu and the western part of the main\nisland of Viti Levu. The fiscal year in Fiji coincides with the calendar year so the government\nhad a good idea what the full year expenditures had been and as a result\n\nMention: \"the PDNA\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"It records how the budget framework absorbed increased expenditure.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006421:9:0", "occurrence_id": "0d6707fde3508ed551bdb2a0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: covers more than 40,000 firms. For exports we use a data set created by the foreign trade office,\n\n\nSECEX ( _Secretaria Comercio Exterior_ ). SECEX provides the universe of registered trade flows at\n\n\nthe firm level. <sup>6</sup>\n\n\nTable 1 shows the main variables used in the analysis; and Tables 2 and 3 show,\n\n\nrespectively, firms export activity, and average values for the main production function variables\n\n\nand wages, according to firms’ exporting status during the sample period. From the figures in Table\n\n\n2 we observe that the proportion of exporting firms in the sample is slightly below 30%. As regards\n\n\nthe variables in Table 3, we proxy capital with assets, and include also electricity and energy as\n\n\nintermediate inputs. We use sector specific producer price indices supplied by the Instituto\n\n\nBrasileiro de Geografia e Estatistica (IBGE) to deflate the variables in the production function and\n\n\nwages, with the exception of labor. The digits level of the sector deflators corresponds to the 2\n\n\ndigits CNAE classification. Unfortunately, the survey PIA _empresa_ has not information on firms’\n\n\nprices to be able to construct firm specific deflators. As can be observed in Table 3, exporters are\n\n\nlarger in terms of output, labor, capital and materials and pay higher wages than non-exporters.\n\n\n[Table 1 around here]\n\n\n[Table 2 around here]\n\n\n[Table 3 around here]\n\n\n**3.** **Production function estimation, firm trade status and TFP**\n\n\nWe assume that firms produce using a Cobb-Douglas technology:\n\n\n_yit_ 0  _llit_  _kkit_  _mmit_  _t_  _it_  _it_ (1)\n\n\n6 There are some differences between PIA and SECEX in identifying exporters. Around 600-700 firms are identified in\n\nMention: \"data set created by the foreign trade office\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The universe of registered firm-level trade flows is held in this dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006421:9:0", "occurrence_id": "19a5eac5e95f36e30f447610", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: covers more than 40,000 firms. For exports we use a data set created by the foreign trade office,\n\n\nSECEX ( _Secretaria Comercio Exterior_ ). SECEX provides the universe of registered trade flows at\n\n\nthe firm level. <sup>6</sup>\n\n\nTable 1 shows the main variables used in the analysis; and Tables 2 and 3 show,\n\n\nrespectively, firms export activity, and average values for the main production function variables\n\n\nand wages, according to firms’ exporting status during the sample period. From the figures in Table\n\n\n2 we observe that the proportion of exporting firms in the sample is slightly below 30%. As regards\n\n\nthe variables in Table 3, we proxy capital with assets, and include also electricity and energy as\n\n\nintermediate inputs. We use sector specific producer price indices supplied by the Instituto\n\n\nBrasileiro de Geografia e Estatistica (IBGE) to deflate the variables in the production function and\n\n\nwages, with the exception of labor. The digits level of the sector deflators corresponds to the 2\n\n\ndigits CNAE classification. Unfortunately, the survey PIA _empresa_ has not information on firms’\n\n\nprices to be able to construct firm specific deflators. As can be observed in Table 3, exporters are\n\n\nlarger in terms of output, labor, capital and materials and pay higher wages than non-exporters.\n\n\n[Table 1 around here]\n\n\n[Table 2 around here]\n\n\n[Table 3 around here]\n\n\n**3.** **Production function estimation, firm trade status and TFP**\n\n\nWe assume that firms produce using a Cobb-Douglas technology:\n\n\n_yit_ 0  _llit_  _kkit_  _mmit_  _t_  _it_  _it_ (1)\n\n\n6 There are some differences between PIA and SECEX in identifying exporters. Around 600-700 firms are identified in\n\nMention: \"SECEX\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"It provides registered trade flows at the firm level.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006421:9:0", "occurrence_id": "4ef6085de30ab9280d1db5bc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: covers more than 40,000 firms. For exports we use a data set created by the foreign trade office,\n\n\nSECEX ( _Secretaria Comercio Exterior_ ). SECEX provides the universe of registered trade flows at\n\n\nthe firm level. <sup>6</sup>\n\n\nTable 1 shows the main variables used in the analysis; and Tables 2 and 3 show,\n\n\nrespectively, firms export activity, and average values for the main production function variables\n\n\nand wages, according to firms’ exporting status during the sample period. From the figures in Table\n\n\n2 we observe that the proportion of exporting firms in the sample is slightly below 30%. As regards\n\n\nthe variables in Table 3, we proxy capital with assets, and include also electricity and energy as\n\n\nintermediate inputs. We use sector specific producer price indices supplied by the Instituto\n\n\nBrasileiro de Geografia e Estatistica (IBGE) to deflate the variables in the production function and\n\n\nwages, with the exception of labor. The digits level of the sector deflators corresponds to the 2\n\n\ndigits CNAE classification. Unfortunately, the survey PIA _empresa_ has not information on firms’\n\n\nprices to be able to construct firm specific deflators. As can be observed in Table 3, exporters are\n\n\nlarger in terms of output, labor, capital and materials and pay higher wages than non-exporters.\n\n\n[Table 1 around here]\n\n\n[Table 2 around here]\n\n\n[Table 3 around here]\n\n\n**3.** **Production function estimation, firm trade status and TFP**\n\n\nWe assume that firms produce using a Cobb-Douglas technology:\n\n\n_yit_ 0  _llit_  _kkit_  _mmit_  _t_  _it_  _it_ (1)\n\n\n6 There are some differences between PIA and SECEX in identifying exporters. Around 600-700 firms are identified in\n\nMention: \"survey PIA _empresa_\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"It lacks firm price information needed for specific deflators.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006424:23:1", "occurrence_id": "e03bb2e164be62e0a288bf3c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  dependent variable is the ratio of the wife’s report of the migrant’s monthly **e** arnings to that of\nthe migrant. Bootstrapped standard errors (with 500 replications) are displayed in parent **h** esis, clustered at\nthe district level.\n\n\n\n*** denotes statistical significance at the 1% level, ** at the 5% level and * at the 10% l **e** vel.\n_Source:_ Authors’ analysis based on data from QSKF.\n\n\n\nReferring to column 2, the positive coefficient on the reported remittance gap is small\n\n\nand statistically not different from zero ( _p_ = 0.95 ). We also performed a similar exercise using\n\n\nthe discrepancy in reported hours worked per week by the migrant, which is unobserved by the\n\n\n22\n\nMention: \"QSKF\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The reported remittance gap is analysed using this household survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006426:17:0", "occurrence_id": "aeb38f0cb0b97c5a29aa28d0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <u>Promoting Democracy in Fragile States</u>\n\n\nconstantly ranks at the very bottom of the United Nations Human Development Index, and\n\n\nthe study areas were especially poor and underserved by infrastructure.\n\n\nThe 2011 election season arrived with much reason to worry about the quality of elec\n\ntoral participation in the countryside. The country still relied on UNMIL as the primary\n\n\nsecurity and infrastructure provider. The Liberian national police, for example, provided\n\n\nvery patchy coverage and were frequently accused of corruption by watchdog organiza\n\ntions and ordinary Liberians (International Crisis Group, 2012). UNMIL was limited in the\n\n\nextent to which its mandate permitted its engagement in the countryside, however, leaving\n\n\nmuch of rural Liberia completely uncovered by state level institutions. In a 2011 survey,\n\n\nSandefur and Siddiqi (2013) found that only four percent of crimes and disputes in ru\n\nral Liberia were taken to the formal justice system. In a 2008-2010 panel study in rural\n\n\nareas Blair, Blattman and Hartman (2012) found that while overt collective violence was\n\n\nrare, property disputes were pervasive, indicators of intra-communal cohesion were deteri\n\norating, and inter-tribal biases and perceptions of inter-ethnic inequity remained endemic.\n\n\nBlair et al. found that key correlates of violence included factors related to ethnic diver\n\nsity, a legacy of the ethnically charged war (Boas and Utas, 2014). Much of the country\n\n\nremained accessible only via unpaved pathways, and nearly the entirety of the country out\n\nside the capital of Monrovia lacked access to a reliable electricity grid or piped water. A\n\n\n2007 survey conducted by the authors found that in rural areas, two-thirds of respondents\n\n\nlacked regular access to radio, the primary form of media access in the country. Word\n\nof-mouth was the primary channel of political information in rural areas. These logistical\n\n\nand communication challenges greatly hampered voter registration efforts (Look, 2011).\n\n\nLiberians and international onlookers worried\n\nMention: \"United Nations Human Development Index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"One country ranks constantly at the very bottom of this index.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006429:13:0", "occurrence_id": "2953345797d49009fa801351", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: county average dibao expenditure. <sup>5</sup> [^5: In the few cases of missing village‐level (county‐level) data we use county‐level (village‐level) information to\nimpute missing values.] This approach effectively assumes an egalitarian distribution\n\n\nwithin villages or within counties of dibao benefits among dibao recipients.\n\n\nThe dibao participation rates in the CHIP rural survey are lower than the aggregate rates\n\n\nimplied by official data. <sup>6</sup> [^6: Gao, Garfinkel and Zhai (2009) find that in the CHIP urban data (for 2002) the rate of dibao participation is also\nlower than the officially reported rate.] To some extent this reflects the selection of provinces in the CHIP\n\n\nsample, but the discrepancy remains even for the nine CHIP provinces (to be discussed in more\n\n\ndetail below). The reason why the CHIP sample has lower dibao participation rates than the\n\n\nofficial data is not clear. It is possible that dibao households are under‐sampled in the CHIP\n\n\nsurvey. Under‐sampling of poor households—which are presumably more likely to be dibao\n\n\nrecipients—is a known feature of the NBS household survey samples from which the CHIP\n\n\nsamples are drawn. It is also possible that some dibao households do not report their dibao\n\n\nparticipation. Households may not be aware that the transfers they received were from the\n\n\ndibao program, or they may not want to disclose their participation in the program. A third\n\n\npossibility is that the official numbers overstate true participation rates. It is widely accepted\n\n\nthat local‐level governments in China massage their statistics so as to appear to comply with\n\n\ncentral government policy targets and in order to obscure local irregularities in program\n\n\nimplementation (Hvistendahl 2013).\n\n\n**IV.** **Patterns of dibao participation, thresholds and transfers**\n\n\nConsistent with the official data in Table 1, our data show substantial expansion of the\n\n\ndibao program since 2007 (Table 3). The mean dibao threshold, calculated using\n\nMention: \"CHIP rural survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Its participation rates fall below those implied by official data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006429:13:0", "occurrence_id": "542ae4be2ff00c2df97cf08b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: county average dibao expenditure. <sup>5</sup> [^5: In the few cases of missing village‐level (county‐level) data we use county‐level (village‐level) information to\nimpute missing values.] This approach effectively assumes an egalitarian distribution\n\n\nwithin villages or within counties of dibao benefits among dibao recipients.\n\n\nThe dibao participation rates in the CHIP rural survey are lower than the aggregate rates\n\n\nimplied by official data. <sup>6</sup> [^6: Gao, Garfinkel and Zhai (2009) find that in the CHIP urban data (for 2002) the rate of dibao participation is also\nlower than the officially reported rate.] To some extent this reflects the selection of provinces in the CHIP\n\n\nsample, but the discrepancy remains even for the nine CHIP provinces (to be discussed in more\n\n\ndetail below). The reason why the CHIP sample has lower dibao participation rates than the\n\n\nofficial data is not clear. It is possible that dibao households are under‐sampled in the CHIP\n\n\nsurvey. Under‐sampling of poor households—which are presumably more likely to be dibao\n\n\nrecipients—is a known feature of the NBS household survey samples from which the CHIP\n\n\nsamples are drawn. It is also possible that some dibao households do not report their dibao\n\n\nparticipation. Households may not be aware that the transfers they received were from the\n\n\ndibao program, or they may not want to disclose their participation in the program. A third\n\n\npossibility is that the official numbers overstate true participation rates. It is widely accepted\n\n\nthat local‐level governments in China massage their statistics so as to appear to comply with\n\n\ncentral government policy targets and in order to obscure local irregularities in program\n\n\nimplementation (Hvistendahl 2013).\n\n\n**IV.** **Patterns of dibao participation, thresholds and transfers**\n\n\nConsistent with the official data in Table 1, our data show substantial expansion of the\n\n\ndibao program since 2007 (Table 3). The mean dibao threshold, calculated using\n\nMention: \"CHIP\n\n\nsurvey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Dibao households may be under-sampled in this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006429:13:0", "occurrence_id": "014fe000b29838c19703e747", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: county average dibao expenditure. <sup>5</sup> [^5: In the few cases of missing village‐level (county‐level) data we use county‐level (village‐level) information to\nimpute missing values.] This approach effectively assumes an egalitarian distribution\n\n\nwithin villages or within counties of dibao benefits among dibao recipients.\n\n\nThe dibao participation rates in the CHIP rural survey are lower than the aggregate rates\n\n\nimplied by official data. <sup>6</sup> [^6: Gao, Garfinkel and Zhai (2009) find that in the CHIP urban data (for 2002) the rate of dibao participation is also\nlower than the officially reported rate.] To some extent this reflects the selection of provinces in the CHIP\n\n\nsample, but the discrepancy remains even for the nine CHIP provinces (to be discussed in more\n\n\ndetail below). The reason why the CHIP sample has lower dibao participation rates than the\n\n\nofficial data is not clear. It is possible that dibao households are under‐sampled in the CHIP\n\n\nsurvey. Under‐sampling of poor households—which are presumably more likely to be dibao\n\n\nrecipients—is a known feature of the NBS household survey samples from which the CHIP\n\n\nsamples are drawn. It is also possible that some dibao households do not report their dibao\n\n\nparticipation. Households may not be aware that the transfers they received were from the\n\n\ndibao program, or they may not want to disclose their participation in the program. A third\n\n\npossibility is that the official numbers overstate true participation rates. It is widely accepted\n\n\nthat local‐level governments in China massage their statistics so as to appear to comply with\n\n\ncentral government policy targets and in order to obscure local irregularities in program\n\n\nimplementation (Hvistendahl 2013).\n\n\n**IV.** **Patterns of dibao participation, thresholds and transfers**\n\n\nConsistent with the official data in Table 1, our data show substantial expansion of the\n\n\ndibao program since 2007 (Table 3). The mean dibao threshold, calculated using\n\nMention: \"NBS household survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its known under-sampling of poor households propagates into the child dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006430:21:1", "occurrence_id": "cd38e920c7580ca3b82e24eb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  but retroceded in 2013. The introduction of performance\nagreements is associated with the implementation of school-level performance agreements.\nWhile the state remains above the regional average, the gap is rapidly closing because Rio de\nJaneiro and Espiritu Santo recorded large improvements.\n\n\nSimilarly, in the case of Rio de Janeiro, there is a clear discontinuity that is concomitant to the\nintroduction of the results agreements and bonus system for school employees in 2010. After\nshowing no change in three rounds of IDEB, the state recorded a sizable increase of 14 percent in\n2011 and 12.5 percent in 2013. As a result, Rio de Janeiro rose from the 23 <sup>rd</sup> position in the\nnational ranking of IDEB 2009 to 5 <sup>th</sup> place in 2013. The state achieved one of Brazil’s most\nremarkable turn-arounds in the education sector.\n\n\n_Before-and-After Comparison of Outcomes_\n\n\nThe comparison of the average measures corresponding to the periods of time before and after\nthe introduction of performance agreements and variable pay shows that in several cases there\nare significant differences in outcomes after the implementation of the intervention (Table 3).\nThe before-and-after design is a quasi-experimental model that compares two groups of\nobservations, before and after time points. The state-years preceding the program are averaged\nto estimate the counterfactual and the later ones to measure results.\n\n\n20\n\nMention: \"national ranking of IDEB 2009\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its ranking position improves markedly by the later year.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006431:34:0", "occurrence_id": "a01fe27108957ce1b68e244e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |Nigerian 2006’s SAM|GTAP database version 8.1|Aggregated 21 GTAP sectors|\n|---|---|---|\n|No. Code<br>Description|No. Code Description|No. Code<br>Description|\n||34<br>nmm Mineral products nec|<br> <br> <br> <br> <br> <br> <br> <br> <br> <br> <br> <br> <br> <br> <br> <br> <br> <br> <br> <br> <br> <br> <br>|\n||35<br>i_s<br>Ferrous metals|35<br>i_s<br>Ferrous metals|\n||36<br>nfm<br>Metals nec|36<br>nfm<br>Metals nec|\n||37<br>fmp<br>Metal products|37<br>fmp<br>Metal products|\n||38<br>mvh<br>Motor vehicles and parts|38<br>mvh<br>Motor vehicles and parts|\n||40<br>ele<br>Electronic equipment|40<br>ele<br>Electronic equipment|\n||41<br>ome<br>Machinery and equipment<br>nec|41<br>ome<br>Machinery and equipment<br>nec|\n||42<br>omf<br>Manufactures nec|42<br>omf<br>Manufactures nec|\n|48 croil<br>Refined oil|32<br>p_c<br>Petroleum, coal products|16<br>RefinedOil\n\nMention: \"GTAP database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Version 8.1 supplies the global counterpart to the national matrix.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006433:27:0", "occurrence_id": "61f4ec6a42bade5d311ab7ba", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ##### Appendix 3: Additional Material\n\n**Average marginal effects of well-being over the lifespan in the synthetic panel regressions using**\n\n\n**LiTS data (2006–2010)**\n\n\n\n\n\n\n\n\n\n\n|Squared<br>(1) full sample<br>5<br>4.5 Prediction<br>4<br>3.5<br>3 Linear<br>2.5<br>2<br>23 28 33 38 43 48 53 58 63 68 73 78 83 88 93 98|Cubic<br>(2) full sample<br>5<br>4.5<br>4<br>3.5<br>3<br>2.5<br>2<br>23 28 33 38 43 48 53 58 63 68 73 78 83 88 93 98|\n|---|---|\n|2<br>2.5<br>3<br>3.5<br>4<br>4.5<br>5<br>Linear Prediction<br>23 28 33 38 43 48 53 58 63 68 73 78 83 88 93 98<br>(3) men|2<br>2.5<br>3<br>3.5<br>4<br>4.5<br>5<br>23 28 33 38 43 48 53 58 63 68 73 78 83 88 93 98<br>(4) men|\n|2<br>2.5<br>3<br>3.5<br>4<br>4.5<br>5<br>Linear Prediction<br>23 28 33 38 43 48 53 58 63 68 73 78 83 88 93 98<br>Age<br>(5) women|2<br>2.5<br>3<\n\nMention: \"LiTS data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"A synthetic panel across these years supports the well-being regressions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006435:5:0", "occurrence_id": "d8673909af42969b1ecdd600", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 4\n\n\n**II.  HOW IS DIVERSIFICATION MEASURED?**\n\n\nMeasures of economic diversification need to look beyond trade, to capture domestic sector\n\n\ndiversification and the underlying dynamic process of structural transformation. _Trade_\n\n\n_diversification_ and _domestic diversification_ are in principle interlinked, the former reflecting\n\n\ndiversification in the external sector, and the latter capturing diversification in the domestic\n\n\nproduction process across sectors. An underlying theme of this paper is that focusing on the\n\n\nentire structure of production paints a more comprehensive and illuminating picture. Therefore,\n\n\nthe two dimensions of diversification are evaluated simultaneously, filling a gap in the existing\n\n\nliterature, which has treated them independently. In addition, the analysis focuses on\n\n\n“diversification spurts,” that is, rapid, sustained, significant spells of diversification.\n\n\nTrade diversification can be achieved along several dimensions. First, diversification may\n\n\noccur across either products or trading partners. Second, product diversification may occur\n\n\nthrough the introduction of new product lines (the _extensive_ margin), or a more balanced mix\n\n\nof existing exports ( _intensive_ margin). Finally, product-quality upgrading represents a slightly\n\n\ndifferent notion and is evidenced by higher prices for existing exports. Our main data source\n\n\nfor trade is an updated version of the UN–NBER data set, which harmonizes UN COMTRADE\n\nbilateral trade flow data at the 4-digit SITC (Rev. 1) level. <sup>3</sup> [^3: The dataset combines importer- and exporter-reported data from COMTRADE to maximize\ncomprehensiveness, while ensuring internal consistency, using the methodology of Asmundson\n(forthcoming).] However, while the existing\n\n\nliterature typically focuses on the post-1988 period, this paper uses data extending back to\n\n\n1962. The extended time dimension turns out to be greatly helpful in examining relationships\n\n\nmore comprehensively.\n\n\nAnalysis of domestic diversification in frontier economies required construction of a new data\n\n\nset. This paper examines diversification in sectoral output and the sectoral allocation of labor\n\n\nusing data from existing and new sources. Existing data sets include measures of value added\n\n\nfor 28 manufacturing sectors, during 1985–2010 (from UNIDO\n\nMention: \"UN–NBER data set\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The main trade source, harmonising bilateral flows at four-digit classification.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006435:5:0", "occurrence_id": "4cf4da4a248b6525a231cc34", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 4\n\n\n**II.  HOW IS DIVERSIFICATION MEASURED?**\n\n\nMeasures of economic diversification need to look beyond trade, to capture domestic sector\n\n\ndiversification and the underlying dynamic process of structural transformation. _Trade_\n\n\n_diversification_ and _domestic diversification_ are in principle interlinked, the former reflecting\n\n\ndiversification in the external sector, and the latter capturing diversification in the domestic\n\n\nproduction process across sectors. An underlying theme of this paper is that focusing on the\n\n\nentire structure of production paints a more comprehensive and illuminating picture. Therefore,\n\n\nthe two dimensions of diversification are evaluated simultaneously, filling a gap in the existing\n\n\nliterature, which has treated them independently. In addition, the analysis focuses on\n\n\n“diversification spurts,” that is, rapid, sustained, significant spells of diversification.\n\n\nTrade diversification can be achieved along several dimensions. First, diversification may\n\n\noccur across either products or trading partners. Second, product diversification may occur\n\n\nthrough the introduction of new product lines (the _extensive_ margin), or a more balanced mix\n\n\nof existing exports ( _intensive_ margin). Finally, product-quality upgrading represents a slightly\n\n\ndifferent notion and is evidenced by higher prices for existing exports. Our main data source\n\n\nfor trade is an updated version of the UN–NBER data set, which harmonizes UN COMTRADE\n\nbilateral trade flow data at the 4-digit SITC (Rev. 1) level. <sup>3</sup> [^3: The dataset combines importer- and exporter-reported data from COMTRADE to maximize\ncomprehensiveness, while ensuring internal consistency, using the methodology of Asmundson\n(forthcoming).] However, while the existing\n\n\nliterature typically focuses on the post-1988 period, this paper uses data extending back to\n\n\n1962. The extended time dimension turns out to be greatly helpful in examining relationships\n\n\nmore comprehensively.\n\n\nAnalysis of domestic diversification in frontier economies required construction of a new data\n\n\nset. This paper examines diversification in sectoral output and the sectoral allocation of labor\n\n\nusing data from existing and new sources. Existing data sets include measures of value added\n\n\nfor 28 manufacturing sectors, during 1985–2010 (from UNIDO\n\nMention: \"UN COMTRADE\n\nbilateral trade flow data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The underlying flows combined into the harmonised dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006438:9:0", "occurrence_id": "de49cf4a30d89c87dddc2f79", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: this is not the case, we use 2000 Census data and other sources to estimate the joint\ndistribution. Appendix A.3 provides further details.\n\n##### **3 Descriptive Statistics**\n\n\nPopulation in Washington, D.C. peaked in the 1950s at about 802,000, declined steadily\nto 572,000 in 2000, and bounced back to 602,000 in 2010. Between 2003 and 2007 it\ngrew from 577,000 up to 586,000, although school-age population declined from 82,000\nto 76,000 according to the Population Division of the U.S. Census Bureau and American\nFact Finder. The city’s racial breakdown has changed as well, going from 28, 65 and 5\npercent white, black and Hispanic in 1990 to 32, 55 and 8 percent respectively in 2007.\nDespite these changes, the city remains geographically segregated by race and income,\nwith large disparities among races. For instance, in 2013 median household income was\n$112,000 for whites, but only $38,000 for blacks and $51,000 for Hispanics.\n\n\n**3.1** **Basic trends in school choice**\n\n\nAs Figure 1A shows, total enrollment declined by about 6,000 students over our sample\nperiod, yet charter enrollment grew by that amount. Since private school enrollment\nremained steady at about 21 percent of total enrollment, market share for public schools\nfell from 66 to 56 percent yet rose from 13 to 22 percent for charters. The number\nof charter school campuses more than doubled, from 27 to 59, whereas the number of\npublic and private school campuses declined slightly due to a few closings and mergers (Figure 1B and Table 1). Over the sample period, 43 percent of private schools\nwere Catholic, 24 percent belonged to other religions and 32 percent were nonsectarian.\nAverage charter reimbursement grew from $7,900 in Fall 2003 to $9,600 in Fall 2007.\n\nMention: \"2000 Census data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Used with other sources to estimate a joint distribution where needed.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006441:12:0", "occurrence_id": "1cdcaa826ce2edbc04bb0bc2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 2. Select Education Outcomes, by Department of Birth**\n\n\nSource: Bolivian 2012 census.\n\n\n**Table 2. Outcomes by Gender and Ethnicity**\n\n\n\nAfroBolivian/\n\nOther\n<u>Ethnicity</u>\n\n\n\nForeign/\nNot Listed\n\n\n\nOutcome\n\n\n\nNon- Other\nAymara Quechua\nIndigenous Indigenous\n\n\n\nLiteracy\nRate\n\n\n\nMale 98.7 97.6 94.1 95.7 97.3 97.1\n\n\n\nRate <u>Female</u> <u>97.0</u> <u>89.6</u> <u>83.4</u> <u>90.7</u> <u>93.4</u> <u>94.8</u>\n\nPrimary Male 86.2 75.9 59.9 70.2 75.8 81.2\nSchool\n\nFemale 81.5 56.4 47.7 62.6 69.0 76.7\n\n<u>Completion</u>\n\n\n\nMale 86.2 75.9 59.9 70.2 75.8 81.2\n\n\n\nSecondary\nSchool\n<u>Completion</u>\n\n\n\nFemale 81.5 56.4 47.7 62.6 69.0 76.7\n\n\nMale 61.6 46.1 30.1 36.1 45.8 52.3\n\n\nFemale 58.1 31.8 25.2 32.0 41.9 48.8\n\n\nMale 10.8 8.6 6.8 7.9 9.0 10.1\n\n\n\nYears of\nEducation\n\n\n\nEducation <u>Female</u> <u>10.0</u> <u>6.1</u> <u>5.1</u> <u>6.7</u> <u>7.9</u> <u>9.4</u>\n\nNote:\n\nMention: \"Bolivian 2012 census\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Education outcomes by department of birth come from this enumeration.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006443:24:0", "occurrence_id": "e2efa6f2a8d6e66b65fd6b20", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **<u>Appendix table 1: Village characteristics used in the regressions</u>**\n\n**Before** **After**\n**<u>Outside</u>** **<u>Inside</u>** **<u>Outside</u>** **<u>Inside</u>**\nTotal population 1,433 2,199 1,554 2,323\nShare of households with access to rural pension 0.000 0.000 0.493 0.657\nShare of households with access to cooperative health scheme 0.529 0.567 0.865 0.892\nAmount of agricultural subsidy (yuan/mu) 68,491 90,275 201,608 268,608\nDistance to the nearest county road (km) 8.382 6.375 8.231 6.141\nShare of asphalt village road 0.046 0.127 0.054 0.123\nShare of cement village road 0.463 0.651 0.757 0.786\nShare of villages with primary school 0.448 0.307 0.272 0.175\n<u>Share of villages with medical stations</u> <u>0.728</u> <u>0.806</u> <u>0.820</u> <u>0.852</u>\n_Source_ : Own computation from 2014 Chengdu village survey based on 529 villages.\n\n\n23\n\nMention: \"2014 Chengdu village survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Based on 529 villages, it underlies the computed statistics.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006455:6:0", "occurrence_id": "8e81f29c0bc9490925c26e25", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: had affected a distinct legal institution in former colonies today, which had been imported by different\ncolonial powers. At the same time, even among former colonies governed by a same colonizer, current\ninstitutional quality greatly varies because of distinctive local culture and cultural norms. There is a\nconnection between degrees of success in the institutional transfer and the current disparities in institutional\nquality in former colonies. For example, for Singapore, South Africa, and Pakistan, Britain was the colonial\nmother country in his paper, but using a resulting aggregated indicator <sup>2</sup> [^2: Originally the World Bank dataset uses six aggregate, perception‐based indicators to measure institutional\nquality. Each indicator ranges from ‐2.50 (least efficient institutional quality) to +2.50 (best quality) and is based on\nempirically collected data sources. The six indicators are interrelated.] to measure the quality of the\ninstructional framework, its scores varied, 1.57, 0.41, and -1.09, respectively. Therefore, countries with\nsimilar colonial influence rarely have the same regulatory framework in their statutory laws.\n\n\n**III.** **Classification of regulations for freight forwarders**\n\n\nThis note focuses on the regulations relevant to freight forwarding business. Logistics services can be\noffered by specialized operators such as trucking companies or customs clearing agents. However, little\nresearch has been done on the freight forwarding business despite its importance as a facilitator for the\nseamless movement of goods. Taking the official description of freight forwarding and logistics services\nby FIATA into account, this note narrows down its definition to help understand the regulatory complexity.\n\n\nDifferent regulators such as government authorities or industry associations have different levels of legal\nenforcement power on logistics operators. Regulations outline the requirements for operators to start their\nbusiness (market entry / access) and to operate their business (operations). In addition, foreign operators\noften encounter additional formal requirements while domestic operators do not, but self-regulations can\nbe equally applied to all operators\n\nMention: \"World Bank dataset\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Originally built from six perception-based indicators of institutional quality.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006432:64:9", "occurrence_id": "89091867c75d9daf7b5f69fa", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ]<br>[0.015]<br>[0.014]|0.026<br>-0.031<br>-0.020<br>[0.085]<br>[0.080]<br>[0.103]|\n|N<br>1,957<br>1,942<br>1,922<br>1,892<br>1,876|1,846<br>1,836<br>1,815|1,617<br>1,423<br>1,403|\n\n\n\n_Notes:_ The predicted values are based on the estimates in Table S5. Statistical significance is *** at 1 percent level, ** at 5 percent level, * at 10 percent level.\n\n\n_Source:_ Authors' analysis based on data from 2003 and 2008 Demographic and Health Surveys for Turkey.\n\n\n63\n\nMention: \"2003 and 2008 Demographic and Health Surveys for Turkey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The analysis is carried out on these two rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006457:18:0", "occurrence_id": "f8e53f5e2b1a05a42796847b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: by the Federal Reserve, there has been a trend toward the overall liberalization of capital\ntransactions, notably on capital and money market instruments. Globally, the world is\nroughly evenly split between countries relying on an exchange rate anchor for monetary\npolicy (mostly small islands and members of currency unions in Africa, for example), and\nthose with more independent monetary policy (including most high-income countries).\n\n\n4. _Cities and livability_ . Apart from the economic arguments suggesting that agglomeration\neconomies can be significant, “smart cities” have become a dominant theme in the recent\neconomic development literature. Fuller and Romer (2014) argue that “Nothing else [other\nthan the urbanization project] will create as many opportunities for social and economic\nprogress.” The New Climate Economy Report 2014 argues that cities are crucial for\nsustainable development and that “the stakes for growth, quality of life and carbon\nemissions could not be higher.”\n\n\nGlaeser (2010) <sup>15</sup> [^15: Glaser (2010) Agglomeration Economics…] shows a strong link between area density and per capita gross\nmetropolitan product at all income levels. Density is strongly correlated with wages and\nproductivity, as well as with future housing price growth. This seems to reflect the benefits\naccruing from labor market pooling and the exchange of ideas, rather than from the benefit\nof reduced transport costs of goods between, for example, suppliers, manufacturers and\ncustomers. Skilled labor pooling seems to be particularly important. Unfortunately, East\nAsian countries have paid scant attention to cities. Among developing East Asia, Bangkok\nis ranked at 117, Beijing at 118, Manila at 136, Ho Chi Minh and Hanoi at 152 and 153 in\nthe Mercer 2015 City Rankings. Only Kuala Lumpur, at 84, among all cities in developing\nAsia, breaks into the top 100 global cities.\n\n\n5. _Cohesion and inequality_ . We had argued that middle-income countries would need to pay\nspecial attention to inequality because the shifting growth and urbanization strategies\nwould likely worsen the\n\nMention: \"the Mercer 2015 City Rankings\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"City liveability positions across developing Asia come from this ranking.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006461:40:0", "occurrence_id": "7cd323c3187fe82fa859e9b3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure A2. Age–wage profile, 2011**\n\n\n_Source:_ Authors’ calculations based on RLMS, 2011.\n\n\n39\n\nMention: \"RLMS, 2011\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The age-wage profile is calculated from this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006462:5:0", "occurrence_id": "bf124246ce93424d7cc3ee6e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: states, and these states lag behind on measures like poverty reduction and other develop\n\nmental outcomes (World Bank 2011b). <sup>1</sup> Understanding the cost of public infrastructure is\n\n\nimportant in particular for these countries in which resources for public investment are of\n\nten scarce. Further, public work contracts, including roads, are subject to substantial levels\n\n\nof corruption. According to Transparency International’s Bribe Payers Survey of over three\n\n\nthousand business executives worldwide, public works contracts and construction is the sec\n\ntor with the highest propensity of paying bribes to officials and other firms (Transparency\n\n\nInternational 2011). As this paper attempts to establish a first set of facts on differences in\n\n\ncosts in low and middle income countries, a focus on the link between corruption and costs\n\n\nis a natural priority. <sup>2</sup>\n\n\nA review by the World Bank’s Transport Research Support Program on the roads sector\n\n\nin conflict countries states that _“...projects_ _that_ _take_ _place_ _in_ _conflict_ _settings_ _would_ _almost_\n\n\n_always_ _be_ _more_ _costly_ _than_ _in_ _other_ _settings_ _because_ _of_ _challenges_ _such_ _as_ _insecurity_ _and_ _low_\n\n\n_government_ _capacity”_ (Rebosio and Wam 2011). Higher costs can be due to the costs of\n\n\nmonitoring an areas security situation, the potential for incurring substantial risks to visit\n\n\nthe construction site, and the possibility of limitations on planning. In addition to pro\n\ntection of the staff working on the particular roads project, firms also risk supplies being\n\n\ncut off due to disruptions of transport networks. <sup>3</sup> Not only the construction but also the\n\n\nprocurement process can be riskier in conflict countries. Rebosio and\n\nMention: \"Bribe Payers Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Over three thousand executives report corruption in public works contracts.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006463:31:0", "occurrence_id": "88c34a44044ff26ed5488120", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _<mark>Notes:</mark>_ <mark>The table presents fuzzy regression discontinuity estimates of the effect of differential</mark>\n<mark>aerial spraying on cultivation around the 10 km cutoff. Columns 1 to 4 pool the years 2006 to</mark>\n<mark>2010; while columns 5 to 8 pool the years 2008 to 2010. Results use different approximations</mark>\n<mark>of the conditional expectation, as indicated in the top of each panel. Moreover, sample</mark>\n<mark>restrictions and the choice of bandwidth is indicated at the bottom of each panel. In all</mark>\n<mark>models we include municipality and year specific intercepts. Standard errors robust against</mark>\n<mark>heteroskedasticity and serial correlation within cells are reported in parenthesis. Estimates</mark>\n<mark>with</mark> <sup>***</sup> <mark>are significant at the 1%, those with</mark> <sup>**</sup> <mark>are significant at the 5%, and those with</mark> <sup>*</sup> <mark>are</mark>\n<mark>significant at the 10%.</mark>\n_<mark>Source: Authors’ analysis based on data from United Nations Office of Drugs and Crime,</mark>_\n_<mark>UNODC.</mark>_\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n|Col1|Table 3. Conditional Differences in Differences Estimate of Being in Sprayed Region.|Col3|Col4|Col5|Col6|Col7|Col8|Col9|Col10|[RS6]|\n|---|---|---|---|---|---|---\n\nMention: \"United Nations Office of Drugs and Crime\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its records underlie the reported analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006473:25:0", "occurrence_id": "4bd042109ae3db9f0e472a13", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Indicators database (2013 edition) for identifying country-level variables that cannot be computed\ndirectly from the surveys.\n\n\nSince the estimation is performed in a first difference setting with Huber-corrected robust standard\nerrors and year dummies, our results control for cohort time-invariant characteristics. All\nregressions are performed separately for males and females. We test with a series of specifications\nand report the model that shows more stability and consistency in the coefficient estimates.\n\n\nTable 1 presents the relationship between the changes in the share of Osow youth in each cohort\nof females at the stage when they shift between ages 12-14 to ages 15-17, from ages 15-17 to 1820, from 18-20 to 21-23, and from 21-23 to 24-26 years of age, respectively (all independent\nvariables refer to changes during the respective years of observation). In this way, we are able to\ntrace the dynamics of the association between the change in the share of Osow youth over the life\ncycle of each specific cohort, and the set of variables characterizing changes in community and\nmacro factors that occur in the respective country exactly during the same time that the cohort age\nshift occurs.\n\n\nThe first variable included is the share of the population that resides in rural areas at each point of\nobservation (from the World Development Indicators database). Since upper Secondary services\nare generally more easily available in urban areas, we would expect that the larger the share of the\npopulation in rural settings, the larger the proportion of adolescents that do not have direct access\nto the education services when transiting from Lower to Upper Secondary school, and the group\nmore likely to drop out. Since employability is normally low at these ages, this would presumably\nbe associated with an increase in the proportion of individuals in the Osow youth group.\n\n\nAccording to our results, this variable is in fact positive and highly significant for the transition at\nthe younger ages from 12-14 to the 15-17 range.\n\nMention: \"World Development Indicators database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The rural population share at each observation comes from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006474:12:0", "occurrence_id": "584b0368c44f7871027043b3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: countries shows that the gender gap in math is present as early as 4th grade (Bharadwaj et al.\n\n\n2012). This is not the case for Mexico. In order to assess whether our sample is representative\n\n\nof the Mexican population, we follow the nationwide cohort of 6th grade ENLACE takers in\n\n\nthe year 2007 over to 9th (in 2010) and 12th grade (in 2013). Girls do consistently better\n\n\nthan boys in language and the gap stays constant throughout the different grades. Neither\n\n\nin 6th grade nor in 9th grade is there evidence of a gender gap in math, but girls’ 12th grade\n\n\nscore in math is 30 points (0.30 _σ_ ) lower than boys’ (Table 1). The gender gap in 12th grade\n\n\nmay be partly explained by differential selection: 28.6% of the boys who took the ENLACE\n\n\n6th grade in 2007 completed the 12 grade exam in 2013, as opposed to 34.9% of the girls.\n\n\n**3.3** **Follow-up** **Data**\n\n\nStudents enrolled in 10th grade in 2009 were supposed to complete high school in 2012. <sup>16</sup> We\n\n\nuse data from the 2012 12th grade ENLACE exam to measure the three main outcomes of\n\n\ninterest: the probability of taking the test, math scores, and Spanish scores. We interpret the\n\n\nprobability of taking the 12th grade test in 2012 as a proxy for the probability of completing\n\n\nupper secondary on time. 61% of the students surveyed at the baseline took part in the\n\n\n12th grade ENLACE exam three years later. Not observing a student who was originally\n\n\nenrolled in EMS in 2009 take part in the 12th grade ENLACE exam in 2012 has four possible\n\n\nexplanations: 1) the student dropped out of school at any point between 9th and 12th grade,\n\n\n2\n\nMention: \"2012 12th grade ENLACE exam\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its results measure test participation, matriculation and achievement.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006466:20:0", "occurrence_id": "deff87dbb810282b0c5c6272", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: fixed location-specific effect. As for the last term ( 𝜀𝑖𝑗 + 𝜑𝑖𝑗 ), it is an error independently\ndistributed across households and has zero mean, so it does not affect the estimation.\n\n\nModels A, B, C and D are estimated as Probit regressions using data from the 61 <sup>st</sup> (2004-05) and\n68 <sup>th</sup> (2011-12) rounds of the NSS employment and unemployment surveys. Observations are at\nthe individual level, but some variables (those indexed by “ _j_ ” as opposed to “ _ij_ ”) need to be\nconstructed at a more aggregate level.\n\n\nThe potential misclassification of urban areas as rural ( 𝐺𝑗 ) is measured at the district level, as the\ndifference between the true share of the urban population in the district the household lives in\nand the urban share estimated based on the NSS. The true urban share is computed on the basis\nof Census data for 2001 and 2011, treating all villages with more than 5,000 inhabitants as urban,\nto take into account the fact that the Census uses a rather restrictive definition of urban areas\ncompared to standard international practice.\n\n\nTo capture the diversity of job opportunities in the local labor market, the employment variable\n( 𝐿𝑗 ) is computed at the substratum level (below district). An innovation of this paper is the\napproach used to construct indicators below the district level and at the level of the village or\ntown, along six ranks in the rural-urban gradation.\n\n\nBoth the 61 <sup>st</sup> and the 68 <sup>th</sup> rounds of the NSS use the Census 2001 as the sampling frame for rural\nareas. <sup>5</sup> The rural sample is representative at the sampling substratum (below district) level.\nLinking the Census to the\n\nMention: \"NSS employment and unemployment surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Two rounds at the individual level supply the observations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006466:20:0", "occurrence_id": "dabb2170c4df41dff9c2556c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: fixed location-specific effect. As for the last term ( 𝜀𝑖𝑗 + 𝜑𝑖𝑗 ), it is an error independently\ndistributed across households and has zero mean, so it does not affect the estimation.\n\n\nModels A, B, C and D are estimated as Probit regressions using data from the 61 <sup>st</sup> (2004-05) and\n68 <sup>th</sup> (2011-12) rounds of the NSS employment and unemployment surveys. Observations are at\nthe individual level, but some variables (those indexed by “ _j_ ” as opposed to “ _ij_ ”) need to be\nconstructed at a more aggregate level.\n\n\nThe potential misclassification of urban areas as rural ( 𝐺𝑗 ) is measured at the district level, as the\ndifference between the true share of the urban population in the district the household lives in\nand the urban share estimated based on the NSS. The true urban share is computed on the basis\nof Census data for 2001 and 2011, treating all villages with more than 5,000 inhabitants as urban,\nto take into account the fact that the Census uses a rather restrictive definition of urban areas\ncompared to standard international practice.\n\n\nTo capture the diversity of job opportunities in the local labor market, the employment variable\n( 𝐿𝑗 ) is computed at the substratum level (below district). An innovation of this paper is the\napproach used to construct indicators below the district level and at the level of the village or\ntown, along six ranks in the rural-urban gradation.\n\n\nBoth the 61 <sup>st</sup> and the 68 <sup>th</sup> rounds of the NSS use the Census 2001 as the sampling frame for rural\nareas. <sup>5</sup> The rural sample is representative at the sampling substratum (below district) level.\nLinking the Census to the\n\nMention: \"NSS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The urban share variable is estimated from this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006466:20:0", "occurrence_id": "3389d6e776bcb0c429d56a57", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: fixed location-specific effect. As for the last term ( 𝜀𝑖𝑗 + 𝜑𝑖𝑗 ), it is an error independently\ndistributed across households and has zero mean, so it does not affect the estimation.\n\n\nModels A, B, C and D are estimated as Probit regressions using data from the 61 <sup>st</sup> (2004-05) and\n68 <sup>th</sup> (2011-12) rounds of the NSS employment and unemployment surveys. Observations are at\nthe individual level, but some variables (those indexed by “ _j_ ” as opposed to “ _ij_ ”) need to be\nconstructed at a more aggregate level.\n\n\nThe potential misclassification of urban areas as rural ( 𝐺𝑗 ) is measured at the district level, as the\ndifference between the true share of the urban population in the district the household lives in\nand the urban share estimated based on the NSS. The true urban share is computed on the basis\nof Census data for 2001 and 2011, treating all villages with more than 5,000 inhabitants as urban,\nto take into account the fact that the Census uses a rather restrictive definition of urban areas\ncompared to standard international practice.\n\n\nTo capture the diversity of job opportunities in the local labor market, the employment variable\n( 𝐿𝑗 ) is computed at the substratum level (below district). An innovation of this paper is the\napproach used to construct indicators below the district level and at the level of the village or\ntown, along six ranks in the rural-urban gradation.\n\n\nBoth the 61 <sup>st</sup> and the 68 <sup>th</sup> rounds of the NSS use the Census 2001 as the sampling frame for rural\nareas. <sup>5</sup> The rural sample is representative at the sampling substratum (below district) level.\nLinking the Census to the\n\nMention: \"Census 2001\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It serves as the rural sampling frame for both survey rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006479:11:0", "occurrence_id": "bd0e0ed18f79a3f3c8dfd20d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ###### **2.4 Auxiliary Data Sources**\n\nWe combine the main sample described in the previous section with four administrative data\n\n\nsources. First, the biannual Mexican school census which collects information on school-level\n\n\nyearly trajectories - e.g. enrollment, failure and drop-out rate, school infrastructure, teachers’ and\n\n\nschool principals’ main characteristics as well as detailed information on the attendance fees that\n\n\nstudents have to incur upon registration in a given school: tools, uniforms, monthly payments, reg\n\nistration and tuition. This information can be conveniently linked with the full list of COMIPEMS\n\n\nschooling options through a unique school identifier ( _Claves de Centros de Trabajo_ - CCT) during\n\n\nthe period 2005-2010. In addition, geocoded information allows us to compute geodesic distances\n\n\nbetween students’ places of residence and the high school programs offered by the COMIPEMS\n\n\nsystem.\n\n\nSecond, we employ a sub-module of the third round of the 2008 national employment survey\n\n\n(ENOE) which records retrospective information on the academic trajectories of individuals in the\n\n\nage group 15-34, who passed at least one year of upper secondary education, including the track\n\n\nattended in high school and whether college education was attended or not. Wage and employment\n\n\ninformation comes from the ENOE that is considered the most reliable source of wage data for\n\n\nMexico. This survey is representative at the State-level and further allows to distinguish between\n\n\nrural, semi-urban and urban areas depending on the population size of the communities of residence\n\n\nof the surveyed individuals. We thus consider as the relevant labor market for the COMIPEMS\n\n\napplicants in the city of Ecatepec the urban sections of the State of Mexico and the Federal District.\n\n\nThird, we use data from a 2009 survey collected on a nationally-representative sample of stu\n\ndents who take a nationally standardized end of high school test (ENLACE). The so called _EN-_\n\nMention: \"Mexican school census\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Biannual records of school-level enrolment and failure trajectories.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006479:11:0", "occurrence_id": "5cd29139dad9a15c432ed8dc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ###### **2.4 Auxiliary Data Sources**\n\nWe combine the main sample described in the previous section with four administrative data\n\n\nsources. First, the biannual Mexican school census which collects information on school-level\n\n\nyearly trajectories - e.g. enrollment, failure and drop-out rate, school infrastructure, teachers’ and\n\n\nschool principals’ main characteristics as well as detailed information on the attendance fees that\n\n\nstudents have to incur upon registration in a given school: tools, uniforms, monthly payments, reg\n\nistration and tuition. This information can be conveniently linked with the full list of COMIPEMS\n\n\nschooling options through a unique school identifier ( _Claves de Centros de Trabajo_ - CCT) during\n\n\nthe period 2005-2010. In addition, geocoded information allows us to compute geodesic distances\n\n\nbetween students’ places of residence and the high school programs offered by the COMIPEMS\n\n\nsystem.\n\n\nSecond, we employ a sub-module of the third round of the 2008 national employment survey\n\n\n(ENOE) which records retrospective information on the academic trajectories of individuals in the\n\n\nage group 15-34, who passed at least one year of upper secondary education, including the track\n\n\nattended in high school and whether college education was attended or not. Wage and employment\n\n\ninformation comes from the ENOE that is considered the most reliable source of wage data for\n\n\nMexico. This survey is representative at the State-level and further allows to distinguish between\n\n\nrural, semi-urban and urban areas depending on the population size of the communities of residence\n\n\nof the surveyed individuals. We thus consider as the relevant labor market for the COMIPEMS\n\n\napplicants in the city of Ecatepec the urban sections of the State of Mexico and the Federal District.\n\n\nThird, we use data from a 2009 survey collected on a nationally-representative sample of stu\n\ndents who take a nationally standardized end of high school test (ENLACE). The so called _EN-_\n\nMention: \"2008 national employment survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its third-round sub-module records retrospective academic trajectories.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006479:11:0", "occurrence_id": "3b50734777afd49504affa86", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ###### **2.4 Auxiliary Data Sources**\n\nWe combine the main sample described in the previous section with four administrative data\n\n\nsources. First, the biannual Mexican school census which collects information on school-level\n\n\nyearly trajectories - e.g. enrollment, failure and drop-out rate, school infrastructure, teachers’ and\n\n\nschool principals’ main characteristics as well as detailed information on the attendance fees that\n\n\nstudents have to incur upon registration in a given school: tools, uniforms, monthly payments, reg\n\nistration and tuition. This information can be conveniently linked with the full list of COMIPEMS\n\n\nschooling options through a unique school identifier ( _Claves de Centros de Trabajo_ - CCT) during\n\n\nthe period 2005-2010. In addition, geocoded information allows us to compute geodesic distances\n\n\nbetween students’ places of residence and the high school programs offered by the COMIPEMS\n\n\nsystem.\n\n\nSecond, we employ a sub-module of the third round of the 2008 national employment survey\n\n\n(ENOE) which records retrospective information on the academic trajectories of individuals in the\n\n\nage group 15-34, who passed at least one year of upper secondary education, including the track\n\n\nattended in high school and whether college education was attended or not. Wage and employment\n\n\ninformation comes from the ENOE that is considered the most reliable source of wage data for\n\n\nMexico. This survey is representative at the State-level and further allows to distinguish between\n\n\nrural, semi-urban and urban areas depending on the population size of the communities of residence\n\n\nof the surveyed individuals. We thus consider as the relevant labor market for the COMIPEMS\n\n\napplicants in the city of Ecatepec the urban sections of the State of Mexico and the Federal District.\n\n\nThird, we use data from a 2009 survey collected on a nationally-representative sample of stu\n\ndents who take a nationally standardized end of high school test (ENLACE). The so called _EN-_\n\nMention: \"ENOE\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Considered the most reliable wage source, repeated as a panel.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006482:25:1", "occurrence_id": "f7175b242f2e55f0511acca1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: s budget by discretionary and Rigidity**\n**shares**\n<u>(percent)</u>\n\n**Paraguay (shares of the budget)**\nDiscretionary Expenditure Rigid Expenditures\n\n\n100%\n\n\n\n80%\n\n\n60%\n\n\n40%\n\n\n20%\n\n\n0%\n\n\n\n2003 2004 2005 2006 2007 2008 2009 2010 2011 2012\n\n\n\n\n\n\n\n80%\n\n\n60%\n\n\n40%\n\n\n20%\n\n\n0%\n\n\n\n2006 2007 2008 2009 2010 2011 2012\n\n\n\nSource: BOOST dataset, Force Toolkit and authors’ calculations Source: BOOST dataset, Force Toolkit and authors’ calculations\n\nMention: \"BOOST dataset\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Annual shares across these years are calculated with this benefits dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006483:5:0", "occurrence_id": "d818cac2335d36be55227c78", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: factor outputs. For example, looking mostly at U.S. public capital stocks, the macro‐level literature finds\nvery large estimates for the elasticity of infrastructure—between 0.20 and 0.40.\n\nConnective infrastructure, such as roads, is likely to have the greatest impact on economic growth.\nImproved connectivity within a country and within a region increases firms’ access to markets for their\ngoods and for cheaper inputs; assists businesses in developing competitive advantages; and provides\nworkers, particularly those in lagging areas, with greater job and income opportunities. Transport typically\nfacilitates agglomerations, which in turn promotes economic growth. For example, Njoh (2009) finds a\nstrong positive relationship between transportation infrastructure and growth based on evidence from\n24 West African countries. As countries develop, economic growth becomes more closely linked to a\n“principle of connectivity” (Biller and Nabi 2013).\n\nOther capital intensive infrastructures such as power are also closely related to economic growth via\nconnectivity. While not directly linked to connectivity (in terms of moving people, goods, and services),\npower facilitates information flows, thereby enabling agglomerations. More importantly, power is a key\ninput in the production process, and power constraints have been identified as a major bottleneck for\ngrowth and job creation in several developing countries. A good example can be found in Nepal. Over the\npast few years, Nepal has faced up to 16 hours of electricity load shedding in dry months – stalling the\nmanufacturing sector and severely curtailing basic services (like health care), other services, and\ncommerce. The economic loss of such a huge power outage must be tremendous. In transport, Nepal has\nsimilar constraints, being a mountainous landlocked country. Yet while geography may be more benign in\nother South Asian countries like Pakistan and India, connective infrastructure is also severely lacking and\nthe existing ones are poorly used. <sup>6</sup>\n\n\n\n\n\n\n\n\n\n_Source:_ World Development Indicators 2012 Online Database, the World Bank\n\nMention: \"World Development Indicators 2012 Online Database, the World Bank\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its edition supplies the infrastructure statistics behind the claim.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006490:14:1", "occurrence_id": "08dfed9f0d21d3d9e34f1ba8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  risk preferences. The KLPS2 survey was\n\n\nlaunched in August of 2007, and 1,180 respondents were surveyed in 2007 prior to the elections.\n\n\nThe crisis led to a two-month suspension of survey activities, which resumed in March of 2008\n\n\nand continued through the end of 2009. We employ two complementary identification strategies.\n\n\nFirst, we estimate the impact of the crisis in a straightforward linear framework, controlling for\n\n\nmonth-of-the-year fixed effects. In these specifications, we also include enumerator fixed effects\n\n\nand controls for sociodemographic characteristics such as respondent age, educational attainment,\n\n\nvalue lottery as a risky alternative.\n\n\n13\n\nMention: \"KLPS2 survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Launched in that year with 1,180 respondents surveyed beforehand.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006493:6:0", "occurrence_id": "8f987cce27e39677a19b12fb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: and zero for 1998 (before the law was passed), and ��� denotes an idiosyncratic error\nterm. For this regression, we also include as control the individual´s age (in weeks). In\nour second regression, �� takes the value of 1 if individual _i_ in cohort c turned age 14 in\nthe first semester of 1999 and 0 if (s)he belongs to the cohort that turned 14 in the second\nhalf of 1998. Table 1 shows the groups under comparison in each case. Because the data\nprovides exact dates of birth, we run the regressions using a 26 week-bandwidth (6month) as can be inferred from table 1.\n\nIn both regression models, the parameter of interest, , provides the intent-totreat (ITT) effect. In other words, the impact of the ban on the cohort hindered from\nparticipating in the formal labor force at age 14.\n\n\n**4. DATA & RESULTS**\nThe sample used in this paper is drawn from the Brazilian household surveys\n( _Pesquisa Nacional por Amostra de Domicilios_ - PNAD) of 1998 and 1999. The PNAD\nis an annual household survey that covers around 100,000 households and about 320,000\nindividuals. It is a major source of microdata in Brazil, and is a nationally representative\nsurvey that contains detailed information on each household’s socio-economic\ncharacteristics, demographic data, household income, and labor force status. The\nBrazilian Bureau of Statistics (IBGE) provides the exact reference date for when data\ncollection is carried out and we used that information to identify the treatment and\ncomparison groups. <sup>9</sup> [^9: According to IBGE, the reference date is the last week of September. There can be some variability in\nthe exact day, but it is usually September 26 or 27.] Our results are exclusive for urban areas.\n\nWe used three measures of child labor: labor force participation rates as a whole\nwithout distinguishing between sectors, participation in the formal labor force, and\n\nMention: \"PNAD\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"An annual household survey covering some 100,000 households.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006498:18:0", "occurrence_id": "0927ece7eaa6ab88160d154b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: significantly compromise the political and administrative feasibility of a VAT (Ebrill et al, 2001).\nAdditionally, WBG/IFC tax compliance cost surveys show that joining the VAT regime substantially\nincreases the tax compliance costs for small businesses. Feedback received from small businesses in\nEuropean Union member countries demonstrates that small businesses consider VAT legislation as\none of the 10 most burdensome EU laws. The compliance challenge is thus increased when VAT\nsystems require extensive documentation, where taxpayers are subject to frequent VAT audits, or\nwhen filing procedures have not been streamlined and simplified. This is the case in a number of ECA\ncountries, as comparative analysis indicates that the time required for VAT compliance in the region\nsubstantially exceeds compliance time in EU countries.\n\n\n**Figure 4: VAT Compliance Time across Regions**\n\n\n\n\n\n\n\n\n\n\n\n\n\n_Source:_ PwC, The Impact of VAT Compliance on Business 2010.\n\n\nA more detailed compliance cost analysis for VAT was conducted in Slovenia in 2001, two years after\nVAT was introduced with a relatively low threshold of SIT 5 million ($22,700), demonstrating the high\ncompliance burden for small businesses.\n\n\n**Table 6: VAT compliance costs as a share of business turnover in Slovenia**\n\n|Taxpayer size(turnover)|Compliance costs as<br>percentage of turnover|\n|---|---|\n|Up toSIT100 million|3.73%|\n|SIT100 million–1 billion|0.73%|\n|AboveSIT1 billion|0.08%|\n\n\n\n_Source:_ Klun, 2003.\n\n\nThe risk to small business in complying with VAT requirements increases further in an environment of\nlow administrative efficiency. Non‐payment of VAT refunds or delays in the processing of refund\nrequests can severely affect the liquidity of the business. Liquidity problems can also arise in case of\n\n\n17\n\nMention: \"WBG/IFC tax compliance cost surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"They show value-added tax registration raises small firm compliance costs.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006500:11:0", "occurrence_id": "4000281446f6a823e3f56372", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: countries with fewer distortions (richer countries) should have more exporters but the implications\n\n\nfor average exporter size and concentration at the top of the size distribution vary, depending on\n\n\nwhich firms are more constrained. If taxes and regulations (e.g., on the labor market and on\n\n\nbusiness entry) become heavy as firms grow larger, only the most productive can overcome these\n\n\ncosts and there would be a “missing middle”: many small firms and a few large firms but few\n\n\nmedium-sized firms. Consistent with a “missing middle”, Tybout (2000) shows a U-shaped firm\n\nsize histogram for a number of developing countries, when firms are allocated to relatively wide\n\n\nsize bins. <sup>10</sup> Tybout (2014) shows that the “missing middle” does not necessarily imply a bimodal\n\n\nfirm-size distribution. Assuming that the undistorted firm-size distribution should be an exact\n\n\nPareto distribution, he compares the actual distribution with an estimated country-specific Pareto\n\n\ndistribution. Using data for a number of Asian countries, the U.S., and Mexico, he finds that the\n\n\nactual firm-size distribution is wider than the estimated Pareto distribution, as represented in Panel\n\n\nA of Figure 1, particularly in developing countries. His results, however, are dependent on\n\n\nallowing the Pareto dispersion parameter to vary across countries, and he estimates a lower Pareto\n\n\nindex (a more skewed firm-size distribution) in higher-income countries. In related work, Alfaro\n\n\net al. (2009) use Dun and Bradstreet’s WorldBase data for 79 countries and find that establishment\n\n\nsize is decreasing in GDP per capita, and interpret this as being consistent with too few medium\n\nsized firms in lower-income countries. <sup>11</sup>\n\n\nIn contrast, Hsieh and Klenow (2009) calculate a hypothetical efficient firm-size\n\n\ndistribution, assuming the marginal products of capital and labor are equalized across all\n\nMention: \"WorldBase data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Establishment size across 79 countries is measured in these data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006502:22:0", "occurrence_id": "f8d00c53cba33abddc7bfc6b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: that informal financing almost always carries associated informal costs (i.e., costs not reflected\n\n\nin interest rates), such as informal gifts and implied obligations to assist lenders in the future,\n\n\nmaking them relatively less attractive to borrowers with formal options. Another reason for\n\n\ntheir greater reliance on formal loans is that better-educated people tend to have fewer siblings\n\n\nand a higher opportunity cost of their time, which would raise their relative costs of using\n\n\ninformal financing. They also have higher-paying jobs in general, and likely steadier incomes,\n\n\nmaking them better candidates for bank financing. The overall effect is therefore that better\n\neducated residents rely more on formal financing.\n\n\nContrasts in the configurations and effectiveness of various social networks on access to\n\n\nfinance shed light on how finance works for Chinese households. Rural households with an\n\n\naffiliation to the Communist Party are 10.2 percentage points more likely to have a bank loan\n\n\n(model 6). For urban households, incidence of affiliation with the Communist Party is relatively\n\n\nhigh (18%, in contrast to 9.5% in rural areas), <sup>20</sup> and party affiliation is not associated with a\n\n\ngreater likelihood of having a bank loan.\n\n\nWhile party membership is associated with greater use of bank loans in rural areas, it is\n\n\nnot associated with more usage of informal loans. These pieces of evidence suggest that\n\n\naffiliation with political power can affect access to formal credit in some rural Chinese contexts,\n\n\nperhaps because state-owned banks have greater influence over credit markets in rural areas.\n\n\nThe number of siblings, a proxy for the extent of private social networks, also influences\n\n\ninformal financing. Each additional sibling is associated with 4.7 and 3.9 percentage point\n\n\ngains in the likelihood of informal borrowing in urban and rural areas, respectively (models 8\n\n\nand 9). Thus, private social networks also play an important role in facilitating informal credit\n\n\nmarkets.\n\n\n20 Figures based on CHFS data.\n\n\n21\n\nMention: \"CHFS data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The informal credit figures rest on this household survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006506:8:0", "occurrence_id": "498b5ba90add8445044941e5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: midyear population according to the World Bank’s World Development Indicators 2011. Costs are\nrecorded exclusive of value added tax and only include the fees associated with completing the\nprocedures to connect a warehouse to electricity. This includes the costs in related to obtaining\nclearances from government agencies, applying for the connection, receiving inspections of both\nthe site and the internal wiring, purchasing material, getting the actual connection works and\npaying a security deposit <sup>7</sup> [^7: Connection costs can be broadly divided into the following categories: often utilities charge a fixed connection fee\nthat spreads the fixed costs of operating a distribution grid over all customers; in addition, customers are required to\npay for the variable costs for each connection, including labor and material costs. Sometimes the costs also include\nfees for inspections that the utility or outside contractors are legally required to carry out. In many economies costs\nalso include payments to other agencies for permits, inspections and approvals.] . Costs associated with electricity consumption are not included.\nInformation from local experts and specific regulations and fee schedules are used as sources for\ncosts. If several local partners provide different estimates, the median reported value is used. In all\ncases, the cost excludes bribes.\n\n\nCost estimates for 86 economies also include the present value of the lost interest earnings on the\nsecurity deposits that utilities require as a guarantee against the possible failure of customers to\npay their consumption bills. Instead of recording the full amount of the guarantee, we record the\npresent value of the losses in interest earnings experienced by the customer because the utility\nholds the security deposit over a prolonged period. In most cases this is until the end of the contract\n(assumed to be after 5 years) <sup>8</sup> [^8: In the few cases in which the security deposit is used to cover the first monthly consumption bills, it is not recorded\nat all.] . To calculate the present value of the lost interest earnings, the end2010 lending\n\nMention: \"World Development Indicators 2011\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Midyear population denominators come from this edition.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006516:43:0", "occurrence_id": "5af9d2a7378f01570c5a01bf", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 42\n\n\n‘achievement’ index? One possibility is to conduct an exit survey of users of health\n\n\nfacilities, the aim being to gather information not only on the health services\n\n\nreceived and the need for them, but also on the living standards of the person’s\n\n\nhousehold; the latter data can be used to construct a wealth index using the same\n\n\nscoring system as in a national survey (like the DHS), so that users can be placed in\n\n\nthe national wealth distribution. As Wagstaff and Waters (2005) report, several\n\n\nstudies in the ‘Reaching the Poor’ volume (Gwatkin et al. 2005) used such an\n\n\napproach. The increased use of electronic personal identification systems to target\n\n\npublic resources on the poor and vulnerable, such as the smart card used in India’s\n\n\naforementioned RSBY program, provide another potential source of information on\n\n\nthe living standards of a user’s household; sometimes this may only indicate\n\n\nwhether the household is above or below a poverty line, but over time more\n\n\ngranular data may be recorded in the system, such as the household’s score on\n\n\nwhatever means test (or proxy means test) was used to determine the position vis-à\n\nvis the poverty line or eligibility for a subsidy.\n\n\nAn additional area for future work is using the index to shed light on policy. A\n\n\ncouple of approaches suggest themselves. One is to use the index to quantify the\n\n\nimpacts of reforms on UHC. Reforms and programs often focus on specific\n\n\nchallenges, sometimes those deemed most important by the international\n\n\ndevelopment community, and will, because of their limited focus, likely affect only\n\n\nsome of the indicators in the UHC index. If one has good evidence on the impacts of\n\n\na reform or program on a subset of indicators, one could use these estimates to see\n\n\n42\n\nMention: \"DHS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its wealth index scoring system is replicated for the smaller survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006519:8:1", "occurrence_id": "80256b2b5806e1937c6127af", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  affiliation. We use\nthis information to construct the household characteristics of the benchmark specification. It must be noted\nthat the monthly consumption reported by the NSS includes paid rent but excludes imputed rent. This biases\ndownward the expenditure of households who own their dwelling units. To address this data limitation, we\nadd information of household dwelling ownership to the set of household characteristics in the analysis.\n\n\nAs for locations, most poverty analyses in India consider the state or the region as the spatial unit of analysis,\nfurther dividing each unit into urban and rural areas. However, that level of aggregation is too high to assess\nlocal externalities. Even data at the district level may not be disaggregated enough for that purpose.\nSeparately identifying individual cities would be difficult too, because the actual boundaries of urban\nagglomerations do not match well the administrative boundaries and classifications used by household\nsurveys (Li et al., 2015). To address these limitations, in this paper we adopt the approach developed by\nChatterjee et al. (2015) to generate estimates below district level. Their approach uses the design of the\nNSS 2011-12 to estimate the population of first-stage sampling units. The resulting characterization of the\nemployment structure across these different population sizes challenges the conventional wisdom of a ruralurban “divide” in India’s context.\n\n\nThe NSS 2011-12 covers all of India except interior villages of Nagaland situated beyond five kilometers\nof a bus route and villages in Andaman and Nicobar Islands. The survey follows a stratified multi-stage\nsampling design. Each district of a state or union territory is stratified into rural and urban areas. In the rural\nstratum, the first stage units are the 2001 census villages; in the urban stratum, they are urban frame survey\nblocks. Within each stratum, first-stage units are ordered by their population and then further stratified. The\nultimate stage units are households, drawn from the selected first stage units of each substrata.\n\n\n7\n\nMention: \"NSS 2011-12\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"Its design lets estimates be generated below district level.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006519:8:1", "occurrence_id": "3b5e394ca89e5cb812c3a7dc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  affiliation. We use\nthis information to construct the household characteristics of the benchmark specification. It must be noted\nthat the monthly consumption reported by the NSS includes paid rent but excludes imputed rent. This biases\ndownward the expenditure of households who own their dwelling units. To address this data limitation, we\nadd information of household dwelling ownership to the set of household characteristics in the analysis.\n\n\nAs for locations, most poverty analyses in India consider the state or the region as the spatial unit of analysis,\nfurther dividing each unit into urban and rural areas. However, that level of aggregation is too high to assess\nlocal externalities. Even data at the district level may not be disaggregated enough for that purpose.\nSeparately identifying individual cities would be difficult too, because the actual boundaries of urban\nagglomerations do not match well the administrative boundaries and classifications used by household\nsurveys (Li et al., 2015). To address these limitations, in this paper we adopt the approach developed by\nChatterjee et al. (2015) to generate estimates below district level. Their approach uses the design of the\nNSS 2011-12 to estimate the population of first-stage sampling units. The resulting characterization of the\nemployment structure across these different population sizes challenges the conventional wisdom of a ruralurban “divide” in India’s context.\n\n\nThe NSS 2011-12 covers all of India except interior villages of Nagaland situated beyond five kilometers\nof a bus route and villages in Andaman and Nicobar Islands. The survey follows a stratified multi-stage\nsampling design. Each district of a state or union territory is stratified into rural and urban areas. In the rural\nstratum, the first stage units are the 2001 census villages; in the urban stratum, they are urban frame survey\nblocks. Within each stratum, first-stage units are ordered by their population and then further stratified. The\nultimate stage units are households, drawn from the selected first stage units of each substrata.\n\n\n7\n\nMention: \"2001 census villages\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"They form the first-stage units in the rural stratum.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006519:8:1", "occurrence_id": "f25ecbe04fd3428d043b046b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  affiliation. We use\nthis information to construct the household characteristics of the benchmark specification. It must be noted\nthat the monthly consumption reported by the NSS includes paid rent but excludes imputed rent. This biases\ndownward the expenditure of households who own their dwelling units. To address this data limitation, we\nadd information of household dwelling ownership to the set of household characteristics in the analysis.\n\n\nAs for locations, most poverty analyses in India consider the state or the region as the spatial unit of analysis,\nfurther dividing each unit into urban and rural areas. However, that level of aggregation is too high to assess\nlocal externalities. Even data at the district level may not be disaggregated enough for that purpose.\nSeparately identifying individual cities would be difficult too, because the actual boundaries of urban\nagglomerations do not match well the administrative boundaries and classifications used by household\nsurveys (Li et al., 2015). To address these limitations, in this paper we adopt the approach developed by\nChatterjee et al. (2015) to generate estimates below district level. Their approach uses the design of the\nNSS 2011-12 to estimate the population of first-stage sampling units. The resulting characterization of the\nemployment structure across these different population sizes challenges the conventional wisdom of a ruralurban “divide” in India’s context.\n\n\nThe NSS 2011-12 covers all of India except interior villages of Nagaland situated beyond five kilometers\nof a bus route and villages in Andaman and Nicobar Islands. The survey follows a stratified multi-stage\nsampling design. Each district of a state or union territory is stratified into rural and urban areas. In the rural\nstratum, the first stage units are the 2001 census villages; in the urban stratum, they are urban frame survey\nblocks. Within each stratum, first-stage units are ordered by their population and then further stratified. The\nultimate stage units are households, drawn from the selected first stage units of each substrata.\n\n\n7\n\nMention: \"urban frame survey\nblocks\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"They form the first-stage units in the urban stratum.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006520:21:0", "occurrence_id": "41fdce675e0c2adb8f4f98e2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Next, we also see how income diversity differ across the poverty categories. High income\ndiversity is generally assumed to make households less vulnerable to shocks, as they have not “put\nall the eggs in one basket”. We summarize this in the following research question: _Do the income_\n_and/or asset poor households have higher environmental reliance and higher income diversity?_\n\nThe overall pattern found in the literature is that the income poor are more reliant on\nenvironmental income in a relative sense, but the non-poor are the most reliant in an absolute sense.\nThis is confirmed by Table 7 (income shares) and Table A7 (absolute income). The _income & asset poor_\nhave the highest environmental income share; 29% of their income is derived from natural\nenvironments compared to 25% for the _income & asset rich_ . <sup>18</sup> [^18: These are the average of the household level income shares, and are different from the shares obtained when first\ncalculating average absolute income (top panel of table) and then calculate income shares.] As also noted in another analysis of the\nPEN data, the differences in income shares across income levels is less pronounced compared to\nwhat several earlier studies suggested (Angelsen et al. 2014).\n\nMost of the environmental income is derived from forested areas (20%) rather than nonforested areas (7%). Among forest income, the three dominant categories are food (30.3%),\nwoodfuel (35.2%) and structural and fiber products (24.9%). For the non-forest environmental\nincomes, the share is higher for food (48.9%) and lower for woodfuel (20.6%) and structural and\nfiber products (9.9%).\n\nCrop income is the single most important income source in all groups (30-31%). <sup>19</sup> The\nlivestock income share is lower for the _income & asset poor_ (11%) compared to\n\nMention: \"PEN data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"A companion analysis notes the flatter income share gradient in these data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006521:13:2", "occurrence_id": "f31d321e9160cda12a2d7da6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ** **3.4** **0.7** **20.9%** **0.2%** **0.6%**\n**World** **408.9** **235.7** **57.7%** **343.3** **206.3** **60.1%** **24.9%** **21.8%**\n\n\n\n**<u>(%)</u>**\n\n\n\n**<u>(%)</u>**\n\n\n\n**<u>Mortality</u>**\n\n\n\nCoastal populations are those living within 100 kilometers (km) of the coast, and excluding countries without\nterritory within 100 km of a coastline measured using an Azimuthal Equidistant (world) projection.\n\n\nHigh infant mortality is 50 or more deaths within the first year of life per 1,000 births (or 500 deaths/10,000 live\nbirths).\n\n\nChild malnutrition counts are the product of malnutrition rates and population counts. Malnutrition rates are\nbased on the percentage of children with a weight‐for‐age z‐score more than two standard deviations below the\nmedian of the NCHS/CDC/WHO International Reference Population.\n\n\nSee the appendix for details on the spatial methods in calculating these coastal population distributions.\n\n\nLow‐income economies are those in which 2013 per capita income was $1,045 or less, lower‐middle‐income\neconomies are those in which 2013 per capita income was between $1,046 and $4,125, upper‐middle‐income\neconomies are those in which 2013 per capita income was between $4,126 and $12,745, and developed countries\nare those in which 2013 per capita income was more than $12,745, as defined by the World Development\nIndicators\n\nMention: \"NCHS/CDC/WHO International Reference Population\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its median defines the standard deviations used for the stunting cutoff.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006532:4:0", "occurrence_id": "260ab0947089a8e808f0a7cc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 1: Urbanization, Productivity, and Economic Growth\n\n\n(a) Urbanization and per capita GDP\n\n\nSource: World Bank. 189 countries. The urbanization rate on the horizontal axis is the percentage share\nof population living in cities in 2012. The vertical axis represents the natural log of GDP per capita in\n2011 U.S. dollar (reproduced from Duranton, 2014b).\n\n\n(b) Urbanization and city-level productivity\n\n\nSource: Global Urban Indicators (1998).\n\n\n3\n\nMention: \"Global Urban Indicators (1998)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Urbanisation and city-level productivity are plotted from this source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006533:11:0", "occurrence_id": "804f1b268600d829965209cc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: simply driven by a changing composition of characteristics of routine and non-routine workers,\n\n\nwe estimate the following set of cross-sectional wage regressions separately for each year, _t_ :\n\n\nln _wi,t_ = _β_ 0 _,t_ + _β_ 1 _,tNR i,t_ + _β_ 2 _,tXi,t_ + _ϵi,t_ for _t ∈{_ 1968 _, . . .,_ 2013 _}_, (1)\n\n\nwhere _NR i,t_ is a dummy variable indicating that individual _i_ works a non-routine job in year _t_ and\n\n\n_Xi,t_ includes a variety of control variables. In particular, we include gender, race, and full time\n\n\nemployment dummies, we control for the weeks worked, a full set of industry fixed effects (50\n\n\nindustries constructed from SIC industry codes by the NBER), as well as fourth order polynomials\n\n\nin age, education, and the interaction of education and age.\n\n\nWe estimate regressions (1) based on individual level data from the annual CPS march sup\nplements and weight by the CPS sampling weights. <sup>13</sup> Panel B of Figure 3 plots the resulting time\n\n\nseries of estimates _β_ <sup>ˆ</sup> 1 _,q_ and the associated 95% confidence intervals based on standard errors that\n\n\nare clustered on industry. These estimates highlight that the rising relative wage for non-routine\n\n\nlabor is not entirely driven by the rising skill premium or by specific industries. The latter obser\n\nvation is of particular importance, as it highlights that the wage premium is not simply due the\n\n\nsteady decline in manufacturing as the estimates _β_ <sup>ˆ</sup> 1 _,q_ are identified from within industry varia\n\ntion. These estimates therefore suggest that part of the\n\nMention: \"CPS march\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its annual supplements with sampling weights carry the individual regressions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006534:4:0", "occurrence_id": "022a515c2fe29b5ac6751ce5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: in 2010—a deep drop in the context of improvements observed in socioeconomic statistics and\n\ngrowth in per capita incomes (see figure 1b).\n\n\n**Figure 1a: GDP per Capita and Satisfaction with Life, 2008-10**\n\n\n8\n\n\n7.5\n\n\n\n7\n\n\n6.5\n\n\n6\n\n\n5.5\n\n\n5\n\n\n4.5\n\n\n4\n\n\n3.5\n\n\n3\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n|ARE<br>QAT<br>SAU<br>BHR<br>JOR<br>DZA<br>IRNTUN LBY<br>DJI LBN<br>EGYIRQ SYR<br>MAR<br>YEMWBG|Col2|Col3|Col4|Col5|Col6|Col7|\n|---|---|---|---|---|---|---|\n||||||||\n\n\n\n5 6 7 8 9 10 11 12\n\n\n**Log GDP per Capita (2005 US dollars)**\n\n\n**Figure 1b: Percentage Growth in GDP and Change in Satisfaction with Life (Weighted**\n**Averages) in 106 Countries, 2005-10**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n_Sources:_ Real GDP per Capita: World Bank Development Indicators; Life Satisfaction: World Database of Happiness.\n_Note:_ Numbers are weighted averages for 147 countries. Abbreviations: ARE=United Arab Emirates; EGY=Egypt;\nJOR=Jordan; LBN=Lebanon; MAR=Morocco; SAU=Saudi Arabia; WBG=West Bank and Gaza; YEM=the\nRepublic of Yemen.\n\n\n3\n\nMention: \"World Bank Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Real output per head comes from this compilation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006534:4:0", "occurrence_id": "8cdc073f56ce89977fb2ccf6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: in 2010—a deep drop in the context of improvements observed in socioeconomic statistics and\n\ngrowth in per capita incomes (see figure 1b).\n\n\n**Figure 1a: GDP per Capita and Satisfaction with Life, 2008-10**\n\n\n8\n\n\n7.5\n\n\n\n7\n\n\n6.5\n\n\n6\n\n\n5.5\n\n\n5\n\n\n4.5\n\n\n4\n\n\n3.5\n\n\n3\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n|ARE<br>QAT<br>SAU<br>BHR<br>JOR<br>DZA<br>IRNTUN LBY<br>DJI LBN<br>EGYIRQ SYR<br>MAR<br>YEMWBG|Col2|Col3|Col4|Col5|Col6|Col7|\n|---|---|---|---|---|---|---|\n||||||||\n\n\n\n5 6 7 8 9 10 11 12\n\n\n**Log GDP per Capita (2005 US dollars)**\n\n\n**Figure 1b: Percentage Growth in GDP and Change in Satisfaction with Life (Weighted**\n**Averages) in 106 Countries, 2005-10**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n_Sources:_ Real GDP per Capita: World Bank Development Indicators; Life Satisfaction: World Database of Happiness.\n_Note:_ Numbers are weighted averages for 147 countries. Abbreviations: ARE=United Arab Emirates; EGY=Egypt;\nJOR=Jordan; LBN=Lebanon; MAR=Morocco; SAU=Saudi Arabia; WBG=West Bank and Gaza; YEM=the\nRepublic of Yemen.\n\n\n3\n\nMention: \"World Database of Happiness\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Life satisfaction averages across 147 countries come from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006537:17:0", "occurrence_id": "84b6b46845d0f4b6a01c7a13", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: statistics (p. 110). This suggests that small and medium size …rms are routinely omitted\n\n\nfrom the IO table samples, and that the industry import intensities inferred are often based\n\n\non data of mostly large …rms. For China, according to the National Input-Output Survey\n\n\nMethods of China (2007) published by the National Bureau of Statistics of the People’s\n\n\nRepublic of China, the sample used to construct the IO tables consists of all large …rms\n\n\nthat have at least 300 million yuan in revenue (about 38 million USD during the sample\n\n\nperiod), along with some small- and medium-sized …rms sampled with unknown proportions\n\n\n(see item 5 on p. 3 about sample selection and item 4 on p. 27 about size cuto¤s). In other\n\n\nwords, the sampling method behind the construction of Chinese IO tables is heavily biased\n\n\ntowards the very large …rms.\n\n\nSecond, recent research by Amiti, Itskhoki and Konings (2014) and Blaum, Lelarge and\n\n\nPeters (2014) shows that large …rms tend to have a higher import-to-sales ratio. This is also\n\n\ncon…rmed by our sample of Chinese …rms. When we regress a …rm’s import intensity on\n\n\n…rm size (measured by log(sales)), controlling for industry-year …xed e¤ects, we …nd that\n\n\ndoubling …rm sales is associated with a 0.5 percentage-point increase in the …rm’s import\n\n\nintensity (signi…cant at the 1% level). <sup>29</sup> Given that …rms’import intensity and DVAR are\n\n\nnegatively correlated, by omitting the smaller …rms, IO tables by construction tend to include\n\n\n…rms with a lower DVAR. Such\n\nMention: \"National Input-Output Survey\n\n\nMethods of China\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its published methods define how the input-output coefficients are derived.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006539:32:1", "occurrence_id": "ab7f43abcf097ae88d135b7d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \nNo discussion on gross job\ndestruction and firm size.\n\n\n\nNo discussion on\nfirm exit rates.\n\n\nNo discussion\non firm exit rates.\n\n\nNo discussion on\nfirm exit rates.\n\n\nSmall firms have\na higher exit\nrates.\n\n\n\nPage and Soderbom\n(2012); various\ncountries.\n\n\n\nWorld Bank Enterprise\nSurveys, and Manufacturing\ncensus data from Ethiopia.\n\n\n\nNo explicit definition\nof SMEs ;firms with\n100+ employees are\ndefined as large; those\nwith fewer are\ncategorized as small\n\n\nNo explicit definition\nof SMEs; various size\ncutoffs provided.\n\n\nNo explicit definition\nof firm size; various\nsize cutoffs provided\nusing firm sales.\n\n\nNo explicit definition\nof SMEs.\n\nMention: \"World Bank Enterprise\nSurveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Combined with one national manufacturing census for the exit rate evidence.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006540:8:0", "occurrence_id": "6a7da01202fa82789d3e996f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: regressions of growth on each of the five individual quintile shares, including them one at a time in their\n\n\ngrowth regressions. They find that higher bottom quintile shares are positively correlated with\n\n\nsubsequent growth, while higher upper quintile shares are negatively correlated with subsequent\n\n\ngrowth. This latter finding is not very surprising given that top (bottom) quintile shares can be thought\n\n\nof as alternative summary measures of inequality that are almost perfectly positively (negatively)\n\n\ncorrelated with the Gini coefficient in their data.\n\n\nFinally, Castelló‐Climent (2010) (hereafter, CC) also documents the effects of inequality on growth in\n\n\na large panel of countries. This paper considers inequality in the distribution of human capital, in\n\n\naddition to the income or consumption inequality measures that are studied in the other three papers\n\n\nreviewed here. Human capital inequality is constructed using the methodology proposed in Castelló and\n\n\nDomenech (2002), which relies on the Barro‐Lee cross‐country data on years of schooling by age groups\n\n\nof the population, and then calculates inequality in education across age groups as the measure of\n\n\nhuman capital inequality. Although this measure captures only human capital inequality across (and not\n\n\nwithin) age groups, an advantage of this measure is that it has much greater country and time coverage\n\n\nthan measures of income and consumption inequality from the UNU‐Wider World Income Inequality\n\n\nDatabase. Since the strongest results in this paper are for the human capital inequality measure, I focus\n\n\non these here. CC finds a strongly significant negative relationship between human capital inequality\n\n\nand subsequent growth in a large panel of countries observed over five‐year periods.\n\n\n**2.2 Estimating Growth Regressions Using the SYS‐GMM Estimator**\n\n\nAll of the studies discussed above estimate variants on the following standard cross‐country growth\n\n\nregression:\n\n\n(1) ��� �� ����� ��′����� ��� ��� ����\n\n\nwhere ��� is log real GDP per capita, �����\n\nMention: \"Barro‐Lee cross‐country data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Years of schooling by age group feed the returns-based human capital measure.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006540:8:0", "occurrence_id": "399c7544f1a45c71144fdf76", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: regressions of growth on each of the five individual quintile shares, including them one at a time in their\n\n\ngrowth regressions. They find that higher bottom quintile shares are positively correlated with\n\n\nsubsequent growth, while higher upper quintile shares are negatively correlated with subsequent\n\n\ngrowth. This latter finding is not very surprising given that top (bottom) quintile shares can be thought\n\n\nof as alternative summary measures of inequality that are almost perfectly positively (negatively)\n\n\ncorrelated with the Gini coefficient in their data.\n\n\nFinally, Castelló‐Climent (2010) (hereafter, CC) also documents the effects of inequality on growth in\n\n\na large panel of countries. This paper considers inequality in the distribution of human capital, in\n\n\naddition to the income or consumption inequality measures that are studied in the other three papers\n\n\nreviewed here. Human capital inequality is constructed using the methodology proposed in Castelló and\n\n\nDomenech (2002), which relies on the Barro‐Lee cross‐country data on years of schooling by age groups\n\n\nof the population, and then calculates inequality in education across age groups as the measure of\n\n\nhuman capital inequality. Although this measure captures only human capital inequality across (and not\n\n\nwithin) age groups, an advantage of this measure is that it has much greater country and time coverage\n\n\nthan measures of income and consumption inequality from the UNU‐Wider World Income Inequality\n\n\nDatabase. Since the strongest results in this paper are for the human capital inequality measure, I focus\n\n\non these here. CC finds a strongly significant negative relationship between human capital inequality\n\n\nand subsequent growth in a large panel of countries observed over five‐year periods.\n\n\n**2.2 Estimating Growth Regressions Using the SYS‐GMM Estimator**\n\n\nAll of the studies discussed above estimate variants on the following standard cross‐country growth\n\n\nregression:\n\n\n(1) ��� �� ����� ��′����� ��� ��� ����\n\n\nwhere ��� is log real GDP per capita, �����\n\nMention: \"UNU‐Wider World Income Inequality\n\n\nDatabase\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its income and consumption inequality measures have weaker coverage.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006541:7:0", "occurrence_id": "870be6b6be1b973ef54cb9c1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: and ranks Tonga 121 <sup>st</sup> out of 190 countries. <sup>4</sup> [^4: All statistics in this section are from the 2013 _Human Development Report_ (UNDP, 2013).] Average incomes in New Zealand are about six\n\n\ntimes as high, with a per capita GNI of $24,400 in 2012. Tonga and New Zealand both rank 26\n\n\nplaces higher in the Human Development Index than they do in terms of GNI, making New\n\n\nZealand the largest positive outlier of very high human development countries and Tonga the\n\n\nsecond largest outlier of countries with medium levels of human development (UNDP, 2013,\n\n\np.27). Hence, this migration flow allows a very large positive income shock, where the higher\n\n\nincomes are not compensating for lower levels of other aspects of human development.\n\n\nEmigration out of Tonga is high, with 30,000 Tongan-born living abroad, mainly in\n\n\nNew Zealand, Australia and the United States. Despite an earlier history of employment\n\n\nmigration to New Zealand, family reunification (mostly marriage) was the main channel of\n\n\naccess in the 1990s following New Zealand’s implementation of a points-based immigration\n\n\nsystem that favors highly skilled migrants. In 2002, New Zealand introduced a new migration\n\n\nprogram, the Pacific Access Category (PAC), that lets a quota of 250 Tongans permanently\n\n\nmigrate each year. Any Tongan citizens aged 18 to 45 who meet certain English, health and\n\n\ncharacter requirements may register. A random ballot selects among applicants, with odds of\n\n\nabout 10% during the 2002-05 ballot years that our sample is drawn from. If their ballot is\n\n\nselected, applicants have six months to obtain a full-time job offer in New Zealand that meets\n\n\nan income threshold similar to the minimum wage. This ensures self-reliance since Tongans\n\n\nare not eligible for most forms of welfare until they reside in New Zealand for two years. After\n\n\na job offer is filed along with a residence application, it typically takes from three to nine\n\n\nmonths to receive residence approval and\n\nMention: \"Human Development Index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Two countries rank far higher here than on income alone.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006545:5:1", "occurrence_id": "9899482f9de6194046d5601a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  GDP by\nindustry and department (expressed in year 1994 constant new soles), and population by department.\n\n\nBecause of gaps in the regional and industry level employment estimates in SIRTOD, we complete this\ndata using the employment information – also at regional and industry levels – found in the National\nHousehold Survey (ENAHO). ENAHO is also our source for data on poverty rates by department. The\npoverty rate is the percentage of population living below a specified threshold of income.\n\n\nThe definitions of the key variables in our study are as follows:\n\n\n\n1. Value Added per worker =\n2. GDP per capita = Value Added\n3. Poverty level =\n\nHere,\n\n\n\nValue Added _i_, _j_ Total employment _i_, _j_\nValue Added _i_ Total population _i_\nPopulation below income threshold _i_ Total population _i_\n\n\n\n_i_  Department, _j_  Industry\n\n\n\n5 We discuss this in more detail in Section 5.\n6 Departments of Lima and Callao are grouped into one region (Lima‐Callao), following the classification used by INEI in the initial\nyears of the sample (early 2000s).\n\n\n4\n\nMention: \"National\nHousehold Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its employment information at regional and industry levels carries the analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006545:5:1", "occurrence_id": "ae2abf2d0b46bb7598fb0ca9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  GDP by\nindustry and department (expressed in year 1994 constant new soles), and population by department.\n\n\nBecause of gaps in the regional and industry level employment estimates in SIRTOD, we complete this\ndata using the employment information – also at regional and industry levels – found in the National\nHousehold Survey (ENAHO). ENAHO is also our source for data on poverty rates by department. The\npoverty rate is the percentage of population living below a specified threshold of income.\n\n\nThe definitions of the key variables in our study are as follows:\n\n\n\n1. Value Added per worker =\n2. GDP per capita = Value Added\n3. Poverty level =\n\nHere,\n\n\n\nValue Added _i_, _j_ Total employment _i_, _j_\nValue Added _i_ Total population _i_\nPopulation below income threshold _i_ Total population _i_\n\n\n\n_i_  Department, _j_  Industry\n\n\n\n5 We discuss this in more detail in Section 5.\n6 Departments of Lima and Callao are grouped into one region (Lima‐Callao), following the classification used by INEI in the initial\nyears of the sample (early 2000s).\n\n\n4\n\nMention: \"ENAHO\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Also the source for poverty rates by department.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006550:12:1", "occurrence_id": "03d71ea5628eecb4ba510a51", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \nUBI 2009 is also a multi‐stage stratified sample drawn from the same listing of enterprise as UBR 2010.\nThis provides us with appropriate sampling weights with the help of which we can obtain consistent\nestimates of the population values of variables interest for the years 2002 and 2009 from the\nobservations of UBI 2002 and UBI 2009, respectively.\n\n\n**_2.2._** **_Broad sectors and comparative advantage industries_**\n\n\nUBR 2001 and UBR 2010 covered all three digit ISIC Rev. 3 industries (with the exception of government\ninstitutions and embassies). So did the samples for UBI 2002 and UBI 2009. Unlike other sectors where\ninformal businesses were also covered, only formal businesses activities were covered for the agricultural\nsector in UBR2001. This was changed in UBR 2010, which covered both formal and informal agricultural\nbusinesses.\n\n\nAs already noted each of the two datasets –that is, UBR 2001 and UBR 2010 on one hand, and UBI2002\nand UBI2009 on the other‐ have been aggregated spatially to the sub‐county level as the unit of\nobservation and functionally to the four‐digit –ISIC industry as the alternative unit of observation.\n\n\nIn describing the pattern of structural change as observed between UBR2001 and UBR2010 we have\nfocused on four major divides, namely,\n\n\n11\n\nMention: \"UBI 2009\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"A multi-stage stratified sample drawn from the same enterprise listing.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006550:12:1", "occurrence_id": "ebf8cc0db8cf5be78611bf00", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \nUBI 2009 is also a multi‐stage stratified sample drawn from the same listing of enterprise as UBR 2010.\nThis provides us with appropriate sampling weights with the help of which we can obtain consistent\nestimates of the population values of variables interest for the years 2002 and 2009 from the\nobservations of UBI 2002 and UBI 2009, respectively.\n\n\n**_2.2._** **_Broad sectors and comparative advantage industries_**\n\n\nUBR 2001 and UBR 2010 covered all three digit ISIC Rev. 3 industries (with the exception of government\ninstitutions and embassies). So did the samples for UBI 2002 and UBI 2009. Unlike other sectors where\ninformal businesses were also covered, only formal businesses activities were covered for the agricultural\nsector in UBR2001. This was changed in UBR 2010, which covered both formal and informal agricultural\nbusinesses.\n\n\nAs already noted each of the two datasets –that is, UBR 2001 and UBR 2010 on one hand, and UBI2002\nand UBI2009 on the other‐ have been aggregated spatially to the sub‐county level as the unit of\nobservation and functionally to the four‐digit –ISIC industry as the alternative unit of observation.\n\n\nIn describing the pattern of structural change as observed between UBR2001 and UBR2010 we have\nfocused on four major divides, namely,\n\n\n11\n\nMention: \"UBR 2010\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its enterprise listing provides the sampling frame and weights.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006550:12:1", "occurrence_id": "a37c316edb40e179766407e3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \nUBI 2009 is also a multi‐stage stratified sample drawn from the same listing of enterprise as UBR 2010.\nThis provides us with appropriate sampling weights with the help of which we can obtain consistent\nestimates of the population values of variables interest for the years 2002 and 2009 from the\nobservations of UBI 2002 and UBI 2009, respectively.\n\n\n**_2.2._** **_Broad sectors and comparative advantage industries_**\n\n\nUBR 2001 and UBR 2010 covered all three digit ISIC Rev. 3 industries (with the exception of government\ninstitutions and embassies). So did the samples for UBI 2002 and UBI 2009. Unlike other sectors where\ninformal businesses were also covered, only formal businesses activities were covered for the agricultural\nsector in UBR2001. This was changed in UBR 2010, which covered both formal and informal agricultural\nbusinesses.\n\n\nAs already noted each of the two datasets –that is, UBR 2001 and UBR 2010 on one hand, and UBI2002\nand UBI2009 on the other‐ have been aggregated spatially to the sub‐county level as the unit of\nobservation and functionally to the four‐digit –ISIC industry as the alternative unit of observation.\n\n\nIn describing the pattern of structural change as observed between UBR2001 and UBR2010 we have\nfocused on four major divides, namely,\n\n\n11\n\nMention: \"UBI 2002\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The earlier round carrying the population values for that year.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006550:12:1", "occurrence_id": "63081cc46b080ec3c3aa33c1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \nUBI 2009 is also a multi‐stage stratified sample drawn from the same listing of enterprise as UBR 2010.\nThis provides us with appropriate sampling weights with the help of which we can obtain consistent\nestimates of the population values of variables interest for the years 2002 and 2009 from the\nobservations of UBI 2002 and UBI 2009, respectively.\n\n\n**_2.2._** **_Broad sectors and comparative advantage industries_**\n\n\nUBR 2001 and UBR 2010 covered all three digit ISIC Rev. 3 industries (with the exception of government\ninstitutions and embassies). So did the samples for UBI 2002 and UBI 2009. Unlike other sectors where\ninformal businesses were also covered, only formal businesses activities were covered for the agricultural\nsector in UBR2001. This was changed in UBR 2010, which covered both formal and informal agricultural\nbusinesses.\n\n\nAs already noted each of the two datasets –that is, UBR 2001 and UBR 2010 on one hand, and UBI2002\nand UBI2009 on the other‐ have been aggregated spatially to the sub‐county level as the unit of\nobservation and functionally to the four‐digit –ISIC industry as the alternative unit of observation.\n\n\nIn describing the pattern of structural change as observed between UBR2001 and UBR2010 we have\nfocused on four major divides, namely,\n\n\n11\n\nMention: \"UBR 2001\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"One of two benchmark rounds aggregated spatially.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006550:12:1", "occurrence_id": "4931786bd894cf66abb0071a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \nUBI 2009 is also a multi‐stage stratified sample drawn from the same listing of enterprise as UBR 2010.\nThis provides us with appropriate sampling weights with the help of which we can obtain consistent\nestimates of the population values of variables interest for the years 2002 and 2009 from the\nobservations of UBI 2002 and UBI 2009, respectively.\n\n\n**_2.2._** **_Broad sectors and comparative advantage industries_**\n\n\nUBR 2001 and UBR 2010 covered all three digit ISIC Rev. 3 industries (with the exception of government\ninstitutions and embassies). So did the samples for UBI 2002 and UBI 2009. Unlike other sectors where\ninformal businesses were also covered, only formal businesses activities were covered for the agricultural\nsector in UBR2001. This was changed in UBR 2010, which covered both formal and informal agricultural\nbusinesses.\n\n\nAs already noted each of the two datasets –that is, UBR 2001 and UBR 2010 on one hand, and UBI2002\nand UBI2009 on the other‐ have been aggregated spatially to the sub‐county level as the unit of\nobservation and functionally to the four‐digit –ISIC industry as the alternative unit of observation.\n\n\nIn describing the pattern of structural change as observed between UBR2001 and UBR2010 we have\nfocused on four major divides, namely,\n\n\n11\n\nMention: \"UBI2002\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Aggregated to the sub-county as the unit of observation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006550:12:1", "occurrence_id": "54d589fcf38785d2455d9a03", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \nUBI 2009 is also a multi‐stage stratified sample drawn from the same listing of enterprise as UBR 2010.\nThis provides us with appropriate sampling weights with the help of which we can obtain consistent\nestimates of the population values of variables interest for the years 2002 and 2009 from the\nobservations of UBI 2002 and UBI 2009, respectively.\n\n\n**_2.2._** **_Broad sectors and comparative advantage industries_**\n\n\nUBR 2001 and UBR 2010 covered all three digit ISIC Rev. 3 industries (with the exception of government\ninstitutions and embassies). So did the samples for UBI 2002 and UBI 2009. Unlike other sectors where\ninformal businesses were also covered, only formal businesses activities were covered for the agricultural\nsector in UBR2001. This was changed in UBR 2010, which covered both formal and informal agricultural\nbusinesses.\n\n\nAs already noted each of the two datasets –that is, UBR 2001 and UBR 2010 on one hand, and UBI2002\nand UBI2009 on the other‐ have been aggregated spatially to the sub‐county level as the unit of\nobservation and functionally to the four‐digit –ISIC industry as the alternative unit of observation.\n\n\nIn describing the pattern of structural change as observed between UBR2001 and UBR2010 we have\nfocused on four major divides, namely,\n\n\n11\n\nMention: \"UBI2009\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The later companion to the earlier round in the spatial aggregation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006551:14:0", "occurrence_id": "4a43b9a4c3454235dabea0a9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ##### **5 Results**\n\n**5.1** **Nation-wide** **effect** **on** **lean-season** **food** **security**\n\n\nWe begin by presenting results from equation (1) which estimates the direct effect of the\n\nprogram on treated households and the indirect effect on untreated households. <sup>12</sup> The sum\n\n\nof the coefficients _β_ 1 + _β_ 2 ( _PWP_ and _PWP_ _∗_ _Top −_ _up_ ) compares households in treatment\n\n\nvillages that were themselves included in PWP to households in control villages. This total\n\n\neffect on top-ups is reported in the tables, along with 95 percent confidence intervals (upper\n\n\nand lower bounds). Table 4 displays the effect of the program on food security measured\n\n\nduring the lean season (in survey round 2). Higher values indicate better food security for\n\n\nthe first five measures and the PCA index, and lower values indicate better food security for\n\n\noutcomes in columns 6-8. Panel A contains results from the full sample. The total effect\n\n\nof the program on top-up households moves in the direction of worse food security for all\n\n\nmeasures: negative coefficients for the first five outcomes, positive coefficients for the next\n\n\nthree outcomes, and a negative coefficient in column (9) for the PCA index. The effect on the\n\n\nPCA index is equivalent to one-tenth of a standard deviation of that outcome for households\n\n\nin control villages, but the coefficient - like those for all but one of the individual outcomes\n\n\n- is not statistically different from zero.\n\n\nHowever, these differences are at least partially driven by pre-existing differences in food\n\n\nsecurity. To account for this imbalance, we first limit the sample to the 159 villages covered\n\n\nby the IHS3 survey (Panel B) and and then include pre-treatment values\n\nMention: \"the IHS3 survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"The sample is limited to its 159 villages to correct an imbalance.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006552:31:0", "occurrence_id": "0b33596afe97ef31c742125f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: #### **Appendix**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n|Table A1. Variable Definitions|Col2|\n|---|---|\n|**Variables**|**Description**|\n|EFL score (old model)|Score generated using the EFL Africa model v2 (data pooled across<br>various African countries)|\n|EFL score (new model)|Score generated using the EFL Global model v1 (first to incorporate non-<br>African data)|\n|Sales|Value of sales reported by the entrepreneurs at the time of the application|\n|Debt-to-sales ratio|Debt-to-sales ratio estimated using data reported by the entrepreneur and<br>data from the SBS|\n|Age|Age of the entrepreneur at the time of the application|\n|Female|Dummy that equals 1 if entrepreneur is female|\n|Marital status<br> Divorced<br> Living with partner<br> Married<br> Separated<br> Single<br> Widowed|As reported by the entrepreneurs at the time of the application<br>Dummy that equals 1 if applicant is divorced<br>Dummy that equals 1 if applicant is living with a partner<br>Dummy that equals 1 if applicant is married<br>Dummy that equals 1 if applicant is separated<br>Dummy that equals 1 if applicant is single<br>Dummy that equals 1 if applicant is widowed|\n|Business sector<br> Agriculture<br> Commerce<br> Other services<br> Manufacturing|As reported by the entrepreneurs at the time of the application<br>Dummy that equals 1 if applicant operates in the sector “agriculture”<br>Dummy that equals 1 if applicant operates in the sector “commerce”<br>Dummy that equals 1\n\nMention: \"data from the SBS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Reported figures complete the debt-to-sales ratio.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006553:9:2", "occurrence_id": "bf6d02e05c7ec58b3c06f935", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: . The mechanism\nin the Aghion et al. (2010) model is through a lower share of long-term investment projects (that\nhave a more positive effect on productivity). For a panel of 21 OECD countries during 1960-2000,\nthe authors examine the effect of commodity price volatility on economic growth. Long-term\ninvestment projects are measured by the ratio of structural investment to total private investment.\n\n\n4 The index of institutional quality that is used in Frankel et al. is based on data provided by International Country Risk Guide on the\ninvestment profile, corruption, the rule of law, and bureaucratic quality.\n\n\n8\n\nMention: \"International Country Risk Guide\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its investment profile, corruption, rule of law and bureaucratic quality series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006554:4:0", "occurrence_id": "e5bb8a1305e965b6a309a620", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: and child mortality, as well as increases in coverage of maternal and child health\n\n\nservices including antenatal cares, contraceptive prevalence and skilled birth attendance\n\n\n(CSA, 2014). <sup>3</sup>\n\n\nNow with the MDG period coming to an end and new targets set in the\n\n\nSustainable Development Goals, studies are investigating not only if the countries\n\n\nprogressed towards achieving the targets, but also if the achievements were inclusive\n\n\n(Wagstaff et al., 2014). Indeed, health inequalities matter for several economic and non\n\neconomic reasons, as supported by vast empirical literature in this area.  However, the\n\n\navailable evidence on health inequality studies is mixed and context specific; there have\n\n\nbeen appreciable and widening inequalities in some countries and improvements in\n\n\nothers (e.g., Barros et al. 2012; Wagstaff et al., 2014). These studies also point out\n\n\nimportant heterogeneities in the results arising from the type of health service or the\n\n\nway it is delivered. For example, Barros et al. (2012) find that those services that are\n\n\ntraditionally delivered at health facilities, such as skilled birth attendant and antenatal\n\n\ncare visits, have more socioeconomic inequalities than those delivered by outreach or\n\n\nmass campaigns. This would point to the role of various constraints that\n\n\ndisproportionately affect health care utilization by the poor. Similarly, a recent multi\n\ncountry study of progress in health MDGs by Wagstaff et al. (2014) finds that the\n\n\npoor are more likely to face health risks, including child undernutrition and mortality,\n\n\n3 According to the 2014 mini DHS Survey, the reductions in undernutrition and child mortality\nlevels from 2000 and 2005 are considerable. Similarly, during the same period, there has been\nconsiderable expansion in maternal health service. However, trend should not be confused\nwith the overall health status.\n\n\n3\n\nMention: \"the 2014 mini DHS Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It records the reductions in undernutrition and child mortality since the baseline years.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006562:20:1", "occurrence_id": "dbff780d8fdd04de002c686d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  cover almost all the municipalities in the country, it lacks the breadth\n\nof observed variables available in the ENCV. Most importantly, it does not capture whether a\n\n\nhousehold receives the Colombia Mayor benefit. Furthermore, the included variables do not allow for\n\nthe construction of Sisben scores.\n\n\nThe ENCV sampled a total of 14,801 households in 2010, 25,364 in 2011, 21,383 in 2012, and 21,565\n\nin 2013. The sampling scheme changed in 2011 and 2012, but was held constant in 2013. The main\n\n\nimplication of the changes in the sampling scheme is that more municipalities are repeated in the last\n\ntwo rounds than in the prior ones. This, however, has no bearing on the validity of our empirical\n\n\napproach. We restrict our sample to households that have at least one eligible member according to\n\nthe age and Sisben criteria—that is, with a sufficiently low Sisben score given the place of residence—\n\n\nand contain either a woman of at least 52 years of age or a man of at least 57 years of age. This yields\n\na total of 15,955 households (2,507 in 2010, 5,618 in 2011, 4,120 in 2012, and 3,710 in 2013). The total\n\n\n19\n\nMention: \"the ENCV\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It carries variables the alternative source lacks, including programme receipt.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006544:5:0", "occurrence_id": "bd70d4b4f6d29bfdd5f70369", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: incremental value when it focuses on areas which are not being addressed multilaterally (Mattoo\n\n\nand Fink, 2004).\n\n\nThere are some broad caveats to our analysis. Market access in many of the countries was not\n\n\npredictable due to a discretionary licensing regime. From banking to transport, entry was\n\n\nrestricted by the explicit and implicit limits on new licenses and the licensing process tended to\n\n\nbe opaque and discretionary. In several ASEAN countries, licenses and foreign equity ownership\n\n\nwere decided on a case-by-case basis, subject to requirements or approvals that involved several\n\n\nregulators and ministries. Some countries in certain sectors had no regulation at all, especially\n\n\nthe lower income countries in the region and pertaining to the supply of services through the\n\n\ncross border and consumption abroad modes. In some of these cases, the supply of services was\n\n\nallowed in practice, while in others it was prohibited. In general, the high level of discretion and\n\n\nthe absence of regulation created a less predictable policy environment and made it hard to\n\n\ndefine and assess the policy regime.\n\n\nSection 2 describes the nature of services trade policy data as well as how it was collected and\n\n\nverified. Section 3 presents the ASEAN policy patterns and places them in a global context.\n\n\nSection 4 takes a closer look at the policy measures used by ASEAN countries, highlighting\n\n\ncertain aspects of the regulatory environment. Section 5 assesses whether ASEAN countries\n\n\nliberalized their policies between 2008 and 2012. Section 6 examines instances of where\n\n\nASEAN countries became more open vis-à-vis each other and provides two examples. Section 7\n\n\ncompares the regional and multilateral commitments of ASEAN countries with actual policy.\n\n\nSection 8 concludes.\n\n\n**2.** **SERVICES TRADE POLICY DATA AND MEASUREMENT**\n\n\nA detailed description of the original World Bank Services Trade Restrictions Database—\n\n\nincluding details on the data collection process, the policy measures covered and the\n\n\nquestionnaire used in the data collection—is provided in Borchert, Gootiiz and Mattoo (2012)\n\n\nand in supplementary\n\nMention: \"World Bank Services Trade Restrictions Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its collection process and policy coverage are described in detail here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006564:24:0", "occurrence_id": "ab98fe54103f410abb459ed0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Figure 6: Isolated PISA Effect of Changes in Selected Parameters in the Production Function,_\n\n_2006-2012_\n\n\n25\n\n\n\n20\n\n\n15\n\n\n10\n\n\n5\n\n\n0\n\n\n‐5\n\n\n‐10\n\n\n‐15\n\n\n‐20\n\n\n\n\n\n\n\n\n\n\n\n_Source_ : Authors’ simulations using estimation results.\n\n\n**VI.** **Conclusions**\nArgentina’s education system, which has shown virtually no improvements in basic\ncognitive abilities as measured by PISA test scores in the last 15 years, will have long-lasting\nnegative effects on the country’s productivity and economic growth prospects. Moreover, since\nthe learning gap between students in public versus private schools increased during the same\nperiod, the education system’s ability to reduce income inequality and promote social mobility is\ncompromised. It is therefore central to Argentina’s development agenda to understand the causes\nbehind these unwelcome trends in learning outcomes.\n\nDespite an important increase in the public resources invested in education inputs, such as\nrises in teacher numbers, teacher salaries, school infrastructure, and laptops, none of it seems to\nhave significantly improved learning outcomes. Our simulation shows that the trends in PISA\nlearning outcomes are explained not by the amount of inputs in the system but by how efficiently\nthose inputs are used in the production of learning. This reinforces the finding from our regression\nanalysis that quantity of teachers (as measured by pupil-teacher ratios) does not correlate with\nlearning, and that other inputs (such as computers and instructional materials) seem to have\nmarginal returns and diminish over time. In fact, the decline in learning outcomes between 2000\nand 2006 is largely explained by the drop in test scores of students in public schools, while much\nof the recovery between 2006 and 2012 is accounted for by improvements in socioeconomic\nvariables such as wealth and mothers’ education levels.\n\nThe results presented here highlight the failure of input-based education policies to\nimprove learning outcomes. This is not to say that education inputs are not important in the\nlearning process, but their impact is limited and seems to\n\nMention: \"PISA test scores\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Basic cognitive abilities show no improvement across fifteen years on this measure.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006569:27:0", "occurrence_id": "6f6279d3b163eb3c296d4893", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: strictions in the land market. We provide several pieces of evidence that relates to the effects\n\n\nof higher migration costs. <sup>25</sup> First, we find that propensity of outmigration is lower (see Figure\n\n\n1 based on population census 2012) and proportion of split households where some members are\n\n\nleft behind to hold on to the land is higher (see FIgure 2 based on HIES 2002) in villages with\n\n\nland retrictions. The census did not provide the out-migration data by gender. To check if female\n\n\nmigration is also less in the areas with higher land restrictions, we use census data on female\n\n\nshare in total population. The census data show that share of female in total population is higher\n\n\nwhen land restrictions are higher (Figure 3). This is true if we focus only on adult women (graph\n\n\nomitted for brevity). Using population census data from 2002 and 2012, we compute the change\n\n\nin female share of population for the same age cohort (15-49 year old in 2002 and 25-59 year old\n\n\nin 2012). Figure 4 plots this variable against proportion of land under restrictions. The positive\n\n\nslope confirms that more female are staying back in areas with more land under restrictions.\n\n\nThird, female share in the labor force –defined from HIES 2002 data – is also positively correlated\n\n\nwith the incidence of land restrictions (see Figure 5). Taken together, these can be interpreted as\n\n\nstrong direct evidence in favor of the proposition that women stay back in the village in response\n\n\nto higher migration costs arising from land restrictions. Finally, using HIES 2002 data, we find\n\n\nthat farm size is negatively correlated with land restrictions, as one would expect if more women\n\n\n(and possibly some men) stay back in the village facing higher migration costs (Figure 6).\n\n\nWe also find suggestive evidence that support the model where heterogeneity in labor force\n\n\nparticipation and the consequences of higher migration costs depend on productivity in home\n\nMention: \"population census 2012\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Out-migration propensity is lower in villages with land restrictions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006569:27:0", "occurrence_id": "5b8c00bcc950736472068ef2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: strictions in the land market. We provide several pieces of evidence that relates to the effects\n\n\nof higher migration costs. <sup>25</sup> First, we find that propensity of outmigration is lower (see Figure\n\n\n1 based on population census 2012) and proportion of split households where some members are\n\n\nleft behind to hold on to the land is higher (see FIgure 2 based on HIES 2002) in villages with\n\n\nland retrictions. The census did not provide the out-migration data by gender. To check if female\n\n\nmigration is also less in the areas with higher land restrictions, we use census data on female\n\n\nshare in total population. The census data show that share of female in total population is higher\n\n\nwhen land restrictions are higher (Figure 3). This is true if we focus only on adult women (graph\n\n\nomitted for brevity). Using population census data from 2002 and 2012, we compute the change\n\n\nin female share of population for the same age cohort (15-49 year old in 2002 and 25-59 year old\n\n\nin 2012). Figure 4 plots this variable against proportion of land under restrictions. The positive\n\n\nslope confirms that more female are staying back in areas with more land under restrictions.\n\n\nThird, female share in the labor force –defined from HIES 2002 data – is also positively correlated\n\n\nwith the incidence of land restrictions (see Figure 5). Taken together, these can be interpreted as\n\n\nstrong direct evidence in favor of the proposition that women stay back in the village in response\n\n\nto higher migration costs arising from land restrictions. Finally, using HIES 2002 data, we find\n\n\nthat farm size is negatively correlated with land restrictions, as one would expect if more women\n\n\n(and possibly some men) stay back in the village facing higher migration costs (Figure 6).\n\n\nWe also find suggestive evidence that support the model where heterogeneity in labor force\n\n\nparticipation and the consequences of higher migration costs depend on productivity in home\n\nMention: \"HIES 2002\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The share of split households holding on to land is higher there.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006569:27:0", "occurrence_id": "d973472bfbc9dc95f8f7278e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: strictions in the land market. We provide several pieces of evidence that relates to the effects\n\n\nof higher migration costs. <sup>25</sup> First, we find that propensity of outmigration is lower (see Figure\n\n\n1 based on population census 2012) and proportion of split households where some members are\n\n\nleft behind to hold on to the land is higher (see FIgure 2 based on HIES 2002) in villages with\n\n\nland retrictions. The census did not provide the out-migration data by gender. To check if female\n\n\nmigration is also less in the areas with higher land restrictions, we use census data on female\n\n\nshare in total population. The census data show that share of female in total population is higher\n\n\nwhen land restrictions are higher (Figure 3). This is true if we focus only on adult women (graph\n\n\nomitted for brevity). Using population census data from 2002 and 2012, we compute the change\n\n\nin female share of population for the same age cohort (15-49 year old in 2002 and 25-59 year old\n\n\nin 2012). Figure 4 plots this variable against proportion of land under restrictions. The positive\n\n\nslope confirms that more female are staying back in areas with more land under restrictions.\n\n\nThird, female share in the labor force –defined from HIES 2002 data – is also positively correlated\n\n\nwith the incidence of land restrictions (see Figure 5). Taken together, these can be interpreted as\n\n\nstrong direct evidence in favor of the proposition that women stay back in the village in response\n\n\nto higher migration costs arising from land restrictions. Finally, using HIES 2002 data, we find\n\n\nthat farm size is negatively correlated with land restrictions, as one would expect if more women\n\n\n(and possibly some men) stay back in the village facing higher migration costs (Figure 6).\n\n\nWe also find suggestive evidence that support the model where heterogeneity in labor force\n\n\nparticipation and the consequences of higher migration costs depend on productivity in home\n\nMention: \"census data on female\n\n\nshare in total population\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Its records show a higher female share where land restrictions bind.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006569:27:0", "occurrence_id": "8a345e2921aaa5996307a4e3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: strictions in the land market. We provide several pieces of evidence that relates to the effects\n\n\nof higher migration costs. <sup>25</sup> First, we find that propensity of outmigration is lower (see Figure\n\n\n1 based on population census 2012) and proportion of split households where some members are\n\n\nleft behind to hold on to the land is higher (see FIgure 2 based on HIES 2002) in villages with\n\n\nland retrictions. The census did not provide the out-migration data by gender. To check if female\n\n\nmigration is also less in the areas with higher land restrictions, we use census data on female\n\n\nshare in total population. The census data show that share of female in total population is higher\n\n\nwhen land restrictions are higher (Figure 3). This is true if we focus only on adult women (graph\n\n\nomitted for brevity). Using population census data from 2002 and 2012, we compute the change\n\n\nin female share of population for the same age cohort (15-49 year old in 2002 and 25-59 year old\n\n\nin 2012). Figure 4 plots this variable against proportion of land under restrictions. The positive\n\n\nslope confirms that more female are staying back in areas with more land under restrictions.\n\n\nThird, female share in the labor force –defined from HIES 2002 data – is also positively correlated\n\n\nwith the incidence of land restrictions (see Figure 5). Taken together, these can be interpreted as\n\n\nstrong direct evidence in favor of the proposition that women stay back in the village in response\n\n\nto higher migration costs arising from land restrictions. Finally, using HIES 2002 data, we find\n\n\nthat farm size is negatively correlated with land restrictions, as one would expect if more women\n\n\n(and possibly some men) stay back in the village facing higher migration costs (Figure 6).\n\n\nWe also find suggestive evidence that support the model where heterogeneity in labor force\n\n\nparticipation and the consequences of higher migration costs depend on productivity in home\n\nMention: \"population census data from 2002 and 2012\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The change in female share for the same age cohort is computed across these.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006569:27:0", "occurrence_id": "e5b882625cb82fe2668a38eb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: strictions in the land market. We provide several pieces of evidence that relates to the effects\n\n\nof higher migration costs. <sup>25</sup> First, we find that propensity of outmigration is lower (see Figure\n\n\n1 based on population census 2012) and proportion of split households where some members are\n\n\nleft behind to hold on to the land is higher (see FIgure 2 based on HIES 2002) in villages with\n\n\nland retrictions. The census did not provide the out-migration data by gender. To check if female\n\n\nmigration is also less in the areas with higher land restrictions, we use census data on female\n\n\nshare in total population. The census data show that share of female in total population is higher\n\n\nwhen land restrictions are higher (Figure 3). This is true if we focus only on adult women (graph\n\n\nomitted for brevity). Using population census data from 2002 and 2012, we compute the change\n\n\nin female share of population for the same age cohort (15-49 year old in 2002 and 25-59 year old\n\n\nin 2012). Figure 4 plots this variable against proportion of land under restrictions. The positive\n\n\nslope confirms that more female are staying back in areas with more land under restrictions.\n\n\nThird, female share in the labor force –defined from HIES 2002 data – is also positively correlated\n\n\nwith the incidence of land restrictions (see Figure 5). Taken together, these can be interpreted as\n\n\nstrong direct evidence in favor of the proposition that women stay back in the village in response\n\n\nto higher migration costs arising from land restrictions. Finally, using HIES 2002 data, we find\n\n\nthat farm size is negatively correlated with land restrictions, as one would expect if more women\n\n\n(and possibly some men) stay back in the village facing higher migration costs (Figure 6).\n\n\nWe also find suggestive evidence that support the model where heterogeneity in labor force\n\n\nparticipation and the consequences of higher migration costs depend on productivity in home\n\nMention: \"HIES 2002 data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Female labour force share is defined from this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006574:29:0", "occurrence_id": "3e0c81d7807145bed7b9c571", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 1: Histogram of raw responses on items in ECERS-R\n\n\nNote: Red outline indicates items that align with the Indonesia ECED Standard\n\n\n28\n\nMention: \"ECERS-R\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its item responses are compared with the national early childhood standard.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006578:8:1", "occurrence_id": "bb3cc5e6b90abcd27c6e01be", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: , and\nAfristat. The survey covers a wide range of socioeconomic variables collected through\nquestionnaires administered to households and communities. The “basic indicators of wellbeing”\nmodule contains data on household composition, labor, education, social capital, health, access to\nservices and credit. The “revenue and expenditure” module includes information on spending,\nconsumption, transfers and income. The household represents the statistical unit of analysis. Of the\n9,557 households surveyed in the 2014 EPCV, 55.3 percent were in urban centers and 44.7 percent\nwere in rural areas. As a secondary source of information, the analysis is based on data from the\n2008 EPCV. This household survey shares the same structure as the 2014 one, and consists of\n9,557 households. The two surveys are cross-sectional representative samples of the underlying\npopulation. In the following paragraphs, a number of descriptive statistics set the stage for the main\nempirical analysis, which will be presented in the next section.\n\n\nMauritanian households are generally organized according to a traditional patriarchal model. Sixtyeight percent of households are headed by men, and 32 percent are headed by women. Household\nsize is clearly correlated with poverty, and poverty incidence increases linearly with the number of\nhousehold members. Households headed by married people tend to both include more children and\nare poorer than households headed by single people. Polygamy is relatively common in Mauritania,\nand polygamous households tend to be among the largest and poorest in the country. Poverty rates\ndeclined among all household types between the 2008 and 2014 surveys, with medium-sized\nhouseholds showing the greatest degree of improvement (Figure 1).\n\n\n18 African Development Fund (2007). Mauritania Appraisal Report, PRECAMF.\n\n\n7\n\nMention: \"Afristat\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"A companion statistical programme behind the survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006578:8:1", "occurrence_id": "d1eca7246f08738020de14bb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: , and\nAfristat. The survey covers a wide range of socioeconomic variables collected through\nquestionnaires administered to households and communities. The “basic indicators of wellbeing”\nmodule contains data on household composition, labor, education, social capital, health, access to\nservices and credit. The “revenue and expenditure” module includes information on spending,\nconsumption, transfers and income. The household represents the statistical unit of analysis. Of the\n9,557 households surveyed in the 2014 EPCV, 55.3 percent were in urban centers and 44.7 percent\nwere in rural areas. As a secondary source of information, the analysis is based on data from the\n2008 EPCV. This household survey shares the same structure as the 2014 one, and consists of\n9,557 households. The two surveys are cross-sectional representative samples of the underlying\npopulation. In the following paragraphs, a number of descriptive statistics set the stage for the main\nempirical analysis, which will be presented in the next section.\n\n\nMauritanian households are generally organized according to a traditional patriarchal model. Sixtyeight percent of households are headed by men, and 32 percent are headed by women. Household\nsize is clearly correlated with poverty, and poverty incidence increases linearly with the number of\nhousehold members. Households headed by married people tend to both include more children and\nare poorer than households headed by single people. Polygamy is relatively common in Mauritania,\nand polygamous households tend to be among the largest and poorest in the country. Poverty rates\ndeclined among all household types between the 2008 and 2014 surveys, with medium-sized\nhouseholds showing the greatest degree of improvement (Figure 1).\n\n\n18 African Development Fund (2007). Mauritania Appraisal Report, PRECAMF.\n\n\n7\n\nMention: \"2014 EPCV\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Just under ten thousand households across urban and rural areas.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006578:8:1", "occurrence_id": "6d0c734d07cd45c919b2bdc3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: , and\nAfristat. The survey covers a wide range of socioeconomic variables collected through\nquestionnaires administered to households and communities. The “basic indicators of wellbeing”\nmodule contains data on household composition, labor, education, social capital, health, access to\nservices and credit. The “revenue and expenditure” module includes information on spending,\nconsumption, transfers and income. The household represents the statistical unit of analysis. Of the\n9,557 households surveyed in the 2014 EPCV, 55.3 percent were in urban centers and 44.7 percent\nwere in rural areas. As a secondary source of information, the analysis is based on data from the\n2008 EPCV. This household survey shares the same structure as the 2014 one, and consists of\n9,557 households. The two surveys are cross-sectional representative samples of the underlying\npopulation. In the following paragraphs, a number of descriptive statistics set the stage for the main\nempirical analysis, which will be presented in the next section.\n\n\nMauritanian households are generally organized according to a traditional patriarchal model. Sixtyeight percent of households are headed by men, and 32 percent are headed by women. Household\nsize is clearly correlated with poverty, and poverty incidence increases linearly with the number of\nhousehold members. Households headed by married people tend to both include more children and\nare poorer than households headed by single people. Polygamy is relatively common in Mauritania,\nand polygamous households tend to be among the largest and poorest in the country. Poverty rates\ndeclined among all household types between the 2008 and 2014 surveys, with medium-sized\nhouseholds showing the greatest degree of improvement (Figure 1).\n\n\n18 African Development Fund (2007). Mauritania Appraisal Report, PRECAMF.\n\n\n7\n\nMention: \"2008 EPCV\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The earlier round with an identical structure serves as a secondary source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006580:8:0", "occurrence_id": "378400e5311f0d8391ae1826", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: renewed relatively frequently. This can explain a preference for long-term debt on the part of\n\n\nfirms, and increased firm volatility if long-term debt is not available when it is preferable. If\n\n\nlong-term finance is undersupplied in a country due to reasons such as poor information or\n\n\ncontract enforcement, firms can either reduce investment in long-term assets or bear\n\n\nadditional liquidity risk (Aghion et al., 2010). To minimize liquidity risk as well as interest\n\n\nrate risk, firms often match the maturity structures of their assets and liabilities (Hart and\n\n\nMoore, 1995). <sup>4</sup> [^4: For empirical evidence on asset and liability maturity matching by firms, see Schiantarelli and Sembenelli\n(1997), Stohs and Mauer (1996), Jaramillo and Schiantarelli (2002), and Schiantarelli and Srivastava (1997).] Limited access to long-term finance inhibits maturity matching when\n\n\ninvesting long-term, possibly resulting in more volatile firm growth.\n\n\nIn summary, long-term debt may reduce firm-level volatility by reducing roll-over risk,\n\n\nwhile conversely a lower ability to monitor and enforce debt contracts may lead to greater\n\n\ninefficiency and risk-taking by firms. We empirically explore these relationships by relating\n\n\nmeasures of firm volatility to proxies for the availability of long-term finance, also\n\n\ncontrolling for the availability of overall external finance in some specifications.\n\n\n**3** **The data**\n\n\nIn this study, we relate measures of firm-level volatility to firm debt maturity structure.\n\n\nThe sample consists of firms in all sectors with the exception of financial firms and firms in\n\n\nthe public sector, as these firms’ capital structure decisions and risk profiles are very different\n\n\nfrom other firms.\n\n\nWe use two measures of firm volatility. In particular, we construct Asset volatility\n\n\n(book) as the standard deviation of the growth rate of the book value of total assets over the\n\n\n1995-2013 period using balance sheet information obtained from the Worldscope database.\n\n\n4 For empirical evidence on asset and liability maturity matching by firms, see Schiantarelli and Sembenelli\n(1997), Stohs and Mauer (\n\nMention: \"the Worldscope database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Balance sheet information across this period comes from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006581:61:2", "occurrence_id": "9d7e30ad5ce5a697237fe2a7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: : the high correlations among endogenous variables and therefore the instruments\nwe use for them could give rise to weak instrument problems. Therefore, in the table, we\nreport various weak-IV statistics to gauge the reasonableness of the 2SLS standard errors.\n\n\n74Because country definitions have changed over time, we concord historical countries to modern entities\nas best we can. For example, Germany today corresponds most closely to the former Federal Republic of\nGermany. Russia today corresponds to the former USSR. And so on. Further, more trade flows in the\nNBER-UN data are zero in 1970 than are zero today, likely due both to true changes from zeros to positive\nvalues over time and differences in reporting thresholds and/or missing data in the two data sources. In\norder to use the whole sample, we replace zeros in 1970 with the smallest values observed in the data.\n\n75Including the instruments separately, without imposing this ratio restriction, yields similar results.\n76Testing the exogeneity of the DVA-Ratio alone, one cannot reject exogeneity.\n\n\n59\n\nMention: \"NBER-UN data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"More zero trade flows are recorded in the earlier year than today.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006583:33:1", "occurrence_id": "8ecc8d9aea12d4238aa84b52", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  (Moody’s, S&P, and Fitch) and has traded corporate\nbonds outstanding for which information is available on Bloomberg. It is important to note that market\nprices factor in implicit government support and spreads between the Indonesian government and PLN\nare relatively tight. For water utilities, however, only very few companies are rated by local rating\nagencies and no traded corporate debt securities are outstanding.\n\n\n**_Resources and capacity_**\n\n\nThe contingent liabilities team at DGFRM consists of three sub-units, including risk analysis, the\nstructuring of guarantee agreements, and risk monitoring and reporting. The head of the team is supported\nby seven staff who rotate between sub-units to balance workloads and build skills across the entire\nspectrum of guarantee risk management. Staff have a general background in finance and economics and\nhave often worked in various units of the debt management office. However, staff mostly do not have\nprevious professional experience in credit risk analysis.\n\n\nThe Risk Management Unit from the Fiscal Policy Office (FPO) has recently been integrated into\nDGFRM. This unit also has responsibilities with respect to managing risks from government guarantees\nand state-owned enterprises. The integration of this team into DGFRM could help strengthen risk\nmanagement capacity. Other government(-related) entities involved in credit risk analysis include the\n\n\n47 Attention has to be given to differences in the institutional setup of guarantees and on-lending. While ability to repay would be the same in both\ncases and only dependent on the credit quality of the respective beneficiary, willingness to repay may differ depending on whether on-lending or\nguarantee agreements are in place.\n\n\n32 | P a g e\n\nMention: \"Bloomberg\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Traded corporate bond information is drawn from this vendor.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006586:30:1", "occurrence_id": "28e076a6ba39368c5f1a0b80", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 0.108<br>0.0367<br>(0.517)<br>(0.0890)<br>(0.0707)<br>0.258***<br>0.161***<br>0.238<br>(0.0465)<br>(0.00126)<br>(0.459)<br> <br> <br> <br>52<br>52<br>52<br>0.086<br>0.959<br>0.994<br> <br> <br> <br> <br>X <br>X <br> <br>X <br>X <br> <br> <br>X|\n\n\nNotes: * p<0.10, ** p<0.05, *** p<0.01.\n_Source_ : LFS 2011 and 2013. World Bank author’s calculations.\n\n|Col1|Table 14. OLS|Col3|\n|---|---|---|\n||Female Headed Households|Male Headed Households|\n||(1)<br>(2)<br>(3)|(1)<br>(2)<br>(3)|\n|<br>Proportion of Foreign-Born in the Population<br> <br>Constant<br> <br> <br>Observations<br>R-squared<br> <br>Year Controls<br>Regional Controls<br>Labor and Demographic Controls|<br> <br> <br>0.898**<br>0.00125<br>0.0348<br>(0.325)<br>(0.209)<\n\nMention: \"LFS 2011 and 2013\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Female head-of-household regressions are computed across these rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006589:4:0", "occurrence_id": "8e29c13e50b1a8cca39370dd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 2: Average Monthly Earnings**\n\n\nA. Less than College B. College\n\n\n\n\n\nNotes: Authors’ calculations using quarterly Labor Force Surveys from 2000 to 2015. Earnings are in constant Mexican Pesos\n2014:1. Earnings refer to full-time workers (at least 30 weekly hours of work) with a valid wage. It includes all workers (men and\nwomen): salaried, self-employment and owners. Earnings below and above the 1 and 99 percentile (within gender and education\ngroup) are dropped. Lines are smoothed by a moving average with a 0.4 weight to the current observation and 0.3 to the previous\nand post observation.\n\n\nThe average hourly wage for college-educated workers declined at a rate of approximately 2\npercent per year between 2001 and 2014. Given that during this period the share of college\ngraduates in the workforce rose from 14 to 18 percent, the decline in relative (and absolute)\nwages may be a simple matter of supply outpacing demand. An alternative explanation (and not\nmutually exclusive) could be that the fall in returns to college education is the outcome of a\nprocess intrinsic to a particular age group. Younger workers, for example, could be experiencing\na decline in their average wage because—as a result of the large expansion of tertiary education\ninstitutions since 1995—their tertiary education is not of the same quality on average as that of\nyounger workers in previous cohorts. For example, in the period 1984 – 1994, university\nenrollment increased by 1.95 percent per year, while during the 1994-2000 period, it increased\nby 5.4 percent per year. The skills acquired through tertiary education could have been of lower\nquality, either because the new post-secondary institutions are of lower quality and/or because\nthe marginal student has lower abilities. We call this the “degraded tertiary effect”. <sup>4</sup> Another\npossibility (again, not mutually exclusive) is that older\n\nMention: \"quarterly Labor Force Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Real monthly earnings across this span are calculated from these rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006591:12:0", "occurrence_id": "c16e08a9eb2ea66d41deb3c6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: different regions allows for two-way trade in each product category, depending on the ease of\n\n\nsubstitution between products from different regions. Land, physical capital, skilled, and unskilled\n\n\nlabor, and in the resource sectors, a natural resource, are used as primary factor inputs into\n\n\nproduction. <sup>3</sup> As all factor inputs other than the natural resource factor are perfectly mobile across\n\n\nsectors and returns adjust to changes in economic conditions, the results from this model are\n\n\nindicative of medium-term outcomes.\n\n\nThe government sector is represented in a stylized fashion. An index of current government\n\n\nexpenditure proxies the welfare derived from the government’s provision of public goods and\n\n\nservices to private households in a country. <sup>4</sup> [^4: The assumptions behind this approach are that the level of public goods provided in the initial equilibrium is\noptimal, preferences for public goods are separable from preferences for private goods, and the utility function for\npublic goods is identical across households within the regional economy (Keller, 1980).] A Cobb-Douglas utility function is used to allocate\n\n\ngovernment spending across composite goods and demand for composite goods is allocated\n\n\nbetween imports and domestically produced goods in a way completely analogous to that of firms\n\n\nand households.\n\n\n_3.2 The data_\n\n\nWe use the GTAP 9 database, aggregated into 13 countries/regions and 16\n\n\nsectors/industries (Appendix Table A.1 and A.2), modified to ensure consistency between Iran’s\n\n\nbilateral trade flows and UN Comtrade data. The aggregation presents individually the countries\n\n\nand groups of countries that (i) play an important role or could affect the outcome of the nuclear\n\n\nnegotiations (Iran, the US, the EU, Russia, Israel); and (ii) are major oil exporting countries or\n\n\ncountry groupings, most of them located in the Middle East and North Africa. The first criterion\n\n\n3 For example, oil fields are inputs into the process of crude oil extraction.\n4 The assumptions behind this approach are that the level\n\nMention: \"GTAP 9 database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Aggregated into 13 regions and 16 sectors for the modelling.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006591:12:0", "occurrence_id": "493238ede2a3689cbcffa3f4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: different regions allows for two-way trade in each product category, depending on the ease of\n\n\nsubstitution between products from different regions. Land, physical capital, skilled, and unskilled\n\n\nlabor, and in the resource sectors, a natural resource, are used as primary factor inputs into\n\n\nproduction. <sup>3</sup> As all factor inputs other than the natural resource factor are perfectly mobile across\n\n\nsectors and returns adjust to changes in economic conditions, the results from this model are\n\n\nindicative of medium-term outcomes.\n\n\nThe government sector is represented in a stylized fashion. An index of current government\n\n\nexpenditure proxies the welfare derived from the government’s provision of public goods and\n\n\nservices to private households in a country. <sup>4</sup> [^4: The assumptions behind this approach are that the level of public goods provided in the initial equilibrium is\noptimal, preferences for public goods are separable from preferences for private goods, and the utility function for\npublic goods is identical across households within the regional economy (Keller, 1980).] A Cobb-Douglas utility function is used to allocate\n\n\ngovernment spending across composite goods and demand for composite goods is allocated\n\n\nbetween imports and domestically produced goods in a way completely analogous to that of firms\n\n\nand households.\n\n\n_3.2 The data_\n\n\nWe use the GTAP 9 database, aggregated into 13 countries/regions and 16\n\n\nsectors/industries (Appendix Table A.1 and A.2), modified to ensure consistency between Iran’s\n\n\nbilateral trade flows and UN Comtrade data. The aggregation presents individually the countries\n\n\nand groups of countries that (i) play an important role or could affect the outcome of the nuclear\n\n\nnegotiations (Iran, the US, the EU, Russia, Israel); and (ii) are major oil exporting countries or\n\n\ncountry groupings, most of them located in the Middle East and North Africa. The first criterion\n\n\n3 For example, oil fields are inputs into the process of crude oil extraction.\n4 The assumptions behind this approach are that the level\n\nMention: \"UN Comtrade data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Used to ensure bilateral flows for one country are consistent.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006593:21:1", "occurrence_id": "e2acc11189a4e1f7566a61ce", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  evidence appears to\nindicate that such losses are large because storage conditions are inadequate, <sup>9</sup> [^9: “As crops rot, millions go hungry in India,” Reuters, July 1, 2012; “2,400 MT wheat rotting in govt granaries for\npast 2 years,” The Times of India, May 7, 2013.] but the actual rate at\nwhich stocks deteriorate is unknown. For private storage costs, following (29), the maximum\nstorage subsidy compatible with a stable steady state is equal to 1.4. Similarly, for public storage\ncosts of US$ 87 per ton, the maximum subsidy is 1.1.\n\n\n7 Private storage costs in India are competitive relative to the global benchmark used here. Estimates of storage\ncosts from National Bulk Handling Corporation, a private entity, are equivalent to about US$16 per ton in general,\nbut a bit higher at about $19.2 in the National Capital Region or port locations (Anil Choudhary, Managing Director\nand CEO, _personal communication)_ . Data from IFC from recent modern grain silos projects indicate a cost of storage\nat about US$ 19.75 per ton in Punjab, where the private developer of storage provides land, and $11.5 per ton in\nMadhya Pradesh, where land is provided by the government. The Food Corporation of India has fixed ceiling rates\nfor storage rental from private providers under the “Private Entrepreneurs Guarantee” (PEG) scheme at Rs 4.78 per\nquintal per month, which translates to only US$ 9.56 per ton (see Department of Food and Public Distribution website\nat http://dfpd.nic.in/?q=node/34). Finally, the base storage rental rates established by the Central Warehousing\nCorporation for private sector bidding for wheat storage under normal access conditions (that is flat terrain with easy\naccess such as may be expected in the national capital region) are about US$ 18 per\n\nMention: \"IFC\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its grain silo project records give a storage cost per tonne.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006602:9:0", "occurrence_id": "bce1d101a5f614ec5e4d99c4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: obstacles to growth, and performance measures. The survey is designed to be representative of a\n\n\ncountry’s private non-agricultural economy and only registered firms with at least five employees\n\n\nare included in the sample.\n\n\nThe data consist of a random sample of 12,000 firms across 27 developing countries in different\n\n\nregions stratified by firm size, location, and sector. The survey year ranges between 2007 and\n\n\n2009. This sample of ES data is selected on the basis of consistent sampling methodology. Surveys\n\n\nfor previous years do not have the same methodology. Police data is obtained from the United\n\n\nNations Survey on Crime Trends and the Operations of Criminal Justice Systems. This data set\n\n\nconsists of an unbalanced panel covering years between 2003 and 2008 for both developing and\n\n\ndeveloped economies. Due to the unbalanced nature of the data set, about 39 countries match\n\n\nbetween the ES firm-level data and the UN country-level police data. Accounting for issues such\n\n\nas missing questions and incorrect data, the sample size reduces to a cross-section of 27 developing\n\n\neconomies. To account for firm-specific unobservables, we use panel data for 644 firms in a subset\n\n\nof 12 Eastern European economies for 2004 and 2007. <sup>2</sup> We would caution that the firms sampled\n\n\nin 2004 do not use the same sampling methodology as the firms in 2007, and data are only available\n\n\nfor Eastern European economies. Details of countries in the sample and their respective survey\n\n\nyears can be found in the first column of table A3 in the appendix.\n\n\n_2.1 Dependent variable_\n\n\nThe dependent variable used is losses due to crime as a percentage of annual sales of a firm. This\n\n\nvariable is derived from the survey question: “In fiscal year [insert fiscal year], what are the\n\n\n2 Panel data are not available for the other countries\n\n\n8\n\nMention: \"ES data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Selected for a consistent sampling methodology across rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006602:9:0", "occurrence_id": "e3a16a91b81d77f9f7c98161", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: obstacles to growth, and performance measures. The survey is designed to be representative of a\n\n\ncountry’s private non-agricultural economy and only registered firms with at least five employees\n\n\nare included in the sample.\n\n\nThe data consist of a random sample of 12,000 firms across 27 developing countries in different\n\n\nregions stratified by firm size, location, and sector. The survey year ranges between 2007 and\n\n\n2009. This sample of ES data is selected on the basis of consistent sampling methodology. Surveys\n\n\nfor previous years do not have the same methodology. Police data is obtained from the United\n\n\nNations Survey on Crime Trends and the Operations of Criminal Justice Systems. This data set\n\n\nconsists of an unbalanced panel covering years between 2003 and 2008 for both developing and\n\n\ndeveloped economies. Due to the unbalanced nature of the data set, about 39 countries match\n\n\nbetween the ES firm-level data and the UN country-level police data. Accounting for issues such\n\n\nas missing questions and incorrect data, the sample size reduces to a cross-section of 27 developing\n\n\neconomies. To account for firm-specific unobservables, we use panel data for 644 firms in a subset\n\n\nof 12 Eastern European economies for 2004 and 2007. <sup>2</sup> We would caution that the firms sampled\n\n\nin 2004 do not use the same sampling methodology as the firms in 2007, and data are only available\n\n\nfor Eastern European economies. Details of countries in the sample and their respective survey\n\n\nyears can be found in the first column of table A3 in the appendix.\n\n\n_2.1 Dependent variable_\n\n\nThe dependent variable used is losses due to crime as a percentage of annual sales of a firm. This\n\n\nvariable is derived from the survey question: “In fiscal year [insert fiscal year], what are the\n\n\n2 Panel data are not available for the other countries\n\n\n8\n\nMention: \"United\n\n\nNations Survey on Crime Trends\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"An unbalanced set of police records across countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006602:9:0", "occurrence_id": "56301397a9601629232a4a3c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: obstacles to growth, and performance measures. The survey is designed to be representative of a\n\n\ncountry’s private non-agricultural economy and only registered firms with at least five employees\n\n\nare included in the sample.\n\n\nThe data consist of a random sample of 12,000 firms across 27 developing countries in different\n\n\nregions stratified by firm size, location, and sector. The survey year ranges between 2007 and\n\n\n2009. This sample of ES data is selected on the basis of consistent sampling methodology. Surveys\n\n\nfor previous years do not have the same methodology. Police data is obtained from the United\n\n\nNations Survey on Crime Trends and the Operations of Criminal Justice Systems. This data set\n\n\nconsists of an unbalanced panel covering years between 2003 and 2008 for both developing and\n\n\ndeveloped economies. Due to the unbalanced nature of the data set, about 39 countries match\n\n\nbetween the ES firm-level data and the UN country-level police data. Accounting for issues such\n\n\nas missing questions and incorrect data, the sample size reduces to a cross-section of 27 developing\n\n\neconomies. To account for firm-specific unobservables, we use panel data for 644 firms in a subset\n\n\nof 12 Eastern European economies for 2004 and 2007. <sup>2</sup> We would caution that the firms sampled\n\n\nin 2004 do not use the same sampling methodology as the firms in 2007, and data are only available\n\n\nfor Eastern European economies. Details of countries in the sample and their respective survey\n\n\nyears can be found in the first column of table A3 in the appendix.\n\n\n_2.1 Dependent variable_\n\n\nThe dependent variable used is losses due to crime as a percentage of annual sales of a firm. This\n\n\nvariable is derived from the survey question: “In fiscal year [insert fiscal year], what are the\n\n\n2 Panel data are not available for the other countries\n\n\n8\n\nMention: \"ES firm-level data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"About 39 countries match with the police series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006602:9:0", "occurrence_id": "9c86716fe2422cc89ae811f3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: obstacles to growth, and performance measures. The survey is designed to be representative of a\n\n\ncountry’s private non-agricultural economy and only registered firms with at least five employees\n\n\nare included in the sample.\n\n\nThe data consist of a random sample of 12,000 firms across 27 developing countries in different\n\n\nregions stratified by firm size, location, and sector. The survey year ranges between 2007 and\n\n\n2009. This sample of ES data is selected on the basis of consistent sampling methodology. Surveys\n\n\nfor previous years do not have the same methodology. Police data is obtained from the United\n\n\nNations Survey on Crime Trends and the Operations of Criminal Justice Systems. This data set\n\n\nconsists of an unbalanced panel covering years between 2003 and 2008 for both developing and\n\n\ndeveloped economies. Due to the unbalanced nature of the data set, about 39 countries match\n\n\nbetween the ES firm-level data and the UN country-level police data. Accounting for issues such\n\n\nas missing questions and incorrect data, the sample size reduces to a cross-section of 27 developing\n\n\neconomies. To account for firm-specific unobservables, we use panel data for 644 firms in a subset\n\n\nof 12 Eastern European economies for 2004 and 2007. <sup>2</sup> We would caution that the firms sampled\n\n\nin 2004 do not use the same sampling methodology as the firms in 2007, and data are only available\n\n\nfor Eastern European economies. Details of countries in the sample and their respective survey\n\n\nyears can be found in the first column of table A3 in the appendix.\n\n\n_2.1 Dependent variable_\n\n\nThe dependent variable used is losses due to crime as a percentage of annual sales of a firm. This\n\n\nvariable is derived from the survey question: “In fiscal year [insert fiscal year], what are the\n\n\n2 Panel data are not available for the other countries\n\n\n8\n\nMention: \"UN country-level police data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The crime side of the cross-national match.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006606:38:0", "occurrence_id": "46b6957bde5143c7e432a976", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 1\n\n\n0.9\n\n\n0.8\n\n\n0.7\n\n\n0.6\n\n\n0.5\n\n\n0.4\n\n\n0.3\n\n\n0.2\n\n\n0.1\n\n\n0\n\n\n\n0.7\n\n\n0.6\n\n\n0.5\n\n\n0.4\n\n\n0.3\n\n\n0.2\n\n\n0.1\n\n\n0\n\n\n\nAweil Bentiu Bor\n\n\nKajok Juba Malakal\n\n\nRumbek Wau\n\n\n\nAweil Bentiu Bor\n\n\nKajok Juba Malakal\n\n\nRumbek Wau\n\n\n\nSource: Authors’ calculations based on WFP data. Source: Authors’ calculations based on WFP data.\n\n\n\nFigure 25: Rolling Standard Deviation of Rates of\nChange of Sesame\n(2007m9-2013m12)\n\n\n0.7\n\n\n0.6\n\n\n0.5\n\n\n0.4\n\n\n0.3\n\n\n0.2\n\n\n0.1\n\n\n0\n\n\nAweil Bentiu Bor\n\n\nKajok Juba Malakal\n\n\nRumbek Wau\n\n\n\nFigure 26: Rolling Standard Deviation of Rates of\nChange of Sorghum\n(2006m6-2013m12)\n\n\n0.5\n\n\n0.45\n\n\n0.4\n\n\n0.35\n\n\n0.3\n\n\n0.25\n\n\n0.2\n\n\n0.15\n\n\n0.1\n\n\n0.05\n\n\n0\n\n\nAweil Bentiu Bor\n\n\nKajok Juba Malakal\n\n\nRumbek Wau\n\n\n\nSource: Authors’ calculations based on WFP data. Source: Authors’ calculations based on WFP data.\n\n\n37\n\nMention: \"WFP data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Market-level price statistics are calculated from this agency's records.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006565:16:0", "occurrence_id": "2bb480bec0b789d7985e1788", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Regression results are presented in the Appendix and are centered on identifying correlates of**\n**downward mobility between 1989 and 2006.** The dependent variable takes a value of one if the individual\nreported moving up or staying in the same position, and zero if it reported downward mobility. We estimated\nprobit regressions to facilitate the interpretation of the results; however, results were qualitatively equivalent\nunder a multivariate regression on three types of movements. A movement is defined (as in section 3) as a\nmovement in any direction of at least two steps, regardless of the initial point. Movements of only one step are\nconsidered as staying in the same step. We excluded households in the two lowest and uppermost steps in 2006\nfrom the sample to avoid biases introduced by the mechanical effect of being unable to move in all possible\ndirections (move down for the upper steps in 2006, and move up for lowest steps in 2006). All regressions and\na thorough discussion of the methodology used are presented in the Appendices.\n\n\n**Perceptions of inequality, fairness and trust matter**\n\n\n**Increases in country income inequality are associated with more negative perceptions of economic**\n**mobility.** Table A2.2 shows that household in countries in which inequality increased between 1989 and 2006\n\n- measured by the ratio of the 80 <sup>th</sup> /20 <sup>th</sup> percentiles– had a higher probability of reporting downward\nmovements in the income distribution (Column (3)). Using an alternative measure of Gini, obtained from the\nWorld Income Inequality Database shows also a negative effect of the increases in inequality with respect to\nthe perception of movements up <sup>9</sup> (Column (4)). This alternative measure, though, is less consistent than the\nobtained from the World Development Indicators and is incomplete for most former Yugoslavia countries,\n\nMention: \"World Income Inequality Database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"An alternative Gini from this database gives a negative inequality effect.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006565:16:0", "occurrence_id": "b6d4614b3324dc4d120f7b65", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Regression results are presented in the Appendix and are centered on identifying correlates of**\n**downward mobility between 1989 and 2006.** The dependent variable takes a value of one if the individual\nreported moving up or staying in the same position, and zero if it reported downward mobility. We estimated\nprobit regressions to facilitate the interpretation of the results; however, results were qualitatively equivalent\nunder a multivariate regression on three types of movements. A movement is defined (as in section 3) as a\nmovement in any direction of at least two steps, regardless of the initial point. Movements of only one step are\nconsidered as staying in the same step. We excluded households in the two lowest and uppermost steps in 2006\nfrom the sample to avoid biases introduced by the mechanical effect of being unable to move in all possible\ndirections (move down for the upper steps in 2006, and move up for lowest steps in 2006). All regressions and\na thorough discussion of the methodology used are presented in the Appendices.\n\n\n**Perceptions of inequality, fairness and trust matter**\n\n\n**Increases in country income inequality are associated with more negative perceptions of economic**\n**mobility.** Table A2.2 shows that household in countries in which inequality increased between 1989 and 2006\n\n- measured by the ratio of the 80 <sup>th</sup> /20 <sup>th</sup> percentiles– had a higher probability of reporting downward\nmovements in the income distribution (Column (3)). Using an alternative measure of Gini, obtained from the\nWorld Income Inequality Database shows also a negative effect of the increases in inequality with respect to\nthe perception of movements up <sup>9</sup> (Column (4)). This alternative measure, though, is less consistent than the\nobtained from the World Development Indicators and is incomplete for most former Yugoslavia countries,\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its measure is more consistent but incomplete for several countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006094:6:0", "occurrence_id": "fcb959ae2e1ed656f0633b1b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The representativeness of the sample varies across countries (Plaza et al., 2011). The\n\n\nsample selected is nationally representative for Nigeria, Senegal, and Uganda. The survey is not,\n\n\nhowever, nationally representative for Burkina Faso and Kenya. For Burkina Faso, the sample is\n\n\nrepresentative of the 10 provinces with the largest concentration of migrant sending households\n\n\n(or receiving remittances). For Kenya, the selected sample is representative of households in the\n\n\ntop 17 districts with the largest concentration of households with migrant members. Data were\n\n\ncollected from about 10,000 households for the five countries combined: 2,100 from Burkina\n\n\nFaso, 1,937 from Kenya, 2,245 from Nigeria, 1,833 from Senegal, and 1,870 from Uganda.\n\n\nThe questionnaire used for the data collection is standardized across all countries.\n\n\nImportantly, aside from questions that are common in household level surveys (e.g., pertaining to\n\n\nhousehold composition, expenditures, living conditions, etc.) the questionnaire gathers\n\n\ninformation on migration and remittances. In particular, it asks whether households have a\n\n\nmember who is currently residing abroad and if the migrant has sent remittances over the past 12\n\n\nmonths. Also, the survey collects information on the amount of remittances received by the\n\n\nhousehold and the method of transfer. Although there are other data sets containing information\n\n\non households’ migrant status and the use of financial services (most notably the World Bank\n\n\nGlobal Findex database, see Demirguc-Kunt and Klapper, 2012) this data set is very rich and\n\n\ncontains critical information that enables us to deal with potential econometric problems\n\n\ncommon to these types of studies. Particularly important is the fact that the questionnaire\n\n\ncontains information on the migrants’ destination countries. This information allows the\n\n\nconstruction of instruments that can be used to control for the potential endogeneity of\n\n\nremittance inflows in the regression of the likelihood of opening a bank account.\n\n\n5\n\nMention: \"World Bank\n\n\nGlobal Findex database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its richness on migrant status and financial use is the benchmark for this dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006607:36:0", "occurrence_id": "4f377ce527b13219004bb190", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Table 3.a. Relation between PACC's Presence and Real Per Capita Household Expenditure from ENIGH 2002‐2008**\n\n\n(1) (2) (3) (4) (5) (6) (7) (8)\n\n\n\nLog Per Cap\nVARIABLES\nExpenditure\n\n\n\nLog Per Cap\nExpenditure\n\n\n\nLog Per Cap\nExpenditure\n\n\n\nLog Per Cap\nExpenditure\n\n\n\nLog Per Cap\nExpenditure\n\n\n\nLog Per Cap\nExpenditure\n\n\n\nLog Per Cap\nExpenditure\n\n\n\nLog Per Cap\nExpenditure\n\n\n\n<mark>PACC Presence (dummy)</mark> <mark>0.00635</mark> <mark>0.0757**</mark> <mark>0.101**</mark> <mark>0.0783**</mark>\n<mark>(0.035)</mark> <mark>(0.032)</mark> <mark>(0.043)</mark> <mark>(0.033)</mark>\nPACC Coverage (% of land sowed) 0.0020 0.0849** 0.0913** 0.0848*\n(0.049) (0.040) (0.046) (0.043)\n<mark>Log of Maize Yield</mark> <mark>0.0870***</mark> <mark>0.0875***</mark> <mark>0.0389*</mark> <mark>0.0409**</mark> <mark>(0.042)</mark> <mark>(0.038)</mark> <mark>‐0.0852**</mark> <mark>‐0.0859**</mark>\n<mark>(0.030)</mark> <mark>(0.\n\nMention: \"ENIGH\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Real per capita household expenditure across these rounds carries the regressions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006613:37:1", "occurrence_id": "f8a45e8aa135366d4ea38d57", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: > <u>X,Y,XY</u>\n\nNotes: Data from the 1996 and 2011 population census at the community (sub-place) level. Each column corresponds to a separate\nOLS regression using a sample of communities within 30 kilometers of the former homeland borders (on both sides). The\ndependent variable is the difference in log per capita income for 2011-1996. Columns 1 to 3 include different functions of the\nlinear distance between communities and the homeland borders. Columns 4 and 5 instead include different functions of the\ngeographic coordinates (lat./lon. – X/Y). All regressions include 1996 controls and municipality fixed effects. 1996 controls\ninclude: linear distance to former homeland border, share of white, colored and Indian/Asian population, share by age group (0-9,\n10-19, 20-39, 40-59, 60+), household head education, share employed, share male, per capita income, total population, share\nconnected to electricity grid, share no refuse, share with piped water. All control variables are interacted with the homeland\nindicator. Regressions are weighted by community population in 1996. Heteroscedasticity robust standard errors in brackets\nclustered at the municipality-homeland level. ***p<0.01, **p<0.05, *p<0.1\n\nMention: \"the 1996 and 2011 population census\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Community-level observations across these two enumerations carry the regressions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006614:35:1", "occurrence_id": "e0e1445c4e459e7388eabdae", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 188 16.4 0.07 0.03 0.5 0.10 5211\n10,189-13,584 32.7 0.10 0.04 0.9 0.14 3982\n13,585-16,980 50.7 0.11 0.05 1.3 0.15 2931\n16,981-23,772 78.4 0.15 0.06 1.9 0.18 3931\n23,773-30,564 135.4 0.18 0.08 2.9 0.20 2312\n30,565-37,356 192.7 0.20 0.09 3.9 0.23 1412\n37,357-44,148 226.6 0.20 0.09 4.5 0.24 907\n44,149-50,940 264.5 0.19 0.09 5.2 0.25 542\n<u>Total</u> <u>75.5</u> <u>0.12</u> <u>0.06</u> <u>1.7</u> <u>0.16</u> <u>21168</u>\n\n\n_Notes:_ This table provides descriptive statistics based on the estimation sample for 2000-2006. Indicators are first\ncomputed at the municipal-year level and then averaged across municipalities and years within each population\nbracket. Data on municipal population aged 4 to 6 used to compute participation in public and private preschool\ncome from the 2000 census.\n\nMention: \"Data on municipal population aged 4 to 6\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The age group behind the preschool participation rates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006614:35:1", "occurrence_id": "87d5fe2a5cf638e77c6ceb12", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 188 16.4 0.07 0.03 0.5 0.10 5211\n10,189-13,584 32.7 0.10 0.04 0.9 0.14 3982\n13,585-16,980 50.7 0.11 0.05 1.3 0.15 2931\n16,981-23,772 78.4 0.15 0.06 1.9 0.18 3931\n23,773-30,564 135.4 0.18 0.08 2.9 0.20 2312\n30,565-37,356 192.7 0.20 0.09 3.9 0.23 1412\n37,357-44,148 226.6 0.20 0.09 4.5 0.24 907\n44,149-50,940 264.5 0.19 0.09 5.2 0.25 542\n<u>Total</u> <u>75.5</u> <u>0.12</u> <u>0.06</u> <u>1.7</u> <u>0.16</u> <u>21168</u>\n\n\n_Notes:_ This table provides descriptive statistics based on the estimation sample for 2000-2006. Indicators are first\ncomputed at the municipal-year level and then averaged across municipalities and years within each population\nbracket. Data on municipal population aged 4 to 6 used to compute participation in public and private preschool\ncome from the 2000 census.\n\nMention: \"2000 census\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The source of the municipal age-group counts.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006620:29:0", "occurrence_id": "e8d3b8d8df100c64dae34ef0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Table 3. Teacher Activity and Reasons for Absence**\n\n\n<u>PANEL A: PHYSICAL VERIFICATION & LOGBOOK</u>\n\n<u>RECORDS OF ABSENCE</u>\n\n\n\nPhysical\nverification (%)\n\n\n\nLog-book\nrecords (today)\n\n\n\nLogbook\nrecords (last\nworking day)\n\n\n\n**Year 2003** 26.29 19.07    **<u>Year 2010</u>** <u>23.64</u> <u>15.94</u> <u>10.24</u>\n\n\n<u>PANEL B: PHYSICAL VERIFICATION & TEACHER ACTIVITY</u>\n\n\n\n**Teacher**\n**Teacher Found in Classroom (%)**\n\n\n\n**Absent (%)**\n\n\n26.29\n\n\n\nActively teaching Passively teaching Not Teaching\n\n\n\n**Found**\n**outside**\n**classroom**\n\n\n\n**<u>(%)</u>**\n\n15.88 9.35\n\n\n\n**Year 2003**\n\n**<u>Year 2010</u>**\n\n\n\n42.93 5.56\n\n\n\n<u>53.08</u>\n\n\n\n<u>4.16</u> <u>8.96</u>\n\n\n\n<u>10.15</u> <u>23.64</u>\n\n\n\n<u>PANEL C. STATED REASONS FOR ABSENCE</u>\n**School** **<u>On Official Duty (%)</u>**\n**Closed (%)** Official other\n\n\n\n**<u>On Official Duty (%)</u>** **Authorized**\n\n\n\n**No Reason**\n\n**(%)**\n\n\n\n**Leave (%)**\n\n\n\n(panchayat\n\nmeetings,\n\npolitical\nmeetings, etc.\n\nMention: \"Log-book\nrecords\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Attendance verification compares physical checks with these records.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006627:10:1", "occurrence_id": "639ef9d39cf58cd8282990fc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  of firm level information, including the employment of the firm,\n\nthe share of foreign ownership, the year of incorporation, revenues, capital investment, the\n\nregion it is located in, and the sector in which it operates. Together, these firms comprise\n53% of employment in Jordan as compared to Labor Force Survey (LFS) data. <sup>15</sup>\n\n\nIn studying the spillover effects of FDI, these Jordanian data have several attractive\n\nfeatures. First, prior to 2000, FDI inflows were very low (below 1% of GDP). Since then,\n\nthe government removed restrictions to FDI in manufacturing and several service sectors\nincluding banking, telecommunication, and retail trade. <sup>16</sup> As a result, net FDI inflows tripled\n\nfrom US$ 0.9 billion in 2000 to US$ 2.8 billion in 2008. On average, these FDI inflows\n\namounted to 10.1 percent of GDP over the decade, a share which ranks Jordan among\n\n\n14We use sample weights for each firm and year making the subset of firms observed in both periods\nrepresentative for the entire population of more than 100,000 firms in the 2006 and 2011 firm census. As we\nfocus on changes over this five-year period, we can only use firms for which data are available in both years.\n\n15Note that LFS data typically exceed firm census employment data significantly in most countries, since\nthey usually better cover informal employment, illegal foreign workers, family workers, etc.\n\n16Note that, transport and professional (accounting and legal) services remain restricted to foreign investors\n(Schiffbauer et al., 2015). This motivates our sample splits between services and manufacturing.\n\n\n8\n\nMention: \"Labor Force Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"The surveyed firms cover 53 percent of employment measured here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006627:10:1", "occurrence_id": "16c520c123f491cc647fe597", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  of firm level information, including the employment of the firm,\n\nthe share of foreign ownership, the year of incorporation, revenues, capital investment, the\n\nregion it is located in, and the sector in which it operates. Together, these firms comprise\n53% of employment in Jordan as compared to Labor Force Survey (LFS) data. <sup>15</sup>\n\n\nIn studying the spillover effects of FDI, these Jordanian data have several attractive\n\nfeatures. First, prior to 2000, FDI inflows were very low (below 1% of GDP). Since then,\n\nthe government removed restrictions to FDI in manufacturing and several service sectors\nincluding banking, telecommunication, and retail trade. <sup>16</sup> As a result, net FDI inflows tripled\n\nfrom US$ 0.9 billion in 2000 to US$ 2.8 billion in 2008. On average, these FDI inflows\n\namounted to 10.1 percent of GDP over the decade, a share which ranks Jordan among\n\n\n14We use sample weights for each firm and year making the subset of firms observed in both periods\nrepresentative for the entire population of more than 100,000 firms in the 2006 and 2011 firm census. As we\nfocus on changes over this five-year period, we can only use firms for which data are available in both years.\n\n15Note that LFS data typically exceed firm census employment data significantly in most countries, since\nthey usually better cover informal employment, illegal foreign workers, family workers, etc.\n\n16Note that, transport and professional (accounting and legal) services remain restricted to foreign investors\n(Schiffbauer et al., 2015). This motivates our sample splits between services and manufacturing.\n\n\n8\n\nMention: \"firm census\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Over one hundred thousand firms enumerated in both years.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006627:10:1", "occurrence_id": "4e0001db140895f374ff9f52", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  of firm level information, including the employment of the firm,\n\nthe share of foreign ownership, the year of incorporation, revenues, capital investment, the\n\nregion it is located in, and the sector in which it operates. Together, these firms comprise\n53% of employment in Jordan as compared to Labor Force Survey (LFS) data. <sup>15</sup>\n\n\nIn studying the spillover effects of FDI, these Jordanian data have several attractive\n\nfeatures. First, prior to 2000, FDI inflows were very low (below 1% of GDP). Since then,\n\nthe government removed restrictions to FDI in manufacturing and several service sectors\nincluding banking, telecommunication, and retail trade. <sup>16</sup> As a result, net FDI inflows tripled\n\nfrom US$ 0.9 billion in 2000 to US$ 2.8 billion in 2008. On average, these FDI inflows\n\namounted to 10.1 percent of GDP over the decade, a share which ranks Jordan among\n\n\n14We use sample weights for each firm and year making the subset of firms observed in both periods\nrepresentative for the entire population of more than 100,000 firms in the 2006 and 2011 firm census. As we\nfocus on changes over this five-year period, we can only use firms for which data are available in both years.\n\n15Note that LFS data typically exceed firm census employment data significantly in most countries, since\nthey usually better cover informal employment, illegal foreign workers, family workers, etc.\n\n16Note that, transport and professional (accounting and legal) services remain restricted to foreign investors\n(Schiffbauer et al., 2015). This motivates our sample splits between services and manufacturing.\n\n\n8\n\nMention: \"firm census employment data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Typically exceeded by survey data, which better cover informal employment.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006631:41:1", "occurrence_id": "2a4688aa641c8a1a65df8b23", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: hold Budget survey<br>(HBS)|1997-2004|38.8|7.5|\n|5|Tunisia|Household Budget<br>Consumption and Living<br>Standards survey (HBCLS)|2005-2010|8.0|4.6|\n|6|Yemen|Household Budget survey<br>(HBS)|1998-2006|31.7|54.5|\n||**Regional average**|**Regional average**||**17.1**|**16.4**|\n|**Panel B: Subjective being**|**Panel B: Subjective being**|**Panel B: Subjective being**||||\n|**No**|**Country**|**Survey name**|**Survey years**|**Dissatisfaction**<br>**(percent)**|**Dissatisfaction**<br>**(percent)**|\n|**No**|**Country**|**Survey name**|**Survey years**|**1st period**|** 2nd period**|\n|1|Algeria|Gallup World Poll|2009-2012|23.2|25.8|\n|2|Egypt|Gallup World Poll|2009-2012|30.2|49.4|\n|3|Iraq|Gallup World Poll|2009-2012|41.7|45.7|\n|4|Jordan|Gallup World Poll|2009-2012|16.1|28.1|\n|5|Lebanon|Gallup World Poll|2009-2012|29.3|40.6|\n|6\n\nMention: \"hold Budget survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"One country's welfare periods rest on this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006631:41:1", "occurrence_id": "e5798dcc2dac701c802292a5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: hold Budget survey<br>(HBS)|1997-2004|38.8|7.5|\n|5|Tunisia|Household Budget<br>Consumption and Living<br>Standards survey (HBCLS)|2005-2010|8.0|4.6|\n|6|Yemen|Household Budget survey<br>(HBS)|1998-2006|31.7|54.5|\n||**Regional average**|**Regional average**||**17.1**|**16.4**|\n|**Panel B: Subjective being**|**Panel B: Subjective being**|**Panel B: Subjective being**||||\n|**No**|**Country**|**Survey name**|**Survey years**|**Dissatisfaction**<br>**(percent)**|**Dissatisfaction**<br>**(percent)**|\n|**No**|**Country**|**Survey name**|**Survey years**|**1st period**|** 2nd period**|\n|1|Algeria|Gallup World Poll|2009-2012|23.2|25.8|\n|2|Egypt|Gallup World Poll|2009-2012|30.2|49.4|\n|3|Iraq|Gallup World Poll|2009-2012|41.7|45.7|\n|4|Jordan|Gallup World Poll|2009-2012|16.1|28.1|\n|5|Lebanon|Gallup World Poll|2009-2012|29.3|40.6|\n|6\n\nMention: \"Gallup World Poll\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Across several countries its two periods give the reported changes.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006632:6:0", "occurrence_id": "bfd101bba9ac657458449640", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: vulnerability of mangroves due to sea level rise in a changing climate; in Section 5, we address\n\n\nthe coastal protection services of mangroves at risk in a changing climate; and in Section 6,\n\n\npresent the limitations of our analysis. Section 7 concludes with a brief discussion of the results.\n\n\n**2. Area Estimates of Coastal Mangroves in Developing Countries with Previous Exposure to**\n\n\n**Tropical Cyclones**\n\n\nFor our analysis, we used information provided by Giri et al. (2010) on the extent and\n\n\ndistribution of mangroves from the global mangrove databases of the USGS: Earth Resources\n\n\nObservation and Science Center. In this database, the status and distributions of mangroves\n\n\nwere mapped using the 30‐m resolution Global Land Survey (GLS) data for 2000 supplemented\n\n\nby Landsat archives. The GLS 2000 mosaics were prepared using images acquired from 1997 to\n\n\n2000. Landsat imagery from the USGS archives was used if GLS data were cloudy. While\n\n\nmapping, each image was normalized for variation in solar angle and earth‐sun distance by\n\n\nconverting the digital number values to the top‐of‐the‐atmosphere reflectance. The results\n\n\nwere validated with other existing global, regional and local data sets (for details, see Giri et al.\n\n\n2010). The USGS database includes a presence or absence grid cells showing the exact location,\n\n\nsize, and shape of the mangroves.\n\n\nIn order to estimate coastal mangrove areas by country, we extracted vector coastline masks\n\n\nfrom SRTM version 2 Surface Water Body Data provided by NASA, and used the country and\n\n\nregion identifiers used by the World Bank. Country boundaries along with mangrove data were\n\n\nused to estimate the extent of coastal mangrove forests, by country. We restricted our analysis\n\n\nto countries with previous exposure to tropical cyclones (UNEP/GRID 2009). A total of 46\n\n\ncountries meet the criteria for inclusion in this study (for country coverage, see Box 1). While\n\n\nother countries have mangrove forests, the absence of cyclones makes their storm protection\n\n\nservice less\n\nMention: \"global mangrove databases\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Mangrove extent and distribution come from this collection.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006632:6:0", "occurrence_id": "056eace4b9deee41bcccdb48", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: vulnerability of mangroves due to sea level rise in a changing climate; in Section 5, we address\n\n\nthe coastal protection services of mangroves at risk in a changing climate; and in Section 6,\n\n\npresent the limitations of our analysis. Section 7 concludes with a brief discussion of the results.\n\n\n**2. Area Estimates of Coastal Mangroves in Developing Countries with Previous Exposure to**\n\n\n**Tropical Cyclones**\n\n\nFor our analysis, we used information provided by Giri et al. (2010) on the extent and\n\n\ndistribution of mangroves from the global mangrove databases of the USGS: Earth Resources\n\n\nObservation and Science Center. In this database, the status and distributions of mangroves\n\n\nwere mapped using the 30‐m resolution Global Land Survey (GLS) data for 2000 supplemented\n\n\nby Landsat archives. The GLS 2000 mosaics were prepared using images acquired from 1997 to\n\n\n2000. Landsat imagery from the USGS archives was used if GLS data were cloudy. While\n\n\nmapping, each image was normalized for variation in solar angle and earth‐sun distance by\n\n\nconverting the digital number values to the top‐of‐the‐atmosphere reflectance. The results\n\n\nwere validated with other existing global, regional and local data sets (for details, see Giri et al.\n\n\n2010). The USGS database includes a presence or absence grid cells showing the exact location,\n\n\nsize, and shape of the mangroves.\n\n\nIn order to estimate coastal mangrove areas by country, we extracted vector coastline masks\n\n\nfrom SRTM version 2 Surface Water Body Data provided by NASA, and used the country and\n\n\nregion identifiers used by the World Bank. Country boundaries along with mangrove data were\n\n\nused to estimate the extent of coastal mangrove forests, by country. We restricted our analysis\n\n\nto countries with previous exposure to tropical cyclones (UNEP/GRID 2009). A total of 46\n\n\ncountries meet the criteria for inclusion in this study (for country coverage, see Box 1). While\n\n\nother countries have mangrove forests, the absence of cyclones makes their storm protection\n\n\nservice less\n\nMention: \"30‐m resolution Global Land Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its mosaics carry the mapping of mangrove status.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006632:6:0", "occurrence_id": "12c4a0d055832c117cec92a5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: vulnerability of mangroves due to sea level rise in a changing climate; in Section 5, we address\n\n\nthe coastal protection services of mangroves at risk in a changing climate; and in Section 6,\n\n\npresent the limitations of our analysis. Section 7 concludes with a brief discussion of the results.\n\n\n**2. Area Estimates of Coastal Mangroves in Developing Countries with Previous Exposure to**\n\n\n**Tropical Cyclones**\n\n\nFor our analysis, we used information provided by Giri et al. (2010) on the extent and\n\n\ndistribution of mangroves from the global mangrove databases of the USGS: Earth Resources\n\n\nObservation and Science Center. In this database, the status and distributions of mangroves\n\n\nwere mapped using the 30‐m resolution Global Land Survey (GLS) data for 2000 supplemented\n\n\nby Landsat archives. The GLS 2000 mosaics were prepared using images acquired from 1997 to\n\n\n2000. Landsat imagery from the USGS archives was used if GLS data were cloudy. While\n\n\nmapping, each image was normalized for variation in solar angle and earth‐sun distance by\n\n\nconverting the digital number values to the top‐of‐the‐atmosphere reflectance. The results\n\n\nwere validated with other existing global, regional and local data sets (for details, see Giri et al.\n\n\n2010). The USGS database includes a presence or absence grid cells showing the exact location,\n\n\nsize, and shape of the mangroves.\n\n\nIn order to estimate coastal mangrove areas by country, we extracted vector coastline masks\n\n\nfrom SRTM version 2 Surface Water Body Data provided by NASA, and used the country and\n\n\nregion identifiers used by the World Bank. Country boundaries along with mangrove data were\n\n\nused to estimate the extent of coastal mangrove forests, by country. We restricted our analysis\n\n\nto countries with previous exposure to tropical cyclones (UNEP/GRID 2009). A total of 46\n\n\ncountries meet the criteria for inclusion in this study (for country coverage, see Box 1). While\n\n\nother countries have mangrove forests, the absence of cyclones makes their storm protection\n\n\nservice less\n\nMention: \"Landsat archives\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Supplement the mosaics; images acquired across these years.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006632:6:0", "occurrence_id": "779ffd7a14078c60fa6cbed4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: vulnerability of mangroves due to sea level rise in a changing climate; in Section 5, we address\n\n\nthe coastal protection services of mangroves at risk in a changing climate; and in Section 6,\n\n\npresent the limitations of our analysis. Section 7 concludes with a brief discussion of the results.\n\n\n**2. Area Estimates of Coastal Mangroves in Developing Countries with Previous Exposure to**\n\n\n**Tropical Cyclones**\n\n\nFor our analysis, we used information provided by Giri et al. (2010) on the extent and\n\n\ndistribution of mangroves from the global mangrove databases of the USGS: Earth Resources\n\n\nObservation and Science Center. In this database, the status and distributions of mangroves\n\n\nwere mapped using the 30‐m resolution Global Land Survey (GLS) data for 2000 supplemented\n\n\nby Landsat archives. The GLS 2000 mosaics were prepared using images acquired from 1997 to\n\n\n2000. Landsat imagery from the USGS archives was used if GLS data were cloudy. While\n\n\nmapping, each image was normalized for variation in solar angle and earth‐sun distance by\n\n\nconverting the digital number values to the top‐of‐the‐atmosphere reflectance. The results\n\n\nwere validated with other existing global, regional and local data sets (for details, see Giri et al.\n\n\n2010). The USGS database includes a presence or absence grid cells showing the exact location,\n\n\nsize, and shape of the mangroves.\n\n\nIn order to estimate coastal mangrove areas by country, we extracted vector coastline masks\n\n\nfrom SRTM version 2 Surface Water Body Data provided by NASA, and used the country and\n\n\nregion identifiers used by the World Bank. Country boundaries along with mangrove data were\n\n\nused to estimate the extent of coastal mangrove forests, by country. We restricted our analysis\n\n\nto countries with previous exposure to tropical cyclones (UNEP/GRID 2009). A total of 46\n\n\ncountries meet the criteria for inclusion in this study (for country coverage, see Box 1). While\n\n\nother countries have mangrove forests, the absence of cyclones makes their storm protection\n\n\nservice less\n\nMention: \"USGS archives\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Used where the primary mosaics were cloudy.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006632:6:0", "occurrence_id": "f1c6658561ad118c3d7a4bbe", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: vulnerability of mangroves due to sea level rise in a changing climate; in Section 5, we address\n\n\nthe coastal protection services of mangroves at risk in a changing climate; and in Section 6,\n\n\npresent the limitations of our analysis. Section 7 concludes with a brief discussion of the results.\n\n\n**2. Area Estimates of Coastal Mangroves in Developing Countries with Previous Exposure to**\n\n\n**Tropical Cyclones**\n\n\nFor our analysis, we used information provided by Giri et al. (2010) on the extent and\n\n\ndistribution of mangroves from the global mangrove databases of the USGS: Earth Resources\n\n\nObservation and Science Center. In this database, the status and distributions of mangroves\n\n\nwere mapped using the 30‐m resolution Global Land Survey (GLS) data for 2000 supplemented\n\n\nby Landsat archives. The GLS 2000 mosaics were prepared using images acquired from 1997 to\n\n\n2000. Landsat imagery from the USGS archives was used if GLS data were cloudy. While\n\n\nmapping, each image was normalized for variation in solar angle and earth‐sun distance by\n\n\nconverting the digital number values to the top‐of‐the‐atmosphere reflectance. The results\n\n\nwere validated with other existing global, regional and local data sets (for details, see Giri et al.\n\n\n2010). The USGS database includes a presence or absence grid cells showing the exact location,\n\n\nsize, and shape of the mangroves.\n\n\nIn order to estimate coastal mangrove areas by country, we extracted vector coastline masks\n\n\nfrom SRTM version 2 Surface Water Body Data provided by NASA, and used the country and\n\n\nregion identifiers used by the World Bank. Country boundaries along with mangrove data were\n\n\nused to estimate the extent of coastal mangrove forests, by country. We restricted our analysis\n\n\nto countries with previous exposure to tropical cyclones (UNEP/GRID 2009). A total of 46\n\n\ncountries meet the criteria for inclusion in this study (for country coverage, see Box 1). While\n\n\nother countries have mangrove forests, the absence of cyclones makes their storm protection\n\n\nservice less\n\nMention: \"USGS database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Holds presence grids giving exact location, size and shape.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006632:6:0", "occurrence_id": "af98ee00e29726a642776f2c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: vulnerability of mangroves due to sea level rise in a changing climate; in Section 5, we address\n\n\nthe coastal protection services of mangroves at risk in a changing climate; and in Section 6,\n\n\npresent the limitations of our analysis. Section 7 concludes with a brief discussion of the results.\n\n\n**2. Area Estimates of Coastal Mangroves in Developing Countries with Previous Exposure to**\n\n\n**Tropical Cyclones**\n\n\nFor our analysis, we used information provided by Giri et al. (2010) on the extent and\n\n\ndistribution of mangroves from the global mangrove databases of the USGS: Earth Resources\n\n\nObservation and Science Center. In this database, the status and distributions of mangroves\n\n\nwere mapped using the 30‐m resolution Global Land Survey (GLS) data for 2000 supplemented\n\n\nby Landsat archives. The GLS 2000 mosaics were prepared using images acquired from 1997 to\n\n\n2000. Landsat imagery from the USGS archives was used if GLS data were cloudy. While\n\n\nmapping, each image was normalized for variation in solar angle and earth‐sun distance by\n\n\nconverting the digital number values to the top‐of‐the‐atmosphere reflectance. The results\n\n\nwere validated with other existing global, regional and local data sets (for details, see Giri et al.\n\n\n2010). The USGS database includes a presence or absence grid cells showing the exact location,\n\n\nsize, and shape of the mangroves.\n\n\nIn order to estimate coastal mangrove areas by country, we extracted vector coastline masks\n\n\nfrom SRTM version 2 Surface Water Body Data provided by NASA, and used the country and\n\n\nregion identifiers used by the World Bank. Country boundaries along with mangrove data were\n\n\nused to estimate the extent of coastal mangrove forests, by country. We restricted our analysis\n\n\nto countries with previous exposure to tropical cyclones (UNEP/GRID 2009). A total of 46\n\n\ncountries meet the criteria for inclusion in this study (for country coverage, see Box 1). While\n\n\nother countries have mangrove forests, the absence of cyclones makes their storm protection\n\n\nservice less\n\nMention: \"SRTM version 2 Surface Water Body Data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Vector coastline masks are extracted from this dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006633:15:0", "occurrence_id": "60cdd0134caab2eff0221fcc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ### **4 Parameterization**\n\nIn this model, labor productivity is a function of the calendar, as well as\n\ntwo depreciation rates, _δr_ and _δm_, and two intensity parameters, _αr_ and _αm_ .\n\nIn this section, I narrow down the parameter space based on three criteria.\n\nFirst, the parameters must imply labor productivity that peaks during the\n\nworkweek, consistent with the patterns of electricity usage documented be\nlow. Second, the variation in productivity over the workweek must be within\n\nthe range of predicted variation in electricity usage per worker. Finally, the\n\nparameters must be such that a year of absence from work is associated\n\nwith a reentry penalty, consistent with magnitudes estimated in the labor\n\nliterature.\n\n\nThe first two criteria are based on predicted variation in electricity per\nworker over the workweek. Variation in electricity usage is informative of\n\nvariation in the utilization rate of capital, which, in the context of the model,\n\nis proportional to variations in _Lt_ (equation 4). I use data on electricity loads\n\nfrom seven regional transmission organizations (RTOs): ERCOT, PJM, ISONE, CAISO, MISO, NYISO and SPP. <sup>11</sup> The aggregate generating capacity of\n\nthese RTOs corresponds to roughly 60% of US totals, though coverage varies\nby RTO. <sup>12</sup> Observations range from January 1st, 2003 and February 15th,\n\n2015. Data are available on an hourly frequency (and on a higher frequency\n\nfor some RTOs) and are aggregated to a daily frequency. To remove long-run\nand seasonal trends, the data is de-trended using a weekly moving average. <sup>13</sup>\n\n\nTo account for potential variation in the number of workers over the\n\nworkweek, I use the American Time Use Survey (ATUS) for\n\nMention: \"American Time Use Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Accounts for variation in the number of workers across the workweek.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006635:6:0", "occurrence_id": "57194997ae244f5820f3f4d1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: transformation has to be used to assign first period characteristics to the second-period observations\n(Inchauste et al., 2012; Azevedo et al., 2013a; Azevedo et al., 2013b).\n\n\nRank preservation implies anonymity, where the identity of the individual experiencing a particular\noutcome (such as income) is irrelevant—a permutation of outcomes between any two individuals in\nany of the two distributions being compared has no effect on the comparison. With this assumption,\none can compare across the same quantiles of two distributions (including distributions at two periods\nof time) _as if_ they are the same observations. But such a comparison may not account for the true\nnature of mobility trajectories (Bourguignon, 2010). <sup>5</sup> [^5: This is most easily seen by a thought experiment of two households with vastly different incomes, where the\nincomes switch perfectly from one period to another (household A’s income in period 1 becomes household\nB’s income in period 2, and vice-versa), while every other household’s income remains constant. With\nanonymity, no change can be detected; whereas in a panel data, there will be change.] By using panel data, where the assumption of\nrank preservation is not needed since households can be observed over both periods, we account for\nthe true trajectories of mobility, which results in more precise estimates of the contribution of each\ncomponent to poverty reduction. As we will see later, the ability to account for mobility is particularly\nimportant in this case, given the vast differences in the distribution of income growth across\nhouseholds that are observed, depending on whether one uses an anonymous or nonanonymous\nmeasure of growth incidence.\n\n\n**3.** **Data and Descriptive Statistics**\n\n\nWe use the India Human Development Survey (IHDS) for 2005 and 2012, a nationally representative\nsurvey that was conducted in all states and union territories of India (with the exception of Andaman\nand Nicobar and Lakshadweep). Both rounds of the\n\nMention: \"India Human Development Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"A nationally representative survey across all states in these two rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006637:38:0", "occurrence_id": "61af243f46850ecb11851d66", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Table 5: Employment growth and credit bureau reform – Enterprise Survey sample**\nThe following regressions estimate equation 2 with robust standard errors clustered at the survey level reported in parentheses.\nRefer to Table 1 for variable definitions. The t-test is under the null hypothesis of the equality of the estimate for the _CB reform_\nvariable in the respective size categories. *** p<0.01, ** p<0.05, * p<0.1\n\n|Col1|(1) (2) (3) (4) Employment growth (%)|\n|---|---|\n|Dataset<br>Sample|Enterprise Survey<br>All firms<br>MSME firms<br>SME firms<br>Large firms|\n|CB reform<br>Log firm size<br>Single establishment<br>Manufacturing<br>Exporter<br>Foreign owned<br>Government owned<br>Log firm age<br>Inflation rate<br>GDP growth rate<br>Constant<br>Observations<br>R-squared<br>Reform countries<br>Control countries<br>Country FE<br>Year FE<br>T-Test<br>p-value|1.282*<br>1.337*<br>1.791**<br>1.113<br>[0.711]<br>[0.787]<br>[0.808]<br>[1.005]<br>-3.394***<br>-4.356***<br>-1.580***<br>-0.882***<br>[0.130]<br>[0.169]<br>[0.173]<br>[0.\n\nMention: \"Enterprise Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Employment growth across firm size classes is regressed on this dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006639:19:0", "occurrence_id": "f649e9e0bc963ac1bc58b47b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 1: Average employment over time in all durable goods industries**\n\n\n\n\n\nNote: The figure displays data from the Mexican Social Security Institute (IMSS). The two vertical lines\ndenote the start and end of the wage subsidy program (January and August 2009).\n\n\n**Figure 2: Average employment over time in matched industries (common support)**\n\n\n\n\n\nNote: The figure displays data from the Mexican Social Security Institute (IMSS). The two vertical lines\ndenote the start and end of the wage subsidy program (January and August 2009).\n\n\n18\n\nMention: \"the Mexican Social Security Institute (IMSS)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its registered employment series spans the subsidy programme window.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006648:14:1", "occurrence_id": "1ed64814109808486d09d285", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: , 2009). World Governance Indicators\nvalues are normalized between -2.5 and 2.5 with a higher value being associated with better\ngovernance. The WGI Voice and Accountability indicator captures the quality of interaction\nbetween citizens and government and is particularly relevant for the MENA region, as well as\nbeing instrumental towards promoting a stable environment for production and investment (World\nBank, 2003a). Among all six WGI indicators, the MENA region averages the lowest in this\nmeasure. Voice refers to the voice of the citizens and their freedom of expression, choice in\ngovernment, and ability to demand action from those in power. Accountability refers to the\naccountability of the government and refers to the presence of sanctions whether through voting,\ninstitutional oversight, or oversight from a free media.\n\n\n13\n\nMention: \"World Governance Indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Normalised between minus 2.5 and 2.5, higher values meaning better governance.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006649:13:0", "occurrence_id": "fc7caad95892e03582266087", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: measurements. Ignoring the weights completely would introduce bias, as it would only be approximately\nunbiased if dwellings were identical in size and equidistant.\n\n\nIn addition to the above concerns weight calculations, another potential issue with this group of methods\nis that there are points in the PSU that would not lead to the selection of any households. Consider the\nshaded area of figure 5. If any of these points were selected, the interviewer would not find any household\nbefore she left the boundaries of the PSU. This issue raises questions for the field protocols. Should\ninterviewers stop at the PSU boundary, or should they continue and select housing units outside of the\nselected PSU? If the former, how would the interviewer know where the PSU boundaries are? If the latter,\nthe probabilities should be adjusted for the fact that the _Ai_ region extends outside of the PSU, which is\nnot straightforward. Additional structures outside of the boundaries of the PSU would need to be\nmapped, requiring additional preparation time. For the purposes of this paper, we mapped all households\nin a 50 meter buffer zone around the PSU\nboundaries. This increased the number of\nstructures required from 309 to 408, 68 to 207,\nand 353 to 724, respectively, nearly doubling\nthe required mapping time if manual\ndelineation is used. A third option would be to\nallow interviewers to travel outside of the PSU\nin search of a selected household, but then\nremove these interviewed households outside\nthe selected PSUs from the data set, because\ntheir probabilities of selection are too complex\nto calculate. This approach preserves the\nprobabilities of selection and is easy for the\ninterviewer to implement, but deleting data is\ninefficient in terms of cost.\n\n\n_3.5_ _Random Walk_\n\n\nThere are many different implementations of\nthe random walk procedure, of which each\ninvokes choosing a starting point within the\nselected area and then proceeding along a\npath, selecting every _k_ <sup>_th_</sup> household. The\n\nAfrobarometer survey.\n\nMention: \"Afrobarometer survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Households selected by a skip interval along a path from a starting area.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006650:7:0", "occurrence_id": "949291bd9cc86c2d8dfa51df", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: About 38-43 percent of sub-sovereign bonds had the same rating as sovereign rating and 36-46\n\n\npercent had lower rating than the sovereign rating—adding up to 74-89 percent of sub-sovereign\n\n\nbonds rated _at or below_ the corresponding sovereign rating. This confirms that the majority of\n\n\nsub-sovereign bonds are subject to a sovereign ceiling. The share of sub-sovereign bonds rated\n\n\n_above_ the sovereign rating is relatively small at 11-26 percent of all bonds (12.3 percent for S&P,\n\n\n25.9 percent for Moody’s and 11.0 percent for Fitch), with average distance above the sovereign\n\n\nrating of 2.4 to 4.0 notches on our rating scale.\n\n\n[ _Insert Table 2 around here_ ]\n\n\n**_2.2 Control variables_**\n\n\nWe use a range of global, country-level, and bond-level variables to account for the factors that\n\n\ncan influence the sovereign and sub-sovereign rating relationship. The summary statistics for\n\n\nthese are presented in table 3. The panel of charts in Figure 2 shows the evolution of the annual\n\n\naverage of global and country-level macroeconomic variables since the early 1990s, together\n\n\nwith the average distance between sovereign and sub-sovereign ratings. Pairwise correlations\n\n\nbetween the distance and macroeconomic variables are provided in the chart.\n\n\n_Global and country-level variables:_\n\n\nStandard set of global and macroeconomic factors are included as determinants of the credit risk\n\n\nof sub-sovereign entities. First, the global financial environment affects the performance of sub\n\nsovereign corporations by changing the financing cost for private firms. Riskiness of the global\n\n\nfinancial market is measured by the Chicago Board Options Exchange’s Volatility Index (VIX\n\n\nindex), also known as the Fear Index (Whaley 2009). Higher global risk can narrow the distance\n\n\nas firms may become dependent on the sovereign during times of turmoil in global financial\n\n\nmarkets.\n\n\nThe first chart in figure 2 shows\n\nMention: \"Volatility Index (VIX\n\n\nindex)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Also called the Fear Index, it measures global financial market risk.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006655:5:2", "occurrence_id": "b7b0d36617bb79c21b8b9818", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  of land<br>area|Most recently available (2005)<br>district GDP data  – Planning<br>Commission, GoI<br> <br>Night-time light intensity data used<br>to help generate missing values:<br>satellite data accessed from National<br>Oceanic and Atmospheric<br>Association (NOAA) -<br>http://ngdc.noaa.gov/eog/dmsp.html|\n|Level of<br>urbanization|Together with economic density,<br>affects a district’s potential<br>ability to benefit from<br>agglomeration economies; and<br>the potential propensity to<br>engage in the production of<br>modern tradable goods and<br>services|% of population<br>living in urban areas,<br>2011|Census of India, 2011|\n\n\n\n4\n\nMention: \"district GDP data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The most recently available subnational output estimates at that date.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006655:5:2", "occurrence_id": "b881ad397e44bad703d85d4c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  of land<br>area|Most recently available (2005)<br>district GDP data  – Planning<br>Commission, GoI<br> <br>Night-time light intensity data used<br>to help generate missing values:<br>satellite data accessed from National<br>Oceanic and Atmospheric<br>Association (NOAA) -<br>http://ngdc.noaa.gov/eog/dmsp.html|\n|Level of<br>urbanization|Together with economic density,<br>affects a district’s potential<br>ability to benefit from<br>agglomeration economies; and<br>the potential propensity to<br>engage in the production of<br>modern tradable goods and<br>services|% of population<br>living in urban areas,<br>2011|Census of India, 2011|\n\n\n\n4\n\nMention: \"Census of India\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The urban population share in that year comes from this enumeration.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006654:10:0", "occurrence_id": "c485cc9f0a18e9234bea6dd6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **3. Data and Main Variables**\n\n\nOur main data source is the World Bank’s Women, Business, and the Law database <sup>4</sup> and the World\n\n\nDevelopment Indicators (WDI). This is complemented with other data sources mentioned below.\n\n\nOur main specifications as well as the IV specification take advantage of the panel nature of the\n\n\ndata by including both country and year fixed effects. The sample is an unbalanced panel of up to\n\n\n95 countries for which data are available. <sup>5</sup> [^5: List of countries: Algeria, Angola, Argentina, Armenia, Azerbaijan, Bangladesh, Belarus, Benin, Bolivia, Botswana, Burkina Faso, Burundi,\nCambodia, Cameroon, Chad, Chile, Colombia, Congo, Dem. Rep., Congo, Rep., Costa Rica, Côte d'Ivoire, Denmark, Djibouti, Dominican\nRepublic, Ecuador, Egypt, Arab Rep., El Salvador, Equatorial Guinea, Ethiopia, Fiji, Gabon, Georgia, Ghana, Guatemala, Guinea, Guyana, Haiti,\nHonduras, Indonesia, Iran, Islamic Rep., Ireland, Kazakhstan, Kyrgyz Republic, Lao PDR, Lebanon, Lesotho, Liberia, Madagascar, Malawi,\nMalaysia, Mali, Mauritania, Mauritius, Mexico, Morocco, Namibia, Nepal, Nicaragua, Niger, Nigeria, Norway, Oman, Pakistan, Panama, Papua\nNew Guinea, Paraguay, Peru, Philippines, Poland, Rwanda, Senegal, Sierra Leone, Slovak Republic, Slovenia, South Africa, Sri Lanka, Sudan,\nSuriname, Swaziland, Sweden, Syrian Arab Republic, Tajikistan, Tanzania, Thailand, Togo, Tunisia, Uganda, Ukraine, Uruguay, Uzbekistan,\nVenezuela, RB, Vietnam, Yemen, Rep., Zambia, and Zimbabwe.] The time period covered is 1990 to 2012. The equation\n\n\nwe estimate is as follows:\n\n\nYit = α + βXit + µZit + YFE + vi +uit\n\n\nSubscripts i,t denote the country and\n\nMention: \"Women, Business, and the Law database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The main data source for the legal indicators analysed.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006654:10:0", "occurrence_id": "fcdaa1c2643c97ecceda4d99", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **3. Data and Main Variables**\n\n\nOur main data source is the World Bank’s Women, Business, and the Law database <sup>4</sup> and the World\n\n\nDevelopment Indicators (WDI). This is complemented with other data sources mentioned below.\n\n\nOur main specifications as well as the IV specification take advantage of the panel nature of the\n\n\ndata by including both country and year fixed effects. The sample is an unbalanced panel of up to\n\n\n95 countries for which data are available. <sup>5</sup> [^5: List of countries: Algeria, Angola, Argentina, Armenia, Azerbaijan, Bangladesh, Belarus, Benin, Bolivia, Botswana, Burkina Faso, Burundi,\nCambodia, Cameroon, Chad, Chile, Colombia, Congo, Dem. Rep., Congo, Rep., Costa Rica, Côte d'Ivoire, Denmark, Djibouti, Dominican\nRepublic, Ecuador, Egypt, Arab Rep., El Salvador, Equatorial Guinea, Ethiopia, Fiji, Gabon, Georgia, Ghana, Guatemala, Guinea, Guyana, Haiti,\nHonduras, Indonesia, Iran, Islamic Rep., Ireland, Kazakhstan, Kyrgyz Republic, Lao PDR, Lebanon, Lesotho, Liberia, Madagascar, Malawi,\nMalaysia, Mali, Mauritania, Mauritius, Mexico, Morocco, Namibia, Nepal, Nicaragua, Niger, Nigeria, Norway, Oman, Pakistan, Panama, Papua\nNew Guinea, Paraguay, Peru, Philippines, Poland, Rwanda, Senegal, Sierra Leone, Slovak Republic, Slovenia, South Africa, Sri Lanka, Sudan,\nSuriname, Swaziland, Sweden, Syrian Arab Republic, Tajikistan, Tanzania, Thailand, Togo, Tunisia, Uganda, Ukraine, Uruguay, Uzbekistan,\nVenezuela, RB, Vietnam, Yemen, Rep., Zambia, and Zimbabwe.] The time period covered is 1990 to 2012. The equation\n\n\nwe estimate is as follows:\n\n\nYit = α + βXit + µZit + YFE + vi +uit\n\n\nSubscripts i,t denote the country and\n\nMention: \"World\n\n\nDevelopment Indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Supplies the accompanying development series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006658:10:2", "occurrence_id": "1904fcb8d6e8dc0905f6de54", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  shrank substantially from the 1950s to the early to mid‐1980s but then rose again, with some impact\non the overall income distribution. The estimated impact is also large enough to explain 20‐40 percent of the gap\nobserved during the 1990s between average consumption growth in NSS data and national accounts data.\n\n\n15 URP, or uniform reference period, is collected for a 30‐day recall period for all consumption items. MRP, or\nmixed reference period, records consumption by combining a 30 day recall period for most items and a 365 day\nrecall for items that are purchased with a lower frequency. The MMRP (modified‐mixed reference period)\naggregate introduced in 2009‐10 and 2011‐12 has further closed the gap. The MMRP method which reduces the\nrecall period for some food items to a 7‐day recall (compared to a 30‐day recall in MRP) brings NSSO’s methods\ncloser to the more common practice of a two week or less recall period for food items in other countries.\n\n\n9\n\nMention: \"NSS data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its consumption growth differs from the national accounts by a known gap.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006660:8:0", "occurrence_id": "32824819024e38f53724864c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: associated 134-by-134 matrices of wage bills and remittance flows for the benchmark year. Also,\nthe GMig2 database provides total population headcounts by region of origin and residence.\n\n\nIn this analysis, the model is implemented with the databases aggregated to consider 24 regions\nand countries and seven sectors. <sup>8</sup> [^8: The 24 countries and regions are listed in Table A1. The seven sectors are Agriculture, Oil and Gas Extraction,\nOther Natural Resource Extraction, Food Processing, Other Manufacturing, Electricity and Gas Distribution, and\nOther Services.] The selection of the regional aggregation aims to balance\ntractability with the desire to separately capture major migration corridors, to separate high-income\nregions from other regions, and to separate as far as possible low-income regions from middleincome regions.\n\n\n**2.3** **Scenarios**\n\n\nThe first of three scenarios considered in this analysis is a baseline of the global economy out to\n2030. The baseline population and labor force paths are drawn from the medium variant\nprojections of the United Nations’ World Population Prospects 2015 Revision (United Nations\n2015). The labor productivity parameters are calibrated residually so that the baseline scenario\nreplicates real GDP growth projections from World Bank (2015b) till 2017 and from the OECD\nprojections documented in IIASA (2013) till 2030. The exogenous baseline migrant worker stocks\nby host region and origin are assumed to evolve in proportion to host regions’ given total labor\nforce path, i.e. the 2007 benchmark shares of workers by origin and skill class in each host region’s\ntotal labor force are preserved. In the subsequent scenarios below, all changes are always relative\nto the baseline. For example, an increase reported in one variable means an increase over the\ncorresponding value of the same variable in the baseline.\n\n\nThe second scenario considers a gradual increase in labor migration from low- and middle-income\n(LMI) regions to high-income countries (HIC), in the spirit of the exercise of Rodrik (2001),\nWalmsley and Winters (\n\nMention: \"GMig2 database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Wage bill and remittance matrices plus population headcounts by region.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006660:8:0", "occurrence_id": "3f92534cce28ac8c9b847afa", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: associated 134-by-134 matrices of wage bills and remittance flows for the benchmark year. Also,\nthe GMig2 database provides total population headcounts by region of origin and residence.\n\n\nIn this analysis, the model is implemented with the databases aggregated to consider 24 regions\nand countries and seven sectors. <sup>8</sup> [^8: The 24 countries and regions are listed in Table A1. The seven sectors are Agriculture, Oil and Gas Extraction,\nOther Natural Resource Extraction, Food Processing, Other Manufacturing, Electricity and Gas Distribution, and\nOther Services.] The selection of the regional aggregation aims to balance\ntractability with the desire to separately capture major migration corridors, to separate high-income\nregions from other regions, and to separate as far as possible low-income regions from middleincome regions.\n\n\n**2.3** **Scenarios**\n\n\nThe first of three scenarios considered in this analysis is a baseline of the global economy out to\n2030. The baseline population and labor force paths are drawn from the medium variant\nprojections of the United Nations’ World Population Prospects 2015 Revision (United Nations\n2015). The labor productivity parameters are calibrated residually so that the baseline scenario\nreplicates real GDP growth projections from World Bank (2015b) till 2017 and from the OECD\nprojections documented in IIASA (2013) till 2030. The exogenous baseline migrant worker stocks\nby host region and origin are assumed to evolve in proportion to host regions’ given total labor\nforce path, i.e. the 2007 benchmark shares of workers by origin and skill class in each host region’s\ntotal labor force are preserved. In the subsequent scenarios below, all changes are always relative\nto the baseline. For example, an increase reported in one variable means an increase over the\ncorresponding value of the same variable in the baseline.\n\n\nThe second scenario considers a gradual increase in labor migration from low- and middle-income\n(LMI) regions to high-income countries (HIC), in the spirit of the exercise of Rodrik (2001),\nWalmsley and Winters (\n\nMention: \"World Population Prospects 2015 Revision\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Medium variant projections supply the population and labour force paths.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006662:3:0", "occurrence_id": "130f86b8888fb3a7128a53b7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: #### **1 Introduction**\n\nVietnam participated in the Programme for International Student Assessment (PISA)\n\nfor the first time in 2012 and its performance has been much higher than other developing\n\ncountries that take part in this OECD led initiative. PISA scores of 15 year-olds in Mathe\nmatics, Reading and Science are calibrated to an OECD mean of 500 and standard deviation\n\nof 100 points. Only a few developing countries take part in PISA, perhaps because most of\n\nthem have results much lower than the OECD countries. In the OECD-PISA 2012 database,\n\nthere are seven countries other than Vietnam with a per capita GDP (in 2010 PPP dollars)\n\nbelow US$ 10,000 - Albania, Colombia, Indonesia, Jordan, Peru, Thailand and Tunisia. At\n\nUS$ 4,098, Vietnam’s GDP per capita is the lowest of this group. Figure 1 indicates a posi\ntive, albeit non-linear correlation between GDP per capita and PISA test scores. Vietnam,\n\nrepresented by a red star, lies much above the other developing countries clustered in the\n\nlower left hand corner of Figure 1. With a mathematics mean score of 511, Vietnam is more\n\naligned to Finland (519) and Switzerland (531), rather than Peru (368) and Colombia (376).\n\n\n**Figure** **1:** **PISA** **2012** **results** **compared** **with** **GDP** **per** **capita**\n\n\n\n\n\n\n|Col1|Shanghai−China<br>Finland<br>Vietnam (511) (519) Switzerland<br>(531)<br>Colombia (376)<br>Peru (368)|Col3|Col4|Col5|Col6|Col7|Col8|\n|---|---|---|---|\n\nMention: \"OECD-PISA 2012 database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Seven lower-income countries with comparable achievement sit alongside it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006667:37:0", "occurrence_id": "b6124028f7886c2303d0677e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: economic and fiscal policy. In Markandya, A., Sainz de Murieta, E. Galarraga, I.\n\n[eds] Handbook on the Economics of Adaptation. Routledge: 417‐436\nMusgrave, R.A. (1959). The theory of public finance. New York: McGraw Hill.\nNoy, I. (2009). The macroeconomic consequences of disasters. _Journal of Development_\n\n_Economics_ 88(2): 221‐231..\nODI&Worldbank (2015). Unlocking the Triple Dividend of DRM. Briefing Note for\n\nWorld Conference on Disaster Risk Reduction.\nOECD (2012). Disaster Risk Assessment and Risk Financing. A G20/OECD\n\nmethodological framework. OECD.\nPhaup, M. and C. Kirschner (2010). Budgeting for Disasters: Focusing on the Good\n\nTimes. _OECD Journal on Budgeting_ 1: 1‐24.\nPriest, G.L (2003). Government Insurance versus Market Insurance. _The Geneva_\n\n_Papers on Risk and Insurance_ 28: 71‐80.\nSchick, A., Polackova Brixi, H. (Eds.). (2002). Government at risk. Washington, DC:\n\nWorld Bank and Oxford University Press.\nScrieciu, S. Chalabi, Z., Belton, V. Mechler, R. (2014). Advancing Methodological\n\nThinking and Practice for Development‐Compatible Climate Policy Planning.\n_Mitigation and Adaptation Strategies for Global Change_ doi.org/10.1007/s11027‐\n013‐9538‐z.\nSurminski, S. (2014). Decision‐making tools. Concept note. Project GFDRR&ODI\n\nReport ‘Realising the resilience dividend: A new business case for disaster risk\nmanagement’. GFDRR&ODI.\nUK Cabinet Office (2015). National Risk Register of Civil Emergencies\n\nMention: \"National Risk Register of Civil Emergencies\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"A register of national emergency risks cited for the business case.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006673:13:0", "occurrence_id": "35c4aa7a5ff5888f8fe96c52", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 8. Estimated coefficient for the dummy indicator of urban SSA by food headings**\n_Note:_ Reported are coefficient estimates for the SSA dummy in column (4) of Table 3.\n_Source:_ Authors’ calculations are based on the 2011 ICP data.\n\n\nAlthough results for the categories of household consumption except for housing rent, household\nconsumption except for housing rent and food, and food would be more robust than the ones for smaller\ncomponents, we also examined other groups of items. Our estimation for housing rent is based on nonadjusted PLIs for countries that collected price data predominantly in urban areas because national-tocapital city price ratios for housing rent are not available. The estimation result shows that housing rent is\n55% more expensive in urban SSA (Table 6). <sup>10</sup> [^10: In its broadest category in the ICP data, the PLI of housing includes not only household expenditure on actual and\nimputed housing rent, but also maintenance and repair of the dwelling and utilities (see Appendix C). We focus on\nonly PLI for housing rent in the present paper.] Other groups of relatively expensive items include\ntransport (42%), communication (46%), and restaurants and hotels (41%). Transport items include\nhousehold expenditure on purchase of vehicles, operation of personal transport equipment, and transport\nservices (see Appendix C). Figure 9 summarizes our results.\n\n\n**<u>Table 6. Estimated coefficient for the indicator of urban SSA for non-food expenditures</u>**\n\n\n\n\n\n|Col1|Housing rent|Transport|Communication|Restaurants & hotels|\n|---|---|---|---|---|\n|Coefficient estimate<br>for SSA dummy|0.546*** <br>(0.188)|0.424** <br>(0.183)|0.457* <br>(0.225\n\nMention: \"2011 ICP data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Household consumption categories are analysed against this comparison programme's prices.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006675:24:6", "occurrence_id": "965602901472d7f502b9084c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  in formal child\ncare arrangement -0.001\n<mark>(0.002)</mark>\n\n<mark>Rounds fixed effects</mark> <mark>Yes</mark> <mark>Yes</mark> <mark>Yes</mark> <mark>Yes</mark> <mark>Yes</mark>\n<mark>Settlement type fixed effects</mark> <mark>Yes</mark> <mark>Yes</mark> <mark>Yes</mark> <mark>Yes</mark> <mark>Yes</mark>\n<mark>Regions fixed effects</mark> <mark>Yes</mark> <mark>Yes</mark> <mark>Yes</mark> <mark>Yes</mark> <mark>Yes</mark>\n\n<mark>Pseudo R2</mark> 0.0851 0.0970 0.116 0.125 0.144\n<u><mark>N</mark></u> <u><mark>3,304</mark></u> <u><mark>2,408</mark></u> <u><mark>2,376</mark></u> <u><mark>2,242</mark></u> <u><mark>1,810</mark></u>\nSource: Russia Longitudinal Monitoring Survey (RLMS), 2004-2012 waves.\n**Note:** Marginal effects are reported, standard errors in parentheses; standard errors clustered by individual.\n*** p<0.01, ** p<0.05, * p<0.1\n\n\n23\n\nMention: \"Russia Longitudinal Monitoring Survey (RLMS), 2004-2012 waves\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Marginal effects across these waves are reported in the table.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006676:18:0", "occurrence_id": "d5f248cc187133d468f95b0e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Education Development Index (IDEB). IDEB was created in 2007 by INEP and compiles\nin a single index the average scores obtained by students on standardized evaluations and\nthe school flow. <sup>41</sup> [^41: School flow is the progression of students in school grades at any given level of education. In IDEB it is\nmeasured by pass rates.] The index is calculated using data from the school census and _Prova_\n_Brasil,_ <sup>42</sup> [^42: For schools and municipalities, IDEB uses _Prova Brasil_, but at the state level and nationwide SAEB is\nused instead.] and it is available at the school, municipal, and state level. This variable is used\nat the school level, for the years 2005, 2007, 2009, and 2011. Finally, we look at dropout\nrates at school level, provided by INEP; these data are available from 2007 to 2011. In\nour analysis, we focus on dropout rates in the first and second cycles of primary schools\n(first to fifth grade and sixth to ninth grades, respectively).\n\nOur empirical analysis also looks at the heterogeneity of impacts across a number\nof program characteristics, such as _Mais Educação_ coverage at the school level\n(percentage of students benefiting from the extended day), spending per student, and main\nthematic areas of activities available in the program. We compute spending per student\nusing administrative data from MEC on the total financial transfer to the school divided\nby the number of students in the school participating in the program. Schools’ thematic\nareas are also provided by MEC/FNDE. The program’s coverage is calculated by dividing\nthe total number of students enrolled in _Mais Educação_ in a given year, gathered from\nMEC, by the total number of enrollments in school census in the previous year. <sup>43</sup> [^43: Reported number of enrollments in schools that joined _Mais Educação_ considers the school census of the\nprevious year.]\n\n\nWe\n\nMention: \"INEP\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The institute that created the education development index.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006676:18:0", "occurrence_id": "9224fe03d8fe2883a9f58b7d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Education Development Index (IDEB). IDEB was created in 2007 by INEP and compiles\nin a single index the average scores obtained by students on standardized evaluations and\nthe school flow. <sup>41</sup> [^41: School flow is the progression of students in school grades at any given level of education. In IDEB it is\nmeasured by pass rates.] The index is calculated using data from the school census and _Prova_\n_Brasil,_ <sup>42</sup> [^42: For schools and municipalities, IDEB uses _Prova Brasil_, but at the state level and nationwide SAEB is\nused instead.] and it is available at the school, municipal, and state level. This variable is used\nat the school level, for the years 2005, 2007, 2009, and 2011. Finally, we look at dropout\nrates at school level, provided by INEP; these data are available from 2007 to 2011. In\nour analysis, we focus on dropout rates in the first and second cycles of primary schools\n(first to fifth grade and sixth to ninth grades, respectively).\n\nOur empirical analysis also looks at the heterogeneity of impacts across a number\nof program characteristics, such as _Mais Educação_ coverage at the school level\n(percentage of students benefiting from the extended day), spending per student, and main\nthematic areas of activities available in the program. We compute spending per student\nusing administrative data from MEC on the total financial transfer to the school divided\nby the number of students in the school participating in the program. Schools’ thematic\nareas are also provided by MEC/FNDE. The program’s coverage is calculated by dividing\nthe total number of students enrolled in _Mais Educação_ in a given year, gathered from\nMEC, by the total number of enrollments in school census in the previous year. <sup>43</sup> [^43: Reported number of enrollments in schools that joined _Mais Educação_ considers the school census of the\nprevious year.]\n\n\nWe\n\nMention: \"school census\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"One of two inputs to the index alongside the assessment.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006676:18:0", "occurrence_id": "664abfa69327365a320bab07", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Education Development Index (IDEB). IDEB was created in 2007 by INEP and compiles\nin a single index the average scores obtained by students on standardized evaluations and\nthe school flow. <sup>41</sup> [^41: School flow is the progression of students in school grades at any given level of education. In IDEB it is\nmeasured by pass rates.] The index is calculated using data from the school census and _Prova_\n_Brasil,_ <sup>42</sup> [^42: For schools and municipalities, IDEB uses _Prova Brasil_, but at the state level and nationwide SAEB is\nused instead.] and it is available at the school, municipal, and state level. This variable is used\nat the school level, for the years 2005, 2007, 2009, and 2011. Finally, we look at dropout\nrates at school level, provided by INEP; these data are available from 2007 to 2011. In\nour analysis, we focus on dropout rates in the first and second cycles of primary schools\n(first to fifth grade and sixth to ninth grades, respectively).\n\nOur empirical analysis also looks at the heterogeneity of impacts across a number\nof program characteristics, such as _Mais Educação_ coverage at the school level\n(percentage of students benefiting from the extended day), spending per student, and main\nthematic areas of activities available in the program. We compute spending per student\nusing administrative data from MEC on the total financial transfer to the school divided\nby the number of students in the school participating in the program. Schools’ thematic\nareas are also provided by MEC/FNDE. The program’s coverage is calculated by dividing\nthe total number of students enrolled in _Mais Educação_ in a given year, gathered from\nMEC, by the total number of enrollments in school census in the previous year. <sup>43</sup> [^43: Reported number of enrollments in schools that joined _Mais Educação_ considers the school census of the\nprevious year.]\n\n\nWe\n\nMention: \"_Prova_\n_Brasil\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The student assessment component of the index.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006676:18:0", "occurrence_id": "99b0196759265cec817ce905", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Education Development Index (IDEB). IDEB was created in 2007 by INEP and compiles\nin a single index the average scores obtained by students on standardized evaluations and\nthe school flow. <sup>41</sup> [^41: School flow is the progression of students in school grades at any given level of education. In IDEB it is\nmeasured by pass rates.] The index is calculated using data from the school census and _Prova_\n_Brasil,_ <sup>42</sup> [^42: For schools and municipalities, IDEB uses _Prova Brasil_, but at the state level and nationwide SAEB is\nused instead.] and it is available at the school, municipal, and state level. This variable is used\nat the school level, for the years 2005, 2007, 2009, and 2011. Finally, we look at dropout\nrates at school level, provided by INEP; these data are available from 2007 to 2011. In\nour analysis, we focus on dropout rates in the first and second cycles of primary schools\n(first to fifth grade and sixth to ninth grades, respectively).\n\nOur empirical analysis also looks at the heterogeneity of impacts across a number\nof program characteristics, such as _Mais Educação_ coverage at the school level\n(percentage of students benefiting from the extended day), spending per student, and main\nthematic areas of activities available in the program. We compute spending per student\nusing administrative data from MEC on the total financial transfer to the school divided\nby the number of students in the school participating in the program. Schools’ thematic\nareas are also provided by MEC/FNDE. The program’s coverage is calculated by dividing\nthe total number of students enrolled in _Mais Educação_ in a given year, gathered from\nMEC, by the total number of enrollments in school census in the previous year. <sup>43</sup> [^43: Reported number of enrollments in schools that joined _Mais Educação_ considers the school census of the\nprevious year.]\n\n\nWe\n\nMention: \"administrative data from MEC\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Total financial transfer per school divided by enrolment.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006682:21:0", "occurrence_id": "b2cf095d963140f069fdc34e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: jobs per acre in Chicago’s CBD in 1956 (McDonald and McMillen, 1990). Table 4b presents analogous\nresults for the 2002 UBR census, where a separate GWR estimation was done to identify these subcenters. <sup>33</sup>\n\n\nTables 4a and 4b suggest that over the last decade the share of CBD in total employment has increased from\n14% to 18% while that in firm count has increased from 19.5% to 20.5%. The share of CBD in total\nemployment is similar to that witnessed by large cities in the US during 1980s. However, since 1960s,\nmetropolitan cities around the developed world are witnessing a decline in the importance of CBD\n(Kenworthy and Laube, 1999; Angel and Blei, 2015a). In contrast to large American cities where\nemployment subcenters contribute to about 15% of employment in 2000, the contribution of Kampala’s\nsubcenters to employment and establishment counts is abysmally small. Potential subcenters in Kampala\naccount for only about 2.4% of employment and 2% of firms. <sup>34</sup> The remaining employment in Kampala,\nwhich is about 80% in 2011, is rather dispersed all across the city. This confirms our preliminary finding\nusing the monocentric model that Kampala is characterized by a mixed land use pattern. <sup>35</sup>\n\n\nIn conclusion, our work on ascertaining the urban spatial structure in Kampala suggests that the city has a\nconcentrated business center, however, beyond the 3 km radius, employment is spatially dispersed. The\nsmall contribution to aggregate employment by the identified subcenters also confirms our intuition that\nKampala is characterized by a mixed land use pattern. At this level of development and income, we do not\nexpect a city like Kampala to have multiple centers of economic activity. However, the degree of\nmonocentricity in Kampala\n\nMention: \"the 2002 UBR census\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Separate geographically weighted estimates identify subcentres here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006683:12:1", "occurrence_id": "147ff7162cb1eec185212b7d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  month, a level deemed to represent “gainful” employment.\nInterestingly, about 15 percent of the sample was already engaged in the same trade for which training they\napplied (denoted as “trade-specific IGA”), indicating that a significant minority of applicants were looking\nto upgrade their existing skills. Though they are not older than the men in the sample, women are more\nlikely to be married and have a child, and have lower employment and earnings at baseline. <sup>31</sup> [^31: More details about the characteristics of the study population are found in a separate baseline report, available upon request\nfrom the authors.]\n\nAs discussed in Section II, the EF provides financial incentives for T&E providers to recruit and train people\nfrom the Dalit and Janajati ethnic groups. For Janajatis, the T&E providers appear to respond to these\nincentives, as 44 percent of the applicants are Janajatis, and they are statistically more likely to be in the\ntreatment group than control. <sup>32</sup> [^32: For reference, the 2001 census indicated that Janajatis made up 37% and Dalits made up 13% of the total Nepali population.] The T&E providers appear to have less success with attracting Dalit\napplicants: only 8 percent of the applicants are Dalits (a bit less than the population average), and they are\nequally divided between the treatment and control groups.\n\n\n**b.** **Estimating the EF Training Program Effects**\nTo estimate causal effects of the EF training program on various outcomes, first we employ a “difference\nin difference” (DID) technique. We will refer to this as the OLS specification in the remainder of this report.\nThe main equation we estimate is:\n\n\n29 The survey team deployed a fixed number of staff to each district center, with a schedule of all events in that district. Events\nconducted by high-capacity T&E providers and in popular trades were more likely to\n\nMention: \"the 2001 census\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its shares for two population groups give the reference population.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006696:24:1", "occurrence_id": "4809b97d233bdcd5eb854d63", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: /u> <u>men</u> <u>15-24</u> <u>upper60</u>\n\nSouth Africa 34.1 0.52 0.70 0.32 0.74\nBotswana 29.0 0.55 0.81 0.23 0.51\nKenya 26.3 1.02 0.57 0.54 0.46\nNigeria 18.4 0.69 0.59 0.23 0.40\nNamibia 16.2 0.15 0.76 0.37 0.32\nCameroon 14.1 0.11 1.09 0.48 0.48\nGhana 12.7 0.61 0.48 0.44 0.29\nMozambique 11.1 0.13 0.74 0.27 0.24\nUganda 7.9 0.38 0.26 1.92 0.28\nRwanda 6.0 0.24 0.76 0.58 0.15\nTanzania 3.5 0.20 1.04 0.60 0.22\nEthiopia 2.7 0.14 0.28 0.54 0.23\n\n\n\n<u>15-24</u>\n\n\n\n<mark>Source: Research ICT Africa surveys (various years). See World Bank (2016). Ratios refer to percentages, e.g.,</mark>\n\n\n<mark>percent rural / percent urban.</mark>\n\n\n<mark>Second, studies reviewed in this paper suggest that basic price and market information systems</mark>\n\n\n<mark>can improve efficiency and welfare. The evidence, though strong, is still limited to certain</mark>\n\n\n<mark>countries and in certain contexts. A number of recent studies have cast doubt on the overall</mark>\n\n\n23\n\nMention: \"ICT Africa surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Ratios across age bands and countries are computed across these rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006699:13:1", "occurrence_id": "ffd9b398e8fccc338f07e3a0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  and large with\n100 or more employees.\n\n\nThe survey included 424 enterprises stratified by sector, region and size, with between 40 and 76\nenterprises in each sector. About half the sample (224 firms) is located in Dar es Salaam and the\nbalance divided rather evenly across each of the remaining regions (each with about 50 firms).\nOverall, small firms make up 63 percent of the sample, medium-size firms 26 percent and large\nfirms 11 percent. However, this firm-size distribution varies considerably across sectors and\nregions. Small firms are mostly found in textiles and garments, fabricated metals, furniture and\nIT, while large firms are more prevalent in food, construction, and transportation. Mbeya and\nZanzibar are dominated by small and medium firms while Dar es Salaam and Arusha have most\nlarge firms.\n\n\nTESS builds on the recently completed 2013 Enterprise Survey for Tanzania, retaining selected\nkey variables on firm characteristics, firm performance and productivity, and adding an expanded\nmodule of questions on employment and skills. It elicited qualitative and quantitative information\nfrom firms about the education and occupational composition of their workforce, their perceptions\nabout skill gaps in education as well as different cognitive, non-cognitive and technical skills and\nthe constraints these pose for their operations, their strategies for addressing skill gaps including\nthe hiring of new workers, use of foreign workers and outsourcing skills, and their in-house and\n\n\n12\n\nMention: \"2013 Enterprise Survey for Tanzania\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The panel builds on it, retaining firm characteristics and performance variables.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006703:6:1", "occurrence_id": "54b71d17ccd9b27cf8a20e7b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: , date of birth, and\n\n\nmaximum schooling level achieved. Only full-time wage earners with complete information\n\n\non demographics and job attributes were kept for the analysis. <sup>2</sup>\n\n\nUsing the unique firm identifiers, we have merged these data with customs records on\n\n\nexport transactions collected by _Secretaria do Com´ercio Externo_ (SECEX) of the Ministry\n\n\nsee Greenaway and Nelson (2000), Bastos and Silva (2009), Falvey, Greenaway and Silva (2008, 2010) and\n\n\nthe references therein.\n\n2In particular, we kept for analysis employees working more than 20 hours per week, aged between 16\n\n\nand 65, and earning at least the national minimum wage.\n\n\n5\n\nMention: \"customs records on\n\n\nexport transactions\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Merged to the firm data through unique identifiers.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006704:8:1", "occurrence_id": "43a94d7db60e2079bc1faac8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 32.7%) compared\nto New York (22.7%) and Los Angeles (27.8%) (HUD 2013; US Census Bureau Fact Finder), which further\ncomplicates the picture of privilege and marginalization along the line of race.\n\nWhy do such statistics matter for understanding the impacts of climate change on poverty? They matter\nbecause, in comparison with burgeoning case studies and comparative analyses on poor people in poor\ncountries, the climate change literature and debates on poor people in developed countries and their\nexperiences with a changing climate have been astoundingly mute. The poor in upper MICs (e.g. China,\nBrazil, South Africa) and HICs (e.g. OECD countries) have been largely invisible, insufficiently accounted\nand represented in international poverty metrics and in‐depth climate change inquiries. The same is true\nfor growing inequalities behind cloaked processes of impoverishment (‘falling into poverty’) and\nmarginalization. Even progressive poverty measures, such as the Multidimensional Poverty Index (UNDP,\n2010) do not capture poverty in HICs. In fact, Chapter 13 on Livelihoods and Poverty in the AR5 noted\nthe limited research on climate change impacts on poor people in more affluent nations as a major gap\nin knowledge (Olsson et al. 2014a). The exception stems from extreme events, especially hurricanes and\nheat waves, during which the ugly face of western/northern poverty all of a sudden becomes shockingly\nvisible. Vignette 1 exposes differential merit in protecting socially vulnerable versus affluent people and\nregions against sea level rise in the United States. A closer look at the 1995 Chicago and 2003 European\nheat waves, provided in Vignette 4, reveals disproportionate and premature death among poor,\ndisadvantaged, and socially isolated segments in society.\n\nOne of the prime reasons why poverty and inequality in MICs and HICs have been understudied is\nrelated to the narratives that encapsulate the essence of ‘the poor’ in affluent nations. In the United\n\nMention: \"US Census Bureau Fact Finder\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Metropolitan shares of an underserved group come from this tool.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006705:18:0", "occurrence_id": "07cfdf479f82a30e19236d9f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 4: Likelihood and intensity of social support from children.\n\n\n**3.1.6.Financial transfers from children to parents**\n\n\nAs noted in previous literature, the upward flow of economic resources within European families is almost\nnegligible. In general less than 2% of respondents in the sample have received financial support from a child.\nThere are a few exceptions. In two countries, Czech Republic and Greece the quota of elderly people receiving\nfinancial help is above 7% - possibly a consequence of weaker public pension provision. In 6 other countries\n\n- Poland, Austria, Estonia, Germany, Italy and Spain – between 3 and 5% of the respondents have received\n\neconomic support from their children.\n\n\n**3.1.7.Intergenerational coresidence**\n\n\nBesides exchanging support while living in separate households, family generations can adopt a further\nstrategy to support each other: co-residence. As it has been shown in previous research, parents support their\nchildren also by letting them to prolong their co-residence in the parental home when they are adult and already\n\nworking, or letting them come back after a divorce or when becoming unemployed. On the other hand, children\nmight support their elderly parents by co-residing with them when they become frail and need assistance. So\nfar, when using the SHARE data, it has proven difficult to separate the two different motives for co-residence\nand, thus, answer the question: who is helping whom through co-residence? Some analyses, though, indicates\nthat the needs of the children are more strongly correlated with intergenerational co-residence that the needs\nof the parents (see for instance Albertini and Kohli 2013). This result suggests that it is more frequently the\ncase that co-residence is a strategy adopted by parents to support their adult children than the other way around.\nMore research is needed, however, to clearly sort out the main reasons behind intergenerational co-residence,\nalso taking into consideration external constraints\n\nMention: \"SHARE data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Its measures make separating two co-residence motives difficult.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006708:9:0", "occurrence_id": "7eccb8e992fa271a6c284a70", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: abolish slavery. Since then, however, with the exception of the 30-year period of autocratic rule\n(1957-1986) under Francois Duvalier (Papa Doc) and his son Jean-Claude Duvalier (Baby Doc),\nHaiti has weathered a succession of short-lived governments. The country has struggled with little\nsuccess to develop the institutional mechanisms and policy fundamentals essential to the\ndevelopment progress.\n\n\n**Figure 1 : Annual GDP Growth vs. Changes in Government, 1971-2013**\n\n\nBetween 1986 and 2014, the country had 18 changes of government (Table 1). In addition,\naccording to the Cross-National Time Series data archive, there were 20 major cabinet changes\nfrom 1986 to 2006 (meaning a replacement of the prime minister or 50 percent of the cabinet\nmembers). Empirical investigations have shown that such cabinet changes are detrimental to\ngrowth (Aisen and Veiga, 2013). These studies estimate that Haiti would have grown by 1.2\npercentage points faster if it had achieved an average level of political stability. <sup>9</sup> In this regard, the\npast decade has been comparatively stable. With security provided by a large UN military presence\n(the United National Stabilization Mission in Haiti—MINUSTAH), two presidents have been\nchosen by election. Stability remains fragile, nevertheless, with frequent changes in government\nand repeated postponements of elections.\n\n9 Aisen and Veiga (2013) examine 169 countries. On average over the period 1980 to 2004, Haiti has seen almost\none cabinet change every year (0.95), while the overall mean in the sample was 0.47. Moving to the sample mean of\npolitical stability, Haiti’s growth rate would have increased by 1.16 percentage points. Calculations from an\naccompanying paper by Antoine et al. (2014).\n\n\n8\n\nMention: \"Cross-National Time Series data archive\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Cabinet changes across that span are counted from this archive.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006718:10:1", "occurrence_id": "6cf809b5e3692fb3f6298772", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 4), the data suggest that people with extreme views tend\n\n\nto live in rural areas (Figures 5), tend to be poor (Figure 6), and predominantly have elementary\n\n\neducation or less (Figure 6). A more detailed breakdown by residence reveals that these people\n\n\ntend to live in small towns and villages (Figure 7). It is striking that all people with extreme\n\n\n7 The number of observations in each figure varies depending on the data available for the combination of variables\nincluded in it and the sample includes all countries in the Gallup World Poll for which there is information on the\nradicalization question (Q1).\n\n\n9\n\nMention: \"the Gallup World Poll\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"All countries with the radicalization question are included.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006720:27:0", "occurrence_id": "aadfd388c548455bc345e1bb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 4**\nSaving for old age around the world and age\n\n\n\n55%\n\n\n50%\n\n\n45%\n\n\n40%\n\n\n35%\n\n\n30%\n\n\n25%\n\n\n20%\n\n\n15%\n\n\n10%\n\n\n5%\n\n\n0%\n\n\n\nWorld East Asia &\n\nPacific\n\n\n\nEurope & Central\n\nAsia\n\n\n\nHigh Income:\n\nOECD\n\n\n\nLatin America &\n\nCaribbean\n\n\n\nMiddle East &\n\nNorth Africa\n\n\n\nSouth Asia Sub-Saharan\n\nAfrica\n\n\n\n<u>Notes: Notes: Weighted averages from the Global FINDEX (2015) data</u>\n\n\n26\n\nMention: \"FINDEX (2015) data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Weighted regional averages of financial access come from this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006731:27:1", "occurrence_id": "88f35022d9786468878c488f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: u>0.136</u> <u>0.702</u>\n<u>Total</u> <u>1</u> <u>0.051</u> <u>0.026</u> <u>0.512</u>\n\n**<u>Panel B. By Broad Source</u>**\n\n\n\nShare of total\n\nimport value\n\n\n\nIHC firms’ share\n\nof source region\n\n\n\nimport value\n<u>cleared under IHC</u>\n\n\n\nShare of total\n\n\n\nIHC share of\n\nIHC firm\n\n\n\n<u>import value</u> <u>cleared under IHC</u> <u>imports</u>\n\nEuropean Union 0.518 0.076 0.039 0.514\n<u>Other</u> <u>0.482</u> <u>0.025</u> <u>0.013</u> <u>0.506</u>\n<u>Total</u> <u>1</u> <u>0.051</u> <u>0.026</u> <u>0.512</u>\nSource: Authors’ calculations based on Serbian customs data.\n\n\n\n<u>import value</u>\n\n\n\nWe now turn to an analysis of 2010 data in order to show that even before the IHC program\n\n\nwas undertaken, IHC firms and non-IHC firms differed in the distribution of customs clearance\n\n\ntimes, in their inspection rates, and in the scale of their imports. Table 3 shows summary statistics\n\n\non customs clearance times in 2010 for firms that do and do not adopt the IHC program by the end\n\n\nof 2013. The striking pattern that emerges is that the distribution of clearance times for firms that\n\n\nadopt the IHC program by 2013 is well below that of firms that do not adopt, at all\n\nMention: \"Serbian customs data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Import values across sectors are calculated from these records.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006729:2:0", "occurrence_id": "d9619b5827225ae26d0bc5ab", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: #### **Using Administrative Data to Assess the Impact and Sustainability of** **Rwanda’s Land Tenure Regularization ¶**\n\nDaniel Ayalew Ali, Klaus Deininger, Marguerite Duponchel\n\n\nWorld Bank, Washington DC\n<u>dali1@worldbank.org,</u> <u>kdeininger@worldbank.org, mduponchel@worldbank.org</u>\n\n\nJEL Classification: O18, Q15, R52\n\n\nKeywords: land tenure regularization, land administration, transaction registration, mortgages, registration fees, Rwanda\n\n\n- We thank the Rwanda National Resource Authority in Kigali, especially Emmanuel Nkurunziza, Grace Nishimwe and Rhona Nyakurama for\ntheir support of the project helpful comments, and data access and Solomon Kyewalabye and Thierry Ngoga for helping us understand the LAIS\ndata and the LTR process. Funding support from the UK’s Department for International Development is gratefully acknowledged.\n\nMention: \"LAIS\ndata\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"inform\", \"summary\": \"Acknowledged help in understanding these records and the titling process.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006750:14:1", "occurrence_id": "c38ae53d4c59c7c53e504b28", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  year.\n\n\nColumn 4 reports the cash transfers’ impact on unconditional school attendance. Given that\n\n\nattendance conditional on enrolment is very high (98.3%) since once enrolled children tend to\n\n\nattend classes, our approach consists of estimating the impact on average school attendance rates\n\n\nof all school age children, unconditional on their enrollment. This is a broad measure of school\n\n\nparticipation with direct policy relevance that accounts for the enrollment and attendance effects\n\n\nand is not confounded by changes in the share of the sample enrolled. We rely on a measure\n\ntaken from school ledgers collected at the school level <sup>17</sup> : for each child, we compute the\n\n\n16 A recent cash transfer program aimed at improving secondary school enrollment in Macedonia randomly gave\nconditional cash transfers to mothers or fathers, and evaluation results also found that specifically targeting mothers\ndid not significantly improve school enrollment (Armand 2014).\n17 We also collected parental self-reports for attendance but the recall period was the last two weeks and since some\nvillages were surveyed after the end of the school year, we consider this measure to be less reliable.\n\n\n13\n\nMention: \"school ledgers\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Enrolment of each child is computed from these school-level registers.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006765:5:0", "occurrence_id": "30f030ddc90626c7d7a40581", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: In modeling the impact of the ECU adopting a CET that mirrors Russia’s WTO commitments through\n2020, it was assumed that those commitments would be fully implemented <sup>4</sup> [^4: There have been concerns that introducing compound tariffs into the CET, adding a specific duty to ad valorem rates, may exceed Russia’s\nbound rates in products such as large-capacity refrigerators. The EU Market Access Database (MAD) also suggests that a total of 1,576 product\nlines have been “corrected” by Russia (abolished or introduced) relative to its 2007 tariff schedule, including tariff sub-codes (presumably\nnational tariff lines below the HS-6 level). This could be characterized as “manipulation” to avoid implementing the tariff commitments. EU\nMAD also states that Russia’s import duties on coated paper and cardboard in the Single Customs Tariff were three times higher than its WTO\ncommitment (15 percent instead of 5 percent).] ; for example, if the current\nCET for a product is higher than Russia's bound commitment for any given year, the applied CET would\nfall to the lower bound rate. Conversely, if the current CET were below the bound rate, presumably there\nwould be no change in the tariff.\n\nThe results suggest that full acceptance of Russia’s WTO tariff bindings through 2020 would increase\ngoods imports by about $3.5 billion (1.1 percent of total imports), measured against a 2012 baseline.\nLower prices for these imports as a result of tariff cuts in affected sectors would benefit Russian\nconsumers by some $371.39 million. The largest increases in import values are estimated for electrical\nand electronic equipment ($592 million, 1.7 percent increase); pharmaceutical products ($365 million, 2.7\npercent); meats ($248 million, 3.4 percent); plastics and plastic products ($238 million, 2.1 percent);\nmachinery ($151 million, 0.3 percent);\n\nMention: \"EU Market Access Database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It reports product lines corrected by the accession country.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006769:4:1", "occurrence_id": "846593fe2c58b354144efdff", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  ethnic networks can be used to reduce employment uncertainty and make receipts of\nfuture remittances more stable and secure via migrants’ marriages at destination.\n\nDe Haan (2007) highlights a lack of studies on the effects of migration on development, inequality,\nand poverty despite increasing population mobility. Albeit quantitative studies have often been\nhindered by lack of adequate data to characterize the migration process (De Brauw et al., 2014)\nand by the inherent impossibility of observing an individual in the two states (migrant and nonmigrant), a number of studies have supported the positive impact of migration on living conditions.\nLall et al. (2006) conclude that migration can be a beneficial phenomenon and migration\nrestrictions should be avoided. Other recent studies, such as Lokshin et al. (2010), find that workrelated migration, both international and domestic, has a positive effect on welfare and reduces\npoverty in Nepal. Beegle et al. (2011) use a 13-year panel survey in Tanzania and show that\nmigration between 1991 and 2004 increased consumption growth by 36 percentage points.\n\n\nComing to the impact of migration on the welfare of Ghanaian households, a series of analyses\npublished by the Ghana Statistical Service in 1995 and based on the 1991 Ghana Migration Survey\nfind that migration induces a decline in household labor supply that is however compensated by\nadditional effort of household members left behind, indicating no loss of output in the short run in\nover half of the interviewed households and an expectation of no drop in the long run (Tutu, 1995).\nThese studies also point to remittances as an important element in origin rural communities to\nsupport the welfare of household members who stayed and to narrow the welfare difference\nbetween rural and urban communities (Asante, 1995).\n\n\nLitchfield and Waddington (2003) estimate a higher consumption level among migrant\nhouseholds, although the difference decreased between 1991/92 and 1998/99, and the finding does\nnot hold when considering non-monetary measures of welfare and the probability\n\nMention: \"1991 Ghana Migration Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its analyses find migration reduces household labour supply with an offset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006770:10:0", "occurrence_id": "2743da21ffdeb0a3ef6d1dd4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: uncertainty. This aspect is crucial to understanding the political economy of reforms and why\n\n\nmost governments in the region have put subsidy reforms at the top of their agenda in recent\n\n\nyears.\n\n\nFigure 2: Commodities and Oil Prices, January 1992–May 2016\n\n\n\n\n\n\n\n\n\n\n\n_Sources:_ Elaborated from http://www.indexmundi.com/ and World Bank Commodities Prices\n\n\ndatabase.\n\n\nThe next issue to consider is the regulated prices that MENA countries were confronted with\n\n\nwhen they were forced to introduce reforms. Table 2 shows the average regional prices for four\n\n\nenergy products—gasoline, diesel, kerosene, and LPG—comparing the main regions of the\n\n\nworld. It is striking how the MENA region distinguishes itself from the others by having, by far,\n\n\nthe lowest average regional prices. For example, the price of gasoline, on average, was US$ 0.67,\n\n\nabout half the price of other regions. The price of LPG was US$ 0.4, which is less than a third of\n\n\nthe average price in South Asia and about a third of the price in East Asia and the Pacific.\n\n\nSimilarly, for diesel and kerosene, the prices in the MENA region were a fraction of the prices in\n\n\nother regions. Hence, the worldwide push for subsidy reforms that took place in virtually all\n\n\nregions of the world since 2010–11 was particularly acute in the MENA region when growth\n\n\ntook a negative turn due to political instability. These two factors together with the economic\n\n\n9\n\nMention: \"World Bank Commodities Prices\n\n\ndatabase\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Commodity price series across this window come from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006771:22:0", "occurrence_id": "48c8ef083f8ac568c1f7cb61", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Libya, where the food subsidy program is very large. In this country and with a universal transfer\n\n\ndesigned to offset the poverty gap, the increase in per capita government revenue can be large\n\n\nbut still below the overall gains in revenues determined by the reforms.\n\n\nFigure 10: Governments’ Revenue Impact of a 30 Percent Reduction in Subsidies on Energy\nProducts, in US$-PPP/capita/year\n\n\n\n**Electricity**\n\n\nTunisia\n\n\nMorocco\n\n\nLibya\n\n\nIran, Islamic Rep.\n\n\nDjibouti\n\n\n0 50 100Values in US$‐PPP150 200 250 300 350 400\n\n\nPer capita equivalent transfer to offset the change in the poverty gap\n\n\nChange in the per capita government revenue\n\n\n**Diesel**\n\n\nTunisia\n\n\nMorocco\n\n\nLibya\n\n\nIran, Islamic Rep.\n\n\nDjibouti\n\n\n0.0 0.5 Values in US$‐PPP1.0 1.5 2.0 2.5 3.0\n\n\nPer capita equivalent transfer to offset the change in the poverty gap\n\n\nChange in the per capita government revenue\n\n\n\nTunisia\n\n\nMorocco\n\n\nLibya\n\n\nIran, Islamic Rep.\n\n\nDjibouti\n\n\n\n**Liquefied petroleum gas**\n\n\n0 10 20 30 40\n\nValues in US$‐PPP\n\n\n\nPer capita equivalent transfer to offset the change in\nthe poverty gap\n\n\nChange in the per capita government revenue\n\n\n**Gasoline**\n\n\nTunisia\n\n\nMorocco\n\n\nLibya\n\n\nIran, Islamic Rep.\n\n\nDjibouti\n\n\n0 50 100 150 200 250 300\n\nValues in US$‐PPP\n\n\nPer capita equivalent transfer to offset the change in\nthe poverty gap\n\n\nChange in the per capita government revenue\n\n\n\n_Source:_ World Bank estimations from Household Budget Surveys.\n\n\n\n21\n\nMention: \"Household Budget Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Government revenue changes per capita are estimated from these rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006774:4:1", "occurrence_id": "bf1b7f6f992c360bd19adf2d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  of GLSS have\n\n\nalso become available, so that the proposed case study paper will be one of the first studies using this\n\n\ndata set. However, only the last four rounds, from 1991/92 (GLSS-3) to 2012/12 (GLSS-6), have been\n\n\nbased on the same questionnaire and are therefore fully comparable.\n\n\nThe availability of comparable and extensive information represents a success on its own. Ghana is\n\n\none of the few countries in Africa that has produced comparable, high-quality household data covering\n\n\nover two decades. This is an important achievement because the availability of such rich and\n\n\ncomparable information beginning in 1991, as well as the quality improvements of the surveys over\n\n\nthe years and the fact that they collect data on both the monetary and the non-monetary dimensions of\n\n\nwelfare, permit the establishment of an accurate picture of inequality and polarization over time,\n\n\nincluding the drivers behind these phenomena.\n\n\n3\n\nMention: \"GLSS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"inform\", \"summary\": \"Later waves becoming available extend what the case study can use.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006776:17:0", "occurrence_id": "a0b40e28f2565e3c15f0cd39", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 100\n\n\n90\n\n\n80\n\n\n70\n\n\n60\n\n\n50\n\n\n40\n\n\n30\n\n\n20\n\n\n10\n\n\n0\n\n\n\nWon't\nmake\nthings\nbetter\n\n\nCould\nmake\nthings\nbetter\n\n\n\nall low\neducation\n\n\n\nfood\ninsecure\n\n\n\nall low\neducation\n\n\n\nfood\ninsecure\n\n\n\n**Uganda** **Nigeria**\n\n\n_Source:_ World Bank (2016) using data from Afrobarometer Round 4.5.2 (Uganda 2011), Round 3.5\n(Nigeria 2007).\n\n\n_Notes_ : The survey question is the following: “Which of the following statements is closest to your view?\nChoose Statement 1 or Statement 2.Statement 1: No matter how you vote, it won’t make things any\nbetter in the future. Statement 2: The way you vote could make things better in the future.”\n\n\nNot only do Ugandan and Nigerian citizens believe in the importance of voting, and vote at high rates,\nbut they also express interest in receiving information about elections. More than 80 percent of\nrespondents said they wanted a little more or a lot more information ahead of the 2007 elections in\nNigeria and the 2011 elections in Uganda (Figure 5). Again, those with less education and food security\nare just as interested in receiving more information as others.\n\n\n**Figure 5. How Much More Information Do Citizens Want in Order to Decide How to Vote, Uganda and**\n**Nigeria**\n\n\n16\n\nMention: \"Afrobarometer Round 4.5.2\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its module on food insecurity carries the two-country comparison.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006777:31:0", "occurrence_id": "10d2e6b2439516e9d0e80973", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 6 Cotton Output per Farmer: Administrative Delivery Data vs. Survey Data, 2009/10 Season**\n\n\nSource: Author’s analysis based on data provided by Fink Masiye.\nNote: This chart juxtaposes the data unerlying Fink Masiye (2015) with the same farmers’ deliveries (in kg) to the cotton buying company\nin the same year, where available. Deliveries are translated from bales into kilos using the average weight of 80kg per bale. Values are not\nlog-transformed in order to retain observations with values equal to zero. To visualize the majority of observations that lie below 1,000 kg,\na subset of the data is displayed on the right.\n\n\n30\n\nMention: \"Administrative Delivery Data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Output per farmer is compared between these records and the survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006778:3:2", "occurrence_id": "b8c38157e6b61b97cd4869ab", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  at the district level (Lanjouw et al., 2013). In addition, this\nstudy computes environmental risks at the commune level <sup>2</sup> to relate it to household information based on the\nVietnam Household Living Standard Surveys (VHLSS) for 2010, 2012, and 2014. Benefiting from the panel\n\n\n1 This number is based on the national poverty line, which is at is $3.49‐a‐day in 2011 PPP terms. By the $1.90‐a‐day\npoverty line, the extreme poverty rate was 3% in 2014.\n2 These are third‐level administrative subdivisions, equivalent to a sub‐district.\n\nMention: \"Vietnam Household Living Standard Surveys (VHLSS)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Household information across these rounds is matched to commune-level data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006779:46:2", "occurrence_id": "976dcbd6b4c26d25c889b375", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  <u>Yes</u> <u>Yes</u> <u>Yes</u> <u>Yes</u> <u>Yes</u> <u>Yes</u> <u>Yes</u>\n\n\nN 2246 2246 2246 2246 2294 2145 2145 2145 2145 2170\n\n\n<u>R2</u> <u>0.082</u> <u>0.090</u> <u>0.080</u> <u>0.083</u> <u>0.069</u> <u>0.074</u> <u>0.081</u> <u>0.068</u> <u>0.082</u> <u>0.047</u>\n\n\nNotes: Table presents coefficients estimated from Panel sub‐sample Fixed Effects model. * 0.10 ** 0.05 *** 0.01 significance level. Values in\nparentheses indicate standard errors, which are corrected for cluster correlation at commune‐level in regressions (1) – (4). Regressions include\nhousehold controls as presented in Table A.5.\nSource: Author’s calculation based on VHLSS 2010, 2012 & 2014 and CRU data.\n\n\n45\n\nMention: \"VHLSS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Household controls in the appendix table are computed across these rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006779:46:2", "occurrence_id": "36748c635fcff6ddfd8398c7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  <u>Yes</u> <u>Yes</u> <u>Yes</u> <u>Yes</u> <u>Yes</u> <u>Yes</u> <u>Yes</u>\n\n\nN 2246 2246 2246 2246 2294 2145 2145 2145 2145 2170\n\n\n<u>R2</u> <u>0.082</u> <u>0.090</u> <u>0.080</u> <u>0.083</u> <u>0.069</u> <u>0.074</u> <u>0.081</u> <u>0.068</u> <u>0.082</u> <u>0.047</u>\n\n\nNotes: Table presents coefficients estimated from Panel sub‐sample Fixed Effects model. * 0.10 ** 0.05 *** 0.01 significance level. Values in\nparentheses indicate standard errors, which are corrected for cluster correlation at commune‐level in regressions (1) – (4). Regressions include\nhousehold controls as presented in Table A.5.\nSource: Author’s calculation based on VHLSS 2010, 2012 & 2014 and CRU data.\n\n\n45\n\nMention: \"CRU data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Weather station observations combined with the household rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006782:17:1", "occurrence_id": "10c73f5d6c0da7dc35644814", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Ecuador is 21% lower than the already underdeveloped\n\n\n18 This is the sector as defined in the Input Output Tables from Ecuador 2007-2013. “Professional, scientific and technical activities.\nAdministrative and support service activities _”_ are ISIC Rev. 4 Sections M (Divisions 69-75) and N (Divisions 77-82).\n\n\n16\n\nMention: \"Input Output Tables\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its sector definitions define the activity group being measured.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006798:9:0", "occurrence_id": "ec4d6caf1570709e112bb55f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **_Indirect Taxes_**\n\n\nVAT is a major component of indirect taxes, contributing to about 36 percent of total tax collection\nin 2014. VAT is levied at a standard rate of 23 percent on most goods and services, but a reduced\nrate of 5 percent applies to certain food, books and magazines or 8 percent is imposed on supplies,\nsuch as medicines, hotel and catering services, certain transport and municipal services. There exists\nalso a rate of 7% for products bought from farmers who pay the tax as a lump sum. A zero-rate applies to\nthe intra-EU community supply of goods, exports of goods, some international transportation and\nrelated services. Usually, in order to apply zero VAT rate additional conditions need to be fulfilled.\nSome financial, medical and cultural services are exempt. In particular, small entrepreneurs with\nyearly revenues below 150 thousand zloty, are exempt from paying VAT and may not deduct the\nVAT paid on inputs. The VAT gap has been estimated to be between 30% and 50% of VAT\ncollected (PwC, 2013).\n\n\nExcise taxes contribute 3.4 percent of GDP and they are levied on goods that are deemed to be a\nharmful for health reasons. The Ministry of Finance in Poland announces every year the excise\ntaxes that apply to products that might be harmful to health or environment. Goods subject to\nexcise duty include energy products, alcoholic beverages and manufactured tobacco products,\nbased on the legislation of the European Union.\n\n\n**Social spending**\n\n\nOverall expenditures in Poland amounted to 42.1 percent of GDP in 2014, down from 45.5 percent\nof GDP in 2010 (Eurostat). A large part of public spending is dedicated to social protection (16.1\npercent of GDP), while education (5.3 percent of GDP) and health (4.6 percent of GDP) are also\nrelatively important. Total social spending in Poland amounts to 26 percent of GDP or 62 percent\n\nMention: \"Eurostat\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Public spending as a share of output across these years comes from this office.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006801:36:0", "occurrence_id": "a65849dc1a185dbdf199716b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **_Annex B: Estimation of GDP Projections to 2030_** <sup>**_3_**</sup> [^3: GDP estimation for 2030 was prepared by Patrick Eozenou (Health Economist, World Bank) for the\nworking paper by Olusoji Adeyi, Caroline Ly, Patrick Eozenou, Allyala Nandakumar, Ariel Pablos‐Mendez and\nTimothy Evans, \"The economic transition of health in Africa: a call for progressive pragmatism to shape the\nfuture of health financing\".]\n\n\n<u>Estimation Method</u>\n\n- Data from 1995 to 2020 are obtained from the World Economic Outlook database.\n\n- Data from 2014 to 2020 are IMF estimates.\n\n- Data from 2020 to 2035 are projections based on the following assumptions:\n\n1. Cobb‐Douglas production function with constant returns to scale: Y = A*K^(1‐\nalpha)*L^(alpha)\n2. Labor share (alpha) = 1/3\n3. TFP growth (A) and Capital growth (K) are constant at steady state\n4. Labor force growth is equal to population growth\n5. Data on population growth are taken from the UN‐DESA World Population Prospects\n2012 (medium fertility assumption)\n\n<u>Steps</u>\n\n1. Derive the projections for annual growth rates of real GDP ‐‐> (World Economic Outlook\ndata base (WEO_NGDP_RPCH)\n2. Rebase the WDI constant GDP series (2011 GDP deflator = 100) in LCU ‐‐>\nWDI_GDP_constant_LCU2011\n3. Derive the projections for real GDP at 2011 prices using projected growth rates (2021‐\n2035) to real GDP estimates from 2020 onward ‐‐> REAL GDP PER CAPITA\n4. Apply WPP 2012 population projections based on medium fertility assumption to derive\nprojected real GDP per capita (expressed in million LCU per capita) ‐‐> REAL GDP PER\nCAPITA\n5. 2011 PPP factors are applied to convert constant GDP in LCU to international $ ‐‐> REAL\nGDP PER CAPITA\n6. A multilevel model with country\n\nMention: \"World Economic Outlook database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Macroeconomic series across this span, including near-term estimates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006801:36:0", "occurrence_id": "d613df50922c2c06b278e9d0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **_Annex B: Estimation of GDP Projections to 2030_** <sup>**_3_**</sup> [^3: GDP estimation for 2030 was prepared by Patrick Eozenou (Health Economist, World Bank) for the\nworking paper by Olusoji Adeyi, Caroline Ly, Patrick Eozenou, Allyala Nandakumar, Ariel Pablos‐Mendez and\nTimothy Evans, \"The economic transition of health in Africa: a call for progressive pragmatism to shape the\nfuture of health financing\".]\n\n\n<u>Estimation Method</u>\n\n- Data from 1995 to 2020 are obtained from the World Economic Outlook database.\n\n- Data from 2014 to 2020 are IMF estimates.\n\n- Data from 2020 to 2035 are projections based on the following assumptions:\n\n1. Cobb‐Douglas production function with constant returns to scale: Y = A*K^(1‐\nalpha)*L^(alpha)\n2. Labor share (alpha) = 1/3\n3. TFP growth (A) and Capital growth (K) are constant at steady state\n4. Labor force growth is equal to population growth\n5. Data on population growth are taken from the UN‐DESA World Population Prospects\n2012 (medium fertility assumption)\n\n<u>Steps</u>\n\n1. Derive the projections for annual growth rates of real GDP ‐‐> (World Economic Outlook\ndata base (WEO_NGDP_RPCH)\n2. Rebase the WDI constant GDP series (2011 GDP deflator = 100) in LCU ‐‐>\nWDI_GDP_constant_LCU2011\n3. Derive the projections for real GDP at 2011 prices using projected growth rates (2021‐\n2035) to real GDP estimates from 2020 onward ‐‐> REAL GDP PER CAPITA\n4. Apply WPP 2012 population projections based on medium fertility assumption to derive\nprojected real GDP per capita (expressed in million LCU per capita) ‐‐> REAL GDP PER\nCAPITA\n5. 2011 PPP factors are applied to convert constant GDP in LCU to international $ ‐‐> REAL\nGDP PER CAPITA\n6. A multilevel model with country\n\nMention: \"UN‐DESA World Population Prospects\n2012\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Medium fertility assumption supplies the population growth input.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006801:36:0", "occurrence_id": "482398e1d821aaced1dbce99", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **_Annex B: Estimation of GDP Projections to 2030_** <sup>**_3_**</sup> [^3: GDP estimation for 2030 was prepared by Patrick Eozenou (Health Economist, World Bank) for the\nworking paper by Olusoji Adeyi, Caroline Ly, Patrick Eozenou, Allyala Nandakumar, Ariel Pablos‐Mendez and\nTimothy Evans, \"The economic transition of health in Africa: a call for progressive pragmatism to shape the\nfuture of health financing\".]\n\n\n<u>Estimation Method</u>\n\n- Data from 1995 to 2020 are obtained from the World Economic Outlook database.\n\n- Data from 2014 to 2020 are IMF estimates.\n\n- Data from 2020 to 2035 are projections based on the following assumptions:\n\n1. Cobb‐Douglas production function with constant returns to scale: Y = A*K^(1‐\nalpha)*L^(alpha)\n2. Labor share (alpha) = 1/3\n3. TFP growth (A) and Capital growth (K) are constant at steady state\n4. Labor force growth is equal to population growth\n5. Data on population growth are taken from the UN‐DESA World Population Prospects\n2012 (medium fertility assumption)\n\n<u>Steps</u>\n\n1. Derive the projections for annual growth rates of real GDP ‐‐> (World Economic Outlook\ndata base (WEO_NGDP_RPCH)\n2. Rebase the WDI constant GDP series (2011 GDP deflator = 100) in LCU ‐‐>\nWDI_GDP_constant_LCU2011\n3. Derive the projections for real GDP at 2011 prices using projected growth rates (2021‐\n2035) to real GDP estimates from 2020 onward ‐‐> REAL GDP PER CAPITA\n4. Apply WPP 2012 population projections based on medium fertility assumption to derive\nprojected real GDP per capita (expressed in million LCU per capita) ‐‐> REAL GDP PER\nCAPITA\n5. 2011 PPP factors are applied to convert constant GDP in LCU to international $ ‐‐> REAL\nGDP PER CAPITA\n6. A multilevel model with country\n\nMention: \"WDI_GDP_constant_LCU2011\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"The constant-price output series rebased into local currency units.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006807:46:0", "occurrence_id": "bc674e3c2d5a51fd311f02a0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Appendix A**\n\n**Table A.1: RDD Regression Results of the Intent-to-Treatment Effects (ITT) – 14-Year-Old Males.**\n**<u>PNAD 1999 -</u>** **_<u>9-Month Bandwidth</u>_**\n\n**<u>Males</u>**\n\n_Paid Work_\n\nITT -0.043** -0.043** -0.043**\n\n(0.02) (0.02) (0.02)\n\nControl Mean 0.15 0.15 0.15\n\n\n_Paid Work - Formal_\n\nITT -0.0049 -0.0051 -0.0052\n\n(0.0041) (0.0039) (0.0039)\nControl Mean <u>0.01</u> <u>0.01</u> <u>0.01</u>\n\n_Paid Work - Informal_\n\nITT -0.043** -0.043** -0.043**\n\n(0.020) (0.020) (0.020)\nControl Mean <u>0.21</u> <u>0.21</u> <u>0.21</u>\n\n_Unpaid Work_\n\nITT 0.0059 0.0058 0.0058\n\n(0.01) (0.01) (0.01)\n\nControl Mean 0.068 0.068 0.068\n\n\n_School Attendance_\n\nITT 0.0035 0.0039 0.0042\n\n(0.02) (0.02) (0.02)\n\nControl Mean 0.90 0.90 0.90\n\n\nN 4549 4549 4549\n\nNote: ** Statistically significant at 5 percent. Clustered standard errors\n\nMention: \"PNAD 1999\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Intent-to-treatment effects for fourteen-year-old males are estimated on this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006809:13:0", "occurrence_id": "c1192732966f67aac631480e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: parametric cross-entropy methods, but keeping it consistent with the United Nations’ population\nprojections and levels of education observed in the household surveys. <sup>11</sup> [^11: The GIDD model also projects the share of skilled and unskilled workers by using initial information on education\navailable in the household surveys, following a methodology that assumes constant enrollment rate across age cohorts.]\n\n\nTo be consistent with the GIDD, LINKAGE is modified to adopt the former’s skilled-unskilled\nlabor definition, whereby a skilled worker is anybody with more than nine years of education,\nand an unskilled worker is anybody with nine years of education, or less. This redefinition\nnecessitates an adjustment of the GTAP data on value added by labor type in production, such\nthat the number of workers of a given skill type in a given sector is consistent in the 2007\nbenchmark year across the two modeling frameworks.\n\n\n**3.2 Scenarios**\n\n\nUsing the simulation framework described above, seven alternative future scenarios are\nconsidered. The scenarios are designed so as to be able to identify the marginal impact of a given\npolicy outcome as it interfaces with demographic change (Table 1). The policy outcomes are\nthose related to the demographic dividend and include the maintenance and improvement of\nemployment ratios, improvements in educational attainment and quality, and changes in savings\nbehavior. To achieve this, first a “business-as-usual” scenario is considered to establish a\nbaseline economic path till 2050 for the SACU economies. In the baseline, educational\nattainment rates are assumed to be constant, the total labor supply grows at the same rate as the\nworking-age population (based on the same rate as the UN WPP’s medium fertility scenario),\nand employment ratios remain constant at 2015 values. The current account balances and\ninvestment shares track the forecasts of World Bank (2015b) till 2017 before converging to longrun values. Labor productivity is assumed to grow in a fashion that real GDP grows till 2017\nfollowing the World Bank (2015b), and following growth projections under the\n\nMention: \"GTAP data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"Value added by labour type is adjusted to the new education cutoff.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006810:9:0", "occurrence_id": "d8440595ef42f5d2d84ef409", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Hanna (2014), from a combination of CPCB online and print records. The dataset spans the whole of\n\n\nIndia and runs from 1986-2004; an observation is a pollution monitor-month.\n\n\nThe national dataset contains information from 470 monitors situated along 162 rivers in 28 states.\n\n\nExcluding all areas outside the Ganga Basin leaves us with 101 unique pollution monitors along 29\n\n\nrivers, in 62 districts representing 10 states; these are mapped in Figure 2. Over our 19-year period\n\n\nof observation, as many as 46 different measures of water quality are recorded at these monitoring\n\n\nstations, but only a few measures are consistently recorded over the entire sample time frame. The\n\n\ndata are noisy, suggesting significant measurement error, and the panel is imbalanced. To mitigate these\n\n\nproblems, we collapse monitor-level observations to the district level and construct moving averages\n\n\nof the data over a four-month window. These steps produce 8,725 district-month observations in the\n\n\nGanga Basin with at least one non-missing value from our pollution measures of interest.\n\n\nOur main indicator of river quality is BOD. This common, broad-based indicator of water pollution\n\n\ncaptures the amount of dissolved oxygen needed by water-borne, aerobic organisms to break down\n\n\norganic material present at a certain temperature (usually 20 degrees Celsius) and over a specific time\n\n\nperiod (usually five days). Its units are milligrams of oxygen consumed per liter (mg/l). Reduction of\n\n\nBOD is the primary goal of waste treatment plants in general (Brown and Caldwell 2001), but BOD is a\n\n\nparticularly good choice for pollution measurement in the setting of Kanpur. Pollution from the tanning\n\n\nprocess primarily comes from two sources: the animal hides themselves, and the chemicals used to tan\n\n\nthem. Both of these sources contain large amounts of organic matter, resulting in abnormally high\n\n\nBOD levels in tannery effluent. According to the United Nations Industrial Development Organization\n\n\n(UNIDO), effluent discharge into surface water typically is required to have BOD below\n\nMention: \"CPCB online and print records\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Combined into a panel of observations spanning the whole country.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006813:14:0", "occurrence_id": "b36d2e5f3b3a793c40371f91", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: For high income OECD economies, Chile, Japan, and Israel have the highest proportions of legal\n\n\ngender disparities while the Slovak Republic, Australia and Ireland have the lowest (figure A7).\n\n\nWhile Chile and Japan have the top two highest proportions of legal gender disparities, the\n\n\ncomposition of legal disparities differs. For Chile it is mostly due to legal disparities in Getting a\n\n\njob, while in Japan it is due to lack of domestic violence legislation. In South Asia, Afghanistan\n\n\nand Pakistan have the highest proportions of legal disparities while Bhutan and India have the\n\n\nlowest. Afghanistan is the only economy in the region with 100 percent legal disparities in three\n\n\nsubcategories - Building credit, Going to court, and Violence (figure A8). In MENA, the Islamic\n\n\nRepublic of Iran has the highest proportions of legal gender disparities followed by Saudi\n\n\nArabia, while Malta and Morocco have the lowest in the region (figure A9).\n\n\n**4.** **Correlations with Other Gender Indices**\n\nThe WBL composite measure of legal disparities provides a clean measure of differential\n\n\ntreatment of men and women before the law. There are two potential critiques of this measure.\n\n\nOn one hand it may be unrelated to gender equality where the law on the books does not reflect\n\n\nthe realities on the ground. Under this scenario, the WBL composite measure of legal disparities\n\n\nwould be uncorrelated with other measures of gender inequality. On the other hand, the measure\n\n\ncould be perfectly correlated with other gender indices, thereby making the WBL measure of\n\n\nlegal disparities redundant. To address both critiques, the correlations between the WBL measure\n\n\nof legal disparities and other gender-related measures are explored. The following gender-related\n\n\nmeasures are considered: CIRI Women's Social Rights, CIRI Women's Economic Rights, CIRI\n\n\nWomen's Political Rights, WEF Gender Gap Index, Social Watch Gender Equity Index, Ratio of\n\n\n13\n\nMention: \"WEF Gender Gap Index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"One of several indices of rights and parity assembled in the list.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006813:14:0", "occurrence_id": "dd0042d984ffdfea56a0c229", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: For high income OECD economies, Chile, Japan, and Israel have the highest proportions of legal\n\n\ngender disparities while the Slovak Republic, Australia and Ireland have the lowest (figure A7).\n\n\nWhile Chile and Japan have the top two highest proportions of legal gender disparities, the\n\n\ncomposition of legal disparities differs. For Chile it is mostly due to legal disparities in Getting a\n\n\njob, while in Japan it is due to lack of domestic violence legislation. In South Asia, Afghanistan\n\n\nand Pakistan have the highest proportions of legal disparities while Bhutan and India have the\n\n\nlowest. Afghanistan is the only economy in the region with 100 percent legal disparities in three\n\n\nsubcategories - Building credit, Going to court, and Violence (figure A8). In MENA, the Islamic\n\n\nRepublic of Iran has the highest proportions of legal gender disparities followed by Saudi\n\n\nArabia, while Malta and Morocco have the lowest in the region (figure A9).\n\n\n**4.** **Correlations with Other Gender Indices**\n\nThe WBL composite measure of legal disparities provides a clean measure of differential\n\n\ntreatment of men and women before the law. There are two potential critiques of this measure.\n\n\nOn one hand it may be unrelated to gender equality where the law on the books does not reflect\n\n\nthe realities on the ground. Under this scenario, the WBL composite measure of legal disparities\n\n\nwould be uncorrelated with other measures of gender inequality. On the other hand, the measure\n\n\ncould be perfectly correlated with other gender indices, thereby making the WBL measure of\n\n\nlegal disparities redundant. To address both critiques, the correlations between the WBL measure\n\n\nof legal disparities and other gender-related measures are explored. The following gender-related\n\n\nmeasures are considered: CIRI Women's Social Rights, CIRI Women's Economic Rights, CIRI\n\n\nWomen's Political Rights, WEF Gender Gap Index, Social Watch Gender Equity Index, Ratio of\n\n\n13\n\nMention: \"Social Watch Gender Equity Index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Another parity index included in the same set.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006814:7:1", "occurrence_id": "24081c8bc86724903f771e84", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ivity), wage competitiveness (labor costs), market\naccess and access to inputs. The logistics measure is from the World Bank Logistics Performance Index\n(LPI), a widely cited international rating system. The measure is influenced by proximity to markets, so\nthere is some overlap between the two indicators. Wages are measured by the World Bank Doing Business\nindicators’ minimum legal wage in dollar terms. Although unit labor costs were preferred as a measure of\nproductivity‐adjusted wages, there is no database with sufficient coverage of countries to be suitable. As\nthe indicator is in dollar terms, exchange rate dynamics affect this indicator, including changes owing to\nlower demand for commodities in recent years. In selecting this as a driver of GVC participation, we also\nacknowledge that higher minimum wages could be a consequence of GVC participation as well as a driver.\nFinally, market access and access to inputs are estimates generated by Kee, Nicita and Olarreaga (2008,\n2009), who calculated two measures of overall trade restrictiveness that account for tariff barriers and\nthe ad‐valorem equivalent of non‐tariff barriers for countries and their trading partners. As these are\naggregate measures focused on removal of trade barriers, they do not fully capture the impact of\npreferential arrangements with certain markets in affecting an industry’s prospects nor do they account\nfor the impacts of deeper integration like investment flows and trade facilitation.\n\n\n5 This measure includes crop, pasture land, timber, non‐timber forest, protected areas, oil, natural gas, coal, and minerals.\n\n\n6\n\nMention: \"the World Bank Doing Business\nindicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its minimum legal wage in dollar terms measures wages here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006815:20:0", "occurrence_id": "45c9bc32de9d25e663bd584a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: households in the bottom of the income distribution disproportionally benefit from an increase in\nthe number of workers in their families, as they move towards fertility transition.\n\nThe realization of the first demographic dividend, led by reductions in child dependency ratios,\ncould facilitate the eradication of global poverty (figure 4). In 1990, East Asia had a higher average\npoverty headcount than South Asia. However, poverty headcount rates in East Asia decreased from\n61 percent to only 7.2 percent between 1990 and 2012. This poverty reduction was paralleled by\nsharp reductions in child dependency ratios in the region. For a country perspective, evidence from\nBangladesh suggests that demographic factors, including age-structure, gender, and regional\ndistributions of populations, accounted for a quarter of the rapid reductions in poverty between\n2000 and 2010 (World Bank 2013). Bangladesh halved its fertility rate between 1971 and 2004,\ngoing from more than 6 children per woman to about 3, and is on track to reach replacement rates\nin the coming decades.\n\n\n**Figure 4 Lower child dependency ratios are associated with lower poverty rates**\n\nPoverty rate and child dependency ratio, percent\n\n\n\n100\n\n\n80\n\n\n60\n\n\n40\n\n\n20\n\n\n0\n\n\n\n1990 2012\n\n\nEAP SAR SSA EAP SAR SSA\n\n\nChild dependency       Poverty headcount\n\n\n\n_Source:_ Authors’ estimates\n_Note_ : Data are from United Nations (2015) and PovcalNet. The poverty headcount rate is based on the $1.90 a day\npoverty line.\n\n\nBecause of the association between fertility and education, income, and life expectancy,\nhouseholds in the top 60 percent of the income distribution tend to have lower child dependency\nratios and to pass through the demographic transition before households in the bottom 40 percent\nin almost all countries for which data are available (figure 5). As fertility rates fall, the\ndemographic structures of the households change and directly affect poverty and shared prosperity,\nparticularly in poor households. Households in the top of the income distribution tend to have\nlower child dependency ratios and to pass through the demographic\n\nMention: \"PovcalNet\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The poverty headcount at the stated daily line is drawn from this tool.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006818:3:1", "occurrence_id": "b3b82f69c4b4a4915a530b5e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  2003,\nVerme et al. 2013, and World Bank, 2012). Such low mobility of human capital is a definite\nsymptom of labor market inefficiency – as human capital cannot be allocated where it could yield\nthe highest returns (Gatti et al., 2013 and World Bank, 2012).\n\nPrevious comparative analysis of labor regulation in the MENA region (see Angel-Urdinola and\nKuddo, 2010 and World Bank, 2013) indicate that some features of labor regulation in Morocco,\nnotably minimum wage policy and payroll taxes, may contribute to elevating the cost of formal\nlabor in the country and to constrain the demand and mobility of formal labor, especially among\nyouth (Betcherman 2012 and World Bank 2012).\n\nIn particular, the level of the non-agricultural minimum wage (SMIG) in Morocco is high relative\nto the average worker’s productivity and well enforced in the formal private sector, especially in\nurban areas (Benhayoun et al. 2001 and Agénor and El Aynaoui, 2003). Compared to international\nbenchmarks, this ratio is the highest in the MENA region, even by OECD standards (AngelUrdinola and Kuddo, 2010). <sup>4</sup> [^4: According to the Doing Business Dataset this ratio was at 0.77 in year 2013 (www.doingbussines.org)] Moreover, between 2007 and 2013, the SMIG increased by 27\npercent while average consumer prices increased by only 13 percent. <sup>5</sup> [^5: International evidence indicates that a 10 percent increase in the level of minimum wage is estimated to reduce youth employment\nby a range of 1 to 4 percent in an average labor market (Angel-Urdinola and Kuddo, 2010).] Also, payroll taxes (paid by\nemployers) and social security contributions (paid by employees) in Morocco approach 25 percent\n\n\n2 Latest available office data from Morocco’s High Planning Commissariat (for year 2015) indicate that overall employment rates\nin the country (population 15-64) are at 42.8 percent\n\nMention: \"Doing Business Dataset\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It gives the ratio at the stated value for that year.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006819:8:2", "occurrence_id": "af9639910154ee9982120db6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  𝑫𝒊𝒋𝒕 is a vector comprising the seven dummy variables for each provision category in the\nDESTA dataset, as described in section 2.2.\n\n\n13 This specification is equivalent to that used in BBF for aggregate trade flows only.\n14 CEPII provides various measures of bilateral distance. The most frequently used are simple distance, which\nmeasures distance from capital city to capital city, and population-weighted distance (both detailed in Mayer\nand Zignago (2011)). Regression results when using the two different measures were very similar (as was\nalso found by Johnson and Noguera (2012)). Therefore, we report results using the simple distance measure\nsince it is more commonly used in the gravity literature.\n\n\n7\n\nMention: \"DESTA dataset\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Seven dummy variables per provision category form the vector used here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006824:19:0", "occurrence_id": "ee848380436e348f137b97ab", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: of the employees. Segmentation is most directly measured by the extent to which the gender of\nowners influences their decision to hire workers of the same gender. For female-owned\nestablishments, for instance, this is measured as the share of female employees in total\nemployment, while for male-led businesses segmentation is computed as the share of male\nemployees in total employment. A shortcoming of this measure, however, is that we need\ninformation on the gender of the owner, which, unfortunately, is unavailable for the organized\nmanufacturing surveys.\n\n\nIn an alternative setting, we view gender segmentation through the perspective of an average\nemployee. For an average employee, we compute the share of people that this employee works\nwith who are of his/her gender. For example, for an average female worker, we measure\nsegmentation as the share of other workers at her establishment that are also female, while we\ncompute a similar metric for an average male employee. In our work, we sometimes also consider\nthe segmentation of the combined group, that is, for an average employee we compute the percent\nof co-workers in the establishment who are of the same gender.\n\n\nSince trade and domestic industry reforms have been widely known to impact primarily the\norganized manufacturing sector (e.g., Topalova, 2007; Tewari and Wilde, 2015), we measure\nsegmentation in our regressions using metrics that are possible to calculate with the ASI data.\nNevertheless, it is useful to clarify here that our industry level regression work includes industry\nlevel information from both the formal and the informal manufacturing sectors, as trade reforms\ncould have indirectly impacted the informal sector via reallocation effects.\n\n\n∆�� in equation (1) is the related change in the industry level reforms, which could take the form\nof change in tariff protection or an index measuring the change in de-reservation of products\nexclusively for SSI. We follow Topalova and Khandelwal (2011) in using three alternative indices\nfor measuring trade protection:\n\nMention: \"ASI data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Only certain segmentation metrics can be computed from these data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006828:19:0", "occurrence_id": "cccb54fae177b13b2d67142c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: IV. DATA\n\n\nThe data used in this analysis come from a range of publically available sources.\n\n\nBecause the written names of mandals and districts are often the unique observation in\n\n\nthe underlying data sets, we successfully merged all data manually for 1,061 mandals\n\n\nfrom 22 districts in AP, about 96 percent of the 1,109 rural mandals found in these 22\n\n\ndistricts in the Indian Population Census of 2001. <sup>19</sup> Definitions, data sources, and\n\n\nsummary statistics for all of the variables used in our analysis can be found in table S1.1\n\n\nin appendix S1.\n\n\nOne major feature of MGNREGS is the pursuit of transparency. To that end, an\n\n\nincredible amount of administrative information about projects and workers is available\n\n\nonline. <sup>20</sup> Website management is handled by the state, with data input directly from the\n\n\nmandal administration and linked to electronic financial transactions data from the\n\n\nrelevant institutions. While one may question the quality of government-reported project\n\n\ndata, a major study on public works projects around the world praises the information\n\n\ntechnology system implemented by AP in particular (Subbarao et al. 2013), providing\n\n\nstrong evidence that we need not be terribly skeptical of the data quality. <sup>21</sup> We\n\n\ndownloaded reports from the website, which include the total amount spent per fiscal\n\n\n19. There are 23 districts in AP; however, Hyderabad, the capital of the state, is excluded because it is an\nentirely urban district and therefore should not benefit from MGNREGS.\n20. On 11 September 2013, we downloaded mandal level spending data by fiscal year from the MGNREGS\nwebsite for AP (http://MGNREGA.ap.gov.in) from the “report” section (reports/reports general/R1.6).\n21. The administrative data in AP are\n\nMention: \"Indian Population Census\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Rural administrative units in the study districts are enumerated here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006831:63:0", "occurrence_id": "292d7877e8326fa8f60fc9c7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 42: NonCU Destinations for Exports of Food and Agriculture**\n\n\n_Source_ : GTIS data.\n_Notes_ : MENA = Middle East-North Africa; nonCU-CIS = members of the Commonwealth of Independent States but not\namong the first three members of EACU; EU = European Union; ROW = Rest of World. Numbered regions: (1) Yaroslavl, (2)\nKaluga, (3) Vladimir, (4) Ivanovo, (5) Perm, (6) Moscow City, (7) Tula, (8) Nizhniy Novgorod, (9) Ryazan, (10) Mari El, (11)\nUdmurtia, (12) Mordovia, (13) Chuvashia [[ (14, 16) Tatarstan, (15) Penza, (17) Ulyanovsk, (18) Saratov, (19, 20) Samara,\n(21) Volgograd, (22) Kalmykia, (23) Adygea, (24) Stavropol, (25) Karachevo-Chercheskia, (26) Kabardino-Balkaria, (27)\nNorth Ossetia, (28) Chechnya, and (29) Ingushethia.\n\n\n0\n\nMention: \"GTIS data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Export destinations for food and agriculture come from these records.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006837:1:0", "occurrence_id": "6aec5aa2e046cc6fbe359e6e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Policy Research Working Paper 7827\n\n#### **Abstract**\n\nGlobalization is creating many new trade and growth\nopportunities, with services trade increasingly becoming an\nissue for export-oriented economies. Services are important\nto country trade strategies, because they represent activities\nin which countries may have a comparative advantage, and\nthey are drivers of competitiveness for the whole economy.\nThis paper uses data from the World Development Indica\ntors, two new databases (the Export in Value-Added database\nfrom the Global Trade Analysis Project, and Trade in Services data), and firm-level data. The paper employs a wide\nrange of indicators to analyze the trade competitiveness of\nthe services sector in the Russian Federation. Since service\nexports are less than would be expected considering Russia’s\nlevel of development, the study finds that the contribution\nof services to export diversification could be heightened\n\n\n\nsignificantly. The scale of Russian business services exports\nis relatively low, although exports of traditional services,\nlike transport and travel, are performing well. Despite the\nrelatively minor importance of exports of modern services,\nthe category of other business services has in recent years\nbeen growing fast, and business services have strengthened\ntheir revealed comparative advantages. Yet Russia still has\nmuch potential for expanding trade in modern services.\nThere is also potential to diversify services exports to other\n\nmarkets, such as France, Germany, Japan, and elsewhere in\nAsia, which today seems underexploited. Finally, although\n\nexports of direct services are low, services such as transport,\ndistribution, finance, and other business services are making\nmajor contributions to other exports, in particular energy.\n\n\n\nThis paper is a product of the Trade and Competitiveness Global Practice Group. It is part of a larger effort by the World\nBank to provide open access to its research and make a contribution to development policy discussions around the world.\nPolicy Research Working Papers are also posted on the Web at http://econ.worldbank.org. The authors may be contacted\nat ssaez@\n\nMention: \"World Development Indica\ntors\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"One of the compilations underlying the competitiveness indicators.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006837:1:0", "occurrence_id": "106aaac925e7b22f6a5c0d2d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Policy Research Working Paper 7827\n\n#### **Abstract**\n\nGlobalization is creating many new trade and growth\nopportunities, with services trade increasingly becoming an\nissue for export-oriented economies. Services are important\nto country trade strategies, because they represent activities\nin which countries may have a comparative advantage, and\nthey are drivers of competitiveness for the whole economy.\nThis paper uses data from the World Development Indica\ntors, two new databases (the Export in Value-Added database\nfrom the Global Trade Analysis Project, and Trade in Services data), and firm-level data. The paper employs a wide\nrange of indicators to analyze the trade competitiveness of\nthe services sector in the Russian Federation. Since service\nexports are less than would be expected considering Russia’s\nlevel of development, the study finds that the contribution\nof services to export diversification could be heightened\n\n\n\nsignificantly. The scale of Russian business services exports\nis relatively low, although exports of traditional services,\nlike transport and travel, are performing well. Despite the\nrelatively minor importance of exports of modern services,\nthe category of other business services has in recent years\nbeen growing fast, and business services have strengthened\ntheir revealed comparative advantages. Yet Russia still has\nmuch potential for expanding trade in modern services.\nThere is also potential to diversify services exports to other\n\nmarkets, such as France, Germany, Japan, and elsewhere in\nAsia, which today seems underexploited. Finally, although\n\nexports of direct services are low, services such as transport,\ndistribution, finance, and other business services are making\nmajor contributions to other exports, in particular energy.\n\n\n\nThis paper is a product of the Trade and Competitiveness Global Practice Group. It is part of a larger effort by the World\nBank to provide open access to its research and make a contribution to development policy discussions around the world.\nPolicy Research Working Papers are also posted on the Web at http://econ.worldbank.org. The authors may be contacted\nat ssaez@\n\nMention: \"Export in Value-Added database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"A newer database on value-added trade used alongside the others.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006837:1:0", "occurrence_id": "bddd34d55795fe0b3d177fe9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Policy Research Working Paper 7827\n\n#### **Abstract**\n\nGlobalization is creating many new trade and growth\nopportunities, with services trade increasingly becoming an\nissue for export-oriented economies. Services are important\nto country trade strategies, because they represent activities\nin which countries may have a comparative advantage, and\nthey are drivers of competitiveness for the whole economy.\nThis paper uses data from the World Development Indica\ntors, two new databases (the Export in Value-Added database\nfrom the Global Trade Analysis Project, and Trade in Services data), and firm-level data. The paper employs a wide\nrange of indicators to analyze the trade competitiveness of\nthe services sector in the Russian Federation. Since service\nexports are less than would be expected considering Russia’s\nlevel of development, the study finds that the contribution\nof services to export diversification could be heightened\n\n\n\nsignificantly. The scale of Russian business services exports\nis relatively low, although exports of traditional services,\nlike transport and travel, are performing well. Despite the\nrelatively minor importance of exports of modern services,\nthe category of other business services has in recent years\nbeen growing fast, and business services have strengthened\ntheir revealed comparative advantages. Yet Russia still has\nmuch potential for expanding trade in modern services.\nThere is also potential to diversify services exports to other\n\nmarkets, such as France, Germany, Japan, and elsewhere in\nAsia, which today seems underexploited. Finally, although\n\nexports of direct services are low, services such as transport,\ndistribution, finance, and other business services are making\nmajor contributions to other exports, in particular energy.\n\n\n\nThis paper is a product of the Trade and Competitiveness Global Practice Group. It is part of a larger effort by the World\nBank to provide open access to its research and make a contribution to development policy discussions around the world.\nPolicy Research Working Papers are also posted on the Web at http://econ.worldbank.org. The authors may be contacted\nat ssaez@\n\nMention: \"Trade in Services data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The services component of the trade measurement.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006838:27:1", "occurrence_id": "53e128b51af7fa824d026f93", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: * 1.43% 1.67% 2.43% 3.44% 3.16%\n\n**OECD** 2.30% 2.19% 2.71% 2.74% 2.58%\nEAP 0.44% 0.49% 1.01% 1.41% 1.27%\n\n**ECA** 1.73% 1.44% 1.79% 1.16% 1.03%\n\n**MENA** 0.04% 1.05% 0.40% 1.18% 0.90%\n\n**<u>SHA</u>** <u>0.24%</u> <u>0.27%</u> <u>0.78%</u> <u>0.82%</u> <u>0.63%</u>\n\nSA 0.28% 0.42% 0.82% 0.69% 0.17%\n\n**<u>LAC</u>** **<u>0.47%</u>** **<u>0.59%</u>** **<u>0.35%</u>** **<u>0.59%</u>** **<u>0.65%</u>**\n\n\n_Source:_ Author’s calculations based on data from Lederman y Saenz (2005) and UNESCO\n**Note:** LAC: Latin America and the Caribbean; Scand.: Scandinavia, EAP: East Asia and the Pacific; ECA:\n\nEurope and Central Asia ; MENA: Middle East and Noth Africa (excluding Israel); SA: South Asia; SHA:\n\nSub-Saharian Africa . All previous groups exclude high income countries.\n\nMention: \"Lederman y Saenz (2005)\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its data are combined with the education body's figures for the comparison.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006840:21:0", "occurrence_id": "0ab1a504a23618cd7e8bdde1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: poverty reduction. Figure 12 summarizes the age composition and average household size by\npercentiles of per capita consumption using the 2010 NLSS data. Poorer households tend to have\na larger household size with more children and fewer working-age adults. On the other hand,\nwealthier households tend to have fewer members but more working-age adults and fewer\nchildren. The contrasting demographic composition of the wealthy and the poor underscores the\nimportant role of population dynamics in poverty reduction and economic development.\n\n\n**Figure 12: Age Composition and Average Household Size by Consumption Percentile**\n\n\n100%\n\n\n\n80%\n\n\n60%\n\n\n40%\n\n\n20%\n\n\n\n\n\n3 7 11 15 19 23 27 31 35 39 43 47 51 55 59 63 67 71 75 79 83 87 91 95 99\n\n\n100%\n\n\n\n80%\n\n\n60%\n\n\n40%\n\n\n20%\n\n\n\n\n\n\n\n**3** **7** **11** **15** **19** **23** **27** **31** **35** **39** **43** **47** **51** **55** **59** **63** **67** **71** **75** **79** **83** **87** **91** **95** **99**\n\n\nConsumption Percentiles\n\n\nSource: Nepal Living Standards Survey 2010\n\n\n20\n\nMention: \"2010 NLSS data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Household composition by consumption percentile is drawn from this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006840:21:0", "occurrence_id": "34b50f7ca9f21fbfecfb8aa6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: poverty reduction. Figure 12 summarizes the age composition and average household size by\npercentiles of per capita consumption using the 2010 NLSS data. Poorer households tend to have\na larger household size with more children and fewer working-age adults. On the other hand,\nwealthier households tend to have fewer members but more working-age adults and fewer\nchildren. The contrasting demographic composition of the wealthy and the poor underscores the\nimportant role of population dynamics in poverty reduction and economic development.\n\n\n**Figure 12: Age Composition and Average Household Size by Consumption Percentile**\n\n\n100%\n\n\n\n80%\n\n\n60%\n\n\n40%\n\n\n20%\n\n\n\n\n\n3 7 11 15 19 23 27 31 35 39 43 47 51 55 59 63 67 71 75 79 83 87 91 95 99\n\n\n100%\n\n\n\n80%\n\n\n60%\n\n\n40%\n\n\n20%\n\n\n\n\n\n\n\n**3** **7** **11** **15** **19** **23** **27** **31** **35** **39** **43** **47** **51** **55** **59** **63** **67** **71** **75** **79** **83** **87** **91** **95** **99**\n\n\nConsumption Percentiles\n\n\nSource: Nepal Living Standards Survey 2010\n\n\n20\n\nMention: \"Nepal Living Standards Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The consumption percentile series in the figure come from this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006846:17:0", "occurrence_id": "1ef34f807b4b6a578f3be4d6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |Figure 15. Exclusion components of welfare<br>aggregate and poverty rates at 5.5 2011 PPPs<br>daily line, %|Figure 16. Exclusion components of welfare<br>aggregate and Gini coefficient|\n|---|---|\n|||\n\n\n\nSource: Authors’ calculation using HEIS 2008-2014.\n<u>Note: Spatially adjusted welfare aggregate is used for calculations.</u>\n\n\n5.4 Welfare aggregate and access to selected durables\n\n\nOne important requirement for the welfare aggregate is its ability to correctly sort out people along the\nwealth ladder. Possession of certain durables is often used as an important indicator of wealth. Figure 17\nshows distribution of cars, computers and dishwashers among the population by expenditure per capita\nquintiles using the preferred spatially adjusted welfare aggregate without durables and health. Clearly, the\npoorest households are less likely to possess computers, cars and dishwashers compared to wealthier ones.\nFor example, literally no one in the bottom quintile has a dishwasher compared to 67 percent owning one\nin the top quintile. The steep gradient in ownership of selected durables across quintiles corresponds well\nwith expectations – as households become better-off they are more likely to own cars, computers and\ndishwashers. This can be viewed as supporting evidence that the constructed welfare aggregate appears to\nbe sorting households in the way one expects.\n\n\n**<u>Figure 17. Access to durables across expenditure per capita quintiles in 2013</u>**\n\n\n16\n\nMention: \"HEIS 2008-2014\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"A spatially adjusted welfare aggregate is computed across these rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006848:40:0", "occurrence_id": "877ee2389b1d8b89e9a1aadd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Table S.3: Panel household fixed effects regression: all households with children, IFLS1-IFLS4.\n\n\n(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)\n<u>Food</u> <u>Food</u> <u>Education</u> <u>Education</u> <u>Health</u> <u>Health</u> <u>Durables</u> <u>Durables</u> <u>Adult</u> <u>Adult</u>\n\n\nMigrant Parent -0.027* -0.007 0.007 0.007 0.008 0.018** -0.050*** -0.074*** -0.023*** -0.047***\n(0.016) (0.020) (0.009) (0.012) (0.007) (0.007) (0.017) (0.026) (0.008) (0.010)\nMigrant Mother -0.049* -0.000 -0.025 0.058** 0.061***\n(0.029) (0.016) (0.018) (0.027) (0.016)\nSize of Household 0.038** 0.037** -0.006 -0.006 -0.020* -0.021* -0.021 -0.020 0.000 0.001\n(0.018) (0.018) (0.013) (0.013) (0.011) (0.011) (0.020) (0.020) (0.010) (0\n\nMention: \"IFLS1-IFLS4\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Panel household fixed effects across these waves carry the regression.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006850:27:1", "occurrence_id": "29acde2d267bd2c11f2637a4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <br>4<br>6<br>Citizen engagement in rulemaking score<br>3_3_CivicParticipation<br>Fitted values||\n\n\n\n_Source:_ Analysis based on data from Citizen Engagement in Rulemaking database, World Bank Group,\nhttp://rulemaking.worldbank.org; the World Justice Project database and the Economist journal’s database.\n_Note:_ All the regressions control for income per capita and are significant at the 1 percent level. WJP = World\nJustice Project.\n\n\nIndeed, countries with a robust, well-functioning democracy also tend to have a highly inclusive\nrulemaking process—and vice versa. Conversely, countries that lack a well-functioning\ngovernment along with political pluralism, respect for civil liberties and freedom of expression\nand belief also tend to exclude the public from rulemaking processes. Researchers emphasize\nthat citizen participation in rulemaking and transparency in governance can enhance democratic\nlegitimacy by bridging the gap between government agencies and their constituencies.\n\n\n26\n\nMention: \"Citizen Engagement in Rulemaking database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Civic participation in rulemaking is measured in this compilation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006850:27:1", "occurrence_id": "0212fa927dd84c5380899785", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <br>4<br>6<br>Citizen engagement in rulemaking score<br>3_3_CivicParticipation<br>Fitted values||\n\n\n\n_Source:_ Analysis based on data from Citizen Engagement in Rulemaking database, World Bank Group,\nhttp://rulemaking.worldbank.org; the World Justice Project database and the Economist journal’s database.\n_Note:_ All the regressions control for income per capita and are significant at the 1 percent level. WJP = World\nJustice Project.\n\n\nIndeed, countries with a robust, well-functioning democracy also tend to have a highly inclusive\nrulemaking process—and vice versa. Conversely, countries that lack a well-functioning\ngovernment along with political pluralism, respect for civil liberties and freedom of expression\nand belief also tend to exclude the public from rulemaking processes. Researchers emphasize\nthat citizen participation in rulemaking and transparency in governance can enhance democratic\nlegitimacy by bridging the gap between government agencies and their constituencies.\n\n\n26\n\nMention: \"World Justice Project database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"One of the rule-of-law compilations used in the regressions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006850:27:1", "occurrence_id": "ee36d657bd29ef37f5761cdd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <br>4<br>6<br>Citizen engagement in rulemaking score<br>3_3_CivicParticipation<br>Fitted values||\n\n\n\n_Source:_ Analysis based on data from Citizen Engagement in Rulemaking database, World Bank Group,\nhttp://rulemaking.worldbank.org; the World Justice Project database and the Economist journal’s database.\n_Note:_ All the regressions control for income per capita and are significant at the 1 percent level. WJP = World\nJustice Project.\n\n\nIndeed, countries with a robust, well-functioning democracy also tend to have a highly inclusive\nrulemaking process—and vice versa. Conversely, countries that lack a well-functioning\ngovernment along with political pluralism, respect for civil liberties and freedom of expression\nand belief also tend to exclude the public from rulemaking processes. Researchers emphasize\nthat citizen participation in rulemaking and transparency in governance can enhance democratic\nlegitimacy by bridging the gap between government agencies and their constituencies.\n\n\n26\n\nMention: \"Economist journal’s database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The third governance compilation in the set.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006852:12:1", "occurrence_id": "dd493535d69d67d7c7c73f02", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  The higher the\nprotection imposed by local labor legislation, the higher the incentives for\ncollecting the bribes and the higher the mismatch and distortion found in the\nlabor market. To capture the institutional characteristics of the labor market,\nwe use the **(Worker’s right)** index from the Human Rights Dataset\n(Cingranelli and Richards, 2010). This index is equal to 0 if workers’ rights\nare not protected and 1 if a full or fair protection is given to workers once\nhired.\n\n\nSeveral papers have investigated the economic consequences of\ncorruption on unemployment and the labor market in general through the\nchannel of foreign direct investment (FDI). The theoretical view is that in a\nhighly corrupt environment, foreign actors are more reluctant to invest and\nto create job opportunities in the local market (corruption increases the cost\nof doing business (often referred to as the corruption tax). This view has\nreceived empirical support from recent research. The perception among\nforeign actors that due to corruption, the allocation of domestic human\ncapital toward the most productive economic sectors is inefficient\nconstitutes an obstacle to the inflow of foreign capital in the economy. We\nuse the variable **(FDI/GDP)** to analyze this hypothesis of the impact of\ncapital inflows on the development of labor market dynamics in the\npresence of corruption practices.\n\n\n\nPage | 11\n\nMention: \"Human Rights Dataset\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its worker's rights index ranges from zero upward.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006855:15:1", "occurrence_id": "e4adb0bd4e406c96f0bfd444", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |22.3|33.8%|\n|Five|53.2|16.2|16.2|19.5|36.7%|\n|Six|59.4|23.5|18.1|15.0|41.2%|\n|Seven|49.1|30.3|15.0|9.6|45.2%|\n|Eight|39.3|36.1|12.0|6.5|47.1%|\n|Nine|26.3|38.6|8.0|4.1|48.2%|\n|Ten or more|53.7|38.8|16.3|8.2|49.1%|\n|Total|328.7|19.5|100.0|100.0|32.1%|\n\n\nSource: GMD, UNDESA, WDI, PovcalNet\n\n\nThe increased risk of extreme child poverty from larger household size is however not clearly associated\nwith ‘multi-generational’ households in which elderly people co-reside with their adult children and\ngrandchildren, a far more common practice in developing countries. Table 12 shows that extreme child\npoverty rates do not differ greatly according to whether children live solely with prime age adults (often\ntheir parents), or in three-generation households, or live solely with elderly people. However, over threequarters of extreme child poverty (78.5 percent) is in two-generation households, suggesting that large\nhouseholds in this group are a major factor in extreme poverty risk. Only a fifth of extremely poor\nchildren live in three-generation households. Overall, these patterns highlight the importance of the\nsensitivity tests of our results to assumptions about economies of scale of households’ expenditure in the\nnext section.\n\n\n14\n\nMention: \"PovcalNet\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"One of four sources behind the poverty aggregate.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006856:34:1", "occurrence_id": "27468210abe87588cbcdcbe6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .5 356.5\n<u>Temperature inversion</u> <u>16,846</u> <u>0.24</u> <u>0.43</u> <u>0</u> <u>1</u>\n_Notes_ : Detailed statistics on wind direction converted to dummy variables representing the eight-point compass are\n\n\nomitted to save space.\n\n\n_Source_ : Data from RMCAB for 1997-1999 and 2008-2010. Author’s calculations.\n\n\n33\n\nMention: \"RMCAB\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Compass-direction wind observations across these periods come from this centre.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006858:5:0", "occurrence_id": "964d8bdc144518332e1316cb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: determinants of decision‐making as time‐invariant and suggests that other large changes in circumstances\n\n\nthat are voluntary in nature (e.g. marriage, childbearing, retirement) may also be unlikely to lead to\n\n\nchanges in preferences and beliefs.\n\n\nThe remainder of the paper is structured as follows: Section 2 describes the context in which this migration\n\n\nflow is occurring, and describes the survey and estimation strategy; Section 3 reports on impacts on risk\n\n\nand time preferences; Section 4 covers the impact on pro‐market beliefs; Section 5 reports on the impact\n\n\non interactions with partners and on household efficiency; Section 6 discusses statistical power and\n\n\nexternal validity issues, and Section 7 concludes.\n\n\n**2.** **Context and Survey**\n\n\n**2.1** **Country Context**\n\n\nThe Kingdom of Tonga is an archipelago of islands in the Pacific, with a resident population of just over\n\n\n100,000. It has a gross national income (GNI) per capita of $5,069 (in 2011 PPP$), slightly less than that of\n\n\nIndia, and ranks 132 <sup>nd</sup> out of 188 countries. <sup>1</sup> [^1: All statistics in this section are from the _2015 Human Development Report_ (UNDP, 2015).] Average income in New Zealand is approximately six times\n\n\nas high, at $32,689. Both countries rank 32 places higher in the Human Development Index than they do\n\n\nin terms of per capita GNI.\n\n\nTonga has been and continues to be a less market‐oriented economy than is the case in New Zealand. The\n\n\nIndex of Economic Freedom ranks New Zealand third in the world, as a free, modern, and competitive\n\n\neconomy, while Tonga is ranked as “mostly unfree” at 95 <sup>th</sup> in the world, between Swaziland and the Kyrgyz\n\n\nRepublic (Heritage Foundation, 2016). Tongan society has historically been highly stratified, with\n\nMention: \"Human Development Index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Both countries rank far higher here than on income alone.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006858:5:0", "occurrence_id": "04d042c007111a346d854636", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: determinants of decision‐making as time‐invariant and suggests that other large changes in circumstances\n\n\nthat are voluntary in nature (e.g. marriage, childbearing, retirement) may also be unlikely to lead to\n\n\nchanges in preferences and beliefs.\n\n\nThe remainder of the paper is structured as follows: Section 2 describes the context in which this migration\n\n\nflow is occurring, and describes the survey and estimation strategy; Section 3 reports on impacts on risk\n\n\nand time preferences; Section 4 covers the impact on pro‐market beliefs; Section 5 reports on the impact\n\n\non interactions with partners and on household efficiency; Section 6 discusses statistical power and\n\n\nexternal validity issues, and Section 7 concludes.\n\n\n**2.** **Context and Survey**\n\n\n**2.1** **Country Context**\n\n\nThe Kingdom of Tonga is an archipelago of islands in the Pacific, with a resident population of just over\n\n\n100,000. It has a gross national income (GNI) per capita of $5,069 (in 2011 PPP$), slightly less than that of\n\n\nIndia, and ranks 132 <sup>nd</sup> out of 188 countries. <sup>1</sup> [^1: All statistics in this section are from the _2015 Human Development Report_ (UNDP, 2015).] Average income in New Zealand is approximately six times\n\n\nas high, at $32,689. Both countries rank 32 places higher in the Human Development Index than they do\n\n\nin terms of per capita GNI.\n\n\nTonga has been and continues to be a less market‐oriented economy than is the case in New Zealand. The\n\n\nIndex of Economic Freedom ranks New Zealand third in the world, as a free, modern, and competitive\n\n\neconomy, while Tonga is ranked as “mostly unfree” at 95 <sup>th</sup> in the world, between Swaziland and the Kyrgyz\n\n\nRepublic (Heritage Foundation, 2016). Tongan society has historically been highly stratified, with\n\nMention: \"Index of Economic Freedom\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its world ranking describes how market-oriented an economy is.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006859:35:0", "occurrence_id": "33e8d3f0720ad22c74b3872a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Table 3. LSMS Surveys and Variables Used in PAT Development and Replicated by Authors\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n|County|Data|Obs.|IRIS selected variables|Poverty rate (%)|\n|---|---|---|---|---|\n|Bolivia|2005 Encuesta de<br>Hogares (EH)|4,086|hhsize, hhsize2, age head, age<br>head2, regions, rural,<br>sublease, brick wall, wood<br>wall, dirt floor, cement floor,<br>fridge, radio, tv, dvd, fan, car,<br>number beds, number<br>kitchens, number computers,<br>sheep|24.03|\n|Malawi|2004-2005 Second<br>Integrated Household<br>Survey (IHS2)|11,280|hhsize, hhsize2, age head, age<br>head2, regions, rural, never<br>married, share of adults with<br>out education, share of adults<br>who can read, number of<br>rooms, cement floor,<br>electricity, flush toilet, soap,<br>bed, bike, music player,<br>coffee table, iron, garden,<br>goats|64.78|\n|East<br>Timor|2001 Timor Leste<br>Living Standards<br>Survey (TLSS)|1,800|hhsize, hhsize2, age head, age<br>head2, regions, rattantin wall,<br>leaf roof, concreter or tile<br>roof, number rooms, private<br>water, shared water, toilet is a<br>bowl or bucket, electricity<\n\nMention: \"Hogares (EH)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its household characteristics carry the poverty rate estimates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006861:8:0", "occurrence_id": "a1a3b110433e8f025b8c9079", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: percent), Ukraine (385 percent), Pakistan (380 percent) and India (370 percent) showed the\n\n\ngreatest increases in high‐skilled emigration between 1990 and 2010.\n\n\nA remarkable and underappreciated component of this high‐skilled migration surge is the role of\n\n\nfemales. Figure 1 shows that the stock of high‐skilled female immigrants in OECD countries grew\n\n\nby 152 percent between 1990 and 2010, from 5.7 to 14.4 million. Indeed, in 2010, the stock of\n\n\nhigh‐skilled female migrants surpassed the stock of high‐skilled male migrants. The root causes\n\n\nof this surge have yet to be traced out fully. Africa and Asia experienced the largest growth of\n\n\nhigh‐skilled female emigration, indicating the potential role of gender inequalities and labor\n\n\nmarket challenges in origin countries as push factors. Evidence is accumulating that differences\n\n\nbetween origin and destination countries in women’s rights underpin such flows (Nejad and\n\n\nYoung 2014). The simple correlation between the growth in high‐skilled female migration and\n\n\nthe UNDP Gender Inequality Index is 0.43.\n\n\nMore broadly, research needs to trace out many features like the detailed occupational\n\n\ndistribution of female immigrants. The available data highlight some interesting early patterns.\n\n\nIn 2010, of the 4.2 million employed high‐skilled females in the United States, 16 percent were\n\n\nemployed as Healthcare Practitioners and Technical Occupations, 13 percent in Office and\n\n\nAdministrative Support Occupations and 12 percent in Education, Training, and Library\n\n\nOccupations. These constituted 66, 68 and 63 percent of the total high‐skilled immigrants in\n\n\nthese occupations, respectively. Notably, while only 3 percent of high‐skilled females were\n\n\nemployed in Healthcare Support Occupations, they represented no less than 80% of foreign‐born\n\n\nworkers in these jobs. Interestingly, the OECD reports that some 37 percent of high‐skilled\n\n\nfemales in the United States were employed in occupations in 2010 that were not commensurate\n\n\nwith their level of education. <sup>5</sup\n\nMention: \"the UNDP Gender Inequality Index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Correlated with growth in high-skilled female out-migration.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006862:8:1", "occurrence_id": "11a4bcc2dfaed94ad12c5b76", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  from different racial backgrounds combined with barriers to\n\n\naccessing credit can generate significant degrees of inequality of opportunity (Ferreira\n\n\nand Gignoux, 2011; Marrero and Rodríguez, 2011). There is also evidence of continued\n\n\ndifferential treatment by race in the U.S. labor market, i.e. racial discrimination (see e.g.\n\n\nBertrand and Mullainathan, 2004; Lang and Manove, 2011). Unequal opportunities\n\n\nbetween men and women are equally objectionable (Hederos et al., 2014; Calo-Blanco\n\n\n8 Unfortunately, the IPUMS USA database does not provide data on the individual components of income.\n9 Note that IO measures the extent to which individual income is driven by individual attributes that are\nbeyond the individual’s control such as race and gender.\n10 Occupation prestige is derived from the “occscore” variable that is constructed by IPUMS USA which\nassigns occupational income scores to observed occupations.\n\n\n7\n\nMention: \"the IPUMS USA database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"It does not provide the individual components of income.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006864:6:0", "occurrence_id": "0a9c6d709cb8c93747e5b12e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: intentions of the states involved and their capacity to carry out treaty provisions, which may be\n\n\nlacking (Brochmann 2012).\n\n\nInspired by the works of Brochmann (2012) and De Stefano et al. (2012), we set out to\n\n\nempirically examine whether basins governed by treaties exhibit resilience, in the form of\n\n\ncooperation as opposed to conflict, in comparison to basins (or basins at one point in time) not\n\n\ngoverned by treaties under conditions of water variability. We make a number of contributions to\n\n\nthe existing literature. In contrast to De Stefano et al.’s (2012) study, we subject a number of\n\n\nvariables to a large-n empirical investigation, examining the relationship among the presence, and\n\n\nnumber, of treaties and water variability as well as level of conflict and cooperation among\n\n\nriparians. In comparison to Brochmann’s (2012) study, which uses the same dependent variable,\n\n\nwe not only consider a longer time period (1948-1999 versus 1948-2008), but we likewise examine\n\n\nwhether water variability impacts basin resilience given the existence (or absence) of a treaty and\n\nnumber of treaties. <sup>3</sup> [^3: Due to the most recent update of the TFDD treaty database (Giordano et al. 2014) and of the water event data\nset (De Stefano et al. 2010), this paper included 220 treaties and 755 water events more than the one by\nBrochmann (2012).] In the data section below, we describe our dependent and independent\n\n\nvariables as well as our controls. Along with the description, we provide hypotheses regarding the\n\n\nexpected impact of each variable.\n\n\n<u>Data</u>\n\n\n_<u>Treaties</u>_\n\nResilience and hydropolitical stability are evidenced when riparians evince more cooperative\n\n\nbehavior relative to conflictive behavior, which can be achieved when a basin is governed by a\n\n\nwater treaty (Wolf, 2007; Brochmann, 2012). In order to examine whether river basins governed\n\n\nby treaties exhibit resilience in the context\n\nMention: \"TFDD treaty database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its most recent update changed the recorded treaty counts.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006865:44:0", "occurrence_id": "dd1480926be645ac2da0d2b3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Table S.6: Net Fuel Exports (as a % of GDP)***\n\n\nSource: World Development Indicators and authors’ own calculations. The blue shaded area represents the post-Arab Spring period.\nNet Fuel Exports (as a % of GDP) are calculated as the difference between Fuel Exports (as a % of GDP) and Fuel Imports (as a % of GDP). * n.a. means not\navailable\n\n\n11\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Net fuel exports as a share of output come from this compilation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006866:9:0", "occurrence_id": "5f94525bb8fcec8042cf00f5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _δKt_ = _λ_ ( _Kt_ <sup>_f_</sup> <sup>_−_</sup> <sup>_Kt−_</sup> <sup>1)</sup> (9)\n\n\nHere _Kt_ and _Kt_ <sup>_f_</sup> <sup>represent</sup> <sup>the</sup> <sup>actual</sup> <sup>and</sup> <sup>optimal</sup> <sup>levels</sup> <sup>of</sup> <sup>capital</sup> <sup>at</sup> <sup>time</sup> <sup>_t_</sup> <sup>,</sup> <sup>while</sup> <sup>the</sup> <sup>_λ_</sup>\n\n\n\nparameter represents how much of the gap between these values is bridged in each time period.\n\n\nLower values of _λ_ imply slower rates of adjustment and, thus, higher adjustment costs. As\n\n\nadjustment costs may differ for different levels of desired investment, Chow tests are conducted\n\nat different points along the investment function to test the continuity of its slope for each type\n\n\nof capital.\n\n##### **3 Data**\n\n\n**3.1** **Overview**\n\n\nWe use a panel of firm-level data for manufacturing firms in the Republic of Ireland. These data\n\n\nare collected by the Irish Central Statistics Office (CSO) via the annual Census of Industrial\n\n\nProduction (CIP). Response to the CIP\n\nMention: \"Census of Industrial\n\n\nProduction (CIP)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"An annual enumeration run by the national statistics office.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006871:37:4", "occurrence_id": "7d2a4ead9636ab50493d5ac9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  trade balance by assumption – balancing regional budget constrains.\nAlternatively, we could pin down wages for some given deficit or surplus.\n\n\n###### **Appendix B. Data sources and construction of controls**\n\nThis appendix provides details on the data used and the data sources. A summary of the key\nvariables and the associated descriptive statistics are given in Table 3 in the main text and in\nTable 8 below.\n\n\n**B.1.** **Data sources.**\n\n\n**Plant-level** **data** **and** **industries.** Our analysis is based on the Annual Survey of Manufacturers (asm) Longitudinal Microdata file. This data cover the years from 1990 to 2010. Our\nfocus is on manufacturing plants only. For every plant we have information on: its primary\n6-digit naics code (the codes are consistent over the 20 year period); its year of establishment;\nits total employment; whether or not it is an exporter in selected years; its sales; the number of\nnon-production and production workers; and its 6-digit postal code. The latter, in combination\nwith the Postal Code Conversion files (pccf), allow us to effectively geo-locate the plants by\nassociating them with the geographical coordinate of their postal code centroids.\n\nThe survey frame of the asm has evolved over time. Early in the period, it was relatively\nstable with, on average, about 32,000 plants per sample year. The sample of plants was restricted to those with total employment (production plus non-production workers) above zero,\nand plants must have sales in excess of $30,000. Also, aggregate records were excluded. These\nrecords represent multiple (typically small) plants without latitudes and longitudes. In 2000,\nhowever, the number of plants in the survey increased substantially as the asm moved from\nits own frame to Statistics\n\nMention: \"Annual Survey of Manufacturers\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Plant-level longitudinal file covering this span.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006871:37:4", "occurrence_id": "9a11e47ab48ffab94d2f0f01", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  trade balance by assumption – balancing regional budget constrains.\nAlternatively, we could pin down wages for some given deficit or surplus.\n\n\n###### **Appendix B. Data sources and construction of controls**\n\nThis appendix provides details on the data used and the data sources. A summary of the key\nvariables and the associated descriptive statistics are given in Table 3 in the main text and in\nTable 8 below.\n\n\n**B.1.** **Data sources.**\n\n\n**Plant-level** **data** **and** **industries.** Our analysis is based on the Annual Survey of Manufacturers (asm) Longitudinal Microdata file. This data cover the years from 1990 to 2010. Our\nfocus is on manufacturing plants only. For every plant we have information on: its primary\n6-digit naics code (the codes are consistent over the 20 year period); its year of establishment;\nits total employment; whether or not it is an exporter in selected years; its sales; the number of\nnon-production and production workers; and its 6-digit postal code. The latter, in combination\nwith the Postal Code Conversion files (pccf), allow us to effectively geo-locate the plants by\nassociating them with the geographical coordinate of their postal code centroids.\n\nThe survey frame of the asm has evolved over time. Early in the period, it was relatively\nstable with, on average, about 32,000 plants per sample year. The sample of plants was restricted to those with total employment (production plus non-production workers) above zero,\nand plants must have sales in excess of $30,000. Also, aggregate records were excluded. These\nrecords represent multiple (typically small) plants without latitudes and longitudes. In 2000,\nhowever, the number of plants in the survey increased substantially as the asm moved from\nits own frame to Statistics\n\nMention: \"Postal Code Conversion files\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"Combined with six-digit postal codes to geo-locate plants.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006872:10:1", "occurrence_id": "80209efd4575ad74bbd2f52c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  indicators of conflict are quite high (table\n1). Not surprisingly, households that report having suffered from theft or violence also have higher\nperceptions of insecurity. Similarly, conflict intensity is higher in provinces experiencing a larger\nnumber of conflict-related incidents. While these two correlation coefficients involve indicators\nfrom the same database, there is also a relatively high and statistically significant correlation\nbetween conflict intensity reported by the provincial database and the perception of insecurity\nestimated based on the NRVA survey. This pattern is corroborated by households’ perception of\nsecurity from the Survey of the Afghan People.\n\n\n**4.** **Modeling conflict intensity**\n\n\nThe econometric specification of Model 1 builds on the analyses by Biddle, Friedman, and Shapiro\n(2012) and Hultman, Kathman, and Shanon (2013). The model focuses on changes in conflict\nintensity at the local level and how they are affected by changes in the presence of troops on the\nground. Conflict intensity is measured by the number of casualties. To allow for richer dynamics,\nthe specification considers two periods: the current year and the previous year. It also differentiates\nbetween international troops and Afghan troops, as their effectiveness at reducing conflict may be\ndifferent. Troops and casualties are deflated by provincial population.\n\n\n9\n\nMention: \"provincial database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Conflict intensity by province is recorded here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006872:10:1", "occurrence_id": "27b54954ac3d63484e6fa92b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  indicators of conflict are quite high (table\n1). Not surprisingly, households that report having suffered from theft or violence also have higher\nperceptions of insecurity. Similarly, conflict intensity is higher in provinces experiencing a larger\nnumber of conflict-related incidents. While these two correlation coefficients involve indicators\nfrom the same database, there is also a relatively high and statistically significant correlation\nbetween conflict intensity reported by the provincial database and the perception of insecurity\nestimated based on the NRVA survey. This pattern is corroborated by households’ perception of\nsecurity from the Survey of the Afghan People.\n\n\n**4.** **Modeling conflict intensity**\n\n\nThe econometric specification of Model 1 builds on the analyses by Biddle, Friedman, and Shapiro\n(2012) and Hultman, Kathman, and Shanon (2013). The model focuses on changes in conflict\nintensity at the local level and how they are affected by changes in the presence of troops on the\nground. Conflict intensity is measured by the number of casualties. To allow for richer dynamics,\nthe specification considers two periods: the current year and the previous year. It also differentiates\nbetween international troops and Afghan troops, as their effectiveness at reducing conflict may be\ndifferent. Troops and casualties are deflated by provincial population.\n\n\n9\n\nMention: \"NRVA survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its responses give the perceived insecurity measure.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006872:10:1", "occurrence_id": "40df566517305dc9c6ffb8b3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  indicators of conflict are quite high (table\n1). Not surprisingly, households that report having suffered from theft or violence also have higher\nperceptions of insecurity. Similarly, conflict intensity is higher in provinces experiencing a larger\nnumber of conflict-related incidents. While these two correlation coefficients involve indicators\nfrom the same database, there is also a relatively high and statistically significant correlation\nbetween conflict intensity reported by the provincial database and the perception of insecurity\nestimated based on the NRVA survey. This pattern is corroborated by households’ perception of\nsecurity from the Survey of the Afghan People.\n\n\n**4.** **Modeling conflict intensity**\n\n\nThe econometric specification of Model 1 builds on the analyses by Biddle, Friedman, and Shapiro\n(2012) and Hultman, Kathman, and Shanon (2013). The model focuses on changes in conflict\nintensity at the local level and how they are affected by changes in the presence of troops on the\nground. Conflict intensity is measured by the number of casualties. To allow for richer dynamics,\nthe specification considers two periods: the current year and the previous year. It also differentiates\nbetween international troops and Afghan troops, as their effectiveness at reducing conflict may be\ndifferent. Troops and casualties are deflated by provincial population.\n\n\n9\n\nMention: \"Survey of the Afghan People\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Household perceptions of security corroborate the pattern.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006876:16:7", "occurrence_id": "d3d16caf3afb4cbfd01c7830", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: </mark></u> <u><mark>22.0</mark></u> <u><mark>21.0</mark></u> <u><mark>58.5</mark></u> <u><mark>46.1</mark></u> <u><mark>38.4</mark></u>\nSource: Author’s own elaboration from enterprise surveys\n\n\n15\n\nMention: \"enterprise surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The elaborated figures rest on these firm interviews.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006880:8:1", "occurrence_id": "80af36f0321877ba1303ba07", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \nThe main child health outcomes in this paper, infant mortality and neonatal mortality, are defined as\n\n\na baby dying before reaching his or her first birthday or month of life, respectively. Children born less\n\n\nthan twelve months prior to their mother’s interview are dropped since it is not possible to know whether\n\n\nthey will survive their first birthday. Also, only children born after their country’s year of independence\n\n\nare included in the sample. This gives a sample of 1,770,705 children born to 501,800 mothers between\n\n\n1967 and 2013 in the 56 countries. The sample average infant mortality rate is 7 percent and the neonatal\n\n\nmortality rate is 3.5 percent, indicating that half of the children that die before turning one year in fact\n\n\nlive less than one month. Figure 3 presents the yearly infant and neonatal mortality rates of the group\n\n\nof HIPCs and non-HIPCs. The differences between the groups are greater for infant mortality than for\n\n\nneonatal mortality and the general trend is decreasing for both mortality measures. Since there appears\n\n\nto be some differences in the trends between the two groups, I control for country-specific linear time\n\n\ntrends in the model specification presented below.\n\n\nTable 3 shows descriptive statistics on infant and neonatal mortality across different sub-groups of\n\n\nthe sample as well as other variables in the analysis (see Section 3.2 for more information on the other\n\n\n5I use the standard DHS unless stated otherwise.\n\n\n7\n\nMention: \"DHS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The standard rounds are used unless another is named.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006882:12:0", "occurrence_id": "484ea63c742dd3c8f8568072", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: $107,000 respectively. <sup>23</sup> [^23: Note that at market exchange rates incomes in Malaysia are considerably lower than abroad. Over this period at\nmarket exchange rates one Malaysian Ringgit was worth 0.32 US dollars, at the PPP conversion rate it was worth 0.69\nUS dollars.] Applicants with an existing job offer declared (offered) salaries that are\n\n\neven somewhat higher. Those applicants who were approved by the REP and returned experienced\n\n\naverage (and median) wage gains of about 11 percent. The interquartile range on these wage gains\n\n\nis 0.79 log points, and 42 percent of people who returned experienced wage losses (in terms of\n\n\nPPP).\n\n\nUsing the Malaysian Labor Force Survey (LFS) we can compare expatriates who return\n\n\nunder the REP with the Malaysian workforce as a whole. REP beneficiaries on average earn 8\n\n\ntimes the salary of an average Malaysian, and median wages are 11 times higher (since REP\n\n\napplicants have a smaller left tail of wages). While it is always hard to accurately capture top\n\nincomes, according to the Malaysian LFS, around 80 percent of REP returnees are in the top 1\n\n\npercent of the Malaysian income distribution. This is not simply a result of higher education levels\n\n\nand occupational choices. For comparison we restrict the sample in the LFS to those with at a least\n\n\na Bachelor’s degree, between the ages 25-64 and employed in the same occupations as the REP\n\n\napplicants. In this restricted sample, REP beneficiaries still earn more than 4 times the comparable\n\n\nMalaysians. Once we control for work experience (abroad and in Malaysia), industry, occupation\n\n\nand education, the REP returnees on average still make 2.5 times as much as other Malaysians. It\n\n\nis clear that the program is able to attract applications from very high-skilled and compensated\n\n\nMalaysians.\n\n###### **2.4. Diaspora and Applicants from the United Kingdom and the United States**\n\n\nWe compare the characteristics of applicants\n\nMention: \"Malaysian Labor Force Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Returning expatriates are compared with the national workforce here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006882:12:0", "occurrence_id": "33268d2dc064569fe8196108", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: $107,000 respectively. <sup>23</sup> [^23: Note that at market exchange rates incomes in Malaysia are considerably lower than abroad. Over this period at\nmarket exchange rates one Malaysian Ringgit was worth 0.32 US dollars, at the PPP conversion rate it was worth 0.69\nUS dollars.] Applicants with an existing job offer declared (offered) salaries that are\n\n\neven somewhat higher. Those applicants who were approved by the REP and returned experienced\n\n\naverage (and median) wage gains of about 11 percent. The interquartile range on these wage gains\n\n\nis 0.79 log points, and 42 percent of people who returned experienced wage losses (in terms of\n\n\nPPP).\n\n\nUsing the Malaysian Labor Force Survey (LFS) we can compare expatriates who return\n\n\nunder the REP with the Malaysian workforce as a whole. REP beneficiaries on average earn 8\n\n\ntimes the salary of an average Malaysian, and median wages are 11 times higher (since REP\n\n\napplicants have a smaller left tail of wages). While it is always hard to accurately capture top\n\nincomes, according to the Malaysian LFS, around 80 percent of REP returnees are in the top 1\n\n\npercent of the Malaysian income distribution. This is not simply a result of higher education levels\n\n\nand occupational choices. For comparison we restrict the sample in the LFS to those with at a least\n\n\na Bachelor’s degree, between the ages 25-64 and employed in the same occupations as the REP\n\n\napplicants. In this restricted sample, REP beneficiaries still earn more than 4 times the comparable\n\n\nMalaysians. Once we control for work experience (abroad and in Malaysia), industry, occupation\n\n\nand education, the REP returnees on average still make 2.5 times as much as other Malaysians. It\n\n\nis clear that the program is able to attract applications from very high-skilled and compensated\n\n\nMalaysians.\n\n###### **2.4. Diaspora and Applicants from the United Kingdom and the United States**\n\n\nWe compare the characteristics of applicants\n\nMention: \"Malaysian LFS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its top-percentile shares describe the returnees' income position.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006874:25:1", "occurrence_id": "55d84a3b7ac121e4b2106f16", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Indeed, there is research (see Tybout 2000 for an\noverview) to indicate that the tax burden falls disproportionately on medium-sized\nfirms in developing countries – small firms often stay informal to avoid taxes and large\nfirms are influential enough to obtain special treatment. Given the country context, it is\noften medium-sized firms that employ labor, both skilled and unskilled, in large\nquantities and thus a high tax burden could strongly and negatively affect employment\ngrowth.\n\nSimilarly the model suggests that a 1 percent increase in licensing fees (measured at\nthe firm level) negatively affects employment growth in equal measure. License fees\nand lengthy license procedures are often barriers to entry and expansion, slowing\neconomy-wide growth (Rodrik and Subramanian 2004). Djankov et al (2002) find that\ncountries with heavier regulations (including high licensing fee burdens) suffer from\nhigher corruption and larger informal economies. Similarly, Klapper et al (2004) find\nthat high costs of entry negatively affects entrepreneurial growth across a variety of\nindustries. According to the 2014 Doing Business Survey, Ethiopia ranks 166 (out of\n189 economies) with regard to the ‘starting a business’ indicator, and licensing fees are\nused in the model as a proxy for entry regulations in Ethiopia.\n\n**Infrastructure** : The infrastructure variables (transport, electricity) and business\nenvironment variables (taxes, licensing) are constructed using firm-level data from the\nLMMIS data set. The variables is averaged across firms at the level of industries and\ncities, and over time.\n\nTransport costs negatively affect firm employment growth, after controlling for firm\n\n\n12 The total number of hours taken to pay taxes has increased from 212 hours in 2004 to 306 hours in\n2014.\n\n\n24\n\nMention: \"2014 Doing Business Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its ranking on starting a business places one country low.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006874:25:1", "occurrence_id": "b9c61565d694264b2c993fbb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Indeed, there is research (see Tybout 2000 for an\noverview) to indicate that the tax burden falls disproportionately on medium-sized\nfirms in developing countries – small firms often stay informal to avoid taxes and large\nfirms are influential enough to obtain special treatment. Given the country context, it is\noften medium-sized firms that employ labor, both skilled and unskilled, in large\nquantities and thus a high tax burden could strongly and negatively affect employment\ngrowth.\n\nSimilarly the model suggests that a 1 percent increase in licensing fees (measured at\nthe firm level) negatively affects employment growth in equal measure. License fees\nand lengthy license procedures are often barriers to entry and expansion, slowing\neconomy-wide growth (Rodrik and Subramanian 2004). Djankov et al (2002) find that\ncountries with heavier regulations (including high licensing fee burdens) suffer from\nhigher corruption and larger informal economies. Similarly, Klapper et al (2004) find\nthat high costs of entry negatively affects entrepreneurial growth across a variety of\nindustries. According to the 2014 Doing Business Survey, Ethiopia ranks 166 (out of\n189 economies) with regard to the ‘starting a business’ indicator, and licensing fees are\nused in the model as a proxy for entry regulations in Ethiopia.\n\n**Infrastructure** : The infrastructure variables (transport, electricity) and business\nenvironment variables (taxes, licensing) are constructed using firm-level data from the\nLMMIS data set. The variables is averaged across firms at the level of industries and\ncities, and over time.\n\nTransport costs negatively affect firm employment growth, after controlling for firm\n\n\n12 The total number of hours taken to pay taxes has increased from 212 hours in 2004 to 306 hours in\n2014.\n\n\n24\n\nMention: \"LMMIS data set\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Firm-level records on taxes and licensing averaged by industry and city.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006884:21:1", "occurrence_id": "77f97fd6eb5b3ceda750609c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  government reduced the\ncorporate and personal income tax rates in 2010, only to raise them substantially as the crisis hit them\nlater. Implicit tax rates on labor increased in all countries after 2009 (Eurostat estimation). Personal\nincome tax rates fell shortly thereafter again, while corporate taxes rose by 2013. Importantly, in order to\nimprove taxation related incentives that bolster asset price cycles, property taxes were changed too. For\nexample, Ireland had abolished the residential property tax in 1978, and since 2013 has adopted a\nresidential property tax (Slack and Bird, 2014). <sup>36</sup> Spain restored the wealth tax. Portugal completed a\n\n\n36 In 2006 an OECD report recommended the introduction of a property tax. The absence of a property tax, apart from stamp\nduty on purchases of second‐hand homes, the halving of capital gains tax to 20%, significant tax incentives which favored\n\n\n20\n\nMention: \"Eurostat estimation\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Implicit tax rates on labour across countries rise after that year here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006885:47:2", "occurrence_id": "db2d1eb06068084592422b0e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .470**\n<u>(0.106)</u> <u>(0.076)</u> <u>(0.110)</u>\n_Note_ . Estimates in bold face are significant at the 5% level or lower. Bootstrap standard errors\nare in parentheses. The dependent variable is the log of hourly earnings. Panel A includes\npotential experience, squared term for experience, and occupation dummies. The sample\nincludes 25- to 64-year-olds (both men and women) in self-employment or wage work.\n_Source_ : STEP Surveys (2014).\n\n\n46\n\nMention: \"STEP Surveys (2014)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Skills of employed adults across these countries are measured here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006886:16:0", "occurrence_id": "6e146f3c55e93506cc9e6dd8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 4. Intensity of computer use at work, workers ages 25 to 64, by employment type**\n**(percent distribution)**\n\n\n_Source_ : STEP Surveys (2014).\n\n\nWith regard to the distribution of workers by the complexity of their computer use, we find\n\n\nthat the largest proportion of computer users at work are using basic applications involving a word\n\n\nprocessor, spreadsheets, making presentations, or doing data entry. This group accounts for 40\n\n\nto 50 percent of computer users in most of the countries with the exception of Kenya and Bolivia\n\n\n(see Figure 5). In Kenya and Bolivia we find that the largest proportion of respondents who use\n\n\ncomputers on their job are using advanced programming skills. The proportion of workers\n\n\nhandling tasks involving basic programming skills like advanced functions in spreadsheets or\n\n\nbookkeeping applications is also substantial across all the countries (from about 15 percent in\n\n\nBolivia and Vietnam to more than 30 percent in Kenya and Ukraine). Finally, we find that less than\n\n\n10 percent of respondents who use computers at work limit their computer usage to browser\n\nbased tasks.\n\n\n15\n\nMention: \"STEP Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Workers by employment type and competence complexity come from here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006887:20:0", "occurrence_id": "48f1f3caf40e105bc5cc88e9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: tribution of transit migrants can now be expressed as\n\n\n       \n\n\n,\n\n\n\n\n\n\n_k̸_ = _U_\n\n\n\n\n\n\n_j̸_ = _i_\n\n\n\n_M_ <sup>_s,i,j,U_</sup> log _p_ <sup>_s,i,j,U_</sup> ( _θ_ ) + _M_ <sup>_s,i,j,S_</sup> log\n\n\n\n\n   \n_p_ <sup>_s,i,j,k_</sup> ( _θ_ )\n\n\n\n(10) log _L_ 1( _θ_ ) =\n\n\n\n\n\n\n_i_\n\n\n\nwhere _k̸_ = _U_ means the destination _k_ is any country other than USA.\n\n\nOur next dataset, the 2010 DIOC provides data on the number of people who migrated within\n\n\nthe last 10 years to OECD destination countries. This allows us to calculate the number of\n\n\nnatives who moved from _j_ to _k_ within the last 10 years if _k_ is an OECD country. We denote\n\nthis matrix _M_ <sup>_s,j,j,k_</sup> . Next, we define _M_ <sup>_s,j,j,N_</sup> as the number of people who move to non-OECD\n\ndestinations (labelled by _N_ collectively) and _M_ <sup>_s,j,j,j_</sup> as the number of people who stayed in\n\n\ntheir birth country _j_ and did not migrate to any country. Both of these are 40 _×_ 1 vectors\n\n\nfor each skill group _s_ . In summary, natives (who were born in _j_ and living in _j_ ) are divided\n\n\ninto the following groups: Staying at\n\nMention: \"2010 DIOC\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Counts of people who migrated within the last decade to destination countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006889:40:0", "occurrence_id": "82ced5a685d1ec28b85829f2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: In the simple theoretical framework presented in Section 3, we emphasize the imported input channel\n\n\nas the main mechanism through which trade liberalization affects firms’ decision to upgrade foreign\n\n\ntechnology. For the sake of simplicity we did not take into account the export side of the story and\n\n\nthe effects of trade liberalization through variations in trade variable costs affecting final goods that are\n\n\nalready well-documented in the theoretical literature (Yeaple, 2005; Bustos, 2011).\n\n\nExpansion of export opportunities due to foreign demand shocks might also increase the incentives\n\n\nfor firms’ to upgrade foreign technology embodied in imported capital goods. Moreover, importing\n\n\nintermediate inputs might lead to higher exports as documented in the empirical literature (Feng et al.,\n\n\n2016; Bas, 2012; Bas and Strauss-Kahn, 2015). Higher export profits would allow to overcome the fixed\n\n\ncost of importing capital goods. If input tariff changes are positively correlated with export performance\n\n\nor with variations in output tariffs set by India’s trading partners, our previous empirical findings might\n\n\nbe just picking up the effects of foreign demand shocks.\n\n\nThe industry-year fixed effects included in the estimations of the previous section already address\n\n\nthis issue since they capture all unobservable shocks at the industry level varying over time. In this\n\n\nsection, we provide additional evidence that foreign demand shocks at the sectoral level captured by\n\n\nexport tariffs are not picking up our results. We control for this alternative explanation by including\n\n\nin the previous specifications the average effectively applied tariff at the 3-digit NIC industry level set\n\n\nby the rest of the world to India (export tariff) during the 1989-1997 period from WITS dataset (World\n\n\nBank). Columns (1) to (3) of Table A5 in the appendix report the results. The effect of export tariff\n\n\nis negative but not significant. The coefficient of interest on the input tariffs remains robust and stable\n\n\nin all specifications when we take into account the role of foreign demand. This finding suggests that\n\n\nthe supply side\n\nMention: \"WITS dataset\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Tariffs faced by one country's exports across this period come from here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006892:11:0", "occurrence_id": "9e181aab4befb5978fc943d8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ### 3. Data and calibration\n\nThis section explores how we assess socioeconomic resilience at the national level. To ensure\ntransparency and replicability – and to allow the reader to test various assumptions – the code that\ncalculates the resilience indicators based on various data sets and that creates all figures presented below\nis available online.\n\n##### 3.1. Poverty definition\n\n\nWe need first to describe inequality and poverty levels in the affected region. We define the poor as the\nindividuals in the bottom quintile in terms of consumption or income. Therefore, the parameter �� is\nequal to 20%. To estimate �� and ��, we use the World Development Indicators database, which provides\nthe income share of the bottom 20%. Figure 3 shows the result, highlighting the large variability – and\nlack of correlation – between inequality and national income level.\n\n\n**Figure 3. Income share of the bottom 20 percent, in percent of GDP, as a function of GDP.**\n\n##### 3.2. Asset vulnerability\n\n\nA few household surveys have examined the percentage of asset losses (that is, asset vulnerability) for\nhouseholds at different income levels. Several caveats are needed before interpreting the above\nestimates of asset vulnerabilities. First, each study has a different definition of “poor” and “non‐poor”,\nwhich does not always fit our 20/80 categorization. Second, vulnerability depends on the type of hazard\nand context in which it occurs. Even within the same country (e.g. Bangladesh in the table below),\nmeasures vary greatly. Third, the methods through which vulnerability estimates are captured also differ\nin each study (hence the range of vulnerabilities varies widely across studies).\n\n\n10\n\nMention: \"World Development Indicators database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"It supplies the income share of the bottom quintile.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006896:8:0", "occurrence_id": "2dc1a33708d581755fb8807a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: (573 firms in Kenya, 289 firms in Tanzania, and 366 in Uganda) with 5 or more employees. The\n\n\nEnterprise Surveys were conducted in all three countries using the same sampling methodology –\n\n\nstratified random sampling – along with a common questionnaire. Three levels of stratification\n\n\nwere employed for each country: firm-size, sector of activity, and location within the country.\n\n\nWeights are used to ensure that the sample is representative of the non-agricultural, non-mining\n\nprivate sector of the economy. <sup>5</sup> [^5: Certain sectors are not covered including: agriculture, mining, and some service sectors such as education,\nhealth, and financial intermediaries. More information on the Enterprise Surveys global methodology is available\non the website http://www.enterprisesurveys.org/.]\n\n\nBefore exploring the relationship between mobile money use and investment, we provide some\n\n\nbasic statistics to explore why firms have or have not adopted mobile money, and what types of\n\n\nfirms tend to adopt mobile money for transactions. The latter has already been explored in depth\n\n\n(Gosavi, 2015) and thus we do not delve into details in the analysis. So why has mobile money\n\n\nbeen adopted by firms? Through the Enterprise Surveys, managers and owners of firms in Kenya,\n\n\nTanzania, and Uganda provided responses as to why they have or have not adopted mobile money.\n\n\nDespite some heterogeneity in the ranking, the most cited reasons for mobile money adoption\n\n\nacross the three countries are reducing costs and risks of transactions and satisfying costumers’\n\n\nrequests (Table 1). Concerning the reasons for non-adoption, the majority of non-adopter firms in\n\n\nKenya cited large payments that were difficult to undertake using mobile money. In Tanzania and\n\n\nUganda the most frequently cited reason is non-adoption by customers or suppliers.\n\n\nWhat types of firms tend to adopt mobile money? In table 2 we present comparisons of averages\n\n\nbetween firms that use mobile money for transactions and those that do not. As expected, firms\n\n\nthat use cell phones for operations are more likely to adopt mobile\n\nMention: \"Enterprise Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Identical stratified sampling across three countries' firm samples.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006902:6:2", "occurrence_id": "be4fd980e2ac2826c26d46c1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: , this series seemed to have little signs of substantial measurement error (though we\n\n\n3. The median endyear of changes in poverty in this sample is 2002.\n4. Without this restriction, around one-third of observations would be derived from Eastern Europe and Central\nAsia and another third from Latin America and the Caribbean.\n\n5. Personal communications with the researchers managing the POVCAL database confirmed that CPI\nadjustments were made for approximately 20 countries for which the weight on food products was deemed to be\nexcessively low. For Africa, the aforementioned World Bank study used the assumption that differences in the food\nbudget share among demographically similar households with the same level of consumption at different points in\ntime would indicate the CPI’s mismeasurement of the true cost of living. Strikingly, the official CPIs seem to have\noverestimated changes in the cost of living for the poor in 13 of the 16 countries examined.\n\n\n5\n\nMention: \"POVCAL database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Managers confirmed price adjustments were made for about twenty countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006911:11:1", "occurrence_id": "9a37fd95a8adae5fff2da2ac", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  reducing the gender enrollment gap since\nimplementing the scholarship program. Figure 3 shows Ministry of Education data on\nenrollment in grades 7–12, aggregated across all public schools in regions 2–6. Panels\n(a)–(b) rescale time so that _t_ = 0 corresponds to the first year of program receipt. Panel (a)\nshows that female enrollment increased relative to the pre-treatment trend after\nintroduction of the program, while male enrollment fell. <sup>12</sup> Panel (b) shows the resulting\nincrease in the female enrollment share. This is suggestive evidence of the program’s\neffect on enrollment, consistent with Gajigo (2014), albeit subject to the potential bias in\nenrollment data discussed previously. Panel (c), which uses calendar time and\ndisaggregates the data by grade, shows that the female enrollment percentage increased\nover time for all grades. It also shows that females comprise a lower share of enrollment\nas grade level increases (with only a few exceptions), meaning that females will be\nunder-represented among test takers relative to their enrollment shares in their\ncorresponding schools.\n\n\nFigure 3(a)-(c): Secondary school enrollment, by gender (a) Enrollment, (b) Female\nenrollment proportion, (c) Female enrollment proportion, by grade. Figure shows\nsecondary school (grade 7–12) enrollment. Data source: Gambia Ministry of Basic and\nSecondary Education.\n\n\n10\n\nMention: \"Ministry of Education data on\nenrollment\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Enrolment in upper grades aggregated across public schools in named regions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006923:24:0", "occurrence_id": "f3a34c2b7ebbaeecab1be896", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Croke et al. (2016)\n\n\nwould only involve the addition of the one multiple dose study the authors identify\n\n\nbut classify as separate from the MDA studies – Stephenson et al. (1993). <sup>23</sup> [^23: Note that we believe that Stephenson et al. (1993) should be included in the set of MDA studies, since\nindividuals were treated without first being screened, as discussed in Appendix A.3.]\n\n\n2. The 2012 Cochrane Review changed this approach, however, introducing a\n\n\ndistinction that effectively distinguished between test-and-treat and MDA studies\n\n\n(Taylor-Robinson et al. 2012). Applying this classification used in the 2012\n\n\nCochrane Review to the studies in TMSDG also leads to the inclusion of\n\n\nStephenson et al. (1993), which was classified as an MDA study by Taylor\n\nRobinson et al. 2012 (or, as a “target population treated” study, using the language\n\n\nin that review) rather than a test-and-treat study.\n\n\nThus, applying either of these procedures from Cochrane Reviews prior to TMSDG results\n\n\nin the addition of Stephenson et al. (1993) to the TMSDG sample. With Stephenson et al.\n\n\n(1993) classified in this way but otherwise using the data set in the TMSDG study, the null\n\n\nhypothesis of a common zero effect is strongly rejected with a p-value of 0.009 (see Table\n\n\n2, row 2).\n\n\n**5.2** **Testing the hypothesis of a common zero effect**\n\n\nWe now turn to the full sample of 22 treatment effect estimates. Table 3 shows that the null\n\n\nhypothesis of a common zero effect is rejected at p<0.001 in the full sample, in areas with\n\n\nprevalence above the thresholds at which WHO recommends MDA (20% prevalence), and\n\n\nin areas where it recommends multiple annual dose MDA (50% prevalence). This result is\n\n\nnot reliant on the addition of any one study to the TMSDG sample: any one of six updates\n\n\nleads to rejection\n\nMention: \"TMSDG study\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its dataset is reclassified to test whether a common effect can be rejected.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006927:10:0", "occurrence_id": "716c33e6ef3aeedf30fd6af7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: transport and communications; financial intermediation; real estate and business activities; public\n\n\nadministration and defense; education; health and social work; other service activities; and private\n\nhouseholds).\n\n\nInformation on socio-demographic characteristics, such as individuals’ age and education are\n\nincluded in the BLFS as well. This allows the construction of a variable for potential labor market\n\n\nexperience. Information on an individual’s location ( _dzongkhag_ and urban or rural area) and his or her\n\ngender and nationality (Bhutanese or not) can also be used. Lastly, the questionnaire contains a series\n\n\nof questions on households’ wealth and exact demographic composition. For precise definitions of all\n\nvariables used in the empirical analysis of Section 4, cf. Table 1.\n\n\n**3.2.** **Analytical Strategy and Econometric Model**\n\n\nThe microeconomic literature has used a wide range of methods to investigate wage differentials\n\nbetween public and private sector workers (cf. Lausev, 2014, for an overview). Among the most\n\n\ncommonly used methods are “single equation” models and “double equation” models. Single equation\n\nmodels are basically wage regressions in the tradition of Mincer (1974) for the whole economy that\n\n\ninclude a public sector dummy variable. For double equation models separate wage regressions are\n\nrun for the public and private sector samples. The main advantage of single equation models is that\n\n\nthey make corrections for self-selection into labor force participation and specific sectors\n\nstraightforward. In the context of analyzing wage differentials between public and private sector,\n\n\ncontrolling for such self-selection is of paramount importance. On the other hand, double equation\n\nmodels are more flexible and allow intercepts and returns to productive characteristics to differ across\n\n\nsectors. They can also be an intermediate step for counterfactual decompositions, i.e., for an\n\ninvestigation of the proportion of the differences in the average levels of hourly wages between public\n\n\nand private sector workers that is\n\nMention: \"the BLFS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its demographic variables allow a potential labour experience measure.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006934:17:0", "occurrence_id": "9523e606ed02b213f6703537", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <u>Table</u> <u>1</u> <u>Remittances</u> <u>-</u> <u>Top</u> <u>recipients</u> <u>in</u> <u>2012</u>\n<u>Ranking</u> <u>Country</u> <u>Billion</u> <u>(US$)</u>\n1 India 70\n2 China 65\n3 Philippines 24\n4 Mexico 24\n5 Nigeria 21\n6 Egypt 18\n7 Pakistan 14\n8 Bangladesh 14\n9 Vietnam 9\n<u>10</u> <u>Lebanon</u> <u>7</u>\nSource: World Bank Report\n\n\n<u>Table</u> <u>2</u> <u>Exchange</u> <u>Rate</u> <u>Regimes</u> <u>-</u> <u>Coarse</u> <u>Classifcationi</u>\n<u>Regime</u> <u>Class</u> <u>Description</u>\n1 No separate legal tender\n1 Pre announced peg or currency board arrangement\n1 Pre announced horizontal band that is narrower than or equal to +/-2%\n1 De facto peg\n2 Pre announced crawling peg\n2 Pre announced crawling band that is narrower than or equal to +/-2%\n2 De factor crawling peg\n2 De facto crawling band that is narrower than or equal to +/-2%\n3 Pre announced crawling band that is wider than or equal to +/-2%\n3 De facto crawling band that is narrower than or equal to +/-5%\n3 Moving band that is narrower than or equal to +/-2% (i.\n\nMention: \"World Bank Report\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The country ranking in the table is taken from this publication.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006941:7:7", "occurrence_id": "6bde97208cf779b89c0c221f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <br>1.1%<br>0.2%<br>No<br>          Others<br>1.5%<br>0.3%<br>No<br>          Non-tax accounts<br>-0.1%<br>0.0%<br>No<br>     Non-tax revenues<br>13.3%<br>2.8%<br>No<br>**Contributions to social security**<br>**6.8%**<br>**1.4%**<br>               From employees<br>6.6%<br>1.4%<br>Yes<br>               From employers<br>0.2%<br>0.0%<br>No<br>**Grants**<br>**0.2%**<br>**0.1%**<br>No|\n||||||\n||||||\n||||||\n||||||\n||||||\n||||||\n||||||\n||||||\n||||||\n||||||\n||||||\n||||||\n||||||\n||||||\n||||||\n||||||\n||||||\n||||||\n||||||\n||||||\n\n\n\n_Source_ : Author’s elaboration based on Direccion de Presupuestos (2014) and the 2009-2015 data on annual tax revenue published by\nChile’s Internal Revenue Service (Servicio de Impuestos Internos, SII).\n\n\n**3\n\nMention: \"Chile’s Internal Revenue Service\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Annual tax revenue across this span is published by this agency.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006946:29:3", "occurrence_id": "ff62733a05f082c5047409bc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: *\n(0.003) (0.003) (0.003)\n\n_>_ 1 deaths in past 6 months -0.015** -0.012* -0.012*\n<u>(0.007)</u> <u>(0.007)</u> <u>(0.007)</u>\n\n\nSource: Author’s calculations from the dataset constructed from the FEPB database and the DoFE database\nNote: The table shows how the migration effect of a death of a migrant in a district-destination cells changes with the number\nof deaths in the district-destination cell in the past six months. The estimates reported are _β_ and _δ_ coefficients from Equation\n(5). All specifications control for the effect of the interaction between current death and actual underlying death rates in the\ndestination countries.\nThe first two columns present the estimates for migrant outflow in the subsequent 6 months, columns (3) and (4) present the\nestimates for migrant outflow in the subsequent 9 months, and columns (5) and (6) present the estimates for migrant outflow\nin the subsequent 12 months.\nOdd numbered columns show the interaction with the number of deaths in the district-destination cell in the past 6 months.\nEven numbered columns show the interact with whether there has been more than one death in district-destination cell in the\npast 6 months.\nThe top panel shows the effect on migration outflow from in the same district-destination cells as the migrant death. The\nsecond panel shows the effect on migration outflow to destinations other than that of the migrant death. The third panel shows\nthe effect on migration outflow from neighboring districts to the same destination as the migrant death.\nFor all panels, standard errors are reported in parenthesis and are clustered at the district level. <sup>_∗∗∗_</sup> : _p_ _<_ 0\n\nMention: \"FEPB database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Combined with the deaths register to build the district-level dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006946:29:3", "occurrence_id": "1e463a832edc376ce76633f8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: *\n(0.003) (0.003) (0.003)\n\n_>_ 1 deaths in past 6 months -0.015** -0.012* -0.012*\n<u>(0.007)</u> <u>(0.007)</u> <u>(0.007)</u>\n\n\nSource: Author’s calculations from the dataset constructed from the FEPB database and the DoFE database\nNote: The table shows how the migration effect of a death of a migrant in a district-destination cells changes with the number\nof deaths in the district-destination cell in the past six months. The estimates reported are _β_ and _δ_ coefficients from Equation\n(5). All specifications control for the effect of the interaction between current death and actual underlying death rates in the\ndestination countries.\nThe first two columns present the estimates for migrant outflow in the subsequent 6 months, columns (3) and (4) present the\nestimates for migrant outflow in the subsequent 9 months, and columns (5) and (6) present the estimates for migrant outflow\nin the subsequent 12 months.\nOdd numbered columns show the interaction with the number of deaths in the district-destination cell in the past 6 months.\nEven numbered columns show the interact with whether there has been more than one death in district-destination cell in the\npast 6 months.\nThe top panel shows the effect on migration outflow from in the same district-destination cells as the migrant death. The\nsecond panel shows the effect on migration outflow to destinations other than that of the migrant death. The third panel shows\nthe effect on migration outflow from neighboring districts to the same destination as the migrant death.\nFor all panels, standard errors are reported in parenthesis and are clustered at the district level. <sup>_∗∗∗_</sup> : _p_ _<_ 0\n\nMention: \"DoFE database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its death registrations identify the events behind the migration effect.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006947:20:0", "occurrence_id": "33a1cb58c9b02f71b6a10616", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 28 percent. According to the estimates, services exports would fall by 62 percent, which implies a 14 and\n46 percentage points drops with respect to the intermediate and “Norway” scenarios. Intermediates value\nadded exports would decrease by about a quarter and foreign value added in the UK’s exports would\ndecrease by one‐third.\n\n\nTo better understand these magnitudes, we use the 2011 data from WIOD to obtain the dollar equivalent\nof the losses for the three different scenarios. It should be noted that these back‐of‐the‐envelope\ncalculations account for the partial effects and do not have general equilibrium or welfare implications\nsince they do not account for possible re‐direction of exports to non‐EU countries (i.e. trade diversion and\ndeflection), price adjustments or changes in wages. Although the largest percentage changes are for\nservice exports, in terms of dollar equivalents we find that export of goods would decrease the most in all\nthree scenarios. The “Norway” case, in which the UK signs an agreement similar to the EEA, suggests that\nexport of goods would decrease by $27.447 billion, while services and value added exports would drop by\n$19.906 billion and $16.445 billion respectively. To put these numbers in perspective, the impact on total\n(goods and services) trade is equal to 1.83 percent of the UK’s GDP and 0.63 percent for value added\nexports. In the worst case scenario, we find that goods exports would decrease by $114.363 billion, which\nis equivalent to 4.41 percent of the UK’s GDP. <sup>11</sup> [^11: The data on GDP are from the World Bank.] For services and value added exports, the values are\nlower, around $78 billion each, which accounts for 3 percent of GDP.\n\n\n_Table 6:_ Changes in UK’s bilateral trade with the EU under different scenarios\n\n\n\n\"no‐agreement\"\n\nscenario\n\nMention: \"WIOD\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Its input-output tables convert model losses into dollar equivalents.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006949:13:2", "occurrence_id": "71137130d9b5d8e50527621a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: >_\n\n- _σ−_ 1 <sup>�</sup> _σ−_ 1\n\n\n\n=\n\n\n\n_S_\n\n\n\n_s_ =1\n\n\n\n_<u>TFPRs</u>_\n_TFPRsi_\n\n\n\n(13)\n\n\n\nEq. 13 shows how within industry misallocation of resources leads to a lower measured TFP.\n\n##### **5 Data Description**\n\n\nOur analysis exploits census data for manufacturing firms in each of the four SSA countries we study:\n\n\nCˆote d’Ivoire (2003-2012), Ethiopia (2011), Ghana (2003), and Kenya (2010). These countries provide\n\n\ncomprehensive and comparable census data. The censuses are nationally representative and both small\n\n\nand large firms in the formal sector are adequately included in all countries. In all four countries, the\n\n\ndata are restricted to manufacturing sector (ISIC Rev 3 15-37).\n\n\nIn what follows, we describe each country’s datasets.\n\n\n**Cˆote** **d’Ivoire** The data source for Cˆote d’Ivoire is balance sheets and income statements associated\n\n\nwith tax reporting. The data is available for all registered firms in the country and contains detailed\n\n\nbalance sheet information on firms’ revenue, employment, cost of labor, book value of fixed assets,\n\n\nintermediate inputs and other firm characteristics. All registered firms are required to report their\n\n\nfinancial statements to the National Statistics Institute (INS), the tax administration (DGI), the court\n\n\nof justice, and the Central Bank (BCEAO), which are reported under the West Africa accounting system\n\n\nstandards, Systeme Comptable Ouest Africain (SYSCOA). INS processed the data after firms hand in\n\n\nhard copies of their forms between March and June following the closing of the fiscal year in December.\n\n\nThe Cˆote d’Ivoire data cover 2003 to 2012.\n\n\n7Following Hsieh and Klenow (2009\n\nMention: \"census data for manufacturing firms\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Manufacturing censuses across four countries carry the productivity analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006949:13:2", "occurrence_id": "ea58a22d382da7b6e34c8a3e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: >_\n\n- _σ−_ 1 <sup>�</sup> _σ−_ 1\n\n\n\n=\n\n\n\n_S_\n\n\n\n_s_ =1\n\n\n\n_<u>TFPRs</u>_\n_TFPRsi_\n\n\n\n(13)\n\n\n\nEq. 13 shows how within industry misallocation of resources leads to a lower measured TFP.\n\n##### **5 Data Description**\n\n\nOur analysis exploits census data for manufacturing firms in each of the four SSA countries we study:\n\n\nCˆote d’Ivoire (2003-2012), Ethiopia (2011), Ghana (2003), and Kenya (2010). These countries provide\n\n\ncomprehensive and comparable census data. The censuses are nationally representative and both small\n\n\nand large firms in the formal sector are adequately included in all countries. In all four countries, the\n\n\ndata are restricted to manufacturing sector (ISIC Rev 3 15-37).\n\n\nIn what follows, we describe each country’s datasets.\n\n\n**Cˆote** **d’Ivoire** The data source for Cˆote d’Ivoire is balance sheets and income statements associated\n\n\nwith tax reporting. The data is available for all registered firms in the country and contains detailed\n\n\nbalance sheet information on firms’ revenue, employment, cost of labor, book value of fixed assets,\n\n\nintermediate inputs and other firm characteristics. All registered firms are required to report their\n\n\nfinancial statements to the National Statistics Institute (INS), the tax administration (DGI), the court\n\n\nof justice, and the Central Bank (BCEAO), which are reported under the West Africa accounting system\n\n\nstandards, Systeme Comptable Ouest Africain (SYSCOA). INS processed the data after firms hand in\n\n\nhard copies of their forms between March and June following the closing of the fiscal year in December.\n\n\nThe Cˆote d’Ivoire data cover 2003 to 2012.\n\n\n7Following Hsieh and Klenow (2009\n\nMention: \"balance sheets and income statements\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Attached to tax reporting for every registered firm in that country.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006951:31:0", "occurrence_id": "7e22205285fe12abcf91e218", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ###### **Appendices** **A.1 Robustness Checks**\n\nIn this section I conduct robustness checks on the main quantitative findings of the\nmodel with respect to two important parameters – the labor share of income - and the\ndepreciation rate of physical capital �. For the benchmark exercise, the labor share of income\nwas assumed to be 51 percent which was sourced from the GTAP database and the\ndepreciation rate was picked to be 3.2 percent which was sourced from the PWT 8.1. The labor\nshare of income for Bangladesh as sourced from the GTAP database is close but somewhat\nlower compared to the mean/median of lower‐middle income countries for which the data is\navailable in the PWT 8.1. In order to test the sensitivity of the main results of the model, I\nrepeat the quantitative exercises with a higher value of labor share of income. The chosen\nhigher value of the labor share – 55 percent, roughly corresponds to the mean labor share of\nlower‐middle income countries. In addition, I also consider two other extreme values of labor\nshare of income – 30 percent and 60 percent, to arrive at a broader range of results.\n\n\nThe following table compares the findings of section 3.1 when alternative values of\nlabor share of income are used for quantitative analysis. For clarity, the results are presented\nusing average growth rate over five year periods. The first set of results refers to the\nbenchmark exercise followed by simulations with other alternatives for labor share.\n\n\nThe resulting growth rate under each scenario is higher than the growth rate in the\ncorresponding scenario in the benchmark exercise when the labor share is reduced to 30\npercent. Equation (8) helps in understanding why this happens. The labor share of income\ndetermines the relative impact of factors pertaining to labor accumulation with respect to\nfactor pertaining to capital accumulation. A higher value of labor share increases the growth\nimpact from the former, for example\n\nMention: \"GTAP database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The labour share of income is sourced from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006951:31:0", "occurrence_id": "b001d99e0090a7bf0e9f9758", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ###### **Appendices** **A.1 Robustness Checks**\n\nIn this section I conduct robustness checks on the main quantitative findings of the\nmodel with respect to two important parameters – the labor share of income - and the\ndepreciation rate of physical capital �. For the benchmark exercise, the labor share of income\nwas assumed to be 51 percent which was sourced from the GTAP database and the\ndepreciation rate was picked to be 3.2 percent which was sourced from the PWT 8.1. The labor\nshare of income for Bangladesh as sourced from the GTAP database is close but somewhat\nlower compared to the mean/median of lower‐middle income countries for which the data is\navailable in the PWT 8.1. In order to test the sensitivity of the main results of the model, I\nrepeat the quantitative exercises with a higher value of labor share of income. The chosen\nhigher value of the labor share – 55 percent, roughly corresponds to the mean labor share of\nlower‐middle income countries. In addition, I also consider two other extreme values of labor\nshare of income – 30 percent and 60 percent, to arrive at a broader range of results.\n\n\nThe following table compares the findings of section 3.1 when alternative values of\nlabor share of income are used for quantitative analysis. For clarity, the results are presented\nusing average growth rate over five year periods. The first set of results refers to the\nbenchmark exercise followed by simulations with other alternatives for labor share.\n\n\nThe resulting growth rate under each scenario is higher than the growth rate in the\ncorresponding scenario in the benchmark exercise when the labor share is reduced to 30\npercent. Equation (8) helps in understanding why this happens. The labor share of income\ndetermines the relative impact of factors pertaining to labor accumulation with respect to\nfactor pertaining to capital accumulation. A higher value of labor share increases the growth\nimpact from the former, for example\n\nMention: \"PWT 8.1\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The depreciation rate is taken from this version.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006954:25:1", "occurrence_id": "40612e9ada7a8e2cab462e83", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: -exporters’ stock markets to\n\n\nchanges in oil prices. Quite surprisingly, a lower WTI oil price hurts the dollar value\n\n\nof the Gulf States’ stock markets but the impact is statistically insignificant: every\n\n\n10% decline in WTI oil price, MSCI Gulf State index drops by 0.55%. The\n\n\nmagnitude is relatively small and not statistically different to zero. Conventional\n\n\n16 <mark>Euro Area; Canada; Japan; Mexico; China; United Kingdom; Taiwan, China; Republic of Korea;</mark>\n<mark>Singapore; Hong Kong SAR, China’ Malaysia; Brazil; Switzerland; Thailand; Philippines;</mark>\n<mark>Australia; Indonesia; India; Israel; Saudi Arabia; Russian Federation; Sweden; Argentina; República</mark>\n<mark>Bolivariana de Venezuela; Chile; and Colombia.</mark> Data are from the Federal Reserve.\n\n\n24\n\nMention: \"MSCI Gulf State index\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its response to oil price declines is estimated here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006954:25:1", "occurrence_id": "28434c1eac5ec19ab6b40ff5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: -exporters’ stock markets to\n\n\nchanges in oil prices. Quite surprisingly, a lower WTI oil price hurts the dollar value\n\n\nof the Gulf States’ stock markets but the impact is statistically insignificant: every\n\n\n10% decline in WTI oil price, MSCI Gulf State index drops by 0.55%. The\n\n\nmagnitude is relatively small and not statistically different to zero. Conventional\n\n\n16 <mark>Euro Area; Canada; Japan; Mexico; China; United Kingdom; Taiwan, China; Republic of Korea;</mark>\n<mark>Singapore; Hong Kong SAR, China’ Malaysia; Brazil; Switzerland; Thailand; Philippines;</mark>\n<mark>Australia; Indonesia; India; Israel; Saudi Arabia; Russian Federation; Sweden; Argentina; República</mark>\n<mark>Bolivariana de Venezuela; Chile; and Colombia.</mark> Data are from the Federal Reserve.\n\n\n24\n\nMention: \"Data are from the Federal Reserve\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Exchange rate and macro series for the listed countries come from this source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006929:23:0", "occurrence_id": "be87ee54e7c72bb69ca4e3aa", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 6: Average Job Quality Ratio Over Time\n\n\n\n\n\n_Notes_ : Authors’ calculations using SEDLAC data. Coun\n\ntries included in each year differ based on availability, but\n\n\ncountry fixed effects are controlled for. Individuals are\n\n\ncounted equally across countries.\n\n\nHonduras, and Peru have a high concentration of workers with “low quality” jobs,\n\n\nbut that do not have “good” jobs either.\n\n\nWe next look at the change in job quality over time and find the average level\n\n\nof job quality in Latin America rose over the sample period. The graph is shown\n\n\nin Figure 6. This graph was constructed by controlling for country specific fixed\n\n\neffects. After a period of stagnation observed during the decade of 1995-2004, job\n\n\nquality steadily rose increasing by roughly 25% between 2004 and 2012. The growth\n\n\nmight have a number of explanations. The first is that unemployment rates fell for\n\n\nmany countries in Latin America, resulting in more workers with a non-zero level of\n\n\njob quality. In addition, the number of successes in each of the job quality dimen\n\nsions also grew. The largest growth came from having both retirement and health\n\n\ninsurance benefits, and receiving an income above the poverty line. Though we do\n\n\n21\n\nMention: \"SEDLAC data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Job quality ratios over time are calculated from these data, coverage varying by year.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006958:26:1", "occurrence_id": "8861b272510abfd499f1f80c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: S. sales, exports and imports of goods and employment.\n\"Sales\" indicates sales of multinational corporations in gross output of U.S. private sector industries. Data\ncovers 2010-2013.\nC. The figure reflects the contribution of global, group-specific, and other factors to the variance of GDP\ngrowth. A dynamic factor model is estimated over the period 1985-2015, using a sample of 106 countries\ngrouped into three regions: advanced economies (AEs), emerging and frontier markets, and other\ndeveloping countries. Variance decompositions are computed for each country and, within each country, for\noutput. Each bar represents the variance share of U.S. and G6 (Canada, France, Germany, Italy, Japan,\nRussia, the United Kingdom and the United States) output growth attributable to the global factor, the\nAE-specific factor, the country-specific factor and the idiosyncratic term.\nD. The figure shows cumulative impulse responses after one year of GDP or industrial production (IP)\ngrowth in the United States following a 1 percentage point increase in GDP or industrial production growth\nin 22 other AEs and 19 EMDEs (13 EMDEs for industrial production). \"Global\" indicates the weighted\naverage impact of AEs and EMDEs. Vertical lines indicate 16th-84th percentile confidence bands. Vector\nautoregression models are estimated for 1998Q1-2016Q2 with four lags. The model includes, in this order,\nglobal GDP or industrial production growth excluding the United States and AE or EMDE, U.S. GDP or\nindustrial production growth, the U.S. 10-year sovereign bond yield plus JP Morgan’s EMBI index and AE\nor EMDE GDP or industrial production growth. The oil price is assumed to be exogenous.\n\nMention: \"JP Morgan’s EMBI index\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"An emerging-market spread used alongside domestic financial variables.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006961:11:1", "occurrence_id": "9a05acdfa4564b66c71a153f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  for years 1981\nand 1991 to extend the panel back in time in some cases.\n\nFigure 3 shows the spatial distribution of these PSUs all over the country. There is at least\none panel village in 70 of the 75 districts. In all of the estimations, I use the NLSS-III sampling\nweights of these PSUs to make my estimates nationally representative. The weighted migration\nrates in the panel are: 9.57 for year 2010, 8.06 for 2008, and 3.47 for year 2001, very close to the\nrates reported in Table 1. A fuller set of descriptive statistics is presented in Table 3. Figure\n4 shows the migration rates to India and non-India destinations for the 3 recent periods. As\ndiscussed above, migration to India tends to be more common from the mid and far-western\nregions of the country whereas migration elsewhere is increasing all over the country.\n\nIn the remainder of this section, I describe the context and relevance of the ‘push’ and ‘pull’\nshocks in this context along with the source of the data.\n\n\n**3.2** **Rainfall** **and** **income**\n\n\nNepal is primarily an agricultural country with farm income heavily depending upon rainfall.\nDuring the 2001/2010 period, the share of agriculture in total GDP hovered between 35-38\npercent (MoAD, 2013). In comparison, the total share of industry and manufacturing in GDP\nremained below 20 percent in the same period. In 2010, 80 percent of the households own\nagricultural land and the same proportion of the labor force is involved in agriculture <sup>13</sup> . Since\nthe fraction of land that is irrigated remains low, agriculture depends heavily on rainfall. Rice\nand maize are the two major crops grown in the country, both of which are grown in the wet\nmonsoon season. Rice is the key staple food and contributes to a fifth of the agricultural GDP\n\nMention: \"NLSS-III\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"Its sampling weights for the retained villages make estimates nationally representative.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006974:11:0", "occurrence_id": "9369e4f68cc3a2fe10e38bb0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: international production networks between “headquarter” and “factory” economies. Headquarter\n\n\neconomies (the United States, Germany and Japan) are the main providers of intermediates\n\n\nessential to exporting and arrange the production networks. Factory economies provide the\n\n\nlabor. Specifically, Baldwin and Lopez-Gonzalez (2015) find that the headquarter-versus\n\nfactory-economy distinction emerges clearly when they analyze the data on backward\n\n\nspecialization, which they refer to as “import to export” (or I2E).\n\n\nBased on this finding, for each country c, industry s and year t, the instrument is computed as the\n\n\naverage foreign value added from the United States, Japan and Germany embodied in exports of\n\n\nindustry s of three countries in the sample that are closest geographically or in income to country\n\nc. <sup>9</sup> The underlying assumption is that technological developments in these three countries (i.e.\n\n\nthe information and communication technology revolution) and declining trade costs globally\n\nwere the main drivers of GVC expansion. <sup>10</sup> The instrument is positively and significantly\n\n\ncorrelated with the backward vertical specialization measure and standard tests rule out its\n\nweakness. <sup>11</sup>\n\n\nThe results in columns 4 and 5 point to a significant effect of the foreign value added embodied\n\nin exports on labor productivity. <sup>12</sup> Focusing on the specification reported in column (5), which\n\n\nmay more effectively address any endogeneity problem than the one in column (4) (see footnote\n\n\n11), a 10 percent increase in the level of GVC-related trade is associated with an increase in\n\n\nlabor productivity of 1.7 percentage points. The impact of GVC-trade on labor productivity is\n\n\nsomewhat larger than that of non-GVC trade.\n\n\n9 Distance data are from CEPII, and per capita PPP income from the International Monetary Fund World Economic\nOutlook.\n10\n\nMention: \"CEPII\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Bilateral distance measures come from this centre.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006978:16:2", "occurrence_id": "c64ccbc7e86c9e78f6355b5b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 937|\n|**World**<br>**Bank**|38.7|8.5|47.2|2.4|49.7|50.3|7,436|\n\n\nSource: IEG data set and staff calculation\n\n\nIn the SAR and EAP regions, a little over half of all projects have reported ERRs in the PAD or ICR (56.1 and\n57.6 percent, respectively). This is higher than for other regions where this ranges from 45.2 to 52\npercent.  In contrast, in SAR and EAP, about 43.9 and 42.4 percent, respectively, of the projects do not\nreport an ERR either at the appraisal or at the completion stage. This number ranges from 48 to 55 percent\nfor other regions (Table 2).\n\n\nModel 1, described above, examines whether the frequency of ERRs has been declining in recent years.\nFigure 1 shows a long‐term declining trend, but also suggests there has been a slight uptick over the past\ndecade. Table 3 shows the declining trends by “high‐CBA” and “low‐CBA” sectors. The IEG study (World\nBank 2010) found that the changes were driven mainly by changes within the “high‐CBA” sectors (i.e.\ntransport, energy, water, agriculture, and urban development) and also a shift toward “low‐CBA” sectors.\nThis paper updates that analysis and finds that this result still holds (Table 4).\n\n\n15\n\nMention: \"IEG data set\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Project evaluation ratings underlying the regional shares.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006980:26:0", "occurrence_id": "9488fcc2016a017e82d8358c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: refugees, mainly from Somalia and South Sudan); and Zaatari in Jordan (with about 80,000 refugees,\nmainly from Syria). There are also a relatively large number of smaller (managed) refugee camps in many\ncountries, which account for about 2.7 million people.\n\n**Figure 16: UNHCR’s Population of Concern by Accommodation end-2015**\n\n\n_Source_ : UNHCR Global Trends 2015\n_Note_ : UNHCR’s Population of Concern includes refugees and people in refugee-like situations, asylum-seekers, IDPs\nand people in IDP like situations protected or assisted by UNHCR, stateless people and other people of concern to\nUNHCR. Excludes Palestinian refugees under UNRWA’s mandate.\n\n\nOver the last decade, women, youth and children have tended to account for at least two-thirds of displaced\npeople—but there is substantial variation across host countries. <sup>50</sup> Demographics of displaced populations\nhave consequences for development in both countries of origin and asylum, affecting the availability of\nskilled and unskilled workers to the labor market, the age structure of the population, and demand for\nservices (e.g. children requiring education services etc.). At the end of 2015, approximately 75 percent of\ndisplaced people were women, youth and children, i.e. 24 percent adult women and 51 percent youth and\nchildren below the age of 18 (UNHCR 2016). <sup>51</sup> The gender profile of displaced populations is consistent\nwith that for the global population, but the available data suggests that displaced populations may be\nyounger on average than the global population. <sup>52</sup> This apparent inconsistency is likely to reflect the\nsignificant variation in the demographics of forced displacement across host countries. For example, in\nGeorgia just over a quarter of IDPs are under the age of 18 and in Germany just over a quarter of refugees\nand asylum-\n\nMention: \"UNHCR Global Trends 2015\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Population of concern by accommodation at the end of that year.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006981:8:0", "occurrence_id": "81504e4dfe87e91142eff56f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: the income, consumption and asset management behavior (Khandker 2012; Khandker and Mahmud\n\n\n2012; Pitt and Khandker 2002).\n\n\n**3.  Seasonality in household borrowing and income: Descriptive evidence**\n\n\nUsing Household Income and Expenditure Surveys (HIES) of 2000 and 2005, Khandker (2012) demon\n\nstrates that seasonality is pronounced in household consumption and income, and that income seasonality\n\n\naffects significantly the seasonality of consumption and poverty beyond and above seasonal variations in\n\n\nconsumption and poverty. This phenomenon is most acute in the Northwest region of Bangladesh\n\n\n(Khandker and Mahmud 2012). Findings also indicate that seasonal variations severely affect poor people\n\n\nwho mainly live on agricultural income, and microcredit has so far played a good role in mitigating sea\n\nsonal hardship. Seasonality still matters, however. Using the 2010 HIES data, Khandker and Samad\n\n\n(2016) show that although the severity of seasonal hunger has declined, consumption and poverty are still\n\n\ninfluenced by the seasonality of income caused by the agriculture cycle.\n\n\nIn order to explore the seasonality of household borrowing and income, we use panel data\n\n\ncollected by the World Bank with help of Bangladesh Institute of Development Studies (BIDS) in\n\n\n1998/99 and Institute of Microfinance (InM) in 2010/11 (see Khandker and Samad 2014 for a detailed\n\n\ndescription of the data set). <sup>2</sup> [^2: While the panel surveys also include data from 1991/92, we could not use them because they are not suitable for calculating\nseason-specific income that would be consistent with incomes calculated from 1998/99 and 2010/11 data. So, the analysis of\nthis paper is restricted to 1998/99 and 2010/11 survey data only. Nevertheless, using the 1991/92 along with the 1998/99\nand 2010/11 panel surveys, our results on the seasonality of income and borrowing still hold.] Using these data we can divide the borrowing sources into three categories\n\n\n(microcredit, formal or commercial, and informal) and the whole year\n\nMention: \"2010 HIES data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Seasonal hunger severity is shown using this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006983:14:1", "occurrence_id": "9d6c9ed6f6508630e49fe35e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .04|0.02|\n|0.47|0.30|0.24|0.17|<br>0.14|0.09|0.06|0.04|0.02|\n|0.55|0.39|0.34|0.27|<br>0.26|0.19|0.17|0.13|0.08|\n\n\n\n_Source:_ Based on data of the 2011 Household Consumption Survey.\n_Note:_ The table shows the share of total consumption of each decile.\n\nTable 9 reports the income gains derived from the reduction in medical expenses for each income decile. The\nreduction in tobacco consumption would have a positive effect on income through the reduction in medical\ntreatments. The income gains would vary between 0.15 and 0.01 percentage points in the case of the lowerbound elasticity assumption, between 0.47 and 0.01 percentage points in the case of the medium-bound\nelasticity, and between 0.55 and 0.03 percentage points in the case of the upper-bound elasticity (figure 2).\nThese results show the importance of the elasticity assumptions; they also stress the relevance to the possible\nelasticity variations across income groups.\n\n\n13\n\nMention: \"the 2011 Household Consumption Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Each decile's share of total consumption is taken from this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006987:30:1", "occurrence_id": "cece833e5fcd3d29b7305922", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <br>property, public goods, social protec-<br>tion, and hate crimes or hate speech.<br>Does not allow for cross-country<br>comparison.|\n|NORMLEX <br>|ILO|NORMLEX is a database of national labor<br>and social security laws, ratification infor-<br>mation, reporting requirements, and com-<br>ments of the ILO’s supervisory bodies.<br>Includes legislation on nondiscrimination in<br>the labor market and ratification of em-<br>ployment discrimination conventions.|Covers only labor legislation and in-<br>ternational instruments and does not<br>cover issues related to institutions,<br>education, property, public goods, so-<br>cial protection, and hate crimes or<br>hate speech. Does not allow for cross-<br>country comparison.|\n|General Surveys <br>(2012, 1996, <br>1986) of the<br>Committee of<br>Experts <br>|ILO|The General Surveys are reports examin-<br>ing law and practice in a particular subject<br>area related to labor in member states that<br>have or have not ratified relevant ILO con-<br>ventions.<br>Includes legislation on nondiscrimination in<br>the labor market.|<br>Covers only labor legislation and does<br>not cover issues related to institu-<br>tions, education, property, public<br>goods, social protection, and hate<br>crimes or hate speech. Does not allow<br>for cross-country comparison.|\n|The OHCHR <br>antidiscrimina-<br>tion database  <br>|Office of the<br>High Com-<\n\nMention: \"NORMLEX\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"A compilation of national labour and social security laws and ratifications.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006987:30:1", "occurrence_id": "a7bc28012304d22e2c786d4a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <br>property, public goods, social protec-<br>tion, and hate crimes or hate speech.<br>Does not allow for cross-country<br>comparison.|\n|NORMLEX <br>|ILO|NORMLEX is a database of national labor<br>and social security laws, ratification infor-<br>mation, reporting requirements, and com-<br>ments of the ILO’s supervisory bodies.<br>Includes legislation on nondiscrimination in<br>the labor market and ratification of em-<br>ployment discrimination conventions.|Covers only labor legislation and in-<br>ternational instruments and does not<br>cover issues related to institutions,<br>education, property, public goods, so-<br>cial protection, and hate crimes or<br>hate speech. Does not allow for cross-<br>country comparison.|\n|General Surveys <br>(2012, 1996, <br>1986) of the<br>Committee of<br>Experts <br>|ILO|The General Surveys are reports examin-<br>ing law and practice in a particular subject<br>area related to labor in member states that<br>have or have not ratified relevant ILO con-<br>ventions.<br>Includes legislation on nondiscrimination in<br>the labor market.|<br>Covers only labor legislation and does<br>not cover issues related to institu-<br>tions, education, property, public<br>goods, social protection, and hate<br>crimes or hate speech. Does not allow<br>for cross-country comparison.|\n|The OHCHR <br>antidiscrimina-<br>tion database  <br>|Office of the<br>High Com-<\n\nMention: \"General Surveys\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Committee of Experts surveys across these years interpret the standards.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006972:23:0", "occurrence_id": "7cb08fc8be51ba7bfc4e26b8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 1. Part‐time employment trends in Europe.**\n\n\nNote: based on estimates from Eurostat, <u>http://ec.europa.eu/eurostat/web/lfs/data/database.</u>\n\n\n22\n\nMention: \"Eurostat\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Part-time employment trends across Europe come from this office's database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007000:22:2", "occurrence_id": "1033adf736bccccf46a982ea", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 0.0259) (0.0227) (0.0227) (0.0227) (0.0258) (0.0260)\n_DVAR*Internet_ 0.0010*\n(0.0006)\n_DVAR*LPI logistics_ 0.0513**\n(0.0211)\n_DVAR*LPI customs_ 0.0550***\n(0.0204)\n_DVAR*LPI overall_ 0.0621***\n(0.0232)\n_DVAR*Time to export_ -0.0053***\n(0.0019)\n_DVAR*Time to import_ -0.0033**\n(0.0014)\n_Constant_ 4.844*** 4.671*** 4.742*** 4.965*** 5.059*** 5.029***\n(0.165) (0.166) (0.157) (0.159) (0.174) (0.181)\nObservations 7,164 7,060 7,060 7,060 7,164 7,164\n<u>R-squared</u> <u>0.872</u> <u>0.879</u> <u>0.879</u> <u>0.879</u> <u>0.872</u> <u>0.872</u>\n\n\nSource: OECD ICIO database. *** p<0.01, ** p<0.05, * p<0.1.\nNotes: Robust standard errors in parentheses. _FVAX_ and _DVAR_ lagged. All level variables in natural logarithms.\n\n\nLooking at the income group specific results, we observe again\n\nMention: \"OECD ICIO database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its estimates carry the reported trade-in-value-added regressions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007007:53:1", "occurrence_id": "aa382c93bd09df2a572a0eb3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  NREGS. However,\n\n\nthe process of district selection was influenced by political considerations due to the huge\n\n\nsize and financial relevance of this program and the rules elaborated by the Planning\n\n\nCommission were not strictly followed.\n\n\n_•_ NREGS introduced in District: This variable is an indicator which equals 1 if a\n\n\nhousehold has access to the NREGS at the district level at the beginning of the\n\n\nagricultural cycle (i.e. when most input allocation decisions have to be taken). The\n\n\nYLS was collected in six districts, four of which are treatment districts: Anan\n\ntapur, Cuddapah, Karimnagar and Mahaboobnagar, and two are control districts\n\n\nSrikakulam and West Godavari.\n\n\nThe period of reference for the 2007 round of interviews is the agricultural year\n\n\n2005/06 (June 2005 to May 2006). According to the Employment Generation Re\n\n10\n\nMention: \"YLS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Collected across six districts, four of them treatment areas.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007008:21:0", "occurrence_id": "9bb70ffa9d7fa929bf5aef3f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: younger and female workers are more likely to lose their job. However, exposure to unemployment is not\n\n\nconcentrated in particular income groups, as these individuals are found across the whole welfare\n\n\ndistribution (see Figure 5).\n\n\n\n\n\nThis result suggests that the ultimate distributive impact of the unemployment increase would depend\n\n\nheavily on the social protection programs available to those who lost their job and, of course, to the whole\n\n\npopulation. However, it is unlikely that these programs and, more generally, public transfers and\n\n\nexpenditure will not suffer from the shrinking oil royalties following the drop in oil price. The oil price\n\n\nshock, indeed, puts government expenditure under pressure and its reduction is likely to have affected\n\n\ndifferently individuals located in different parts of the welfare distribution.\n\n\nTaking advantage of the availability of data on public transfers in a different survey, the RLMS (Russian\n\n\nLongitudinal Monitoring Survey), our second simulation estimates the effect of a 10 percent cut in public\n\n\ntransfers. The impact on poverty and shared prosperity depends on the incidence of the transfers along\n\n\nthe welfare distribution: if poor households depend relatively more than the richer ones on public\n\n\ntransfers, the final effect of the oil shock could be reversed once the cut of transfers is taken into account.\n\n\nMost of the beneficiaries of public transfers <sup>21</sup> and pensions are indeed located in the first decile of the\n\n\nconsumption distribution (see Figure 6) while the share is much smaller for households located in the top\n\n\n30 percent of the distribution.\n\n\n21 Transfers are defined as the sum of children, rent, utilities, fuel and unemployment benefits.\n\n\n20\n\nMention: \"RLMS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Public transfer data in this second survey drive an alternative simulation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007008:21:0", "occurrence_id": "0a275a029e7fec0b3a3f495c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: younger and female workers are more likely to lose their job. However, exposure to unemployment is not\n\n\nconcentrated in particular income groups, as these individuals are found across the whole welfare\n\n\ndistribution (see Figure 5).\n\n\n\n\n\nThis result suggests that the ultimate distributive impact of the unemployment increase would depend\n\n\nheavily on the social protection programs available to those who lost their job and, of course, to the whole\n\n\npopulation. However, it is unlikely that these programs and, more generally, public transfers and\n\n\nexpenditure will not suffer from the shrinking oil royalties following the drop in oil price. The oil price\n\n\nshock, indeed, puts government expenditure under pressure and its reduction is likely to have affected\n\n\ndifferently individuals located in different parts of the welfare distribution.\n\n\nTaking advantage of the availability of data on public transfers in a different survey, the RLMS (Russian\n\n\nLongitudinal Monitoring Survey), our second simulation estimates the effect of a 10 percent cut in public\n\n\ntransfers. The impact on poverty and shared prosperity depends on the incidence of the transfers along\n\n\nthe welfare distribution: if poor households depend relatively more than the richer ones on public\n\n\ntransfers, the final effect of the oil shock could be reversed once the cut of transfers is taken into account.\n\n\nMost of the beneficiaries of public transfers <sup>21</sup> and pensions are indeed located in the first decile of the\n\n\nconsumption distribution (see Figure 6) while the share is much smaller for households located in the top\n\n\n30 percent of the distribution.\n\n\n21 Transfers are defined as the sum of children, rent, utilities, fuel and unemployment benefits.\n\n\n20\n\nMention: \"Russian\n\n\nLongitudinal Monitoring Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The full name of the same panel used for the transfer simulation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007010:30:0", "occurrence_id": "041d43c6d155a9550201f860", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 4: Education and experience premium by year.**\n\n\n\n\n\n\n\n\n\n\n\n\n\nNote. Each line represents the predictions of a yearly regression that includes a quartic polynomial in education, a quartic polynomial\nin potential experience, a dummy variable for workers below the minimum wage, two dummies for formality status, two race\ndummies, a gender dummy, four region dummies and an indicator of work in rural areas (see Table 2). The year 1995 includes 1995\nand 1996 PNAD samples, the year 2003 includes 2002 and 2003 PNAD sample, and the year 2012 included 2011 and 2012 PNAD\nsample. Source: _Pesquisa Nacional por Amostra de Domicílios_ PNAD.\n\n\n29\n\nMention: \"PNAD samples\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"Adjacent rounds are pooled into single analysis years.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007016:6:0", "occurrence_id": "c23b8312cae41e017d9b2483", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: efforts in the realm of policy and programming, at the local and national levels; failure to do so\n\n\nwill merely prolong the precarious role of women in the economic life of the nation.\n\n### 2. Female Labor Force Trends in Post-Liberalization India\n\n\n2.1 Database and Method\n\n\nConfining our analysis to the post-liberalization era in India, we consider the NSSO data sets for\n\n\nall rounds between Round-50 (1993-94) and the latest available, Round-68, that was collected in\n\n\n2011-12. In order to calculate labor force participation trends, we utilize information on Usual\n\n\nPrincipal and Subsidiary Status (UPSS) of the activity classification provided in these data sets. <sup>2</sup>\n\n\nTotal FLFP rates, as well as those for rural and urban areas are provided below in Table 2. For the\n\n\npurposes of this study, we consider the working age population as inclusive of those in the age\n\n\ngroup of 15 years and above.\n\n\n<u>Table 2: Female Labor Force Participation Rates, Using UPSS for Ages 15 and Above (%)</u>\nRural Urban Total\n1993-94 49.0 23.7 42.6\n1999-00 46.0 21.0 39.5\n2004-05 49.4 24.4 42.7\n2009-10 37.8 19.4 32.6\n2011-12 35.8 20.5 31.2\n\n\n2 Activity status’ is the activity which a person reports for a given reference period, with regard to their participation in economic\nand non-economic activities. According to this, a person could be in one of the following three broad activity statuses during a\nreference period: (i) working or being engaged in economic activity (work) as defined above; (ii) being not engaged in economic\nactivity (work) but either making tangible efforts to seek 'work' or being available for 'work\n\nMention: \"NSSO data sets\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Every round across this span is considered for the analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007025:10:1", "occurrence_id": "d8589eed9153ab1e52d712e0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  1990s is due to new countries that were\n\n\nformed in the 1990s, and part is due to a rise in the volume of international loans. In\n\n\nthis respect, the migration stock in 1990 can be regarded as a predetermined variable\n\n\nfor country pairs that had no lending activity in the 1980s and started to lend in the\n\n\n1990s. <sup>8</sup>\n\n\nWe complete our dataset with gravity variables taken from Aviat and Coeurdacier\n\n\n(2007) and CEPII (CHELEM dataset). Trade data are taken from Feenstra et al.\n\n\n(2005).\n\n\n7. In 2010, the number of country pairs reached 1,908.\n8. However, our results are fully robust to the use of data for the 2000-2010 period. Note that\nmeasurement errors might be larger for the earlier period due to the appearance of new countries in the\n1990s.\n\n\n9\n\nMention: \"CHELEM dataset\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its gravity variables are used with the trade data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007029:4:0", "occurrence_id": "dbab7d8a535316211d4f2aaf", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: economics and how recent developments in economics can actually offer new instruments to model\n\n\nforced displacement. This is done from the perspective of utility theory and modeling choice. Next, we\n\n\ndiscuss data issues that are currently preventing economists from working on forced displacement.\n\n\nSection 5 will provide a concrete example of how economics research can serve policy makers and\n\n\nimprove the lives of refugees, IDPs and their hosts. Section 6 concludes.\n\n#### 2. Some facts\n\n\nThe United Nations High Commissioner for Refugees (UNHCR) reports that at the end of 2015 there were\n\n\n65.3 million forcibly displaced people worldwide. <sup>5</sup> This number includes 21.3 m. refugees and about 10\n\n\nm. stateless people with the rest mostly being accounted for by Internally Displaced Persons (IDPs). The\n\n\nforcibly displaced population accounts therefore for about 0.9 percent of the world population estimated\n\n\nat 7.3 bn people in 2015. The latest estimate of forcibly displaced people is the largest on record, since\n\n\nthe creation of the UNHCR in 1950. The growth in these numbers is evidently associated with episodes of\n\n\nconflict and violence. The Somalia, South‐Sudan and Syrian conflicts have been responsible for the major\n\n\noutflows of refugees and movements of IDPs over the past decade, with the Syrian conflict being\n\n\nresponsible for the latest spike (Figure 1).\n\n\n_Figure 1 ‐ Number of Forcibly Displaced Persons (1951‐2015)_\n\n\nSource: Constructed from UNHCR population data <u>(http://popstats.unhcr.org/en/time_series). Note:</u>\n\n\n2015 data are mid‐year and lower than end of year data.\n\n\n5 http://www.unhcr.org/en‐us/figures‐at‐a‐glance.html.\n\n\n3\n\nMention: \"UNHCR population data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Displaced totals since the mid-twentieth century, with the latest year mid-period.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006994:45:0", "occurrence_id": "6cdd3f39af2a4408f3918243", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Table 7: Regression coefficients of individual nutritional outcomes on the DHS household wealth**\n\n**index and household consumption per person**\n\n\nDHS LSMS\n\n\n\nWeight-for\n<u>height</u> <u>BMI</u>\n\n\n\nHeight-for\n<u>age</u>\n\n\n\nWeight-for\n<u>height</u>\n\n\n\n<u>BMI</u>\n\n\n\nHeight-for\n<u>age</u>\n\n\n\nBenin 0.194*** 0.189*** 0.080***\nBurkina Faso 0.279*** 0.281*** 0.097*** n/a 0.377*** -0.016\n\nBurundi 0.242*** 0.505*** 0.115***\nCameroon 0.285*** 0.451*** 0.257***\nCongo, Rep. 0.265*** 0.292*** 0.051**\nCôte d'Ivoire 0.203*** 0.279*** 0.043\nCongo, Dem. Rep. 0.276*** 0.378*** 0.073***\nEthiopia 0.352*** 0.374*** 0.229*** n/a 0.037 0.025\n\nGabon 0.182*** 0.397*** 0.056**\nGambia 0.208*** 0.297*** 0.061**\nGhana 0.385*** 0.299*** 0.070*** 0.120*** 0.178** 0.172\n\nGuinea 0.284*** 0.285*** 0.023\nKenya 0.331*** 0.257*** 0.210***\nLesotho 0.263***\n\nMention: \"DHS LSMS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Nutritional outcomes are reported side by side for the two survey programmes.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007005:27:0", "occurrence_id": "c35b4b7b6b3028b4bfbfaea1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: the GIS road network and the 93 nodes, we used the shortest path routing algorithm to compute number\nof kilometers, travel time, and total road user cost for each origin-destination pair (or route).\n\n\n_<u>Table 5. Terrain Classes and Percentage of Land Area Peru</u>_\n\n**<u>Class Description</u>** **<u>Peru</u>**\n**1** Plains 0.11\n**2** Mid-altitude plains 0.00\n**3** High-altitude plains 0.00\n**4** Lowlands 10.92\n**5** Rugged lowlands 17.41\n**6** Platforms (very low plateaus) 1.48\n**7** Low plateaus 0.05\n**8** Mid-altitude plateaus 0.00\n**9** High plateaus 0.72\n**10** Very high plateaus 0.14\n**11** Hills 20.83\n**12** Low mountains 7.44\n**13** Mid-altitude mountains 10.37\n**14** High mountains 16.81\n**<u>15</u>** <u>Very high mountains</u> <u>13.70</u>\n_Source:_ Authors’ calculations.\n\n###### 6.2. GIS Data Limitations\n\nA number of limitations associated with the input data sets and the methods used in this analysis should\nbe taken into account in the interpretation of results.\n\n\n_Scale of roads_ . The scale of the source data (that is, the level of detail) for roads is unknown, but there are\nobvious differences between the countries and in comparison with large-scale data sets such as Open\nStreet Map. The smaller the scale, the less likely the features in the\n\nMention: \"Open\nStreet Map\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"A large-scale reference layer for judging feature coverage at smaller scales.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007036:8:1", "occurrence_id": "0209cfb412531eab6312fd63", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  _Indicators_ database. We also create a dummy variable capturing whether the country\n\n\nis under an IMF assistance program for the year that the ease of doing business is assessed. This\n\n\ninformation is taken from the IMF website.\n\n\nIn Table 1 we report some descriptive statistics of the main variables contained in our data\n\n\nset for the top 20 DPL recipients (countries that received 10 or more loans). All measures of\n\n\nperformance, including the average _Doing_ _Business_ score, its change, and the growth rate of per\n\n\ncapita GDP are computed over the period starting in the year the country received its first DPL\n\n\n5 Linkages with the _Doing_ _Business_ score could be thought of as even tighter for those DPLs explicitly targeting\nprivate sector development. However, while about 420 DPLs, of the 690 operations in our data set, have their\nmajor sector classified as ”General Public Administration”, ”Central Government Administration”, and ”Sub-national\nGovernment Administration”, the theme ”Infrastructure services for private sector development” appears for only 68\noperations. Of these, 30 are in Africa, 15 in East Asia and Pacific, 8 in Europe and Central Asia, 6 in South Asia, 5\nin Latin America and Carribean and 4 in Middle-East and North Africa. In all regions the size of these DPLs varies\nsignificantly, from a few million to hundreds of millions.\n\n\n6\n\nMention: \"_Indicators_ database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Used alongside a constructed flag for programme participation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007042:4:0", "occurrence_id": "921bbf00d6ea4f237b4c7d5d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ordinary least squares (OLS) and fixed-effect models using employer-employee matched data\n\n\nshows correlations between personality traits and wages both within and across formal sector\n\n\nfirms.\n\n\nThis analysis aims to make three main contributions to the literature on non-cognitive skills and\n\n\nlabor market outcomes. First, we present findings regarding the association between personality\n\n\ntraits and labor market outcomes among labor force participants in Bangladesh. To the best of our\n\n\nknowledge, this is the first comprehensive assessment of personality traits and their relationship\n\nto wages in the South Asian context. <sup>2</sup> [^2: Similar questions were administered in Sri Lanka via the STEP survey (Dundar _et al_ . 2014) but we could not find\nany analysis of the relationship between non-cognitive skills and employment outcomes.] With an estimated population of 1.7 billion in 2014, <sup>3</sup> [^3: http://data.worldbank.org/region/SAS.] and\n\n\nwith the labor force set to grow by 1 million to 1.2 million new workers every month over the next\n\ntwo decades, <sup>4</sup> [^4: http://blogs.worldbank.org/category/tags/south-asia-workforce.] South Asia presents unique challenges to researchers and policy makers striving to\n\n\nunderstand the characteristics and skill sets of the region’s rapidly growing workforce. Bangladesh\n\n\npresents an especially salient example of the changing employment dynamics in South Asia, as\n\n\nthe share of the labor force employed in agriculture shrank from 51 percent in 2000 to 45 percent\n\nin 2013. <sup>5</sup> [^5: Bangladesh Labor force surveys.] Moreover, the Government of Bangladesh has made skills development a core priority,\n\n\nas described in its 2011 National Skills Development Policy.\n\n\nSecond, we use employer-employee matched data taken from the 2012 Enterprise Skills Survey\n\n\n(ESS), which sampled 500 formal sector employers and employees. Unlike household-based labor\n\n\nforce surveys, employer-employee matched data allow an estimation of\n\nMention: \"STEP survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Similar questions were administered there, though unanalysed for this relationship.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007042:4:0", "occurrence_id": "f3a53084d3ba5d0598a96f9a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ordinary least squares (OLS) and fixed-effect models using employer-employee matched data\n\n\nshows correlations between personality traits and wages both within and across formal sector\n\n\nfirms.\n\n\nThis analysis aims to make three main contributions to the literature on non-cognitive skills and\n\n\nlabor market outcomes. First, we present findings regarding the association between personality\n\n\ntraits and labor market outcomes among labor force participants in Bangladesh. To the best of our\n\n\nknowledge, this is the first comprehensive assessment of personality traits and their relationship\n\nto wages in the South Asian context. <sup>2</sup> [^2: Similar questions were administered in Sri Lanka via the STEP survey (Dundar _et al_ . 2014) but we could not find\nany analysis of the relationship between non-cognitive skills and employment outcomes.] With an estimated population of 1.7 billion in 2014, <sup>3</sup> [^3: http://data.worldbank.org/region/SAS.] and\n\n\nwith the labor force set to grow by 1 million to 1.2 million new workers every month over the next\n\ntwo decades, <sup>4</sup> [^4: http://blogs.worldbank.org/category/tags/south-asia-workforce.] South Asia presents unique challenges to researchers and policy makers striving to\n\n\nunderstand the characteristics and skill sets of the region’s rapidly growing workforce. Bangladesh\n\n\npresents an especially salient example of the changing employment dynamics in South Asia, as\n\n\nthe share of the labor force employed in agriculture shrank from 51 percent in 2000 to 45 percent\n\nin 2013. <sup>5</sup> [^5: Bangladesh Labor force surveys.] Moreover, the Government of Bangladesh has made skills development a core priority,\n\n\nas described in its 2011 National Skills Development Policy.\n\n\nSecond, we use employer-employee matched data taken from the 2012 Enterprise Skills Survey\n\n\n(ESS), which sampled 500 formal sector employers and employees. Unlike household-based labor\n\n\nforce surveys, employer-employee matched data allow an estimation of\n\nMention: \"2012 Enterprise Skills Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Employer-employee matched data on five hundred formal sector employers.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007046:14:0", "occurrence_id": "ce89d8c1038f15cdc226f195", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: institution for 15 years or more—and are clearly a significant limitation when we apply the tools of cohort\n\n\nanalysis to a single firm. <sup>22</sup>\n\n\n**Figure 3: Exit Rates from the WBG between 1987 and 2015**\n\n\n13\n\n\n12\n\n\n11\n\n\n10\n\n\n9\n\n\n8\n\n\n7\n\n\n6\n\n\n5\n\n\nA final characterization of the data is in terms of salaries. Table 1 shows the mean _real_ salaries of\n\n\nemployees at each grade over time and to focus on changes, we compare all salaries to a base of 100 for\n\n\ngrade GA in 1987. <sup>23</sup> To preserve anonymity, we leave blank the cells where there are too few employees\n\n\nand do not present results for grade GK. The table first demonstrates considerable variation in salaries\n\n\nwithin each grade. Typically, there is a difference of 20-40% between the 10 <sup>th</sup> percentile and the 90 <sup>th</sup>\n\n\npercentile of the within-grade salary distribution. Second, salaries in higher grades exhibit large\n\n\n22 If b is the exponential decay rate, x the initial stock of employees and t the number of years, then the existing employee stock\nwith halve when xb <sup>t</sup> =x/2 or, tln(b)=ln(0.5) or t = ln(0.5)/ln(b). With an attrition rate of 9%, b=0.91 yielding 8 years for halving the\nemployee stock.\n23 We use the consumer price index for the U.S. to deflate nominal salaries—note that price increases in the BaltimoreWashington area have been higher than for the U.S. as a whole, so using the BW CPI will lower real wage increases further over\nthe period of our data.\n\n13\n\nMention: \"BW CPI\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"A local price index that dampens real wage growth relative to the national one.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007051:10:0", "occurrence_id": "f5a956543b7387a6b95c691c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: to serve in Korea. de Walque (2007) also imposes this sample restriction. <sup>8</sup> [^8: We also perform analyses using 1930-1956 birth cohorts in sensitivity analyses reported in section 5.4. This sample would have\nbeen completely ineligible to serve in WWII, which is an important restriction as we only want to consider the post WWII\ninduction system, created in 1948.] Because we are\n\n\ninterested in educational attainment, we limit our sample to those over age 25 at interview as\n\n\nmost individuals have completed their schooling by this age. We further restrict our sample to\n\n\nthose with valid measurements of height and weight because we use these to create BMI. <sup>9</sup> [^9: To be consistent across the years of NHIS, we drop individuals below 59 inches and above 76 inches, and those weighing less\nthan 100 pounds and more than 285 pounds, the most restrictive data publishing policies used by NHIS over the years of our\nsample.]\n\n\nIn certain analyses, we use a restricted-use version of the NHIS which provides us with\n\n\ndata on state of birth (blinded), <sup>10</sup> [^10: This measure allows us to determine individuals who were born in the same state, permitting us to estimate models including\nstate fixed effects. However, this measure is only available for a subsample of our data.] whether and in which specific year an individual died if he or\n\n\nshe died by 2011 (NHIS-Mortality Linked File), and his or her age of death. These mortality data\n\n\nwere collected by matching NHIS participants annually to the National Death Index system\n\n\ndeath certificates by a combination of social security number, gender, first and last name, date of\n\n\nbirth, or birth month and year.\n\n\n**3.2.Selective Service Reports**\n\n\nWe use annual (1955-1966) and semi-annual (1967-1975) reports of the Director of the\n\n\nSelective Service for induction eligibility, the number of monthly inductions,\n\nMention: \"NHIS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"A restricted-use version adds the blinded state of birth.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007051:10:0", "occurrence_id": "cfb545ad6deb690c4de459bd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: to serve in Korea. de Walque (2007) also imposes this sample restriction. <sup>8</sup> [^8: We also perform analyses using 1930-1956 birth cohorts in sensitivity analyses reported in section 5.4. This sample would have\nbeen completely ineligible to serve in WWII, which is an important restriction as we only want to consider the post WWII\ninduction system, created in 1948.] Because we are\n\n\ninterested in educational attainment, we limit our sample to those over age 25 at interview as\n\n\nmost individuals have completed their schooling by this age. We further restrict our sample to\n\n\nthose with valid measurements of height and weight because we use these to create BMI. <sup>9</sup> [^9: To be consistent across the years of NHIS, we drop individuals below 59 inches and above 76 inches, and those weighing less\nthan 100 pounds and more than 285 pounds, the most restrictive data publishing policies used by NHIS over the years of our\nsample.]\n\n\nIn certain analyses, we use a restricted-use version of the NHIS which provides us with\n\n\ndata on state of birth (blinded), <sup>10</sup> [^10: This measure allows us to determine individuals who were born in the same state, permitting us to estimate models including\nstate fixed effects. However, this measure is only available for a subsample of our data.] whether and in which specific year an individual died if he or\n\n\nshe died by 2011 (NHIS-Mortality Linked File), and his or her age of death. These mortality data\n\n\nwere collected by matching NHIS participants annually to the National Death Index system\n\n\ndeath certificates by a combination of social security number, gender, first and last name, date of\n\n\nbirth, or birth month and year.\n\n\n**3.2.Selective Service Reports**\n\n\nWe use annual (1955-1966) and semi-annual (1967-1975) reports of the Director of the\n\n\nSelective Service for induction eligibility, the number of monthly inductions,\n\nMention: \"NHIS-Mortality Linked File\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"It records whether and in which year a participant had died.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007051:10:0", "occurrence_id": "77b3cdac4fc1a29db9d6a963", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: to serve in Korea. de Walque (2007) also imposes this sample restriction. <sup>8</sup> [^8: We also perform analyses using 1930-1956 birth cohorts in sensitivity analyses reported in section 5.4. This sample would have\nbeen completely ineligible to serve in WWII, which is an important restriction as we only want to consider the post WWII\ninduction system, created in 1948.] Because we are\n\n\ninterested in educational attainment, we limit our sample to those over age 25 at interview as\n\n\nmost individuals have completed their schooling by this age. We further restrict our sample to\n\n\nthose with valid measurements of height and weight because we use these to create BMI. <sup>9</sup> [^9: To be consistent across the years of NHIS, we drop individuals below 59 inches and above 76 inches, and those weighing less\nthan 100 pounds and more than 285 pounds, the most restrictive data publishing policies used by NHIS over the years of our\nsample.]\n\n\nIn certain analyses, we use a restricted-use version of the NHIS which provides us with\n\n\ndata on state of birth (blinded), <sup>10</sup> [^10: This measure allows us to determine individuals who were born in the same state, permitting us to estimate models including\nstate fixed effects. However, this measure is only available for a subsample of our data.] whether and in which specific year an individual died if he or\n\n\nshe died by 2011 (NHIS-Mortality Linked File), and his or her age of death. These mortality data\n\n\nwere collected by matching NHIS participants annually to the National Death Index system\n\n\ndeath certificates by a combination of social security number, gender, first and last name, date of\n\n\nbirth, or birth month and year.\n\n\n**3.2.Selective Service Reports**\n\n\nWe use annual (1955-1966) and semi-annual (1967-1975) reports of the Director of the\n\n\nSelective Service for induction eligibility, the number of monthly inductions,\n\nMention: \"National Death Index system\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Death certificates matched annually using identifiers.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007053:6:0", "occurrence_id": "a48c8fd69294a64114021c19", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _CROSSOVERS_ _5_\n\n\nmotivation from the Uganda study, but we drew on a question administered in\nthe WEDP baseline survey that asked the entrepreneur, “Are most enterprises\nin your business sector owned by men or women?” If more than 75 percent of\nresponses from the entire baseline sample (2,369 female entrepreneurs) were that\nmen owned most enterprises in their business sector, we defined that sector as\nmale dominated. We used the entire core baseline sample of 2,369 entrepreneurs\nto determine this definition of a male-dominated sector and defined a woman\nwho had a business in any of these sectors as a “crossover.” Using this definition\nof a male-dominated sector, we established 164 crossover businesses, with the\nremaining 626 businesses in the sample considered noncrossover.\n\n\nFigure 1. Sectors classified as male dominated\n\n\nAs a robustness check, we examined the sector label names that have been\nclassified as male dominated based on this definition. We believe that the sectors\n\n\nownership was more than 75 percent, because the overrepresentation could have been a reflection of the\nspecific sample, which is often not representative of the entire population of firms in the country.\n\nMention: \"the WEDP baseline survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its question on sector ownership was reused in the Uganda design.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007056:16:1", "occurrence_id": "25741dd1880871799bda07f6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  another\nrural area. <sup>6</sup> Of the 2,111 individuals who did not move, 331 live in an area that got reclassified\nfrom rural to urban and we drop them from our analysis to avoid confusion. Virtually everyone\nin the rural areas is in the informal sector (only 3.5 percent rely on a formal wage job), so this\ngroup corresponds closely to the rural population we have in mind in the model. In the towns,\n18.9 percent live in a household that relies on formal wage employment. This rises to 22.2\npercent in the cities. <sup>7</sup>\n\n\nThe 2010 income distribution of these individuals could essentially be thought of as the\nmigration equilibrium outcome of the rural population of Kagera in 1991 across the 3\ndestinations, the rural area (i.e. no migration or another rural area), the secondary towns and the\ncities and across the informal and formal sectors within the two latter destinations. <sup>8</sup> [^8: Intrahousehold decision making is abstracted from.] It provides\nan ideal backdrop to give empirical content to the findings of the model under the different\nscenarios considered (as summarized in Table 1) and explore the model’s empirical relevance.\n\n\n6 We distinguish between urban and rural centers using the 2002 census classification. Urban centers with more than\n500,000 inhabitants are considered cities, the others are towns. Using this definition, the cities are Dar es Salaam\n(the economic capital of Tanzania with an estimated 4.36 million people according to the 2012 census) and Mwanza\n(with an estimated 0.7 million inhabitants)\n7 Of the formally wage employed, 70 and 80 percent work in the private sector in towns and cities, respectively. In\ntowns, 21 percent also work in the public sector, while that share is 11 percent in the cities. The remaining 7 percent\nwork for NGOs or religious organizations, in both the towns and the city\n\nMention: \"2002 census\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its classification separates urban from rural centres.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007056:16:1", "occurrence_id": "393949565510c5ac8a64281d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  another\nrural area. <sup>6</sup> Of the 2,111 individuals who did not move, 331 live in an area that got reclassified\nfrom rural to urban and we drop them from our analysis to avoid confusion. Virtually everyone\nin the rural areas is in the informal sector (only 3.5 percent rely on a formal wage job), so this\ngroup corresponds closely to the rural population we have in mind in the model. In the towns,\n18.9 percent live in a household that relies on formal wage employment. This rises to 22.2\npercent in the cities. <sup>7</sup>\n\n\nThe 2010 income distribution of these individuals could essentially be thought of as the\nmigration equilibrium outcome of the rural population of Kagera in 1991 across the 3\ndestinations, the rural area (i.e. no migration or another rural area), the secondary towns and the\ncities and across the informal and formal sectors within the two latter destinations. <sup>8</sup> [^8: Intrahousehold decision making is abstracted from.] It provides\nan ideal backdrop to give empirical content to the findings of the model under the different\nscenarios considered (as summarized in Table 1) and explore the model’s empirical relevance.\n\n\n6 We distinguish between urban and rural centers using the 2002 census classification. Urban centers with more than\n500,000 inhabitants are considered cities, the others are towns. Using this definition, the cities are Dar es Salaam\n(the economic capital of Tanzania with an estimated 4.36 million people according to the 2012 census) and Mwanza\n(with an estimated 0.7 million inhabitants)\n7 Of the formally wage employed, 70 and 80 percent work in the private sector in towns and cities, respectively. In\ntowns, 21 percent also work in the public sector, while that share is 11 percent in the cities. The remaining 7 percent\nwork for NGOs or religious organizations, in both the towns and the city\n\nMention: \"2012 census\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Estimated populations of the two largest cities come from this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007058:35:1", "occurrence_id": "037e66461910ade73eeed3ad", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .1 51.5 17.5 2.9 100\nEmployer 27.1 52.5 17.5 2.8 100\nSelf-employed 27.9 51.1 17.9 3.2 100\nUnemployed 30.6 51.5 14.9 2.9 100\n_City size_\n100,000 to 1,000,000 26.2 51.7 19.1 2.9 100\nMore than 1,000,000 29.4 51.2 16.8 2.6 100\n<u>Under 100,000</u> <u>26.5</u> <u>52.7</u> <u>18.5</u> <u>2.4</u> <u>100</u>\nSource: Author’s elaboration based on Latinobarómetro.\n\n\n34\n\nMention: \"Latinobarómetro\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Regional attitude shares are elaborated from this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007032:30:0", "occurrence_id": "9ef6034709f73f772418b8b6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Table O.3** African and other countries seeing structural change in the 21st century\n\n\nLabor productivity growth (percent)\n\nCountry Total Within Structural\n\nBotswana\n\n1990–2000 1.1 1.7 –0.6\n\n2000–2010 2.7 4.9 –2.2\n\nGhana\n\n1992–2000 1.0 –0.9 2.0\n\n2000–2006 4.5 6.0 –1.5\n\n2006–2010 2.7 0.0 2.6\n\nNigeria\n\n1996–1999 –0.8 –7.1 6.3\n\n1999–2005 4.4 6.2 –1.8\n\n2005–2009 4.1 –1.6 5.7\n\nNigeria, excluding oil and gas\n\n1996–1999 0.8 –2.0 2.8\n\n1999–2005 4.8 9.4 –4.6\n\n2005–2009 7.0 2.6 4.4\n\nZambia\n\n1991–2002 –2.0 0.0 –2.0\n\n2002–2010 3.6 1.8 1.8\n\nIndia\n\n1990–1999 2.9 1.7 1.3\n\n2000–2004 6.5 6.2 0.3\n\nVietnam\n\n1990–2008 5.1 1.9 3.1\n\n1990–2000 5.2 1.0 4.2\n\n2000–2008 4.9 2.7 2.2\n\nBrazil\n\n1995–2005 0.8 0.6 0.2\n\n1990–2005 0.8 0.8 –0.0\n\n1993/1995–2007/2008 0.5 0.3 0.2\n\n\nSource: _Botswana_ —Value-added and employment data are from the Groningen Growth and Development Centre Africa\nSector Base; _Ghana_ —Economic Survey of Ghana 1961–1982; population and housing censuses 1960, 1970, 1984,\n2000, and 2010; Ghana Living Standard Survey 1991–1992 and 2005–2006; Singal and Nartey (\n\nMention: \"Groningen Growth and Development Centre Africa\nSector Base\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Value added and employment for one country come from this sector base.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007032:30:0", "occurrence_id": "98dc41ce8194e57782bdc04a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Table O.3** African and other countries seeing structural change in the 21st century\n\n\nLabor productivity growth (percent)\n\nCountry Total Within Structural\n\nBotswana\n\n1990–2000 1.1 1.7 –0.6\n\n2000–2010 2.7 4.9 –2.2\n\nGhana\n\n1992��2000 1.0 –0.9 2.0\n\n2000–2006 4.5 6.0 –1.5\n\n2006–2010 2.7 0.0 2.6\n\nNigeria\n\n1996–1999 –0.8 –7.1 6.3\n\n1999–2005 4.4 6.2 –1.8\n\n2005–2009 4.1 –1.6 5.7\n\nNigeria, excluding oil and gas\n\n1996–1999 0.8 –2.0 2.8\n\n1999–2005 4.8 9.4 –4.6\n\n2005–2009 7.0 2.6 4.4\n\nZambia\n\n1991–2002 –2.0 0.0 –2.0\n\n2002–2010 3.6 1.8 1.8\n\nIndia\n\n1990–1999 2.9 1.7 1.3\n\n2000–2004 6.5 6.2 0.3\n\nVietnam\n\n1990–2008 5.1 1.9 3.1\n\n1990–2000 5.2 1.0 4.2\n\n2000–2008 4.9 2.7 2.2\n\nBrazil\n\n1995–2005 0.8 0.6 0.2\n\n1990–2005 0.8 0.8 –0.0\n\n1993/1995–2007/2008 0.5 0.3 0.2\n\n\nSource: _Botswana_ —Value-added and employment data are from the Groningen Growth and Development Centre Africa\nSector Base; _Ghana_ —Economic Survey of Ghana 1961–1982; population and housing censuses 1960, 1970, 1984,\n2000, and 2010; Ghana Living Standard Survey 1991–1992 and 2005–2006; Singal and Nartey (\n\nMention: \"Economic Survey of Ghana\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its output across this span gives the accounting series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007032:30:0", "occurrence_id": "b9d6ae2a1b1ec7a792aa4dd5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Table O.3** African and other countries seeing structural change in the 21st century\n\n\nLabor productivity growth (percent)\n\nCountry Total Within Structural\n\nBotswana\n\n1990–2000 1.1 1.7 –0.6\n\n2000–2010 2.7 4.9 –2.2\n\nGhana\n\n1992–2000 1.0 –0.9 2.0\n\n2000–2006 4.5 6.0 –1.5\n\n2006–2010 2.7 0.0 2.6\n\nNigeria\n\n1996–1999 –0.8 –7.1 6.3\n\n1999–2005 4.4 6.2 –1.8\n\n2005–2009 4.1 –1.6 5.7\n\nNigeria, excluding oil and gas\n\n1996–1999 0.8 –2.0 2.8\n\n1999–2005 4.8 9.4 –4.6\n\n2005–2009 7.0 2.6 4.4\n\nZambia\n\n1991–2002 –2.0 0.0 –2.0\n\n2002–2010 3.6 1.8 1.8\n\nIndia\n\n1990–1999 2.9 1.7 1.3\n\n2000–2004 6.5 6.2 0.3\n\nVietnam\n\n1990–2008 5.1 1.9 3.1\n\n1990–2000 5.2 1.0 4.2\n\n2000–2008 4.9 2.7 2.2\n\nBrazil\n\n1995–2005 0.8 0.6 0.2\n\n1990–2005 0.8 0.8 –0.0\n\n1993/1995–2007/2008 0.5 0.3 0.2\n\n\nSource: _Botswana_ —Value-added and employment data are from the Groningen Growth and Development Centre Africa\nSector Base; _Ghana_ —Economic Survey of Ghana 1961–1982; population and housing censuses 1960, 1970, 1984,\n2000, and 2010; Ghana Living Standard Survey 1991–1992 and 2005–2006; Singal and Nartey (\n\nMention: \"population and housing censuses\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Five enumerations supply the population denominators.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007032:30:0", "occurrence_id": "113bc67f2ee46ca13820ae02", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Table O.3** African and other countries seeing structural change in the 21st century\n\n\nLabor productivity growth (percent)\n\nCountry Total Within Structural\n\nBotswana\n\n1990–2000 1.1 1.7 –0.6\n\n2000–2010 2.7 4.9 –2.2\n\nGhana\n\n1992–2000 1.0 –0.9 2.0\n\n2000–2006 4.5 6.0 –1.5\n\n2006–2010 2.7 0.0 2.6\n\nNigeria\n\n1996–1999 –0.8 –7.1 6.3\n\n1999–2005 4.4 6.2 –1.8\n\n2005–2009 4.1 –1.6 5.7\n\nNigeria, excluding oil and gas\n\n1996–1999 0.8 –2.0 2.8\n\n1999–2005 4.8 9.4 –4.6\n\n2005–2009 7.0 2.6 4.4\n\nZambia\n\n1991–2002 –2.0 0.0 –2.0\n\n2002–2010 3.6 1.8 1.8\n\nIndia\n\n1990–1999 2.9 1.7 1.3\n\n2000–2004 6.5 6.2 0.3\n\nVietnam\n\n1990–2008 5.1 1.9 3.1\n\n1990–2000 5.2 1.0 4.2\n\n2000–2008 4.9 2.7 2.2\n\nBrazil\n\n1995–2005 0.8 0.6 0.2\n\n1990–2005 0.8 0.8 –0.0\n\n1993/1995–2007/2008 0.5 0.3 0.2\n\n\nSource: _Botswana_ —Value-added and employment data are from the Groningen Growth and Development Centre Africa\nSector Base; _Ghana_ —Economic Survey of Ghana 1961–1982; population and housing censuses 1960, 1970, 1984,\n2000, and 2010; Ghana Living Standard Survey 1991–1992 and 2005–2006; Singal and Nartey (\n\nMention: \"Ghana Living Standard Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Two rounds give the household welfare series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007063:1:0", "occurrence_id": "7f1260e1f3cad22b66042efc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Policy Research Working Paper 8078\n\n### **Abstract**\n\nThere is relatively little systematic evidence on the links\n\nbetween procurement systems and outcomes such as competition and corruption levels. This paper adds to the evidence\nbase, using data on almost 34,000 firms from the World\nBank’s Enterprise Surveys, in 88 countries that also have\nprocurement systems data from Public Expenditure and\nFinancial Accountability (PEFA) assessments. The analysis\nfinds that in countries with more transparent procurement\n\nsystems, where exceptions to open competition in tendering\nmust be explicitly justified, firms are more likely to participate in public procurement markets. Moreover, firms report\npaying fewer and smaller kickbacks to officials in countries\nwith more transparent procurement systems, effective and\nindependent complaint mechanisms, and more effective\n\n\n\nexternal auditing systems. These findings—particularly on\nkickbacks—are robust to the inclusion of many controls\nand to a range of sensitivity tests. The study finds evidence\nthat better procurement systems matter more for smaller\nfirms’ participation in procurement markets and payment\nof kickbacks to obtain contracts, consistent with the view\nthat information and transactions costs that are incurred in\nlearning about bidding opportunities and fulfilling bidding\nrequirements are more onerous for smaller firms. Falsification tests show that other, non-procurement indicators\nfrom the PEFA assessments are not associated with procurement outcomes, and that the PEFA procurement indicators\nare not strongly associated with other “governance”-related\noutcomes in firm surveys that are unrelated to procurement.\n\n\n\nThis paper is a product of the Macroeconomics and Growth Team, Development Research Group. It is part of a larger\neffort by the World Bank to provide open access to its research and make a contribution to development policy discussions\naround the world. Policy Research Working Papers are also posted on the Web at http://econ.worldbank.org. The authors\nmay be contacted at sknack@worldbank.org.\n\n\n_The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about\n\nMention: \"Enterprise Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Almost thirty-four thousand firms across 88 countries, matched to procurement data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007064:13:0", "occurrence_id": "e65da9d370a64a518e567605", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: - autonomy in decisions about household/own agricultural production; decision-making power\nover agricultural productive assets (WEAI <sup>16</sup> [^16: WEAI=Women’s Empowerment in Agriculture Index. It includes (i) women’s role in household production\ndecisions; (ii) women’s role in household decisions about use of income; (iii) women’s rights, jointly or individually,\nto own, use and control assets; (iv) women’s leadership in the community; and (v) time use. It also focuses on\nwhether the choice is the women’s own, or is imposed or coerced (one made primarily to avoid negative\nconsequences).], discussed in various publications including Malapit\nand Quisumbing 2015, and Alkire et al. 2013)\n\n  - time use/ allocation of household chores between men and women (WEAI, Garikipati et al 2008)\n\n  - reproductive health views and behaviors (Bandiera et al. 2014)\n\n  - belongs to women’s groups/speaks up in public (WEAI).\n\n\nIn forming an index, some authors use the “relative deprivations” approach followed by Alkire et al.\n(2013) and Garikipati (2008), assigning a score to each dimension and then aggregating the dimensions\ninto a simple index. While providing a clean variable for use in multivariate statistical analysis, this\napproach does not help in understanding which dimensions matter for a specific outcome. In addition,\nPeterman et al. (2015) show that, not surprisingly, how women are ranked on empowerment is highly\ndependent on how the index is constructed. In addition, when Hanmer and Klugman (2016) analyzed\nthe correlations among DHS indicators on attitudes and behaviors that represent dimensions of\nempowerment, they found fairly low correlations, indicating that an index of these variables may be\nproblematic.\n\n\nCan measurement be standardized across countries? The DHS surveys have, for some time, used a\nstandardized set of questions in low-income countries around the world to measure women’\n\nMention: \"DHS surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"A standardized questionnaire across low-income countries aids comparability.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007070:26:0", "occurrence_id": "1cc2cb9090b8b859e4e98eb2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 2. Total Employment in High and Low Financially Dependent Sectors (Index, 2008=100.0)**\n\n\nSource: based on Eurostat National Accounts, detailed sectoral breakdown. Sample includes all countries in EU28. High\n\n\nFD sectors are those in the highest quartile of the distribution of the FD variable calculated using US data (see main text for full\n\n\ndescription of this variable).\n\n\n25\n\nMention: \"Eurostat National Accounts\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Detailed sectoral breakdown across member states in that base year.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007075:15:2", "occurrence_id": "73ff1b4411e7ff91cdc5d89d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: br>0.00<br>0.19<br>0.37<br>0.70<br>0.77<br>0.83<br>1.00|\n\n\n\nNotice first that the correlation between our corrected and uncorrected measure is about 0.87, while\nthe correlation between our measure and Pukthuanthong and Roll’s (2009) is about 0.70. When we\nlook at the correlation of the FGI with the other measures, the measure that displays the highest\ncorrelation with our FGI is Chin and Ito’s _de jure_ one. Among the Lane and Milesi-Ferretti based\nmeasures, instead, the one that displays the larger correlation is the ratio of foreign portfolio equity\nliabilities to GDP (FPEL). The finding is not surprising given our focus on stock market behaviors.\n\nIn order to see how our own measure compares with the other financial integration measures (IMs),\nin table 6, we replicate column 1 in table 3 using all the different measures on the same data set. <sup>17</sup> [^17: This explains why the number of observations dropped from 19,294 to 16,792. The LMF data set is available up to\n2014; hence, the estimates in this table include the 1993-2015 period (given that the IM enters with a lag).] The\nresults are presented in table 6 below, where all variables are standardized, so that the different\ncoefficients can be easily compared.\n\nIt is worth noticing that, using the stock-based measures from Lane and Milesi-Ferretti’s data set,\nfinancial globalization seems to have little effect on volatility, and if it has one, it is an amplifying one.\nContrary, the FGI and CI are negatively correlated with volatility. The fact that the results with our\nmeasure and CI’s are similar was not expected, given the _de jure_ nature of the latter index.\n\n\n16 The index is based on the restrictions on cross-border financial transactions reported in the IMF’\n\nMention: \"LMF data set\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Available through the later year, covering this estimation window.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007075:15:2", "occurrence_id": "e4b833acff2e58cde0c52354", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: br>0.00<br>0.19<br>0.37<br>0.70<br>0.77<br>0.83<br>1.00|\n\n\n\nNotice first that the correlation between our corrected and uncorrected measure is about 0.87, while\nthe correlation between our measure and Pukthuanthong and Roll’s (2009) is about 0.70. When we\nlook at the correlation of the FGI with the other measures, the measure that displays the highest\ncorrelation with our FGI is Chin and Ito’s _de jure_ one. Among the Lane and Milesi-Ferretti based\nmeasures, instead, the one that displays the larger correlation is the ratio of foreign portfolio equity\nliabilities to GDP (FPEL). The finding is not surprising given our focus on stock market behaviors.\n\nIn order to see how our own measure compares with the other financial integration measures (IMs),\nin table 6, we replicate column 1 in table 3 using all the different measures on the same data set. <sup>17</sup> [^17: This explains why the number of observations dropped from 19,294 to 16,792. The LMF data set is available up to\n2014; hence, the estimates in this table include the 1993-2015 period (given that the IM enters with a lag).] The\nresults are presented in table 6 below, where all variables are standardized, so that the different\ncoefficients can be easily compared.\n\nIt is worth noticing that, using the stock-based measures from Lane and Milesi-Ferretti’s data set,\nfinancial globalization seems to have little effect on volatility, and if it has one, it is an amplifying one.\nContrary, the FGI and CI are negatively correlated with volatility. The fact that the results with our\nmeasure and CI’s are similar was not expected, given the _de jure_ nature of the latter index.\n\n\n16 The index is based on the restrictions on cross-border financial transactions reported in the IMF’\n\nMention: \"Lane and Milesi-Ferretti’s data set\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its stock-based measures give the financial globalisation effect.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007082:26:8", "occurrence_id": "80423435fd63e97417e3dfaa", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: br>0.006<br> <br> <br>0.612<br>0.004<br> <br> <br>0.663<br>0.008<br>|<br> <br> <br>0.319<br> <br>0.301<br> <br>0.281<br> <br>0.277<br> <br>0.275<br> <br> <br> <br> <br> <br> <br> <br> <br> <br> <br> <br> <br> <br> <br> <br>|\n\n\nNote: The estimates for 2013 are not comparable due to a change in the reference period for the _health2_\nindicator (see text for details). The standard errors allow for design effects due to stratification and clustering.\nSource: APIS survey data sets for rounds 2004 through 2013; and Balisacan (2015).\n\n\n25\n\nMention: \"APIS survey data sets for rounds 2004 through 2013\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Design effects from stratification and clustering are handled in these rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007087:16:0", "occurrence_id": "1732add3b82833e0fe90f2a9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: error that results from re-scaling all incomes by the GDP per capita growth 𝑟𝑔0,−1 instead of the\n\nincome growth 𝑟𝑦0,−1. This is the equivalent to the _Pass-through Error_ (PTE); this error will get\n\n\ncloser to zero whenever we use a pass-through similar to the actual one. Finally, [𝑃(𝜇𝑔0, 𝐿−1′) −\n\n\n𝑃(𝜇𝑔0, 𝐿−2′)] is the error that arises from using information from the past that does not reflect the\n\n\npresent information. This error is equivalent to the _Informational error_ (IE) in the PE method.\n\n\n**4** **Data and empirical approach for validating poverty nowcasts**\n\n\n_4.1_ _Per capita household income from harmonized household surveys in LAC_\n\nIn order to measure country-specific and regional poverty in this paper, we need to increase cross\n\n\ncountry comparability of welfare measures and poverty lines. Since the LAC region is the focus\n\n\nof the paper, we use the $4 per person per day moderate poverty line in 2005 PPP to measure\n\n\npoverty. In addition, we use the SEDLAC micro data as the primary source of the comparable\n\n\nwelfare aggregate. The SEDLAC project is a harmonized micro database of LAC’s main\n\n\nhouseholds’ surveys produced by the poverty group at the World Bank in partnership with the\n\n\nCenter for Distributive, Labor, and Social Studies (CEDLAS, for its acronym in Spanish) at the\n\nUniversidad Nacional de La Plata in Argentina. <sup>12</sup> [^12: See Castaneda et al. (2016), Bourguignon (2015), and Gasparini, Cicowiez, and Escudero (2013) for a more detailed\ndescription of the SEDLAC project.] The main goal of this project is to increase cross\n\ncountry comparability of many socioeconomic indicators, including total household per capita\n\n\nincome, from more than 300 household surveys within 18 countries spanning more than 20 years\n\nof surveys. <sup>13</sup>\n\n\nThe SEDLAC project has been increasingly used by researchers in the\n\nMention: \"SEDLAC micro data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The primary source of the comparable welfare aggregate at the stated line.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007089:27:1", "occurrence_id": "74f83bf03e90d4af096c0184", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 042\n<u>5</u> <u>-0.146</u> <u>0.035</u> <u>-0.002</u> <u>0.041</u> <u>0.023</u> <u>0.082</u> <u>0.152</u> <u>0.017</u>\nSource: Own estimations based on IHSES 2012\n\n\n**<u>Table 10:</u>** <u>Own-Price Elasticities of Ration Items by quintile of per-capita consumption and area, 2012</u>\n\n<u>Ration products</u> <u>Equivalent Free Market Product</u>\n\n\n\nBrown\n\n\n\nBrown Vegetable\n\nRice Sugar\n<u>Flour</u> <u>Oil</u>\n\n\n\nCereal Rice Sweets Oils\n<u>Oil</u>\n\n\n\nKurdistan\n\n1 -0.032 -0.038 -0.006 -0.005 0.010 -0.004 -0.013 -1.034\n2 -0.037 -0.046 -0.007 -0.007 0.006 -0.002 -0.005 -0.941\n3 -0.040 -0.044 -0.007 -0.007 0.003 -0.004 -0.013 -0.457\n4 -0.048 -0.055 -0.009 -0.009 0.005 0.000 -0.011 -0.541\n5 -0.076 -0.080 -0.016 -0.015 0.006 0.002 -0.010 -0.398\nRest of Iraq\n\n1 -0.030 -0.029 -0.005 -0.004 0.009 -0.007 -\n\nMention: \"IHSES 2012\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Own-price elasticities of ration items by quintile are estimated on this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007091:28:5", "occurrence_id": "47db0d2d2aa41a3f7c29f413", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <br>-0.0022<br>0.0023*<br>0.01**<br>(-4.41)<br>(-0.91)<br>(2.45)<br>(2.93)|\n|Observations<br>Adjusted R-squared|12584<br>0.302|8637<br>8637<br>8637<br>8637<br>0.087<br>0.122<br>0.118<br>0.135|8637<br>8637<br>8637<br>8637<br>0.205<br>0.090<br>0.024<br>0.086|\n\n\nNotes: (1) Source: Author's calculations based on data collected by Social Observatory, World Bank and Government of Bihar. (2) SC is the omitted caste\ngroup. (3) Each column represents a separate regression wherein an `empowerment indicator’ is regressed on variables that are reported as well as\nadditional controls: age, age squared, marital status of the woman, age at marriage, and a dummy variable for some schooling of the female respondent; a\ndummy variable for whether it is female-headed household, per-capital expenditure, per-capita expenditure squared, landholdings, education of the\nhousehold head, number of members in the household and panchayat level fixed-effects. The female employment is run for all female adults in the sample\n(individual level). (4) We report robust standard errors in the brackets. (5) We report the level of significance: * p value < .05, ** p value < .01 and *** p\nvalue < .001.\n\n\n\n\n\n27\n\nMention: \"Social Observatory, World Bank and Government of Bihar\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Field data collected jointly by these three bodies carry the regressions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007059:4:0", "occurrence_id": "9517c8d32b8e47dcad2f7394", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: As consanguinity may be linked to a wide range of human development indicators, we use a unified\n\n\nframework to lay out the decision to enter into consanguineous marriages and potential effects on child\n\n\ndevelopment. We then explore the relationship between consanguinity and two important child\n\n\ndevelopment outcomes: child cognitive ability and stature. The paper makes use of a unique household\n\n\nsurvey from Pakistan that includes information on the marriage patterns of all household members as\n\n\nwell as their parents (regardless of whether they are still alive or present in the household). Human\n\n\ndevelopment outcomes include height and weight measurements for all household members age 5 and\n\n\nabove, cognitive tests administered to children and adults, and a set of questions useful for modeling the\n\n\nmarriage decisions of the parents of household members, including grandparent land ownership status\n\n\nand mortality.\n\n\nThe next section briefly reviews motivations for marriage to a relative, most frequently a first or second\n\n\ncousin, and existing evidence on human development impacts. Section 3 presents a modified version of\n\n\nBecker’s marriage model that is used to motivate the empirical analysis. Data sources are described by\n\n\nSection 4, followed by a discussion of empirical results and conclusions.\n\n\n**2. Consanguineous Marriages**\n\n\n**2.1 Why Marry a Relative?**\n\n\nHigh levels of consanguinity in Pakistan have been well documented. Using the 1990/91 Pakistan\n\n\nDemographic and Health Survey, Hussain and Bittles (1998) estimate that 60 percent of marriages are\n\n\nconsanguineous unions, and that the incidence of consanguinity had remained unchanged over the\n\n\nprevious three to four decades. <sup>2</sup> The decision to marry a relative may be driven by cultural or economic\n\n\nfactors or both.\n\n\nIn some societies, there is a belief that compatibility between husband and wife, as well as between the\n\n\nbride and the rest of the household, will be maximized by marriages among kin (Khlat et al. 1986). Further,\n\n\nin societies\n\nMention: \"household\n\n\nsurvey from Pakistan\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"It records marriage patterns for all household members alongside child outcomes.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007059:4:0", "occurrence_id": "f2c795c4d25664fecaeae298", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: As consanguinity may be linked to a wide range of human development indicators, we use a unified\n\n\nframework to lay out the decision to enter into consanguineous marriages and potential effects on child\n\n\ndevelopment. We then explore the relationship between consanguinity and two important child\n\n\ndevelopment outcomes: child cognitive ability and stature. The paper makes use of a unique household\n\n\nsurvey from Pakistan that includes information on the marriage patterns of all household members as\n\n\nwell as their parents (regardless of whether they are still alive or present in the household). Human\n\n\ndevelopment outcomes include height and weight measurements for all household members age 5 and\n\n\nabove, cognitive tests administered to children and adults, and a set of questions useful for modeling the\n\n\nmarriage decisions of the parents of household members, including grandparent land ownership status\n\n\nand mortality.\n\n\nThe next section briefly reviews motivations for marriage to a relative, most frequently a first or second\n\n\ncousin, and existing evidence on human development impacts. Section 3 presents a modified version of\n\n\nBecker’s marriage model that is used to motivate the empirical analysis. Data sources are described by\n\n\nSection 4, followed by a discussion of empirical results and conclusions.\n\n\n**2. Consanguineous Marriages**\n\n\n**2.1 Why Marry a Relative?**\n\n\nHigh levels of consanguinity in Pakistan have been well documented. Using the 1990/91 Pakistan\n\n\nDemographic and Health Survey, Hussain and Bittles (1998) estimate that 60 percent of marriages are\n\n\nconsanguineous unions, and that the incidence of consanguinity had remained unchanged over the\n\n\nprevious three to four decades. <sup>2</sup> The decision to marry a relative may be driven by cultural or economic\n\n\nfactors or both.\n\n\nIn some societies, there is a belief that compatibility between husband and wife, as well as between the\n\n\nbride and the rest of the household, will be maximized by marriages among kin (Khlat et al. 1986). Further,\n\n\nin societies\n\nMention: \"Pakistan\n\n\nDemographic and Health Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its estimates find most marriages are consanguineous.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007072:26:0", "occurrence_id": "af3c2f3be2258e5bd7306d20", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: regressions include the full set of regressors and upazila and year fixed effects.\n\n\nThe first three columns of Table 3 report the fixed effects regression results for employment by all firms\n\n\n(using population census data) and the last three for employment by the large firms (using data from\n\n\neconomic censuses). In order to facilitate comparison between the economic census data on large firms and\n\n\nthe population census data on all firms, the employment categories are kept at a fairly aggregate level,\n\n\ndistinguishing only between manufacturing and services. <sup>23</sup> The top panel reports the results for the STE\n\n\nsample and the bottom for the rural sample. The number of periods in the census panel data is two and\n\n\neconomic census is three. This is because 1990 census data are used to define the initial characteristics of\n\n\nupazilas, and the economic census has an additional data point for 2006 which is a non-population census\n\n\nyear.\n\n\n**Evidence** **on** **All** **Firms** **(Irrespective** **of** **Size)**\n\n\nThe evidence in the first three columns on all firms shows an interesting pattern: the estimated effect of\n\n\na positive rainfall shock is positive and statistically significant on the manufacturing and services sectors,\n\n\nwhile the effect is negative on agricultural employment, although it is not statistically significant at the\n\n\n10 percent level. This pattern of estimates is observed in both the rural and the STE samples, with the\n\n\nmagnitude of the effect larger for manufacturing and smaller for services in the case of the rural sample.\n\n\nThe negative effect on agriculture, although imprecisely estimated, suggests labor reallocation away from\n\n\nfarming. Note that according to the theoretical analysis, given a labor endowment _L_ <sup>¯</sup>, a higher agricultural\n\n\nproductivity results in labor reallocation away from agriculture when _Ewθ_ _>_ 1. Given that\n\nMention: \"population census data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Employment across all firms is measured from these enumerations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007072:26:0", "occurrence_id": "a0534aecde9b016b66843a55", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: regressions include the full set of regressors and upazila and year fixed effects.\n\n\nThe first three columns of Table 3 report the fixed effects regression results for employment by all firms\n\n\n(using population census data) and the last three for employment by the large firms (using data from\n\n\neconomic censuses). In order to facilitate comparison between the economic census data on large firms and\n\n\nthe population census data on all firms, the employment categories are kept at a fairly aggregate level,\n\n\ndistinguishing only between manufacturing and services. <sup>23</sup> The top panel reports the results for the STE\n\n\nsample and the bottom for the rural sample. The number of periods in the census panel data is two and\n\n\neconomic census is three. This is because 1990 census data are used to define the initial characteristics of\n\n\nupazilas, and the economic census has an additional data point for 2006 which is a non-population census\n\n\nyear.\n\n\n**Evidence** **on** **All** **Firms** **(Irrespective** **of** **Size)**\n\n\nThe evidence in the first three columns on all firms shows an interesting pattern: the estimated effect of\n\n\na positive rainfall shock is positive and statistically significant on the manufacturing and services sectors,\n\n\nwhile the effect is negative on agricultural employment, although it is not statistically significant at the\n\n\n10 percent level. This pattern of estimates is observed in both the rural and the STE samples, with the\n\n\nmagnitude of the effect larger for manufacturing and smaller for services in the case of the rural sample.\n\n\nThe negative effect on agriculture, although imprecisely estimated, suggests labor reallocation away from\n\n\nfarming. Note that according to the theoretical analysis, given a labor endowment _L_ <sup>¯</sup>, a higher agricultural\n\n\nproductivity results in labor reallocation away from agriculture when _Ewθ_ _>_ 1. Given that\n\nMention: \"economic censuses\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Large-firm employment is measured here instead.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007072:26:0", "occurrence_id": "0b8608f327e102ca707c84a8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: regressions include the full set of regressors and upazila and year fixed effects.\n\n\nThe first three columns of Table 3 report the fixed effects regression results for employment by all firms\n\n\n(using population census data) and the last three for employment by the large firms (using data from\n\n\neconomic censuses). In order to facilitate comparison between the economic census data on large firms and\n\n\nthe population census data on all firms, the employment categories are kept at a fairly aggregate level,\n\n\ndistinguishing only between manufacturing and services. <sup>23</sup> The top panel reports the results for the STE\n\n\nsample and the bottom for the rural sample. The number of periods in the census panel data is two and\n\n\neconomic census is three. This is because 1990 census data are used to define the initial characteristics of\n\n\nupazilas, and the economic census has an additional data point for 2006 which is a non-population census\n\n\nyear.\n\n\n**Evidence** **on** **All** **Firms** **(Irrespective** **of** **Size)**\n\n\nThe evidence in the first three columns on all firms shows an interesting pattern: the estimated effect of\n\n\na positive rainfall shock is positive and statistically significant on the manufacturing and services sectors,\n\n\nwhile the effect is negative on agricultural employment, although it is not statistically significant at the\n\n\n10 percent level. This pattern of estimates is observed in both the rural and the STE samples, with the\n\n\nmagnitude of the effect larger for manufacturing and smaller for services in the case of the rural sample.\n\n\nThe negative effect on agriculture, although imprecisely estimated, suggests labor reallocation away from\n\n\nfarming. Note that according to the theoretical analysis, given a labor endowment _L_ <sup>¯</sup>, a higher agricultural\n\n\nproductivity results in labor reallocation away from agriculture when _Ewθ_ _>_ 1. Given that\n\nMention: \"economic census\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"One more period than the rural panel, because the baseline year is used differently.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007072:26:0", "occurrence_id": "e41b53cba7f3b4b2c8c48c50", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: regressions include the full set of regressors and upazila and year fixed effects.\n\n\nThe first three columns of Table 3 report the fixed effects regression results for employment by all firms\n\n\n(using population census data) and the last three for employment by the large firms (using data from\n\n\neconomic censuses). In order to facilitate comparison between the economic census data on large firms and\n\n\nthe population census data on all firms, the employment categories are kept at a fairly aggregate level,\n\n\ndistinguishing only between manufacturing and services. <sup>23</sup> The top panel reports the results for the STE\n\n\nsample and the bottom for the rural sample. The number of periods in the census panel data is two and\n\n\neconomic census is three. This is because 1990 census data are used to define the initial characteristics of\n\n\nupazilas, and the economic census has an additional data point for 2006 which is a non-population census\n\n\nyear.\n\n\n**Evidence** **on** **All** **Firms** **(Irrespective** **of** **Size)**\n\n\nThe evidence in the first three columns on all firms shows an interesting pattern: the estimated effect of\n\n\na positive rainfall shock is positive and statistically significant on the manufacturing and services sectors,\n\n\nwhile the effect is negative on agricultural employment, although it is not statistically significant at the\n\n\n10 percent level. This pattern of estimates is observed in both the rural and the STE samples, with the\n\n\nmagnitude of the effect larger for manufacturing and smaller for services in the case of the rural sample.\n\n\nThe negative effect on agriculture, although imprecisely estimated, suggests labor reallocation away from\n\n\nfarming. Note that according to the theoretical analysis, given a labor endowment _L_ <sup>¯</sup>, a higher agricultural\n\n\nproductivity results in labor reallocation away from agriculture when _Ewθ_ _>_ 1. Given that\n\nMention: \"1990 census data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Used to define initial characteristics of the administrative units.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007102:23:5", "occurrence_id": "517654854365e14960115dc6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: br>-6,276.02<br>***<br>-5,761.33<br>***<br>-8,179.54<br>***<br> <br> <br> <br> <br> <br> <br> <br> <br> <br> <br>**3.94**<br> <br>**3.72**<br> <br>**3.21**<br> <br>1,063<br> <br>1,063<br> <br>1,063<br> <br>1,063<br>|\n\n\n\n_Source: Authors’ computation based on ERHS 2004-2009_\n_*** p<0.01, ** p<0.05, * p<0.1_\n\n\n22\n\nMention: \"ERHS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Computations across these rounds carry the reported estimates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007107:5:2", "occurrence_id": "1a93cf41c3e958fdf7bd5bf9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  we estimate the global impact of AMR with the features described above:\na well-defined global CGE model that has dynamics and makes full use of the GTAP database, the\ninclusion of both microeconomic assessments on health and other channels of shock, and analysis\n\n\n4\n\nMention: \"the GTAP database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The global model with dynamics is built on this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007109:29:0", "occurrence_id": "a0b737e45afaf02d6f3f67c3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **5.6** **Testing** **for** **different** **size** **of** **forecasting** **windows**\n\n\nThe IMF produces 5-year forecasts in each WEO edition. We use this five yearly forecast as our default\n\n\ntesting horizon, since it captures the maximum information regarding a country’s growth trajectory in any\n\n\ngiven year. In order to check the robustness of our results we can narrow this forecast window by excluding\n\n\nouter years from our variables and re-run our main specification.\n\n\nTable 16 below presents our results using forecast windows of 1, 2, 3 and 4 years in addition to our\n\n\nbaseline estimates using the 5 year window. For windows shorter than 4-years we fail to find a significant\n\n\neffect. This suggests that shorter windows we are unable to measure with precision our negative effect on the\n\n\nforecast error. The magnitude and significance is maximized at the longest window available in the WEO\n\n\nforecasts.\n\n\n(1) (2) (3) (4) (5)\nDiff 1y Diff 2y Diff 3y Diff 4y Diff 5y\n\n\nLag disc dum\n\n(disc in past 2 years (excl. t=0)) -0.599 -0.527 -0.350 -0.597 <sup>_∗∗_</sup> -0.833 <sup>_∗∗∗_</sup>\n\n<u>(0.509)</u> <u>(0.381)</u> <u>(0.305)</u> <u>(0.295)</u> <u>(0.308)</u>\nDep var 1 yr window 2 yr window 3 yr window 4 yr window 5 yr window\nFixed effects CFE YFE CFE YFE CFE Y\n\nMention: \"WEO\n\n\nforecasts\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its longest available horizon gives the largest effect.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007116:31:1", "occurrence_id": "6866a11f7eaf768a081398bb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: >_\n\n\nThis approach to measuring decision-making roles has not been changed or adapted substantially\n\n\nover time, with its use increasing, especially in large-scale surveys in developing countries. This focus on\n\n\nthe main decision-maker stems from the notion that the greater number of decisions an individual is\n\n\ninvolved in, the greater control they have over their own life (Kishor 2005). The nationally-representative\n\n\nDemographic and Health Surveys (DHS), conducted in over 90 countries, form the largest source of data\n\n\non women’s participation in decision-making. Initially, the surveys used slightly different question\n\nwordings depending on the topic at hand. For example, questions regarding decision-making over health\n\n\ncare, household purchases, consumption, and visiting friends or relatives would ask “who in your family\n\n\nusually has the final say on [X]”, while questions regarding control over use of earnings would ask “who\n\n\nmainly decides how the money you earn will be used?” More recently, the DHS decision-making questions\n\n\nhave been standardized across countries and decision-making areas (‘domains’), asking the respondent\n\n\n“who usually makes decisions about [X]” (see Box 8 below).\n\n\n30\n\nMention: \"Demographic and Health Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Run in over ninety countries, the largest source on the measured outcomes.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007120:8:0", "occurrence_id": "42b64f5ce98b9a2b22ac017a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **<u>Table 1. Ownership status in percentage by region (share within regions) in Egypt 2012/13</u>**\n\n\n\n\n\n\n\n\n\n\n\n\n\n|Ownership status|Region|Col3|Col4|Col5|Col6|\n|---|---|---|---|---|---|\n|**_Ownership status_**|**Metropolitan**|**Lower**<br>**Urban**|**Lower**<br>**Rural**|**Upper**<br>**Urban**|**Upper**<br>**Rural**|\n|**Rent ‐ OLD**|31.7|13.9|0.6|18.7|1.7|\n|**Rent ‐ NEW**|9.6|14.4|2.1|9.8|1.0|\n|**Owner**|51.7|56.9|79.4|57.1|82.7|\n|**In‐kind/donation**|7.0|14.8|17.9|14.5|14.7|\n\n\nSource: HIECS, 2012/13\nNote: Rent‐OLD refers to subsidized tenants and Rent‐NEW refers to tenants.\n\nThe effects of rental regulation are clear when we look at the values of rents declared by households in\nregulated units and compared these to the declared rent for “new law” tenants. Looking at the\ndistributions of rent paid (Figure 1), it is clear rents paid by households living in unregulated dwellings is\nhigher, on average. Moreover, the distribution of rents paid by households living in regulated dwellings is\nalmost entirely to the left of the distribution of the rents among\n\nMention: \"HIECS, 2012/13\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Consumption shares by tenure category come from this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007122:13:0", "occurrence_id": "c60c23dbde6565ecbe5520bd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: likely be the first one in a developed country asking questions about sexual orientation and\n[gender identity. In Uruguay, an effort is underway to conduct a census among transgender](http://www.mides.gub.uy/innovaportal/v/61348/3/innova.front/numeros-para-la-inclusion)\npeople.\n\n##### **4. Methodological overview: How developmental outcomes for LGBTI people** **have been measured through surveys**\n\n\nIn the following section, we will discuss four surveys of LGBTI people that have attempted\nto assess development outcomes: the previously mentioned survey in Nepal, a survey in\nIndia; the FRA survey from Europe; and the largest survey of intersex people from\nAustralia. For each survey, we will focus on the methodologies, constraints, and lessons.\n\n**Nepal**\n\nThe Williams Institute and UNDP conducted one of the <u>[most recent surveys on sexual](http://williamsinstitute.law.ucla.edu/wp-content/uploads/Nepal-Survey-Oct-2014.pdf)</u>\n<u>[orientation, gender identity and/or expression](http://williamsinstitute.law.ucla.edu/wp-content/uploads/Nepal-Survey-Oct-2014.pdf)</u> in cooperation with a local Nepalese NGO,\nBlue Diamond Society. <sup>36</sup> [^36: This included a feedback on the methodology, survey tool and data analysis.] Respondents were asked a set of questions, including open-ended\nquestions, on their sexual orientation, gender identity and/or expression covering the selfidentification, behavioral, and attraction dimensions of all three. A number of survey\nenumerators from the Blue Diamond Society were recruited, trained and ultimately able to\nreach 1,178 respondents. <sup>37</sup>\n\nMost of the respondents were recruited through local\n\nMention: \"survey in Nepal\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"One of several attempts to measure outcomes for this population.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007122:13:0", "occurrence_id": "9905f72334e76daf258cd41a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: likely be the first one in a developed country asking questions about sexual orientation and\n[gender identity. In Uruguay, an effort is underway to conduct a census among transgender](http://www.mides.gub.uy/innovaportal/v/61348/3/innova.front/numeros-para-la-inclusion)\npeople.\n\n##### **4. Methodological overview: How developmental outcomes for LGBTI people** **have been measured through surveys**\n\n\nIn the following section, we will discuss four surveys of LGBTI people that have attempted\nto assess development outcomes: the previously mentioned survey in Nepal, a survey in\nIndia; the FRA survey from Europe; and the largest survey of intersex people from\nAustralia. For each survey, we will focus on the methodologies, constraints, and lessons.\n\n**Nepal**\n\nThe Williams Institute and UNDP conducted one of the <u>[most recent surveys on sexual](http://williamsinstitute.law.ucla.edu/wp-content/uploads/Nepal-Survey-Oct-2014.pdf)</u>\n<u>[orientation, gender identity and/or expression](http://williamsinstitute.law.ucla.edu/wp-content/uploads/Nepal-Survey-Oct-2014.pdf)</u> in cooperation with a local Nepalese NGO,\nBlue Diamond Society. <sup>36</sup> [^36: This included a feedback on the methodology, survey tool and data analysis.] Respondents were asked a set of questions, including open-ended\nquestions, on their sexual orientation, gender identity and/or expression covering the selfidentification, behavioral, and attraction dimensions of all three. A number of survey\nenumerators from the Blue Diamond Society were recruited, trained and ultimately able to\nreach 1,178 respondents. <sup>37</sup>\n\nMost of the respondents were recruited through local\n\nMention: \"FRA survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The European instance in the set of comparable exercises.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007122:13:0", "occurrence_id": "357569836a971a2b32e5f6e6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: likely be the first one in a developed country asking questions about sexual orientation and\n[gender identity. In Uruguay, an effort is underway to conduct a census among transgender](http://www.mides.gub.uy/innovaportal/v/61348/3/innova.front/numeros-para-la-inclusion)\npeople.\n\n##### **4. Methodological overview: How developmental outcomes for LGBTI people** **have been measured through surveys**\n\n\nIn the following section, we will discuss four surveys of LGBTI people that have attempted\nto assess development outcomes: the previously mentioned survey in Nepal, a survey in\nIndia; the FRA survey from Europe; and the largest survey of intersex people from\nAustralia. For each survey, we will focus on the methodologies, constraints, and lessons.\n\n**Nepal**\n\nThe Williams Institute and UNDP conducted one of the <u>[most recent surveys on sexual](http://williamsinstitute.law.ucla.edu/wp-content/uploads/Nepal-Survey-Oct-2014.pdf)</u>\n<u>[orientation, gender identity and/or expression](http://williamsinstitute.law.ucla.edu/wp-content/uploads/Nepal-Survey-Oct-2014.pdf)</u> in cooperation with a local Nepalese NGO,\nBlue Diamond Society. <sup>36</sup> [^36: This included a feedback on the methodology, survey tool and data analysis.] Respondents were asked a set of questions, including open-ended\nquestions, on their sexual orientation, gender identity and/or expression covering the selfidentification, behavioral, and attraction dimensions of all three. A number of survey\nenumerators from the Blue Diamond Society were recruited, trained and ultimately able to\nreach 1,178 respondents. <sup>37</sup>\n\nMost of the respondents were recruited through local\n\nMention: \"largest survey of intersex people from\nAustralia\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The biggest such exercise, from a third country.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007128:8:3", "occurrence_id": "b3c54784cb5bc35cb9bc9b77", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ,m,t_ are respectively the online and total exports from\n\n\ncountry _x_ to country _m_ in year _t_ . _βx,m_ and _αx,m_ control for time-invariant country-pair specific\n\n\ndeterminants of trade flows such as distance, colonial linkages, common language, etc. _βx,t_ and _αx,t_\n\n\nare year and country specific shocks in the exporting country. We use the estimate in equations (2)\n\n(7) to predict the log of bilateral online and total exports. To estimate the impact of online trade\n\n\non the wage gap, we need to aggregate the bilateral exports flows at the exporter and year level.\n\n\nWe first take the exponential of the predicted bilateral trade flows estimated by equations (2)-(7).\n\n\nWe then aggregate these trade flows at the exporter and year level by summing the trade flows over\n\n\nall importers each year. After taking logs we use these six variables to instrument the log share of\n\n\nonline and total exports in GDP.\n\n##### **3 Data**\n\n\nThe variable of interest is the wage gap between skilled and unskilled workers. We borrow the\n\n\ndata and methodology from Cruz and Milet (2017). The data source is the International Income\n\n\nDistribution Database (I2D2) put together by the World Bank which compiles more than 1,000\n\n\nhousehold and labor force surveys from 164 countries over the period 1960-2014 covering more\n\n\nthan 120 million individuals. The data have been harmonized to the extent possible ensuring that\n\nconceptual variables have similar meaning, allowing for cross-country analysis. <sup>4</sup> For the purpose of\n\n\nthis study, we selected a restricted sample of 22 countries for which sufficient data are available. <sup>5</sup>\n\n\n4See Montenegro and Hirn (2009) and Montenegro and Patrinos (2014) for more information about\n\nMention: \"I2D2\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Compiles over a thousand household and labour surveys.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007132:9:0", "occurrence_id": "ee16183973c3ccb3d09ea126", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: weighted by population at the district level and are obtained by overlaying the gridded CRU TS\n\n\nclimate data with population and district-level grid maps.\n\n\nData on demographics and socioeconomics at the individual, household and district levels are\n\n\ndrawn from three main sources. The first one is the 2008-09 income and expenditure household\n\n\nsurvey (known as IOF, for its acronym in Portuguese) collected by the National Institute of\n\n\nStatistics of Mozambique. The IOF survey is representative at the province level and, in addition\n\n\nto providing detailed data on expenditures (including self-consumption) and incomes, it contains\n\n\na wealth of information on sociodemographic characteristics, education, health, children’s\n\n\nanthropometrics, labor market indicators and housing conditions. The second source of data is the\n\n\n2011 Demographic and Health Survey (DHS), which provides a range of variables in aspects\n\n\nrelated to population, health and nutrition. Lastly, we used a 10% random sample of the 2007\n\n\npopulation census to obtain additional information on human capital indicators.\n\n\n_Rainfall shocks_\n\n\nMozambique has a tropical climate that is characterized by two seasons. The wet (rainy)\n\n\nseason goes from October to March, and is characterized by heavier rainfall between December\n\n\nand March. Most cyclones are also experienced over the same period, especially in the coastal\n\n\nareas. The dry season extends from April to September, and is particularly pronounced in the south\n\n\nof the country, making it more prone to drought. Due to Mozambique’s location in the tropical\n\n\nzone, temperatures do not fluctuate much within and across seasons.\n\n\nRainfall shocks are prevalent across the country. We define them as district-level annual\n\n\ncumulative rainfall events between 1950 and 2014 that were two standard deviations above (for\n\n\nfloods) or below (for droughts) the historical annual mean for the corresponding district. The\n\n\nannual historical mean is calculated over the 1901-2014 period for each of the 128 districts\n\n\ncomprising the 11 provinces of Mozambique. It is observed that extreme precipitation (whether in\n\n\ndeficit\n\nMention: \"CRU TS\n\n\nclimate data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Gridded observations overlaid with population and district maps.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007132:9:0", "occurrence_id": "1e442dc4f954158cb51b455d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: weighted by population at the district level and are obtained by overlaying the gridded CRU TS\n\n\nclimate data with population and district-level grid maps.\n\n\nData on demographics and socioeconomics at the individual, household and district levels are\n\n\ndrawn from three main sources. The first one is the 2008-09 income and expenditure household\n\n\nsurvey (known as IOF, for its acronym in Portuguese) collected by the National Institute of\n\n\nStatistics of Mozambique. The IOF survey is representative at the province level and, in addition\n\n\nto providing detailed data on expenditures (including self-consumption) and incomes, it contains\n\n\na wealth of information on sociodemographic characteristics, education, health, children’s\n\n\nanthropometrics, labor market indicators and housing conditions. The second source of data is the\n\n\n2011 Demographic and Health Survey (DHS), which provides a range of variables in aspects\n\n\nrelated to population, health and nutrition. Lastly, we used a 10% random sample of the 2007\n\n\npopulation census to obtain additional information on human capital indicators.\n\n\n_Rainfall shocks_\n\n\nMozambique has a tropical climate that is characterized by two seasons. The wet (rainy)\n\n\nseason goes from October to March, and is characterized by heavier rainfall between December\n\n\nand March. Most cyclones are also experienced over the same period, especially in the coastal\n\n\nareas. The dry season extends from April to September, and is particularly pronounced in the south\n\n\nof the country, making it more prone to drought. Due to Mozambique’s location in the tropical\n\n\nzone, temperatures do not fluctuate much within and across seasons.\n\n\nRainfall shocks are prevalent across the country. We define them as district-level annual\n\n\ncumulative rainfall events between 1950 and 2014 that were two standard deviations above (for\n\n\nfloods) or below (for droughts) the historical annual mean for the corresponding district. The\n\n\nannual historical mean is calculated over the 1901-2014 period for each of the 128 districts\n\n\ncomprising the 11 provinces of Mozambique. It is observed that extreme precipitation (whether in\n\n\ndeficit\n\nMention: \"2008-09 income and expenditure household\n\n\nsurvey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The first of three sources for household and district welfare.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007132:9:0", "occurrence_id": "66a3e8d996f3f52edb734e06", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: weighted by population at the district level and are obtained by overlaying the gridded CRU TS\n\n\nclimate data with population and district-level grid maps.\n\n\nData on demographics and socioeconomics at the individual, household and district levels are\n\n\ndrawn from three main sources. The first one is the 2008-09 income and expenditure household\n\n\nsurvey (known as IOF, for its acronym in Portuguese) collected by the National Institute of\n\n\nStatistics of Mozambique. The IOF survey is representative at the province level and, in addition\n\n\nto providing detailed data on expenditures (including self-consumption) and incomes, it contains\n\n\na wealth of information on sociodemographic characteristics, education, health, children’s\n\n\nanthropometrics, labor market indicators and housing conditions. The second source of data is the\n\n\n2011 Demographic and Health Survey (DHS), which provides a range of variables in aspects\n\n\nrelated to population, health and nutrition. Lastly, we used a 10% random sample of the 2007\n\n\npopulation census to obtain additional information on human capital indicators.\n\n\n_Rainfall shocks_\n\n\nMozambique has a tropical climate that is characterized by two seasons. The wet (rainy)\n\n\nseason goes from October to March, and is characterized by heavier rainfall between December\n\n\nand March. Most cyclones are also experienced over the same period, especially in the coastal\n\n\nareas. The dry season extends from April to September, and is particularly pronounced in the south\n\n\nof the country, making it more prone to drought. Due to Mozambique’s location in the tropical\n\n\nzone, temperatures do not fluctuate much within and across seasons.\n\n\nRainfall shocks are prevalent across the country. We define them as district-level annual\n\n\ncumulative rainfall events between 1950 and 2014 that were two standard deviations above (for\n\n\nfloods) or below (for droughts) the historical annual mean for the corresponding district. The\n\n\nannual historical mean is calculated over the 1901-2014 period for each of the 128 districts\n\n\ncomprising the 11 provinces of Mozambique. It is observed that extreme precipitation (whether in\n\n\ndeficit\n\nMention: \"2011 Demographic and Health Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Labour market, housing, population and health variables.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007132:9:0", "occurrence_id": "80672898db5ed4edbab64882", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: weighted by population at the district level and are obtained by overlaying the gridded CRU TS\n\n\nclimate data with population and district-level grid maps.\n\n\nData on demographics and socioeconomics at the individual, household and district levels are\n\n\ndrawn from three main sources. The first one is the 2008-09 income and expenditure household\n\n\nsurvey (known as IOF, for its acronym in Portuguese) collected by the National Institute of\n\n\nStatistics of Mozambique. The IOF survey is representative at the province level and, in addition\n\n\nto providing detailed data on expenditures (including self-consumption) and incomes, it contains\n\n\na wealth of information on sociodemographic characteristics, education, health, children’s\n\n\nanthropometrics, labor market indicators and housing conditions. The second source of data is the\n\n\n2011 Demographic and Health Survey (DHS), which provides a range of variables in aspects\n\n\nrelated to population, health and nutrition. Lastly, we used a 10% random sample of the 2007\n\n\npopulation census to obtain additional information on human capital indicators.\n\n\n_Rainfall shocks_\n\n\nMozambique has a tropical climate that is characterized by two seasons. The wet (rainy)\n\n\nseason goes from October to March, and is characterized by heavier rainfall between December\n\n\nand March. Most cyclones are also experienced over the same period, especially in the coastal\n\n\nareas. The dry season extends from April to September, and is particularly pronounced in the south\n\n\nof the country, making it more prone to drought. Due to Mozambique’s location in the tropical\n\n\nzone, temperatures do not fluctuate much within and across seasons.\n\n\nRainfall shocks are prevalent across the country. We define them as district-level annual\n\n\ncumulative rainfall events between 1950 and 2014 that were two standard deviations above (for\n\n\nfloods) or below (for droughts) the historical annual mean for the corresponding district. The\n\n\nannual historical mean is calculated over the 1901-2014 period for each of the 128 districts\n\n\ncomprising the 11 provinces of Mozambique. It is observed that extreme precipitation (whether in\n\n\ndeficit\n\nMention: \"2007\n\n\npopulation census\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"A ten percent random sample used for human capital indicators.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007133:50:2", "occurrence_id": "b9cd56975ea9171e58046bac", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  (e.g. exports of final\ngoods)\n\n\nShare of a country's exports and imports with its top-3 trading partners. These top-3\npartners are defined as the three partners with the largest value of bilateral total trade in a\ngiven year.\n\nShare of a country's exports and imports with the top-3 most central countries in the global\ntrade network. The top-3 most central countries in the global trade network are those with\nthe greatest value of the random walk betweenness centrality measure developed by\nNewman (2005) and Fisher and Vega-Redondo (2006). This classification is conducted\nseparately for every year in the sample period.\n\n\nShare of a country's exports and imports with countries in the core and in the inner\nperiphery of the global trade network. Core countries are those ranked above the 95 <sup>th</sup>\n\npercentile of the cross-country rankings given by the random walk betweenness centrality\nmeasure developed by Newman (2005) and Fisher and Vega-Redondo (2006). Innerperiphery countries are ranked in the 70-94 <sup>th</sup> percentiles of the rakings. This classification is\nconducted separately for every year in the sample period.\n\n\n\nCalculations based on 3-digit SITC\nRevision 2 data of Feenstra et al.\n(2005), updated with Comtrade data\n\n\nCalculations based on 2-digit SITC\nRevision 2 data of Feenstra et al.\n(2005), updated with Comtrade data\n\n\nCalculations based on 4-digit SITC\nRevision 2 data of Feenstra et al.\n(2005), updated with Comtrade data\n\n\nCalculations based on 4-digit SITC\nRevision 2 data of Feenstra et al.\n(2005), updated with Comtrade data\n\n\nCalculations based on DOTS\n\n\nCalculations based on DOTS\n\n\nCalculations based on DOTS\n\nMention: \"Comtrade data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Updates the earlier bilateral flow compilation at the stated classification.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007140:19:0", "occurrence_id": "5e4f258f5b8b91a020b733d4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **6.4 Life expectancy at older ages**\n\n\nHazan (2012) argues that life expectancy at birth exhibits more variation across economies\n\n\nand cohorts due to high infant and child mortality. He suggests that life expectancy at age five\n\n\nwill be more appropriate to capture true effect of health on human capital investment decisions.\n\n\nAlthough the model laid out in section 4 shows that life expectancy at age 5 or 10 is endogenous\n\n\nbecause it will reflect parental decisions on health investments, it is still useful to examine the\n\n\nsensitivity of our estimates to alternative estimates of life expectancy. As life expectancy at\n\n\nexact ages is not available before 1950 for many economies, we use samples of birth cohorts\n\n\nbeginning in 1945 and 1940. The life expectancies are only available for birth cohorts in 5-year\n\n\nage ranges.\n\n\nTable 5 reports that the effect of life expectancy at higher ages on time spent in school is\n\n\nconsistently positive and statistically significant, unlike Hazan’s (2012) finding of no\n\n\nrelationship. A one-year gain in life expectancy at age 5, 10, and 15 increases time in school by\n\n\n0.185 year, 0.17 year, and 0.129 year, respectively. Parents’ life expectancy turns out to be\n\n\npositive and statistically significant in specifications with life expectancy at birth, but not\n\n\nconsistently for life expectancy at higher ages. Note that the value of parental life expectancy\n\n\nfalls, at least in precision, as we measure life expectancy at higher ages, which probably indicates\n\n\nthat parents’ endowment is not as crucial as it is in early childhood.\n\n\n**6.5 Robustness checks**\n\n\nThis section incorporates several economy-cohort specific measures to check the robustness\n\n\nof the effects of life expectancy at birth on completed years of schooling. To investigate the\n\n\nquality of institutions and political regimes at the time of one’s birth, we utilize a polity measure\n\n\nthat ranks economies by their strength of democratic institutions. We use Polity IV data, which\n\n\n18\n\nMention: \"Polity IV data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Ranks economies by the strength of democratic institutions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007144:36:3", "occurrence_id": "0f6db925f13fdd1b74c78ccc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 9|14.6|11.3|13.2|18.3|15.5|12.0|14.8|\n|Decent jobs|67.1|66.3|58.0|53.5|46.3|54.9|59.6|53.5|47.4|48.0|\n|Underemployment|70.6|65.9|56.9|49.6|43.0|59.9|56.9|46.4|40.3|44.3|\n|Termination of usual activity|31.4|21.3|23.7|18.4|17.2|25.0|19.5|17.3|13.8|12.2|\n|Access to health services|23.1|12.3|12.8|12.7|15.7|14.2|15.1|11.1|11.1|2.8|\n|Quality of health services|57.8|38.3|34.1|18.7|9.5|31.6|37.0|30.3|16.4|7.2|\n|Affordability of health services|25.6|27.4|19.2|7.9|0.6|33.7|16.8|16.3|9.5|3.8|\n\n\n\n_Source_ : ILCS data.\n\n\n_Note_ : * VP = Very poor, P = Poor, BM = Below middle, M = Middle, and OM = Over Middle. Rich is not presented\nbecause not enough households defined themselves in this category.\n\n\n35\n\nMention: \"ILCS data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Perceived quality of health services by welfare class comes from these data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007150:3:2", "occurrence_id": "a5ea8f141aec3a198e678087", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  percent) over the same period (Uganda Bureau of</mark>\n<mark>Statistics 2014). Yet some of this expansion is due to a redefinition of administrative boundaries for urban</mark>\n<mark>areas. An alternative measure of urbanization that is comparable across countries, the agglomeration</mark>\nindex, suggests that Uganda’s urban share is even higher – at 25 percent (World Bank 2012).\n\n\n<mark>While there are some studies that examine migration in Uganda, most rely on cross-sectional data that</mark>\n<mark>pose methodological challenges for identification. Moreover, none of these analyses provides causal</mark>\n<mark>evidence of the impact of migration on welfare. In this paper, we examine the relationship between spatial</mark>\n<mark>and economic mobility in Uganda using data from 2005 through 2012. The paper makes two principal</mark>\n<mark>contributions to the literature. First, it contributes to the policy and academic literature on internal</mark>\n<mark>migration and structural transformation in developing countries through the use of multiple waves of</mark>\n<mark>panel evidence. The use of longitudinal data confers an advantage over cross-sectional analyses which,</mark>\n<mark>though more common in the literature, fail to properly account for time-invariant factors that can</mark>\n\n\n3 <mark>While economic considerations lead many of Uganda’s migrants to move, other factors also drive migration</mark>\n<mark>decisions. For example, insecurity and conflict, particularly in the North of the country during the 2000s, prompted</mark>\n<mark>the displacement and forced migration of large segments of the rural population (Mulumba and Olema 2009). A</mark>\n<mark>period of reverse migration then followed, with an influx of displaced residents returning to\n\nMention: \"data from 2005 through 2012\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Spatial and economic mobility are related across these years.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007151:16:0", "occurrence_id": "b8ea74cec763cd103d085593", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Table 3.2 reports the five Brazilian occupations with the strongest requirements for each\n\n\nbroad activity bundle and the five Brazilian occupations with the weakest requirements for\n\neach broad activity bundle. Car washers and window washers are among the most routine\n\n\noccupations, while orchestral musicians and missionaries are among the least routine\n\n\noccupations in Brazil. By contrast, psychiatrists are highly non-routine task intensive and\n\n\nhand knitters are among the least non-routine task intensive occupations in Brazil.\n\n\nOur main goal is to assess how intensively such activity bundles are used by Brazilian\n\n\nestablishments, firms, and industries, and whether the intensity changes over time in\n\nresponse to the adoption of digital technologies. The final step in calculating the dependent\n\n\nvariables in our analysis aggregates over all workers (and their respective occupations) in an\n\n\nindustry-municipality in a given year. Since the occupation information from O*NET is\n\n\nspecific to the year 2000, the time variation in these dependent variables results as\n\nemployers shift the workforce composition over time. In the first column of Table 3.3, we\n\n\nreport average values across all establishments for the main dependent variable—the ratio\n\n\nof non-routine to routine tasks. We note that the relative value of non-routine to routine tasks\n\nis always less than one, signaling Brazil’s comparative advantage in lower-skilled tasks due\n\n\nto an abundance of lower-skilled labor. However, over the 10-year time horizon, the relative\n\n\nimportance of non-routine tasks in the workforce has steadily increased, by about two\n\npercentage points.\n\n\n**3.3.** **Digital technologies data**\n\n\n**Rollout of internet services** Beginning in 1999, the Brazilian Census Bureau (IBGE)\n\n\nhas conducted an annual survey of all Brazilian municipalities, the _Pesquisa de Informações_\n\n\n_Básicas Municipais_ (known as MUNIC), on the structure and functions of municipal\n\ninstitutions. In the years 1999, 2001, 2005, and 2006\n\nMention: \"O*NET\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Occupation information specific to that year anchors the task measures.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007151:16:0", "occurrence_id": "35252c250eed3ca8ad4937b3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Table 3.2 reports the five Brazilian occupations with the strongest requirements for each\n\n\nbroad activity bundle and the five Brazilian occupations with the weakest requirements for\n\neach broad activity bundle. Car washers and window washers are among the most routine\n\n\noccupations, while orchestral musicians and missionaries are among the least routine\n\n\noccupations in Brazil. By contrast, psychiatrists are highly non-routine task intensive and\n\n\nhand knitters are among the least non-routine task intensive occupations in Brazil.\n\n\nOur main goal is to assess how intensively such activity bundles are used by Brazilian\n\n\nestablishments, firms, and industries, and whether the intensity changes over time in\n\nresponse to the adoption of digital technologies. The final step in calculating the dependent\n\n\nvariables in our analysis aggregates over all workers (and their respective occupations) in an\n\n\nindustry-municipality in a given year. Since the occupation information from O*NET is\n\n\nspecific to the year 2000, the time variation in these dependent variables results as\n\nemployers shift the workforce composition over time. In the first column of Table 3.3, we\n\n\nreport average values across all establishments for the main dependent variable—the ratio\n\n\nof non-routine to routine tasks. We note that the relative value of non-routine to routine tasks\n\nis always less than one, signaling Brazil’s comparative advantage in lower-skilled tasks due\n\n\nto an abundance of lower-skilled labor. However, over the 10-year time horizon, the relative\n\n\nimportance of non-routine tasks in the workforce has steadily increased, by about two\n\npercentage points.\n\n\n**3.3.** **Digital technologies data**\n\n\n**Rollout of internet services** Beginning in 1999, the Brazilian Census Bureau (IBGE)\n\n\nhas conducted an annual survey of all Brazilian municipalities, the _Pesquisa de Informações_\n\n\n_Básicas Municipais_ (known as MUNIC), on the structure and functions of municipal\n\ninstitutions. In the years 1999, 2001, 2005, and 2006\n\nMention: \"MUNIC\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its municipal survey describes the structure and functions of local institutions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007152:26:0", "occurrence_id": "600c19a82461fb6adb45eae6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: clouds could have traveled past the closest districts by the time an image was taken. Regardless, the\nresults are uniform in that all affected districts are neighbors. Overall, the model seems to give a fair\npicture of when and where the eruptions were at their most intense, although the ground level intensity\ncan be hard to specify.\n\n\nTable 15: 10 Most Impacted Provinces\n\n\n<u>Province</u> <u>Year</u> <u>Intensity</u>\nJawa Tengah 2010 34.641\nYogyakarta 2010 18.607\nSulawesi Utara 2004 16.224\nJawa Barat 2010 10.271\nSulawesi Utara 2007 6.387\nSulawesi Utara 2008 4.861\nNusa Tenggara Timur 2014 2.356\nAceh 2014 2.134\nSulawesi Utara 2005 0.684\n<u>Jawa</u> <u>Timur</u> <u>2010</u> <u>0.660</u>\nNote: Multiplied by 1,000\n\n\nTable 16: Aggregated Volcano Intensity Data by Province and Year\n\n\n<u>Province</u> <u>2004</u> <u>2005</u> <u>2007</u> <u>2008</u> <u>2010</u> <u>2014</u>\nAceh 2.134\nBanten 0.163\nJawa Barat 10.271\nJawa Tengah 34.641\nJawa Timur 0.660 0.432\nNusa Tenggara Timur 2.356\nPapua 0.061\nSulawesi Utara 16.224 0.684 6.387 4.861\nSumatera Utara 0.487\n<u>Yogyakarta</u> <u>18.607</u>\n\n\n25\n\nMention: \"Aggregated Volcano Intensity Data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Hazard intensity by province and year is tabulated from this aggregation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007161:19:2", "occurrence_id": "c99da71b6e7aaddac108e7dc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  63 42,307 10,261 24.3 1,144 2.7\nMangochi (MI) 1,530 878 242 104,871 44,607 42.5 7,729 7.4\nMulanje (MJ) 165 49 5 23,833 760 3.2 8 Mwanza (MN) 148 30 25 10,709 1,569 14.7 705 6.6\nNeno (NE) 10 - 1 232 - - 18 7.6\nNsanje (NJ) 68 5 9 5,198 140 2.7 40 0.8\nPhalombe (PE) 9 - - 87 - - - Thyolo (TO) 111 19 7 4,390 118 2.7 35 0.8\nZomba (ZA) 339 161 19 20,813 2,267 10.9 192 0.9\n_Subtotal South_ 3,399 1,495 427 246,523 60,502 24.5 10,741 4.4\n**Total Malawi**\n_<u>Total Malawi</u>_ <u>35,140</u> <u>24,474</u> <u>9,676</u> <u>1,348,763</u> <u>644,236</u> <u>47.8</u> <u>137,064</u> <u>10.2</u>\n_Source:_ Own computation from the National Geographical Estates Database.\n\n\n18\n\nMention: \"National Geographical Estates Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Land area and share statistics are computed from this registry.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007162:3:0", "occurrence_id": "da7a1eb3b6880fedef0f1f1e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: - 2 \n\n**I.** **INTRODUCTION**\n\n\nRural access is among the most important infrastructure elements to stimulate economic\n\n\ngrowth in rural and remote areas. According to the Rural Access Index developed by Roberts\n\n\net al. (2006), in Africa only 34 percent of the rural populations have access to an all-season\n\n\nroad within an approximate 2-km walking distance. Many farmers and local businesses are\n\n\nstill not well connected to the road network and do not have good access to markets. Rural\n\n\naccess is particularly important from the poverty reduction point of view. The majority of the\n\n\nrural populations are living below the poverty lines, simply engaging in subsistence farming\n\n\nwith few advanced inputs used. Rural access is also critical for people’s access to social\n\n\nfacilities, such as schools and hospitals.\n\n\nThe literature is generally supportive of the impacts of rural accessibility. In the short term,\n\n\ntransport costs and travel time can be reduced by improved road conditions (e.g., Lokshin\n\n\nand Yemtsov 2005; Danida 2010). Over the longer term, agricultural productivity would\n\n\nlikely be increased (Khandker et al. 2009; Bell and Van Dillen, 2012), and firms would\n\n\nbecome more profitable with more jobs created (Mu and van de Walle, 2011). Poverty could\n\n\nthen be alleviated (Dercon et al., 2008; Khandker and Koolwal, 2011).\n\n\nRoad financing is a matter of particular concern in many developing countries. Despite the\n\n\nfact that there are significant unmet needs for road infrastructure, available resources are\n\n\nlimited. In Africa, it is estimated that about US$12.7 billion would be needed to rehabilitate,\n\n\nupgrade and maintain the road network to meet minimum connectivity standards (AICD,\n\n\n2009). Maintenance of rural roads, of which the traffic level is low at less than 200 vehicles\n\n\nper day, tends to be neglected under fiscal pressure. On the other hand, many rural roads in\n\nMention: \"Rural Access Index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"It reports the rural population without access on a continent.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007164:46:0", "occurrence_id": "20ca53d28ba8448681755684", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **<u>Table A12. Excise rates on coffee and non-alcoholic beverages</u>**\n\n\n\n\n\n\n\n\n\n\n|Coffee Non-alcoholic beverages|Col2|\n|---|---|\n|roasted coffee: 6 kn/kg|water, including mineral and sparkling water,<br>sweetened or aromatized, except fruit juices and<br>nectars: 40 kn/hl|\n|substitutes for roasted coffee that contain coffee: 6<br>kn/kg|other beverages with less than 1.2% of alcohol<br>volume, except mixtures of beer and non-alcoholic<br>beverages with more than 0.5% alcohol volume:<br>40 kn/hl|\n|non-alcoholic beverages containing coffee or<br>extract/essence/concentrate of coffee: 6 kn/kg|syrups and concentrates for preparation of non-<br>alcoholic beverages, except fruit syrups: 240 kn/hl|\n|extracts, essences and concentrates of coffee: 20<br>kn/kg|powders and granules for preparation of non-<br>alcoholic beverages: 400 kn/100 kg|\n\n\n\nInformation from the HBS that we use to calculate the excises consist of expenditures and quantities, both\ndirectly reported by households. The only item which we were not able to identify in the HBS is “nonalcoholic beverages containing coffee or extract, essence or concentrate of coffee”. For all other items we\nwere able to find the corresponding item in the HBS.\n\n\n**2.3 Other indirect taxes**\n\n\n**_Special tax on the premiums of insurance from automobile responsibility_**\n\n\nThis tax is paid on the premiums of the so-called “insurance from automobile responsibility”. The tax base\nis the value of premium, and the rate is 15 percent. In the case of casco-type insurance of road vehicles,\n\nMention: \"the HBS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Expenditures and quantities from this survey give the tax bases.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007165:7:0", "occurrence_id": "01d89c200a1bab40b98959cc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Cargo reservation schemes still exist under the UN Liner Code, but regulation has reduced the\nextent of anticompetitive practices such as price-fixing agreements and maritime conferences. <sup>1</sup> [^1: Competition laws such as the United States’ Ocean Shipping Reform Act of 1999, European regulation\nthat abolished block exemptions from shipping conferences, and the emergence of large shipowners in\nthe 1980s and 1990s weakened conferences (WTO 2010).]\nOther forms of protection, such as barriers to investment in maritime transport service, still exist\n(one example is the limitations foreign investors face when establishing local offices). Bertho,\nBochert, and Mattoo (2014) find that policy barriers reduce trade by 28–46 percent through\nhigher transport costs.\n\n\nPort efficiency has received attention in the literature as an important determinant of the level\nof and changes in transport costs. The evidence shows a strong negative link between port\nefficiency and shipping costs (Micco and Pérez 2002; Sanchez and others 2003; Clark, Dollar, and\nMicco 2004; Wilmsmeier, Hoffmann, and Sanchez 2006). However, there is no agreement on the\nmeasure of port efficiency. Consequently, the coefficient on efficiency in maritime transport cost\nestimations varies significantly across papers.\n\n\nClark, Dollar, and Micco (2004) incorporate ad hoc measures of port efficiency derived from the\nWorld Bank’s _Global Competitiveness Report_ . They construct an index ranking of port efficiency\nfrom survey responses of cargo handling firms to the statement “Port facilities and inland\nwaterways are extensive and efficient” (1 = strongly disagree, 7 = strongly agree). They find that\nfor the average country, an increase in port efficiency from the 25th percentile to the 75th\npercentile reduces costs by about 10–15 percent, everything else equal. Wilmsmeier, Hoffmann,\nand Sanchez (2006) estimate the effects of port efficiency on transport costs for 16 Latin\nAmerican countries in 2002 using data from the International Trade Database, which is\nmaintained by the United Nations. They note that port efficiency need not lead\n\nMention: \"International Trade Database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its records give the transport cost efficiency estimates for sixteen countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007174:34:1", "occurrence_id": "2748c00a51b1e5eaa0d57ac0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: , 2010). The vignette is thus designed to focus on health inputs of\nvaccination and bed nets for all children to highlight the public good characteristics of the proposed\nspending.\n\n\nPublic employment or government jobs, on the other hand, have been documented as a classic form of\nprovision of private benefits in weak institutional contexts where public bureaucracies are not\nsufficiently professionalized and rules-based (Wantchekon, 2003; Robinson and Verdier, 2003;\nChandra, 2004; Kitschelt and Wilkinson, 2007; Grindle 2012; Gibson, Hoffman and Jablonski 2015).\nThe dismantling of political patronage in government jobs is at the heart of accounts of transition from\nweak to strong institutions, such as in the history of the United States during the Progressive Era\n(Glaeser and Goldin, 2006). Furthermore, there is systematic correlation between failures in delivering\nbroad public services and the provision of patronage jobs, and clientelist forms of vote mobilization\nmore generally (Callen et al, 2014; Khemani, 2015; Andersen et al, 2015). That is, when government\njobs are provided as political patronage, the job-holders have weak incentives to exert effort and perform\nwell in delivering public services (World Bank, 2004).\n\n\n20 Such questions are contained in the Afrobarometer surveys, for example.\n\n\n33\n\nMention: \"the Afrobarometer surveys\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"They contain the service delivery questions referenced.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007175:25:0", "occurrence_id": "b342a9de0b0cf9e16e5bcee6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Source: Our own estimation based on household surveys and O*NET\n\n\nIn summary, during the last two decades, the workplace in Argentina and Uruguay has\n\n\nexperienced some changes in terms of the type of tasks which workers perform in their jobs. In\n\n\nthis sense, we can detect an increase in the relative importance of non-routine cognitive tasks and\n\n\na reduction in the intensity of manual tasks. In contrast with that seen in developed countries,\n\n\nwork intensive in routine cognitive tasks has shown signs of growth. These variations correspond\n\n\nto various explanatory factors. On one hand, the process of technological change which\n\n\nincentivizes the modification of the optimum combination of tasks within an occupation. At the\n\n\nsame time, and for the case of Uruguay, there is also still an effect of sectoral change, that is, the\n\n\nmovement of work between branches of activity, moving from those which are more intensive in\n\n\nmanual tasks to sectors with a greater emphasis of occupations intensive in cognitive tasks.\n\n\nAll changes are enhanced by the educational expansion that took place in the labor force, since\n\n\nas the workers increase their levels of qualification, they increase their chances of dedicating a\n\n\nlarger proportion of their time to the performance of cognitive tasks. In any case, it is the young\n\n\ngenerations who dominate these trends, which suggests that they have a greater capacity for\n\n\nadaptation than the older generations.\n\n\n4.4 Polarization of the labor market\n\n\nThe process of adopting new production technologies based on automation, robotics and digital\n\n\ncommunication would permit an increase in global productivity and therefore economic\n\n\nefficiency. However, that also implies a risk from the point of view of distribution. That being\n\n\nthrough the generation of a possible polarization of the labor market, since the transition from\n\n\n                              - 24\n\nMention: \"O*NET\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Combined with household surveys to estimate workplace task change.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007176:21:0", "occurrence_id": "0bc63e345e587e833c89612b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 10. Incidence of cash and in-kind benefits by deciles**\n\n\n\n3.5\n\n\n3.0\n\n\n2.5\n\n\n2.0\n\n\n1.5\n\n\n1.0\n\n\n0.5\n\n\n\n0.30\n\n\n0.25\n\n\n0.20\n\n\n0.15\n\n\n0.10\n\n\n0.05\n\n\n\n0.00\n1 2 3 4 5 6 7 8 9 10\n\n\n\n\n\n50\n\n45\n\n40\n\n35\n\n30\n\n25\n\n20\n\n15\n\n10\n\n5\n\n0\n\n\n\n\n\n3.0\n\n\n2.5\n\n\n2.0\n\n\n1.5\n\n\n1.0\n\n\n0.5\n\n\n0.0\n\n\n\n0.0\n1 2 3 4 5 6 7 8 9 10\n\n\n\n% of disposable income net of intervention (right axis)\nUSD PPP per day\n\n\n\n% of disposable income net of intervention (right axis)\nUSD PPP per day\n\n\n\n**(a) benefits** **(b) privileges**\n\n\n\n20\n\n18\n\n16\n\n14\n\n12\n\n10\n\n8\n\n6\n\n4\n\n2\n\n\n\n\n\n250\n\n\n200\n\n\n150\n\n\n100\n\n\n50\n\n\n0\n\n\n\n0\n1 2 3 4 5 6 7 8 9 10\n\n\n\n% of disposable income net of intervention (right axis)\nUSD PPP per day\n\n**(c) pensions**\n_Note._ The ratio of pensions to the disposable income net of pensions for the first decile is not calculated as related net income for this\ndecile is close to 0.\n_Source_ : own estimates based on HBS data.\n\n**3.5.** **Indirect subsidies**\n\n\nSubsidies for utilities and urban public transportation are two major types of indirect subsidies in Belarus. Utilities\nsubsidies amount to 2% of GDP, with around 1% covered by the cross-subsidization by the enterprises, and the\nother 1% coming directly from the budget (IMF, 2016). As of 2015, subsidies on utilities were available to everyone\nautomatically in the form of subsidized tariffs in the utility bill (since September 2016 direct subsidies to the utilities\nare also available). Some households and apartments, however, are not eligible for the subsidy. These are the\n\nMention: \"HBS data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Decile related net income is estimated from these data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007178:7:0", "occurrence_id": "a3e72d36d0961c2249b86d20", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: distance from the managerial frontier. <sup>6</sup> [^6: The subjective scale ranges from 1-10 and the objective from 1-5. We have simply divided the former by two and this\nmay lead to a misalignment of means. However, the downward slope would remain however we center the data.] Firms increasingly do not seem to know what they do not\n\n\nknow about the frontier technology of even running their firm -- they are, in fact, only “more or less\n\ncompetent” as Lundvall put it- and they are certainly not using full information to locate it.\n\n\n_Optimality and the role of the market_\n\n\nIt is not obvious that these deviations from rationality dictate discarding the market and the\n\n\nneoclassical framework that simplistically attempts to model it. It is important to stress that “more or\n\nless competent” does not imply “incompetent,” that “bounded rationality” does not imply\n\n\n“irrationality.” Interviews with firms suggest that however badly they score on the World Management\n\nSurvey, they do broadly adopt practices that are consistent with profit maximization, that is, they weigh\n\n\nthe costs of inputs, think of how best to combine complementary factors to produce output, etc., and\n\nthey respond to incentives set by the market. Even while questioning what can be said about the\n\n\noverall “optimality” of the resulting market outcome- a concern shared by many neoclassical\n\neconomists schooled in the theory of the second best- it is not obvious that government functionaries\n\n\n(or international bureaucrats) have so much better information or capability to replace it. There are\n\nwriters in the NIS literature who appear to broadly accept this view. Despite the scale and ubiquity of\n\n\nthe market failures, many of which Stiglitz details in his Nobel lecture, Cimoli, Dosi, Nelson and\n\nStiglitz (2009) make a plea for consensus based on a “pragmatic view of markets.” Chaminade and\n\n\nEdquist (2010) working firmly in the NIS tradition, argue that “Innovation policy – or other kinds of\n\npublic intervention – should\n\nMention: \"the World Management\n\nSurvey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Firms scoring poorly still adopt practices consistent with profit maximisation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007180:19:1", "occurrence_id": "846ae31a109c88eb69a0f7e7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  with the contextual control variables) but for a sample of the same\ncountries. The OECD-IADB and our PCA measures cover 60 countries in common. The coverage\noverlap between our measure and that of the ILO is 71 countries. Running our multivariate\ncorrelations for our measure of digital technology use by businesses in a country, we find that our\nDoing Business PCA measure of employment protection remains negative and significant (at 1\npercent) to this change in sample (see Table 6).\n\n\nHowever, as discussed in prior sections, the composite measures of employment protection are\nuseful, but blunt instruments for analysis. It is likely that different dimensions of employment\nregulation will affect firms’ decisions in different ways. This is clear in Alesina, et al., (2015) and in\nthe differing signs on the coefficients we estimate for the statutory minimum wage and employment\nprotection. There is further nuance in the effects of regulatory instruments to be found. We\nexamine this possibility by replacing our PCA composite measure of employment protection with\nthe four separate measures constructed from the Doing Business LMR data. Of the four, the\nstrongest and most consistently significantly, negatively associated with firms’ digital technology use,\nis our measure of procedural difficulty of dismissing workers ( _ldif_firing_idx_ ). This contrasts with our\nPCA measure of the financial costs of dismissal, which is positive but does not appear statistically\nsignificant in any of the specifications.\n\n\nAs with the level of the statutory minimum wage, the possible economic explanation for this\nstatistical relationship is straightforward. All other things equal, more onerous procedures on firms’\nchoices of productive inputs impede businesses’ ability to adopt and adjust to new technologies. If\nfirms are more constrained in their human resource decisions, they will find it more difficult to\nembed new technologies into their production models, to adopt the processes that the new\ntechnologies entail, and to find the complementary labor and human capital they require to reach a\nnew optimal level of operation. And although none\n\nMention: \"Doing Business LMR data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Four separate measures of employment regulation are constructed from it.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007169:29:2", "occurrence_id": "fd2c625c009636fe0e08aa0a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  36 507 1 7.10% 2.80%\n2015-1 0 524 0 0.00%\n2015-2 3 565 0 0.50% 0.00%\n_<u>Total</u>_ _<u>178</u>_ _<u>2975</u>_ _<u>12</u>_ _<u>6.00%</u>_ _<u>6.70%</u>_\n\n\nNotes: in all panels the table shows information calculated based on the FVA lab sample data set (1 <sup>st</sup>, 3 <sup>rd</sup>, 4 <sup>th</sup>, and 5 <sup>th</sup> columns) and\nthe EXIM data set (2 <sup>nd</sup> and 4 <sup>th</sup> columns). The data sets are combined based on the broad product group.\n\n\nTable 3 also shows that in the second semester of 2014 the sampling rate increased and the non\n\ncompliance rate increased, reaching its maximum value in the sample period. Imports of food of animal\n\n\norigin represent the largest share, covering approximately 93% of the total number of shipments and of the\n\n\ntotal number of shipments sampled of the combination of food of animal origin and animal feeds. Imports\n\n\n28\n\nMention: \"FVA lab sample data set\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Supplies the odd-numbered columns of the comparison table.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007169:29:2", "occurrence_id": "4eab41794d5f58fa978d0818", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  36 507 1 7.10% 2.80%\n2015-1 0 524 0 0.00%\n2015-2 3 565 0 0.50% 0.00%\n_<u>Total</u>_ _<u>178</u>_ _<u>2975</u>_ _<u>12</u>_ _<u>6.00%</u>_ _<u>6.70%</u>_\n\n\nNotes: in all panels the table shows information calculated based on the FVA lab sample data set (1 <sup>st</sup>, 3 <sup>rd</sup>, 4 <sup>th</sup>, and 5 <sup>th</sup> columns) and\nthe EXIM data set (2 <sup>nd</sup> and 4 <sup>th</sup> columns). The data sets are combined based on the broad product group.\n\n\nTable 3 also shows that in the second semester of 2014 the sampling rate increased and the non\n\ncompliance rate increased, reaching its maximum value in the sample period. Imports of food of animal\n\n\norigin represent the largest share, covering approximately 93% of the total number of shipments and of the\n\n\ntotal number of shipments sampled of the combination of food of animal origin and animal feeds. Imports\n\n\n28\n\nMention: \"EXIM data set\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Combined with the other dataset on a common basis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007185:9:0", "occurrence_id": "4e46203f173f779e924d9578", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 1: The spatial distribution of firms in Mexico (2004 and 2014)\n\n\n_Sources_ : DeLorme, aaa and denue (inegi). _Note_ : Universe of all formal firm in 2004 (top left panel) and\n\n\n2014 (top right panel). Manufacturing in 2004 (bottom left panel) and 2014 (bottom right panel).\n\n\nthe nearest available year before which we have firm data in the denue and in the Economic\nCensuses (1986, 1994, 1999, 2004, 2009 and 2014), leaving us with a road panel for the years\n1985, 1993, 1999, 2004, 2008 and 2016. <sup>6</sup> We grouped the aaa road classes into four categories,\nwhich we label ‘Multilane divided’, ‘Two lanes or Divided’, ‘Pavement’, and ‘Gravel and Earth\nRoad’.\n\nFigure 2 represents the evolution of the road network for the years 1985, 1999, 2008, and\n2016. As can be seen, major road improvements took place in the middle of the studied period,\nbetween 1999 and 2008. This is confirmed by table 13 in Appendix A, which provides road\nlength by road category for all six years during the 1985-2016 period. <sup>7</sup>\n\n\n6We use the roads information for 2016 as the aaa map for recent years before or in 2015 were not available to\nus.\n\n7Observe that official statistics provided by inegi for Mexico provide significantly greater total road lengths\n\n\n7\n\nMention: \"Economic\nCensuses\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Firm data are matched to the nearest preceding enumeration in this series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007185:9:0", "occurrence_id": "1caed312cfdefe493ccec98c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 1: The spatial distribution of firms in Mexico (2004 and 2014)\n\n\n_Sources_ : DeLorme, aaa and denue (inegi). _Note_ : Universe of all formal firm in 2004 (top left panel) and\n\n\n2014 (top right panel). Manufacturing in 2004 (bottom left panel) and 2014 (bottom right panel).\n\n\nthe nearest available year before which we have firm data in the denue and in the Economic\nCensuses (1986, 1994, 1999, 2004, 2009 and 2014), leaving us with a road panel for the years\n1985, 1993, 1999, 2004, 2008 and 2016. <sup>6</sup> We grouped the aaa road classes into four categories,\nwhich we label ‘Multilane divided’, ‘Two lanes or Divided’, ‘Pavement’, and ‘Gravel and Earth\nRoad’.\n\nFigure 2 represents the evolution of the road network for the years 1985, 1999, 2008, and\n2016. As can be seen, major road improvements took place in the middle of the studied period,\nbetween 1999 and 2008. This is confirmed by table 13 in Appendix A, which provides road\nlength by road category for all six years during the 1985-2016 period. <sup>7</sup>\n\n\n6We use the roads information for 2016 as the aaa map for recent years before or in 2015 were not available to\nus.\n\n7Observe that official statistics provided by inegi for Mexico provide significantly greater total road lengths\n\n\n7\n\nMention: \"aaa map\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"The recent-year road layer standing in for missing earlier maps.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007188:10:2", "occurrence_id": "f3a31369296b77a7bab79fbc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  evidence about the most effective and efficient policy\nchoice for the interaction of federal and state level programs. Even though, the same could be argued for the interaction between PAM\nand the PROSPERA Elderly Component, the operation rules of both programs clearly establish that they cannot be combined. The lack\nof empirical evidence can partially be attributed to the limited data from federal and state-level governments on potential and actual\nbeneficiaries. SEDESOL is currently developing the Integrated Social Information System (SIS), which may allow further research to\nsoon be conducted in order to gauge what is the most effective and efficient policy choice for different government level programs\ncoordination.\n9Currently, the benefit level is not indexed to inflation. Thus, it has decreased in real terms during periods when its nominal level has\nremained constant.\n\n\n9\n\nMention: \"Integrated Social Information System (SIS)\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Under development; may soon allow research on beneficiary targeting.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007191:22:0", "occurrence_id": "332dc5177de9c27259701ee9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 4: Changes in poverty rates and migration\n\n\n\n\n\nSource: Author’s calculations from the 2001 and 2011 Census and Small Area Estimates of poverty.\nNote: This figure shows how changes in _illaka_ level migration rates is associated with changes in _illaka_ level poverty\nrates. The top panel plots the changes in Gulf-Malaysia migration rates and the bottom panel plots changes in India\nmigration rates in the horizontal axes. Each point represents an _illaka_ with the area proportional to the population\nin 2001. The line shows linear regression estimate of the relationship.\n\n\n21\n\nMention: \"2001 and 2011 Census\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Poverty rate and migration changes at sub-district level come from these rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007191:22:0", "occurrence_id": "b4cd3aa1c25f8127e504067d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 4: Changes in poverty rates and migration\n\n\n\n\n\nSource: Author’s calculations from the 2001 and 2011 Census and Small Area Estimates of poverty.\nNote: This figure shows how changes in _illaka_ level migration rates is associated with changes in _illaka_ level poverty\nrates. The top panel plots the changes in Gulf-Malaysia migration rates and the bottom panel plots changes in India\nmigration rates in the horizontal axes. Each point represents an _illaka_ with the area proportional to the population\nin 2001. The line shows linear regression estimate of the relationship.\n\n\n21\n\nMention: \"Small Area Estimates of poverty\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Modelled small-area welfare figures paired with the enumerations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007194:6:1", "occurrence_id": "2c09de50397e1914e083491d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  an up to date snapshot of\npower sector reform globally based on numerous indicators on the power sector structure and\ninstitutional arrangements in 2015. A second important source of data is the World Bank’s Private\nParticipation in Infrastructure database, <sup>7</sup> which records all developing country transactions involving the\nprivate sector going back into the 1990s, and makes it possible to understand how private sector\nparticipation has evolved over the same period. Important historic reference points are provided by\ncomparable earlier surveys conducted by Besant‐Jones (2006) and ESMAP (1999).\n\n\nThe paper will first individually examine the diffusion of each of the four dimensions of power sector\nreform: regulation, restructuring, competition and private sector participation. Along each of these four\ndimensions, a number of stages of reform are identified, which make it possible to examine diffusion\ntrends over time as well as the geographic patterns of uptake. The prevalence of power sector reform in\ndeveloping countries is compared to that in developed countries over the same time frame.\n\n\nThe final section of the paper turns to drawing out the main conclusions and insights from looking across\nexperiences on all four dimensions of reform. The ways in which different types of reforms are packaged\n\n\n1 Gassner, K., A. Popov, and N. Pushak. 2008a. “Does the private sector deliver on its promises? Evidence from a\nglobal study in water and electricity”. _Gridlines 44512_ . PPIAF, The World Bank, Washington D.C.\n2 Nagayama, H. 2010. “Impacts on investments, and transmission/distribution loss through power sector reforms”.\n_Energy Policy_, 38(7), 3453‐3467.\n3 Erdogdu, E. 2014. “Investment, security of supply and sustainability in the aftermath of three decades of power\nsector reform”. _Renewable and Sustainable Energy Reviews_, 31, 1‐8.\n4 Jama\n\nMention: \"Private\nParticipation in Infrastructure database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Records developing-country transactions involving private provision.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007210:10:1", "occurrence_id": "7e60bdce1e90871ede7d2a2a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  17.4\n\n\n\nLatin America and 17.4\n\nCaribbean 61.6% 15.8% 11.5%\n\n<u>Middle East and N. Africa</u> <u>5.3</u> <u>52.8%</u> <u>23.7%</u> <u>17.2%</u>\n**<u>\"Developing World\"</u>** <u>205.5</u> <u>92.0%</u> <u>55.5%</u> <u>45.9%</u>\n\n\n\nCaribbean\n\n\n\n**Table 2. Microfinance heterogeneity**\n\n\nCustomers by region, gender, and poverty level\nData as of December 31, 2013.\nAdapted from Table 7, Microcredit Summit Campaign, _State of the Summit Report 2015_\n\n\n8 For a recent survey on microfinance in India (arguing that there’s “still much to do”), see Sinha\net al (2017). For an inquiry into SHGs in South India (and further references), see <mark>Guérin</mark> et al.\n(2010). See also www.nabard.org.\n9 Since some SHGs are accounted for independently in the Microcredit Summit data, the\nNABARD numbers net out the overlapping institutions. The SHG fraction of the whole is thus\nslightly higher (by about 3 percent) when the other SHGs are also included (Microcredit Summit\n2015).\n\n\n9\n\nMention: \"Microcredit Summit data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Some groups counted here are netted out of the national figures.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007219:27:0", "occurrence_id": "adcc2691f2fcf7636609d8d4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: de Hoyos, R., V. A. García-Moreno, and H. A. Patrinos (2017). The impact of an accountability\n\n\nintervention with diagnostic feedback: Evidence from Mexico. _Economics_ _of_ _Education_\n\n_Review_ _58_, 123–140.\n\n\nde Hoyos, R., P. A. Holland, and S. Troiano (2015). Understanding the trends in learning\n\n\noutcomes in Argentina, 2000 to 2012. (Policy Research Working Paper No. 7518). The\n\nWorld Bank. Washington.\n\n\nDiNIECE (2013). Redefiniciones normativas y desafíos de la educación secundaria en\n\nArgentina. Acuerdos federales en un sistema descentralizado. La educación en debate.\n\nBuenos Aires, Argentina: Dirección Nacional de Información y Evaluación de la Calidad\n\nEducativa (DiNIECE).\n\n\nDiNIEE (2015). Anuario Estadístico 2015. Buenos Aires, Argentina: Dirección Nacional de\n\n\nInformación de la Calidad Educativa (DiNIECE).\n\n\nDuflo, E., J. Berry, S. Mukerji, and M. Shotland (2015). A wide angle view of learning:\n\nEvaluation of the CCE and LEP programmes in Haryana, India. (Impact Evaluation Report\n\nNo. 22). International Initiative for Impact Evaluation (3ie). New Delhi, India.\n\n\nDuro, E., M. Scasso, S. Bonelli, A. Hoszowski, R. Cortés, C. Giacometti, and V. Volman\n\n\n(2016). Operativo Nacional de Evaluación (ONE) 2013: Diagnóstico y consideraciones\n\nmetodológicas necesarias para el análisis y difusión de sus resultados. Ciudad Autónoma\n\nde Buenos Aires, Argentina: Secret\n\nMention: \"Anuario Estadístico 2015\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The national education statistics yearbook behind the indicators.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007220:13:0", "occurrence_id": "a6ead6df35a06a828fe6ef29", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The last element of information we use in our analysis is a measure of the intensity with\n\n\nwhich each industry uses the internet. The internet is likely to be more important for some\n\n\nindustries than others due to differences in their products, production processes and transactions\n\n\nprocesses. We consider three measures, each with its own limitations, but each providing some\n\n\ninformation on the intensity with which industries use the internet. Our first measure is the share\n\n\nof post and telecommunications expenses in total intermediate expenses at the 2-digit industry\n\nlevel for China. This measure is designated as _China share of post and telecommunications_, and\n\n\nbased on an input-output table for China in 1998, obtained from the World Input-Output Database.\n\n\nOur second measure is the share of information and communications technology (ICT) capital\n\n\nservices in total capital (the sum of ICT and non-ICT capital) services at the 2-digit industry level\n\n\nfor the United States in 1998. This measure is designated as _US share of ICT capital services_, and\n\n\nbased on U.S. data in the KLEMS database. This measure of sectoral ICT intensity follows most\n\nprevious studies on ICT and productivity, starting with Stiroh (2002). <sup>16</sup> [^16: In contrast to Stiroh (2002) we do not exclude IT-producing industries from our analysis―they naturally exhibit some of the\nhighest shares of ICT capital services.] By focusing on the U.S.\n\n\nthis measure can be viewed as providing information on the technological characteristics of the\n\n\nindustry, and its use alleviates concerns that may arise with the China share of post and\n\n\ntelecommunications being endogenous to firm performance in China. A third measure, which we\n\n\nwill only use in conjunction with the _number of .CN domains_ as a measure of internet penetration,\n\n\nis the share of firms with a website at the 3-digit industry level in Vietnam in 2007. This measure\n\n\nis designated as _Vietnam share of firms with websites_, and based on the 2007 Vietnam Enterprise\n\nCensus. <sup>17</sup> Vietnam’s production\n\nMention: \"World Input-Output Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its table for one country in that year gives the first measure.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007220:13:0", "occurrence_id": "2e4cd8976f68263cb862de38", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The last element of information we use in our analysis is a measure of the intensity with\n\n\nwhich each industry uses the internet. The internet is likely to be more important for some\n\n\nindustries than others due to differences in their products, production processes and transactions\n\n\nprocesses. We consider three measures, each with its own limitations, but each providing some\n\n\ninformation on the intensity with which industries use the internet. Our first measure is the share\n\n\nof post and telecommunications expenses in total intermediate expenses at the 2-digit industry\n\nlevel for China. This measure is designated as _China share of post and telecommunications_, and\n\n\nbased on an input-output table for China in 1998, obtained from the World Input-Output Database.\n\n\nOur second measure is the share of information and communications technology (ICT) capital\n\n\nservices in total capital (the sum of ICT and non-ICT capital) services at the 2-digit industry level\n\n\nfor the United States in 1998. This measure is designated as _US share of ICT capital services_, and\n\n\nbased on U.S. data in the KLEMS database. This measure of sectoral ICT intensity follows most\n\nprevious studies on ICT and productivity, starting with Stiroh (2002). <sup>16</sup> [^16: In contrast to Stiroh (2002) we do not exclude IT-producing industries from our analysis―they naturally exhibit some of the\nhighest shares of ICT capital services.] By focusing on the U.S.\n\n\nthis measure can be viewed as providing information on the technological characteristics of the\n\n\nindustry, and its use alleviates concerns that may arise with the China share of post and\n\n\ntelecommunications being endogenous to firm performance in China. A third measure, which we\n\n\nwill only use in conjunction with the _number of .CN domains_ as a measure of internet penetration,\n\n\nis the share of firms with a website at the 3-digit industry level in Vietnam in 2007. This measure\n\n\nis designated as _Vietnam share of firms with websites_, and based on the 2007 Vietnam Enterprise\n\nCensus. <sup>17</sup> Vietnam’s production\n\nMention: \"KLEMS database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its national data give sectoral intensity of capital services.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007220:13:0", "occurrence_id": "b371dc961acd73c6b8422ae8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The last element of information we use in our analysis is a measure of the intensity with\n\n\nwhich each industry uses the internet. The internet is likely to be more important for some\n\n\nindustries than others due to differences in their products, production processes and transactions\n\n\nprocesses. We consider three measures, each with its own limitations, but each providing some\n\n\ninformation on the intensity with which industries use the internet. Our first measure is the share\n\n\nof post and telecommunications expenses in total intermediate expenses at the 2-digit industry\n\nlevel for China. This measure is designated as _China share of post and telecommunications_, and\n\n\nbased on an input-output table for China in 1998, obtained from the World Input-Output Database.\n\n\nOur second measure is the share of information and communications technology (ICT) capital\n\n\nservices in total capital (the sum of ICT and non-ICT capital) services at the 2-digit industry level\n\n\nfor the United States in 1998. This measure is designated as _US share of ICT capital services_, and\n\n\nbased on U.S. data in the KLEMS database. This measure of sectoral ICT intensity follows most\n\nprevious studies on ICT and productivity, starting with Stiroh (2002). <sup>16</sup> [^16: In contrast to Stiroh (2002) we do not exclude IT-producing industries from our analysis―they naturally exhibit some of the\nhighest shares of ICT capital services.] By focusing on the U.S.\n\n\nthis measure can be viewed as providing information on the technological characteristics of the\n\n\nindustry, and its use alleviates concerns that may arise with the China share of post and\n\n\ntelecommunications being endogenous to firm performance in China. A third measure, which we\n\n\nwill only use in conjunction with the _number of .CN domains_ as a measure of internet penetration,\n\n\nis the share of firms with a website at the 3-digit industry level in Vietnam in 2007. This measure\n\n\nis designated as _Vietnam share of firms with websites_, and based on the 2007 Vietnam Enterprise\n\nCensus. <sup>17</sup> Vietnam’s production\n\nMention: \"2007 Vietnam Enterprise\n\nCensus\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The share of firms with websites comes from this enumeration.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007190:41:1", "occurrence_id": "f517dad7f9e13ffb356df13b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  (see section A.3 in Annex A for a discussion of the use of satellite im\n\nagery). Oil fields were not linked to sites in the VIIRS output where no oil stack infrastructure\n\n\nwas observed in daytime imagery, even if the site’s coordinates were within the boundary (see\n\n\nAnnex A section A.3 for discussion of infrastructure).\n\n\n8Four observations for the whole country of Syria were added to the final data for robustness check.\n[9Data were obtained from http://www.energy-redefined.com/](http://www.energy-redefined.com/)\n\n\n40\n\nMention: \"VIIRS output\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Night-time imagery used to locate flaring sites, filtered against daytime checks.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007204:14:1", "occurrence_id": "f2017280719bf600f0b6646c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: br>-- <br>22.9 <br>8.37 <br>-- <br>3.25 <br>1.90 <br>0.03 <br>7.6 <br>2.2 <br>0.6 <br>5.9 <br>0.4 <br>0.4|18.7% <br>18.5% <br>14.4% <br>7.0% <br>4.1% <br>1.6% <br>-- <br>-- <br>-- <br>-- <br>-- <br>-- <br>12.4% <br>4.6% <br>-- <br>1.8% <br>1.0% <br>0.0% <br>4.1% <br>1.2% <br>0.3% <br>3.2% <br>0.2% <br>0.2%|<br> <br> <br> <br>Yes <br>No <br> <br> <br> <br> <br> <br> <br> <br>Yes <br> <br>Yes <br>No <br>No <br> <br>No <br>No <br>No <br>Yes <br>|\n\n\n_Source_ : Republic of Zambia Annual 2015 Economic Report (2016) and Republic of Zambia 2015 Budget Address, both from the\nMinistry of Finance.\n_Note:_ Revenue collections included may not be fully allocated within the LCMS (2015) for various reasons. See sec. 3 for detail on\nthe allocative methods and assumptions.\n\nAt approximately 14 percent of GDP, Zambia’\n\nMention: \"Republic of Zambia Annual 2015 Economic Report\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Fiscal allocations and collections are tabulated in this report.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007204:14:1", "occurrence_id": "bebe23dd5c4629d0e588728d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: br>-- <br>22.9 <br>8.37 <br>-- <br>3.25 <br>1.90 <br>0.03 <br>7.6 <br>2.2 <br>0.6 <br>5.9 <br>0.4 <br>0.4|18.7% <br>18.5% <br>14.4% <br>7.0% <br>4.1% <br>1.6% <br>-- <br>-- <br>-- <br>-- <br>-- <br>-- <br>12.4% <br>4.6% <br>-- <br>1.8% <br>1.0% <br>0.0% <br>4.1% <br>1.2% <br>0.3% <br>3.2% <br>0.2% <br>0.2%|<br> <br> <br> <br>Yes <br>No <br> <br> <br> <br> <br> <br> <br> <br>Yes <br> <br>Yes <br>No <br>No <br> <br>No <br>No <br>No <br>Yes <br>|\n\n\n_Source_ : Republic of Zambia Annual 2015 Economic Report (2016) and Republic of Zambia 2015 Budget Address, both from the\nMinistry of Finance.\n_Note:_ Revenue collections included may not be fully allocated within the LCMS (2015) for various reasons. See sec. 3 for detail on\nthe allocative methods and assumptions.\n\nAt approximately 14 percent of GDP, Zambia’\n\nMention: \"LCMS (2015)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Collections may not be fully allocated within it; assumptions are documented.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007221:7:1", "occurrence_id": "d98358857c2f2090ba5a1ec4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  and equity.\nThis rating consists of five sub-components: gender equality, equity of public resource use,\nbuilding human resources, social protection and labor and environmental sustainability.\nCountries are rated each year on each of these five dimensions, on a one to six scale. The\nresulting cluster score is the simple arithmetic mean of the five components.\n\n\n6 The CPIA is arguably the most appropriate policy measure available at the Bank, because its\ncontent reflects the views of World Bank management and staff regarding what policies are most\nconducive to poverty reduction and the effective use of aid resources. Second, it is the most relevant\navailable cross-country indicator of the policies World Bank country teams are attempting to\nimprove when they design policy loans. Third, CPIA ratings have also been used in empirical\nanalyses in academic literature outside the World Bank (see, e.g., Collier and Hoeffler, 2004).\n\n\nNevertheless, we also searched for alternative measures of social policy quality (not social\ndevelopment outcomes), but failed to find useful indicators with the same level of granularity and\nconsistency and availability of data. Stockholm University’s Social Policy Indicators (SPIN)\nproject gathers time-series data of social policy legislation. Unfortunately, that data set is limited\nto (mostly) high-income countries. Likewise, the OECD collects a wide variety of social indicators,\nbut again mostly for its member states.\n\n\n6\n\nMention: \"Social Policy Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"A time-series compilation of social policy legislation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007232:25:1", "occurrence_id": "98ed5e59203316fe2a7673b1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  gender, age group, minority group membership, and other\nindicators of diversity\n\nG4‐LA13 Ratio of basic salary and remuneration of women to men by employee category, by\nsignificant locations of operation\n\n\n\n‐0.132 0.236 **0.690*** 0.000**\n\n\n0.020 0.854 **0.296*** **0.057**\n\n\n0.055 0.627 **0.269*** **0.096**\n\n\n0.026 0.813 0.246 0.122\n\n\n\n<mark>G4‐LA14</mark> <mark>Percentage of new suppliers that were screened using labor practices criteria</mark> <mark>0.011</mark> <mark>0.922</mark> **<mark>0.703*** 0.000</mark>**\nG4‐LA15 Significant actual and potential negative impacts for labor practices in the supply chain 0.033 0.776 **0.533*** 0.005**\nand actions taken\n\n\n\n**G4‐LA16 Number of grievances about labor practices filed, addressed, and resolved through**\n\n**<u>formal grievance mechanisms</u>**\n\n\n\n**0.229**** **0.034** ‐0.016 0.919\n\n\n\nSource: Own calculations. p*<0.1, p**<0.05, p***<0.01. Note: Variables refer to 2015 GRI data (see section 2.2).\n\nAll regressions control for country‐sector fixed effects. Standard errors are robust to heteroscedasticity.\n\n\nFirms also tend to disclose little information on social and human rights issues (Figure 6). Less than one in\nfour MNEs (22.6%) discloses information on\n\nMention: \"2015 GRI data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The regression variables are taken from this edition of the disclosure dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007242:8:1", "occurrence_id": "e82d0f3af2e2559a93f22951", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  were defined as those that achieved a\ndirect impact on one of the four functional objectives proposed by Andrews et al. A result was\ncategorized as ‘form’ if it does not provide evidence of a direct contribution to one of the functional\nindicators (see Figure 3). Within the 77 PFM‐specific indicators that we identified, we found a total of\n51 ‘form’ results and 26 ‘function’ results. In short, almost 70% of what projects measure or aim for is\nchange in terms of whether institutions _look_ like their functioning counterparts (i.e., have the requisite\nstructures, policies, systems and laws in place). We also find that World Bank PFM projects had\nconsiderably more success in achieving these ‘form’ results: 44 of our 77 project indicators were\nclassed as met. Of these 44 successes, only 10 of them (23%) were classed as functional. So while 67%\nof what was aimed at in terms of change in ‘form’ was met, only 38% of our ‘functional’ aims were\nmet. It appears that demonstrable improvements in actual performance are far harder to achieve than\nchanges that are primarily regulative, procedural or systems‐oriented. A recent mid‐term review for a\nBank‐supported PFM project observes how this divergence plays out in practice:\n\n\nTechnical solutions (capacity building, procurement of hardware and software) seem to be\nprogressing very well but attempts to improve the control environment are painfully slow –\nbank reconciliation, commitment control, cash management, unauthorized access to IFMIS\n(World Bank 2015).\n\n\nTaken as a whole, this distinction means that _only 23% of our projects had a demonstrable impact on_\n_underlying performance_ . This is in marked contrast to the initial 57% rate of success with which we\nbegan.\n\n\n12 Most of the projects captured both indicators and related results in a table at the start of the project’s\nImplementation Completion Report (ICR)\n\nMention: \"IFMIS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Control weaknesses in this financial management system are catalogued.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007247:15:0", "occurrence_id": "b2e8052ba796018ca6edcef5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **c.** **Married Couple with Second Earner Moving into Employment**\n\n\nA second set of scenarios in the TBM considers earning incentives for second earners in the\nhousehold when the first is already working. Take a household with 1 child where the principal\nearner income is set at the MW: in the model, the family net income schedule shifts according to\nthe wage of the second earner, who, in 24 percent of the cases in Poland, tends to be the mother. <sup>15</sup>\n\n\n**Table 2: Gender of highest earning member by family type (for families with two earners)**\n\n\n\n\n\n\n\n|Sex|Overall (%)|Married, with 1 child<br>(%)|Married, with 2 children<br>(%)|\n|---|---|---|---|\n|Male|74|72|75|\n|Female|26|28|25|\n\n\n_Source:_ Authors’ calculations with HBS 2014 data.\n\n\n\n\n\nThis allows representing the trade-offs faced by families when deciding if the second earner should\nstay home looking after the children or enter the labor force. Under the current set of taxes and\nbenefits, Figure 7 shows that for the second earner, moving into work leads to only minimal\nwelfare gains for the household, even before accounting for childcare and other costs. In the most\nextreme cases, when the second earner makes between 21 percent and 46 percent of the AW (which\nis equivalent to a full-time MW job), the household will be actually worse off financially if the\nmother works rather than if she stays home. <sup>16</sup> If the second earner moves from inactivity to MW\nemployment, there is a moderate surplus in family income (PLN 4,800 per year), yet this does not\naccount for the fact that it is likely that additional\n\nMention: \"HBS 2014 data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Gender shares of an activity are calculated from this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007250:9:0", "occurrence_id": "321cef3347df0f9d906a66af", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 1: Mexico’s elevations\n\n\n(a) _Raster_ _picture_ _of_ _Mexico’s_ _elevations_\n\n\n(b) _Elevations_ _translated_ _into_ _polygons_ _(municipalities)_\n\n\nSource: Iacovone et al. (2016) with Geographical data from INEGI.\nNote: The authors used ArcMap to translate elevations into averages for\nmunicipalities.\n\n##### **4 Data and Descriptive Statistics**\n\n\nThe data used were obtained from microdata of Mexico’s 2009 and 2014 Economic Censuses, INEGI. Out of\n\n\nthe 4,230,745 establishments included in the 2014 Census, we were able to match almost 2,159,804 with the\n\n\nprevious census, that is, around 50% of the total universe in 2013. The main reason behind this match rate\n\n\nis the dynamics of firm demography. The module regarding Science, Technology use and innovation, from\n\n\nwhich we construct the data regarding ICT use is not applied to microenterprises, which account for 95%\n\n\n8\n\nMention: \"INEGI\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Geographical data later translated into municipal elevation averages.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007253:21:0", "occurrence_id": "49f023798a25b5946b2286c6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Overall, this micro-level evidence of how one large shock to security affects firm behavior\n\n\nmeasureable in CDR data motivates our shift to a large panel data analysis.\n\n###### **4.2 District-Month Panel Results**\n\n\nNext, we exploit our panel dataset of CDR measures and GTD killings to explore the gen\n\neralized relationship between local changes in district-level insecurity on firm and employee\n\n\npresence. As our main dependent variables, we construct aggregate totals of the number\n\n\nof active corporate line firms and total of individual corporate subscribers in each district\n\nmonth covered by our CDR data. We convert these into logged values to minimize the\n\n\nimpact of extreme outliers and ease interpretation of coefficients in percentage terms. <sup>23</sup> To\n\n\nconstruct our main independent variable, we first calculate the total number of killings in\n\n\na district-month and select the top 1% of this distribution (23 or more killings). We then\n\n\ncreate an indicator variable, Major Violent Event, that equals 1 if 23 or more killings took\n\n\nplace in that month in that district; we count 111 such events distributed across 54 districts\n\n\nacross the country and appearing in 41 of our 45 months of data.\n\n\nWe begin by estimating the relationship between the count of firms present in a district\n\n\nand recent killings using the following semi-log model:\n\n\n_ln_ ( _Ydt_ ) = _β_ 1 _MajorV iolentEventdt−_ 1 + _θd_ + _δt_ + _γd ∗_ _t_ + _µd ∗_ _t_ <sup>2</sup> + _ϵdt_ (1)\n\n\nwhere _ln_ ( _Ydt_ ) is the log of total active firms in district _d_ and month _t_, 1 _MajorV iolentEventdt−_ 1\n\n\nis the indicator variable for 23\n\nMention: \"CDR data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Incident counts measuring the effect of a security shock on firms.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007253:21:0", "occurrence_id": "a9e6c8d97af4d51f58cf1dcd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Overall, this micro-level evidence of how one large shock to security affects firm behavior\n\n\nmeasureable in CDR data motivates our shift to a large panel data analysis.\n\n###### **4.2 District-Month Panel Results**\n\n\nNext, we exploit our panel dataset of CDR measures and GTD killings to explore the gen\n\neralized relationship between local changes in district-level insecurity on firm and employee\n\n\npresence. As our main dependent variables, we construct aggregate totals of the number\n\n\nof active corporate line firms and total of individual corporate subscribers in each district\n\nmonth covered by our CDR data. We convert these into logged values to minimize the\n\n\nimpact of extreme outliers and ease interpretation of coefficients in percentage terms. <sup>23</sup> To\n\n\nconstruct our main independent variable, we first calculate the total number of killings in\n\n\na district-month and select the top 1% of this distribution (23 or more killings). We then\n\n\ncreate an indicator variable, Major Violent Event, that equals 1 if 23 or more killings took\n\n\nplace in that month in that district; we count 111 such events distributed across 54 districts\n\n\nacross the country and appearing in 41 of our 45 months of data.\n\n\nWe begin by estimating the relationship between the count of firms present in a district\n\n\nand recent killings using the following semi-log model:\n\n\n_ln_ ( _Ydt_ ) = _β_ 1 _MajorV iolentEventdt−_ 1 + _θd_ + _δt_ + _γd ∗_ _t_ + _µd ∗_ _t_ <sup>2</sup> + _ϵdt_ (1)\n\n\nwhere _ln_ ( _Ydt_ ) is the log of total active firms in district _d_ and month _t_, 1 _MajorV iolentEventdt−_ 1\n\n\nis the indicator variable for 23\n\nMention: \"panel dataset of CDR measures and GTD killings\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"District-month panel combining both conflict series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007255:10:0", "occurrence_id": "c1130eb8acad2eb2b1669ff9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **<u>Figure</u>** **<u>6</u>** <u>Investment</u> <u>Returns</u> <u>for</u> <u>Equities,</u> <u>Bonds</u> <u>and</u> <u>Bill</u> <u>1900-2017</u>\n\n\nSource: Dimson, Marsh, and Staunton (2017)\n\n\ntered. But for the future, assumptions must be made. These are inherently difficult given\n\nthat actual returns can vary greatly within countries over time and between countries.\n\n\nOne key message for using the tool is to explore a range of scenarios - and to engage\n\nfully with the potential distribution of outcomes that are generated by any stochastic\n\nsimulation tool. An excellent source of returns data is provided by the London Business\n\nSchool-Credit Suisse Global Investment Return Yearbook. This provides real return data\n\nfor equities, government bonds and government Treasury bills for 23 countries over the\n\nperiod 1900-2016. The key averages are shown in Figure 6 for the whole time period and\n\ntwo sub-periods: one a period of strong equity returns and the other the period 2000-2016\n\nin which equities under-perform bonds.\n\n\nSection 4, below reviews the historic returns of the main variables and the assumptions\n\nchosen. For equities, the historic returns in India have been very high, with an equity\n\nrisk premium over government bonds of 9%. This is used to generate an initial scenario\n\nto show how the model works. But given that this is not likely to be maintained into\n\nthe future the paper also looks at much lower assumptions for the equity premium of\n\n5% and 3% respectively. It is worth noting, however, that as the LBS-Credit Suisse\n\nReport shows, an equity premium of 9% is not unprecedented over extended periods \nthough it is certainly not\n\nMention: \"London Business\n\nSchool-Credit Suisse Global Investment Return Yearbook\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Provides long-run real return data across twenty-three countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007255:10:0", "occurrence_id": "73100fc8b083ea8ed42c4474", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **<u>Figure</u>** **<u>6</u>** <u>Investment</u> <u>Returns</u> <u>for</u> <u>Equities,</u> <u>Bonds</u> <u>and</u> <u>Bill</u> <u>1900-2017</u>\n\n\nSource: Dimson, Marsh, and Staunton (2017)\n\n\ntered. But for the future, assumptions must be made. These are inherently difficult given\n\nthat actual returns can vary greatly within countries over time and between countries.\n\n\nOne key message for using the tool is to explore a range of scenarios - and to engage\n\nfully with the potential distribution of outcomes that are generated by any stochastic\n\nsimulation tool. An excellent source of returns data is provided by the London Business\n\nSchool-Credit Suisse Global Investment Return Yearbook. This provides real return data\n\nfor equities, government bonds and government Treasury bills for 23 countries over the\n\nperiod 1900-2016. The key averages are shown in Figure 6 for the whole time period and\n\ntwo sub-periods: one a period of strong equity returns and the other the period 2000-2016\n\nin which equities under-perform bonds.\n\n\nSection 4, below reviews the historic returns of the main variables and the assumptions\n\nchosen. For equities, the historic returns in India have been very high, with an equity\n\nrisk premium over government bonds of 9%. This is used to generate an initial scenario\n\nto show how the model works. But given that this is not likely to be maintained into\n\nthe future the paper also looks at much lower assumptions for the equity premium of\n\n5% and 3% respectively. It is worth noting, however, that as the LBS-Credit Suisse\n\nReport shows, an equity premium of 9% is not unprecedented over extended periods \nthough it is certainly not\n\nMention: \"real return data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Equity, bond and bill returns over the stated span.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007255:10:0", "occurrence_id": "96dfa5a92afc8dbb996b9870", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **<u>Figure</u>** **<u>6</u>** <u>Investment</u> <u>Returns</u> <u>for</u> <u>Equities,</u> <u>Bonds</u> <u>and</u> <u>Bill</u> <u>1900-2017</u>\n\n\nSource: Dimson, Marsh, and Staunton (2017)\n\n\ntered. But for the future, assumptions must be made. These are inherently difficult given\n\nthat actual returns can vary greatly within countries over time and between countries.\n\n\nOne key message for using the tool is to explore a range of scenarios - and to engage\n\nfully with the potential distribution of outcomes that are generated by any stochastic\n\nsimulation tool. An excellent source of returns data is provided by the London Business\n\nSchool-Credit Suisse Global Investment Return Yearbook. This provides real return data\n\nfor equities, government bonds and government Treasury bills for 23 countries over the\n\nperiod 1900-2016. The key averages are shown in Figure 6 for the whole time period and\n\ntwo sub-periods: one a period of strong equity returns and the other the period 2000-2016\n\nin which equities under-perform bonds.\n\n\nSection 4, below reviews the historic returns of the main variables and the assumptions\n\nchosen. For equities, the historic returns in India have been very high, with an equity\n\nrisk premium over government bonds of 9%. This is used to generate an initial scenario\n\nto show how the model works. But given that this is not likely to be maintained into\n\nthe future the paper also looks at much lower assumptions for the equity premium of\n\n5% and 3% respectively. It is worth noting, however, that as the LBS-Credit Suisse\n\nReport shows, an equity premium of 9% is not unprecedented over extended periods \nthough it is certainly not\n\nMention: \"LBS-Credit Suisse\n\nReport\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its long history shows the higher premium is not unprecedented.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007259:13:0", "occurrence_id": "c84fdeb60debcf25beedb799", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 12\n\n\nus to better capture the change in labor productivity due to structural reforms. Second, although\n\n\nthe TFP, the most used measure of productivity, accounts for the level of capital and the\n\n\nproduction technology, we are unable to capture the change in TFP stemming from structural\n\n\nreforms as the WBES database collects information only on the current stock of capital. The\n\n\ninformation on the past stock of capital required for calculating retrospective TFP is not\n\n\nreported. Recall that the standardized WBES data set used in this study has a pseudo panel\n\n\nstructure. <sup>12</sup> [^12: The lack of panel data on firms limits the possibility to use robust measures of productivity such as the TFP\nfrom Levinsohn-Petrin (Levinsohn and Petrin, 2003).] The sensitivity analysis explores the robustness of the findings using TFP and\n\n\nvalue-added per worker as alternative measures of productivity.\n\n\nLabor productivity growth is computed as the annual average growth of labor productivity over\n\n\nthe last three years. Firms are asked during the surveys to report their total annual sales and\n\nfull-time employees at the end of the previous fiscal year (�−1) and three years ago (�−3),\n\n\nrespectively. At each period, labor productivity �� is computed as the ratio of total annual\n\n\nsales over total permanent full-time employees. Following Davis and Haltiwanger (1999) and\n\n\nWorld Bank (2016), we compute labor productivity growth by dividing the change in\n\nproductivity between �−1 and �−3 ������ – ������ by the average value of initial and final\n\n\n\nlabor productivity\n\n\n\n<u>������ � ������</u>\n\n. This approach helps to reduce the influence of outliers.\n\n   \n\n\nFurther, since there were two full years between the two points in time, we calculate the annual\n\n\naverage labor productivity growth as follows:\n\n\n<u>�</u> <u>������ – ������<\n\nMention: \"WBES database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Only the current capital stock is collected, so structural TFP change is missed.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007259:13:0", "occurrence_id": "cc3158776a284807a5c9261b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 12\n\n\nus to better capture the change in labor productivity due to structural reforms. Second, although\n\n\nthe TFP, the most used measure of productivity, accounts for the level of capital and the\n\n\nproduction technology, we are unable to capture the change in TFP stemming from structural\n\n\nreforms as the WBES database collects information only on the current stock of capital. The\n\n\ninformation on the past stock of capital required for calculating retrospective TFP is not\n\n\nreported. Recall that the standardized WBES data set used in this study has a pseudo panel\n\n\nstructure. <sup>12</sup> [^12: The lack of panel data on firms limits the possibility to use robust measures of productivity such as the TFP\nfrom Levinsohn-Petrin (Levinsohn and Petrin, 2003).] The sensitivity analysis explores the robustness of the findings using TFP and\n\n\nvalue-added per worker as alternative measures of productivity.\n\n\nLabor productivity growth is computed as the annual average growth of labor productivity over\n\n\nthe last three years. Firms are asked during the surveys to report their total annual sales and\n\nfull-time employees at the end of the previous fiscal year (�−1) and three years ago (�−3),\n\n\nrespectively. At each period, labor productivity �� is computed as the ratio of total annual\n\n\nsales over total permanent full-time employees. Following Davis and Haltiwanger (1999) and\n\n\nWorld Bank (2016), we compute labor productivity growth by dividing the change in\n\nproductivity between �−1 and �−3 ������ – ������ by the average value of initial and final\n\n\n\nlabor productivity\n\n\n\n<u>������ � ������</u>\n\n. This approach helps to reduce the influence of outliers.\n\n   \n\n\nFurther, since there were two full years between the two points in time, we calculate the annual\n\n\naverage labor productivity growth as follows:\n\n\n<u>�</u> <u>������ – ������<\n\nMention: \"WBES data set\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"The standardised version has a pseudo-panel structure.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007265:24:1", "occurrence_id": "ec97615bc97193995fa037a8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \neducation performance. Among middle income countries, those in Eastern Europe and Central Asia\n\n\nperform the best. Developing countries perform worse in both primary and secondary education\n\n\nthan developed countries, and have much larger variance, especially at the primary level. While\n\n\nvariation is high, and the top-performing country in Sub-Saharan African still performs lower\n\n\nthan the lowest performing country in developed economies.\n\n\nAn important note when interpreting these results is that they might capture a series of selection\n\n\neffects. One such selection effect is driven by immigration. If a country tends to attract the most\n\n\nable students from around the world, it will have high mean scores. For instance, PISA 2009\n\n\nresults showed that approximately 15 percent of pupils from Singapore have an immigrant\n\n\nbackground, while this proportion is more than 40 percent in Qatar, which might drive mean\n\n\nscores up due to immigration rather than school quality alone. We can check this empirically\n\n\n23\n\nMention: \"PISA 2009\n\n\nresults\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its results show the share of pupils with an immigrant background in one city-state.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007267:21:0", "occurrence_id": "d12ff7fedd843d39614f6cf4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **6. Summary and Conclusions**\n\n\nIn this paper, we have constructed a spatial panel database of cyclone tracks and coastal\n\n\nlandfalls in Bangladesh, West Bengal and Odisha. For 1960-2016, our tracking data come from\n\n\nthe Bangladesh Meteorological Department (BMD) for cyclones with Bangladesh landfalls and\n\n\nthe India Meteorological Department (IMD) for cyclones with landfalls in West Bengal and\n\n\nOdisha. For storms before 1960, our data come from country reports in the global IBTrACS\n\n\ndatabase maintained by the Global Data Center for Meteorology of the US National Oceanic and\n\n\nAtmospheric Administration. We determine each storm’s landfall as its first intersection with an\n\n\nordered line of coastline points, spaced at .001 decimal degrees (approximately 100 meters). We\n\n\nidentify each cyclone’s primary coastal impact zone as the set of coastline points lying within the\n\n\ncyclone’s radial distance from its landfall point to its zone of maximum wind speed. Although we\n\n\nconfine our exercise to Bangladesh and the two Indian coastal states, our methods are general and\n\n\ncould be expanded to an arbitrarily-large set of coastal locations.\n\n\nWe use the database for an analysis of cyclone strike locations and impact zones in seven\n\n\nhistorical periods: 1877-1900, 1901-1920, 1921-1940, 1941-1960, 1961-1980, 1981-2000 and\n\n\n2001-2016. Although we believe that our results have independent interest, we highlight their\n\n\nimplications for socioeconomic analysis of household and community responses to cyclonic storm\n\n\nrisks. Episodic, short-run responses to some individual storms are well-documented. Such\n\n\nanecdotal evidence certainly provides many useful insights, but georeferenced panel data are\n\n\nrequired for systematic comparative research on coastal adaptation.\n\n\nIn this context, robust assessment of causal factors requires significant variation in cyclone\n\n\nimpacts, both across coastal areas and over time. The results of our seven-period analysis indicate\n\n\nthat such variation characterizes the coastal zone. During each 20-year period, some areas have\n\n\n20\n\nMention: \"IBTrACS\n\n\ndatabase\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Country reports on earlier storms are held in this global archive.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007269:6:0", "occurrence_id": "8aa6b5a273ce3a689d239305", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: (EAM from its acronym in Spanish.). EAM is a census of manufacturing firms in Colombia that employ\nat least 10 workers or have an output value about 73,000 USD. <sup>2</sup> [^2: 137.2 million COP in the year 2013 and exchange rate of 1,867 COP per USD, which is the average\nexchange rate for that year.] It gathers very detailed information\nabout employment levels and costs, asset levels and investment, output and intermediate materials\nuse for production. Starting in the year 2008 it also collects an ICT module where information about\ninternet use and technology can be found. In the sample, 6,999 (48.3%) of the plants have\ninformation for every year and were used as a balanced sample to check the robustness of the\nresults. I also use broadband speed data at the state level kindly provided by the Communications\nRegulation Commission for the years 2008‐2014. The sample used to estimate total factor\nproductivity was a subsample of these firms and it is described in detail in that section.\n\n\nICT use in the Colombian manufacturing sector\nIn this section I describe the results about ICT use in Colombian firms from the information collected\nin EAM special modules for the years 2008 through 2014. The pooled sample summary statistics for\nthe five‐year period are discussed. In this section the analysis is broken down by firm size and I\nreport information collected in the ICT module about the use of computers at the firm, internet\naccess, and the type and speed of internet connections.\n\n\n**Table 1 Distribution of firms by total employment size**\n\n<u>Full sample period</u>\n\n<u>Size</u> <u>Frequency</u> <u>Percent</u> <u>Cum.</u>\n\n[0,50] 91,861 72.07 72.07\n(50,200] 22,776 17.87 89.94\n(\n\nMention: \"EAM special modules\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its modules across these years record firm technology use.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007278:10:1", "occurrence_id": "9e3bad6da4cf37b84b9ff26e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: -0.54<br>-0.52<br>-0.52<br>-0.52<br>-0.45<br>-0.44<br>-0.46|-0.13<br>-0.33<br>-0.53|\n\n\n\n_Source:_ Estimates based on data of the household budget surveys 2012–15.\n\nMeanwhile, the upper-bound elasticity tends to reflect a longer-term scenario, echoing the effect the tobacco\ntax would have on younger people. After a few decades, these groups will be the majority of smokers. Thus,\nthe total average effect of the price increase is more accurately approximated by the upper-bound price elasticity.\nTo test the robustness of our calculation, we re-estimate elasticities eliminating different years from the sample.\nResults of such calculations are presented in Annex I. Annex II presents an assessment of the stability of the\nsurvey’s calculated deciles. We find that the income deciles are relatively stable in terms of demographic\ncharacteristics. In addition, we tested and confirmed that the elasticity estimations do not depend on a specific\nsurvey year, as the exclusion of different survey rounds from the estimation yields similar results.\n\nThe literature presents different views on the most accurate way to translate the reduction in tobacco\nconsumption into the reduction of medical expenses. One argument is that tobacco price increases reduce\nconsumption on average, but price increase does not necessarily lead to adult tobacco cessation in a similar\nway,[10] as some people may quit, while others may just reduce the number of cigarettes smoked per day. In\naddition, there are two two important points to underline. First, it has been shown that only quitting – and not\nsmoking less -brings substantial positive effects on medical well-being[2,38-40]. Second, tobacco cessation has\na stronger health effect for younger groups of the population[41]. This\n\nMention: \"the household budget surveys 2012–15\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Elasticity estimates are computed across these rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007286:40:0", "occurrence_id": "8fe011ff7f15d1063a929605", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Appendix Figure 9. Shapley decomposition of per capita growth in SSA**\n\n\nSource: Author’s estimation based on data described in Section 2.\n\n\n39\n\nMention: \"data described in Section 2\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The section's construction feeds the regional growth decomposition.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007295:40:1", "occurrence_id": "721de7f215070b059aad72aa", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: /u> _<u><mark>α</mark></u>_ <u><mark>,</mark></u> _<u><mark>β</mark></u>_ <u><mark>=2)</mark></u> <u><mark>0.150</mark></u> <u><mark>0.196</mark></u>\nNote that ��0 for all multidimensional poverty measures reported in the Table.\nSource: Author’s calculations from Third National Family Health Survey data.\n\nMention: \"Third National Family Health Survey data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Multidimensional poverty measures are calculated from this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007299:19:1", "occurrence_id": "f632f163c369816be55ec76d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  a channel is diluted by a factor of _N_\n\n\nonce landownership passes from a father to his _N_ sons, which strikes us as extreme since,\n\n\nat a minimum, brothers are likely to cooperate on land-related matters and in many cases\n\n\neven designate one of their own as operator of their joint holdings. Figure 3 illustrates the\n\n\ndistribution of family landholdings on all head outlets and all tail outlets pooled together.\n\n\nAverage family landownership on head outlets is 8.4 acres, compared to 8.9 acres on tail\n\n\noutlets; the 98th percentiles are 45.8 and 47.5 acres, respectively. Thus, families with large\n\n\n22In some cases, outlets had actually been closed years ago but the outlet lists provided to us by the\nirrigation department and used for sampling had not been updated.\n\n\n17\n\nMention: \"the\nirrigation department\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Its outlet lists used for sampling were not updated, so closed outlets remained.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007304:10:1", "occurrence_id": "bb698560c56306e7a7a9f77e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  India’s low-income states had a much higher share of\ninternal out-migrants than the richer states, such as Maharashtra, Gujarat and Tamil Nadu. This\ntrend of relatively less developed states such as Bihar and Uttar Pradesh having high net outmigration is shared by the CMM, which was developed to capture labor migration. This metric\nalso shows that the relatively more developed states, including Goa, Delhi, Maharashtra, Gujarat,\nTamil Nadu, Kerala and Karnataka, have the highest net in-migration.\n\n\n\n**Figure 8:** “ **Inter-state” migrants and per capita**\n**income, by state, 2011-12**\n\n\n\n**Figure 9: “Intra-state” migrants and per**\n**capita income, by state, 2011-12**\n\n\n\nSource: Authors’ calculation based on the IHDS Source: Authors’ calculation based on the IHDS\n\n\n**Figure 10: Distribution of migrant flows across**\n**states, 2011-12**\n\n\n9\n\nMention: \"the IHDS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"State-level income and migrant distributions are computed from this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007305:4:1", "occurrence_id": "430cdefa64892194cc7882a6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  **DATA**\n\n\n**National Water and Sanitation Survey (NWSS) 2015**\n\n\nData on access to WASH variables come from the 2015 NWSS household survey. The household\n\n\nsurvey was conducted by the Federal Ministry of Water Resources, which interviewed 201,842\n\nhouseholds across 36 states in Nigeria (Figure 1). <sup>5</sup> The survey asked questions relating to\n\n\n4 <mark>All surveyed households and water service points were georeferenced in the surveys to provide latitude and longitude coordinates.</mark>\n<mark>Water schemes were also georeferenced using their centroid location, although it should be noted that in many cases these schemes</mark>\n<mark>occupy a significant area and so the use of a single central location is a potentially crude approximation of their true spatial extent</mark>\n<mark>and coverage.</mark>\n\n5 See a more detailed description at Andres et al. (2017).\n\n\n3\n\nMention: \"National Water and Sanitation Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The named round supplies the access variables.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007305:4:1", "occurrence_id": "d8caa0de19864888e11ea4e8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  **DATA**\n\n\n**National Water and Sanitation Survey (NWSS) 2015**\n\n\nData on access to WASH variables come from the 2015 NWSS household survey. The household\n\n\nsurvey was conducted by the Federal Ministry of Water Resources, which interviewed 201,842\n\nhouseholds across 36 states in Nigeria (Figure 1). <sup>5</sup> The survey asked questions relating to\n\n\n4 <mark>All surveyed households and water service points were georeferenced in the surveys to provide latitude and longitude coordinates.</mark>\n<mark>Water schemes were also georeferenced using their centroid location, although it should be noted that in many cases these schemes</mark>\n<mark>occupy a significant area and so the use of a single central location is a potentially crude approximation of their true spatial extent</mark>\n<mark>and coverage.</mark>\n\n5 See a more detailed description at Andres et al. (2017).\n\n\n3\n\nMention: \"2015 NWSS household survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"A household survey run by the water ministry.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007306:17:0", "occurrence_id": "a7338696eaad9331adce9898", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <mark>education or to take on a job. Families ended up prioritizing men’s work.</mark> <sup>33</sup> [^33: Contrary to that, Chioda (2016) shows that highly educated women tend to see work more as a\ncareer than merely a source of income.] <mark>This</mark>\n<mark>arrangement is reinforced not only by men’s socially expected role as providers, but</mark>\n<mark>also with participants’ expectation that men will earn higher wages.</mark> <sup>34</sup> [^34: The World Values Survey 2014 shows that in Brazil 16.8 percent of interviewees agree that when\njobs are scarce, men should have more right to a job than women.] <mark>It shows that,</mark>\n<mark>even when women value having a paid job, power dynamics within households and</mark>\n<mark>negotiating the organization of domestic work is still a major barrier to equal</mark>\n<mark>participation in the work force.</mark>\n\n\n**<mark>Additionally, women with small children mentioned experiences of discrimination.</mark>**\n<mark>Women reported being asked about whether they have a partner and small children</mark>\n<mark>during job interviews and not being considered for the positions after their positive</mark>\n<mark>responses.</mark> While it remains a speculation, the level of detail to which the interviewee,\nin the following quotation, was inquired about her childcare responsibilities suggests\nthat discrimination based on gender and care responsibilities may account for some of\nyoung women’s difficulties in accessing jobs.\n\n\n_Q7: P: I was interviewed last week and... I think I lost the opportunity, the owner told me he had really_\n_liked me and all, but I lost the opportunity when\n\nMention: \"World Values Survey 2014\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its item on job priority gives the share agreeing men should earn more.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007307:45:0", "occurrence_id": "ed4a0d7996b4a9ff2487e650", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: where all days _t_ of a year and all nightlight cells _i_ in province or district _p_ are aggregated.\n\n\n**A.4** **Tsunami** **Index**\n\n\nThe final disaster damage index constructed is for the 2004 Christmas tsunami. There is\n\n\nlittle local district level damage data available, so it was decided to use the methodology\n\n\nfrom Heger (2016), whereby inundation maps are used to construct a district level damage\n\n\nindex assuming uniform damage across all flooded areas.\n\n\nTo construct an inundation map of the affected areas, a map based on MODIS satellite\n\n\npictures from Anderson _et al._ (2004) is used. In terms of the intensity of the flood, there is no\n\n\nexisting data on that, but a uniform flood intensity across all flooded areas is assumed, just\n\n\nas in Heger (2016). Using a base map from DFO, spatial algorithms were used to detect the\n\n\ndifference in color between inundated and non-inundated areas. This process started with\n\n\noverlying the base map on a regular shapefile of Indonesia, then detecting the specific color\n\n\nof inundated areas, before constructing a new shapefile where all inundated areas (areas with\n\n\nthe same color) have value 1 and all other areas have value 0.\n\n\nFinally, the weights are  - again  - defined as nightlight in the cell over total nightlight in\n\n\nthe province. Assuming a damage of 75 percent in the inundated cells, yields a final damage\n\n\nindex formula:\n\n\n_TD_ _i_ = _Wi,p,t · D,_ _i_ = 1 _, . . ., I_ (13)\n\n\nwhere _TD_ _i_ is the province weighted damage from nightlight cell _i_, _Wi,p,t_ is the same weight\n\n\n44\n\nMention: \"MODIS satellite\n\n\npictures\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its imagery builds the inundation map of the affected areas.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007308:32:0", "occurrence_id": "ce191eacb72e090a1221c8f8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Economic household typology All women Women 20‐34\n\n\nPoor Non‐ Poor Non‐\n<u>poor</u> <u>poor</u>\n\nMale earner ‐ with children & earner dep (21.0%) 41.7% 28.1% 39.8% 28.2%\n\n\nHead couple earner ‐ with children only (10.2%) 10.6% 15.0% 10.4% 14.4%\n\n\nMultiple earners ‐ with children & earner dep (7.1%) 15.0% 12.2% 15.0% 13.3%\n\n\nNo earners ‐ with senior dep (6.7%) 0.9% 4.3% 0.2% 0.3%\n\n\nTotal (45%) 68.3% 59.5% 65.3% 56.2%\n\n\nNote: Numbers in parenthesis refer to the share of each typology in the total number of households; table shows the\nfour typologies with the highest shares among total households. Sample of 89 countries for Demographic typology,\nand 71 countries for Economic Typology\nSource: WB Staff’s calculations based on GMD.\n\n\nThe previous sub‐sections examined the share of different types of households that are poor,\nclassified both according to their demographic and economic composition. On the demographic side, in\ngeneral households with children are more likely to be poor than similar households without children.\nHowever, of the households with children, the most common types of households ‐‐ those with children\nand an adult couple, with or without other adults present ‐‐ have a smaller share in poverty. On the\neconomic side, households with both adult and child dependents are disproportionately poor, particularly\nthose with no earners, a single female earner, or to a lesser extent, a single male earner. Among households\nwithout grown dependents, single female earner households are the poorest.\n\n\nThis final section considers the extent to which these results are robust when controlling for\nadditional\n\nMention: \"GMD\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Calculations across many countries on household typology rest on this dataset.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007309:8:0", "occurrence_id": "6180ad25e101eb64eec6ceae", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 4: Lorenz Curve and Inequality Coefficients by Region using NPS**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n|2008/09|2010/11|2012/13|\n|---|---|---|\n|<br>**Gini**<br>**p90/p10**<br>**Income shares**<br>|<br>**Gini**<br>**p90/p10**<br>**Income shares**<br>|<br>**Gini**<br>**p90/p10**<br>**Income shares**<br>|\n|<br> <br> <br>Low.<br>quintile<br>Top<br>quintile<br> <br> <br> <br>Low<br>quintile<br>Top<br>quintile<br> <br> <br> <br>Low<br>quintile<br>Top<br>quintile<br>|<br> <br> <br>Low.<br>quintile<br>Top<br>quintile<br> <br> <br> <br>Low<br>quintile<br>Top<br>quintile<br> <br> <br> <br>Low<br>quintile<br>Top<br>quintile<br>|<br> <br> <br>Low.<br>quintile<br>Top<br>quintile<br> <br> <br> <br>Low<br>quintile<br>Top<br>quintile<br> <br> <br> <br>Low<br>quintile<\n\nMention: \"NPS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Regional Lorenz curves and inequality coefficients across three rounds.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007311:42:1", "occurrence_id": "fe35bac9671e42c760f5f964", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 6 1.4 0.2 3.5\n\n\n\n_Without Health Insurance (%)_ 99.9 99.7 14.3 99.0\n\n\n_<u>Without Social Insurance (%)</u>_ <u>99.9</u> <u>99.6</u> <u>16.0</u> <u>99.3</u>\nSource: Analysis of 2014 Labor Force Survey\n\n\n**<u>Table A7: Number and Percentage of Jobs in the Wage Sector by Selected Characteristics</u>**\n\n**Have Written Contract** **Have Health Insurance** **Have Social Insurance** **#** **%**\nNo No No 7,261,841 38.3\nNo No Yes 11,100 0.1\nNo Yes No 28,636 0.2\nNo Yes Yes 42,950 0.2\nYes No No 1,542,468 8.1\nYes No Yes 56,727 0.3\nYes Yes No 260,187 1.4\n<u>Yes</u> <u>Yes</u> <u>Yes</u> <u>9,398,338</u> <u>49.6</u>\n\nSource: Analysis of 2014 Labor Force Survey\n\n\n41\n\nMention: \"2014 Labor Force Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Coverage and job counts in the wage sector are analysed from this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007313:30:2", "occurrence_id": "a26068885174b00735ff5c4c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  errors in parentheses. <sup>*</sup> _p_ < 0.10, <sup>**</sup> _p_ < 0.05, <sup>***</sup> _p_ < 0.01\n\n**- “Food expenditure”** includes daily and holiday expense on eating, drinks and smoking. **“Daily nonfood**\n**expenditure”** contains expense on stuffs serving for daily living such as: gas, coal, washing powder, shampoo, books.\n**“Annual nonfood expenditure”** covers the irregular spending like: clothing, shoes, bicycle tires, toys, and watches.\n**“Value of vehicle (cars or motorbikes) purchases”** is calculated at current price and only include vehicles which\nhouseholds bought in 2001 and 2000 for VHLSS of 2002 and in 2002 and 2003 of VHLSS of 2004. **“Value of**\n**household durable good purchases”** calculated as similarly as value of vehicle purchases, but exclude value of\nvehicle purchases. **“Value of vehicle parts”** is annual expenditure on spare parts of car and motorbikes in the past 12\nmonths. It is noteworthy that all the values are deflated at January 2002.\n\n\nControl variables are Household head (HH)’s ethnicity, HH age, HH gender, HH vocational education\ndummy, HH high school dummy, HH middle school dummy, HH primary school dummy, spouse vocational education\ndummy, spouse high school dummy, spouse middle school dummy, spouse primary school dummy, household size,\nand number of working age members. All of these household characteristics are reported 2002.\n\n\n29\n\nMention: \"VHLSS of 2002\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"Its purchase window is defined by the two preceding years.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007317:27:0", "occurrence_id": "a6725edcfc8474bca2e796aa", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Source:_ Authors’ calculation based on NWSS (2015) survey data.\n\n\n26\n\nMention: \"NWSS (2015) survey data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Access statistics are calculated from this round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007323:9:1", "occurrence_id": "d7f3c1ca5702ced47c8af101", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \nlivestock production are collected as part of the AgSS but it is not straightforward to include these in the\nagricultural output.\n\n##### **2.3. Manufacturing output**\n\nA census of large and medium sized manufacturing establishments is conducted every year. An\nestablishment is considered eligible for this survey if it has more than 10 employees and uses electricity.\nThe survey collects information on output, assets, operating costs and employment. The town of each\nestablishment is recorded and in some cases the zone. By matching towns to zones, zonal\nmanufacturing output can be estimated.\n\n\nThese estimates do not include manufacturing output of smaller firms. Nationally, this is a small\nproportion of manufacturing output. Soderbom (2012) compares micro‐manufacturing firms in Ethiopia\nwith larger firms included in the annual census and shows that the value added of larger manufacturing\nfirms is 8 times that of firms with fewer than 10 employees. Focusing only on the larger firms for an\nestimate of manufacturing output thus captures a large share of the manufacturing output in Ethiopia.\nHowever, it may be the case that the smaller manufacturing firms matter more for poverty reduction.\nOur regression estimation strategy allows for the share of manufacturing output produced by small\nfirms to vary across zones and to change with time, but it relies on the growth rate of manufacturing\noutput of small firms to be constant within a given zone across the full period 1996‐2011.\n\n\n8\n\nMention: \"AgSS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Livestock production is collected here but hard to include in output.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007323:9:1", "occurrence_id": "935db99849c02cabfd6747d3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \nlivestock production are collected as part of the AgSS but it is not straightforward to include these in the\nagricultural output.\n\n##### **2.3. Manufacturing output**\n\nA census of large and medium sized manufacturing establishments is conducted every year. An\nestablishment is considered eligible for this survey if it has more than 10 employees and uses electricity.\nThe survey collects information on output, assets, operating costs and employment. The town of each\nestablishment is recorded and in some cases the zone. By matching towns to zones, zonal\nmanufacturing output can be estimated.\n\n\nThese estimates do not include manufacturing output of smaller firms. Nationally, this is a small\nproportion of manufacturing output. Soderbom (2012) compares micro‐manufacturing firms in Ethiopia\nwith larger firms included in the annual census and shows that the value added of larger manufacturing\nfirms is 8 times that of firms with fewer than 10 employees. Focusing only on the larger firms for an\nestimate of manufacturing output thus captures a large share of the manufacturing output in Ethiopia.\nHowever, it may be the case that the smaller manufacturing firms matter more for poverty reduction.\nOur regression estimation strategy allows for the share of manufacturing output produced by small\nfirms to vary across zones and to change with time, but it relies on the growth rate of manufacturing\noutput of small firms to be constant within a given zone across the full period 1996‐2011.\n\n\n8\n\nMention: \"census of large and medium sized manufacturing establishments\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"An annual enumeration with an eligibility threshold by size.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007326:8:0", "occurrence_id": "1628dbc7360ccf9311d4315c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: using an index that assigned higher priority to individuals between the ages of 16 and 24, and\n\n\nwomen heads of household. The prioritization index also had a component that included variables\n\n\nrelating to household income, to give higher priority to the poorest applicants.\n\n\nFinally, the local committees of each municipality verified and confirmed the prioritized list of\n\n\nparticipants.\n\n\nThe program reached its peak in 2012, covering 27,992 beneficiaries, representing an allocation\n\n\nof 0.08 percent of GDP (Figure 3). Around 40 percent of the beneficiaries were youngsters in the\n\n\n16–24-year-old range, and more than 70 percent were females. The most active years were 2011\n\n\nand 2013, and in 2015 the program was discontinued.\n\n\nAn impact evaluation of the program showed that it was effective in attracting individuals who\n\n\nwere most in need, as well as in increasing labor force participation (in particular among the youth,\n\n\nwomen, and those relatively well educated), and readiness to start a new job (Beneke de Sanfeliu\n\n\nand Acosta 2014) in the short run. The evaluation also highlighted that program participants seem\n\n\nto be more satisfied in their communities with the public areas, many of which were improved by\n\n\nthe assets built through the program, especially among the persons who will probably make most\n\n\nuse of them, the youth and adults. Finally, the participants demonstrated increased civic\n\n\nparticipation in their community, whether in community organizations or in positions of\n\n\nresponsibility. However, the evaluation did not find any improving effects on social interactions\n\n\nin the neighborhood.\n\n\n**Figure 3: PATI: Number of beneficiaries and spending as a share of GDP 2010–2014**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n_Source:_ World Bank, ASPIRE database.\n\n\n7\n\nMention: \"ASPIRE database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Programme coverage and spending as shares of output come from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007330:35:0", "occurrence_id": "e78d9bf2227fe3c4c4a641b4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Table 2: Outcomes and data sources\n\n\n\n\n\n|Outcome<br>Family|Indicator|Level of Analy-<br>sis|Data Source|\n|---|---|---|---|\n|Individual<br>and<br>house-<br>hold security|Crime victimization<br>In-migrant or resettlement status|HH<br>HH|HH Survey<br>HH Survey|\n|Community<br>stability|Infrastructural regeneration|Community|Chief Survey|\n|Economic re-<br>vitalization|Livelihood<br>Income<br>Alimentary consumption<br>Higher order consumption|Individual<br>HH<br>HH<br>HH|HH survey<br>HH survey<br>HH survey<br>HH survey|\n|Social<br>revi-<br>talization|Community organization participation<br>Political participation|Individual<br>Individual|HH survey<br>HH survey|\n\n\nxxxiv\n\nMention: \"HH Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Household-level indicators of victimisation and resettlement status.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007330:35:0", "occurrence_id": "dcf75385461b3e282dd4d2eb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Table 2: Outcomes and data sources\n\n\n\n\n\n|Outcome<br>Family|Indicator|Level of Analy-<br>sis|Data Source|\n|---|---|---|---|\n|Individual<br>and<br>house-<br>hold security|Crime victimization<br>In-migrant or resettlement status|HH<br>HH|HH Survey<br>HH Survey|\n|Community<br>stability|Infrastructural regeneration|Community|Chief Survey|\n|Economic re-<br>vitalization|Livelihood<br>Income<br>Alimentary consumption<br>Higher order consumption|Individual<br>HH<br>HH<br>HH|HH survey<br>HH survey<br>HH survey<br>HH survey|\n|Social<br>revi-<br>talization|Community organization participation<br>Political participation|Individual<br>Individual|HH survey<br>HH survey|\n\n\nxxxiv\n\nMention: \"Chief Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Community-level indicators of stability and regeneration.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007334:4:0", "occurrence_id": "7eb60b6b857ef876142a4db6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: January 2015 to February 2016 representing the pre-downturn period; February 2016 to\nOctober 2016 representing the interim period prior to the caps; and October 2016 to September\n2017 representing the one-year period after the caps for which data are analyzed in this study. <sup>2</sup> [^2: The bill was signed in August 2016, and came into full effect in mid-September 2016.]\n\nSeveral important findings emerge from the quantitative analysis. First, the data confirm a\nstatistically significant decline in aggregate lending in the Kenyan economy, with a 3.1%\ndecline after February 2016 from an aggregate base of Ksh. 167.65 billion (US $1.64 billion)\nin the pre-February 2016 period, and a further 2.7% decline after the interest rate caps.\nCorrespondingly, there is a sizable increase in Non-Performing Loans (NPLs), with the size of\nabnormal loans seeing an increase of almost 53% after February 2016 from an aggregate base\nof Ksh. 25.67 billion (US $0.25 billion), and an additional 8% after the caps. The increase in\nNPLs is exhibited across nearly every major sector in the economy.\n\nThis precipitous rise in NPLs is reflective of both the prevailing economic and market downturn\nas well as the interest rate caps. Our microeconomic analysis finds that after the caps,\ncommercial banks started issuing loans of shorter maturity, thus likely contributing to the\nincrease in NPLs. Furthermore, according to two recent Central Bank of Kenya (CBK)\nCommercial Banks’ Credit Surveys for the second and third quarters of 2017, the upward trend\nin NPLs was expected to continue into the fourth quarter of the year and beyond due to the\npolitical uncertainty and economic downturn in the country. The CBK surveys further confirm\ncommercial banks responded to these changes by elevating their risk mitigation measures,\neffectively rationing out many small and unsecured borrowers. As a result, 54% of the\nrespondents revealed that interest rate caps had negatively affected their lending to SMEs.\n\nOur data analysis\n\nMention: \"Commercial Banks’ Credit Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its quarterly rounds report the expected rise in non-performing loans.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007334:4:0", "occurrence_id": "56275f0e2aaaa910ecc29ba0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: January 2015 to February 2016 representing the pre-downturn period; February 2016 to\nOctober 2016 representing the interim period prior to the caps; and October 2016 to September\n2017 representing the one-year period after the caps for which data are analyzed in this study. <sup>2</sup> [^2: The bill was signed in August 2016, and came into full effect in mid-September 2016.]\n\nSeveral important findings emerge from the quantitative analysis. First, the data confirm a\nstatistically significant decline in aggregate lending in the Kenyan economy, with a 3.1%\ndecline after February 2016 from an aggregate base of Ksh. 167.65 billion (US $1.64 billion)\nin the pre-February 2016 period, and a further 2.7% decline after the interest rate caps.\nCorrespondingly, there is a sizable increase in Non-Performing Loans (NPLs), with the size of\nabnormal loans seeing an increase of almost 53% after February 2016 from an aggregate base\nof Ksh. 25.67 billion (US $0.25 billion), and an additional 8% after the caps. The increase in\nNPLs is exhibited across nearly every major sector in the economy.\n\nThis precipitous rise in NPLs is reflective of both the prevailing economic and market downturn\nas well as the interest rate caps. Our microeconomic analysis finds that after the caps,\ncommercial banks started issuing loans of shorter maturity, thus likely contributing to the\nincrease in NPLs. Furthermore, according to two recent Central Bank of Kenya (CBK)\nCommercial Banks’ Credit Surveys for the second and third quarters of 2017, the upward trend\nin NPLs was expected to continue into the fourth quarter of the year and beyond due to the\npolitical uncertainty and economic downturn in the country. The CBK surveys further confirm\ncommercial banks responded to these changes by elevating their risk mitigation measures,\neffectively rationing out many small and unsecured borrowers. As a result, 54% of the\nrespondents revealed that interest rate caps had negatively affected their lending to SMEs.\n\nOur data analysis\n\nMention: \"CBK surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"They confirm banks raised risk standards in response.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007335:29:2", "occurrence_id": "81e3de642d2c0caf29c36cb8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  The scale parameter _h_ controls how lobbying e\u001bort translates into lobbying expenditure. For\n\n\nthis reason, I choose to match the average lobbying expenditure to sales from the microdata, which\n\n\nis equal to 0.08 % .\n\n\nThe common component associated to the tax bene\u001ct _φ_ is targeted to match the average e\u001bective\n\n\ntax rate of non-lobbying rms in Compustat, which is equal to 21.4 % . The parameter _ν_ controls\n\n\n42This moment is taken from Wol\u001b (2012). Using data from the Survey of Consumer Finance (SCF) he computed\nwealth distributions for a series of periods of time. For an intuition of the relationship between wealth concentration and persistence of the process refer to Moll (2014). To calibrate these parameters, I exploit the fact that the\nautocorrelation of the process depends on _ψ_ and that given _ψ,_ the stationary variance is given by 2 <sup>_<u>σ</u>_</sup> _ψ_ <sup>2</sup> <sup>.</sup>\n\n43We can think of _m_ as the fraction of potential agents that could start a private rm. In general, these agents will\nnot have the connections or the information necessary to lobby at congress, at least for the short run.\n\n44For a discussion of the main rm level characteristics of private rms versus public rms in the U.S. see Davis\net al. (2007) and Asker et al. (2011).\n\n\n28\n\nMention: \"Survey of Consumer Finance\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its data give the wealth distributions used for the moment parameter.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007338:3:1", "occurrence_id": "933070d45a9e10d6bd0b822a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  of skilled\nworkers, managers, and unskilled workers as well as supply chain relationships as a\nubiquitous effect of the war. The key conclusion of this analysis is that the damage to social\nand human capital is significant, and rebuilding that social and human capital is likely a\ngreater challenge than the bricks and mortar.\n\n\nDiscussion of the sample, methodology, and missing firms\n\n\nIt was extremely challenging to conduct a firm survey in Syria in early 2017 due to the\nongoing insecurity across the country. For this reason, the survey was conducted by phone.\nThe survey was designed to be completed in approximately 30 minutes and the questions\nfocused on changes that firms have experienced since the beginning of the crisis in 2010.\nThe questions covered the following topic areas: major problems and constraints to firms,\nsales, supplies and costs, finance, and employment. (See Annex B for the questionnaire.)\n\n\nThe survey was designed as a follow‐up to the 2009 Enterprise Survey of Syria. The 2009\nsurvey included 508 firms and the 2017 survey attempted to re‐survey all 508 firms which\nwere surveyed in 2009. Table 1 below provides a breakdown of the response outcomes for\n\n\n1 This paper and the supporting analysis was undertaken by Nabila Assaf, Kinley Salmon, and David C.\nFrancis based on a survey conducted in April and May 2017.\n\n\n2\n\nMention: \"2009 Enterprise Survey of Syria\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its five hundred firms form the frame the later survey re-interviews.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007332:3:1", "occurrence_id": "a1c42dd97543a3ef25826879", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  of tax complexity, weak or ineffective tax\nenforcement, and lack of incentives to carry out business transactions and make payments through formal\nbanking channels are among the factors affecting the level of informality. Another important factor is the\nlevel of trust taxpayers have in government. <sup>7</sup> [^7: Forthcoming: “Innovations in Tax Compliance”, Verhoeven, M, et al, World Bank Working Paper, 2018.\n8. These factors are Voice and Accountability, Political Stability and Absence of Violence/Terrorism, Government Effectiveness,\nRule of Law, Regulatory Quality, and Control of Corruption.] These factors can all be influenced by tax policies and tax\nadministration measures. Other factors likely to impact the level of the shadow economy include the state\nof the official economy, labor market participation rates, quality of public-sector services, and a “culture of\ntolerance,” but as these factors are beyond the control of tax administrations they are not discussed in this\npaper. In general, a strategy to address the factors that can help reduce the size of the shadow economy\nneeds to be comprehensive, involving all stakeholders, for e.g., the central bank, the banking sector, the\nministry of finance or economy, and the tax administration.\n\nOther governance-related factors considered include the six Worldwide Governance Indicators (WGI). <sup>8</sup> Of\nthese, two, Regulatory Quality and Control of Corruption, appear to have a measurable impact on the\nshadow economy and the level of taxation. We also explored the impact of the competitiveness factors\nanalyzed by the World Economic Forum report; of the 12 factors considered there, one, Technological\nReadiness, has significant bearing on the shadow economy and taxation.\n\nFirst, we analyzed the relationship between the ratios of tax collection to gross domestic product (GDP)\nand the ratios of the shadow economy to GDP, to confirm the presence of a negative relationship: high\nlevels of the shadow economy correlate with low levels of tax collection. This\n\nMention: \"Worldwide Governance Indicators (WGI)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Six dimensions, of which two bear on tax administration.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007349:6:1", "occurrence_id": "85f3ab9d1381499b86583805", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: u> <u>1,273</u> <u>38.1</u> <u>742</u> <u>58.3</u>\nSource: Liberia Institute of Statistics and Geo-Information Services.\n\n\n\nEconomic activity is highly concentrated in Monrovia, where more than 1.2 million people,\n\n\nor about 30 percent of the total population and 55 percent of the total urban population, live.\n\n\nThe distribution of firm location is even more skewed (figure 1). The government’s business\n\n\nregistry database indicates that more than 41,000 firms were officially registered in Liberia as\n\n\nof 2016, of which approximately 80 percent were located in Montserrado County, where the\n\n\ncapital city, Monrovia, is (table 2). Thus, formal firms are almost exclusively concentrated in\n\n\nMonrovia, but even while Liberia has been experiencing rapid urbanization, many people\n\n\nlive in rural and remote areas.\n\n\nThe skewness of firm location is higher than in other countries (figure 2). This does not mean\n\n\nthat there are few businesses in other areas; there are many informal businesses, but they\n\n\nhave not been registered. As discussed above, informality is significant in the Liberian\n\n\neconomy, particularly in rural areas.\n\nMention: \"business\n\n\nregistry database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Over forty-one thousand firms officially registered as of that date.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007362:11:0", "occurrence_id": "1ca88eec70d8567ad0885ee8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: defining as \"pivot\" a household of five members, two adults and three children (one\n\n\nunder-five years child, one child between five and nine, and one over nine child). This\n\n\nsecond approach has been retained for poverty estimation.\n\n\nTo perform sensitivity analyses, Equation (1) could be written in accordance with the\n\n\nformulation of Cutler and Katz (1992) as follows:\n\n\n##### eq   na  nc \n\n\n\n\n#####   na  nc                                                 (4)\n\n\n##### N eq   na  nc\n\n\n\nIn this case, the effect of the household size is captured by  while the effect of the\n\nhousehold composition is captured by a single discount factor  .\n\n\nThe study is based on data from the World Bank’s African Database. Only surveys over\n\n\nthe period 2007-2012 are considered, which represents 35 African countries and accounts\n\n\nfor around 75 percent of the total population of Sub-Saharan Africa. The main variables\n\n\ninclude household consumption and other socioeconomic and demographic\n\n\ncharacteristics of households.\n\n\n**4 Results**\n\n**4.1 Poverty estimates and age group comparisons**\n\n\nTable 1 shows initial and adjusted poverty rates in Africa by various age groups. Poverty\n\n\nis based on the international poverty line ($1.90 a day, 2011 PPP). With the initial\n\n\npoverty based on the per capita approach, there are some statistically significant\n\n\ndifferences between various age groups of children, as well as between children and\n\n\nadults. In fact, the poverty incidence appears to be higher for under-five children,\n\n\nfollowed by the poverty incidence for children aged 5 to 9 years. All together, all under\n\n\n18 children have a poverty incidence of 50.4 percent, which is higher than that of adults\n\n\nby 11.4 percentage points. When poverty is adjusted using the FAO/WHO equivalence\n\n\nscale, gaps between groups of children disappear\n\nMention: \"World Bank’s African Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Only surveys in the stated window are used, covering thirty-five countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007365:37:2", "occurrence_id": "93a21d2366e68498920c6c39", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 125) (0.144)\nHow to learn work process - by yourself 0.009 −0.010\n\n(0.158) (0.118)\nHow to learn work process - from all 0.073 −0.018\n\n(0.088) (0.077)\nReported to TVI during in-company training\n(1=yes) 0.021 0.022\n\n(0.039) (0.039)\nTimes visited TVI teacher during incompany training (1=yes) 0.006** 0.002\n\n(0.003) (0.002)\n<u>Observations</u> <u>837</u> <u>374</u> <u>374</u> <u>407</u> <u>365</u> <u>418</u>\n_Source: DTS Tracking Survey, 2015, World Bank. World Bank calculations._\n_Note: Standard errors in parentheses. * p<0.1; ** p<0.05; *** p<0.01_\n\n\n36\n\nMention: \"DTS Tracking Survey, 2015\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Its tracking rounds carry the reported estimates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007369:26:2", "occurrence_id": "597b2304e7a6c7670c9fb563", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ‐specific barriers to\nentry for some fields. One indication of such differences between genders would be a higher fraction of\nwomen working outside of their field of study. We use data on field‐of‐study available in the World Bank’s\nSTEP survey to check whether we observe a mismatch between field of study and occupation more often\n\n\n25\n\nMention: \"STEP survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its field-of-study module allows the mismatch check.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007372:10:0", "occurrence_id": "e3013249cf0043dffc0052e5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 70,000 ha are in Kigali city. While nationally 64% of parcels (66% of area) are used for agriculture and\n\n\n12% (8% of area) for residential purposes, in Kigali 49% (56% of land area) are used for agriculture and\n\n\n35% (17% of area) for residences. For 2013–16, sales were registered for some 85,000 properties, almost\n\nhalf in Kigali and some 20,000 in our study area. Property prices averaged US$28 per m <sup>2</sup> for residential\n\nand US$4 per m <sup>2</sup> for agricultural land nationwide and US$40 (US$ 7.5) or US$ 51 (US$ 11) per m <sup>2</sup> for\n\n\nresidential (agricultural) land in Kigali and our study area, respectively.\n\n\nTo compare potential to actual revenue from land and more effective collection of current lease fees, we\n\n\ndraw on 2015 parcel-level data for land tax revenue from the Rwanda Revenue Authority (RRA). Although\n\n\nRRA and RNRA use the same UPI to identify properties, so that in principle RRA could use the cadastral\n\n\nmap to automatically bill owners for land lease fees owed (as assessed at the time of registration), this has\n\n\nnot been done. Instead lease fee charges are still based on self-declaration. Table 4, based on using the UPI\n\n\nto link RRA and LAIS data, illustrates that only about 40% (45,000) of the 112,000 residential parcels in\n\n\nKigali were subject to land lease fees and 37,000 did so. In urban Kigali in 2015, only US$1.95 million of\n\n\nUS$6.74 million owed on residential properties was collected. Thus, in urban Kigali alone, closing the\n\ncollection gap could more than triple revenue from land use fees. <sup\n\nMention: \"2015 parcel-level data for land tax revenue\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Parcel-level revenue records used to assess collection effectiveness.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007372:10:0", "occurrence_id": "26070dae0d9d5f74fbe7f204", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 70,000 ha are in Kigali city. While nationally 64% of parcels (66% of area) are used for agriculture and\n\n\n12% (8% of area) for residential purposes, in Kigali 49% (56% of land area) are used for agriculture and\n\n\n35% (17% of area) for residences. For 2013–16, sales were registered for some 85,000 properties, almost\n\nhalf in Kigali and some 20,000 in our study area. Property prices averaged US$28 per m <sup>2</sup> for residential\n\nand US$4 per m <sup>2</sup> for agricultural land nationwide and US$40 (US$ 7.5) or US$ 51 (US$ 11) per m <sup>2</sup> for\n\n\nresidential (agricultural) land in Kigali and our study area, respectively.\n\n\nTo compare potential to actual revenue from land and more effective collection of current lease fees, we\n\n\ndraw on 2015 parcel-level data for land tax revenue from the Rwanda Revenue Authority (RRA). Although\n\n\nRRA and RNRA use the same UPI to identify properties, so that in principle RRA could use the cadastral\n\n\nmap to automatically bill owners for land lease fees owed (as assessed at the time of registration), this has\n\n\nnot been done. Instead lease fee charges are still based on self-declaration. Table 4, based on using the UPI\n\n\nto link RRA and LAIS data, illustrates that only about 40% (45,000) of the 112,000 residential parcels in\n\n\nKigali were subject to land lease fees and 37,000 did so. In urban Kigali in 2015, only US$1.95 million of\n\n\nUS$6.74 million owed on residential properties was collected. Thus, in urban Kigali alone, closing the\n\ncollection gap could more than triple revenue from land use fees. <sup\n\nMention: \"LAIS data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Linked through a common identifier to the revenue records.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007373:7:0", "occurrence_id": "8966bcc8a57c7f3091eb366e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |Figure 1: Trends in intra‐regional trade in the ECA|Col2|\n|---|---|\n|a. ECA’s share of global trade has been falling|b. Intra‐regional trade growing faster than extra‐<br>regional trade|\n|<br>25%<br>27%<br>29%<br>31%<br>33%<br>35%<br>37%<br>1996<br>2000<br>2004<br>2008<br>2012<br>ECA Trade as a Share of Total World<br>Trade (exports and imports)|<br>0<br>2000<br>4000<br>6000<br>8000<br>10000<br>1996 1998 2000 2002 2004 2006 2008 2010 2012 2014<br>Intra‐ ECA trade vs Extra ECA trade<br>(billions of dollars)<br>INTRA‐ECA<br>EXTRA‐ECA|\n\n\n\nSource: Author’s calculations based on UNCTAD and World Bank Development Indicators\n\n\nWhile there are potentially hundreds of ways countries can connect with varying implications for the\ntransfer of ideas, this paper focuses on six types of economic connections: trade, FDI, migration,\ninformation and communication technologies (ICT), air transport and portfolio financial flows. A summary\nof the data used is provided in Table 1. While other forms of connectivity may also be important for how\nknowledge transfers between countries and for economic growth, these data are the only ones available\non a global and country to country basis. Subsequent sections will drill down into additional layers of\nconnectivity (for example, land transport in ECA) as well as unique aspects of the\n\nMention: \"UNCTAD and World Bank Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Trade values within and between regions are computed from these compilations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007377:5:1", "occurrence_id": "765c05149bf1f83eb742937d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  same likelihood of making payments for security, the investing\n\n\nfirms allocate a larger share of their sales towards security arrangements. Second, we find a similar\n\n\nphenomenon with respect to firms’ access to finance. Firms with bank accounts and loan applications face\n\n\nhigher risks and uncertainty. This further lends credence to the thought that participation acts as a signal to\n\n\nexpropriating parties that see these firms as better targets having larger surpluses from firm operations. Our\n\n\nthird finding corresponds to the formal status of firms. While it is true for the Afghan case that formal firms\n\n\nare larger, both in terms of workforce and sales, they too are disproportionately affected by risks and\n\n\nuncertainty. All the above findings suggest that policy interventions in a volatile economic environment for\n\n\na country like Afghanistan are not independent of consequent risks and uncertainty. In other words, some\n\n\nbenefits of such policies are potentially seized via crime and corruption. Finally, we also find that a large\n\n\nsegment of firms headed by female managers perceive crime, corruption, and political instability as being\n\n\nserious roadblocks to operation.\n\n\nIn our final analysis, we turn our attention towards estimating the economic loss from crime-related\n\n\nevents. The World Bank enterprise survey reports firms’ loss due to such events together with firms’\n\n\n4\n\nMention: \"World Bank enterprise survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Reports firms' losses from crime-related events.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007378:15:2", "occurrence_id": "32f31a8ea9f9adf190029813", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \nYears of implementation, cont. ‐1.153 ‐1.153 ‐0.220\n(1.494) (1.301) (1.630)\nGovernment implemented, dummy ‐0.0219** ‐0.0219*** ‐0.0259**\n(0.00870) (0.00758) (0.0120)\nSample size, continuous ‐0.00227 ‐0.00227 ‐0.00167\n(0.00389) (0.00339) (0.00370)\nUrban interventions, categorical ‐0.0141 ‐0.0141* ‐0.0175\n(0.00925) (0.00806) (0.0110)\nConstant 8.906 8.906 9.821\n<u>(11.36)</u> <u>(9.895)</u> <u>(12.39)</u>\n\nObservations 136 135 104\n<u>R‐squared</u> <u>0.621</u> <u>0.573</u> <u>0.650</u>\n\n\n\n<u>(pooled effects)</u>\n\n\n\n_Source:_ World Bank GWSP‐SIEF WASH Impact Evaluation Database.\n_Note:_ Robust standard errors in parentheses. *** p<0.01, ** p<0.05, * p<0.1.\n\n\n14\n\nMention: \"SIEF WASH Impact Evaluation Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Pooled impact estimates across water and sanitation interventions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007346:13:0", "occurrence_id": "bf49922880b11df21e584b8b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: #### **4 Data**\n\nThe data sets compiled for this analysis include household characteristics and energy consumption from three waves of a panel survey, geographical data on the electricity transmission and distribution network, monthly energy prices by region, and financial data for the\nelectricity distribution utility. The combined data sets provide a comprehensive perspective\non the energy sector in Uganda.\n\n###### **4.1 Household survey data**\n\n\nThe Uganda National Panel Survey (UNPS) is a household panel that began in 2009–10.\nThere were three waves in the original panel: 2009–10, 2010–11, and 2011–12. The sample\nframe for this survey was the national household survey in 2005–06. <sup>15</sup>\n\nThe survey recorded both the total amount paid and the quantity of energy consumed\nfor each energy commodity. For kerosene, firewood, gasoline and charcoal, the quantity and\nthe amount paid were asked in two separate sections of the survey: the energy module and\nthe consumption module. This repetition provides a consistency check on the responses.\n\nDuring the period covered by the panel, there was a gradual increase in the number of\nhouseholds with electricity connections, a substantial increase in the number of households\nusing solar installations for lighting, and a small decrease in the number of households using\nkerosene for lighting.\n\nIn 2011, the dominant fuel used for lighting was kerosene (Table 1). Even a significant\nnumber of households with an electricity connection still reported using kerosene for lighting,\neither in combination with electricity or on its own. Generator ownership is uncommon, and\neven the households with a generator appear to use it more as a backup rather than as a\nprimary source of energy. By 2011, solar was a more popular form of o↵-grid generation\nthan diesel or gasoline generators.\n\n###### **4.2 Electricity coverage**\n\n\nThe decision by a household to obtain an electricity\n\nMention: \"Uganda National Panel Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"A household panel beginning in that year with three original waves.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007385:10:0", "occurrence_id": "36c71f1603bbb581dd5511ed", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: three large countries, namely lower middle‐income India; upper middle‐income Brazil; and high‐\nincome United States. The poorest 40% of the populations in these three countries enjoy very different\nincome levels; the mean income for that group in Brazil is equivalent to the ninth decile in India,\nroughly, and an analogous comparison holds between the United States and Brazil. Whereas the\nannual mean income among the poorest 40% of US citizens is $8,861 (in 2011 PPP), it is $1,819 in Brazil,\nand $664 in India – about 13 times less than in the richer country. <sup>13</sup>\n\n\nOur summary description of shared prosperity trends focuses on the 2008‐2013 period, and draws on\na World Bank data set known as the Global Shared Prosperity Database (GSPD). This data set consists\nof growth spells for the income or consumption aggregate for the bottom 40% and for the mean,\nobtained from intertemporally comparable national household surveys. A first round of data included\nspells circa 2006‐11 and was described in World Bank (2015a) and World Bank (2015b). The second\nand third rounds focused on spells circa 2007‐12 and 2008‐13 and were reported in World Bank (2016a\nand 2016b).\n\n\nThe latest available wave of the GSPD covers 83 countries and provides these two annualized growth\nrates (for the bottom 40% and the overall mean) for a five‐year period circa 2008‐13. <sup>14</sup> Though this\nsample includes fewer than half of the world’s countries, it does nevertheless cover 75% of the world’s\npopulation in 2013. The geographical coverage across regions is unbalanced: of these 83 countries, 24\nbelong to one single region, Eastern Europe and Central Asia (EECA), while Industrialized countries and\nLatin America and the Caribbean (LAC) follow in the\n\nMention: \"Global Shared Prosperity Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Growth spells for income or consumption aggregates across the period.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007385:10:0", "occurrence_id": "655d82dd94e8331ab111675e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: three large countries, namely lower middle‐income India; upper middle‐income Brazil; and high‐\nincome United States. The poorest 40% of the populations in these three countries enjoy very different\nincome levels; the mean income for that group in Brazil is equivalent to the ninth decile in India,\nroughly, and an analogous comparison holds between the United States and Brazil. Whereas the\nannual mean income among the poorest 40% of US citizens is $8,861 (in 2011 PPP), it is $1,819 in Brazil,\nand $664 in India – about 13 times less than in the richer country. <sup>13</sup>\n\n\nOur summary description of shared prosperity trends focuses on the 2008‐2013 period, and draws on\na World Bank data set known as the Global Shared Prosperity Database (GSPD). This data set consists\nof growth spells for the income or consumption aggregate for the bottom 40% and for the mean,\nobtained from intertemporally comparable national household surveys. A first round of data included\nspells circa 2006‐11 and was described in World Bank (2015a) and World Bank (2015b). The second\nand third rounds focused on spells circa 2007‐12 and 2008‐13 and were reported in World Bank (2016a\nand 2016b).\n\n\nThe latest available wave of the GSPD covers 83 countries and provides these two annualized growth\nrates (for the bottom 40% and the overall mean) for a five‐year period circa 2008‐13. <sup>14</sup> Though this\nsample includes fewer than half of the world’s countries, it does nevertheless cover 75% of the world’s\npopulation in 2013. The geographical coverage across regions is unbalanced: of these 83 countries, 24\nbelong to one single region, Eastern Europe and Central Asia (EECA), while Industrialized countries and\nLatin America and the Caribbean (LAC) follow in the\n\nMention: \"GSPD\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its latest wave covers 83 countries with two annualised growth rates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007389:23:0", "occurrence_id": "3b7d2ae2c0550f8379824787", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: first and last date of receiving the social support. In 2014: about 59% of households were first\n\n\ntime beneficiaries. The remaining 41% of households have received some form of social\n\n\nassistance in the past, where the maximum length is 313 months (i.e., about 26 years). A large\n\n\nfraction of the beneficiary households are skewed towards a year or less assistance durations,\n\n\nsuggesting that social protection programs are becoming popular and expansive recently in\n\n\nNiger. To further understand these social interventions, Figure 5b illustrates the maximum span\n\n\nthat households have been receiving social assistance across the eight Nigerien regions. Spatial\n\n\ndifferences exist. The Dosso region records the highest followed by Niamey, Diffa, Zinder and\n\n\nTillabEri.\n\n\n**Figure 5a: Distribution: length of receiving social** **Figure 5b: Understanding social protection across**\n**assistance** **regions**\n\n\n**Notes:** Source: World Bank Household Panel, 2014\nNiger. The figure counts the number of months\nbetween the first and last time HH received assistance\nIn principle, the figure does not say anything about\nwhether HH receives assistance continuously; it only\ncaptures the span.\n\n\n22\n\nMention: \"World Bank Household Panel, 2014\nNiger\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Months between first and last transfer receipt are counted from this panel.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007393:39:0", "occurrence_id": "730431da0701f1da7dd5bbf6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 3: ENLACE test scores and secondary school outcomes\n\n\n\n\n\nNotes: The graph plots local means of secondary school outcomes by ENLACE test score\npercentile in grade 6 in 2007. The solid line shows a linear fit estimated using the grouped\ndata. Panel (a) reports the probability of on-time graduation from grades 9 and 12, proxied\nby siting in the Enlace exam in those grades. Panel (b) reports Enlace test scores in grades\n9 and 12 (normalised with mean 0 and SD 1) conditional on taking the Enlace exam in 2010\nand 2013. Data: ENLACE panel.\n\n\n38\n\nMention: \"ENLACE panel\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Standardised scores conditional on taking the exam in both years.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007395:14:0", "occurrence_id": "2de7793e844d466f711efd52", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: In practice, it has proved difficult in the developing world to implement regulatory\nframeworks that meet all the good practice criteria for regulatory governance.\n\n\nEberhard (2006), who provides a summary of the challenges and problems that utility\nregulation faces in developing countries, and particularly Africa, highlights, among others,\nthe “lack of regulatory commitment: political expediency and the limits of\nindependence”. Furthermore, even when there are separate regulatory institutions with\nlegal mandates for tariff setting and other regulatory decisions, government can still exert\npressure, undermining regulatory independence. Other challenges identified are: lack of\ntransparency, participation and accountability; institutional fragility; and lack of\ncapacity/competency, which compromises regulatory substance. Eberhard (2006) suggests\nthat regulatory performance can be improved through mandatory, periodic, independent\nreviews of regulators’ performance and impact; through building or strengthening the\n“demand” for regulatory transparency and participation, by consumers, investors, etc.;\nand through capacity building, involving experienced and qualified staff.\n\n\nThe World Bank’s Africa Infrastructure Country Diagnostic assesses the infrastructure\ninstitutional frameworks for 24 countries in Africa scoring the quality of governance of the\nregulatory agencies out of 100 (The World Bank 2010). The results show that electricity\nsector regulators perform relatively well compared to those in other infrastructure\nsectors. However, in absolute terms, the overall average score for the quality of\ngovernance of electricity sector regulators is under 50. The weakest dimension of\ngovernance was autonomy (or independence of decision-making) and the strongest\ntransparency (captured in terms of publishing information) (Figure 4).\n\n\n**Figure 4: Institutional progress on regulation – Regulatory quality**\n\n\nSource: The World Bank (2010).Figure 4.3. Panel (b)\n\n\n13\n\nMention: \"Africa Infrastructure Country Diagnostic\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Scores institutional frameworks across twenty-four countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007396:23:0", "occurrence_id": "eb8b41e0d2e9922ea16bf7c4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: an integral part of geographically-targeted poverty reduction strategies. Our review shows that capacity\n\n\ndevelopment is integral to improving productivity, stronger and equitable forest and land rights strengthen\n\n\nagency, and without effective public institutions and services, the poor cannot be expected to use their\n\n\naccess to forest resources to escape poverty. Of course, communities have demands that go beyond income\n\nand the non-cash contributions of forests can be very important. <sup>26</sup> Our focus on gender analyses also points\n\n\nto the importance of understanding the implications of rural development strategies for women and men.\n\n\nWhile the role forests play in enhancing rural resilience and food security is not fully addressed in\n\n\nthis paper, social and environmental safeguards will certainly need to go hand in hand with any forest\n\n\ninterventions that seek to reduce poverty. The FAO, CIFOR, IFRI and the WBG recently launched a global\n\n\neffort to develop a standard survey instrument, the Forestry Living Standards Measurement Survey, to\n\n\nmeasure forest use and the wellbeing of forest-dependent communities (FAO _et al_ ., 2016). Combining such\n\n\nfield-based surveys with participatory gender-disaggregated information and satellite data may ensure that\n\n\nforest sustainability and poverty are simultaneously monitored.\n\n\nOver time, poverty reduction in forest areas will likely be no different than what is seen in\n\n\nagricultural areas: ‘off-forest’ jobs and migration (Hecht _et al._ 2016) will play a significant role in changing\n\n\nthe relationship between forests and people. We see this in trends in forest villages in middle-income\n\n\ncountries such as Turkey and Albania (World Bank 2016b, IOM 2016). Thus, strategies that enable\n\n\ncommunities to seek these jobs while simultaneously strengthening their ability to benefit from forest\n\n\nresources in a sustainable manner will be key.\n\n\n22\n\nMention: \"Forestry Living Standards Measurement Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"A newly launched standard instrument for forest-dependent communities.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007398:15:1", "occurrence_id": "bc0cf83a5202a2778ff75607", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  grammatical gender) is provided in our Data Construction\n\n\nAppendix.\n\n\n**3.1.3** **Identifying** **Gender** **Languages**\n\n\nFor each mother tongue in the _Ethnologue_ database, we attempt to code two variables\n\n\ncharacterizing the language’s grammatical gender structure. First, we create an indicator\n\n\n13Data from the WALS must be used with caution because they were compiled by the linguist Greville\nCorbett; as discussed above, Corbett advocates the use of a non-standard definition of grammatical gender\nthat includes systems of anaphoric pronominal agreement (Corbett 1991). This is particularly problematic\nwhen one combines the WALS with other data sources that do not classify systems of pronominal agreement\nbased on the gender of the referent as examples of grammatical gender. We address this by excluding WALS\ndata on languages that are classified as “strictly semantic” (i.e. agreement class can always be inferred from\nthe meaning of the noun) since Corbett considers pronominal agreement an example of such a system. We\nrely on other sources to classify those languages. Languages that are classified in the WALS as either lacking\na grammatical gender system of having a system that is “semantic and formal” are unambiguous.\n\n\n14\n\nMention: \"_Ethnologue_ database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Each mother tongue is coded for grammatical gender structure from this catalogue.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007398:15:1", "occurrence_id": "957181b3b5cf0411247e4dbb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  grammatical gender) is provided in our Data Construction\n\n\nAppendix.\n\n\n**3.1.3** **Identifying** **Gender** **Languages**\n\n\nFor each mother tongue in the _Ethnologue_ database, we attempt to code two variables\n\n\ncharacterizing the language’s grammatical gender structure. First, we create an indicator\n\n\n13Data from the WALS must be used with caution because they were compiled by the linguist Greville\nCorbett; as discussed above, Corbett advocates the use of a non-standard definition of grammatical gender\nthat includes systems of anaphoric pronominal agreement (Corbett 1991). This is particularly problematic\nwhen one combines the WALS with other data sources that do not classify systems of pronominal agreement\nbased on the gender of the referent as examples of grammatical gender. We address this by excluding WALS\ndata on languages that are classified as “strictly semantic” (i.e. agreement class can always be inferred from\nthe meaning of the noun) since Corbett considers pronominal agreement an example of such a system. We\nrely on other sources to classify those languages. Languages that are classified in the WALS as either lacking\na grammatical gender system of having a system that is “semantic and formal” are unambiguous.\n\n\n14\n\nMention: \"WALS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its compiled gender data must be used with caution.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007401:49:0", "occurrence_id": "23bd4cfde48fb53ae2d702b6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Appendix Table S4. Model With Interaction Between Inequality and Initial Income\n\n(WDI Data)\n\n\nDependent Variable is: ln(yit)\n\n(1) (2) (3)\n\n\n\nGiniit 40.52***\n(7.11)\n\n\nGiniit -5.48***\n*ln(yi1980) (0.99)\n\n\n\n21.88***\n(3.69)\n\n\n-2.98***\n\n(0.50)\n\n\n\n25.05***\n(5.67)\n\n\n\nGiniit -3.39***\n*ln(yit-1) (0.79)\n\n\n\nln(yit-1) 0.53***\n(0.08)\n\n\n\n0.43***\n(0.10)\n\n\n\nKleibergen Paap F-Statistic 72 100 70\n\nCountry FE Yes Yes Yes\n\nTime FE Yes Yes Yes\n\nObservations 495 490 554\n\n\nNote: The method of estimation is two-stage least squares. Bootstrapped standard errors are shown in parentheses. The instrument for _Gini_ it is\n_Residual Giniit = Giniit – αln(yit)_, where α measures the effect that ln(yit) has on _Gini_ it.\n\n\n48\n\nMention: \"WDI Data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Inequality and initial income enter the growth regression from this source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007402:8:1", "occurrence_id": "0301e4417154d7ded46b4e53", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  social accounting matrix (SAM) of year 2007. The social accounting matrixes\n\nof later years (e.g., 2012) are also constructed by some studies. <sup>9</sup> We used the 2007 SAM by design,\n\n\n8 The Stone-Geary function is used to model household demand when households are assumed to have a subsistence\n\n\nlevel of consumption irrespective of the prices of goods for subsistence consumption and household income.\n\n\nHouseholds then make a choice for further consumption. It uses the linear expenditure system (LES) for estimating\n\n\nconsumer demand. The LES functional form was first introduced by British economist Sir John Richard Nicholas\n\n\nStone (Stone, 1954) and explained by Ragner Frisch (Frisch, 1954), the Norwegian economist and the co-recipient of\n\n\nthe first Nobel Memorial Prize in Economic Sciences in 1969.\n\n\n9 Macroeconomics & Fiscal Management Global Practice, South Asia Region, Economic Impacts of Hydropower\nInvestments, World Bank, April 2017, unpublished document.\n\n\n7\n\nMention: \"the 2007 SAM\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Chosen by design over later constructions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007405:3:0", "occurrence_id": "f51ed526020291d136f7c2e7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ##### INTRODUCTION\n\nUkraine scores closer to the world’s poorest economies on most measures of perceived corruption\nthan it does to the European Union, which it aspires to join. Ukraine was ranked in the 20th\npercentile worldwide in the World Governance Indicators (WGI) Control of Corruption indicator in\n2016, well below Poland, Romania, and averages for lower-middle-income countries and SubSaharan Africa. Transparency International’s 2016 Corruption Perception Index (CPI) ranked\nUkraine 131st out of 167 countries; only Kyrgyzstan, Tajikistan, Uzbekistan and Turkmenistan in\nthe Europe and Central Asia region performed worse. Various attempts have been made to\nquantify the cost of corruption in Ukraine, with estimates ranging from 4 to 10 percent of GDP\nannually between 2007-2011. <sup>1</sup>\n\n\n**Figure 1. Control of Corruption Ukraine vs comparators**\n\n\n<u>Source: Word Governance Indicators</u>\n\n\nIn this paper we focus on rent seeking as distinct from corruption. Rent seeking is the manipulation\nof public institutions to obtain economic rent – the portion of income paid to a factor of production\nin excess of what is needed to keep it employed in its current use – without the creation of new\nwealth. The conventional definition of corruption assumes that it is always a crime. <sup>2</sup> Rent seeking is\nsometimes legal. Both corruption and rent seeking misappropriate public resources for personal\nbenefit to the detriment of society. In Ukraine, rent seeking includes the award of public resources\nto companies through tax exemptions, direct subsidies and procurement contracts to connected\ncompanies <sup>3</sup> that cannot be justified in terms of the economic benefits to society as a whole. The rent\nseeking activities provide a basis for the existence of so-called “crony capitalism”.\n\n\nWe call the model of economic governance in which policy decisions are subject to undue influence\nof a small number of businesses “\n\nMention: \"World Governance Indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its control-of-corruption percentile places one country low.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007405:3:0", "occurrence_id": "b3ea371ef277c18ce709d0f8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ##### INTRODUCTION\n\nUkraine scores closer to the world’s poorest economies on most measures of perceived corruption\nthan it does to the European Union, which it aspires to join. Ukraine was ranked in the 20th\npercentile worldwide in the World Governance Indicators (WGI) Control of Corruption indicator in\n2016, well below Poland, Romania, and averages for lower-middle-income countries and SubSaharan Africa. Transparency International’s 2016 Corruption Perception Index (CPI) ranked\nUkraine 131st out of 167 countries; only Kyrgyzstan, Tajikistan, Uzbekistan and Turkmenistan in\nthe Europe and Central Asia region performed worse. Various attempts have been made to\nquantify the cost of corruption in Ukraine, with estimates ranging from 4 to 10 percent of GDP\nannually between 2007-2011. <sup>1</sup>\n\n\n**Figure 1. Control of Corruption Ukraine vs comparators**\n\n\n<u>Source: Word Governance Indicators</u>\n\n\nIn this paper we focus on rent seeking as distinct from corruption. Rent seeking is the manipulation\nof public institutions to obtain economic rent – the portion of income paid to a factor of production\nin excess of what is needed to keep it employed in its current use – without the creation of new\nwealth. The conventional definition of corruption assumes that it is always a crime. <sup>2</sup> Rent seeking is\nsometimes legal. Both corruption and rent seeking misappropriate public resources for personal\nbenefit to the detriment of society. In Ukraine, rent seeking includes the award of public resources\nto companies through tax exemptions, direct subsidies and procurement contracts to connected\ncompanies <sup>3</sup> that cannot be justified in terms of the economic benefits to society as a whole. The rent\nseeking activities provide a basis for the existence of so-called “crony capitalism”.\n\n\nWe call the model of economic governance in which policy decisions are subject to undue influence\nof a small number of businesses “\n\nMention: \"2016 Corruption Perception Index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its ranking of 131st out of 167 countries is the comparison used.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007405:3:0", "occurrence_id": "be795160c6847d74030490a7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ##### INTRODUCTION\n\nUkraine scores closer to the world’s poorest economies on most measures of perceived corruption\nthan it does to the European Union, which it aspires to join. Ukraine was ranked in the 20th\npercentile worldwide in the World Governance Indicators (WGI) Control of Corruption indicator in\n2016, well below Poland, Romania, and averages for lower-middle-income countries and SubSaharan Africa. Transparency International’s 2016 Corruption Perception Index (CPI) ranked\nUkraine 131st out of 167 countries; only Kyrgyzstan, Tajikistan, Uzbekistan and Turkmenistan in\nthe Europe and Central Asia region performed worse. Various attempts have been made to\nquantify the cost of corruption in Ukraine, with estimates ranging from 4 to 10 percent of GDP\nannually between 2007-2011. <sup>1</sup>\n\n\n**Figure 1. Control of Corruption Ukraine vs comparators**\n\n\n<u>Source: Word Governance Indicators</u>\n\n\nIn this paper we focus on rent seeking as distinct from corruption. Rent seeking is the manipulation\nof public institutions to obtain economic rent – the portion of income paid to a factor of production\nin excess of what is needed to keep it employed in its current use – without the creation of new\nwealth. The conventional definition of corruption assumes that it is always a crime. <sup>2</sup> Rent seeking is\nsometimes legal. Both corruption and rent seeking misappropriate public resources for personal\nbenefit to the detriment of society. In Ukraine, rent seeking includes the award of public resources\nto companies through tax exemptions, direct subsidies and procurement contracts to connected\ncompanies <sup>3</sup> that cannot be justified in terms of the economic benefits to society as a whole. The rent\nseeking activities provide a basis for the existence of so-called “crony capitalism”.\n\n\nWe call the model of economic governance in which policy decisions are subject to undue influence\nof a small number of businesses “\n\nMention: \"Word Governance Indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The comparator chart on control of corruption is drawn from here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007407:9:0", "occurrence_id": "d6f0aa9be26f28d3d610dc0c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 7\n\n\nThe importance of record keeping, in the Nigerian context, stems from several factors.\n\n\nFirst and foremost is the fact that the Government of Nigeria has struggled to reliably track how\n\n\nmuch it spends on primary health care. This is partly because classifications of expenditure in the\n\n\nhealth sector tend to be too broad to capture such granular but vital information, but also because\n\n\nthe resources (cash and in-kind) that flow to primary health care in the country are heavily\n\n\nfragmented, with no system in place to aggregate them into central databases. This means that it\n\n\nis important to devise alternative ways of tracking income and expenditure in the sector. In an\n\n\neffort to help Nigeria with that development, the World Bank introduced an innovative\n\n\ncontinuous Public Expenditure Tracking Survey (PETS) and Resource Tracking (RT) exercise in\n\n\ntwo states (Ekiti and Niger), which recorded and aggregated information on resource flows in\n\n\nreal time, rather than following the more established practice of retrospectively generating such\n\ndata. In both states, the intervention and the monitoring were conducted by the same local firm. <sup>3</sup> [^3: The local firm carrying out this work was Hanovia Medical Ltd., Nigeria.]\n\n\nThe need is especially great at the primary facility level, for which no entity above the\n\n\nfacility level – including local government and state authorities – has reliable information on\n\n\nincome and expenditure streams. One of the central issues identified by previous studies is the\n\n\nweakness of accountability structures at the facility level – a gap that a World Bank-financed\n\n\nintervention is attempting to address. The main contribution of the PETS/RT has been to design\n\n\nand introduce relatively simple-to-fill-out forms (spreadsheets; other resource tracking tools) to\n\n\ntrack various income and expenditure streams. This includes, for example, creation of a\n\n\nstandardized cash book form (for the recording of cash receipts from user charges e.g.,\n\n\nconsultation fees, laboratory tests, and drug sales and expenditures such as due to purchases of\n\n\ndrugs, equipment, materials, or supplies\n\nMention: \"continuous Public Expenditure Tracking Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"An innovative continuous tracking exercise run in two states.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007409:17:1", "occurrence_id": "958366991659bec073241b85", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: s. <sup>31</sup> Also, I exclude years 2008, 2009, and 2010 in which\ntrade flows were heavily affected by the global financial crisis. This is also\nto be consistent with the literature in using 5-year intervals. The resulting\ndataset has 56 countries and 4 years (1995, 2000, 2005, and 2011). <sup>32</sup>\n\nFor data on trade agreements, I use the Economic Integration Agreement (EIA) database by Baier and Bergstrand (2007) and the Design of\nTrade Agreements (DESTA) database by D¨ur et al. (2014). The latest\nversion of the EIA database covers the period 1950–2012. They code\nthe trade agreements as following: (1) non-reciprocal preferential trade\narrangements, (2) preferential trade arrangements, (3) free trade areas, (4)\ncustoms union, (5) common market, and (6) economic union. The dummy\nvariable for the existence of an RTA equals one if a country pair has an\nFTA or stronger, as often used in the literature.\n\n\n292015 edition.\n30The wide coverage of developing and developed countries makes the OECD ICIO\ndatabase most suitable for the analysis of heterogeneous effects of services trade agreements by income level. Alternative data sources for ICIOs include the World Input Output Database (WIOD) which provides more detailed sectoral data but limited mostly\nto advanced economies.\n\n31These are Brunei Darussalam, Cyprus, Hong Kong SAR, China, Luxembourg, and\nSaudi Arabia. See Appendix A.1 for the full list of countries.\n\n32The baseline regressions include years 1995, 2000, 2005, and 2011 instead of 2010\nfor two reasons: (i) In 2010, many countries’ trade flows were still heavily affected by\nthe crisis, and (ii) by using 2011 instead of 2010, the number of services agreements\nincluded in the sample increases significantly\n\nMention: \"Economic Integration Agreement (EIA) database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Trade agreement membership across these years comes from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007409:17:1", "occurrence_id": "2e893c4989fb5fd53141c87c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: s. <sup>31</sup> Also, I exclude years 2008, 2009, and 2010 in which\ntrade flows were heavily affected by the global financial crisis. This is also\nto be consistent with the literature in using 5-year intervals. The resulting\ndataset has 56 countries and 4 years (1995, 2000, 2005, and 2011). <sup>32</sup>\n\nFor data on trade agreements, I use the Economic Integration Agreement (EIA) database by Baier and Bergstrand (2007) and the Design of\nTrade Agreements (DESTA) database by D¨ur et al. (2014). The latest\nversion of the EIA database covers the period 1950–2012. They code\nthe trade agreements as following: (1) non-reciprocal preferential trade\narrangements, (2) preferential trade arrangements, (3) free trade areas, (4)\ncustoms union, (5) common market, and (6) economic union. The dummy\nvariable for the existence of an RTA equals one if a country pair has an\nFTA or stronger, as often used in the literature.\n\n\n292015 edition.\n30The wide coverage of developing and developed countries makes the OECD ICIO\ndatabase most suitable for the analysis of heterogeneous effects of services trade agreements by income level. Alternative data sources for ICIOs include the World Input Output Database (WIOD) which provides more detailed sectoral data but limited mostly\nto advanced economies.\n\n31These are Brunei Darussalam, Cyprus, Hong Kong SAR, China, Luxembourg, and\nSaudi Arabia. See Appendix A.1 for the full list of countries.\n\n32The baseline regressions include years 1995, 2000, 2005, and 2011 instead of 2010\nfor two reasons: (i) In 2010, many countries’ trade flows were still heavily affected by\nthe crisis, and (ii) by using 2011 instead of 2010, the number of services agreements\nincluded in the sample increases significantly\n\nMention: \"Design of\nTrade Agreements (DESTA) database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its provisions detail complements the membership data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007409:17:1", "occurrence_id": "17c692e2bbdcf5c1f33572af", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: s. <sup>31</sup> Also, I exclude years 2008, 2009, and 2010 in which\ntrade flows were heavily affected by the global financial crisis. This is also\nto be consistent with the literature in using 5-year intervals. The resulting\ndataset has 56 countries and 4 years (1995, 2000, 2005, and 2011). <sup>32</sup>\n\nFor data on trade agreements, I use the Economic Integration Agreement (EIA) database by Baier and Bergstrand (2007) and the Design of\nTrade Agreements (DESTA) database by D¨ur et al. (2014). The latest\nversion of the EIA database covers the period 1950–2012. They code\nthe trade agreements as following: (1) non-reciprocal preferential trade\narrangements, (2) preferential trade arrangements, (3) free trade areas, (4)\ncustoms union, (5) common market, and (6) economic union. The dummy\nvariable for the existence of an RTA equals one if a country pair has an\nFTA or stronger, as often used in the literature.\n\n\n292015 edition.\n30The wide coverage of developing and developed countries makes the OECD ICIO\ndatabase most suitable for the analysis of heterogeneous effects of services trade agreements by income level. Alternative data sources for ICIOs include the World Input Output Database (WIOD) which provides more detailed sectoral data but limited mostly\nto advanced economies.\n\n31These are Brunei Darussalam, Cyprus, Hong Kong SAR, China, Luxembourg, and\nSaudi Arabia. See Appendix A.1 for the full list of countries.\n\n32The baseline regressions include years 1995, 2000, 2005, and 2011 instead of 2010\nfor two reasons: (i) In 2010, many countries’ trade flows were still heavily affected by\nthe crisis, and (ii) by using 2011 instead of 2010, the number of services agreements\nincluded in the sample increases significantly\n\nMention: \"OECD ICIO\ndatabase\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its wide coverage suits the heterogeneous effects analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007412:20:0", "occurrence_id": "3f8ef6a6abab0682dd44ab27", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 3: Costs of financial intermediation and financial depth**\n\n\n_Source_ : Global Financial Development Database.\n\n\n**Figure 4: Costs of financial intermediation and financial inclusion**\n\n\n_Source_ : World Development Indicators; Financial Access Survey.\n\n\n19\n\nMention: \"Global Financial Development Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Intermediation cost and financial depth series come from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007412:20:0", "occurrence_id": "0e881dddd80c771d10171beb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 3: Costs of financial intermediation and financial depth**\n\n\n_Source_ : Global Financial Development Database.\n\n\n**Figure 4: Costs of financial intermediation and financial inclusion**\n\n\n_Source_ : World Development Indicators; Financial Access Survey.\n\n\n19\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Supplies the inclusion series alongside the access survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007412:20:0", "occurrence_id": "ec13fcfa0f61be473fc7344a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 3: Costs of financial intermediation and financial depth**\n\n\n_Source_ : Global Financial Development Database.\n\n\n**Figure 4: Costs of financial intermediation and financial inclusion**\n\n\n_Source_ : World Development Indicators; Financial Access Survey.\n\n\n19\n\nMention: \"Financial Access Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its access indicators complete the inclusion picture.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007417:5:1", "occurrence_id": "387d11e02a4216837be8631a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  85 percent of\n\n\nfarmers engage in rice production. Using the spatial distribution of rice production based on\n\n\nthe 2010 SPAM, the latest district-level rice production data for the period of 2013-15 are\n\n\ndisaggregated into 3,198 locations or pixels. This is our dependent variable, _Y_ ( **Figure 3** ).\n\n\n1 It is available on the Internet at MapSpaM.info.\n\nMention: \"2010 SPAM\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its spatial distribution of rice production anchors the district allocation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007419:6:0", "occurrence_id": "68ad3bcc92f5bd8f38636a09", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: - 5 \n\nMadagascar possesses an official road network of 31,640 km, of which only about 5,600 km\n\n\nare paved. While about 70 percent of the paved roads are in good or fair condition, the vast\n\n\nmajority of unpaved roads are in poor condition ( **Figure 2** ). Because of the poor condition of\n\n\nthe feeder road network, people’s transport mobility is extremely limited in rural areas. The\n\n\nRural Access Index, which measures the share of rural population who live in within 2 km of\n\n\na road in good condition, is estimated at 11.4 percent in Madagascar, leaving 17 million rural\n\n\npeople unconnected ( **Figure 3** ).\n\n\n**Figure 1. District-level poverty estimate** **Figure 2. Official road network**\n\n\nSource: World Bank estimate. Source: Madagascar Road Authority.\n\n\nAccording to the official firm registry database, as of 2017, there existed 1,309\n\n\nagrobusinesses in the country. Output-related agrobusinesses, including large plantations,\n\n\ncollectors, processing companies, and exporters, are dominant, which amount to 902\n\n\ncompanies in the database. On the other hand, 399 input-related agrobusinesses, also exist,\n\nsuch as fertilizer and other input dealers and equipment suppliers. <sup>2</sup> The spatial distribution of\n\n\nagrobusinesses is highly skewed ( **Figure 4** ). Among the agribusinesses, 569 firms or nearly\n\n\n2 This classification may be ambiguous. In the database, each agrobusiness is categorized as either an input- or\noutput-related firm based on the description of its primary business activity. In practice, however, some of them\nare likely to engage in both activities.\n\nMention: \"official firm registry database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Counts the agribusinesses registered as of that year.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007421:12:0", "occurrence_id": "c9e1917afe98dc0cd2f1dcf6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: - at time - ; <sup>14</sup> [^14: Data on BITs come from UNCTAD.] ������ is a dichotomous variable that takes the value of 1 whenever there is a\n\npreferential trade agreement, either active or inactive, between two countries at time - ; <sup>15</sup> [^15: The variable ������ is taken from Egger and Larch (2008).] - s are\n\n\nsets of country‐pair, importer‐time and exporter‐time fixed effects.\n\n\nb. Baseline results\nTable 2 reports the coefficients of total depth, core depth and PCA depth for the regressions\n\n\nusing DVA and FVA as dependent variables. All coefficients of depth are positive and significant.\n\n\nAdding a policy area is associated to an average increase 0.4 percent of total domestic value\n\n\nadded and an average increase of 0.26 percent of foreign value added. The coefficients increase\n\n\nsubstantially when looking at core depth only, suggesting that those policy areas are particularly\n\n\nimportant as they reduce the governance gap between countries in areas that are relevant for\n\n\nGVC‐related trade. <sup>16</sup> [^16: See Baldwin (2011) and WTO (2011).] Also, the coefficients of PCA depth show a similar pattern to the one of total\n\n\nnumber of provisions. Control variables such as BITs have the expected sign, suggesting that\n\n\nsigning BITs has a positive impact on domestic and foreign value added in exports. <sup>17</sup> [^17: The magnitude of the BITs coefficient needs to be interpreted carefully, given that these agreements often focus\non specific sectors or areas. Therefore, in our regressions, which are aggregated at the country level, their impact\nmight be positively biased.] The PTA\n\n\ndummy is non‐significant in most of the estimations. This variable controls for the presence of\n\n\nshallow PTAs and for agreements no longer in force for which we have no information\n\nMention: \"UNCTAD\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Bilateral investment treaty data come from this body.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007424:4:0", "occurrence_id": "f5aacffcac37d58438990ce8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 1. Developments in commodity markets**\n\n\n**A. Real commodity prices** **B. Cumulative growth in GDP, population,**\n**energy and metals consumption, 1996-2016**\n\n\n\nUS$ 2010=100\n\n160\n140\n120\n100\n80\n60\n40\n20\n0\n\n\n\n\n\nPercent\n\n180\n150\n120\n90\n60\n30\n0\n\n\n\n**C. Contribution to average annual growth in**\n**metals consumption**\n\n\n\n**D. Contribution to average annual growth in**\n**energy consumption**\n\n\n\nPercentage points\n\n3\n\n\n2\n\n\n1\n\n\n0\n\n\n-1\n\n\n\nPercentage points\n\n5\n\n\n4\n\n\n3\n\n\n2\n\n\n1\n\n\n0\n\n\n-1\n\n\n\nChina India\nOther EM7 Other EMDEs\n\n\n\nChina\nIndia\nOther EM7\n\n\n\n1996-2008 2008-2016\n\n\n\n1996-2008 2008-16\n\n\n\n**E. EM7 share of commodity consumption** **F. Commodity intensity of output**\n\n\n\nPercent of world total\n\n60 2016 1996\n\n\n40\n\n\n20\n\n\n0\n\n\n\nIndex, 1965 = 100\n\n120\n\n110\n\n100\n\n90\n\n80\n\n70\n\n\n\n60\n\n50\n\n\n\nEnergy ex China Metals ex China\n\n\n\n\n\nSources: BP Statistical Review, U.S. Department of Agriculture, World Bank, World Bureau of Metals Statistics.\nA. Deflated using the manufacturing unit value index from the January 2018 edition of the Global Economic Prospects report.\nB.-D. Metals aggregate includes aluminum, copper, lead, nickel, tin, and zinc. Energy aggregate includes coal, crude oil, natural gas, nuclear,\nand renewables.\nC.D. “AEs” stands for advanced economies. Other EM7 includes Brazil, Indonesia, Mexico, Russia, and Turkey.\nC. “Other AEs” contains 18 advanced economies. Other EMDEs contains 32 countries.\nD. “Other AEs” contains 17 advanced economies. Other EMDEs contains 31 countries.\nE. Other EM7 includes Brazil, India, Indonesia, Mexico, Russia, and Turkey.\nF. Commodity intensity calculated as global energy and metals use (in volumes) relative\n\nMention: \"Global Economic Prospects report\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its manufacturing unit value index deflates the commodity series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007428:10:1", "occurrence_id": "c6228a8bdaf7c77e018f8ae1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  potential employer would wait until a qualified applicant appears. This means\nthat labor demand intersects labor supply at the upward sloping portion of the supply curve.\n\n\n  - Business climate interventions assume that _the supply of youth_ with the characteristics needed\nto succeed in employment is adequate, but the business climate needs to improve to generate\nmore and better opportunities for youth. These interventions try to move out the demand\ncurve along the supply curve.\n\n\nMost youth employment interventions are supply side ones, and of these, almost all involve some kind\nof training. <sup>8</sup> [^8: 82 percent of the interventions listed in the ILO’s Youth Employment Program Inventory are training programs.] As discussed in Section 2, the assumptions behind supply‐side interventions for wage\nemployment are certainly open to question in low‐income countries.\n\n\nBecause multiple constraints can be involved, youth employment issues are often addressed by several\ninterventions. This complexity often makes program evaluation difficult, especially impact evaluation,\nas it is not clear which component was most effective in alleviating which constraint, and what might\nbe the most cost‐effective way to move forward.\n\n\nA critical challenge in experimental research on employment interventions is measuring displacement.\nMost studies have not even tried to do so. Although displacement can be difficult to measure in an\nimpact evaluation, a few have been able to do it. A Ugandan study measured displacement by\nrandomizing the firms which participated in the project and received vocational training graduates or\ninterns (Alfonsi et al., 2017). This study did find complete displacement. A study in France measured\ndisplacement effects by varying the intensity of the treatment across areas (Crepon et al., 2012). While\nthe study did find that the employment program helped treated individuals within the treatment area\nfind a job “slightly faster” than those who were not in the study, when the authors compared areas\nwhere most youth were allowed to participate (treated) with areas where few youth were allowed to\n\n\n7 Most firm‐level interventions\n\nMention: \"ILO’s Youth Employment Program Inventory\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Most listed interventions are training programmes.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007431:46:0", "occurrence_id": "e61b3e4e8719413d8ab9b218", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Table A2: Skilled labor value added contained in Chinese exports, 1995-2011 (relative to**\n\n**total value added, % of total)**\n\n\n**<u>Sector</u>** **<u>1995</u>** **<u>2011</u>**\nAgr, Forestry, Fisheries 0.80 3.52\nBeverages and Tobacco Products 12.00 13.45\nConstruction 14.29 17.87\nChemical, Rubber, Plastic Products 16.70 18.39\nElectricity, Gas, Water 34.49 29.71\nEnergy Extraction 11.75 18.72\nMetal Products 15.00 18.27\nFerrous Metals 15.10 18.50\nLeather Products 13.60 13.65\nWood Products 11.30 13.13\nMachinery and Equipment nec 17.00 19.43\nMetals nec 15.90 19.08\nMineral Products nec 14.70 18.53\nManufactures nec 10.90 13.97\nMinerals nec 12.40 17.30\nPubAdmin/Defence/Health/Educat 60.50 52.85\nOther Private Services 40.80 34.56\nProcessed Foods 16.90 9.27\nPaper Products, Publishing 15.50 17.30\nTrade and Transport Services 18.50 19.31\nTextiles 14.00 14.63\nTransport Equipment 15.10 18.68\nWearing Apparel 12.10 15.24\n**<u>Total</u>** **<u>15.93</u>** **<u>17.88</u>**\n\n\nNotes: Labor Content of Exports (LACEX) database, World Bank. The skilled domestic labor value added embodied\nin a sector's exports, including the skilled wages paid directly for the production of the sector's exports and indirectly\nvia the production of economy-wide inputs for the sector's exports\n\nMention: \"Labor Content of Exports (LACEX) database, World Bank\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Skilled domestic value added embodied in sectoral exports comes from here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007413:16:0", "occurrence_id": "8f6cdc383f184f15685ff60d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Table 2 – Financial Dependency and BNDES Credit on Distortions: Main results**\n\n\n<u>(1)</u> <u>(2)</u>\n\n\n<u>VARIABLES</u> <u>ln(1 - tyit)</u> <u>ln(1 + tkit)</u>\n\n\n(a) BNDES Creditit 0.0195*** -0.0346***\n(0.00450) (0.00464)\n(b) BNDES Cred * Fin.Depit 0.00183 -0.00797**\n(0.00338) (0.00349)\n(c) Financial Depit -0.0391*** 0.129***\n(0.00186) (0.00192)\n\n\nObservations 104,092 104,092\nNumber of ii 36,598 36,598\nR-squared 0.138 0.113\nBNDES Cred. Contrib to Var -0.13% 0.09%\nMedian Fin Dep. 0.166 0.166\nBNDES Credit Eff. 0.0198 -0.0359\n<u>p-value</u> <u>0.000</u> <u>0.000</u>\n\n\nNote: authors´ calculation based on raw IBGE manufacturing data, 2003-2014. BNDES credit is a dummy\nwhich takes value 1 if firm _i_ used BNDES/FINAME credit in year _t_ and zero otherwise; Financial Dep equals\ninvestment outlays / cash flow, where cash flow is value added minus wage bill; Firm and Year dummies\nincluded; lagged log revenue and log value added included as controls. BNDES Cred. Contrib. is the proportion\nof the variance of the dependent variable explained by BNDES Credit variables, following\n\nMention: \"IBGE manufacturing data, 2003-2014\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Raw establishment records across this span carry the productivity estimates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007435:54:1", "occurrence_id": "097e079a43c31607b4336664", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: >Boundary</u> <u>5</u> <u>km</u> <u>radius</u> <u>50</u> <u>km</u> <u>radius</u>\nN 3359370 3359370 689745 689745 688275 688275\n<u>r2</u> <u>0.95</u> <u>0.96</u> <u>0.94</u> <u>0.95</u> <u>0.94</u> <u>0.97</u>\n_<u>∗p <</u>_ 0 _._ 10 _,∗∗_ _p <_ 0 _._ 05 _,∗∗∗_ _p <_ 0 _._ 01\n\nThe table shows difference-in-differences estimates of the impact of new village roads on local forest cover, under\nalternate sample definitions. We define forest cover as log village forest cover; the data source is Vegetation Continuous\nFields. Specifications are identical to those in Table 3 with the following changes. Column 1 includes villages that did\nnot receive PMGSY roads at any time as part of the control group. Column 2 adds village-specific time trends to show\nthat the positive treatment estimate in Column 1 is driven by differential trends in never-treated villages. Columns\n3 and 4 repeat these two specifications with the standard set of villages to show that village-specific time trends do\nnot affect our main estimates. Column 5 estimates the standard specification from Table 3 with treated villages only,\nbut the dependent variable includes forest cover in a 5km radius from the village centroid. Column 6 uses a 50km\ncentroid. Award Period is an indicator variable that takes the value one for years after a road contract was awarded\nand before the\n\nMention: \"Vegetation Continuous\nFields\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Village forest cover is defined from this satellite product.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007436:42:0", "occurrence_id": "ad8acd19d2b744c6af049f6d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: the government and non‐government forces. Violence against civilians was committed in 28 percent\n\n\nof all events (Figure 16).\n\n\n**Figure 16: Conflict events by type**\n\n\nBattle‐Government\n\nregains territory\n\n2%\n\n\n28%\n\n\nBattle‐No change of\n\n\n\nStrategic\ndevelopment\n\n10%\n\n\n7%\n\n\n\nterritory\n\n46%\n\n\n\n\n\n1%\n\n\n\nbase established\n\n0%\n\n\n\nDistance to a deadly conflict event is used to generate an external conflict exposure variable. The\n\n\naverages of latitude and longitude of all households in a cluster in the AGI survey are used to compute\n\n\ncluster GPS coordinates. By merging the girls’ households GPS coordinates with the conflict event GPS\n\n\ncoordinates, the distance between each cluster‐conflict event pair is calculated. The continuous\n\n\nindicator is the normalized sum of the distances of all fatal conflict events within a radius of 5 km from\n\n\nthe cluster.\n\n\n41\n\nMention: \"AGI survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"Household coordinates within clusters are averaged to give cluster locations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007457:7:0", "occurrence_id": "0fe913aabc1f93f12e032a5c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: eliminated as those bands were water‐vapor absorption bands/for overlapping regions/had no\ninformation (Barry 2001; Datt et al. 2003). Bad bands were removed while converting Digital\nNumber (DN) value to radiance using the radiometric calibration tool. The output was converted\nto band‐interleaved‐by line (BIL) radiance image with floating point values as Fast Line‐of‐sight\nAtmospheric Analysis of Spectral Hypercube (FLAASH) correction module use BIL format. After\nremoval of 67 bad bands, a total of 175 calibrated bands were obtained and used for further\nprocessing.\n\n\nThese 175 calibrated bands then underwent visual and statistical examination; and signal‐\nto‐noise ratios were determined for all bands. At this stage, several other corrections were made,\nincluding de‐striping and atmospheric corrections. During the acquisition of Hyperion data,\nvertical striping occurs at times due to poor calibration of push broom sensors. For this analysis,\nde‐striping was performed by filling up the DN value of the gap line with an average DN value\nfrom the previous and the next column (Farooq and Govil 2014). Atmospheric correction was\nperformed using the FLAASH package available in ENVI <sup>TM</sup> . These corrections provided calibrated\nimage spectra of mangrove species (Figure 3).\n\n\n**Figure 3: Comparison of pre‐ and post‐correction images**\n\n\nMangrove species‐specific image spectra were then generated from corrected Hyperion\nimagery for a subset of GPS locations where field information on existing mangroves was\ncollected (Figure 4). Absorption and peak reflectance detected from the spectral profile indicated\nthe characteristics of each species. All remaining Hyperion images were then classified with those\ndetected spectral signatures using the Spectral Angle Mapper (SAM) in ENVI <sup>TM</sup> .\n\n\n6\n\nMention: \"Hyperion data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Vertical striping from sensor calibration requires de-striping corrections.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007459:4:0", "occurrence_id": "6a18d79f3c859d6645511929", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: In this paper, use a dataset of publicly listed firms in the United States from the Compustat\n\n\ndatabase to identify star firms and the industries in which these firms are more likely to appear in.\n\n\nWe examine the role of intangible capital, competition (Herfindahl-Hirschman industry concentra\n\ntion index), and market power in influencing star firm status. We use three measures of market\n\n\npower - a firm-level measure of operating markups, firm-level market share, and a raw measure of\n\n\nfirm size. <sup>4</sup> Finally, we examine the persistence of the star firms’ performance and whether they\n\n\ndiffer in their investment and output per unit of capital compared to other firms in the economy.\n\n\nFollowing Furman and Orszag [2015], we define star firms as firms in the top 10% of Return\n\n\non Invested Capital (ROIC) in the US in a particular year. <sup>5</sup> Using conventional ROIC measures,\n\n\nin Figure 1, we find that there has indeed been a run-up in the ROIC of the top decile of large,\n\n\nnon-financial sector, publicly-listed U.S. firms. <sup>6</sup> Over the period 1965-2015, the ratio of the 90 <sup>_th_</sup>\n\n\npercentile ROIC firm to the median ROIC firm has increased by over 69%. Importantly, we find\n\n\nthat the star firms whose returns are diverging from the rest of the firms are in industries that\n\n\nrequire high cognitive skills and that in these industries average returns are higher. In industries\n\n\nwhere the tasks involve routine manual skills and which score low on non-routine cognitive and\n\n\ncomplex problem solving skills, we see lower returns and don’t see the star firms pulling away from\n\n\nthe rest.\n\n\nHowever, conventional return metrics do not capitalize research and development, brand\n\nMention: \"Compustat\n\n\ndatabase\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Publicly listed firms in one country are identified from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007460:31:0", "occurrence_id": "92046a7552be87584ce1a5f7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 11: Effect of FV on exchange rate pass-through\n\n\n_Source_ : World Input-Output tables (2013 release), BIS and authors’ calculations.\n\n\n_Note_ : The Figure reports percentiles of _FV_ with respective values in the second row of the x-axis and the corresponding\n\n\nexchange rate elasticity on the vertical axis. Elasticities are computed from the specification in column (6) of Table 2,\n\n\nincluding origin-sector-destination fixed effects and time dummies.\n\n\n**5.2.2** **Export** **volume** **elasticities**\n\n\nTable 3 reports the results for the export volumes regression. <sup>26</sup> The pooled OLS reveals\n\n\na dampening effect of _FV_ on the bilateral exchange rate elasticity and the positive effect\n\n\nof _IV_ on the elasticity of exports to the direct partner’s nominal effective exchange rate\n\n\n(NEER). When we exploit the within sector-destination (second column) or the sector\n\norigin-destination (fourth column) variability and control for common time effects, we\n\n\nobtain a negative and significant coefficient for the interaction terms of the bilateral ex\n\nchange rate with _FV_ and _RDV_ and an insignificant effect of _IV_ on the exchange rate\n\n\nelasticity.\n\n\nInterestingly, the coefficient of the destination’s NEER vis-à-vis its trading partners is\n\n\nnot significant, whereas its interaction with _IV_ is positive and significant, meaning that\n\nsector _s_ <sup>_′_</sup> exports are more sensitive to its direct trading partners nominal effective exchange\n\n\nrate the higher is the quantity of exports that are intermediated by the latter. The results\n\n\nare confirmed even when controlling for time-varying characteristics of countries/currencies\n\n\nwith country-time dummies, although the interaction of _IV_ with the NEER of the desti\n\nnation market becomes statistically insignificant, but remains\n\nMention: \"World Input-Output tables (2013 release)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its linkage coefficients give the pass-through exposure.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007461:20:0", "occurrence_id": "e642d99d57c02ae964138d38", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 19\n\n\nWe compute, for each country, time paths of childhood stunting and cognition without the\n\n\nnutrition program, to which we apply the program effects 𝑑𝑆��𝑡�𝜕𝐷⁄ ��𝑡� and 𝑑𝐶��𝑡�𝜕𝐷⁄ ��𝑡�. We\n\n\nget the counterfactual childhood stunting time path by setting the 2015 childhood stunting rate equal\n\n\nto the latest JME childhood stunting rate for the country in question, and then assuming that before\n\n\nand after 2015 stunting falls at an annual rate of 1.5% (the rate we computed in section 2 above). For\n\n\nthe counterfactual cognition time path, we assume zero change in the absence of the nutrition\n\n\nprogram, and assume the z-score is initially zero. As in a recent World Bank report (World Bank\n\n\n2016), we assume the program goes to scale gradually over a 10-year period between 2016 and 2025,\n\n\nachieving Bhutta et al.’s estimated 20% reduction (compared to the counterfactual) in 2025. This\n\n\ngives us, after 10 years, a value of 𝑑𝑆��𝑡�𝜕𝐷⁄ ��𝑡� (in terms of year-to-year changes) equal to -3.7%.\n\n\nWe assume the program is maintained at the same scale thereafter, so the stunting rate remains at\n\n\n20% below the counterfactual rate for all periods after 2025.\n\n\nWe estimate the change in the cognition z-score attributable to the nutrition program,\n\n\n𝑑𝐶��𝑡�𝑑𝐷⁄ ��𝑡�, by multiplying, for each intervention in the Bhutta et al. package, the estimated\n\n\ncognition effect of the intervention (obtained from Table 3) by 0.9 minus the fraction of children\n\n\ncurrently covered by the intervention. (The effect size is relevant for going from 0% to 100%,\n\n\nwhereas the program takes intervention coverage from its current rate to 90%.) We use the\n\n\nestimated mean cognition effects from the meta-analyses summarized in Table 3. The current\n\n\nintervention coverage rates are from various sources. We assume the effects\n\nMention: \"JME childhood stunting rate\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The latest rate for each country sets the model's starting level.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007463:14:1", "occurrence_id": "8ef76fe1d82158bcee9fdcbf", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  with no education or with primary education, (ii) farmers with secondary or tertiary\neducation, (iii) nonfarmers with no education or with primary education, and (iv) nonfarmers with\nsecondary or tertiary education. The positions of individuals (the sons and daughters) are also\ndefined in the same way as those of the fathers. However, contrary to Cogneau and Mesplé-Somps\n(2008), the unemployed, students, and retirees are not included because the outcome variable here\nis the labor earnings of individuals, that is, the sample captures only individuals who, at the time\nof the survey, reported they were employed.\n\nThe second data source is the Polity IV data set measuring the MEPVs in the country. The MEPV\nvariable of interest is the magnitude of all societal MEPVs. This includes episodes of civil and\nethnic violence as well warfare involving the country during a year. The scale of the magnitude\n\n\n16 Total labor earnings is defined here as the sum of all income derived from primary and secondary employment and\nself-employment for individuals. In the construction of the labor earnings aggregates, the bottom 1 percent and top 1\npercent were trimmed to correct for outliers.\n\n\n13\n\nMention: \"Polity IV data set\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its measure of political competition magnitude is the second data source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007468:29:0", "occurrence_id": "41577c24c05326f329c2ab20", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The overall satisfaction results across all users in various economies is high on average. Actually,\nusers of in-house and commercial systems yield very similar results where LIC and LMIC as well as\nHIC show generally a high satisfaction with their DMS compared to UMIC countries.\n\nAnother interesting observation is that when the respondents were asked about the overall\nsatisfaction with their current DMS, the results were on average higher than the average satisfaction\ncalculated as a result of each individual DMS feature regarding connection, integration, etc.\n\n\n**Figure 8: Satisfaction with the Debt Management System: Individual Attributes**\n\n\nSource: World Bank Survey on Debt Management Systems\n\n\nDespite the general satisfaction with the systems, the users still indicated a few areas that their\ncurrent DMS is not covering. Respondents using a commercial system emphasized the lack of\nintegration with the central bank to receive treasury debt issuance auction results. According to the\nsurvey results, system and IT limitations were listed as the main reasons for the lack of integration.\nOther reasons may include organizational policies and resource capacity, which the survey did not\ncover. Other issues raised in the survey were the connectivity with other platforms and lack of STP.\nUsers were also asked regarding features they would like to see in their system that are not essential\nand they currently do not have. Improved report generation and reliability were listed as the most\ndesired feature that needs improvement. Speed, deal capture and connection with other platforms\nwere also brought up.\n\nPerhaps for these and possibly more reasons, 72 percent of all users indicated their current software\ncould use an upgrade while 19 percent prefer an entirely new system. Around 70 percent of users\nbelieve upgrading their existing system or installing a new system would cost less than a million\nwhile 22 percent believe the cost would be above two million dollars. Interestingly, the amount that\nthe users indicated their respective institution could allocate toward such effort does not deviate too\nmuch with the cost. 80 percent of respondents believe their institution could allocate up to one\nmillion and 15 percent\n\nMention: \"World Bank Survey on Debt Management Systems\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"User satisfaction with national debt management systems is reported here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007433:50:0", "occurrence_id": "a4cdda0de4af8587a312f7d7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: beginning until the end of the period, it cannot capture the fast rise in _TFP_ in the\n\nrst decade of the acceleration nor it can it explain the decline thereafter. As said,\n\nthe smooth impulse implied by the calibrated reform, leads to a more protracted\n\ngrowth in aggregate productivity and cannot generate contractions.\n\n\nFigure C.3: Development Dynamics: Chile's Growth Acceleration 1984-2011\n\n\n\n|Col1|Col2|Col3|Col4|Col5|TFP|Col7|Col8|Col9|\n|---|---|---|---|---|---|---|---|---|\n||||||||||\n|||Mo<br>~~Da~~|del<br>~~ta~~||||||\n||||||||||\n||||||||||\n||||||||||\n||||||||||\n||||||||||\n||||||||||\n||||||||||\n||||||||||\n\n\n1985 1995 2005\n\n\nYear\n\n\n**Employment Share Innovation**\n\n\n0 5 10 15 20 25 30\n\n\nYear\n\n\n\n**Investment Rate**\n\n\n1985 1995 2005\n\n\nYear\n\n\n**Av Size**\n\n\n1985 1995 2005\n\n\nYear\n\n\n\nNote: The data for _TFP_ is constructed from the Penn World Tables Database (Zeileis 2021). We construct\n_TFP_ using _rgdpna_ as the measure of real _GDP,_ the product of the population and the human capital index\n( _pop∗hc_ ) as the measure of the labor input, and _rkna_ as the measure of the capital stock. We\n\nMention: \"Penn World Tables Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Real output and inputs from this database construct the productivity measure.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007480:5:0", "occurrence_id": "ed16cc0686a8793540d72f26", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: perceive that in utility or welfare metrics, its advantages outweigh its drawbacks\n\n\n(Binswanger, Deininger, and Fe 1995). Yet to our knowledge, this has received\n\n\nlittle attention in the literature.\n\n\nIn this article, we contribute to the literature on the economic consequences\n\n\nof land fragmentation, taking these three fundamental issues seriously. Building\n\n\non insights found in Ali, Deininger, and Ronchi (2018) and Veljanoska (2016),\n\n\nwe address issues of endogeneity by exploiting a unique natural experiment,\n\n\nEthiopia’s history of land reform and allocation which we argue represents an\n\n\nexogenous source of household-level land fragmentation. Like Ali, Deininger,\n\n\nand Ronchi (2018), we are able to construct multiple complementary fragmen\n\ntation measures, reflecting number of parcels, size and geographical dispersion.\n\n\nUnlike most studies, however, our data source, Ethiopia’s Living Standards and\n\n\nMeasurement Study-Integrated Survey on Agriculture (LSMS-ISA), contains in\n\nformation on household food security outcomes allowing us to assess the impact\n\n\nof land fragmentation on welfare metrics in terms of food security.\n\n\nWe find that in Ethiopia, land fragmentation reduces food insecurity. To\n\n\nunderstand why, the paper unpacks the risk diversification mechanism. Consis\n\ntent with Shaw’s (1963) conjecture, land fragmentation mitigates the adverse\n\n\neffects of low rainfall on food security. Households with diverse parcel charac\n\nteristics in terms of slope, elevation and wetness can grow a diverse set of crops.\n\n\nBy allowing farmers to create a diverse crop portfolio, land fragmentation helps\n\n\nmitigate weather risks, improving food security outcomes.\n\n##### **The evolution of access to land in Ethiopia**\n\n\nOur identification strategy hinges on our assertion that land access and frag\n\nmentation are exogenous to farmer ability. Justifying this claim first requires\n\n\nunderstanding the evolution of access to land in the diverse regions of Ethiopia.\n\n\n3\n\nMention: \"Living Standards and\n\n\nMeasurement Study-Integrated Survey on Agriculture\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"It carries household food security outcomes alongside plot information.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007484:21:1", "occurrence_id": "ad8570b76d338c15d28593c8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  grant) becomes available, if firms are not credit constrained, they should always use the cheaper credit\nto pay o↵more expensive loans. In fact, they should draw down the expensive\nloans to zero if credit is freely available. In Appendix Table E1, we examine\ndata on borrowing for school and household accounts of school owner households.\nWhile there is limited borrowing for investing in the school, over 20% of school\nowner households do borrow (presumably for personal reasons). Yet, we find no\nstatistically significant declines in borrowing at the school or household level as\na result of our intervention.\n\nWe now consider whether these expenditure changes a↵ected school revenues.\nSince schools may not always be able to fully collect fees from students, we use two\nrevenue measures: (i) posted revenues based on posted fees and enrollment (cols\n2-4), calculated as the sum of revenues expected from each grade as given by the\ngrade-specific monthly tuition fee multiplied by the grade-level enrollment; and\n(ii) collected revenues as reported by the school (cols 5-7). <sup>26</sup> To obtain the latter\nmeasure, we inspected the school account books and computed revenues actually\ncollected in the month prior to the survey. <sup>27</sup> While this measure captures revenue\n\n\n26Posted revenues are available for rounds 1,2, and 4, and collected revenues are available from rounds\n2-5. We use baseline posted revenues as the control variable in all revenue regressions.\n27Over 90% of schools have registers for fee payment collection, and for the remainder, we record\n\nMention: \"school account books\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Inspected directly to compute revenues actually collected before the survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007487:6:0", "occurrence_id": "49f364f617fa28d740651817", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <mark>up to their bound levels and can take another member to dispute settlement only when it increases its</mark>\n\n\n<mark>applied tariff above the bound level.</mark> The size of water measures the possibility of a country to freely\n\n\nincrease its applied MFN tariffs up to the bound rate without incurring into a dispute at the WTO. The\n\n\nsimple presence of water makes trade policy less predictable and therefore more uncertain.\n\n\nIn our analysis we take commitments under PTAs into account by setting the bound rate equal to the\n\n\npreferential tariff for those country pairs that have signed a preferential agreement. Specifically, the\n\n\nlevel of water is equal to zero for country pairs belonging to the same PTA unless preferential rates are\n\nhigher than MFN. <sup>6</sup> [^6: The number of cases in which preferential rates are above MFN rates is very small.] Algebraically, we define water in product - for a country pair �� as follows:\n\n\n�������� �� <sup>�������� �������</sup> <sup>������, 0�        ∀��∈���</sup>\n����� ������� ������,             ��������� <sup>(1)</sup>\n\n\nData on MFN applied rates and WTO bound rates are obtained from Groppo and Piermartini (2014).\n\n\nTheir database combines information on MFN applied tariffs from the WTO's Integrated Data Base\n\n(IDB) and UNCTAD's Trade Analysis and Information System (TRAINS). <sup>7</sup> [^7: The authors use TRAINS as the primary source for tariff data and IDB to fill the missing values.] The latter database is also\n\n\nused to extract data on effectively applied tariffs for country pairs belonging to a PTA\n\nMention: \"Integrated Data Base\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Applied most-favoured tariff lines come from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007487:6:0", "occurrence_id": "968f7618c605e1d0c3a61e99", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <mark>up to their bound levels and can take another member to dispute settlement only when it increases its</mark>\n\n\n<mark>applied tariff above the bound level.</mark> The size of water measures the possibility of a country to freely\n\n\nincrease its applied MFN tariffs up to the bound rate without incurring into a dispute at the WTO. The\n\n\nsimple presence of water makes trade policy less predictable and therefore more uncertain.\n\n\nIn our analysis we take commitments under PTAs into account by setting the bound rate equal to the\n\n\npreferential tariff for those country pairs that have signed a preferential agreement. Specifically, the\n\n\nlevel of water is equal to zero for country pairs belonging to the same PTA unless preferential rates are\n\nhigher than MFN. <sup>6</sup> [^6: The number of cases in which preferential rates are above MFN rates is very small.] Algebraically, we define water in product - for a country pair �� as follows:\n\n\n�������� �� <sup>�������� �������</sup> <sup>������, 0�        ∀��∈���</sup>\n����� ������� ������,             ��������� <sup>(1)</sup>\n\n\nData on MFN applied rates and WTO bound rates are obtained from Groppo and Piermartini (2014).\n\n\nTheir database combines information on MFN applied tariffs from the WTO's Integrated Data Base\n\n(IDB) and UNCTAD's Trade Analysis and Information System (TRAINS). <sup>7</sup> [^7: The authors use TRAINS as the primary source for tariff data and IDB to fill the missing values.] The latter database is also\n\n\nused to extract data on effectively applied tariffs for country pairs belonging to a PTA\n\nMention: \"Trade Analysis and Information System\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The primary source for the tariff information in the combined database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007488:23:0", "occurrence_id": "90eba1127fe002caa1d467ed", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 8: Official National Poverty Ratio, GDP per capita and middle**\n**class size between 2003 and 2015.**\n\n\nSource: World Bank, World Development Indicators.\n\n\nWas economic growth the only source of the middle-class expansion, as it\n\n\nseems from the graph? To answer this question, we decomposed the changes in the\n\n\nsize of the middle class into a “location” effect and a “shape” effect, using a\n\n\nnonparametric version <sup>11</sup> of the methodology based on Datt and Ravaillon’s (1992)\n\n\ndecomposition. The location effect measures the change in the middle-class share\n\n\nthat can be attributed to balanced growth, corresponding to an equal relative\n\n\nincrease of each household income. This growth, equal to an increase in the location\n\n\nparameter of the distribution (the mean or the median income), generates a shift in\n\nthe income density in a distributionally neutral fashion. The shape effect refers to\n\n\nthe change in the middle class attributable to changes in the income curve holding\n\n\nthe mean/median constant. In other words, this is the change that would have\n\n\noccurred if only the observed change in the shape of the income distribution had\n\n\noccurred without any shift in the mean/median of the curve. Therefore, this effect\n\n\n11 The method is presented in Massari et al. (2009).\n\n22\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Output per head and middle-class size across this window come from here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007496:9:1", "occurrence_id": "8eb4f38d91a81765a190da66", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  B of Table 1 also reports account usage six months\n\n\nafter the 2015 transfer, using administrative data. We find that individuals who received\n\n\nlarge transfers (either in cash or via direct deposit) made significantly more transactions (p\n\nvalue is 0.022). Although the overall mean is low, individuals who received the large transfer\n\n\nwere also more likely to correctly report the fees for the basic account and to trust NBS\n\n\nmore. All these variables were expected to be affected by treatment. While other variables\n\nin panel B could also have changed with the transfer, we find that they did not: excluding\n\n\nthe percentage of individuals who correctly predicted the fees in the basic NBS account, the\n\n\nnumber of transactions in the six and 12 months after the large transfer, and the percentage\n\n\nwho trust NBS, the p-value of the F-test that the remaining characteristics in panel B are\n\n\njointly zero is 0.419. In particular, while households that received the transfer (particularly in\n\n\ncash) seem to have larger balances in 2016 than those without the transfer, these differences\n\n\nare not statistically significant (p-value is 0.375). The structural model described in the next\n\n\nsection uses balance data to explain the decision of account ownership and for this reason it\n\n\ncan only be estimated with individuals with available balance data. Although we requested\n\n\nbalance data for the full sample of 747 individuals with accounts, the bank only provided\n\n\n8\n\nMention: \"administrative data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Account usage six months after the transfer is observed here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007497:46:1", "occurrence_id": "84eb30ae63b0f321980510a1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ) data or on World Bank’s WDI CPI\n\n\nseries.\n\n\n_Computing the incidence of catastrophic and impoverishing expenditures_\n\n\nThe incidence rates of catastrophic and impoverishing health expenditures are computed\n\n\nusing (the code underlying) the Stata module FPRO (Eozenou and Wagstaff 2018). Where health\n\n\nexpenditure data are individual-level, the data are aggregated to the household level. A household is\n\n\ndefined as incurring catastrophic expenses if its out-of-pocket health expenditures strictly exceed the\n\n\n47 PovcalNet can be accessed at http://iresearch.worldbank.org/PovcalNet/povOnDemand.aspx.\n\nMention: \"WDI\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its consumer price series deflates the health expenditure.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007497:46:1", "occurrence_id": "42f1759d56c173caf56787e4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ) data or on World Bank’s WDI CPI\n\n\nseries.\n\n\n_Computing the incidence of catastrophic and impoverishing expenditures_\n\n\nThe incidence rates of catastrophic and impoverishing health expenditures are computed\n\n\nusing (the code underlying) the Stata module FPRO (Eozenou and Wagstaff 2018). Where health\n\n\nexpenditure data are individual-level, the data are aggregated to the household level. A household is\n\n\ndefined as incurring catastrophic expenses if its out-of-pocket health expenditures strictly exceed the\n\n\n47 PovcalNet can be accessed at http://iresearch.worldbank.org/PovcalNet/povOnDemand.aspx.\n\nMention: \"PovcalNet\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The threshold for catastrophic expense is drawn from this tool.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007500:8:1", "occurrence_id": "e7b70d0143dbc9580682667f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  household resources increase.\nWhile a quarter of the poorest households consume bidis, this is only true of 6 percent of the richest\nhouseholds. For smokeless products, consumption is similar, except among the top deciles. This is\nconsistent with other studies describing the socioeconomic gradient in tobacco use; the gradient is not\nseen in cigarette consumption, but is noticeable in the consumption of bidis and smokeless products\n(Barkat et al. 2012). On average, the consumption of smokeless products accounts for the largest\nshare, 1.4 percent of the total household budget. Cigarettes follow at an average expenditure share of\n1.3 percent. The cigarette budget share is even across the distribution.\n\nThe median quantity of cigarettes consumed was 152 a month per adult man age 15 years or older;\nthe corresponding number of bidis was 348 per month per adult man (figure 5). <sup>9</sup> Assuming daily\nsmoking, this is equivalent to an average of 5 cigarettes and 12 bidis per adult man per day. These\nhousehold figures are consistent with data on individuals from the Global Adult Tobacco Survey\namong adult men (Barkat et al. 2012). The poorest decile consumed about 20 fewer cigarettes per\nadult man relative to the richest decile (132 compared with 152 cigarettes per adult man per month).\nHowever, among deciles 4–10, the median quantities were similar. One may see the opposite pattern\nin the case of bidis, that is, the richest consume slightly fewer bidis.\n\n9 The data do not allow the identification of the household members who are smoking; so, these results should\nbe interpreted with care.\n\n\n7\n\nMention: \"Global Adult Tobacco Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Individual figures for adult men are consistent with the household totals.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007481:27:1", "occurrence_id": "b6c3910278d2dd854623cac8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: </u>\nn 3,756 3,757 3,758 3,754 3,758 3,440 3,758 3,732\nAdj. _R_ <sup>2</sup> 0.79 0.744 0.689 0.675 0.717 0.68 0.668 0.687\n\n\na _∗∗∗p <_ 0 _._ 01 _, ∗∗_ _p <_ 0 _._ 05 _, ∗p <_ 0 _._ 1. In all columns, country effects have been controlled for and standard\nerrors have been clustered by country. In all columns, the dependent variable is the estimated location\npremium (measured in natural logs) from a series of country-level first-stage regressions after controlling\nfor observable worker characteristics within the broad sample (all wage/salary employees aged from 14 to\n65 years old) and survey-year fixed effects. Worker characteristics include age, age-squared, marital status,\ngender, and years of schooling. The source of the population data is the Gridded Population of the World\n(GPW), v4.\n\n\nsmaller than for Private -v- Public Sector and Formal -v- Informal workers, differences can also be\n\n\nobserved in the estimated elasticity of the city premium with respect to average years of schooling\n\n\nfor Young -v- Old and Male -v- Female workers. Old workers appear to experience stronger human\n\n\ncapital externalities than Young workers, while Male workers benefit from stronger human capital\n\n\nexternalities than Female workers.\n\n\nIn the case of market access, the estimated elasticities are close to zero for public sector and\n\n\ninformal workers, and statistically insignificant. Market access benefits more the productivity of\n\n\nworkers in Formal and Private firms. This would be consistent with these types of firms selling to\n\n\nother regions in the country more and benefitting more from cheaper inputs brought from elsewhere.\n\n\nMarket access also has an insignificant impact for both Old and Female\n\nMention: \"Gridded Population of the World\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Population by marital status, gender and schooling on a grid, version four.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007493:10:1", "occurrence_id": "71d415d47072ac3415598314", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  to bequeath,\nsell, rent out, use as collateral and invest in the asset (Kilic and Moylan 2016; Doss et al 2017). Third, the\nunderlying data are collected by interviewing multiple respondents per household (typically one adult male\nand one adult female). As we will discuss further in the next section, this approach of soliciting data on\nasset ownership, while operationally demanding, avoids potential biases from the use of proxy respondents\n(Doss et al 2008, 2017; Kilic and Moylan 2016).\n\n\nThe Gender Land and Asset Survey (Jacobs et al 2011; Kes et al 2011) was conducted in three locations\nin South Africa and Uganda in 2009. Across all three sites, women are found to be less likely to claim\nownership of land and housing than men. Joint ownership of immovable assets (land, housing) is prevalent\nin all study locations, particularly for women, but does not always carry significant advantages (compared\nwith non-ownership) in terms of decision-making (Jacobs and Kes 2015). The study further documents that\nreported ownership (based on an individual’s perception) often does not go in hand with formal\ndocumentation. It should be noted though that due to the limited geographic coverage of the data, the\nestimates are not nationally representative.\n\n\nThe Gender Asset Gap Project collected nationally (or, in one case, state-level) representative data on\na broad range of assets in Ghana, Ecuador and Karnataka, India, in 2010. As shown in Deere et al (2013),\nmarried women hold 19 percent of couple wealth in Ghana, which is less than in Ecuador but more than in\nKarnataka, India. These gender wealth gaps are largely driven by immovable assets – agricultural land, the\nprincipal residence and other real estate, which in Ghana are often owned individually by men. This\nreinforces the notion that land and housing property are an important driver of the overall distribution of\nwealth in developing countries\n\nMention: \"Gender Land and Asset Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Run in three locations; not nationally representative.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007493:10:1", "occurrence_id": "cd16a317c68af5deb0dd74ec", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  to bequeath,\nsell, rent out, use as collateral and invest in the asset (Kilic and Moylan 2016; Doss et al 2017). Third, the\nunderlying data are collected by interviewing multiple respondents per household (typically one adult male\nand one adult female). As we will discuss further in the next section, this approach of soliciting data on\nasset ownership, while operationally demanding, avoids potential biases from the use of proxy respondents\n(Doss et al 2008, 2017; Kilic and Moylan 2016).\n\n\nThe Gender Land and Asset Survey (Jacobs et al 2011; Kes et al 2011) was conducted in three locations\nin South Africa and Uganda in 2009. Across all three sites, women are found to be less likely to claim\nownership of land and housing than men. Joint ownership of immovable assets (land, housing) is prevalent\nin all study locations, particularly for women, but does not always carry significant advantages (compared\nwith non-ownership) in terms of decision-making (Jacobs and Kes 2015). The study further documents that\nreported ownership (based on an individual’s perception) often does not go in hand with formal\ndocumentation. It should be noted though that due to the limited geographic coverage of the data, the\nestimates are not nationally representative.\n\n\nThe Gender Asset Gap Project collected nationally (or, in one case, state-level) representative data on\na broad range of assets in Ghana, Ecuador and Karnataka, India, in 2010. As shown in Deere et al (2013),\nmarried women hold 19 percent of couple wealth in Ghana, which is less than in Ecuador but more than in\nKarnataka, India. These gender wealth gaps are largely driven by immovable assets – agricultural land, the\nprincipal residence and other real estate, which in Ghana are often owned individually by men. This\nreinforces the notion that land and housing property are an important driver of the overall distribution of\nwealth in developing countries\n\nMention: \"Gender Asset Gap Project\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Nationally representative data on a broad range of assets.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007509:20:0", "occurrence_id": "ad89465d8d81b761d34c1d2e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The fact that the units of the assessment of learning matter is not in and of itself a\n\n\nproblem for the _LAYS_ approach. It does mean, however, that _LAYS_ should be thought of not just\n\n\n“average years of schooling measured in terms of the learning productivity of the top performer”\n\n\nbut as “average years of schooling measured in terms of the productivity of the top performer\n\n\n_according to the metric used to determine that productivity_ .” But given that the alternative unit\n\nfree measure yields results in line with the basic _LAYS_ results, in practical terms that distinction\n\n\nmay not be as significant as it first seems.\n\n\n**_4.2 The assessment used: Using PISA instead of TIMSS_**\n\n\nTIMSS is not the only assessment that could be used to calculate _LAYS_ . PISA is another\n\n\ninternational assessment with wide coverage: in 2015, the PISA assessment covered 72\n\n\ncountries and economies (35 of which are in the OECD). PISA is an assessment of 15-year-olds\n\n\nin secondary school. Since these students are not necessarily all in the same grade, here we take\n\n\na slightly different approach to calculating _LAYS_, using age rather than grade when we define the\n\n\naverage learning profile. So, for example, we now divide the score on the assessment by 9 (in\n\n\nthe equivalent of Equation 3), since this is the number of years of learning that a student who\n\n\nstarted learning at age 6 would have acquired by age 15. For alternative calculations that allow\n\n\nfor learning to start before age 6, we could then proceed as before (that is, as per Equation 4) and\n\n\nadd in additional years of learning. Implementing _LAYS_ in this way using PISA 2015\n\n\nmathematics scores yields the _LAYS_ estimates shown in Figure 4.2, which are generally speaking\n\n\nin line with the results that use only TIMSS.\n\n\nwas already large using either the average or the “\n\nMention: \"PISA 2015\n\n\nmathematics scores\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Its mathematics results give the learning-adjusted years measure.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007511:43:1", "occurrence_id": "457a2cbb696f152f26bb16d6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  regression of\n\n\nheight on adult survival results in a slope coefficient of 19 and a standard error of 3.6, which is\n\n\nextremely close to the Weil (2007) baseline estimate of 19.2. In Sub‐Saharan Africa, the slope\n\n\ncoefficient is close to zero and very imprecisely estimated. This likely reflects the large swings in adult\n\n\nmortality rates due to the AIDS epidemic. To assess this, the relationship is re‐estimated for all\n\n\ncountries, but excluding observations above the Sub‐Saharan Africa median for aids‐related death rates.\n\n\nThis gives a very similar estimate of 18.3 with a standard error of 3.4.  These estimates lie in the same\n\n\nvicinity as the baseline estimate in Weil (2007). This evidence suggests that the Weil (2007) estimate of\n\n𝛽������,��� �19.2 is reasonable to use in the baseline “return” to ASR of 𝛾��� �𝛾������ \n𝛽������,��� �0.034 �19.2 �0.65 .\n\n\n22 All of the discussion here relies on the literature on the returns to height. While less extensive, there are well‐\nidentified econometric estimates of the economic returns to a number of other measures of health, including iron\ndeficiency anemia, malaria, and hookworm infestation. Subsequent updates of this background paper will\ninvestigate how the estimates of overall health to productivity change when these other estimates of individual\nhealth conditions and their returns are used in this framework.\n23 The data come from the Health Equity and Financial Protection (HEFPI) Project at the World Bank. Patrick\nEozenou and Adam Wagstaff kindly made this data available.\n\n\n42\n\nMention: \"Health Equity and Financial Protection (HEFPI) Project\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its household data carry the catastrophic expenditure framework.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007517:46:0", "occurrence_id": "49eca6c25ecefbec3c318017", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Animal / veg\noil / fat / wax\n\n\n\nCrude materials\n\nexcluding fuel\n\n\n\nMachinery /\n\nTransport\nEquipment\n\n\n\nChemicals /\nProducts n.e.s.\n\n\n\nManufactured\n\ngoods\n\n\n\nBeverages and\n\ntobacco\n\n\n\nFood and Live\n\nAnimals\n\n\n\nMiscellaneous\nManuf. Articles\n\n\n\nCommodities\n\nn.e.s.\n\n\n\n20\n\n\n10\n\n\n0\n\n\n-10\n\n\n-20\n\n\n-30\n\n\n-40\n\n\n-50\n\n\n-60\n\n\n-70\n\n\n\nMineral fuel /\n\nlubricants\n\n\n\n\n\nNote: Data is from UN Comtrade database. Export growth is in nominal USD. Years are calendar years. The\nbars denote the difference between the average growth rate between 2003 and 2008 and the average growth\nrate between 2012 and 2016.\n\n\n  - India’s export destinations are diversified too. The largest share of exports from India is\n\ndestined for the Middle East (approximately 20 percent in 2016), and within it the largest share\nis exported to the United Arab Emirates. The United States is the second largest destination,\naccounting for 16 percent of India’s exports, followed by China (including Hong Kong SAR,\nChina), Sub-Saharan Africa and Europe. We disaggregate Indian exports across its 8 main\ntrading partners and the rest of the world for 2003 to 2017. <sup>48</sup> We note that the export\nslowdown experienced by India after 2013 was across destinations, as exports slowed to most\nof India’s main trading partners in the Middle East, the UAE and Saudi Arabia, but also to the\nUS, China and Singapore.\n\n\n**Figure 28: Destinations of Indian Exports**\n\n28 A: Contribution to Export Growth 28 B: Contribution to Export Growth: Middle\nEast and the USA\n\n\n48 India’s main trading partners, in order of share in total exports, are the USA, the United Arab Emirates, Hong Kong\nSAR, China, China, the United Kingdom, Singapore, Germany, and Saudi Arabia.\n\n\n45\n\nMention: \"UN Comtrade database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Nominal export growth by product group comes from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007518:22:1", "occurrence_id": "b1c3586b35eb5d75d59e2a5b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  by improving\nallocative efficiency, enhancing productive efficiency, and boosting innovation. Aghion, Akcigit\nand Howitt (2014) also provide a compelling framework linking market competition to\nproductivity growth through creative destruction, while Conway et al (2006) and Alesina et al.\n(2005) trace the positive impact of pro-competition product market regulation on productivity, and\nBarone and Cingano (2011) find empirical evidence that pro-competition reforms in input sectors\nbenefit productivity of the overall economy. Furthermore, the introduction of anti-cartel policy is\nalso related to higher labor productivity growth. Symeonidis (2008) records a 20 to 30 p.p. lower\nlabor productivity growth in industries with cartels than in industries without cartels. Some\ngovernment interventions unduly inhibit firm’s ability or incentives to compete. The OECD\nProduct Market Regulation (PMR) indicator measures the degree to which policies or regulatory\nmeasures promote or inhibit competition in areas of the product market where competition is\nviable. PMR data for Brazil in 2013 suggests that policies and regulations in key sectors are less\nconducive to competition than in comparator countries.\n\n\nWhile other BRICS countries have promoted reforms over the last decade, the overall\nrestrictiveness of product market regulation in Brazil remained unchanged between 2008 and 2013.\nThe general degree of restrictiveness of product market regulations in Brazil has not changed since\nbetween 2008 and 2013 and compares less favorably to other economies which on average have\nmoved towards a more pro-competitive regulatory framework (Figure 24). In fact, while others\n\n37 _Competition policy_ encompasses a set of policies and laws that ensure competition in the marketplace is not\nrestricted in such a way as to reduce economic welfare (Motta, 2004).\n38 For more details on what competition policy encompasses, see Kitzmueller and Licetti (2012).\n\n\n21\n\nMention: \"PMR data for Brazil in 2013\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its sector regulation scores show policies less conducive to competition.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007520:37:1", "occurrence_id": "f024ed5fe02a643334ad0bff", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  on a\nsophisticated approach to assessing road project lifecycle costs, namely the Highway\nDevelopment and Management (HDM-III/4) tool.\n\n\nThe latter involves consideration of Chile’s transport policy goals. Road density in Chile lags behind\nother Latin America and OECD countries, partly due to low overall population density but high\nconcentration in Santiago and Valparaiso. The road network’s size (approximately 78,000 kms in\n2010) has not significantly changed since 1990, though the share of paved roads has increased\nfrom 13.8 to 23.3 percent. This pilot application of IPF considers inter-urban road projects\nintended to extend the road network and link urban areas. Specifically, the pilot was applied to 50\nprojects in the BIP that met basic requirements (mentioned below) to be included in the IPF pilot,\nincluding HDM-3 data and the existence of a CBA. <sup>15</sup> Road repositioning projects were excluded\nfrom the pilot.\n\n\nRoad Transport Project Sample\n\n\nThe sample of 50 interurban road transport projects to which IPF was applied includes only a set\nthat has reached Execution status in the SNI. Additionally, this set has passed the basic\nprofitability test of having IRRs of at least 6% and have been judged to align with transport policy\ngoals.\n\n\n14 The Organic Law on Financial Administration of the State, Article 19, requires only that a project’s profitability be\nanalyzed, but does not require project profitability. The Organic Constitutional Law on Government and Regional\nAdministration (19175), Article 75 states that projects financed with regional funds require an RS RATE.\n15 Special classifications exempt some projects from requiring CBA (e.g., projects in extreme zones, projects with\nexpected volumes of less than 400 vehicles per day).\n\n\n36\n\nMention: \"HDM-3 data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Required for inclusion in the pilot alongside a cost-benefit analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007522:24:5", "occurrence_id": "a42cb009cde874bd8971f863", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: on the median values of𝐼𝑌<br>⁄  and𝐼<br>𝐼<br>⁄ . <br>|For a general description of the sources of data, see Table 3. The sample size for all variables is guided by the consistent availability for<br>individual countries of PWT 8.1 data for labor share, capital‐to‐output ratio, human capital growth and TFP growth. See Appendix 3, Table<br>A3.1 for median values of other key variables used in the simulations.<br>Countries are classified according to the 2018 World Bank classification of countries by income for the 2017 calendar year.<br>* Multiply by 100 (except values in ppts) to obtain parameter/variable values in percentage share or growth terms (%).<br>Note:<br>_(1) _For𝛽, individual country values below 0.45 were increased to 0.45 and those above 0.70 reduced to 0.70.<br>_(2) _For𝐾଴𝑌଴ <br>⁄<br>, individual country values below 1.5 were increased to 1.5 and those above 3.5 reduced to 3.5.<br>_(3)\n\nMention: \"PWT 8.1 data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its series on factor shares and productivity growth are selected for coverage.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007525:6:0", "occurrence_id": "c0fc277637b4d20891ffd301", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: host countries rises by 10 percent. In fact, the extent of gravity in FDI is comparable to the extent of\ngravity in trade that has been found in the literature (see, for example, Head and Mayer, 2013).\n\nYeaple (2003a) extends earlier work by exploring an interaction between country and industry\ndeterminants of FDI and offers empirical support to both market access and comparative advantage\nmotives. Specifically, he uses U.S. affiliate sales data in 39 countries and 50 manufacturing industries\nfrom the BEA Benchmark Survey of 1994 and finds that U.S. multinational firms from unskilledlabor intensive industries tend to invest in unskilled-labor abundant countries, a result consistent\nwith the hypothesis that countries' factor endowment differences lead to vertical FDI. Unlike in\nBrainard (1997), the role of transportation cost is found to be negative and statistically insignificant,\ndeparting from the expected sign in the context of horizontal FDI.\n\nIrarrazabal, Moxnes and Opromolla (2013) exploit the source of gravity in FDI by introducing intrafirm trade into the framework of Helpman, Melitz and Yeaple (2004) and generating gravity\nrelationships for both exports and multinational production. The model rationalizes gravity in FDI\nby assuming that the headquarters produce a firm-specific tangible or intangible input that is\nrequired for production in any location and trade in such input is subject to trade costs including\ntransportation costs or tariffs in the case of physical inputs and cultural and geographic remoteness\nfrom the headquarters in the case of headquarters services. The results suggest that intra-firm trade\nappears to play a crucial role in shaping the geography of multinational production; to justify the\ngravity observed, the affiliate's cost share related to input purchases from the headquarters must be\nabout 90 percent. As suggested in the paper, this result may have captured other mechanisms that\nare dampening firms' multinational production as trade costs increase, such as imperfect\ntransmission of technology between parents and affiliates either because of imperfect codifi\n\nMention: \"BEA Benchmark Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Affiliate sales across thirty-nine countries and fifty industries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007516:22:0", "occurrence_id": "4e4851984cea1c20b9d78415", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Source: Primary Research_\n\n**F.** **Business Environment: Doing Business and Global Rankings**\n\n\nDespite a thriving ecosystem of entrepreneurs and a private sector-led economy, Kenya has made a\nrelatively mediocre showing in global investment climate rankings. Specifically, as shown in Table 12\nbelow, the country has consistently ranked in the bottom half of both the World Bank’s Doing Business\nReport and the Heritage Foundation’s Economic Freedom Index. It has also ranked in the bottom quartile\nof Transparency International’s Corruption Perception Index.\n\n\nKenya's ranking in Transparency International’s Corruption Perception Index is worthy of note because it\nhighlights the kinds of environmental challenges facing enterprises in Kenya. Widespread corruption can\nact as a leach on an economy and the companies within it, and SMEs lack the resources and networks to\nnavigate the ecosystem of graft unscathed. Fund managers in Kenya actively screen for corrupt practices\nwhen conducting due diligence since irregularities exert both seen and unseen costs on businesses. Given\nthat a meaningful part of the value-add of private equity firms is their focus on corporate governance,\nanything that detracts from these efforts can negatively impact the value of an investee company. Moreover,\nproblems at investee companies can impair the operations and damage the reputation of the private equity\nfirms that are shareholders, either minority or majority, in these businesses.\n\n\n21\n\nMention: \"Doing Business\nReport\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"One country ranks consistently in the bottom half of this ranking.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007516:22:0", "occurrence_id": "4ed2c0ef496997f6af588480", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Source: Primary Research_\n\n**F.** **Business Environment: Doing Business and Global Rankings**\n\n\nDespite a thriving ecosystem of entrepreneurs and a private sector-led economy, Kenya has made a\nrelatively mediocre showing in global investment climate rankings. Specifically, as shown in Table 12\nbelow, the country has consistently ranked in the bottom half of both the World Bank’s Doing Business\nReport and the Heritage Foundation’s Economic Freedom Index. It has also ranked in the bottom quartile\nof Transparency International’s Corruption Perception Index.\n\n\nKenya's ranking in Transparency International’s Corruption Perception Index is worthy of note because it\nhighlights the kinds of environmental challenges facing enterprises in Kenya. Widespread corruption can\nact as a leach on an economy and the companies within it, and SMEs lack the resources and networks to\nnavigate the ecosystem of graft unscathed. Fund managers in Kenya actively screen for corrupt practices\nwhen conducting due diligence since irregularities exert both seen and unseen costs on businesses. Given\nthat a meaningful part of the value-add of private equity firms is their focus on corporate governance,\nanything that detracts from these efforts can negatively impact the value of an investee company. Moreover,\nproblems at investee companies can impair the operations and damage the reputation of the private equity\nfirms that are shareholders, either minority or majority, in these businesses.\n\n\n21\n\nMention: \"Economic Freedom Index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"A second ranking placing the same country low.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007516:22:0", "occurrence_id": "f8f290473c58392e44b3428e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Source: Primary Research_\n\n**F.** **Business Environment: Doing Business and Global Rankings**\n\n\nDespite a thriving ecosystem of entrepreneurs and a private sector-led economy, Kenya has made a\nrelatively mediocre showing in global investment climate rankings. Specifically, as shown in Table 12\nbelow, the country has consistently ranked in the bottom half of both the World Bank’s Doing Business\nReport and the Heritage Foundation’s Economic Freedom Index. It has also ranked in the bottom quartile\nof Transparency International’s Corruption Perception Index.\n\n\nKenya's ranking in Transparency International’s Corruption Perception Index is worthy of note because it\nhighlights the kinds of environmental challenges facing enterprises in Kenya. Widespread corruption can\nact as a leach on an economy and the companies within it, and SMEs lack the resources and networks to\nnavigate the ecosystem of graft unscathed. Fund managers in Kenya actively screen for corrupt practices\nwhen conducting due diligence since irregularities exert both seen and unseen costs on businesses. Given\nthat a meaningful part of the value-add of private equity firms is their focus on corporate governance,\nanything that detracts from these efforts can negatively impact the value of an investee company. Moreover,\nproblems at investee companies can impair the operations and damage the reputation of the private equity\nfirms that are shareholders, either minority or majority, in these businesses.\n\n\n21\n\nMention: \"Corruption Perception Index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its bottom quartile placement is the third comparison.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007530:36:3", "occurrence_id": "9938d9c1544912e875ef985e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  -0.374 -0.159 -2.162\n5 - richest -0.373 -0.199 -0.071 -0.643\n<u>Total</u> <u>-1.678</u> <u>-0.382</u> <u>-0.525</u> <u>-2.579</u>\n<u>Panel C: Share of maize, cassava and rice in food budget across income quintiles</u>\nQuintiles\nItem 1 2 3 4 5\nMaize 18.46 25.29 24.72 19.24 7.92\nCassava 13.75 9.46 6.21 3.38 1.25\n<u>Rice</u> <u>4.19</u> <u>4.90</u> <u>4.60</u> <u>4.33</u> <u>3.02</u>\n_Source_ : World Bank based on IOF-2014/15 and IAS-2015\n\n\n35\n\nMention: \"IOF-2014/15\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The income and expenditure round behind the reported figures.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007530:36:3", "occurrence_id": "c1cfcb9b49bc0e4ba8cd43d2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  -0.374 -0.159 -2.162\n5 - richest -0.373 -0.199 -0.071 -0.643\n<u>Total</u> <u>-1.678</u> <u>-0.382</u> <u>-0.525</u> <u>-2.579</u>\n<u>Panel C: Share of maize, cassava and rice in food budget across income quintiles</u>\nQuintiles\nItem 1 2 3 4 5\nMaize 18.46 25.29 24.72 19.24 7.92\nCassava 13.75 9.46 6.21 3.38 1.25\n<u>Rice</u> <u>4.19</u> <u>4.90</u> <u>4.60</u> <u>4.33</u> <u>3.02</u>\n_Source_ : World Bank based on IOF-2014/15 and IAS-2015\n\n\n35\n\nMention: \"IAS-2015\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The accompanying agricultural round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007533:7:0", "occurrence_id": "75356a461a8968d9bf7f63a6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: and academics in the relationship between concentration in the banking sector and financial\n\n\nstability, because market concentration may have played a role in the run-up to the crisis, and,\n\n\nmore importantly, has tended to increase ever since, raising stability concerns (Committee on the\n\n\nGlobal Financial System, 2018).\n\n\nOur empirical analysis rests on a data set comprising available cross-country, annual\n\n\nobservations for 68 countries over the period 1997-2015. Financial stability is proxied by the\n\n\nprobability of entering both a systemic and a non-systemic banking crisis. Data are drawn from an\n\n\nupdated version of the Reinhart and Rogoff (2009) dataset, which gives 42 crisis episodes during\n\n\nthe sample period. Crisis data are given by a binary dummy that equals 1 if country _i_ at time _t_ has\n\n\nentered a crisis, and zero otherwise. Concentration is measured by the share of banking assets held\n\n\nby the three largest banks. However, we also study the impact of alternative concentration\n\n\nmeasures, such as the share of banking system assets concentrated in the five largest banks. We\n\n\nproxy for the mediators through which concentration affects financial stability by using bank return\n\n\non assets (ROA) for _profitability_ ; real lending rate for the _cost of credit_ ; bank foreign assets plus\n\n\nbank foreign liabilities to banking system assets for _diversification_ ; and the (log) number of banks\n\n\nfor the _ease of monitoring._ Following Beck et al. (2006), we estimate a binomial logit model, and\n\n\ncontrol for the following variables: real GDP growth, inflation, nominal exchange rate\n\n\ndepreciation, current account balance, real domestic credit growth, and (log) real GDP per capita.\n\n\nIn order to disentangle the role of each mediator through which concentration is expected to impact\n\n\nfinancial stability, we estimate a model where proxies for the mediators are allowed to interact\n\n\nwith the level of concentration. This allows us to estimate the marginal effects of the mediators\n\n\nconditional upon selected\n\nMention: \"Reinhart and Rogoff (2009) dataset\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"An updated version gives the crisis episodes in the sample period.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007533:7:0", "occurrence_id": "eacd1ce70a22f5f2d0e417d8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: and academics in the relationship between concentration in the banking sector and financial\n\n\nstability, because market concentration may have played a role in the run-up to the crisis, and,\n\n\nmore importantly, has tended to increase ever since, raising stability concerns (Committee on the\n\n\nGlobal Financial System, 2018).\n\n\nOur empirical analysis rests on a data set comprising available cross-country, annual\n\n\nobservations for 68 countries over the period 1997-2015. Financial stability is proxied by the\n\n\nprobability of entering both a systemic and a non-systemic banking crisis. Data are drawn from an\n\n\nupdated version of the Reinhart and Rogoff (2009) dataset, which gives 42 crisis episodes during\n\n\nthe sample period. Crisis data are given by a binary dummy that equals 1 if country _i_ at time _t_ has\n\n\nentered a crisis, and zero otherwise. Concentration is measured by the share of banking assets held\n\n\nby the three largest banks. However, we also study the impact of alternative concentration\n\n\nmeasures, such as the share of banking system assets concentrated in the five largest banks. We\n\n\nproxy for the mediators through which concentration affects financial stability by using bank return\n\n\non assets (ROA) for _profitability_ ; real lending rate for the _cost of credit_ ; bank foreign assets plus\n\n\nbank foreign liabilities to banking system assets for _diversification_ ; and the (log) number of banks\n\n\nfor the _ease of monitoring._ Following Beck et al. (2006), we estimate a binomial logit model, and\n\n\ncontrol for the following variables: real GDP growth, inflation, nominal exchange rate\n\n\ndepreciation, current account balance, real domestic credit growth, and (log) real GDP per capita.\n\n\nIn order to disentangle the role of each mediator through which concentration is expected to impact\n\n\nfinancial stability, we estimate a model where proxies for the mediators are allowed to interact\n\n\nwith the level of concentration. This allows us to estimate the marginal effects of the mediators\n\n\nconditional upon selected\n\nMention: \"Crisis data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"A binary flag for entry into crisis in each country-year.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007535:4:1", "occurrence_id": "2caa6cf74914dde327d837c6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  (AUM) totaled US$3.8 trillion in June 2014, with US$1.8\ntrillion in dry powder <sup>5</sup> [^5: Capital committed but not yet paid-in and deployed.] as of December 2014. <sup>6</sup> [^6: Preqin 2015.] Almost 60 percent of PE/VC AUM was accounted for by\nNorth America; while Europe comprised almost a quarter and Asia 13 percent of total AUM. In 2014 PE/VC\nfunds raised a total of US$495 billion in capital, down 6 percent from the $528 billion raised in 2013.\nAccording to data from Preqin, global fund raising for private equity in 2015 is expected to jump, projected\nto surpass US$790 billion. Investment deal flow remained active in 2014 with 5,305 completed PE deals\ntotaling US$996 billion of capital invested, which marked an 8-year high, according to Pitchbook. <sup>7</sup> [^7: Source: Pitchbook, http://pitchbook.com/pr_20150109_2.html.] Venture\ncapital comprised US$86 billion in deal value in 2014; while buyout activity was US$332 billion. <sup>8</sup> [^8: Preqin 2015.] 2014\nsaw the recent trend of increasing numbers of firms focusing on minority and growth investments, add-on\ninvestments, and liquidity continue, with US$445 billion in capital exited from such investments, up 46\npercent from 2013’s US$305 billion. <sup>9</sup> [^9: Source: Pitchbook.]\n\n\n2 This section draws from “Private Equity and Venture Capital in SMEs in Developing Countries: The Role for Technical\nAssistance,” World Bank Policy Research Working Paper 6827 co-authored by Shanthi Divakaran, Patrick McGinnis, and\nMasood Shariff. It also draws from Private Equity: Fund Types, Risk and Returns, and Regulation. Douglas Cumming, 2010.\n3 See\n\nMention: \"Preqin\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Fund raising totals and projections for private equity come from this vendor.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007539:5:0", "occurrence_id": "009a41f5b457d18bfda16bfa", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: following unexpected contingencies and thus bail out efficient exporters that experience\n\n\nfinancial difficulties (Huberman and Kahn, 1988). While differentiating these two channels is\n\n\nnotoriously hard, in the second part of the paper we try to gain intuition into the mechanisms\n\n\nthat drive the baseline results.\n\n\nIn a first set of tests, we build on the prediction of the banking literature that banks’\n\n\nsoft budget constraints are more severe in institutional settings in which banks are more\n\n\nsusceptible to the pressure of external parties (e.g., the government) to rollover credit to\n\n\nbusinesses, and less subject to independent monitoring of bank auditors and supervisors\n\n\non their decisions to grant such rollover. Matching information from our main databases\n\n\nwith data from the World Bank Banking Supervision Survey, we obtain that in countries\n\n\nwhere banks are less independent from the pressures of external parties and where bank\n\n\nmanagers are subject to less rigorous scrutiny of bank auditors and supervisors that prevent\n\n\nthe release of misleading information, domestic banks reduce the dynamism of the export\n\n\nsector relative to stock market investors and to foreign banks (i.e., they reduce exporters’\n\n\nentry, exit and turnover, in line with the soft budget constraint hypothesis).\n\n\nIn a second set of tests, we exploit the prediction of the banking literature that, to\n\n\nthe extent that banks’ behavior instead reflects long-termism, this benefit should manifest\n\n\nitself especially following negative shocks (e.g., crises or recessions) or in settings where\n\n\nex ante it is harder to write detailed contracts that can account for unexpected shocks.\n\n\nHowever, we uncover no evidence that domestic banks moderate the exit of exporters more\n\n\nduring recessions or financial crises than in tranquil times. Moreover, the differential effect\n\n\nof foreign banks is no more pronounced for products characterized by higher informational\n\n\nopacity or contractual complexity (measured according to the methodologies in Rauch,\n\n\n1999, and Nunn, 2007). Overall these findings cast doubts on the benign interpretation that\n\n\ndomestic banks are a buffer against the realization of unforeseen shocks over the\n\nMention: \"World Bank Banking Supervision Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Matched to the main databases to measure supervisory independence.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007541:27:0", "occurrence_id": "766845edf8c346c56969bb71", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 26\n\n\n**<u>Table 6. Simulation Results – Effect of Treaty-Shopping</u>**\n\nColumn (1) (2) (3)\n\nNo DTT with\nSubsample All countries DTT with investment hub\n<u>investment hub</u>\nDiversion (in%) 36.00 92.71 30.20\nFDI Gain (in%) 0.96 2.90 0.76\nCIT Revenue Loss (in%) 5.13 14.48 4.17\nNotes: Based on the conceptual framework and estimation results of Table 5, this Table presents simulation\n<u>results in percent, differentiating between three country groups.</u>\n\n\n**VI.  CONCLUSION**\n\n\nWe analyze the cost and benefits of concluding Double Tax Treaties (DTTs) with investment\nhubs on FDI and government revenues for a sample of African economies. Our results imply\nthat these economies are likely facing a substantial reduction in revenue when signing a\ntreaty with investment hubs. At the same time, we find no evidence substantiating\nexpectations of a corresponding benefit in terms of additional investment generation. The\nvalue of tax treaties to attract investors thus remains uncertain at best, especially where\ninvestments tend to be aimed at a country’s natural resources or specific market.\n\nFurther analysis is needed to assess the large magnitudes of treaty related base erosion and\nprofit shifting in Sub-Saharan Africa identified in our analysis. While micro-level firm level\ninformation on MNEs remains very limited in commercial databases for many developing\ncountries, including most economies in Sub-Saharan Africa (Platform for Collaboration on\nTax 2017), future research drawing on micro level administrative information, including\nCountry-by-Country reports of large MNEs, could potentially help fill this gap\n\nTo assess the robustness of our findings and evaluate their potential underlying drivers, we\nexamine the role of treaty shopping in driving nominal investment flows and provide initial\nempirical evidence on the importance of treaty-shopping in the region\n\nMention: \"Country-by-Country reports of large MNEs\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Future research could use these micro-level administrative reports.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007544:26:2", "occurrence_id": "43147bd36ebdc005c75c524f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .12 0.08 0.04\nUpper-bound elasticity 1.24 0.80 0.55 0.44 0.39 0.39 0.30 0.25 0.18 0.09\n\n_Source:_ Estimates based on Russia Longitudinal Monitoring Survey–HSE 2016.\n_Note:_ The table reports the share of total consumption for each decile.\n\n\n**Table D.4 – Years of Working Life Lost (%), by Decile**\n\n\n_Price shock scenario_ _1_ _2_ _3_ _4_ _5_ _6_ _7_ _8_ _9_ _10_\nLow-bound elasticity 0.03 0.01 0.00 -0.01 -0.02 -0.02 -0.02 -0.03 -0.03 -0.04\nMedium elasticity 0.20 0.21 0.17 0.17 0.17 0.21 0.19 0.20 0.18 0.20\n<u>Upper-bound elasticity</u> <u>0.38</u> <u>0.41</u> <u>0.35</u> <u>0.35</u> <u>0.36</u> <u>0.43</u> <u>0.41</u> <u>0.42</u> <u>0.40</u> <u>0.45</u>\n\n_Source:_ Estimates based on Russia Longitudinal Monitoring Survey–HSE 2016.\n_Note:_ The table reports the share of total individual consumption for each decile. Years of life lost have been estimated using all\ndeaths from tobacco-related diseases.\n\n\n**Table D.5 - Net Effect on Household Expenditures (%), by Decile*\n\nMention: \"Russia Longitudinal Monitoring Survey–HSE 2016\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Decile shares of total consumption are estimated from this wave.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007545:40:1", "occurrence_id": "19df9d2720ae48831e224109", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .g. see Jedwab and Storeygard\n2015; and Blankespoor et al. 2016).\n\n\n28 The following road data are used in the travel time model: road categories from DeLorme (Primary, Secondary and\nTertiary); road surface type (paved and non‐paved) and road surface condition (very good, good, fair, poor and very\npoor).\n\n\n39\n\nMention: \"road categories from DeLorme\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Primary, secondary and tertiary classes feed the travel time model.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007546:25:0", "occurrence_id": "6aca5f532401628ef99b8a22", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: These results are in stark contrast with those reported in recent studies. Predictions based on different\ncutoffs for population size and population density yield urbanization rates that range from 47.2 to 78.0\npercent (IDFC Institute 2015). A measure based on an agglomeration index that combines population\ndensity, population of the nearby main center, and travel time to that center suggests an urbanization\nrate of 55.3 percent (Uchida and Nelson 2009; World Bank 2015). And measures based on satellite imagery\nand agglomeration criteria produce urbanization rates of from 37.5 to 63.1 percent (Denis and Marius‐\nGnanou 2011). All of these results exceed our highest estimate.\n\n\n_Gaps at the subnational level_\n\n\nAlthough we find no major discrepancy with the official urbanization rate for India as a whole, our results\nsuggest that the 2011 Census of India misclassified large numbers of units and people. Based on the\nbenchmark specification, the misclassification affected 62 million people.\n\n\nThe misclassification is substantial among units falling within three of India’s administrative types (figure\n8). Among statutory urban units, the gap is highest for other towns. According to our assessment, 1,317\nof 2,861 lack sufficient urban characteristics and so should be considered rural. Thus 17 million people are\nincorrectly considered urban residents.\n\n\n**Figure 8. Misclassification of urbanization in India is highest among three administrative types**\n\n\nMisclassification is also substantial among the census towns of India. These units are administratively\nrural, but they meet three criteria to be identified as urban areas according to the 2011 Census of India.\nThey have more than 5,000 inhabitants, a population density exceeding 400 persons per square kilometer,\nand over 75 percent of their male working population is engaged in non‐agricultural activities. We find\n\n\n24\n\nMention: \"the 2011 Census of India\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Its urbanisation rate suggests large numbers of misclassified units.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007547:27:1", "occurrence_id": "4f9d3dbe9369a545e3fe0549", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  using Fitzgerald _et al._ ’s method are qualitatively\n\n\nsimilar to the results shown in Table 1 and Table 3, while those using Wooldridge’s method display\n\n\n19 We use the unweighted HBS data because population weights are not available. Official estimates by the Russian\nNational Statistical Agency “Rosstat” also suggest that the Gini coefficient remains stable between 1994-2015 around\n0.41 (Rosstat, 2016). See also Yemtsov (2008) for discussion on other issues related to reweighting and non-response\nwith the HBS data.\n\n\n26\n\nMention: \"unweighted HBS data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"Used unweighted because population weights are unavailable.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007550:43:0", "occurrence_id": "70395a256ddb8c2c56f3f171", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: is a slack measure, 𝜋���� represents the imported inflation, while 𝜋������ is the growth rate of the\n\naverage gross wage.\n\n\nThe following measures of slack were considered: (1) output gap – NBR estimates based on the\n\n\nmedium term analysis and projection model; (2) output gap – determined by applying the\n\n\nHodrick-Prescott filter to the series of real GDP; (3) ILO unemployment gap – NBR estimates\n\n\nbased on the medium term analysis and projection model; (4) the growth rate of the number of\n\n\nunemployed persons (ILO data).\n\n\nThe following measures as proxy for imported inflation were considered: (1) HICP inflation in\n\n\nthe euro area; (2) the growth rate of an effective external price index (a composite indicator\n\n\nquantifying the impact of euro area and US price dynamics on domestic inflation, according to the\n\n\nshares held by the two currencies in Romania’s foreign trade invoices); (3) the growth rate of the\n\n\neuro area export deflator; (4) the growth rate of the Romanian unit value of imports excluding\n\n\nfuels and motor vehicles, adjusted for the EURRON exchange rate dynamics.\n\n\n**_Estimation_**\n\n\nThe equation was set at quarterly frequency and the estimation was performed for the period\n\n\n2000-2016, using the least-square method for the fixed-coefficients version of the equation. In\n\n\ncase of the time-varying coefficients version, the equation was estimated using Bayesian methods,\n\n\nnamely a Gibbs sampler, with Inverted-Gamma priors for all of the parameters of the model.\n\n\n**2. A semi-structural model**\n\n\nTo further check the robustness of the results, a small-scale semi-structural model was also\n\n\nconsidered.\n\n\n**_Methodological framework_**\n\n\nThe model consisted of three main equations: the equation of aggregate demand, i.e. output gap\n\n\nequation (IS curve); the equation of core inflation (Phillips curve), with\n\nMention: \"ILO data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The growth rate of the unemployed enters the projection model.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007556:7:0", "occurrence_id": "ea3742b4799c9d74f6155e0a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 2: U.S. MFN and Column 2 tariffs\n\n\n\n(A) A comparative snapshot of U.S. tariffs\n\n\n\n(B) U.S. Column 2 tariffs\n\n\n\n3500\n\n\n3000\n\n\n2500\n\n\n2000\n\n\n1500\n\n\n1000\n\n\n500\n\n\n0\n\n\n\nPercent\n\n40\n35\n30\n25\n20\n15\n10\n5\n0\n\n\n\nRate increase\nColumn 2 rate\n\n\n\n100%\n90%\n80%\n70%\n60%\n50%\n40%\n30%\n20%\n10%\n0%\n\n\n\nFrequency Cumulative %\n\n\n\n**Tariff rate**\n\n\nSource: USITC.\nA. Trade weighted averages.\nB. Based on the 2018 8-digit HTSUS harmonized tariff schedule of the United States and 11,796 product\nlines. Specific tariffs are not represented. Tariffs on washing machines and solar panels: above-quota\nsafeguard tariffs to be applied in the first year. Tariffs on autos and parts are not yet official.\n\n\nimports from countries of the Soviet Union and others under communist control, except\nYugoslavia. The Jackson–Vanik amendment to the Trade Act of 1974 withheld NTR status\nfrom communist, non-market economies that restricted the freedom of emigration and other\nhuman rights.\n\n\nChina, the biggest in the group of non-market economies, was first granted non-discriminatory\nMFN status in 1980. This was, however, subject to annual renewal and even in the absence\nof actual changes in U.S. tariffs, introduced a great deal of uncertainty and political controversy in US-China trade relations. Every year between 1990 and 2001, the U.S. Congress\nintroduced, but failed to pass, legislation to revoke China’s NTR status. Finally, in 2000,\nChina was granted permanent normal trade relations (PNTR) status, also facilitated by its\naccession to the WTO in 2001. Close to two decades after granting China PNTR status, there\nis still substantial controversy that surrounds China’s non-market economy status as well as\nthe impact of China’s WTO accession and PNTR status on the decline in U\n\nMention: \"2018 8-digit HTSUS harmonized tariff schedule\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Trade-weighted averages across nearly twelve thousand product lines.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007558:20:1", "occurrence_id": "b6a34963c22b3a52b39b6fde", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  naturally, strongly correlated with FDI share at the\nindustry level, and its inclusion yields very similar results to column (2).\n\n\nIt is also possible that HGFs are simply more prevalent in more innovative or more high‐tech sectors. In\ncolumn (4) we control for this by the share of innovative firms at the industry‐ level, calculated from the\nCommunity Innovation Survey (only available for the whole country, rather than at a geographically\ndisaggregated way). We find stronger productivity growth in industries with more innovative firms and\nin industries supplying more innovative industries. The regressions yield positive and mostly significant\ncoefficients for the HGF measures, but they are substantially smaller than in the baseline model. Finally,\nin column (5), we control for the industry’s technology by including the average productivity of the\nvertically related sectors. This is calculated from our data, hence available at the same level of\naggregation as the HGF measures. We find that firms supplied by and supplying industries with more\nproductive firms experience higher productivity growth. The downstream HGF measure again becomes\nsmaller but remains significant.\n\n\nWe find that HGF share in industries is indeed correlated with technology level and international\nintegration. These industry‐level variables explain a significant part of the association between HGF\nshare in vertically related industries and productivity growth, but even conditional on these factors\nmore HGF share seems to be associated with a premium in productivity growth.\n\n\n**Table 10 Controlling for other industry characteristics**\n\n\n(1) (2) (3) (4) (5)\n**Dependent:** **Labor Productivity (3‐year change)**\n\n\n19\n\nMention: \"Community Innovation Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Industry shares of innovative firms, available only at national level.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007559:43:0", "occurrence_id": "15847d8f3e824ba3d3f7733c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ##### **Annex 2: Explanation of TFP methodology**\n\nThe firm‐level data for estimating our TFP measures were retrieved from the Ministry of Science, Industry\nand Technology (MoSIT) Enterpreneurial Information System (EIS) database. Data in this database are for\nthe period 2007‐2014 for manufacturing firms only. No data were available at the affiliate level to estimate\ntheir separate production functions. <sup>17</sup> [^17: However, it remains unclear whether the variable reported at the manufacturing firm level also comprises activities\nof the affiliate, which may result that the productivity figures could include some services activities.]\n\nFirm‐level TFP measures used in this paper are computed in different ways. Over the years, various\nmethodologies have been developed in the international economics literature, with Olley and Pakes (2003)\nand Levinsohn and Petrin (2008) as the most commonly used ones. Several papers that are very close to\nour line of research, such as Fernandes and Paunov (2012) and Arnold _et al_ . (2015), have used the TFP\nestimation developed by Ackerberg _et al_ . (2006; 2015).\n\nAlthough all three approaches correct for the endogeneity of input choices, including the choice of services\nas inputs, for the TFP estimates, the Ackerberg specification improves the former two TFP methods by\ncorrecting for potential collinearity problems, which could otherwise occur from a distorting factor with\nregards to the identification of the variable input coefficients. Ackerberg _et al_ . (2006; 2015) also provide a\ncorrection for the timing of the input choice decision.\n\nThis paper also uses the Hsieh and Klenow (2009, 2014) way of estimating TFP, which is relatively new and\nfollows a different structure. More precisely, in their methodology the authors follow a structural\napproach using a Cobb‐Douglas estimate with commonly used labor shares to explicitly take into account\npolicy distortions that would otherwise affect traditional TFP measures in their revenue as opposed to\nphysical TFP.\n\nWe estimate all four approaches\n\nMention: \"Enterpreneurial Information System\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Manufacturing records across this period come from this system.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007570:5:1", "occurrence_id": "bcaa2298672afcf31070b96f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  study by Atuesta et al. (2018) found a clear link between job accessibility and\nhousing values in Mexico City.\n\n\nThe rest of this paper is structured as follows: Section 2 describes the context about job\naccessibility challenges in Nairobi. Section 3 discusses the methodology by explaining various data\nsets and analytical methods employed in this study. Section 4 presents the results on the job\naccessibility calculated for the city, the levels of job accessibility by different household groups,\nand hedonic regressions. Section 5 concludes with a brief summary of findings.\n\n\n**2. Context: Spatial challenges in Nairobi**\n\n\nRecently collected surveys help to grasp spatial challenges in Nairobi. Most Nairobi residents are\nrent-paying tenants, and low-cost rental units are mainly located in informal residential areas.\nAccording to the Cities Baseline Survey (explained in Section 3), the distribution of housing rents\nin Nairobi is skewed with a small group of very high rents (Panels A and B in Figure 1). The\naverage housing rent among the surveyed rental units is K Sh 5,315. <sup>3, 4</sup> A quarter of rental units\nhave a rent of lower than K Sh 1,800; half the rental units have rents less than K Sh 2,500; and 75\npercent of rental units have rents less than K Sh 4,400. About 10 percent of units have very high\nrents (more than K Sh 30,000). The mean rent value in informal settlements (K Sh 3,600) is less\nthan half of that in formal residential areas (K Sh 9,000), and the majority of cheap rental units\n(less than K Sh 2,500) in the city are located in informal residential areas.\n\n\n3 The housing rents used in this study include utilities.\n4 K Sh 5,315 was roughly equivalent to US$61 as of 2013.\n\n4\n\nMention: \"the Cities Baseline Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its rental distribution shows low-cost units concentrated in informal areas.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007573:5:0", "occurrence_id": "4f2143816f99a96e5bd8cd26", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Dolls, Fuest, Peichl (2012) and Dolls, Fuest, Peichl, Wittneben (2018) analyze the capacity of\nEuropean tax benefit systems to act as an automatic stabilizer in times of income changes and how\nthis capacity has been affected by policy changes.\n\n##### 2. Assessing vertical redistribution\n\n\n**2.1 Measuring redistribution**\n\n\nRedistribution is typically measured using the Reynolds-Smolensky index of redistribution as the\ndifference between the Gini coefficients of market income and disposable income:\n\n\n𝑅𝑆��𝐺𝑖𝑛𝑖 <sup>������</sup> �𝐺𝑖𝑛𝑖 <sup>����������</sup>             \n\nFigure 1, which is based on the OECD Income Distribution and Poverty Database <sup>3</sup> [^3: Available at http://stats.oecd.org/Index.aspx?DataSetCode=IDD.], shows the\nReynolds-Smolensky index of redistribution for a selection of countries. The European Union’s\naverage level of redistribution equals 21 Gini points. This is almost twice as large as the\nredistribution occurring in the United States, amounting to 11 Gini points.\n\n\nThe Gini difference is a measure of the redistributive effectiveness of taxes and transfers at a given\npoint in time: the larger the difference, the larger the impact of taxes and transfers on reducing\ninequality in market incomes. Market incomes include gross labor incomes and earnings from\nemployment and self-employment (both permanent and temporary jobs), capital income, incomes\nfrom property as well as private pensions. Disposable incomes equal market income minus direct\ntaxes and social insurance contributions (for both employees and self-employed), plus any\nadditional transfers (such as pensions, means tested benefits and non-means tested benefits). Direct\ntaxes include for example personal income taxes, taxes on capital dividends and interests, as well\nas property taxes. <sup>4</sup> Incomes are observed at the household level and weighted with the modified\nOECD equivalence scale. The Gini coefficients for\n\nMention: \"OECD Income Distribution and Poverty Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The inequality figure is built from this compilation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007574:39:1", "occurrence_id": "bccddbc050da236e21b1977b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ** : 2015 and 2016 waves of the Rural-Urban Migration in China (RUMiC) Survey, and RUMiC Information\nIntervention Survey, 2016.\n**Notes** :An estimated city-level treatment effect is plotted against the relative premium for the urban employee\nhealth insurance and pension programs in the city. The relative premium is calculated as the premium required\nfor self-employed workers to participate divided by the city-wide average monthly earnings of migrants. The citylevel treatment effects are coefficients from the health insurance and pension participation models estimated with\ncity-specific information treatment indicators (city indicators interacted with the information treatment) in models\nthat control for the city fixed effects.\n\n\n38\n\nMention: \"Rural-Urban Migration in China (RUMiC) Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Two waves plus an information intervention round.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007576:15:1", "occurrence_id": "c0723d2fe50f2d221c8bf5f2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 05 and NSS 2011–12. These surveys were conducted between\nJuly 2004 and June 2005 and between July 2011 and June 2012, respectively (NSSO 2005, 2012). They\nreport household consumption information on an itemized form. We use monthly consumption based\non the mixed recall period and divide by household size to compute the monthly nominal household\nexpenditure per capita.\n\n\nTo account for spatial price variations, we deflate the resulting nominal expenditure by a year‐specific\nsubnational deflator that differs across state and between rural and urban areas. To allow for\ncomparability across years, we further deflate the resulting expenditures by the nationwide consumer\nprice index.\n\n\nOne caveat of using household‐level surveys is their vulnerability to low response rates among the rich\nand to underreporting of expenditures among those who respond (Ravallion 2003). As a result, survey‐\nbased consumption aggregates for nationally representative samples typically fall short of private\nconsumption as measured by national accounts. With the underestimation being larger in better‐off\nlocations, the convergence coefficient ( 𝛽 <sup>�</sup> or 𝛽 <sup>�</sup> ) could be biased in the direction of greater apparent\nconvergence. However, the discrepancy between levels of consumption as measured by India’s NSS and\nnational accounts has stabilized since the late 2000s, which mitigates the risk of bias for the period\nunder consideration in this paper (2004 to 2011).\n\n\nWhen using living standards as performance indicator to study drivers of growth, the finding can be\nbiased by self‐selection and sorting. Households are heterogeneous in their mixture of skills, energy and\nentrepreneurship, and the most productive often drawn to more productive places (Combes, Duranton\nand Gobillon 2008; Combes et al. 2010). To account for this possibility, we consider the growth in\n\n\n13\n\nMention: \"NSSO\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Itemised household consumption from two survey years.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007576:15:1", "occurrence_id": "01a175408d7038d4dafc3b43", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 05 and NSS 2011–12. These surveys were conducted between\nJuly 2004 and June 2005 and between July 2011 and June 2012, respectively (NSSO 2005, 2012). They\nreport household consumption information on an itemized form. We use monthly consumption based\non the mixed recall period and divide by household size to compute the monthly nominal household\nexpenditure per capita.\n\n\nTo account for spatial price variations, we deflate the resulting nominal expenditure by a year‐specific\nsubnational deflator that differs across state and between rural and urban areas. To allow for\ncomparability across years, we further deflate the resulting expenditures by the nationwide consumer\nprice index.\n\n\nOne caveat of using household‐level surveys is their vulnerability to low response rates among the rich\nand to underreporting of expenditures among those who respond (Ravallion 2003). As a result, survey‐\nbased consumption aggregates for nationally representative samples typically fall short of private\nconsumption as measured by national accounts. With the underestimation being larger in better‐off\nlocations, the convergence coefficient ( 𝛽 <sup>�</sup> or 𝛽 <sup>�</sup> ) could be biased in the direction of greater apparent\nconvergence. However, the discrepancy between levels of consumption as measured by India’s NSS and\nnational accounts has stabilized since the late 2000s, which mitigates the risk of bias for the period\nunder consideration in this paper (2004 to 2011).\n\n\nWhen using living standards as performance indicator to study drivers of growth, the finding can be\nbiased by self‐selection and sorting. Households are heterogeneous in their mixture of skills, energy and\nentrepreneurship, and the most productive often drawn to more productive places (Combes, Duranton\nand Gobillon 2008; Combes et al. 2010). To account for this possibility, we consider the growth in\n\n\n13\n\nMention: \"nationwide consumer\nprice index\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"Used to deflate expenditures for comparability across years.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007576:15:1", "occurrence_id": "dc898c4229fae334cbd79723", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 05 and NSS 2011–12. These surveys were conducted between\nJuly 2004 and June 2005 and between July 2011 and June 2012, respectively (NSSO 2005, 2012). They\nreport household consumption information on an itemized form. We use monthly consumption based\non the mixed recall period and divide by household size to compute the monthly nominal household\nexpenditure per capita.\n\n\nTo account for spatial price variations, we deflate the resulting nominal expenditure by a year‐specific\nsubnational deflator that differs across state and between rural and urban areas. To allow for\ncomparability across years, we further deflate the resulting expenditures by the nationwide consumer\nprice index.\n\n\nOne caveat of using household‐level surveys is their vulnerability to low response rates among the rich\nand to underreporting of expenditures among those who respond (Ravallion 2003). As a result, survey‐\nbased consumption aggregates for nationally representative samples typically fall short of private\nconsumption as measured by national accounts. With the underestimation being larger in better‐off\nlocations, the convergence coefficient ( 𝛽 <sup>�</sup> or 𝛽 <sup>�</sup> ) could be biased in the direction of greater apparent\nconvergence. However, the discrepancy between levels of consumption as measured by India’s NSS and\nnational accounts has stabilized since the late 2000s, which mitigates the risk of bias for the period\nunder consideration in this paper (2004 to 2011).\n\n\nWhen using living standards as performance indicator to study drivers of growth, the finding can be\nbiased by self‐selection and sorting. Households are heterogeneous in their mixture of skills, energy and\nentrepreneurship, and the most productive often drawn to more productive places (Combes, Duranton\nand Gobillon 2008; Combes et al. 2010). To account for this possibility, we consider the growth in\n\n\n13\n\nMention: \"national accounts\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Survey samples fall short of the private consumption measured here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007563:29:1", "occurrence_id": "16e712b39f2cd163c636d70a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **\n\n\nSource: International Financial Statistics Database ‐ International Monetary Fund.\n\n\nDefinition of inflation: annual percentage change in the cost to the average consumer of acquiring a basket of goods\n\n\nand services. Percentage change is calculated based on the consumer price index. May be fixed or changed at\n\n\nspecified intervals, such as yearly.\n\n\nLink for download: <u>http://data.imf.org/?sk=388DFA60‐1D26‐4ADE‐B505‐A05A558D9A42&sId=1479329334655</u>\n\n\n  - **Gross Domestic Product**\n\n\nSource: Penn World Tables 9.0.\n\n\nDefinition: Real GDP at constant 2011 national prices (in mil. 2011US$).\n\n\nPaper associated: Feenstra, Robert C., Robert Inklaar and Marcel P. Timmer. 2015. \"The Next Generation of the Penn\n\n\nWorld Table.\" _American Economic Review_ 105(10): 3150‐3182.\n\n\nLink for download: <u>https://www.rug.nl/ggdc/productivity/pwt/</u>\n\n\n28\n\nMention: \"International Financial Statistics Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The inflation series are taken from this fund's database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007563:29:1", "occurrence_id": "ff7db6ea2899e4ddc03cfec0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **\n\n\nSource: International Financial Statistics Database ‐ International Monetary Fund.\n\n\nDefinition of inflation: annual percentage change in the cost to the average consumer of acquiring a basket of goods\n\n\nand services. Percentage change is calculated based on the consumer price index. May be fixed or changed at\n\n\nspecified intervals, such as yearly.\n\n\nLink for download: <u>http://data.imf.org/?sk=388DFA60‐1D26‐4ADE‐B505‐A05A558D9A42&sId=1479329334655</u>\n\n\n  - **Gross Domestic Product**\n\n\nSource: Penn World Tables 9.0.\n\n\nDefinition: Real GDP at constant 2011 national prices (in mil. 2011US$).\n\n\nPaper associated: Feenstra, Robert C., Robert Inklaar and Marcel P. Timmer. 2015. \"The Next Generation of the Penn\n\n\nWorld Table.\" _American Economic Review_ 105(10): 3150‐3182.\n\n\nLink for download: <u>https://www.rug.nl/ggdc/productivity/pwt/</u>\n\n\n28\n\nMention: \"Penn World Tables\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Real output at constant national prices in millions of dollars.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007578:23:1", "occurrence_id": "535dd30b504c44db937b42f9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  data. Russian potential employment data and forecast\n\n\nare calculated using population and smoothed employment data in each age group obtained from\n\n\nILO as the fitted value fails to capture Russia’s low labor market participation rate for the aged.\n\n\nRussian capital stock data are taken from the capital stock index growth rate from the Russian\n\n\nFederation Statistical Service as it is more up-to-date than the Penn World table. The perpetual\n\n\ninventory method estimation that was employed for other countries generated similar results.\n\n\n22\n\nMention: \"population and smoothed employment data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"By age group, used where the fitted value misses a country's pattern.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007578:23:1", "occurrence_id": "99e694edc79e2b1e231b567b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  data. Russian potential employment data and forecast\n\n\nare calculated using population and smoothed employment data in each age group obtained from\n\n\nILO as the fitted value fails to capture Russia’s low labor market participation rate for the aged.\n\n\nRussian capital stock data are taken from the capital stock index growth rate from the Russian\n\n\nFederation Statistical Service as it is more up-to-date than the Penn World table. The perpetual\n\n\ninventory method estimation that was employed for other countries generated similar results.\n\n\n22\n\nMention: \"capital stock index growth rate\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"More up to date than the alternative international source.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007578:23:1", "occurrence_id": "7abe72956fb65ee70e437ca4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  data. Russian potential employment data and forecast\n\n\nare calculated using population and smoothed employment data in each age group obtained from\n\n\nILO as the fitted value fails to capture Russia’s low labor market participation rate for the aged.\n\n\nRussian capital stock data are taken from the capital stock index growth rate from the Russian\n\n\nFederation Statistical Service as it is more up-to-date than the Penn World table. The perpetual\n\n\ninventory method estimation that was employed for other countries generated similar results.\n\n\n22\n\nMention: \"Penn World table\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Less current than the national statistical service for this variable.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007586:66:0", "occurrence_id": "540bd3b6446eb1b4574b156a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: #### **9 Annexes**\n\n###### **9.1 Survey Data - Construction of Individual Selection Variables**\n\n_Public_ _Sector_ _Motivation_ _(PSM)_ - In order to measure bureaucrats’ and politicians’ moti\n\nvation to serve in the public sector, we use Perry’s Public Sector Motivation Index (Perry,\n\n\n1996). Annex I presents the version used in the survey in Uganda: it is composed of 36\n\n\nquestions measuring motivation in six different dimensions – Attraction to Policy Making,\n\n\nCommitment to Public Interest, Social Justice, Civic Duty, Compassion and Self-Sacrifice.\n\n\nAs discussed by Perry (1996), each dimension captures distinct motivations to serve iden\n\ntified previously in the public administration literature. Interviewees are asked how much\n\n\nthey agree with each of the questions on a scale of 1 (Strongly Disagree) to 5 (Strongly\n\n\nAgree). We follow the established methodology in the personnel economics literature, as\n\n\nin Dal Bo, Finan and Rossi (2013) and Callen et al. (2016): we first recode all questions so\n\n\nthat higher values are associated with normatively positive answers (i.e. higher agreement\n\n\nwith “I think there are many public causes which are worth defending” but disagreement\n\n\nwith “I hardly think of the welfare of people I do not know personally”), and then create\n\n\nan index for each dimension by taking simple averages of its answers. Finally, we create\n\n\nan aggregate PSM Index for each individual by taking an average of the z-scores of each of\n\n\nthe six dimensions.\n\n\n_Behavioral and Personality Traits_   - The module used to assess behavioral and person\n\nality traits is the same as the one used in the World Bank STEP (Skills Towards Employa\n\nbility and Productivity) surveys . Annex II presents the 25 questions used to measure eight\n\n\ndistinct dimensions of personality\n\nMention: \"World Bank STEP\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its personality and behavioural instrument is reused here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007589:16:0", "occurrence_id": "38e5f322c4dddb5897ab8030", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: distributing across the cells of the ‘USE’ matrix <sup>3</sup> [^3: The ‘USE’ matrix in a SAM refers to inputs of different goods and services for the production of various goods and\nservices.] the total value of domestically produced goods\n\n\nand services consumed by the productive sectors (intermediate consumption), using technology\n\ncoefficients from the GTAP <sup>4</sup> [^4: GTAP is a global network of researchers who conduct quantitative analysis of international economic policy issues,\nespecially trade policy. A key output of the network is a global economic database, covering many sectors and all\nparts of the world. It is coordinated by the Center for Global Trade Analysis, Department of Agricultural Economics,\nPurdue University, West Lafayette, IN.] 2007 SAM for Bangladesh. The use of 2007 GTAP coefficients\n\n\nmeans ignoring possible structural change between 2007 and 2012 in the Bangladesh economy.\n\n\nAll elasticity parameters are taken from Timilsina and Shrestha (2007). As it is difficult to\n\n\nempirically estimate the elasticity of substitution across commodities for individual countries,\n\n\nusing values from the literature is the norm. Data on government transfers to BPDB were obtained\n\n\nfrom the Medium Term Macroeconomic Policy Statement of the Ministry of Finance, Government\n\n\nof Bangladesh, while quantity forecasts for natural gas and LNG were obtained from Petrobangla\n\n\n(2012), with natural gas price data for 2015 for Bangladesh coming from Petrobangla and LNG\n\n\nprice data for India from Sen (2015).\n\n\n**4.** **Results and Discussion**\n\n\nIn this section, we first discuss results from the CGE model simulations of electricity\n\n\nsubsidy removal followed by results from simulations in which the price of natural gas is increased\n\n\nto its opportunity cost. Under each set of results, aggregate level results, such as impacts on GDP,\n\n\ntotal sectoral outputs, total household demand, total exports, total imports, total investment,\n\n\ndisposable income, the trade balance, and government income are discussed first; this is followed\n\n\nby a discussion\n\nMention: \"GTAP\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its technology coefficients give intermediate consumption linkages.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007590:63:0", "occurrence_id": "c8626407450d828342eb015b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: must contain at least one metropolitan area of above 300,000 people as defined by the UN\n\n\n(2015); third, the sum of the UN metropolitan area or areas within our polygon must be at\n\n\nleast 800,000 people.\n\n\nWe use the intersection with the UN metropolitan areas data as it provides an external\n\n\ncheck on the Landscan data. One issue we noticed in India and China, in particular, was\n\n\nthat large swathes of seemingly high density rural areas had been combined into gigantic\n\n\nurban areas, which we were certain were not urban areas in reality. Presumably this was\n\n\noccurring because Landscan is smearing people in these regions into agricultural areas.\n\n\nHence, the UN allows us to sense check the data against reported evidence.\n\n\nApplying our initial density-cutoffs to create contiguous polygons, leaves us with a set\n\n\nof 13,638 metropolitan area polygons. After we overlay this with the coordinates of the\n\n\nmetropolitan areas above 300,000 in the 2015 UN data, we get 1,544 polygons which match\n\n\nand we therefore keep. Only 54 of the total 1,692 UN points above 300,000 did not match\n\n\nwith our polygons. In the case that several metropolitan areas fall into one polygon, the\n\n\nUN population is summed and the name of the largest metropolitan area is kept. Once we\n\n\napply the total population threshold of above 800,000, we are finally left with a set of 599\n\n\ncities worldwide, with 451 in the developing world.\n\n\n**B.1.3** **Creating Core, Fringe and Settlement Polygons for Africa**\n\n\nFor separate analyses which we do just on Africa in Sections 4-6, we define urbanized\n\n\nareas using similar but distinct density and population threshold cutoffs. We define three\n\n\nunique urbanized areas: cores, fringes, and LD settlements. A city is the consolidation of\n\n\nan urban core and its surrounding peripheral fringe, whereas a settlement is a unique area.\n\n\nWe define each area as follows:\n\n\n1. Cores are defined\n\nMention: \"UN metropolitan areas data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"An external check on the population surface.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007590:63:0", "occurrence_id": "53d5a0ba4cae15f63562f443", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: must contain at least one metropolitan area of above 300,000 people as defined by the UN\n\n\n(2015); third, the sum of the UN metropolitan area or areas within our polygon must be at\n\n\nleast 800,000 people.\n\n\nWe use the intersection with the UN metropolitan areas data as it provides an external\n\n\ncheck on the Landscan data. One issue we noticed in India and China, in particular, was\n\n\nthat large swathes of seemingly high density rural areas had been combined into gigantic\n\n\nurban areas, which we were certain were not urban areas in reality. Presumably this was\n\n\noccurring because Landscan is smearing people in these regions into agricultural areas.\n\n\nHence, the UN allows us to sense check the data against reported evidence.\n\n\nApplying our initial density-cutoffs to create contiguous polygons, leaves us with a set\n\n\nof 13,638 metropolitan area polygons. After we overlay this with the coordinates of the\n\n\nmetropolitan areas above 300,000 in the 2015 UN data, we get 1,544 polygons which match\n\n\nand we therefore keep. Only 54 of the total 1,692 UN points above 300,000 did not match\n\n\nwith our polygons. In the case that several metropolitan areas fall into one polygon, the\n\n\nUN population is summed and the name of the largest metropolitan area is kept. Once we\n\n\napply the total population threshold of above 800,000, we are finally left with a set of 599\n\n\ncities worldwide, with 451 in the developing world.\n\n\n**B.1.3** **Creating Core, Fringe and Settlement Polygons for Africa**\n\n\nFor separate analyses which we do just on Africa in Sections 4-6, we define urbanized\n\n\nareas using similar but distinct density and population threshold cutoffs. We define three\n\n\nunique urbanized areas: cores, fringes, and LD settlements. A city is the consolidation of\n\n\nan urban core and its surrounding peripheral fringe, whereas a settlement is a unique area.\n\n\nWe define each area as follows:\n\n\n1. Cores are defined\n\nMention: \"Landscan data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The population surface whose large gaps are noted in two countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007592:23:0", "occurrence_id": "06ff73cdbc0e6b8a1aab89f1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Infrastructure Hub Project Pipeline, a database of future and current government\ninfrastructure projects that allows projects to be tracked from conception to completion\n(https://pipeline.gihub.org/).\n\n\n_Deep and liquid secondary market_\n\n\nA deep and liquid secondary market is important for the development of the _sukuk_ market\nin general and the green _sukuk_ market in particular. The liquidity of an instrument is\ngenerally measured by trading volume, trading frequency, bid-ask spreads, quote sizes,\ntrade sizes, and price impact coefficients (Petrella and Resti 2017). Tight bid-ask spreads\ngive investors an indication on the efficiency of a market; significant widening of the\nspread may be a signal of tension in the market.\n\n\n_Sukuk_ investors face a liquidity risk arising from the underdevelopment of the secondary\nmarket, which reflects the buy and hold attitude of investors, the lack of _sukuk_ supply and\nnew players in the market, the nontradability of certain _sukuk_ structures, and the\nunderdevelopment of domestic fixed-income markets. Secondary market _sukuk_ volumes\ncould improve as more and more African countries—home to a quarter of the world’s\nMuslims—enter the _sukuk_ market (Thomson Reuters 2017). <sup>19</sup> [^19: Côte d’Ivoire, The Gambia, Nigeria, d Senegal, South Africa, Sudan, and Togo have issued _sukuk_ ; Niger and Kenya\nare planning to do so.] Larger issuances, a wider\nand more diversified investor base, maturity, and global interest rate volatility determine\nthe perception of secondary market liquidity of a bond or a _sukuk_ (Nienhaus and Karatas\n2017). Other factors that could foster the development of a vibrant, deep, and liquid\nsecondary market for green _sukuk_ include preferential regulatory treatment, a transparent\nand well-established price discovery mechanism, and greater _sukuk_ supply.\n\n**_Financial innovation_**\n\n\nFinancial innovation consists of offering a diversified range of instruments that align\n\nMention: \"Infrastructure Hub Project Pipeline\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"A register of current and planned government infrastructure projects.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007594:8:2", "occurrence_id": "a8eeae90436cc1785cedd640", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: sup> 6/2013\n4/20118/2012 <sup>(1)</sup>\n\n\n\n101,662\n\n\n\n1,832\n\n\n\n7,180\n\n\n\n16,341\n\n\n\n2/2010 - 2012 <sup>(3)</sup> 7/2011- 9/2009– 2010 <sup>(6)</sup> 7/2010- 7/2012 1/2011 <sup>(2)</sup> 6/2012 <sup>(4)</sup> 9/2010 <sup>(5)</sup> 6/2011 <sup>(7)</sup> 6/2013 <sup>(8)</sup>\n\n\n\nSource: Authors’ own estimates based on South Asia Harmonized Micro Dataset (SARMD).\n<mark>(1)National Risk and Vulnerability Assessment 2011 - 2012, Afghanistan Living Conditions Survey Report. Helmand and Khost provinces were</mark>\n<mark>excluded from consumption information</mark>\n(2) Statistical Division of the Ministry of Planning (2011) Bangladesh Household Income and Expenditures Survey: Key Findings and Report.\n(3) Asian Development Bank (2013) Bhutan Living Standard Survey 2012 report.\n(4) All the States and Union Territories except Andaman & Nicobar Islands, and the remote areas of Nagaland. (Indian Central Statistical Office: No\nDesign and Estimation Procedure of NSS 68th Round)\n(5) More precisely: Malè: Sept 2009 - Jan 2010; Atolls: Feb 2010 - Sept 2010. Source: Department of National Planning (2012) Household Income a\nExpenditures Survey 2009/10. Findings\n(6) Central Bureau of Statistics (2011) Nepal Living Standard Survey – highlights\n(7) Statistic Division of the Government of Pakistan - PSLM report\n(8) Sri Lanka Department of Census and Statistics (\n\nMention: \"South Asia Harmonized Micro Dataset\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Regional estimates are built on this harmonised collection.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007594:8:2", "occurrence_id": "21025064183224a61101c0d1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: sup> 6/2013\n4/20118/2012 <sup>(1)</sup>\n\n\n\n101,662\n\n\n\n1,832\n\n\n\n7,180\n\n\n\n16,341\n\n\n\n2/2010 - 2012 <sup>(3)</sup> 7/2011- 9/2009– 2010 <sup>(6)</sup> 7/2010- 7/2012 1/2011 <sup>(2)</sup> 6/2012 <sup>(4)</sup> 9/2010 <sup>(5)</sup> 6/2011 <sup>(7)</sup> 6/2013 <sup>(8)</sup>\n\n\n\nSource: Authors’ own estimates based on South Asia Harmonized Micro Dataset (SARMD).\n<mark>(1)National Risk and Vulnerability Assessment 2011 - 2012, Afghanistan Living Conditions Survey Report. Helmand and Khost provinces were</mark>\n<mark>excluded from consumption information</mark>\n(2) Statistical Division of the Ministry of Planning (2011) Bangladesh Household Income and Expenditures Survey: Key Findings and Report.\n(3) Asian Development Bank (2013) Bhutan Living Standard Survey 2012 report.\n(4) All the States and Union Territories except Andaman & Nicobar Islands, and the remote areas of Nagaland. (Indian Central Statistical Office: No\nDesign and Estimation Procedure of NSS 68th Round)\n(5) More precisely: Malè: Sept 2009 - Jan 2010; Atolls: Feb 2010 - Sept 2010. Source: Department of National Planning (2012) Household Income a\nExpenditures Survey 2009/10. Findings\n(6) Central Bureau of Statistics (2011) Nepal Living Standard Survey – highlights\n(7) Statistic Division of the Government of Pakistan - PSLM report\n(8) Sri Lanka Department of Census and Statistics (\n\nMention: \"Afghanistan Living Conditions Survey Report\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Two provinces excluded from the consumption information.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007594:8:2", "occurrence_id": "246470d1328fe11c12b9f357", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: sup> 6/2013\n4/20118/2012 <sup>(1)</sup>\n\n\n\n101,662\n\n\n\n1,832\n\n\n\n7,180\n\n\n\n16,341\n\n\n\n2/2010 - 2012 <sup>(3)</sup> 7/2011- 9/2009– 2010 <sup>(6)</sup> 7/2010- 7/2012 1/2011 <sup>(2)</sup> 6/2012 <sup>(4)</sup> 9/2010 <sup>(5)</sup> 6/2011 <sup>(7)</sup> 6/2013 <sup>(8)</sup>\n\n\n\nSource: Authors’ own estimates based on South Asia Harmonized Micro Dataset (SARMD).\n<mark>(1)National Risk and Vulnerability Assessment 2011 - 2012, Afghanistan Living Conditions Survey Report. Helmand and Khost provinces were</mark>\n<mark>excluded from consumption information</mark>\n(2) Statistical Division of the Ministry of Planning (2011) Bangladesh Household Income and Expenditures Survey: Key Findings and Report.\n(3) Asian Development Bank (2013) Bhutan Living Standard Survey 2012 report.\n(4) All the States and Union Territories except Andaman & Nicobar Islands, and the remote areas of Nagaland. (Indian Central Statistical Office: No\nDesign and Estimation Procedure of NSS 68th Round)\n(5) More precisely: Malè: Sept 2009 - Jan 2010; Atolls: Feb 2010 - Sept 2010. Source: Department of National Planning (2012) Household Income a\nExpenditures Survey 2009/10. Findings\n(6) Central Bureau of Statistics (2011) Nepal Living Standard Survey – highlights\n(7) Statistic Division of the Government of Pakistan - PSLM report\n(8) Sri Lanka Department of Census and Statistics (\n\nMention: \"Bangladesh Household Income and Expenditures Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its key findings give the national welfare figures.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007594:8:2", "occurrence_id": "e065ea0b2f588beebb697208", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: sup> 6/2013\n4/20118/2012 <sup>(1)</sup>\n\n\n\n101,662\n\n\n\n1,832\n\n\n\n7,180\n\n\n\n16,341\n\n\n\n2/2010 - 2012 <sup>(3)</sup> 7/2011- 9/2009– 2010 <sup>(6)</sup> 7/2010- 7/2012 1/2011 <sup>(2)</sup> 6/2012 <sup>(4)</sup> 9/2010 <sup>(5)</sup> 6/2011 <sup>(7)</sup> 6/2013 <sup>(8)</sup>\n\n\n\nSource: Authors’ own estimates based on South Asia Harmonized Micro Dataset (SARMD).\n<mark>(1)National Risk and Vulnerability Assessment 2011 - 2012, Afghanistan Living Conditions Survey Report. Helmand and Khost provinces were</mark>\n<mark>excluded from consumption information</mark>\n(2) Statistical Division of the Ministry of Planning (2011) Bangladesh Household Income and Expenditures Survey: Key Findings and Report.\n(3) Asian Development Bank (2013) Bhutan Living Standard Survey 2012 report.\n(4) All the States and Union Territories except Andaman & Nicobar Islands, and the remote areas of Nagaland. (Indian Central Statistical Office: No\nDesign and Estimation Procedure of NSS 68th Round)\n(5) More precisely: Malè: Sept 2009 - Jan 2010; Atolls: Feb 2010 - Sept 2010. Source: Department of National Planning (2012) Household Income a\nExpenditures Survey 2009/10. Findings\n(6) Central Bureau of Statistics (2011) Nepal Living Standard Survey – highlights\n(7) Statistic Division of the Government of Pakistan - PSLM report\n(8) Sri Lanka Department of Census and Statistics (\n\nMention: \"Bhutan Living Standard Survey 2012 report\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The published results behind the Bhutan figures.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007594:8:2", "occurrence_id": "131ce08d4e37c6a7ff049d5d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: sup> 6/2013\n4/20118/2012 <sup>(1)</sup>\n\n\n\n101,662\n\n\n\n1,832\n\n\n\n7,180\n\n\n\n16,341\n\n\n\n2/2010 - 2012 <sup>(3)</sup> 7/2011- 9/2009– 2010 <sup>(6)</sup> 7/2010- 7/2012 1/2011 <sup>(2)</sup> 6/2012 <sup>(4)</sup> 9/2010 <sup>(5)</sup> 6/2011 <sup>(7)</sup> 6/2013 <sup>(8)</sup>\n\n\n\nSource: Authors’ own estimates based on South Asia Harmonized Micro Dataset (SARMD).\n<mark>(1)National Risk and Vulnerability Assessment 2011 - 2012, Afghanistan Living Conditions Survey Report. Helmand and Khost provinces were</mark>\n<mark>excluded from consumption information</mark>\n(2) Statistical Division of the Ministry of Planning (2011) Bangladesh Household Income and Expenditures Survey: Key Findings and Report.\n(3) Asian Development Bank (2013) Bhutan Living Standard Survey 2012 report.\n(4) All the States and Union Territories except Andaman & Nicobar Islands, and the remote areas of Nagaland. (Indian Central Statistical Office: No\nDesign and Estimation Procedure of NSS 68th Round)\n(5) More precisely: Malè: Sept 2009 - Jan 2010; Atolls: Feb 2010 - Sept 2010. Source: Department of National Planning (2012) Household Income a\nExpenditures Survey 2009/10. Findings\n(6) Central Bureau of Statistics (2011) Nepal Living Standard Survey – highlights\n(7) Statistic Division of the Government of Pakistan - PSLM report\n(8) Sri Lanka Department of Census and Statistics (\n\nMention: \"Household Income a\nExpenditures Survey 2009/10\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Fieldwork across atolls in the stated months.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007552:10:0", "occurrence_id": "284522d554ae34f333bfc9bb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **S** ource: Penn World Table 9.0. Human capital 𝐻 in the Penn World Table 9.0 is assumed to enter\nproduction as in the following production function 𝑌= 𝐴𝐾 <sup>𝛼</sup> (𝐻𝑙) <sup>1−𝛼</sup> . Years of schooling for Indonesia\nbetween 2011 and 2014 are extrapolated. Output, capital and TFP are measured at constant national\nprices.\n\n\n**3.** **Labor allocation across sectors**\n\n\nIn EAP, labor productivity improvement was driven largely by labor productivity growth within\nsectors (Figure 7). <sup>3</sup> The three sectors we consider are agriculture, industry, and services. The\nreallocation of labor away from less-productive sectors to more productive sectors (“static\nreallocation”) also contributed to improving productivity, especially in Indonesia, Thailand and\nVietnam. Limited evidence of faster productivity growth (“dynamic reallocation”) is available\nonly for China.\n\n\n**Figure 7 Growth in labor productivity: contributions of intersectoral reallocation and**\n\n**within-sector productivity growth (annualized, percent)**\n\n\n3 This is consistent with the evidence from Foster-McGregor and Verspagen (2016). The authors note that withinsector productivity growth explains most of the productivity change for middle to high-income countries.\nConversely, structural transformation plays a larger role as economies move from low- to middle-income status.\nThis can be explained by the fact that high-income economies tend to display lower dispersion in productivity across\nsectors.\n\n\n9\n\nMention: \"Penn World Table\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its human capital measure enters production in the stated way.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007579:18:0", "occurrence_id": "9381327cd9d5112921ed2109", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Source: Authors’ calculations using DHS and MICS._\n\n\n_4.3: Regional Analysis_\n\n\nUntil now we have focused on headcount and gap measures only at the country level.\n\n\nExtending this analysis to the regional level may provide further insight on malnutrition across the\n\n\nworld. Therefore, we examine the regional averages of malnutrition. As survey data are not\n\n\navailable every year for most countries, only a few countries in a particular region have a survey\n\n\nin a given year, with some regions having no survey conducted in certain years.\n\n\nThere are two common practices for calculating regional averages in such cases: (1) modeling\n\n\nmethods and (2) aggregating over a range of years. An example of modeling methods closely\n\n\nrelated to this study is the UNICEF-WHO-World Bank joint child malnutrition estimates (JME)\n\n\n(UNICEF, WHO, and the World Bank 2018). The JME adopts linear mixed-effect models\n\n\nallowing for random effects at the country level and for heterogeneous covariance structures. One\n\n\nmodel is fitted for each region or country group for calculating its aggregated number. Such\n\n\nmodeling methods are beyond the scope of this study. For simplicity, we chose to calculate\n\n\nregional averages for a range of years following the exercise in Nguyen and Wodon (2015). Thus,\n\n\nwe create regional averages for five-year periods: 1993-1997, 1998-2002, 2003-2007, and 2008\n\n2012. We use these five-year periods so that the middle years of these ranges,1995, 2000, 2005,\n\n\n17\n\nMention: \"DHS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Combined with the other programme for the regional welfare measures.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007579:18:0", "occurrence_id": "997bfa5e34ee729a6308cabf", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Source: Authors’ calculations using DHS and MICS._\n\n\n_4.3: Regional Analysis_\n\n\nUntil now we have focused on headcount and gap measures only at the country level.\n\n\nExtending this analysis to the regional level may provide further insight on malnutrition across the\n\n\nworld. Therefore, we examine the regional averages of malnutrition. As survey data are not\n\n\navailable every year for most countries, only a few countries in a particular region have a survey\n\n\nin a given year, with some regions having no survey conducted in certain years.\n\n\nThere are two common practices for calculating regional averages in such cases: (1) modeling\n\n\nmethods and (2) aggregating over a range of years. An example of modeling methods closely\n\n\nrelated to this study is the UNICEF-WHO-World Bank joint child malnutrition estimates (JME)\n\n\n(UNICEF, WHO, and the World Bank 2018). The JME adopts linear mixed-effect models\n\n\nallowing for random effects at the country level and for heterogeneous covariance structures. One\n\n\nmodel is fitted for each region or country group for calculating its aggregated number. Such\n\n\nmodeling methods are beyond the scope of this study. For simplicity, we chose to calculate\n\n\nregional averages for a range of years following the exercise in Nguyen and Wodon (2015). Thus,\n\n\nwe create regional averages for five-year periods: 1993-1997, 1998-2002, 2003-2007, and 2008\n\n2012. We use these five-year periods so that the middle years of these ranges,1995, 2000, 2005,\n\n\n17\n\nMention: \"MICS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The second programme in the joint calculation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007583:38:0", "occurrence_id": "f4c5d47e6d2165be7916aaad", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 17. Cumulative tax multiplier evaluated after two years: Linear vs. non-linear multipliers\n\n\nNotes: Country …xed e¤ect panel regression with linear and quadratic trends.\n\n\n6.5 Non-linearities in action III: How do taxes a¤ect incentives to work and invest?\n\n\nIn this subsection, we analyze whether the non-linear distortionary and disincentive-based theoretical\n\n\narguments and the empirical results obtained above are actually re‡ected on the perceived e¤ect of\n\n\ntaxes on incentives to work and invest. Based on a survey of a representative sample of business\n\n\nleaders in 142 countries, the Global Competitiveness Index elaborated by the World Economic Forum\n\n\nincludes speci…c questions such as (i) “[i]n your country, to what extent do taxes reduce the incentive\n\n\nto work?”and (ii) “[i]n your country, to what extent do taxes reduce the incentive to invest?”.\n\n\nFigure 18. Relation between VAT rate and the perceived (survey-based) e¤ect of taxes on the\n\n\nincentives to work and invest\n\n\nPanel A. Relation between VAT rate Panel B. Relation between VAT rate\n\n\nand the perceived e¤ect of taxes and the perceived e¤ect of taxes\n\n\non the incentives to work on the incentives to invest\n\n\nNotes: Regression line includes linear and quadratic terms.\n\n\n37\n\nMention: \"Global Competitiveness Index\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Survey of business leaders in 142 countries with specific items.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007599:11:0", "occurrence_id": "6ec84c8e10ef16745daef139", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: distortions, revenue productivity would be the same for all firms in a sector and across all sectors.\n\n\nThis implies that the greater the dispersion of TFPR, the greater the degree of misallocation. Table\n\n\n2 presents three measures of dispersion of TFPR. <sup>6</sup>\n\n\n**Table 2. Measures of Dispersion of TFPR, 1998–2013**\n\n\n**_1998_** **_2003_** **_2008_** **_2013_**\n\nStandard deviation 0.95 0.98 1.08 1.11\np75–p25 1.23 1.25 1.39 1.39\n<u>p90–p10</u> <u>2.39</u> <u>2.44</u> <u>2.73</u> <u>2.82</u>\n_Source:_ Author’s computations based on Economic Census data.\n_Note_ : _pX_ refers to the _X_ th percentile on the TFPR distribution. TFPR refers to the Total Revenue Productivity\n\n\nTwo facts follow from the analysis of TFPR. First, there is large dispersion in revenue productivity\n\n\nacross firms, implying substantial misallocation of resources. For example, in 2013, a firm in the\n\n\n75th percentile of the revenue productivity distribution was 39 percent more productive than one\n\n\nin the 25th percentile. <sup>7</sup> Second, over the 15 years considered, this dispersion has persisted (in fact,\n\n\nincreased slightly), indicating persistence in the misallocation of capital and labor.\n\n\n**Misallocation Has Specific Patterns Across Firms**\n\n\nMisallocation in Mexico shows specific patterns. To identify these patterns, revenue productivity\n\n\nis compared across firm types. Table 3 shows the results of OLS regressions of the _log_ of firm _i_\n\nMention: \"Economic Census data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Percentiles of the dispersion measure are computed from these enumerations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007602:29:0", "occurrence_id": "2d78091ffa9c03ff9d36ca68", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: implementable, replicable, and incentive compatible. Furthermore, from an institutional\n\n\nviewpoint, the SPI is designed to be better aligned with other capacity assessment tools. For\n\n\nexample, it is consistent with the Systematic Country Diagnostics (SCD), a new assessment tool\n\n\nthat the World Bank recently developed to identify a country’s priority development areas.\n\n\nOn the other hand, the increased numbers of indicators and countries covered for the SPI\n\n\nrequire more data compared to the SCI. The data for the SPI come from different sources, ranging\n\n\nfrom established databases with international organizations to national statistical agencies’\n\n\nwebsites. Given this heterogeneity of data sources, some data inconsistency may be expected. The\n\n\ncomplexities with the former are not just simply linearly related to the increase in quantities, but\n\n\nalso concern data quality challenges such as missing data. Missing indicators for less-developed\n\n\ncountries can represent a quite different data situation from missing indicators for richer countries.\n\n\nWhile the former may suggest weak statistical capacity, the latter might be explained by the little\n\n\nneed of tracking certain indicators. A specific example is child stunting. In the SPI database, we\n\n\nassign maximum scores to richer countries (including the OECD and the EU) for this indicator\n\n\nassuming that richer countries have achieved the desired target of reducing child stunting and\n\n\nprovide the benchmark against which progress for developing countries can be measured.\n\n\n**5. Some Empirical Illustrations**\n\n\n**5.1. Data**\n\n\nWe have constructed the database for the SPI for 2016. Efforts are underway to construct\n\n\nindicators for the SPI after 2016, as well as going backward several years for better comparison.\n\n\nWe provide below some empirical illustrations using the available SPI data and various other data\n\n\nsources including the World Bank’s World Development Indicators database (World Bank,\n\n\n28\n\nMention: \"SPI database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Maximum scores are assigned to richer countries for some indicators.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007602:29:0", "occurrence_id": "2e881674b1b7ac0cf0fd7f7a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: implementable, replicable, and incentive compatible. Furthermore, from an institutional\n\n\nviewpoint, the SPI is designed to be better aligned with other capacity assessment tools. For\n\n\nexample, it is consistent with the Systematic Country Diagnostics (SCD), a new assessment tool\n\n\nthat the World Bank recently developed to identify a country’s priority development areas.\n\n\nOn the other hand, the increased numbers of indicators and countries covered for the SPI\n\n\nrequire more data compared to the SCI. The data for the SPI come from different sources, ranging\n\n\nfrom established databases with international organizations to national statistical agencies’\n\n\nwebsites. Given this heterogeneity of data sources, some data inconsistency may be expected. The\n\n\ncomplexities with the former are not just simply linearly related to the increase in quantities, but\n\n\nalso concern data quality challenges such as missing data. Missing indicators for less-developed\n\n\ncountries can represent a quite different data situation from missing indicators for richer countries.\n\n\nWhile the former may suggest weak statistical capacity, the latter might be explained by the little\n\n\nneed of tracking certain indicators. A specific example is child stunting. In the SPI database, we\n\n\nassign maximum scores to richer countries (including the OECD and the EU) for this indicator\n\n\nassuming that richer countries have achieved the desired target of reducing child stunting and\n\n\nprovide the benchmark against which progress for developing countries can be measured.\n\n\n**5. Some Empirical Illustrations**\n\n\n**5.1. Data**\n\n\nWe have constructed the database for the SPI for 2016. Efforts are underway to construct\n\n\nindicators for the SPI after 2016, as well as going backward several years for better comparison.\n\n\nWe provide below some empirical illustrations using the available SPI data and various other data\n\n\nsources including the World Bank’s World Development Indicators database (World Bank,\n\n\n28\n\nMention: \"SPI data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Used with other compilations in the empirical illustrations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007602:29:0", "occurrence_id": "745d3e0200161d1a0d50e610", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: implementable, replicable, and incentive compatible. Furthermore, from an institutional\n\n\nviewpoint, the SPI is designed to be better aligned with other capacity assessment tools. For\n\n\nexample, it is consistent with the Systematic Country Diagnostics (SCD), a new assessment tool\n\n\nthat the World Bank recently developed to identify a country’s priority development areas.\n\n\nOn the other hand, the increased numbers of indicators and countries covered for the SPI\n\n\nrequire more data compared to the SCI. The data for the SPI come from different sources, ranging\n\n\nfrom established databases with international organizations to national statistical agencies’\n\n\nwebsites. Given this heterogeneity of data sources, some data inconsistency may be expected. The\n\n\ncomplexities with the former are not just simply linearly related to the increase in quantities, but\n\n\nalso concern data quality challenges such as missing data. Missing indicators for less-developed\n\n\ncountries can represent a quite different data situation from missing indicators for richer countries.\n\n\nWhile the former may suggest weak statistical capacity, the latter might be explained by the little\n\n\nneed of tracking certain indicators. A specific example is child stunting. In the SPI database, we\n\n\nassign maximum scores to richer countries (including the OECD and the EU) for this indicator\n\n\nassuming that richer countries have achieved the desired target of reducing child stunting and\n\n\nprovide the benchmark against which progress for developing countries can be measured.\n\n\n**5. Some Empirical Illustrations**\n\n\n**5.1. Data**\n\n\nWe have constructed the database for the SPI for 2016. Efforts are underway to construct\n\n\nindicators for the SPI after 2016, as well as going backward several years for better comparison.\n\n\nWe provide below some empirical illustrations using the available SPI data and various other data\n\n\nsources including the World Bank’s World Development Indicators database (World Bank,\n\n\n28\n\nMention: \"World Development Indicators database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"One of the accompanying compilations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007603:26:1", "occurrence_id": "edb8e05db24b746fb1d04d47", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  of Latin American and Caribbean Cities 2012 Towards a new urban\n\n\ntransition. Nairobi, Kenia.\n\n\nWu, J., Ma, L., Li, W., Peng, J., & Liu, H. (2014). Dynamics of urban density in China: Estimations\n\n\nbased on DMSP/OLS nighttime light data. IEEE Journal of Selected Topics in Applied Earth\n\n\nObservations and Remote Sensing, 7(10), 4266-4275.\n\nMention: \"DMSP/OLS nighttime light data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its imagery gives the urban density estimates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007606:31:1", "occurrence_id": "a1277ba8f5874cc9898a1787", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: up>𝑆</sup> �� �𝑆���∗𝑃���/ ∑��� <sup>𝑆</sup> �� �𝑆���\n𝑃 <sup>��</sup> ��∑��� <sup>𝑆</sup> �� �𝑆���∗𝑃���/ ∑��� <sup>𝑆</sup> �� �𝑆���\nThe motivation for this final modification is that in the previous decomposition methods all\nexpending sectors contributed, by definition, positively to labor productivity, even if they are\nbelow average productivity levels or growth rates.\n\n**Data**\n\nWe use data from: the GGDC 10-sector database (Timmer et al., 2015). The data set includes\ninformation on real value-added and employment in agriculture, four industrial sectors, three\nmarket services sectors, government services, and a final sector consisting of other services\nfrom 1960 until 2011 (see Table B.1). The GGDC10 covers 42 countries, including 11 African\ncountries (see appendix Table B.2). <sup>29</sup>\n\nTo compare Ghana’s pattern of growth-affecting structural change to other countries, ideally,\nsector value added should be deflated by multilateral purchasing power parities on the sectorlevel to account for variation in relative prices across tradable and non-tradable sectors (Inklaar\nand Timmer, 2014). Unfortunately, however, the GGDC10 does not include relative sectorlevel prices. <sup>30</sup> [^30: For the year 2005, sector-level PPPs for the 11 African countries are integrated in the African Sector Database\n(Vries et al., 2015). For the same year, The Productivity Level Database provided by Groningen gives information\non relative prices for 42 economies, but South Africa is the only African country. Unfortunately, both series differ\nin their methodology and cannot be combined.] We therefore stick to a comparison of the components\n\nMention: \"GGDC 10-sector database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Real value added and employment across ten sectors.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007606:31:1", "occurrence_id": "1c9c074f96e023cdbda6c314", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: up>𝑆</sup> �� �𝑆���∗𝑃���/ ∑��� <sup>𝑆</sup> �� �𝑆���\n𝑃 <sup>��</sup> ��∑��� <sup>𝑆</sup> �� �𝑆���∗𝑃���/ ∑��� <sup>𝑆</sup> �� �𝑆���\nThe motivation for this final modification is that in the previous decomposition methods all\nexpending sectors contributed, by definition, positively to labor productivity, even if they are\nbelow average productivity levels or growth rates.\n\n**Data**\n\nWe use data from: the GGDC 10-sector database (Timmer et al., 2015). The data set includes\ninformation on real value-added and employment in agriculture, four industrial sectors, three\nmarket services sectors, government services, and a final sector consisting of other services\nfrom 1960 until 2011 (see Table B.1). The GGDC10 covers 42 countries, including 11 African\ncountries (see appendix Table B.2). <sup>29</sup>\n\nTo compare Ghana’s pattern of growth-affecting structural change to other countries, ideally,\nsector value added should be deflated by multilateral purchasing power parities on the sectorlevel to account for variation in relative prices across tradable and non-tradable sectors (Inklaar\nand Timmer, 2014). Unfortunately, however, the GGDC10 does not include relative sectorlevel prices. <sup>30</sup> [^30: For the year 2005, sector-level PPPs for the 11 African countries are integrated in the African Sector Database\n(Vries et al., 2015). For the same year, The Productivity Level Database provided by Groningen gives information\non relative prices for 42 economies, but South Africa is the only African country. Unfortunately, both series differ\nin their methodology and cannot be combined.] We therefore stick to a comparison of the components\n\nMention: \"GGDC10\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Covers forty-two countries including eleven African ones.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007606:31:1", "occurrence_id": "60421eb144e0a099328a34e3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: up>𝑆</sup> �� �𝑆���∗𝑃���/ ∑��� <sup>𝑆</sup> �� �𝑆���\n𝑃 <sup>��</sup> ��∑��� <sup>𝑆</sup> �� �𝑆���∗𝑃���/ ∑��� <sup>𝑆</sup> �� �𝑆���\nThe motivation for this final modification is that in the previous decomposition methods all\nexpending sectors contributed, by definition, positively to labor productivity, even if they are\nbelow average productivity levels or growth rates.\n\n**Data**\n\nWe use data from: the GGDC 10-sector database (Timmer et al., 2015). The data set includes\ninformation on real value-added and employment in agriculture, four industrial sectors, three\nmarket services sectors, government services, and a final sector consisting of other services\nfrom 1960 until 2011 (see Table B.1). The GGDC10 covers 42 countries, including 11 African\ncountries (see appendix Table B.2). <sup>29</sup>\n\nTo compare Ghana’s pattern of growth-affecting structural change to other countries, ideally,\nsector value added should be deflated by multilateral purchasing power parities on the sectorlevel to account for variation in relative prices across tradable and non-tradable sectors (Inklaar\nand Timmer, 2014). Unfortunately, however, the GGDC10 does not include relative sectorlevel prices. <sup>30</sup> [^30: For the year 2005, sector-level PPPs for the 11 African countries are integrated in the African Sector Database\n(Vries et al., 2015). For the same year, The Productivity Level Database provided by Groningen gives information\non relative prices for 42 economies, but South Africa is the only African country. Unfortunately, both series differ\nin their methodology and cannot be combined.] We therefore stick to a comparison of the components\n\nMention: \"African Sector Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Sector-level purchasing power parities for the African countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007606:31:1", "occurrence_id": "fbe045720970960cd85b64e9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: up>𝑆</sup> �� �𝑆���∗𝑃���/ ∑��� <sup>𝑆</sup> �� �𝑆���\n𝑃 <sup>��</sup> ��∑��� <sup>𝑆</sup> �� �𝑆���∗𝑃���/ ∑��� <sup>𝑆</sup> �� �𝑆���\nThe motivation for this final modification is that in the previous decomposition methods all\nexpending sectors contributed, by definition, positively to labor productivity, even if they are\nbelow average productivity levels or growth rates.\n\n**Data**\n\nWe use data from: the GGDC 10-sector database (Timmer et al., 2015). The data set includes\ninformation on real value-added and employment in agriculture, four industrial sectors, three\nmarket services sectors, government services, and a final sector consisting of other services\nfrom 1960 until 2011 (see Table B.1). The GGDC10 covers 42 countries, including 11 African\ncountries (see appendix Table B.2). <sup>29</sup>\n\nTo compare Ghana’s pattern of growth-affecting structural change to other countries, ideally,\nsector value added should be deflated by multilateral purchasing power parities on the sectorlevel to account for variation in relative prices across tradable and non-tradable sectors (Inklaar\nand Timmer, 2014). Unfortunately, however, the GGDC10 does not include relative sectorlevel prices. <sup>30</sup> [^30: For the year 2005, sector-level PPPs for the 11 African countries are integrated in the African Sector Database\n(Vries et al., 2015). For the same year, The Productivity Level Database provided by Groningen gives information\non relative prices for 42 economies, but South Africa is the only African country. Unfortunately, both series differ\nin their methodology and cannot be combined.] We therefore stick to a comparison of the components\n\nMention: \"Productivity Level Database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Relative prices for forty-two economies, though one country is missing.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007597:4:0", "occurrence_id": "cf5c11ac3baecc332e1ddcac", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: targeted to populations in greatest need and tailored to local conditions, thereby promoting the\n\n\nefficiency and impact of program efforts.\n\n\nThe emergence of standardized cross‐sectional household surveys such as the Demographic and\n\n\nHealth Surveys (DHS) Program, <sup>19</sup> which now routinely include georeferencing (i.e., latitude and\n\n\nlongitude coordinates), provides a new opportunity to understand subnational geographic variations\n\n\nin a wide range of socioeconomic, health and demographic indicators, including stunting. A recent\n\n\nDHS survey has been undertaken in Tanzania <sup>15</sup> and, like nearly all others, is configured to provide\n\n\nrepresentative estimates of stunting rates at the first subnational unit level (regions in Tanzania), but\n\n\nnot below. However, geospatial methods have been established that can use the georeferenced\n\n\nsurvey data alongside relevant spatially gridded covariate data to reliably estimate DHS indicators at\n\n\nmuch finer levels of spatial disaggregation. <sup>20,21</sup>\n\n\nThe use of georeferenced survey data with geospatial models to generate interpolated surfaces is\n\n\nperhaps most established in the field of infectious diseases. <sup>22,23,24,25,26</sup> In recent years, these methods\n\n\nhave also been developed and applied to a broader set of social, demographic, and health contexts.\n\n\nThis includes the use of both DHS and World Bank Living Standards Measurement Survey data, among\n\n\nother sources, to create high‐resolution poverty maps for a range of Sub‐Saharan African\n\n\ncountries. <sup>27,28,29,30</sup> Most recently, the method has been applied to mapping water and sanitation\n\n\nindicators in Nigeria based on the 2015 Nigeria Water and Sanitation Survey. <sup>31</sup> This study describes\n\n\nthe use of such an approach to generate a new high‐resolution map of stunting prevalence in Tanzania.\n\n\n**Methods**\n\n\n_\n\nMention: \"Demographic and\n\n\nHealth Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Standardized cross-sectional surveys that now routinely carry coordinates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007597:4:0", "occurrence_id": "cde7fd55e52f57323e3a4a62", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: targeted to populations in greatest need and tailored to local conditions, thereby promoting the\n\n\nefficiency and impact of program efforts.\n\n\nThe emergence of standardized cross‐sectional household surveys such as the Demographic and\n\n\nHealth Surveys (DHS) Program, <sup>19</sup> which now routinely include georeferencing (i.e., latitude and\n\n\nlongitude coordinates), provides a new opportunity to understand subnational geographic variations\n\n\nin a wide range of socioeconomic, health and demographic indicators, including stunting. A recent\n\n\nDHS survey has been undertaken in Tanzania <sup>15</sup> and, like nearly all others, is configured to provide\n\n\nrepresentative estimates of stunting rates at the first subnational unit level (regions in Tanzania), but\n\n\nnot below. However, geospatial methods have been established that can use the georeferenced\n\n\nsurvey data alongside relevant spatially gridded covariate data to reliably estimate DHS indicators at\n\n\nmuch finer levels of spatial disaggregation. <sup>20,21</sup>\n\n\nThe use of georeferenced survey data with geospatial models to generate interpolated surfaces is\n\n\nperhaps most established in the field of infectious diseases. <sup>22,23,24,25,26</sup> In recent years, these methods\n\n\nhave also been developed and applied to a broader set of social, demographic, and health contexts.\n\n\nThis includes the use of both DHS and World Bank Living Standards Measurement Survey data, among\n\n\nother sources, to create high‐resolution poverty maps for a range of Sub‐Saharan African\n\n\ncountries. <sup>27,28,29,30</sup> Most recently, the method has been applied to mapping water and sanitation\n\n\nindicators in Nigeria based on the 2015 Nigeria Water and Sanitation Survey. <sup>31</sup> This study describes\n\n\nthe use of such an approach to generate a new high‐resolution map of stunting prevalence in Tanzania.\n\n\n**Methods**\n\n\n_\n\nMention: \"DHS survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"A recent national round with the standard configuration.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007597:4:0", "occurrence_id": "52f7f314f8230a2324c21825", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: targeted to populations in greatest need and tailored to local conditions, thereby promoting the\n\n\nefficiency and impact of program efforts.\n\n\nThe emergence of standardized cross‐sectional household surveys such as the Demographic and\n\n\nHealth Surveys (DHS) Program, <sup>19</sup> which now routinely include georeferencing (i.e., latitude and\n\n\nlongitude coordinates), provides a new opportunity to understand subnational geographic variations\n\n\nin a wide range of socioeconomic, health and demographic indicators, including stunting. A recent\n\n\nDHS survey has been undertaken in Tanzania <sup>15</sup> and, like nearly all others, is configured to provide\n\n\nrepresentative estimates of stunting rates at the first subnational unit level (regions in Tanzania), but\n\n\nnot below. However, geospatial methods have been established that can use the georeferenced\n\n\nsurvey data alongside relevant spatially gridded covariate data to reliably estimate DHS indicators at\n\n\nmuch finer levels of spatial disaggregation. <sup>20,21</sup>\n\n\nThe use of georeferenced survey data with geospatial models to generate interpolated surfaces is\n\n\nperhaps most established in the field of infectious diseases. <sup>22,23,24,25,26</sup> In recent years, these methods\n\n\nhave also been developed and applied to a broader set of social, demographic, and health contexts.\n\n\nThis includes the use of both DHS and World Bank Living Standards Measurement Survey data, among\n\n\nother sources, to create high‐resolution poverty maps for a range of Sub‐Saharan African\n\n\ncountries. <sup>27,28,29,30</sup> Most recently, the method has been applied to mapping water and sanitation\n\n\nindicators in Nigeria based on the 2015 Nigeria Water and Sanitation Survey. <sup>31</sup> This study describes\n\n\nthe use of such an approach to generate a new high‐resolution map of stunting prevalence in Tanzania.\n\n\n**Methods**\n\n\n_\n\nMention: \"World Bank Living Standards Measurement Survey data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Combined with the health surveys to make high-resolution poverty maps.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007597:4:0", "occurrence_id": "88681a4591ff59d8a2cf00ed", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: targeted to populations in greatest need and tailored to local conditions, thereby promoting the\n\n\nefficiency and impact of program efforts.\n\n\nThe emergence of standardized cross‐sectional household surveys such as the Demographic and\n\n\nHealth Surveys (DHS) Program, <sup>19</sup> which now routinely include georeferencing (i.e., latitude and\n\n\nlongitude coordinates), provides a new opportunity to understand subnational geographic variations\n\n\nin a wide range of socioeconomic, health and demographic indicators, including stunting. A recent\n\n\nDHS survey has been undertaken in Tanzania <sup>15</sup> and, like nearly all others, is configured to provide\n\n\nrepresentative estimates of stunting rates at the first subnational unit level (regions in Tanzania), but\n\n\nnot below. However, geospatial methods have been established that can use the georeferenced\n\n\nsurvey data alongside relevant spatially gridded covariate data to reliably estimate DHS indicators at\n\n\nmuch finer levels of spatial disaggregation. <sup>20,21</sup>\n\n\nThe use of georeferenced survey data with geospatial models to generate interpolated surfaces is\n\n\nperhaps most established in the field of infectious diseases. <sup>22,23,24,25,26</sup> In recent years, these methods\n\n\nhave also been developed and applied to a broader set of social, demographic, and health contexts.\n\n\nThis includes the use of both DHS and World Bank Living Standards Measurement Survey data, among\n\n\nother sources, to create high‐resolution poverty maps for a range of Sub‐Saharan African\n\n\ncountries. <sup>27,28,29,30</sup> Most recently, the method has been applied to mapping water and sanitation\n\n\nindicators in Nigeria based on the 2015 Nigeria Water and Sanitation Survey. <sup>31</sup> This study describes\n\n\nthe use of such an approach to generate a new high‐resolution map of stunting prevalence in Tanzania.\n\n\n**Methods**\n\n\n_\n\nMention: \"2015 Nigeria Water and Sanitation Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its indicators are mapped with the same approach.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007601:10:0", "occurrence_id": "a8631d7b25d5a434e27bf790", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: captured in a separate question with a prompt for the respondent describing what constitutes an absent\n\n\nmember.\n\n\n_Figure 1 Five-arm randomized control trial testing various methods of measuring food consumption._\n\n\nBelow, we describe in greater detail the differences between treatment arms, with a focus on how we\n\n\nmeasure FAFH for adults in the household. With the exception of T1, adult FAFH was collected in an\n\n\nentirely separate module specifically designed for each of the experimental arms. Adult FAFH was\n\n\ncollected for nine separate meal events, differentiating main meals (breakfast, lunch and dinner), snacks\n\n\n(morning, afternoon, and evening), and drinks (bottled water, alcohol, other drinks). This structure was\n\n\nmeant to assist the respondent in thinking in more detail about FAFH, and to ensure they did not omit any\n\n\nFAFH consumption. The explicit mention of snacks, for example, is particularly important as most of the\n\n\nsnacking is likely to take place out of the home. The module also asked for the most common place of\n\n\nconsumption, and collected the value of all food consumed, whether purchased or received for free.\n\n\nQuestions were designed so that there was no double-counting of expenditures (e.g. for family meals outside\n\n\nthe home paid for by one member). More information on each of the experimental treatment arms can be\n\n\nfound in Appendix 9.1.\n\n\nUnlike the newly designed modules, treatment arm T1, the 7-day “one-line” recall, followed the current\n\n\nand standard methodology of the VHLSS, where households were asked to recall their food consumption\n\n\nin one question: “How much has your household consumed over the past 7 days of outdoor meals and drinks\n\n\n9\n\nMention: \"VHLSS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its standard single-question food recall is followed here.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007611:10:1", "occurrence_id": "f91feefde7cc2fb1ad3a3b62", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  resource share of value\nadded. To measure the growth in agricultural output as a factor of production we use the FAO’s\nindex of agricultural production (both food and non-food) measured in real terms.\n\nWhile this approach to agriculture may appear artificial, it exactly parallels how timber\nproduction is treated in TFP growth estimation. The wood itself is considered to be the natural\nresource, and the growth rate of the quantity of timber produced is calculated year on year based\nupon timber production measured in cubic meters. The share of timber production in value added\nis the associated resource rent on timber harvest – which, arguably, could also be viewed as a\ntype of land rent. <sup>4</sup>\n\n_Labor and the self-employed_\nIn developing countries, particularly low-income countries, self-employment is a large share of\ntotal labor inputs– in Sub-Saharan Africa, for example, the share of formal wages in value added\ncan be as low as 10%. For the share of labor in the equations above, 𝑠� as measured in the SNA\nis too low because it includes only formal employment. The Penn World Table data on total\ncompensation of labor add other components of value added (part of mixed income in developed\ncountries, agricultural value added in low-income countries) to the SNA figures for the\ncompensation of employees. As a result, the sum of returns to capital, labor and natural resources\n\n\n4 Note that there is some risk of double-counting the contribution of forests to TFP growth because,\naccording to SNA principles, _plantation_ forest outputs are measured as part of production in the\nagriculture sector.\n\n\n9\n\nMention: \"Penn World Table data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Total labour compensation adds components the national accounts omit.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007612:4:0", "occurrence_id": "dfaa99ebb5ef3ecbc13fc689", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: phenomena in the literature on armed conflict (Regan and Norton, 2005). <sup>1</sup> Following Sharp\n\n\n(1973, 2005) and Chenoweth and Ulfeder (2017), we define nonviolent uprisings as campaigns\n\n\nwhich are typically organized by “activists, public figures and civilians, who attempt to change\n\n\nthe status quo by coercing the opponent through disruption and pressure”, ultimately eroding\n\n\npublic support for the regime. The relationship between grievances and uprisings is expected to\n\n\nbe stronger in the case of peaceful than violent uprisings because the participation in peaceful\n\n\ncivil uprisings is associated with a lower risk of violent government retaliation than\n\n\nparticipation in armed conflict. Furthermore, participation in civil resistance does not require\n\n\nthe legitimization of the use of violence, nor the resources to buy weaponry. Hence, barriers to\n\n\nparticipation in peaceful events are lower than those in armed conflicts and as a result, success\n\n\nexpectations are likely to play a smaller role (Klandermans, 1996).\n\n\n_Figure 1 The number of demonstrations relative to the number of armed conflicts between 1995_\n_and 2015_\n\n\nData sources: CNTS (2015) for the number of demonstrations and Pettersson & Wallensteen (2015) for the number\nof armed conflicts.\n\nChenoweth and Ulfelder (2017) provide a first systematic examination of the extent to\n\n\nwhich grievances can explain nonviolent uprisings. They find that poverty can explain only a\n\n\n1 For instance, when nonviolent resistance fails, opposition groups often use this as a justification to escalate to\ncivil war (Regan and Norton, 2005).\n\n\n3\n\nMention: \"CNTS (2015)\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Counts of demonstrations across the two decades.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007613:28:3", "occurrence_id": "4284ff83c91411c03de4e005", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: * 2.247** 2.078* 0.486** 2.447** 0.960 0.77\n\n\n**IT** 0.718 0.988 15.013** 0.534 0.342 0.262**\n\n\n**hotel&restaurants** 1.758 2.123* 0.136* 2.108 0.878 1.195\n\n\n**construction** 0.932 0.660+ 0.324* 1.234 1.357 0.764\n\n\n**Transport** 0.679 0.727 0.332** 0.834 1.01 0.177**\n\n\n**<u>other services</u>** <u>4.790*</u> <u>1.905</u> <u>5.145</u> <u>4.307**</u> <u>0.812</u> <u>1.513</u>\n\n\n**Constant** 0.010** 0.017** 24.531** 0.013** 0.022** 0.050**\n\nYes Yes Yes Yes Yes\n\n**<u>Country Fixed effects</u>** <u>Yes</u>\n\nSource: own elaboration based on ES dataset. Note: logit regression applies sampling weights. The following countries were included in the\nanalysis: Cambodia, Indonesia, Lao PDR, Malaysia, Myanmar, the Philippines, Thailand, and Vietnam\n\n**+ p<0.10, * p<0.05, ** p<0.01**\n\n\n27\n\nMention: \"ES dataset\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Weighted logit estimates across the listed countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007620:49:0", "occurrence_id": "7830d63fca6592160596b292", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Openness of the\n_openness_ Export of goods and services (% of GDP) World Bank (WB)\n<u>economy</u>\n\n<u>GDP per capita</u> _<u>gdppc</u>_ <u>GDP per capita (constant 2010 US$)</u> <u>WB</u>\n\nWorld Federation\n\n\n\nMarket capitalization _capitalization_ Market capitalization of listed domestic companies (% of GDP)\n\n\nStocks traded _stock_turnover_ Stocks traded, turnover ratio of domestic shares (%)\nturnover ratio\n\n\n\nof Exchanges\n\n<u>database</u>\n\nWorld Federation\n\nof Exchanges\n\n<u>database</u>\n\n\nIMF\n\n\n\nNet foreign assets _nfa_\n\n\n\nNet foreign assets are the sum of foreign assets held by monetary\nauthorities and deposit money banks, less their foreign liabilities.\n<u>Data is in current local currency.</u>\n\n\n\nClaims on central Claims on central government (% of GDP) - includes loans to\n_claims_gov_gdp_ IMF\n<u>government</u> <u>central government institutions, net of deposits.</u>\n\n\n\n**_<u>Monetary Policy Variables</u>_**\nCentral Bank\nIndependence <u>Garriga - weighted</u>\n\nCentral Bank\nIndependence <u>Garriga - average</u>\n\n\n\nMeasure of Central Bank Independence according to Ana\n_cbi_G_\nCarolina Garriga - weighted average\n\n\nMeasure of Central Bank Independence according to Ana\n_cbi_G_a_\nCarolina Garriga - average\n\n\n\nCentral Bank Measure of Central Bank Independence according to Dincer and\n_cbi_D&E_\nIndependence - D&E Eichengreen\n\n\nCentral Bank Measure of Central Bank Independence according to Dincer and\n_cbt_D&E_\nTransparency Eichengreen\n\n\nValue 1 if the country uses inflation targeting as the monetary\n\nMention: \"IMF\n\n\n\nNet foreign assets\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Foreign assets held by monetary authorities plus those of banks.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007623:16:1", "occurrence_id": "4ab4063dd12d4970a8f3feb1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  a recession). The country fixed effects\ncapture any country-specific characteristics that may affect the outcomes of interest, including national\npriorities for the HD sectors and/or an appetite for HD lending. They also ensure that the identifying\nvariation comes from countries that transition from IDA to IBRD (since the fixed effects will be\ncollinear for countries that remain in the same status throughout). Given the fact that IDA graduation\npolicy based on the GNI per capita cut-off came into existence in 1987, for our main results, we\nrestrict the analysis to data from 1988-2015. In our robustness checks, we report the results from the\nfull sample of years – 1961-2015.\n\n\n**5.** **Regression results**\n\n\nTable 4A reports the marginal effects of estimating Probit regressions of the likelihood of any project\n(Columns 1-4) and the likelihood of any HD project (Columns 5-8). For each outcome variable, the\nfirst column reports the results from estimating equation 1 without controls and without country fixed\neffects. The second column then adds the control variables. The third column then reports results\nincluding country fixed effects but no controls. The fourth column then adds the control variables.\nAll regressions include fiscal year fixed effects. Results for other variables are reported in the same\nformat. We primarily focus our discussion on the fourth column, which includes year and country\nfixed effects, as well as the control variables, but present all results for completeness.\n\n\n13 In other versions of the results, we also included controls for other bilateral and multilateral net flows (DAC data from\nOECD) as well as countries’ own spending levels in the HD sectors (data from IMF). We chose not to include these\ncovariates because the data are incomplete (resulting in a large loss of number of observations) and because these data\ndid not have much explanatory power. Results are available upon request.\n\n\n15\n\nMention: \"data from IMF\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Country spending in the priority sectors, excluded for incompleteness.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007624:59:0", "occurrence_id": "d60c06bc2729e642bc815287", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ###### APPENDIX 8 : Mitigation Strategies: Protected Areas\n\nProtected areas can provide an important\ntool for mitigating environmental risks from\nBRI rail and road projects. Establishment or\nincreased enforcement of a protected area\nnear a transportation corridor can help\nreduce the potential impact from land use\nchange from human settlement, illegal timber\nextraction, and wildlife poaching. Figure 16\nshows that, while there are many protected\nareas along existing roads and rail in CICPEC\nas well as those under construction or\nplanned, they are strikingly absent from the\nLao People’s Democratic Republic. <sup>17</sup> This\nclearly flags the need to consider this as part\nof a mitigation strategy for the planned rail\nline that will pass through areas of high\nendemism for threatened biodiversity.\n\n\n\nEstablishing a protected area does not\nprovide automatic safeguard against human\nsettlement and destructive activities such as\nwildlife poaching and illegal logging. A road or\nrail bisecting a protected area will create an\nedge effect that can penetrate deep into the\nzone. Moreover, the protected status is only\nas strong as its enforcement. Consider the\ncase of Snuol Wildlife Reserve in Cambodia.\nLandsat imagery reveals that in less than a\ndecade after a road was constructed passing\ndirectly through the park, extensive clearing\noccurred within the park along the road in part\ndue to poor enforcement (Clements et al. 2014;\nFigure 17).\n\n\nTwo of the planned BRI projects – a segment of\nthe Thai Nam Tok‐ Thanbyuzayat rail (Figure 10\nin Appendix 3) and the Cambodian Sihanoukville\nPort road (Figure 11 in Appendix 6) – both pass\nthrough existing national parks. The degree to\nwhich these parks can provide sufficient\n\n\n\n\n\n17 While there are a few areas protected within Lao PDR, none is within the IUCN Categories I‐IV with its higher\nlevels of protection.\n\n\n\n58\n\nMention: \"Landsat imagery\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Reveals forest loss within a decade of a road being built.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007630:25:0", "occurrence_id": "147ea0159bac9180f1fe0c1a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: own children or children from one’s reference group whose conditions may be similar,\n\n\nprobably receive greater weights in one’s evaluation of children’s future chances in\n\n\ngeneral, how exactly respondents frame their answers to this survey question is not\n\n\nclear. Still, the results can be cautiously viewed as reinforcing the main results of the\n\n\nanalysis, as well as the predictions of theoretical models discussed in section 2, that\n\n\nlink inequality in access to key opportunities in life to lower expectations of future\n\n\nmobility; moreover, these lower expectations may be relevant not only in the context\n\n\nof the immediate future, but rather may persist over time, and across generations,\n\n\nthus perpetuating existing inequality traps.\n\n\n_7.2._ _Inequality_ _in_ _access_ _to_ _opportunities_ _and_ _redistributive_ _preferences_\n\n\nIf access to key opportunities in life is perceived to be unfair, on account to whom\n\n\nyou know, then resulting inequalities in society are likely to be similarly perceived to\n\n\nbe unfair. This is especially likely to be the case for those who do not have connec\n\ntions that may facilitate access to opportunities such as jobs or university education.\n\n\nWhen inequalities are perceived to be unfair, individuals may be more likely to desire\n\n\na greater amount of redistribution in society to compensate for such unfair distribu\n\ntion of rewards (see Cojocaru, 2014 for empirical evidence of the link between fairness\n\n\nand inequality aversion based on LiTS data). It is possible to explore with LiTS data\n\n\nwhether the “Rawlsian” inequality of opportunity examined in this analysis is asso\n\nciated with stronger redistributive preferences. The latter are elicited in the survey\n\n\nquestionnaire by way of the following statement “The gap between the rich and the\n\n\npoor in our country should be reduced”, with which the respondents can either agree\n\n\nor disagree on a 5-step Likert scale that ranges from “strongly disagree” to “strongly\n\n\nagree”\n\nMention: \"LiTS data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Its experimental items give fairness and inequality aversion.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007634:5:1", "occurrence_id": "304c1437c8db1da37cda6ce4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: . The actual implementation of new water projects is facilitated by the LGAs under the\n\n\nPresident’s Office of Regional and Local Government (PORALG).\n\n\nOne of the more ambitious projects that went hand-in-hand with the country’s effort in improving\n\n\nwater access was the creation of its Water Point Mapping (WPM) database. With the assistance of Water\n\n\nAid which has previously done similar work in Malawi (Stoupy & Sudgen, 2003), the country\n\n\nimplemented the WPM with the goal of gathering geographical data on all improved water points in an\n\n\narea, in addition to management, technical and demographic information. The information collected was\n\n\nused to create a baseline of accurate, reliable and up to date information on all improved water points in\n\n\naddition to serving as a benchmark for future monitoring efforts, improving decision making as well as\n\n\nallocation of resources for rural water supply services (Welle, 2010).\n\n\nThrough the implementation of the first phase of the WSDP and by extension, the RWSSP, an\n\n\nadditional 8,285 water points have been installed covering almost 1.9 million additional beneficiaries.\n\n\nThis has increased rural water coverage from 55 percent in 2007 – roughly at the start of the program – to\n\n\n5 The figures reported in JMP (2015) were lower than those reported in the 2002 Housing Census. While population growth\npressures could have influenced the figures, the difference is most likely due to the different definition of improved water points\nthat were adopted by the two databases. Under the WSDP, the WPM have since adopted the definition of improved water points\nthat is consistent with the internationally accepted definition by WHO/UNICEF (2000). The figures presented in JMP (2015)\nwould be based on this definition.\n6 Goals that were set partly through the commitments of the National Strategy for Growth and Reduction of Poverty also known\nas MKUKUTA and by the Millennium Development Goals (MDG) respectively.\n\n\n4\n\nMention: \"Water Point Mapping (WPM) database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Created alongside national efforts to improve access reporting.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007634:5:1", "occurrence_id": "3a69adc236dcaa0311cac0b5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: . The actual implementation of new water projects is facilitated by the LGAs under the\n\n\nPresident’s Office of Regional and Local Government (PORALG).\n\n\nOne of the more ambitious projects that went hand-in-hand with the country’s effort in improving\n\n\nwater access was the creation of its Water Point Mapping (WPM) database. With the assistance of Water\n\n\nAid which has previously done similar work in Malawi (Stoupy & Sudgen, 2003), the country\n\n\nimplemented the WPM with the goal of gathering geographical data on all improved water points in an\n\n\narea, in addition to management, technical and demographic information. The information collected was\n\n\nused to create a baseline of accurate, reliable and up to date information on all improved water points in\n\n\naddition to serving as a benchmark for future monitoring efforts, improving decision making as well as\n\n\nallocation of resources for rural water supply services (Welle, 2010).\n\n\nThrough the implementation of the first phase of the WSDP and by extension, the RWSSP, an\n\n\nadditional 8,285 water points have been installed covering almost 1.9 million additional beneficiaries.\n\n\nThis has increased rural water coverage from 55 percent in 2007 – roughly at the start of the program – to\n\n\n5 The figures reported in JMP (2015) were lower than those reported in the 2002 Housing Census. While population growth\npressures could have influenced the figures, the difference is most likely due to the different definition of improved water points\nthat were adopted by the two databases. Under the WSDP, the WPM have since adopted the definition of improved water points\nthat is consistent with the internationally accepted definition by WHO/UNICEF (2000). The figures presented in JMP (2015)\nwould be based on this definition.\n6 Goals that were set partly through the commitments of the National Strategy for Growth and Reduction of Poverty also known\nas MKUKUTA and by the Millennium Development Goals (MDG) respectively.\n\n\n4\n\nMention: \"2002 Housing Census\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Its figures exceed the later international estimate.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007640:5:0", "occurrence_id": "c09bf5d7d2f0f3562093df3d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: performance for 775 cities, from 2000 to 2015. It includes capital cities and cities with at least 400,000\n\n\ninhabitants. located in 140 countries across all regions and all income levels. <sup>4</sup> [^4: As mentioned in the World Bank Report (2016), income levels are defined by the World Bank Group’s per capita\nGDP cut‐off lines as of 2012: low income (below $1,035), lower‐middle income ($1,036‐$4,085), upper‐middle\nincome ($4,086‐$12,615), and high income (above $12,616).]\n\n\nSecond, we combine Oxford Economics Dataset with city centroids data from the Global Rural‐Urban\n\n\nMapping Project data set (GRUMP) to geo‐reference the cities. Since we did not have the associated\n\n\ngeographic limits for each of the cities in the original Oxford Economics Dataset, we chose to match each\n\n\ncity with the corresponding city centroid to avoid using arbitrary limits. Since there was no common\n\n\nidentification for the cities in both data sets, we use the name of the city (and country) for matching\n\n\npurposes. Because of this limitation, not all cities were perfectly matched. After manually fixing more than\n\n\n200 city names to improve the matching, we end up with 769 geo‐referenced cities, distributed by region\n\n\nas follows: 27.5% in East Asia Pacific, 22.5% in OECD countries, 12.5% in Latin America, 11.5% in South\n\n\nAsia, 9.5% in Sub‐Saharan Africa, 9% in Europe and Central Asia, and 6.5% in Middle East and North Africa.\n\n\nThe six cities in the Oxford Dataset that we were unable to match are: Baixada Santista and Norte\n\n\nCatarinense (Brazil), Erdos (China), Male (Maldives), Marseille (France), and Podgorica (Montenegro).\n\n\nThird,\n\nMention: \"Oxford Economics Dataset\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"City-level output and income data combined with coordinates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007640:5:0", "occurrence_id": "d1f536d3e4a055faff6df892", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: performance for 775 cities, from 2000 to 2015. It includes capital cities and cities with at least 400,000\n\n\ninhabitants. located in 140 countries across all regions and all income levels. <sup>4</sup> [^4: As mentioned in the World Bank Report (2016), income levels are defined by the World Bank Group’s per capita\nGDP cut‐off lines as of 2012: low income (below $1,035), lower‐middle income ($1,036‐$4,085), upper‐middle\nincome ($4,086‐$12,615), and high income (above $12,616).]\n\n\nSecond, we combine Oxford Economics Dataset with city centroids data from the Global Rural‐Urban\n\n\nMapping Project data set (GRUMP) to geo‐reference the cities. Since we did not have the associated\n\n\ngeographic limits for each of the cities in the original Oxford Economics Dataset, we chose to match each\n\n\ncity with the corresponding city centroid to avoid using arbitrary limits. Since there was no common\n\n\nidentification for the cities in both data sets, we use the name of the city (and country) for matching\n\n\npurposes. Because of this limitation, not all cities were perfectly matched. After manually fixing more than\n\n\n200 city names to improve the matching, we end up with 769 geo‐referenced cities, distributed by region\n\n\nas follows: 27.5% in East Asia Pacific, 22.5% in OECD countries, 12.5% in Latin America, 11.5% in South\n\n\nAsia, 9.5% in Sub‐Saharan Africa, 9% in Europe and Central Asia, and 6.5% in Middle East and North Africa.\n\n\nThe six cities in the Oxford Dataset that we were unable to match are: Baixada Santista and Norte\n\n\nCatarinense (Brazil), Erdos (China), Male (Maldives), Marseille (France), and Podgorica (Montenegro).\n\n\nThird,\n\nMention: \"Global Rural‐Urban\n\n\nMapping Project data set\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"City centroids used to geo-reference the sample.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007640:5:0", "occurrence_id": "cae395077c38a942288e6e0b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: performance for 775 cities, from 2000 to 2015. It includes capital cities and cities with at least 400,000\n\n\ninhabitants. located in 140 countries across all regions and all income levels. <sup>4</sup> [^4: As mentioned in the World Bank Report (2016), income levels are defined by the World Bank Group’s per capita\nGDP cut‐off lines as of 2012: low income (below $1,035), lower‐middle income ($1,036‐$4,085), upper‐middle\nincome ($4,086‐$12,615), and high income (above $12,616).]\n\n\nSecond, we combine Oxford Economics Dataset with city centroids data from the Global Rural‐Urban\n\n\nMapping Project data set (GRUMP) to geo‐reference the cities. Since we did not have the associated\n\n\ngeographic limits for each of the cities in the original Oxford Economics Dataset, we chose to match each\n\n\ncity with the corresponding city centroid to avoid using arbitrary limits. Since there was no common\n\n\nidentification for the cities in both data sets, we use the name of the city (and country) for matching\n\n\npurposes. Because of this limitation, not all cities were perfectly matched. After manually fixing more than\n\n\n200 city names to improve the matching, we end up with 769 geo‐referenced cities, distributed by region\n\n\nas follows: 27.5% in East Asia Pacific, 22.5% in OECD countries, 12.5% in Latin America, 11.5% in South\n\n\nAsia, 9.5% in Sub‐Saharan Africa, 9% in Europe and Central Asia, and 6.5% in Middle East and North Africa.\n\n\nThe six cities in the Oxford Dataset that we were unable to match are: Baixada Santista and Norte\n\n\nCatarinense (Brazil), Erdos (China), Male (Maldives), Marseille (France), and Podgorica (Montenegro).\n\n\nThird,\n\nMention: \"Oxford Dataset\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Six cities could not be matched to the other sources.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007607:8:0", "occurrence_id": "922bcdca39a3cf542234eca7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Existing sources of publicly available data are rather limited with respect to these three criteria. We\nresorted to the LIS Cross-National Data Center in Luxembourg (http://www.lisdatacenter.org/),\nwhich allowed us to process data from four countries (Italy, Germany, France and Switzerland), while\na fifth country was obtained from accessing the original provider (United Kingdom –\n<u>https://www.understandingsociety.ac.uk/).</u>\n\nThe surveys we have used are therefore the following:\n**Italy:** Survey on Household Incomes and Wealth (SHIW), collected by the Bank of Italy – 11 surveys,\n\ncovering the period 1993-2014 (information on parental background is not available before the\nstarting date – originally consisting of 112,690 individuals, which reduces to 107,846 when\nconsidering non-missing information.\n**Germany:** German Socio-economic Panel (SOEP) – 11 surveys, covering the period 1984-2013 –\n\noriginally including 156,338 individuals, then reduced to 133,467 in case of non-missing\ninformation.\n**France:** Household Budget Survey (HBS), conducted by the Banque de France) – 6 surveys, covering\n\nthe period 1978-2005 – originally consisting of 97,306 individuals, declining to 89,119 when\nmissing information is excluded.\n**Switzerland:** Swiss Household Panel (SHP) – 6 surveys, covering the period 1999-2014 – originally\n\nconsisting of 43,102 individuals, which then decline to 31,273 valid observations.\n**United Kingdom:** starts as British Household Panel (BHPS), replaced after 2009 by the\n\nUnderstanding Society-Household Longitudinal Survey (UKHLS) – considers 24 waves over\nthe period 1991-2014 – originally consisting of 434,253 individuals, which then decline to\n308,625 valid observations.\n\nOur selection rules include\n\nMention: \"LIS Cross-National Data Center\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The centre granted access to process the harmonised national surveys.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007607:8:0", "occurrence_id": "b4845db92e8ef2b7981af9cb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Existing sources of publicly available data are rather limited with respect to these three criteria. We\nresorted to the LIS Cross-National Data Center in Luxembourg (http://www.lisdatacenter.org/),\nwhich allowed us to process data from four countries (Italy, Germany, France and Switzerland), while\na fifth country was obtained from accessing the original provider (United Kingdom –\n<u>https://www.understandingsociety.ac.uk/).</u>\n\nThe surveys we have used are therefore the following:\n**Italy:** Survey on Household Incomes and Wealth (SHIW), collected by the Bank of Italy – 11 surveys,\n\ncovering the period 1993-2014 (information on parental background is not available before the\nstarting date – originally consisting of 112,690 individuals, which reduces to 107,846 when\nconsidering non-missing information.\n**Germany:** German Socio-economic Panel (SOEP) – 11 surveys, covering the period 1984-2013 –\n\noriginally including 156,338 individuals, then reduced to 133,467 in case of non-missing\ninformation.\n**France:** Household Budget Survey (HBS), conducted by the Banque de France) – 6 surveys, covering\n\nthe period 1978-2005 – originally consisting of 97,306 individuals, declining to 89,119 when\nmissing information is excluded.\n**Switzerland:** Swiss Household Panel (SHP) – 6 surveys, covering the period 1999-2014 – originally\n\nconsisting of 43,102 individuals, which then decline to 31,273 valid observations.\n**United Kingdom:** starts as British Household Panel (BHPS), replaced after 2009 by the\n\nUnderstanding Society-Household Longitudinal Survey (UKHLS) – considers 24 waves over\nthe period 1991-2014 – originally consisting of 434,253 individuals, which then decline to\n308,625 valid observations.\n\nOur selection rules include\n\nMention: \"Survey on Household Incomes and Wealth\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Eleven surveys across this period for one country.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007607:8:0", "occurrence_id": "22e51570a7ce0ffd747fcada", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Existing sources of publicly available data are rather limited with respect to these three criteria. We\nresorted to the LIS Cross-National Data Center in Luxembourg (http://www.lisdatacenter.org/),\nwhich allowed us to process data from four countries (Italy, Germany, France and Switzerland), while\na fifth country was obtained from accessing the original provider (United Kingdom –\n<u>https://www.understandingsociety.ac.uk/).</u>\n\nThe surveys we have used are therefore the following:\n**Italy:** Survey on Household Incomes and Wealth (SHIW), collected by the Bank of Italy – 11 surveys,\n\ncovering the period 1993-2014 (information on parental background is not available before the\nstarting date – originally consisting of 112,690 individuals, which reduces to 107,846 when\nconsidering non-missing information.\n**Germany:** German Socio-economic Panel (SOEP) – 11 surveys, covering the period 1984-2013 –\n\noriginally including 156,338 individuals, then reduced to 133,467 in case of non-missing\ninformation.\n**France:** Household Budget Survey (HBS), conducted by the Banque de France) – 6 surveys, covering\n\nthe period 1978-2005 – originally consisting of 97,306 individuals, declining to 89,119 when\nmissing information is excluded.\n**Switzerland:** Swiss Household Panel (SHP) – 6 surveys, covering the period 1999-2014 – originally\n\nconsisting of 43,102 individuals, which then decline to 31,273 valid observations.\n**United Kingdom:** starts as British Household Panel (BHPS), replaced after 2009 by the\n\nUnderstanding Society-Household Longitudinal Survey (UKHLS) – considers 24 waves over\nthe period 1991-2014 – originally consisting of 434,253 individuals, which then decline to\n308,625 valid observations.\n\nOur selection rules include\n\nMention: \"German Socio-economic Panel\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Eleven surveys covering this span with over 150,000 individuals.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007607:8:0", "occurrence_id": "fd87b32c2367c1f35a9e44f9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Existing sources of publicly available data are rather limited with respect to these three criteria. We\nresorted to the LIS Cross-National Data Center in Luxembourg (http://www.lisdatacenter.org/),\nwhich allowed us to process data from four countries (Italy, Germany, France and Switzerland), while\na fifth country was obtained from accessing the original provider (United Kingdom –\n<u>https://www.understandingsociety.ac.uk/).</u>\n\nThe surveys we have used are therefore the following:\n**Italy:** Survey on Household Incomes and Wealth (SHIW), collected by the Bank of Italy – 11 surveys,\n\ncovering the period 1993-2014 (information on parental background is not available before the\nstarting date – originally consisting of 112,690 individuals, which reduces to 107,846 when\nconsidering non-missing information.\n**Germany:** German Socio-economic Panel (SOEP) – 11 surveys, covering the period 1984-2013 –\n\noriginally including 156,338 individuals, then reduced to 133,467 in case of non-missing\ninformation.\n**France:** Household Budget Survey (HBS), conducted by the Banque de France) – 6 surveys, covering\n\nthe period 1978-2005 – originally consisting of 97,306 individuals, declining to 89,119 when\nmissing information is excluded.\n**Switzerland:** Swiss Household Panel (SHP) – 6 surveys, covering the period 1999-2014 – originally\n\nconsisting of 43,102 individuals, which then decline to 31,273 valid observations.\n**United Kingdom:** starts as British Household Panel (BHPS), replaced after 2009 by the\n\nUnderstanding Society-Household Longitudinal Survey (UKHLS) – considers 24 waves over\nthe period 1991-2014 – originally consisting of 434,253 individuals, which then decline to\n308,625 valid observations.\n\nOur selection rules include\n\nMention: \"Household Budget Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Six surveys conducted by the national central bank.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007607:8:0", "occurrence_id": "1018b6d9dc177b1943dfb161", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Existing sources of publicly available data are rather limited with respect to these three criteria. We\nresorted to the LIS Cross-National Data Center in Luxembourg (http://www.lisdatacenter.org/),\nwhich allowed us to process data from four countries (Italy, Germany, France and Switzerland), while\na fifth country was obtained from accessing the original provider (United Kingdom –\n<u>https://www.understandingsociety.ac.uk/).</u>\n\nThe surveys we have used are therefore the following:\n**Italy:** Survey on Household Incomes and Wealth (SHIW), collected by the Bank of Italy – 11 surveys,\n\ncovering the period 1993-2014 (information on parental background is not available before the\nstarting date – originally consisting of 112,690 individuals, which reduces to 107,846 when\nconsidering non-missing information.\n**Germany:** German Socio-economic Panel (SOEP) – 11 surveys, covering the period 1984-2013 –\n\noriginally including 156,338 individuals, then reduced to 133,467 in case of non-missing\ninformation.\n**France:** Household Budget Survey (HBS), conducted by the Banque de France) – 6 surveys, covering\n\nthe period 1978-2005 – originally consisting of 97,306 individuals, declining to 89,119 when\nmissing information is excluded.\n**Switzerland:** Swiss Household Panel (SHP) – 6 surveys, covering the period 1999-2014 – originally\n\nconsisting of 43,102 individuals, which then decline to 31,273 valid observations.\n**United Kingdom:** starts as British Household Panel (BHPS), replaced after 2009 by the\n\nUnderstanding Society-Household Longitudinal Survey (UKHLS) – considers 24 waves over\nthe period 1991-2014 – originally consisting of 434,253 individuals, which then decline to\n308,625 valid observations.\n\nOur selection rules include\n\nMention: \"Swiss Household Panel\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Six surveys across this span.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007607:8:0", "occurrence_id": "9182a205c579202cb27af679", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Existing sources of publicly available data are rather limited with respect to these three criteria. We\nresorted to the LIS Cross-National Data Center in Luxembourg (http://www.lisdatacenter.org/),\nwhich allowed us to process data from four countries (Italy, Germany, France and Switzerland), while\na fifth country was obtained from accessing the original provider (United Kingdom –\n<u>https://www.understandingsociety.ac.uk/).</u>\n\nThe surveys we have used are therefore the following:\n**Italy:** Survey on Household Incomes and Wealth (SHIW), collected by the Bank of Italy – 11 surveys,\n\ncovering the period 1993-2014 (information on parental background is not available before the\nstarting date – originally consisting of 112,690 individuals, which reduces to 107,846 when\nconsidering non-missing information.\n**Germany:** German Socio-economic Panel (SOEP) – 11 surveys, covering the period 1984-2013 –\n\noriginally including 156,338 individuals, then reduced to 133,467 in case of non-missing\ninformation.\n**France:** Household Budget Survey (HBS), conducted by the Banque de France) – 6 surveys, covering\n\nthe period 1978-2005 – originally consisting of 97,306 individuals, declining to 89,119 when\nmissing information is excluded.\n**Switzerland:** Swiss Household Panel (SHP) – 6 surveys, covering the period 1999-2014 – originally\n\nconsisting of 43,102 individuals, which then decline to 31,273 valid observations.\n**United Kingdom:** starts as British Household Panel (BHPS), replaced after 2009 by the\n\nUnderstanding Society-Household Longitudinal Survey (UKHLS) – considers 24 waves over\nthe period 1991-2014 – originally consisting of 434,253 individuals, which then decline to\n308,625 valid observations.\n\nOur selection rules include\n\nMention: \"British Household Panel\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The starting British panel before its replacement.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007607:8:0", "occurrence_id": "afca0543539f11038efd902b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Existing sources of publicly available data are rather limited with respect to these three criteria. We\nresorted to the LIS Cross-National Data Center in Luxembourg (http://www.lisdatacenter.org/),\nwhich allowed us to process data from four countries (Italy, Germany, France and Switzerland), while\na fifth country was obtained from accessing the original provider (United Kingdom –\n<u>https://www.understandingsociety.ac.uk/).</u>\n\nThe surveys we have used are therefore the following:\n**Italy:** Survey on Household Incomes and Wealth (SHIW), collected by the Bank of Italy – 11 surveys,\n\ncovering the period 1993-2014 (information on parental background is not available before the\nstarting date – originally consisting of 112,690 individuals, which reduces to 107,846 when\nconsidering non-missing information.\n**Germany:** German Socio-economic Panel (SOEP) – 11 surveys, covering the period 1984-2013 –\n\noriginally including 156,338 individuals, then reduced to 133,467 in case of non-missing\ninformation.\n**France:** Household Budget Survey (HBS), conducted by the Banque de France) – 6 surveys, covering\n\nthe period 1978-2005 – originally consisting of 97,306 individuals, declining to 89,119 when\nmissing information is excluded.\n**Switzerland:** Swiss Household Panel (SHP) – 6 surveys, covering the period 1999-2014 – originally\n\nconsisting of 43,102 individuals, which then decline to 31,273 valid observations.\n**United Kingdom:** starts as British Household Panel (BHPS), replaced after 2009 by the\n\nUnderstanding Society-Household Longitudinal Survey (UKHLS) – considers 24 waves over\nthe period 1991-2014 – originally consisting of 434,253 individuals, which then decline to\n308,625 valid observations.\n\nOur selection rules include\n\nMention: \"Understanding Society-Household Longitudinal Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Twenty-four waves across this span, succeeding the earlier panel.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007649:8:1", "occurrence_id": "854c369be2ba2d2cf2cd92ab", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \n2000).\n\n\n2.3 Cost Effectiveness of Learning Interventions\n\n\nBuilding on the impact metrics, this paper also presents cost‐effectiveness analysis and benefit‐cost ratios\nof a range of programs with available cost data from J‐PAL (2014). The cost‐effectiveness of selected\ninterventions is estimated in terms of EYOS per US$100. For instance, we first multiply the point estimate\nof improved learning by ∆𝐿 to get the amount of EYOS for each participant by the program. We then divide\nthe present value of total costs of the program (including costs to both the implementer and the\nbeneficiary) by the number of participants to determine the program cost per beneficiary. And we divide\nthe amount of EYOS by the cost per individual, then multiply by 100 to express the outcome per a US$100\ninvestment.\n\n\nBenefit‐cost ratios or return to investment (ROI) ratios are calculated as the NPV of lifetime increased\nwage income divided by the NPV of the program costs per individual. All conversions and calculations\nfollow the methodology described previously in this section.\n\n##### 3. Data\n\n\nThe primary data used in this paper are the World Bank’s STEP Skills Measurement Program. The STEP\nhousehold surveys measure skills of the adult population in low‐ and middle‐income countries and provide\ncomparable international scores of reading proficiency on the scale of OECD’s assessment of adult skills,\nthe PIAAC (World Bank 2016a). We use the STEP survey data collected between March 2012 and July 2014\nin five countries with available literacy assessment data: Bolivia, Colombia, Ghana, Kenya, and Vietnam. <sup>9</sup> [^9: Data are also available for Armenia, Ukraine and Georgia but not included for the reason that over 50 percent of\nthe population in these countries have more than 12 years of formal education, which dramatically reduces the\nsample size for the analysis in this paper.]\n\n\n8 Wages are highly likely to increase over time and assuming no wage growth is\n\nMention: \"STEP Skills Measurement Program\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The primary data measuring adult skills in lower-income countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007649:8:1", "occurrence_id": "12b7c5744e4b5cc46f5c9dd4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \n2000).\n\n\n2.3 Cost Effectiveness of Learning Interventions\n\n\nBuilding on the impact metrics, this paper also presents cost‐effectiveness analysis and benefit‐cost ratios\nof a range of programs with available cost data from J‐PAL (2014). The cost‐effectiveness of selected\ninterventions is estimated in terms of EYOS per US$100. For instance, we first multiply the point estimate\nof improved learning by ∆𝐿 to get the amount of EYOS for each participant by the program. We then divide\nthe present value of total costs of the program (including costs to both the implementer and the\nbeneficiary) by the number of participants to determine the program cost per beneficiary. And we divide\nthe amount of EYOS by the cost per individual, then multiply by 100 to express the outcome per a US$100\ninvestment.\n\n\nBenefit‐cost ratios or return to investment (ROI) ratios are calculated as the NPV of lifetime increased\nwage income divided by the NPV of the program costs per individual. All conversions and calculations\nfollow the methodology described previously in this section.\n\n##### 3. Data\n\n\nThe primary data used in this paper are the World Bank’s STEP Skills Measurement Program. The STEP\nhousehold surveys measure skills of the adult population in low‐ and middle‐income countries and provide\ncomparable international scores of reading proficiency on the scale of OECD’s assessment of adult skills,\nthe PIAAC (World Bank 2016a). We use the STEP survey data collected between March 2012 and July 2014\nin five countries with available literacy assessment data: Bolivia, Colombia, Ghana, Kenya, and Vietnam. <sup>9</sup> [^9: Data are also available for Armenia, Ukraine and Georgia but not included for the reason that over 50 percent of\nthe population in these countries have more than 12 years of formal education, which dramatically reduces the\nsample size for the analysis in this paper.]\n\n\n8 Wages are highly likely to increase over time and assuming no wage growth is\n\nMention: \"STEP\nhousehold surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"They measure skills of the adult population and provide context variables.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007649:8:1", "occurrence_id": "6665bac2c172c4e3d1a77063", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \n2000).\n\n\n2.3 Cost Effectiveness of Learning Interventions\n\n\nBuilding on the impact metrics, this paper also presents cost‐effectiveness analysis and benefit‐cost ratios\nof a range of programs with available cost data from J‐PAL (2014). The cost‐effectiveness of selected\ninterventions is estimated in terms of EYOS per US$100. For instance, we first multiply the point estimate\nof improved learning by ∆𝐿 to get the amount of EYOS for each participant by the program. We then divide\nthe present value of total costs of the program (including costs to both the implementer and the\nbeneficiary) by the number of participants to determine the program cost per beneficiary. And we divide\nthe amount of EYOS by the cost per individual, then multiply by 100 to express the outcome per a US$100\ninvestment.\n\n\nBenefit‐cost ratios or return to investment (ROI) ratios are calculated as the NPV of lifetime increased\nwage income divided by the NPV of the program costs per individual. All conversions and calculations\nfollow the methodology described previously in this section.\n\n##### 3. Data\n\n\nThe primary data used in this paper are the World Bank’s STEP Skills Measurement Program. The STEP\nhousehold surveys measure skills of the adult population in low‐ and middle‐income countries and provide\ncomparable international scores of reading proficiency on the scale of OECD’s assessment of adult skills,\nthe PIAAC (World Bank 2016a). We use the STEP survey data collected between March 2012 and July 2014\nin five countries with available literacy assessment data: Bolivia, Colombia, Ghana, Kenya, and Vietnam. <sup>9</sup> [^9: Data are also available for Armenia, Ukraine and Georgia but not included for the reason that over 50 percent of\nthe population in these countries have more than 12 years of formal education, which dramatically reduces the\nsample size for the analysis in this paper.]\n\n\n8 Wages are highly likely to increase over time and assuming no wage growth is\n\nMention: \"PIAAC\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The OECD scale on which reading proficiency is reported.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007649:8:1", "occurrence_id": "b5d04b6c9bea66b2e33a60bd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \n2000).\n\n\n2.3 Cost Effectiveness of Learning Interventions\n\n\nBuilding on the impact metrics, this paper also presents cost‐effectiveness analysis and benefit‐cost ratios\nof a range of programs with available cost data from J‐PAL (2014). The cost‐effectiveness of selected\ninterventions is estimated in terms of EYOS per US$100. For instance, we first multiply the point estimate\nof improved learning by ∆𝐿 to get the amount of EYOS for each participant by the program. We then divide\nthe present value of total costs of the program (including costs to both the implementer and the\nbeneficiary) by the number of participants to determine the program cost per beneficiary. And we divide\nthe amount of EYOS by the cost per individual, then multiply by 100 to express the outcome per a US$100\ninvestment.\n\n\nBenefit‐cost ratios or return to investment (ROI) ratios are calculated as the NPV of lifetime increased\nwage income divided by the NPV of the program costs per individual. All conversions and calculations\nfollow the methodology described previously in this section.\n\n##### 3. Data\n\n\nThe primary data used in this paper are the World Bank’s STEP Skills Measurement Program. The STEP\nhousehold surveys measure skills of the adult population in low‐ and middle‐income countries and provide\ncomparable international scores of reading proficiency on the scale of OECD’s assessment of adult skills,\nthe PIAAC (World Bank 2016a). We use the STEP survey data collected between March 2012 and July 2014\nin five countries with available literacy assessment data: Bolivia, Colombia, Ghana, Kenya, and Vietnam. <sup>9</sup> [^9: Data are also available for Armenia, Ukraine and Georgia but not included for the reason that over 50 percent of\nthe population in these countries have more than 12 years of formal education, which dramatically reduces the\nsample size for the analysis in this paper.]\n\n\n8 Wages are highly likely to increase over time and assuming no wage growth is\n\nMention: \"STEP survey data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Collected across this window in five countries with literacy measures.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007650:5:1", "occurrence_id": "0a62865bb1f6664095d6bb79", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: , and occupation; and the distribution of public and private\nsector earnings, and the public sector earnings premium across the earnings distribution. We use the\ncountry‐level regional and income classifications from the World Development Indicators (WDI)\ndatabase. A selected list of variables and their description is given in Annex 2.\n\n\nWe applied a variety of criteria to determine which surveys to include, to ensure sufficient sample sizes\nfor the main variables. We excluded from the sample surveys with more than 30 percent missing\nobservations on sector of employment (public sector or private sector), wages, or occupation. We also\nexcluded from the analysis countries with odd data; for example, in Haiti 100 percent of the\nrespondents claimed to be working in the private sector, implying that Haiti has no public sector, which\nis clearly wrong. We used the most recent acceptable survey available for each country.\n\n\nThe resulting data set contains surveys covering the period from 2009 to 2016, for which we have 91\ncountries with data on public employment and 68 countries with data on wages. These include 31\ncountries in Sub‐Saharan Africa (SSA), 16 in Latin America and the Caribbean (LAC), 15 in Eastern\nEurope and Central Asia (EECA), 8 in East Asia and the Pacific (EAP), 10 in high‐income Western Europe\nand North America, 6 in the Middle East and North Africa (MENA) and 5 in South Asia (SA). See Annex\n1 for the list of countries.\n\n\n1 <u>https://datacatalog.worldbank.org/dataset/worldwide‐bureaucracy‐indicators.</u>\n2 Where data exist for a country in the LIS but not in the I2D2, we use the LIS; where data exist for a country in\nboth the LIS and I2D2, we use the I2D2.\n\n\n4\n\nMention: \"World Development Indicators (WDI)\ndatabase\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Supplies the regional and income classifications.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007650:5:1", "occurrence_id": "c63f5aaa0478b4d95965103d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: , and occupation; and the distribution of public and private\nsector earnings, and the public sector earnings premium across the earnings distribution. We use the\ncountry‐level regional and income classifications from the World Development Indicators (WDI)\ndatabase. A selected list of variables and their description is given in Annex 2.\n\n\nWe applied a variety of criteria to determine which surveys to include, to ensure sufficient sample sizes\nfor the main variables. We excluded from the sample surveys with more than 30 percent missing\nobservations on sector of employment (public sector or private sector), wages, or occupation. We also\nexcluded from the analysis countries with odd data; for example, in Haiti 100 percent of the\nrespondents claimed to be working in the private sector, implying that Haiti has no public sector, which\nis clearly wrong. We used the most recent acceptable survey available for each country.\n\n\nThe resulting data set contains surveys covering the period from 2009 to 2016, for which we have 91\ncountries with data on public employment and 68 countries with data on wages. These include 31\ncountries in Sub‐Saharan Africa (SSA), 16 in Latin America and the Caribbean (LAC), 15 in Eastern\nEurope and Central Asia (EECA), 8 in East Asia and the Pacific (EAP), 10 in high‐income Western Europe\nand North America, 6 in the Middle East and North Africa (MENA) and 5 in South Asia (SA). See Annex\n1 for the list of countries.\n\n\n1 <u>https://datacatalog.worldbank.org/dataset/worldwide‐bureaucracy‐indicators.</u>\n2 Where data exist for a country in the LIS but not in the I2D2, we use the LIS; where data exist for a country in\nboth the LIS and I2D2, we use the I2D2.\n\n\n4\n\nMention: \"worldwide‐bureaucracy‐indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The catalogue behind the bureaucracy measures.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007650:5:1", "occurrence_id": "ea44025420445cd23a96a6a1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: , and occupation; and the distribution of public and private\nsector earnings, and the public sector earnings premium across the earnings distribution. We use the\ncountry‐level regional and income classifications from the World Development Indicators (WDI)\ndatabase. A selected list of variables and their description is given in Annex 2.\n\n\nWe applied a variety of criteria to determine which surveys to include, to ensure sufficient sample sizes\nfor the main variables. We excluded from the sample surveys with more than 30 percent missing\nobservations on sector of employment (public sector or private sector), wages, or occupation. We also\nexcluded from the analysis countries with odd data; for example, in Haiti 100 percent of the\nrespondents claimed to be working in the private sector, implying that Haiti has no public sector, which\nis clearly wrong. We used the most recent acceptable survey available for each country.\n\n\nThe resulting data set contains surveys covering the period from 2009 to 2016, for which we have 91\ncountries with data on public employment and 68 countries with data on wages. These include 31\ncountries in Sub‐Saharan Africa (SSA), 16 in Latin America and the Caribbean (LAC), 15 in Eastern\nEurope and Central Asia (EECA), 8 in East Asia and the Pacific (EAP), 10 in high‐income Western Europe\nand North America, 6 in the Middle East and North Africa (MENA) and 5 in South Asia (SA). See Annex\n1 for the list of countries.\n\n\n1 <u>https://datacatalog.worldbank.org/dataset/worldwide‐bureaucracy‐indicators.</u>\n2 Where data exist for a country in the LIS but not in the I2D2, we use the LIS; where data exist for a country in\nboth the LIS and I2D2, we use the I2D2.\n\n\n4\n\nMention: \"I2D2\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Preferred over the alternative where both cover a country.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007656:45:1", "occurrence_id": "66e79ac9432f8dec73f3601b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  518-50\n\n\n[43] glennerster.r, t.suri and s.bhogale (2016) The Implications of the Ebola Outbreak\n\n\non Markets, Traders, and Food Security in Sierra Leone, IGC Policy Brief.\n\n\n[44] goldin.c and l.f.katz (2002) “The Power of the Pill: Oral Contraceptives and Women’s\n\n\nCareer and Marriage Decisions,” _Journal_ _of_ _Political_ _Economy_ 110: 730-70.\n\n\n[45] gollier.c and j.w.pratt (1996) “Risk Vulnerability and the Tempering Effect of Back\n\nground Risk,” _Econometrica_ 64: 1109-23.\n\n\n[46] guarnieri.e and h.rainer (2018) Female Empowerment and Male Backlash, CESifo\n\n\nWP7009.\n\n\n[47] haushofer.j, j.shapiro, c.ringdal, x.y.wang (2018) Income Changes and Intimate\n\n\nPartner Violence: Evidence from Unconditional Cash Transfers in Kenya, mimeo Princeton.\n\n\n[48] hes.s, d.jaimovich and m.scheundeln (2018) Development Projects and Economic Net\n\nworks: Lessons From Rural Gambia, mimeo Goethe University.\n\n\n[49] himelein.k, m.testaverde, a.turay and s.turay (2015) The Socio-economic Impacts\n\n\nof Ebola in Sierra Leone: Results from a High Frequency Cell Phone Survey (Round Three),\n\n\nWashington DC: The World Bank.\n\n\n43\n\nMention: \"High Frequency Cell Phone Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its third round reports the socioeconomic impacts of the epidemic.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007659:4:0", "occurrence_id": "994f891c17302fb38590eba0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: During the global financial crisis we observe an increase in the poverty headcount in a number of low\n\nincome and lower-middle-income economies. For example, the average poverty headcount ratio at $1.90 in\n\n\nlow-income countries increases 11 percentage points — from 48.6% in 2009 to 60% in 2013. Some middle\n\nincome countries, including Croatia, Georgia and Romania in Eastern Europe and Honduras and Panama in\n\n\nCentral America, experienced a significant increase in the poverty headcount.\n\n\nWe use the World Bank’s data on new business formation to investigate the channel through which\n\n\nbusiness regulation may affect poverty. These annual data are collected from 143 company registrars on the\n\n\nnumber of newly registered firms. New business entry is defined as the number of newly registered corporations\n\n\nper 1,000 working-age people (those ages 15–64). The units of measurement are private, formal sector\n\n\ncompanies with limited liability. Data extend from 2005 to 2015.\n\n\nAs proxy for the regulatory attitude of governments, we use the World Bank’s Doing Business data,\n\n\nfirst published in 2003. The methodology for data collection is described in Djankov et al. (2002) and updated\n\n\nin Djankov (2016). We use the Doing Business data in two ways. First, we employ the overall score, which\n\n\ncombines 10 distinct groups of measures that constitute the Doing Business index. These relate to the legal\n\n\nprotections and regulatory processes that an average private business needs to go through in order to operate\n\n\n(Table 1).\n\n\nSecond, we use measures in four areas – starting a business, acquiring licenses, getting credit and\n\n\nenforcing contracts – as specific measures of business regulation. All measures are positively correlated, with\n\n\nthe correlation ranging between 0.1810 and 0.7430, with low correlation between contract enforcement and\n\n\nacquiring licenses and high correlation between getting credit and the total aggregate _Doing Business_ measure.\n\n\nProperty registration has the highest correlation with contract enforcement (0.4947\n\nMention: \"Doing Business data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"A proxy for regulatory attitude, first published in that year.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007660:56:0", "occurrence_id": "395aeb0ca8e95285c6da3746", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Table 10: First stage regressions corresponding to Table 4\n\n\n(1)\nBlack population growth\n\n\n_Panel_ _A:_ _Population_ _growth_ _rates_ _(1991-1996)_\nlog distance -0.940\n<u>(0.593)</u>\n\n\n_Panel_ _B:_ _Population_ _growth_ _rates_ _(1991-2001)_\nlog distance -0.007 <sup>_∗∗∗_</sup>\n\n<u>(0.001)</u>\n\n\n_Panel_ _C:_ _Population_ _growth_ _rates_ _(1991-2011)_\nlog distance -0.004 <sup>_∗∗∗_</sup>\n\n<u>(0.001)</u>\n\n\nFixed effects Yes\n\n\nControls Yes\n\n\nObservations 2093\n\n\n_Notes._ This Table displays estimates of equation (1) in the main\ntext. Each cell presents estimates from a separate regression. The\nstandard errors are clustered on the municipality level and presented\nin parentheses. There are 201 clusters. All columns are estimated\nusing OLS where the natural log of distance to the nearest homeland\nis the variable of interest. The outcome variable is absolute black\npopulation growth in the relevant time period divided by the overall\npopulation. The relevant time periods are 1991-1996 in Panel A,\n1991-2001 in Panel B and 1991-2011 in Panel C. Controls include\nvariables on education, income, population group, population density\nand employment in 1991. There are nine provinces for which fixed\neffects are included. Coefficients that are statistically significant at\nthe 90% level of confidence are marked with a *; at the 95% level, a\n**; and at the 99% level, a ***. Standard errors in parentheses.\n_Source_ : Authors’ analysis based on South African census data.\n\n\n55\n\nMention: \"South African census data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Subnational outcomes are analysed from these enumerations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007661:30:0", "occurrence_id": "1726ff030e059fedc00dcfff", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 19. Kakwani indices by types of interventions in Ukraine and peers.**\n\n1.0\n\n\n0.8\n\n\n0.6\n\n\n0.4\n\n\n0.2\n\n\n0.0\n\n\n-0.2\n\n\n\n-0.4\n\n\n\nDirect Taxes Pension\nContributions\n\n\n\nPension Pensions Direct Indirect Taxes Indirect In-kind Health In-kind\nContributions Transfers Subsidies Education\n\nBelarus-2015 Poland-2014 Russia-2014 Ukraine-2016\n\n\n\nPensions Direct\nTransfers\n\n\n\nIndirect Taxes Indirect\nSubsidies\n\n\n\nNote: Kakwani index is a measure of progressivity. The higher the value is the more progressive is the transfer.\nSource: authors’ calculations using HLCS-2016 data for Ukraine and respective CEQ analysis for other countries: Belarus – Bornukova et\nal., 2017; Poland - <mark>Goraus and Inchauste, 2016; Russia – Popova et al., 2018.</mark>\n\n\n**6.** **Conclusions**\n\n\nThe fiscal incidence analysis based on the CEQ methodology reveals that the tax-benefit system in Ukraine is quite\npro-poor in the sense that the first 6 deciles of population are net recipients from the system and the top 4 deciles\nare net payers into the system if pensions are treated as government transfers. The poorest decile almost completely\ndepends on transfers and other benefits., first of all on pensions. Most of the components of the tax-benefit system\nin Ukraine (pensions, other direct transfers, direct taxes and pension contributions and in-kind health benefits) are\nprogressive in the sense that their share in final income decreases (increases in case of taxes) with market income\nlevel. Indirect taxes, subsidies and in-kind education benefits have less pronounced redistributive effects by\ncomparison. The relative importance of each intervention varies significantly by income groups. If decomposed\nby types of households, there are only two categories of net payers - households with no children and no\npensioners, and families that consist of\n\nMention: \"HLCS-2016 data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Transfer progressivity is calculated from this round for one country.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007668:16:1", "occurrence_id": "cac2e1cb855f27ddf3e2ba84", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  dollars)_\n\n\n40\n35\n30\n25\n20\n15\n10\n5\n0\n\n\n\n\n\n\n\n\n\n\n\n**Labor productivity**\n_(Median, in thousands of 2009 dollars)_\n\n\n8\n\n6\n\n4\n\n2\n\n0\n\n\n\n\n\n\n\n\n\n\n\n\n\n_Source:_ Calculations using World Bank Enterprise Survey data.\n_Note:_ Labor productivity is the ratio of value added by the number of workers.\n\n\nThese conditions are reflected in lower productivity premia for pioneers in FCS: higher returns are\noffset by important costs and uncertainty. Estimates from the World Bank Enterprise Survey\nsuggest that the productivity of pioneering firms in FCS is lower than that of their peers in nonFCS. As illustrated in Figure 4, the median value added per worker in pioneering firms in FCS\n($6,986) is lower than that of pioneering firms in non-FCS ($11,651).\n\n\n15\n\nMention: \"World Bank Enterprise Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its estimates show pioneering firms in fragile settings are less productive.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007671:25:0", "occurrence_id": "c7c02f043ca000c530c4432a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: criminal activity and corruption. <sup>67</sup> [^67: Project Wickenby in Australia illustrates what can be achieved by a coordinated cross-agency approach to the\nproblem of offshore tax evasion. For more information visit the ATO website:\nhttps://www.ato.gov.au/General/The-fight-against-tax-crime/News-and-results/Project-Wickenby-hasdelivered/.] Tax administrations can also benefit from the exchange of\ninformation about the types of structures and entities that are being exploited for the purposes\nof offshore tax evasion. This type of intelligence sharing is something that tax administrations\nthat are members of the OECD’s Forum on Tax Administration (FTA) have undertaken and\nthe FTA’s current work program includes a project on “Tackling offshore evasion: intelligent\nuse and assurance of CRS data”. <sup>68</sup> [^68: For more information about the FTA and its current projects visit: http://www.oecd.org/tax/forum-on-taxadministration/about/.] Regional groupings of tax administrations, such as the\nAfrican Tax Administration Forum, the Inter-American Center of Tax Administrations and the\nIntra-European Organisation of Tax Administrations, also help to encourage the sharing of\nintelligence between their members, which include many developing economies.\n\n\nIn addition to the data that are being exchanged under the Common Reporting Standard,\ncountries are beginning to develop registers of beneficial ownership. The UK government plans\nto launch a public beneficial ownership register in 2021. EU member states will make public\nthe ownership of legal entities such as companies by January 2020 and other countries around\nthe world have also committed to put in place public registers of ownership. <sup>69</sup> [^69: For a recent review of registers of beneficial ownership see:\nhttps://researchbriefings.parliament.uk/ResearchBriefing/Summary/CBP-8259#fullreport.] Increased\ntransparency about ownership will clearly help tax administrations across the world to detect\nand deter offshore tax evasion.\n\n\nAddressing profit diversion by MNEs is more complex as\n\nMention: \"CRS data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"A project on the intelligent use and assurance of these exchange data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007665:4:6", "occurrence_id": "f5a4332208ab5170640dcd90", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: /sup> <sup>**𝑘**</sup> <sup>𝐸</sup> <sup>**𝑗**</sup> **𝑗** **𝑗** **𝑗**\n\nwhich I use to estimate the parameters by employing the iterated non-linear seemingly unrelated 𝛾 **𝑗** 𝑃 **𝑟** 𝑃 **𝑟**\nregression method. <sup>4</sup> [^4: Details in chapter 14.9 of Greene (2011).] I choose industry as the base sector and transform the constrained\nparameters such that the estimation becomes unconstrained. <sup>5</sup> [^5: The appendix contains details of the parameter restrictions and transformation scheme.]\n\nThe quantitative analysis requires data on intermediate shares and sectoral prices. I use\nthe OECD input-output database (OECD, 2017b) to collect the intermediate shares for many\ncountries. To obtain the sectoral prices, I utilize the final consumption expenditure data from the\nOECD national accounts database (OECD, 2017a). The database contains expenditures by sectors\nin both current and constant prices (country-specific base year) in national currencies, which I\nuse to construct price series indexed to unity for a base year for each country. Finally, I adjust the\nprice series using the harmonized price data for 2005 from the GGDC Productivity Level Database\n(Inklaar & Timmer, 2014) to make prices comparable across countries and time. A limitation of\nthe price series is that it corresponds to the consumption side of the economy. The consumer\nprices differ from the producer prices as the former include certain margins and distortions.\n\nAll sources contain disaggregated data into finer sector classification, which I aggregate\nup to three sectors – agriculture (a), industry (m) and services (s), using the International\nStandard Industrial Classification of All Economic Activities, Revision 3 (ISIC Rev. 3.1, 2002). The\naggregate expenditure is normalized to\n\nMention: \"OECD input-output database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Intermediate shares across many countries come from this database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007665:4:6", "occurrence_id": "f21ec2fde850100b0e2b4cd4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: /sup> <sup>**𝑘**</sup> <sup>𝐸</sup> <sup>**𝑗**</sup> **𝑗** **𝑗** **𝑗**\n\nwhich I use to estimate the parameters by employing the iterated non-linear seemingly unrelated 𝛾 **𝑗** 𝑃 **𝑟** 𝑃 **𝑟**\nregression method. <sup>4</sup> [^4: Details in chapter 14.9 of Greene (2011).] I choose industry as the base sector and transform the constrained\nparameters such that the estimation becomes unconstrained. <sup>5</sup> [^5: The appendix contains details of the parameter restrictions and transformation scheme.]\n\nThe quantitative analysis requires data on intermediate shares and sectoral prices. I use\nthe OECD input-output database (OECD, 2017b) to collect the intermediate shares for many\ncountries. To obtain the sectoral prices, I utilize the final consumption expenditure data from the\nOECD national accounts database (OECD, 2017a). The database contains expenditures by sectors\nin both current and constant prices (country-specific base year) in national currencies, which I\nuse to construct price series indexed to unity for a base year for each country. Finally, I adjust the\nprice series using the harmonized price data for 2005 from the GGDC Productivity Level Database\n(Inklaar & Timmer, 2014) to make prices comparable across countries and time. A limitation of\nthe price series is that it corresponds to the consumption side of the economy. The consumer\nprices differ from the producer prices as the former include certain margins and distortions.\n\nAll sources contain disaggregated data into finer sector classification, which I aggregate\nup to three sectors – agriculture (a), industry (m) and services (s), using the International\nStandard Industrial Classification of All Economic Activities, Revision 3 (ISIC Rev. 3.1, 2002). The\naggregate expenditure is normalized to\n\nMention: \"OECD national accounts database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Final consumption by sector in current and constant prices.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007665:4:6", "occurrence_id": "69790554431bd914deefafc5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: /sup> <sup>**𝑘**</sup> <sup>𝐸</sup> <sup>**𝑗**</sup> **𝑗** **𝑗** **𝑗**\n\nwhich I use to estimate the parameters by employing the iterated non-linear seemingly unrelated 𝛾 **𝑗** 𝑃 **𝑟** 𝑃 **𝑟**\nregression method. <sup>4</sup> [^4: Details in chapter 14.9 of Greene (2011).] I choose industry as the base sector and transform the constrained\nparameters such that the estimation becomes unconstrained. <sup>5</sup> [^5: The appendix contains details of the parameter restrictions and transformation scheme.]\n\nThe quantitative analysis requires data on intermediate shares and sectoral prices. I use\nthe OECD input-output database (OECD, 2017b) to collect the intermediate shares for many\ncountries. To obtain the sectoral prices, I utilize the final consumption expenditure data from the\nOECD national accounts database (OECD, 2017a). The database contains expenditures by sectors\nin both current and constant prices (country-specific base year) in national currencies, which I\nuse to construct price series indexed to unity for a base year for each country. Finally, I adjust the\nprice series using the harmonized price data for 2005 from the GGDC Productivity Level Database\n(Inklaar & Timmer, 2014) to make prices comparable across countries and time. A limitation of\nthe price series is that it corresponds to the consumption side of the economy. The consumer\nprices differ from the producer prices as the former include certain margins and distortions.\n\nAll sources contain disaggregated data into finer sector classification, which I aggregate\nup to three sectors – agriculture (a), industry (m) and services (s), using the International\nStandard Industrial Classification of All Economic Activities, Revision 3 (ISIC Rev. 3.1, 2002). The\naggregate expenditure is normalized to\n\nMention: \"GGDC Productivity Level Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"Harmonised price data make prices comparable across countries and time.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007676:12:0", "occurrence_id": "a24f71f6d516952adbce9e1a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 1: Cumulative productivity distributions\n\n\nNote: Survey weights are applied in the computation of the data presented above\n\n\nIn section 1, we outlined concerns that have been raised at the macro level that productivity growth has\n\n\nbeen lagging in the region, and the micro data from the WBES provide further justification of these\n\n\nconcerns. Table 3 shows the evolution of labor productivity and total factor productivity across the seven\n\n\ncountries covered in our analysis, for the three rounds of WBES data that are available for these countries.\n\n\nWhile average labor productivity increased from 2006 to 2010, it subsequently declined in 2017; for the\n\n\nmedian establishment, there was a minor increase between 2010 and 2017. Total factor productivity also\n\n\nincreased from 2006 to 2010, but subsequently declined—this holds at both the mean and the median\n\n\nvalues.\n\n\n11\n\nMention: \"micro data from the WBES\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Firm-level evidence corroborating the regional productivity concern.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007676:12:0", "occurrence_id": "6a7d349f130043c2d6065009", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Figure 1: Cumulative productivity distributions\n\n\nNote: Survey weights are applied in the computation of the data presented above\n\n\nIn section 1, we outlined concerns that have been raised at the macro level that productivity growth has\n\n\nbeen lagging in the region, and the micro data from the WBES provide further justification of these\n\n\nconcerns. Table 3 shows the evolution of labor productivity and total factor productivity across the seven\n\n\ncountries covered in our analysis, for the three rounds of WBES data that are available for these countries.\n\n\nWhile average labor productivity increased from 2006 to 2010, it subsequently declined in 2017; for the\n\n\nmedian establishment, there was a minor increase between 2010 and 2017. Total factor productivity also\n\n\nincreased from 2006 to 2010, but subsequently declined—this holds at both the mean and the median\n\n\nvalues.\n\n\n11\n\nMention: \"WBES data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Three rounds across seven countries give the productivity evolution.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000061:13:0", "occurrence_id": "ce0adb5cc4fb7a286f04fcc7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |Col1|Nombre del proceso/Subproceso: Prevención y Protección|Código: PP-P01-F10|\n|---|---|---|\n||**ALERTA TEMPRANA **|Versión: 01|\n||**ALERTA TEMPRANA **|Vigente desde:<br>07/09/2018|\n\n\n\ndel proceso de producción, distribución y comercialización de drogas ilegales buscan la\ndistribución de ingresos por medio del tráfico de estupefacientes. Estos grupos hoy se trenzan\nen una disputa territorial por el monopolio de la comercialización de sustancias alucinógenas a\ntravés de la violencia ejercida contra el grupo adversario, dinámica que impacta en el conjunto\nde la población asentada en los sectores donde ejercen presencia.\n\n**<mark>2.1 E</mark>** **scenario de riesgo y situación actual.**\n\n\ny rurales.\n\nEl informe elaborado por el Sistema Integrado de Monitoreo de Cultivos Ilícitos (Simci), de la\nOficina de las Naciones Unidas contra la Droga y el Delito, denominado Monitoreo de territorios\nafectados por cultivos ilícitos 2020, evidencia el aumento año tras año de la cantidad de cocaína\n\n\n14\n\nMention: \"Monitoreo de territorios\nafectados por cultivos ilícitos 2020\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Su evidencia muestra el aumento anual de la cantidad de cocaína.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000062:6:1", "occurrence_id": "779bca62fd12c69f18977590", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  la protection de l’enfance. Le FTS\na été amélioré pour pouvoir proposer des données plus précises et granulaires, dont des données\nsur le financement de secteurs spécifiques dans le cadre de programmes multisectoriels. Toutefois, le\nFTS n’identifie pas totalement les données de financement pour les Réponses nationales et régionales\ndédiées aux réfugiés. Pour intégrer cette question, le rapport utilise également des données du Suivi du\nfinancement des réfugiés (RFT), piloté par le HCR, ainsi que des données fournies par le HCR directement.\nL’analyse englobe 26 Plans de réponse humanitaire (PRH), le Plan de réponse conjoint pour les Rohingyas\nau Bangladesh, le Plan régional d’aide pour les réfugiés et les migrants du Venezuela et 17 plans nationaux\npour les réfugiés regroupés au sein de cinq Plans de réponse pour les réfugiés (PRR). Les besoins de\nfinancement de la protection de l’enfance et les fonds reçus ont été calculés à l’aide des données du FTS\net du RFT, ce qui permet d’offrir un aperçu global tout en évitant les doublons dans le calcul. Toutes les\ndonnées utilisées dans le rapport sont exactes en date du 9 juillet 2024.\n\nMention: \"FTS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Mejorado para ofrecer datos más precisos y granulares.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000062:6:1", "occurrence_id": "6f042369aadd6097573841dd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  la protection de l’enfance. Le FTS\na été amélioré pour pouvoir proposer des données plus précises et granulaires, dont des données\nsur le financement de secteurs spécifiques dans le cadre de programmes multisectoriels. Toutefois, le\nFTS n’identifie pas totalement les données de financement pour les Réponses nationales et régionales\ndédiées aux réfugiés. Pour intégrer cette question, le rapport utilise également des données du Suivi du\nfinancement des réfugiés (RFT), piloté par le HCR, ainsi que des données fournies par le HCR directement.\nL’analyse englobe 26 Plans de réponse humanitaire (PRH), le Plan de réponse conjoint pour les Rohingyas\nau Bangladesh, le Plan régional d’aide pour les réfugiés et les migrants du Venezuela et 17 plans nationaux\npour les réfugiés regroupés au sein de cinq Plans de réponse pour les réfugiés (PRR). Les besoins de\nfinancement de la protection de l’enfance et les fonds reçus ont été calculés à l’aide des données du FTS\net du RFT, ce qui permet d’offrir un aperçu global tout en évitant les doublons dans le calcul. Toutes les\ndonnées utilisées dans le rapport sont exactes en date du 9 juillet 2024.\n\nMention: \"données du Suivi du\nfinancement des réfugiés\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Intègre la question du financement consacré aux réfugiés.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000062:6:1", "occurrence_id": "d94327d600671726c3d6b535", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  la protection de l’enfance. Le FTS\na été amélioré pour pouvoir proposer des données plus précises et granulaires, dont des données\nsur le financement de secteurs spécifiques dans le cadre de programmes multisectoriels. Toutefois, le\nFTS n’identifie pas totalement les données de financement pour les Réponses nationales et régionales\ndédiées aux réfugiés. Pour intégrer cette question, le rapport utilise également des données du Suivi du\nfinancement des réfugiés (RFT), piloté par le HCR, ainsi que des données fournies par le HCR directement.\nL’analyse englobe 26 Plans de réponse humanitaire (PRH), le Plan de réponse conjoint pour les Rohingyas\nau Bangladesh, le Plan régional d’aide pour les réfugiés et les migrants du Venezuela et 17 plans nationaux\npour les réfugiés regroupés au sein de cinq Plans de réponse pour les réfugiés (PRR). Les besoins de\nfinancement de la protection de l’enfance et les fonds reçus ont été calculés à l’aide des données du FTS\net du RFT, ce qui permet d’offrir un aperçu global tout en évitant les doublons dans le calcul. Toutes les\ndonnées utilisées dans le rapport sont exactes en date du 9 juillet 2024.\n\nMention: \"données fournies par le HCR directement\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Complètent les sources documentaires pour les plans de réponse.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000063:1:0", "occurrence_id": "958e32a9a3e771aa9ea09036", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ## Resumen ejecutivo\n\nEn 2023, las necesidades humanitarias alcanzaron niveles nunca vistos, debido al aumento\nde los conflictos armados, a desplazamientos forzados sin precedentes, a la emergencia\nclimática y a las catástrofes derivadas de peligros naturales. A finales del 2023, 363 millones\nde personas precisaron asistencia humanitaria. De ellas, 245 millones fueron destinatarias de planes\nde respuesta coordinados por las Naciones Unidas: un aumento del 7 % con respecto a los datos\npreliminares del Panorama Global Humanitario de 2023. Estas crisis crecientes hicieron que las\nnecesidades de financiación humanitaria aumentaran hasta 56 700 millones de dólares a finales\nde 2023. Pese a que los niveles de financiación fueron de los más altos de la historia, la tasa de\nfinanciación para los llamamientos coordinados por las Naciones Unidas fue solo del 43 %, la más\nbaja de la historia. Los menores serán los más perjudicados por este déficit de financiación, ya que\nse ven afectados de manera desproporcionada por las crisis humanitarias.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nAnálisis de la financiación destinada a la protección de la niñez y la adolescencia en la acción humanitaria en 2023 2\n\nMention: \"Panorama Global Humanitario\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Sus datos preliminares sirven de base para el aumento porcentual.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000065:8:3", "occurrence_id": "847aaa008deb4ddafd79a734", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Congolais ont quitté\nvolontairement l’Angola par crainte d’être expulsés : 121 hommes, 101\nfemmes et 106 enfants (51 garçons et 55 filles).\n\n\n**<mark>KASAI CENTRAL</mark>** <sup>**7**</sup>\n\n- Au cours de ce mois de juillet 2024, la province a connu la persistance\nd’un conflit foncier entre deux groupements voisins dans le territoire de\nDimbelenge. En effet, tant sur la ville de Kananga que dans les cinq\nterritoires, une insécurité grandissante a été enregistrée,\nparticulièrement dans les communes de Katoka et Nganza où des\nbandits non autrement identifiés sévissent en visitant les ménages\n\n\n7 <u>Rapport mensuel de monitoring de protection Kasai central mois de juillet 2024 UNHCR et</u>\n<u>Vibosa</u>\n\n\n\nSi vous avez des commentaires, questions, ou données supplémentaires, veuillez contacter :\n[Steve Ndikumwenayo (ndikumwe@unhcr.org) ou Lorraine de Limelette (lorraine.delimelette@nrc.no)](mailto:ndikumwe@unhcr.org) 9\n\nMention: \"Rapport mensuel de monitoring de protection Kasai central\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Ses observations mensuelles documentent les incidents de protection.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000069:74:1", "occurrence_id": "383c2513ee793c11950d2dbd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  gauge\nor set the qualitative range of indicators at sectoral\nlevel.\n\n\nIssues related to M&E were also frequently raised\nby key informants during interviews. Several\nrespondents noted that cash Post-Distribution\nMonitoring (PDM) exercises typically focus on\nexpenditure or market-related issues and rarely\ndelve into outcomes of cash. The fact that multipurpose cash cuts across sectors adds a layer of\ncomplication and means that monitoring a range of\noutcomes across several sectors is needed. This is\nhowever largely perceived as difficult and unusual,\n\n\n\nand in the words of a key informant, “we don’t see\nmuch of this type of monitoring coming from the\nfield”.\n\n\nIndeed, there were frequent mentions that\nregardless of the fact that multi-purpose cash\nand related monitoring can enable a better\nunderstanding of household need prioritization\npatterns, and of the effect of cash injections across\ndifferent aspects of well-being, M&E indicators for\nmulti-purpose cash are largely limited to those with\nwhich agencies are more familiar and are easier to\ncollect (e. g. food security indicators). Other, more\ncomplex and unusual indicators, such as access\nto housing, are typically neglected. In addition,\ncash monitoring systems tend to predominantly\nrely on economic benchmarks, often neglecting\nto consider the environmental externalities\nassociated with these solutions. For example, in\nthe WASH sector there is a need to understand\nnot only whether domestic water was purchased\nand in what quantity, but also assess the quality of\nwater purchased, the sources from which it was\nobtained (water kiosk, truck, etc. ) and associated\nenvironmental risks.\n\n\nIn Greece, the few available PDMs showed very\nlimited, if any, disaggregation of cash expenditures\n(e. g. hygiene), and some reports grouped and\npresented different, largely unrelated cash\nexpenditures categories together (e. g. food and\nclothes). From available PDM reports it was not\npossible to gain a comprehensive picture of cash\nexpenditures and outcomes, or draw comparisons\nbetween cash expenditures and outcomes in\ncamps versus\n\nMention: \"PDMs\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"The few available exercises showed only very limited disaggregation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000072:17:1", "occurrence_id": "8b266181e1a4a3046e2b6ed4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  complaints (A) and on other\ninformation related to the police and judicial response to racist violence, including\nmeasures adopted by the Ministry of Justice, Transparency and Human Rights (B).\n\n\n**Α. Analysis of RVRN data on complaints**\n\n\nAt the time of their recording by RVRN organisations, 39 incidents out of 102 had been\nreported to the police.\n\n\n14 of those incidents related to assaults against refugees and immigrants and 5 against\ndefenders, while only 6 incidents concerning LGBTQI persons had been reported to\nthe police at the time. In 19 cases, the victims stated that they had not reported the\nincident to the police but that they intended to do so. In 22 cases, the victims stated\nthat they would not take further steps due to fear; or lack of trust in the authorities; or\nbecause they did not want to reveal their sexual orientation; or because they did not\nconsider the incident serious enough to become involved in criminal proceedings.\n\n\n18\n\nMention: \"RVRN data on complaints\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Tracks incidents at the moment partner organisations record them.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000076:3:2", "occurrence_id": "ff0a3c4ac6541f620af9e077", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ción en\nlas tasas de abandonos indica que las personas que\nhuyen de violencia están cada vez más, considerando\na Guatemala como su destino final. En el sistema de\nrefugio de México, por ejemplo, las tasas de\nabandono o desistencia han variado de 36% en 2014\na 27% en 2016 .\n\n\n\n**3**\n\n\n\n1. ACNUR, Estadísticas de Población, disponible en: http://popstats.unhcr.org/en/asylum_seekers_monthly\n\nMention: \"Estadísticas de Población\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Sus series mensuales de solicitantes de asilo dan las variaciones.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000080:16:0", "occurrence_id": "ce5c6a593aa0c79e045a6574", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ##### **Livelihoods**\n\n**Livelihood sector outcomes and the**\n**contribution of cash**\n\n\nThere is limited available data in returnee\nmonitoring reports about use of the repatriation\ncash grant to support livelihood opportunities upon\nreturn. UNHCR Afghanistan (2016b) finds that only\n15% of returnees surveyed made investments in\n“sustainable livelihoods”. By contrast, the 2009\nevaluation of the repatriation cash grant found that\nthe majority, over 75%, were able to use part of the\ncash grant to start small businesses, from as little as\n$20 to as much as $1,400 (Altai Consulting, 2009).\nThe same evaluation however already observed far\nlower “levels of entrepreneurship” in the returnees’\nsample (2007 and 2008 returnees) compared to\nthat of the 2005 evaluation (Altai Consulting, 2005,\nin Altai Consulting, 2009). One possible explanation\nsuggested was the worsening macro-economic\ncontext: “opportunities to create new businesses\nwere higher in the initial years of development,\nsince the country was missing mostly everything,\nincluding many small shops and services. Seven\nyears later, it has become more difficult to\nidentify niches and create new businesses” (Altai\nConsulting, 2009: 23).\n\n\nThe worsening economy in a context of escalating\nconflict and chronic political impasse (Samuel Hall\net al., 2018; Afghanistan, 2016b), as well as the\ninflux of nearly 370,000 returnees from Pakistan\nin 2016 alone, are among the contributing factors\nof an increasingly contracted labour market.\nThus for the thousands of Afghan refugees who\nreturned in 2016, finding a job or setting up a\nsmall business through the cash grant appears\nmore challenging than it was a few years ago.\nThe upcoming evaluation of the repatriation cash\ngrant commissioned by UNHCR at the end of 2017\nmay usefully shed light on this trend and allow\ncomparisons of different returnee cohorts surveyed\nsince 2005.\n\n\nThe findings of FGDs likewise indicate\n\nMention: \"returnee\nmonitoring reports\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Little information on the use of the livelihoods grant.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000080:16:0", "occurrence_id": "1caeb4826683b28d81512e86", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ##### **Livelihoods**\n\n**Livelihood sector outcomes and the**\n**contribution of cash**\n\n\nThere is limited available data in returnee\nmonitoring reports about use of the repatriation\ncash grant to support livelihood opportunities upon\nreturn. UNHCR Afghanistan (2016b) finds that only\n15% of returnees surveyed made investments in\n“sustainable livelihoods”. By contrast, the 2009\nevaluation of the repatriation cash grant found that\nthe majority, over 75%, were able to use part of the\ncash grant to start small businesses, from as little as\n$20 to as much as $1,400 (Altai Consulting, 2009).\nThe same evaluation however already observed far\nlower “levels of entrepreneurship” in the returnees’\nsample (2007 and 2008 returnees) compared to\nthat of the 2005 evaluation (Altai Consulting, 2005,\nin Altai Consulting, 2009). One possible explanation\nsuggested was the worsening macro-economic\ncontext: “opportunities to create new businesses\nwere higher in the initial years of development,\nsince the country was missing mostly everything,\nincluding many small shops and services. Seven\nyears later, it has become more difficult to\nidentify niches and create new businesses” (Altai\nConsulting, 2009: 23).\n\n\nThe worsening economy in a context of escalating\nconflict and chronic political impasse (Samuel Hall\net al., 2018; Afghanistan, 2016b), as well as the\ninflux of nearly 370,000 returnees from Pakistan\nin 2016 alone, are among the contributing factors\nof an increasingly contracted labour market.\nThus for the thousands of Afghan refugees who\nreturned in 2016, finding a job or setting up a\nsmall business through the cash grant appears\nmore challenging than it was a few years ago.\nThe upcoming evaluation of the repatriation cash\ngrant commissioned by UNHCR at the end of 2017\nmay usefully shed light on this trend and allow\ncomparisons of different returnee cohorts surveyed\nsince 2005.\n\n\nThe findings of FGDs likewise indicate\n\nMention: \"FGDs\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its findings point to outcomes similar to the survey data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000084:6:0", "occurrence_id": "8131144fb4f9d179f1614f1a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Page 7\n182 Incidents de protection collectés \nRépartition des incidents par typologies\nRépartition des typologies de violation de droits\nRépartition des victimes par type de  populations\nRépartition des victimes par âge et par sexe\nRépartition des incidents par auteurs\n3.  Tahoua\n0%;0\n0%;0\n1% ; 9 \n99% ;1173\nFille\nGarçon\nFemme\nHomme\nAdulte 100%\nEnfant 0%\nRapport Mensuel de Monitoring de Protection | Régions de Tillabéri -Tahoua   JUILLET 2022\na. Répartition des incidents par typologie:  \nAu total 182 incidents ont été collectés, documentés et rapportés au cours du \nmois de juillet 2022 à travers le mécanisme de monitoring de protection mis \nen place dans les départements de Bagaroua, Tahoua, Tassara, Madaoua, \nIllela, Konni, Malbaza, Abalak, Tchintabraden et Tillia, dans la région de \nTahoua repartis en neuf (09) typologies :\n \nExtorsions de biens : 115 cas sur 182 collectés au total ;\n \nViols et pillages : 29 cas sur 182 collectés au total ;\n \nAgressions physiques : 24 cas sur 182 collectés au total ;\n \nMenaces : 05 cas sur 182 collectés au total ;\n \nIncursions et attaques : 03 cas sur 182 collectés au total ;\n \nConflit intercommunautaire : 03 cas sur 182 collectés au total ;\n \nEnlèvement : 01 cas sur 182 collectés au total ;\n \nAssassinat/meurtre : 01 cas sur 182 collectés au total ;\n \nArrestation : 01 cas sur 182 collectés au total.\nb.  Répartition des typologies de violation des droits: \nTrois (03) types violations de droits ont été enregistrés dans le cadre du\n\nMention: \"Rapport Mensuel de Monitoring de Protection\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Ses relevés mensuels répartissent les incidents par typologie et par région.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000101:1:2", "occurrence_id": "8c37f861161a99c658ea77cc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  devait les accompagner, il y\na lieu de noter que le partenaire a élaboré son plan de distribution lui-même\net constat fait, certaines aires de santé n’ont pas bénéficié des intrants et\nmême celles qui avaient bénéficié la quantité était insuffisante, raison pour\nla quelle certaines structures sont tombées en rupture un mois seulement\naprès le départ du partenaire.\n\n\n\n**b) Encodeur du BCZS**\n\n\nDans la province du Tanganyika en général et la zone de santé de\nkabalo en particulier, la qualité des données nutritionnelles constitue un\nvrai problème dans la planification des interventions en l’absence des\nenquêtes fiables, le DHIS2 qui pouvait nous aider nous donne souvent\ndes informations qui ne sont pas en phase avec la réalité du terrain.\n\n\nAprès échange, il été fait remarquer que la zone encode sur base de\nSNIS transmis par les aires de santé et non dans les masques PCIMA,\nseulement que l’encodeur doit être en contact avec le superviseur\nnutrition pour que les informations soient conformes aux masques avant\nde les mettre dans l’outils DHIS2, ce qui n’est pas toujours le cas dans\nla zone.\n\n**c) SEHA** :\n\nL’entretien avec le superviseur Eau Hygiène et assainissement a tourné\nautour de la couverture en eau propre et potable dans la zone de santé,\nà en croire ce dernier, la zone de santé de Kabalo a une couverture des\nvillages ayant des points d’eau améliorés est seulement de 31\n\nMention: \"DHIS2\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Ses informations sont souvent incompatibles avec la planification en l'absence d'enquêtes fiables.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000109:8:0", "occurrence_id": "098e051c133c0f1012b320de", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Figure 9: Food Insecurity and the household hunger score_\n\n\n**Annex: Data and Methodology**\n\n\nThe environment surrounding displaced populations in Somalia in terms of welfare, food insecurity, and\nemployment is changing rapidly. It is difficult to monitor the situation frequently due to insecurity in many\nparts of the country. There is therefore need for innovative, cost-effective, and time-saving data collection\nmethods and techniques. Socioeconomic data on IDPs was last collected nationally in 2017 through the\nSomali High Frequency Survey (SHFS). The Somali Health and Demographic Survey (SHDS) 2018-19—\nanother large-scale household survey—did not include displaced populations.\n\n\nThe World Bank (WB) in collaboration with the United Nations High Commissioner for Refugees (UNHCR),\nthe National Bureau of Statistics (NBS) in Somalia and the World Bank-UNHCR Joint Data Center on Forced\nDisplacement (JDC) conducted a high-frequency phone-based survey (HFPS). The first round of the survey\nwas conducted between November 2021 and March 2022 and second round was conducted in July-August\n2022. The sample consists of about 500 households from host communities, IDPs living in camps, IDPs\nliving outside camps, refugees, and refugee returnees.\n\n\n9\n\nMention: \"high-frequency phone-based survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"First round conducted in this window by the common centre.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000113:7:1", "occurrence_id": "aec568608663667c296020b3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: aire et au recrutement forcé. La violence armée résultant</mark>\n<mark>d’interventions militaires ou de tensions intercommunautaires continue</mark>\n<mark>de perturber les activités agricoles et commerciales.</mark>\n\n<mark>Selon les résultats du 20ème cycle d’analyse IPC, 27 millions de</mark>\n<mark>personnes sont en situation d’insécurité alimentaire aigüe parmi</mark>\n<mark>lesquelles 20,9 millions sont en phase de crise (phase 3) et 6,1 millions</mark>\n<mark>en phase d’urgence (phase 4). Dans la province de l'Ituri, la majeure</mark>\n<mark>partie de la population, en particulier dans les zones touchées par le</mark>\n<mark>conflit, vit en phase d’urgence (phase 4) et phase de crise (phase 3). Sur</mark>\n<mark>les 70 zones analysées dans la toute première analyse de la malnutrition</mark>\n<mark>aigüe de l’IPC, 8 ont des niveaux de GAM supérieurs à 15% et 42 zones</mark>\n<mark>ont des GAM entre 10% et 15%, ce qui met en évidence une situation</mark>\n<mark>nutritionnelle critique</mark> <sup>4</sup> [^4: IPC. 2021. Acute Malnutrition Democratic Republic of Congo. November\n2021. (Also available at:] <mark>.</mark>\n\n<mark>La situation demeure préoccupante dans ces zones. Une personne sur</mark>\n<mark>quatre est affectée par l’insécurité alimentaire. Ces personnes\n\nMention: \"20ème cycle d’analyse IPC\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Ses résultats donnent le nombre de personnes en insécurité alimentaire.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000116:1:0", "occurrence_id": "711db2d8a8ff51b02a6c732a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Though the voluntary character of the return was emphasized by the authorities during the\nplanning phase, the Protection cluster believed in the importance to follow the SOPs on\nreturn as a guide for the humanitarian when providing support to the authorities in the\nreturn process and in the importance of consulting the population to better ascertain the\nintentions and needs of the population.\n\n\n**II.** **Methodology**\n\n\nBased on the information received from FDMA on the denotification of the 50 tehsil, data\nwithin UNHCR database of registered IDPs were compared and 4549 were found as from Ali\nSherzai tribe.\n\n\nThe protection cluster established a representative sample capturing a significant portion of\nthe displaced population from Central Kurrum from Ali Sherzai tribe. IVAP provided the\nphone numbers of the registered families and IOM conducted phone calls with the IDPs.\n\n\n259 families were consulted and the RIS was conducted by 7 enumerators who contacted\ndisplaced families from a call centre using the contact information available through the\nIVAP records. This RIS was conducted using a specific tool/questionnaire developed in 2012\nfor previous consultations and slightly adapted in 2013. <sup>1</sup> [^1: See Annex1]\n\n\nThe enumerators were trained on 23rd January by protection cluster coordinator on the\ntool as well as on basic principles of confidentiality, informed consent and interviewing\ntechniques. The data collection was conducted from 29.1. to 5.2.\n\n\nIn total, 259 interviews were conducted- 98 families currently residing in Hangu, 94 in\nKohat, 42 in Peshawar, 17 in New Durrani, 4 in Nowshera, 3 in Charsadda, 1 in Tank.\n\n\n**III.** **Main findings of the Return Intention Survey 2014**\n\n\n**1)** **Profile of the interviewed population**\n\n\nMostly heads of families were interviewed during the return intention survey (89%), only in\n28 cases other family members were interviewed, depending on\n\nMention: \"UNHCR database of registered IDPs\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Cross-checked against cancellation information to identify the people concerned.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000116:1:0", "occurrence_id": "364abca7a21c37969fabb64a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Though the voluntary character of the return was emphasized by the authorities during the\nplanning phase, the Protection cluster believed in the importance to follow the SOPs on\nreturn as a guide for the humanitarian when providing support to the authorities in the\nreturn process and in the importance of consulting the population to better ascertain the\nintentions and needs of the population.\n\n\n**II.** **Methodology**\n\n\nBased on the information received from FDMA on the denotification of the 50 tehsil, data\nwithin UNHCR database of registered IDPs were compared and 4549 were found as from Ali\nSherzai tribe.\n\n\nThe protection cluster established a representative sample capturing a significant portion of\nthe displaced population from Central Kurrum from Ali Sherzai tribe. IVAP provided the\nphone numbers of the registered families and IOM conducted phone calls with the IDPs.\n\n\n259 families were consulted and the RIS was conducted by 7 enumerators who contacted\ndisplaced families from a call centre using the contact information available through the\nIVAP records. This RIS was conducted using a specific tool/questionnaire developed in 2012\nfor previous consultations and slightly adapted in 2013. <sup>1</sup> [^1: See Annex1]\n\n\nThe enumerators were trained on 23rd January by protection cluster coordinator on the\ntool as well as on basic principles of confidentiality, informed consent and interviewing\ntechniques. The data collection was conducted from 29.1. to 5.2.\n\n\nIn total, 259 interviews were conducted- 98 families currently residing in Hangu, 94 in\nKohat, 42 in Peshawar, 17 in New Durrani, 4 in Nowshera, 3 in Charsadda, 1 in Tank.\n\n\n**III.** **Main findings of the Return Intention Survey 2014**\n\n\n**1)** **Profile of the interviewed population**\n\n\nMostly heads of families were interviewed during the return intention survey (89%), only in\n28 cases other family members were interviewed, depending on\n\nMention: \"IVAP records\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Contact details of displaced families are used to call them back.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000116:1:0", "occurrence_id": "885fbfb7fe710a03a29ff648", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Though the voluntary character of the return was emphasized by the authorities during the\nplanning phase, the Protection cluster believed in the importance to follow the SOPs on\nreturn as a guide for the humanitarian when providing support to the authorities in the\nreturn process and in the importance of consulting the population to better ascertain the\nintentions and needs of the population.\n\n\n**II.** **Methodology**\n\n\nBased on the information received from FDMA on the denotification of the 50 tehsil, data\nwithin UNHCR database of registered IDPs were compared and 4549 were found as from Ali\nSherzai tribe.\n\n\nThe protection cluster established a representative sample capturing a significant portion of\nthe displaced population from Central Kurrum from Ali Sherzai tribe. IVAP provided the\nphone numbers of the registered families and IOM conducted phone calls with the IDPs.\n\n\n259 families were consulted and the RIS was conducted by 7 enumerators who contacted\ndisplaced families from a call centre using the contact information available through the\nIVAP records. This RIS was conducted using a specific tool/questionnaire developed in 2012\nfor previous consultations and slightly adapted in 2013. <sup>1</sup> [^1: See Annex1]\n\n\nThe enumerators were trained on 23rd January by protection cluster coordinator on the\ntool as well as on basic principles of confidentiality, informed consent and interviewing\ntechniques. The data collection was conducted from 29.1. to 5.2.\n\n\nIn total, 259 interviews were conducted- 98 families currently residing in Hangu, 94 in\nKohat, 42 in Peshawar, 17 in New Durrani, 4 in Nowshera, 3 in Charsadda, 1 in Tank.\n\n\n**III.** **Main findings of the Return Intention Survey 2014**\n\n\n**1)** **Profile of the interviewed population**\n\n\nMostly heads of families were interviewed during the return intention survey (89%), only in\n28 cases other family members were interviewed, depending on\n\nMention: \"Return Intention Survey 2014\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its results describe the profile of the respondents.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000127:8:1", "occurrence_id": "41861d3197946f52a2568d53", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: u>Services de soutiens existants</u>**\n\n\nL’agriculture et l’élevage au Niger bénéficient de plus en plus de systèmes\nd’information et de service de vulgarisation utilisant les nouvelles technologies d’information et de la communication. Ces services peuvent faciliter\nl’accès à des informations techniques et de marché permettant aux producteurs de non seulement améliorer leur rendement mais aussi et surtout\nd’adopter la meilleure stratégie d’écoulement de leur production. A titre illustratif, trois services de soutiens seront décrits dans ce paragraphe, à\nsavoir:\n\n\nn nLe système d’information sur les marchés et prix des produits maraîchers\n\nn nLe Système d’Information sur le Marché du bétail (SIM Bétail)\n\nn nLe programme e-extension Niger.\n\n\n**2.2.1 Système d’information sur les marchés des produits**\n**maraichers**\n\n\nLe système d’information sur les marchés et prix des produits maraîchers\nest une initiative du RECA et du Programme Nigéro-Allemand de Promotion\nde l’agriculture Productive (Promap). C’est un système d’information sur les\nprix des produits maraichers au profit des producteurs de la région de Tillabéry ( **Figure 2** ). Ce système d’information a fonctionné de 2007 à 2013 et\na couvert 18 Marchés. Les informations sur les prix des produits maraîchers\n\n\n\n5\n\nMention: \"informations sur les prix des produits maraîchers\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Le système a couvert dix-huit marchés sur cette période.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000129:1:0", "occurrence_id": "c773b8703620101b9cfd70da", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **are people with disabilities** and **17% are older people** . Among older people around 50%\nare persons with disabilities which exacerbate the risks that this category of refugees\nfaces.\nThis briefing note aims to provide an overview of the situation of older refugees in\nMoldova and the gaps and needs they face, as well as the current efforts being made by\nhumanitarian stakeholders to address these challenges.\n\nFrom the **Socio-Economic Insights Survey** (SEIS) conducted by the IMPACT Initiative in1\nAugust 2024, the following **four main areas** were identified as significantly impacting\n**older refugees** compared to other groups:\n\n**Access to Information:** Older refugees face significant challenges in\naccessing information compared to other refugee groups. Many lack the\nnecessary devices to connect online, and the available information is often\nnot presented in formats that accommodate their needs.\n\n\n**Healthcare Access:** Access to healthcare and to medicines emerged as\nurgent priorities for older refugees, with a significantly higher proportion\nidentifying these needs as compared to younger refugee groups.\n\n\n**Support Channels:** Refugees aged 75 and above are less likely to have access\nto secure and private channels for seeking support, or reporting concerns,\nincluding sensitive community issues (only 75% reported access compared to\nan average of 95% of other age groups).\n\n\n**Social Assistance:** Seventy percent (70%) of refugees aged 75 and above\nreported the need to have better access to social assistance services to\nimprove their socio-economic situation in Moldova as compared to the other\nage groups which is less then 50%.\n\nMention: \"Socio-Economic Insights Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its findings feed the report's key conclusions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000139:6:0", "occurrence_id": "2ee89adf3ce9ffbddeecdc5d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: DTM Mali – Décembre 2024\n\n### **INTRODUCTION**\n\n\nDurant l’année 2024, le suivi des tendances de déplacement était complexe, à la suite des violences armées,\naux tensions intercommunautaires et à des inondations, avec des mouvements de retour, y compris avortés,\ndes déplacements de courte durée ont continué à être signalés dans différentes régions du pays. A la fin de\nl’année 2024, une proportion importante de personnes déplacées internes vivaient encore dans les centres\nurbains du pays dans des conditions de protection et d’assistance inadéquates, le reste de la population déplacée\nétant dispersée dans les zones rurales dans des familles d’accueils ou sur des sites spontanés. L’amélioration\nsignificative des conditions de sécurité dans certaines localités a permis à de nombreuses personnes déplacées\ninternes (PDI) à regagner leurs foyers au Centre et au Nord du pays, mais de nombreux obstacles persistent,\nles empêchant de trouver des solutions durables à leur déplacement.\n\n\nLe transfert du programme matrice de suivi des déplacements DTM (Displacement Tracking Matrix)  au\nGouvernement par l’OIM, en novembre 2014 et la signature de deux accords de partenariat entre le\nGouvernement et l’UNHCR pour la gestion des sites PDIs et  l’enregistrement des rapatriés, ont permis au\nGouvernement du Mali d’obtenir une meilleure compréhension des caractéristiques des populations\ndéplacées et retournées, ainsi que de mettre au jour leur\n\nMention: \"Displacement Tracking Matrix\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Transféré au gouvernement par l'organisation en ce mois.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000140:6:1", "occurrence_id": "eab3cb9a77db374dc4a6be57", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  “I believe you know the ECOWAS protocol encourages\nfree movement of persons, goods and services. But for those going to Europe, we are not, as an\nadministration, agreeing with those who continue to defy the Sahara Desert and the Mediterranean\nbecause they feel there are greener pastures there, whether they are prepared for it or not. We do not\nsupport anything illegal and indiscipline. You must recall that about six weeks ago, we repatriated\nabout 3,000 Nigerians that were stuck in Libya on their way to Europe. You must have read in\nthe newspapers and watch on the television the number of Nigerians lost in the Mediterranean”.\n\n\n**Thousands** **displaced** **in** **Nigeria** **due** **to** **floods:** In September 2018 there were\nseries of floods in particular in rural areas in Niger state and also Kwara, Benue, Kogi,\nAdamawa, Taraba, Kebbi, Bayelsa, Edo, Anambra, Rivers and Delta states. <u>[Thousands of](https://www.bbc.com/news/world-africa-4554669)</u>\n<u>[people were displaced and vast swathes of farmlands were destroyed by](https://www.bbc.com/news/world-africa-4554669)</u> <u>the floods in</u>\n<u>central and southern Nigeria, according to the BBC's Is'haq Khalid in the capital, Abuja.</u>\n\n\nMonthly Trends Analysis West Africa 7\n\nMention: \"Monthly Trends Analysis West Africa\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its monthly entries describe how the situation evolved.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000141:24:1", "occurrence_id": "081cf46b6b67ee6adecdc6b7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  the decision-making process to end\ndisplacement.\n\n\nLaw 1448/2011 created the Unit for the Attention and\nIntegral Reparation of the Victims (UARIV, Spanish\nacronym), the government agency responsible for\ncoordinating the National System of Attention and\nReparation for Victims regarding the implementation\nof public policy to assist and provide comprehensive\nreparation to victims. The Victims Unit has within its\nmandate the task of guaranteeing the operation of\nthe National Information Network for assistance and\nreparation to Victims, including the interoperability of\nthe various relevant information systems. The Unit also\nmanages the Single Victims Registry, overseeing the\nintegrity of all existing records about displacement and\nhuman rights violations due to armed conflict.\n\n\n\n**22** Contributed by Oscar Iván Rico Valencia (Coordinador Grupo de Atención a Víctimas en el Exterior) as well as Katizza Carvajal and Karen\nChavez from the SDGs Indicators Working Group at DANE (the National Statistics Office in Colombia).\n\n**23** [https://secretariageneral.gov.co/transparencia/control/informaci%C3%B3n-poblaci%C3%B3n-vulnerable/ley-387-1997](https://secretariageneral.gov.co/transparencia/control/informaci%C3%B3n-poblaci%C3%B3n-vulnerable/ley-387-1997)\n\n**24** https://www.unidadvictimas.gov.co/en\n\n\n25 DATA DISAGGREGATION OF SDG INDICATORS BY FORCED DISPLACEMENT\n\nMention: \"Single Victims Registry\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The unit manages this registry and preserves the integrity of its records.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000149:1:5", "occurrence_id": "baa2740cd8dbb7877d72830b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  2024, en raison de périodes de sécheresse, des inondations, des chocs\néconomiques, des conflits et de l’insécurité.\n\n\nEn **Afghanistan**, malgré une baisse de 4,1 millions grâce à l’aide humanitaire,\nau soutien à l’agriculture, aux bonnes récoltes et à une réduction de\nl’inflation, 15,8 millions de personnes restent en insécurité alimentaire aiguë.\nDouze pays et territoires représentés sur la Figure 4 (tous sauf le Myanmar,\nla Namibie, le Pakistan et la Palestine) connaissent des crises alimentaires\npersistantes ou récurrentes, puisqu’ils ont été inclus dans les neuf éditions\ndu GRFC. Les 35 pays confrontés à des crises alimentaires persistantes ou\nrécurrentes représentent entre 80 et 94 pour cent de la population dans des\nniveaux élevés d’insécurité alimentaire aiguë chaque année.\n\n\n\nfig. 4 **Pays et territoires avec les nombres et proportions les plus élevés de**\n**personnes confrontées à des niveaux élevés d’insécurité alimentaire aiguë.**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n**2** Dans les zones où des analyses IPC/CH n’étaient pas disponibles, il n’a pas été possible de déterminer la répartition par phase des 68,2 millions de personnes confrontées à des niveaux élevés d’insécurité alimentaire aiguë.\n\n\n**2**\n\nMention: \"GRFC\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Ses neuf éditions couvrent les pays en crise alimentaire persistante.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000151:7:2", "occurrence_id": "8688b1a3972415b9a421d4ae", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \npossess a deeper understanding of local needs. They\nsuggested a more inclusive approach to planning\nand execution to ensure a fairer and more effective\ndistribution. This concern over exclusion was\nechoed in the survey findings, where 33.34% of\nrespondents acknowledged that humanitarian\norganizations were effective in reducing the risk of\nexclusion for minority groups, while 30.48%\nindicated they were only somewhat effective, and\n21.4% felt that exclusion mitigation efforts were not\neffective at all.\n\n\nThe FGDs also highlighted issues of **fairness versus**\n**effectiveness** **in** **aid** **delivery.** While some\nparticipants felt that the aid distribution was fair in\ntheory, they pointed out disparities in practice due\nto logistical issues and favoritism, resulting in certain\nregions or communities receiving disproportionate\namounts of assistance. This sentiment aligns with\nthe survey data showing that 39.85% of\nrespondents reported being inadequately informed\nabout available aid, suggesting that aid is not\nreaching those who need it most. Additionally, key\ninformants reinforced this view, with 53.85% rating\nthe current aid distribution process as only\n“somewhat effective” and 23.08% rating it as “not\neffective at all” for marginalized groups.\n\n\n\nUNHCR Somalia / June 2025 8\n\nMention: \"FGDs\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"They raise the tension between equity and efficiency of assistance.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000158:2:0", "occurrence_id": "72fe0c09516586370c34a83c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ## **Executive summary**\n###### **Key Trends**\n\nThere was little change in the overall conflict during February and March. Fighting characterised by airstrikes and shelling continued\non the major frontlines in Hajja, Al-Jawf, Nihm, Marib, Taiz, Al-Bayda and Lahj, including regular reports of civilian casualties, both on\nand away from frontline areas. Meanwhile, missiles continued to be launched across the border into Saudi Arabia, including the\ncapital Riyadh, which prompted retaliatory airstrikes on locations in Yemen, including civilian areas.\n\nHowever, in the five target governorates there was an overall slowdown in hostilities, especially in Sana’a governorate and Marib. As\na result, the civilian impact from conflict events in the five governorates fell in February and March compared to the prior two\nmonths.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nCIMP recorded 252 incidents with civilian impact compared to 322 in the previous reporting period, a 21% decrease in incidents.\n\n\nSa’ada again saw the most incidents, 141, more than twice as many as the next highest, Al-Hudaydah, where 65 incidents were\nrecorded.\n\n\nAs a result of this decrease in incidents, the reporting period saw a corresponding fall in the number of civilian casualties, which\nreduced from 818 to 439, a 46% decrease.\n\n\nIn particular, the deadliness of the incidents decreased, with the number of civilian fatalities declining from 507 to 202 (-61%),\nalmost three times the fall in incidents.\n\n\nHowever, although the number of civilian casualties fell, the number of child and female casualties increased. The percentage of\ncivilian casualties that were children or women more than doubled from 16% to 36%.\n\n\nAs with the previous reporting period, the vast majority of incidents were the result of airstrikes, which caused 160 incidents\n(63\n\nMention: \"CIMP\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Records civil impact incidents per reporting period.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000161:7:0", "occurrence_id": "eabd544c082f3d4c296ff0e6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: MOUVEMENTS MIXTES > **Tchad** / JUILLET-SEPTEMBRE 2021\n\n\n**CONTACTS**\n\n**Iris Blom**, Deputy Representative (Protection), Tchad\n\nblom@unhcr.org\n\n\n**Babacar Samb**, Senior Protection Officer, Tchad\n\nsamb@unhcr.org\n\n\n**Terri Moussa Adam**, Associate Protection (Mouvements Mixtes), Tchad\n\nmoussat@unhcr.org\n\n\n**Filip Hilgert**, Information Management Officer, Tchad\n\nhilgert@unhcr.org\n\n\n**LIENS**\n<u><mark>[www.unhcr.org](http://www.unhcr.org/)</mark></u> <mark>|</mark> <u><mark>[UNHCR Data Portal - Chad](https://data2.unhcr.org/en/country/tcd)</mark></u> <mark>|</mark> <u><mark>[UNHCR Chad](https://reporting.unhcr.org/node/2533)</mark></u>\n\n<u>[Twitter](https://twitter.com/UnhcrTchad)</u> | <u>[Facebook](https://www.facebook.com/UNHCRTchad/)</u>\n\n\nwww.unhcr.org 8\n\nMention: \"UNHCR Data Portal - Chad\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Le portail national fournit les chiffres de référence sur les réfugiés.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000173:10:0", "occurrence_id": "15680574288149ef8caee9e9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: EVENTOS DE SALUD PÚBLICA EN LOS MUNICIPIOS MÁS AFECTADOS POR EL DESPLAZAMIENTO Y CONFINAMIENTO\n\n\n\n**Casos / 1.000 nacidos vivos**\n\n\n≤ 32,7\n\n\n32,8 – 55,6\n\n\n55,7 – 68,1\n\n\n≥68,2\n\n\n\n**Fuentes:** Ministerio de Salud y Protección Social. SISPRO. Cubo\nSivigila 2022, Instituto Nacional de Salud. Tablero de monitoreo\nde morbilidad materna extrema 2023 y DANE. Nacimientos\n2022 y 2023 preliminares (consultada en abril 2024). 2023p=\nCifras preliminares\n\n\n\nwww.acnur.org/pais/colombia 11\n\nMention: \"DANE\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Sus cifras de nacimientos complementan el tablero de mortalidad materna.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000178:12:0", "occurrence_id": "63cdeb341fe6e0a0e6609fbd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: #### Définition de vulnérabilité\n\nLa vulnérabilité des réfugiés est identifiée comme une **combinaison de plusieurs dimensions socio-**\n**économiques**, capturées dans la base de données SR pour chaque ménage et validées avec des\nconsultations communautaires avec les réfugiés.\n\n\nChaque ménage a son propre niveau de vulnérabilité basé sur 5 dimensions:\n\n\n\n\n\n\n\n\n\n\n\n_d’éducation_\n\n\n- _Scolarisation_\n\n\n_des enfants_\n\n\n\n_d’enfants,_\n\n\n_personnes_\n\n\n_âgées,_\n\n\n_handicapés et_\n\n\n_malades, vs._\n\n\n_nombre_\n\n\n_d’adultes_\n\n\n_actifs_\n\n\n\n**Spécifiques**\n\n\n- _Présence d'_\n\n\n_handicaps ou_\n\n\n_de maladies_\n\n\n_chroniques_\n\n\n\n\n- _Participation_\n\n\n_au travail_\n\n\n- _Différences_\n\n\n_entre les_\n\n\n_sexes_\n\n\n\n\n- _Consommation_\n\n\n_alimentaire_\n\nMention: \"base de données SR\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Capture plusieurs dimensions socio-économiques par ménage, validées localement.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000182:7:0", "occurrence_id": "a7e13cba6f73f20da5f18d38", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 3. Economic Opportunities\n\n\nReview to increase security in camps for boys and girls.\n\n\nThe Refugees Proclamation accords refugees the same rights as nationals in terms of access to justice,\nincluding access to legal counselling and assistance as per Ethiopian legislation.\n\n\nIn practice, access to justice is limited for both refugees and host communities because of capacity\nconstraints and physical distance to relevant institutions, although mobile courts are operational in some\nof the refugee camps. Traditional dispute resolution mechanisms are more easily accessible and,\nparticularly in the Somali region, refugees access them in the same way as host communities. Findings\nfrom the 2020 JNS Survey seem to indicate that, overall, refugees experience more legal problems than\nhost communities and nationals living elsewhere in the country, rate these problems as more serious, and\nare less likely to resolve their problems.\n\n\nThe Refugees Proclamation mandates ARRA “to ensure that recognized refugee and asylum-seeker\nwomen shall have equal enjoyment of rights and protections enshrined under relevant laws in particular,\nspecific measures are taken to protect them from gender-based violence”. Various relevant policies to\n[prevent and address gender-based violence (GBV) are in place. These include the 1995 Constitution, the](about:blank)\n<u>[2004 Criminal Code](https://www.ilo.org/dyn/natlex/docs/ELECTRONIC/70993/75092/F1429731028/ETH70993.pdf)</u> and the <u>[2000 revised Family Code.](https://www.refworld.org/pdfid/4c0ccc052.pdf)</u> <u>[Standard Operating Procedures for the Response](http://catalog.iifphc.org/Record/359)</u>\n<u>[and Prevention of Sexual Violence](http://catalog.iifphc.org/Record/359)</\n\nMention: \"2020 JNS Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its results indicate more legal problems among refugees.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000183:5:1", "occurrence_id": "462fc2e175d21ee77c45393c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  ‘crisis‘ ADCS, 32%\n\nmore than in 2014. The percentage of HHs buying food\n\non credit and reducing essential nonfood expenses such\n\nas health or education was more than double that of\n\n2014 and triple 2013. Spending savings, selling goods\n\nand assets, and withdrawing children from school were\n\nalso more common.\n\n\nThe gap between monthly expenditures and income was\n\nestimated at $300. The percentage of HHs with debts\n\nwas up from 81% in 2014 to 89% in 2015 with HHs\n\nmainly borrowing money to buy food followed by paying\n\nrent and covering health expenses. The amount of\n\nmoney owed rocketed too: on average, HHs with debts\n\nowed $842 compared with $674 in 2014. This national\n\naverage figure has been skewed by that of HHs in BML\n\nregion, where the mean debt average was $1,151. At\n\ndistrict level HHs in El Meten, Beirut, Baabda, Bcharre\nBatroun, Aley and Chouf owed more than $1,000 on\n\naverage.\n\n\n9\n\nMention: \"ADCS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its entries track buying food on credit and reducing meals.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000142:16:0", "occurrence_id": "09034fb5e96af9c67ab92b65", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: accommodation. <sup>10</sup> [^10: These 8,500 are from a survey of one third of municipalities in Serbia. It can be assumed then, that the number\nis larger; how large, depends on the representativeness of the municipalities selected, both in terms of poverty,\nand in terms of the size of their refugee population.]\n\n\n_Refugees outside collective centres_\n\n\n49. The most recent study of the remaining refugees in Serbia, prepared in 2008\nby the Commissariat for Refugees, paints a bleak picture of the poverty,\nunemployment and poor housing conditions they experience. This is to be\nexpected, as the 61,000 that remain registered as refugees, 15 years after their\narrival, are generally those who are least able to support themselves. Without the\nsupport that they are receiving from the government, international organisations\nand local NGOs, their situation would evidently be even worse.\n\n\n50. At the same time, there is a need for a degree of caution in assessing the\nsituation of the remaining refugees. According to the Commissariat study, for\nexample, just under 20 per cent of the refugees surveyed suffer from a chronic\nillness.\n\n\n51. One year previously, however, a World Bank-sponsored Living Standards\nMeasurement Study reported that “nearly one third of the population in Serbia (32\npercent) reported that they suffered from a long-lasting disease or a health\nproblem.” These figures would suggest (a) that the health of refugees is rather\nbetter than that of the general population, or (b) that the indicators employed by\nthe two studies to assess the extent of ‘chronic illness’ and ‘long-lasting disease’ are\nquite different.\n\n\n52. While the latter interpretation would appear to have most validity, it is\ninteresting to note that the Living Standards Measurement Study was able to\ndisaggregate refugees and IDPs from its general study of the population, and to\ncompare data from 2002 with information collected in 2007. As the graph below\nindicates, the evidence appears to show a significant reduction in\n\nMention: \"Living Standards\nMeasurement Study\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its previous-year report describes the share of the population affected.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000187:34:0", "occurrence_id": "0820eeb7f1ce9c7249da2d70", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: DTM Mali – Décembre 2023\n\n\n\n\n\n\n\n\n\n\n\n|Regions-Cercles-<br>Communes|Menages|Enfant ( < 18 ans )|Col4|Adulte (18 - 59 ans )|Col6|Col7|Col8|Personne âgée ( > 59 ans )|Col10|Col11|Total<br>Femme|Total<br>Homme|Total<br>Individus|\n|---|---|---|---|---|---|---|---|---|---|---|---|---|---|\n|**Regions-Cercles-**<br>**Communes**|**Menages**|**F **|**M **|**Total**<br>**Enfant**|**F **|**M **|**Total**<br>**Adulte**|**F **|**M **|**Total**<br>**Personne**|**Total**<br>**Personne**|**Total**<br>**Personne**|**Total**<br>**Personne**|\n|KORO|1941|2171|1918|4089|1886|741|2627|132|137|269|4189|2796|6985|\n|MADOUGOU|35|45|39|84|28|7|35|7|0|7|80|46|126|\n|PEL MAOUDE|\n\nMention: \"DTM Mali – Décembre 2023\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Dénombre les ménages par âge dans les subdivisions administratives.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000194:0:5", "occurrence_id": "da467fc1917e277e5735826f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  pour des réponses durables aux crises ; (iii) un déficit de stock de contingence pour répondre aux<br>situations d’urgence dans un bref délai (Source :Profil humanitaire de la région de Gao, mai 2019).<br> <br>La**région de Ménaka** demeure instable, avec une intensification des affrontements armés, des opérations militaires et des incursions d’hommes armés dans les villages<br>entrainant de nombreux déplacements et une exacerbation des besoins humanitaires. Ménaka est particulièrement marquée par le banditisme à grande échelle qui touche<br>les humanitaires et la population civile.  Les risques de protection ont atteint un niveau inquiétant alors que l’accès humanitaire se rétrécit progressivement dans les zones<br>sud-est de la région. Par ailleurs, 61% de la population (soit 44 409 personnes) est en insécurité alimentaire. 4 658 enfants de moins de cinq ans et FEFA sont à risque de<br>malnutrition aigüe globale. 61% de la population (soit 44 623 personnes) n’ont pas accès à l’eau potable et 50% des 174 écoles sont non fonctionnelles, privant près de 8 000<br>élèves de leur droit à l’éducation. 75% des CSRef et 22% des CSCom sont non fonctionnels. Les principaux défis humanitaires de la région sont : (i) des difficultés d’accès pour<br>l’acheminement de l\n\nMention: \"Profil humanitaire de la région de Gao\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Décrit les capacités de réponse aux situations d'urgence de la région.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000203:3:0", "occurrence_id": "428b87e00ea3ba05c4193e5a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: `o` Respondents were selected randomly from the ProGres database out of the total pool of beneficiaries\nwho received the assistance from August to November 2021.\n\n`o` With a 95% confidence level, and 7% for confidence interval, the sample size needed for the total of\nhouseholds (HH) assisted was 182. Nevertheless, a total of 400 HH were selected to address the high\nmobility and lack of phone numbers on record.\n\n`o` A total of 174 households answered, out of which 141 were headed by women.\n\n\n- Limitations and challenges faced:\n\n\n`o` PoCs are usually registered at the border state of Roraima; however, many individuals continue their\njourney to the country resulting, on most occasions, in a change of phone number. As data is collected\nusing phone calls, this represents a real challenge. An alternative was to reach respondents through\nWhatsApp for those that were not initially answering.\n\n`o` This monitoring exercise was carried out during the COVID-19 pandemic, under the impact of restrictive\nmeasures to combat the spread of the disease.\n### 3.KEY FINDINGS\n\n**Household demographics**\n\n\n\n\n- Average household size 3.8\n\n- Nationalities: 97% Venezuelans and 3% other\n\nnationalities\n\n- Number of people by age group who live in the\n\nsame household:\n\n`o` 46% of families are composed of adults (18 to\n\n59 years old).\n\n`o` 33% include children between 5 and 17 years\n\nold.\n\n\n\n\n|Number of respondents disaggregated by age<br>and gender|Col2|Col3|Col4|\n|---|---|---|---|\n|**Age**|**Age**|**Gender**|**Gender**|\n|**# Cases**|**Age range**|**# Cases**|**Gender*\n\nMention: \"ProGres database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Respondents were drawn at random among registered beneficiaries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000215:4:2", "occurrence_id": "02821d4e3348bbb0a92b218d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  engagement with humanitarian organisations. The analysis of these difficulties forms the\nbasis of key pragmatic recommendations for humanitarian actors.\n\n## **2. Brief methodology**\n\n\nKey informant interviews (KIIs) were conducted with internally displaced persons with disabilities residing in 11 sub\ndistricts in inadequate living situations throughout northwest Syria over the phone by humanitarian staff who have a\nbackground in disability and inclusion. For those key informants who have hearing and communication difficulties, a\nsupport person was present to aid the interaction over the phone and often video call was used.\n\n\nThe KII’s were conducted during the time of severe restrictions on movements to staff and beneficiaries due to COVID19 related safety measures. Inclusion criterion for participants was defined as having significant functional difficulty with\none or more areas of daily functioning congruent with the Washington Group short set domains of functioning which\nare: communication, hearing, cognition, self-care, mobility and vision. <sup>8</sup>\n\n\nFor further details related to the methodology, including consent procedures and study limitations, please refer to Annex\nA.\n\n\n2 HNO (2019) Syrian Arab Republic\n3 CCCM Cluster, ISIMM data (September 2020); see also HNAP, October 2020 Mobility and Needs Monitoring NSAG & TBAF controlled areas (November 2020).\n4 HNAP (2020) Summer Report Series – Disability Overview\n5 Ibid\n6 HNAP (2019) IDP insight: Disability\n7 Ibid\n[8 Washington Group on Disability Statistics (2001). Washington Group Short Set of questions. Retrieved from: https://www.washingtongroup-](https://www.washingtongroup-disability.com/question-)\n<u>[disability.com/question-sets/wg-short-set-on-functioning-wg-ss/](https://www.washingtongroup-disability.com/question-)</u>\n\n5\n\nMention: \"ISIMM data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Mobility and cluster needs tracking serves as the reference.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000221:3:0", "occurrence_id": "3f503b4935c7a48e10e681bc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ## Secondary Data Review\n# GENDER-BASED VIOLENCE – Mozambique: Cyclone Idai and Floods\n\nAPRIL 2019\n\n\n#### Crisis Impact and Pre-Crisis Baseline BASIC NEEDS\n\n<mark>GBV does not happen in a silo and is never a standalone incident. Disasters, conflicts,</mark>\n<mark>pre-crisis vulnerabilities, stressed living conditions, and a lack of resources are all</mark>\n<mark>exacerbating factors for the manifestation of GBV and related negative coping</mark>\n<mark>mechanisms. Reduced availability and accessibility of basic services (source of income;</mark>\n<mark>health facilities, food, market etc.) as a result of the cyclone and floods increase the</mark>\n<mark>vulnerabilities of women and girls.</mark>\n\n**<mark>Water, Sanitation, and Hygiene:</mark>** <mark>According to the</mark> Demographic Health Survey <u>[(2015),](https://dhsprogram.com/pubs/pdf/AIS12/AIS12.pdf)</u>\napproximately 23.7% of people in Sofala did not have access to an improved water\nsource. A further 23.1% used a hand pump. In previous years, during periods of drought,\nit was noted that during the water collection activities, girls faced increased risk of\n[confrontations with wild animals and GBV (CARE, 2016).](https://www.humanitarianresponse.info/en/operations/southern-eastern-africa/document/ares-rapid-gender-analysis-commitment-addressing-gender)\n\n_<mark>Water</mark>_\n<mark>[Over 50% of displacement sites assessed have water sources on site (DTM R\n\nMention: \"Demographic Health Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its entries give access to water and sanitation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000226:4:3", "occurrence_id": "4c918ad7ad2bd9e55fdb74ac", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: tour-volontaire-des-migrants-137-maliens-de-retour-de-la-libye.html)</u>\n18 and 30, including women with children. One returnee,\nMrs. Ndiaye Aminata Boiré, said: \"I went to Libya since 2007.\nI was really tired there. That's why I decided to go back to my\ncountry to do something else. Even though I was not in jail\nin Libya, what I endured was more than a prison. I'm back.\nMy husband stayed there\". This is the 16th voluntary return\nflight 2018 and was the result of a collaboration between\nthe Malian Ministry of Foreign Affairs and the International\nOrganization for Migration (IOM).\n\n\nMonthly Trends Analysis West Africa 5\n\nMention: \"Monthly Trends Analysis West Africa\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Monthly analysis produced with the migration organisation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000227:6:1", "occurrence_id": "f356020e0cd7b853aee1cd85", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  the flooding due to Hurricane Katrina, in August 2005, necessitated\nthe evacuation of hundreds of thousands inhabitants of New Orleans while tens of\nthousands of others, primarily Afro-Americans, remained trapped in the city due to a\nlack of transport amenities (Cresswell 2006).\n\n\nWhile we know approximately the number of persons affected by flooding worldwide\n(106 million, on average, between 2000 and 2005 according to the _International_\n_Disaster Database_ ), and by hurricanes (38 million), the total number of people\nthreatened by an eventual increase of this kind of disaster is, however, very difficult to\nestimate (EM-DAT). No climate model is able to predict with accuracy whether or not\nthe affected zones will be densely populated and whether the damage will have tragic\nconsequences.\n\n\nApart from this difficulty of forecasting, the studies carried out after such events tend\nto relativize their effects in terms of migration in general, and long-term migration in\nparticular. Living mainly in poor countries, the victims have little mobility (Lonergan\n\n\n5\n\nMention: \"EM-DAT\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Reference for the difficult estimation of this disaster type.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000240:3:0", "occurrence_id": "ec50cafa6f84e391f38aaf0c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ##### **Outcomes**\n\n\n- More than 63 % of families had enough winter items such as blankets, heaters, mattresses, and cloths\n\n- Most of the households (95.2%) mentioned that they were able to find the items and services needed in the markets\nand shops\n\n- Most of the respondents (87.1%) mentioned that they had spent the full amount received from UNHCR by the time\nof the interview.\n\n- The top three expenditures as rated by respondents were food (1st), rent (2nd), firewood, or fuel for cooking or\nheating (3rd).\n\n- Most of the respondents (98.7%) agreed that the cash assistance improved their living conditions, reduced their\nfinancial burden (98.95%), and reduced feelings of stress (98.7%).\n\n- Most households (72%) were able to meet half or less than half of their basic needs\n\n##### **Socioeconomic conditions and well being**\n\n\n- The top three sources of income were cash assistance through ATM from humanitarian organizations, credit debts,\nand income for work, which was similar to 2019-2020 PDM results\n\n- The majority of families (87.9%) had borrowed money in the last three months, which was similar to 2019-2020 PDM\nresults\n\n- The average overall debt amount that has not been paid back was 1,846,817 LBP which was higher than the average\ndebt for WINCAP only assisted group (1,475,605) in 2019-2020 PDM results\n\n- The majority of interviewed households (63.9%) indicated being dissatisfied or very dissatisfied with their lives, that\nthey feel their standard of living is getting worse (71.7%), and that they worry about the money always or most of the\ntime (88.4%).\n\n- About 99.2% of households had at least one food\n\nMention: \"2019-2020 PDM results\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Its results serve as reference for debt and incomes.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000218:0:0", "occurrence_id": "01c813b4b108bbd6d5a4ba09", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The displacement crisis in Somalia is complex and multifaceted, driven by conflict and climatic shocks. This displacement is\noften characterized with protection risks <sup>i</sup> faced by the displaced populations. With particular focus on Kismaayo district, the\nmain drivers of displacement include conflict and drought as per the Protection and Return Monitoring Network (PRMN) <sup>ii</sup> in\nSomalia as of October, before the occurrence of El Niño flooding. The most prevalent protection risks currently being faced by\nthe populations in Kismaayo district are:\n\n      - **Risk 1: Discrimination and stigmatization, denial of resources, opportunities, services and/or humanitarian access.**\n\n      - **Risk 2: Forced recruitment and association of children in armed forces and groups.**\n\n      - **Risk 3: Gender based violence.**\n\n      - **Risk 4: Child and forced family separation.**\n\nThe Protection Analysis Update for Kismaayo is the first of a series of analysis that the Protection Cluster will be publishing\nduring 2024 focusing on priority districts. Kismaayo was selected as the first of these series as the ATMIS drawdown is already\nin process and humanitarian actors are assessing the impact associated to the drawdown, on the protection response and\nhumanitarian needs of the displaced population. In addition, the Jubaland State government has historically showed both,\ninterest, and commitment on facilitating the work of protection humanitarian actors in ensuring protection of civilians and\ndisplaced communities. The new potential scenarios to happen in Kismaayo linked to the ATMIS drawdown, justified this joint\nprotection analysis, conducted by the Protection Cluster and its partners to define the protection response strategy and\npriorities in the district for the coming period.\n\nMention: \"Protection and Return Monitoring Network\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Identifies conflict and drought as drivers of displacement.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000242:2:4", "occurrence_id": "8524010ea2dc0df4b02b4cc2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: mienky ubytovania boli obzvlášť náročné pre najzraniteľnejšie\nskupiny. Veľká časť hromadných ubytovacích zariadení v prieskume uviedla, že ubytováva osoby s\nrôznymi špecifickými potrebami, vrátane starších osôb (61 %), osamelých rodičov alebo opatrovateľov\n(24 %), osôb so zdravotným znevýhodnením (20 %) alebo osôb s vážnymi zdravotnými ťažkosťami (12\n%). Napriek tomu väčšina zariadení (68 z 83 v prieskume) uviedla, že nemá vybavenie na zabezpečenie\npotrieb starších osôb a osôb so zdravotným znevýhodnením, 36 % zariadení v prieskume uviedlo, že nie\nsú prístupné pre osoby so zdravotným znevýhodnením, 35 % nemalo bezbariérové sprchy a 59 % nemalo\nbezbariérové toalety. <sup>23</sup> [^23: UNHCR, IOM (2024), Site Mapping and Monitoring [dokument v príprave].]\n\n\nSlovenská vláda dôsledne zabezpečovala potrebné predĺženia príspevkov za ubytovanie. <sup>24</sup> Avšak\nnačasovanie týchto predĺžení — nieked\n\nMention: \"Site Mapping and Monitoring\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Ses relevés décrivent l'accessibilité des services dans les sites.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000250:18:0", "occurrence_id": "96176b0254d5bda7f58e2ac7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <u>23</u>\n\n\n\nAn estimated number of persons who entered Serbia exceeds 16,000. <sup>17</sup> According to SCRM, more than 18,000 of new accommodations were provided into asylum\nand reception centres in the same period. <sup>18</sup> [^18: Migration Open House in Obrenovac materials, March 26, the Commissariat for Refugees and Migration of\nthe Republic of Serbia] On the other hand, the number of refugees, migrants and asylum seekers in Serbia in accommodation centres and improvised shelters, hostels and similar, monthly ranged from approximately 3,0004,000 persons (4,500 in December 2018) <sup>19</sup> [^19: UNHCR Inter-Agency Operational updates 2018, https://data2.unhcr.org/], many staying in Serbia for prolonged\nperiod, some over two years.\n\n\n\nWith the coming of spring, the\nnumber of new arrivals increased\nsteadily and peaked during the\nsummer. Influx of Pakistani men\ntravelling from Greece, who had\nstayed a considerable time in\nthis EU country, arrived in Serbia\nthrough North Macedonia. During\nthe second part of the year, a significant influx of unaccompanied\nchildren was observed, entering\nfrom the direction of Bulgaria and\nNorth Macedonia. It is estimated\nthat less than a third of this population remained in the country\nby the end of 2018. <sup>20</sup> [^20: UNHCR, Desperate Journeys, January-December 2018, https://data2.unhcr.org/en/documents/\ndownload/67712#_ga=2.186617261.588054035.1557418235-1885394486.1557155396] Among the\nobserved new arrivals HCIT identified, every third child at exit\npoints was UASC.\n\n\n\n\n\nThroughout the year, CRPC and HCIT, in coordination with UNHCR, observed and\nfocused on new arrivals who had entered Serbia during the past few days or during\nthe past month. The new arrivals were counselled and interviewed, some were assisted on the spot and\n\nMention: \"UNHCR Inter-Agency Operational updates 2018\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its updates describe prolonged lengths of stay.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000254:1:2", "occurrence_id": "34e0bdccb8f5515cb763c7c8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: aux\nde mobilité en dehors des périodes de distribution pourrait donc justifier le fait que certains ménages soient\ninjoignables au moment de l’enquête.\n#### **_Les premiers résultats indiquent que le mVAM permet de collecter des_** **_informations fiables sur la consommation alimentaire…_**\n\n\n\nLes résultats indiquent que seulement 4% des ménages ont\nune consommation alimentaire pauvre (3% pour le PDM),\n24% ont une consommation alimentaire limite et 72% ont une\nconsommation acceptable. Selon les données du PDM\n(décembre 2014) et du mVAM (janvier 2015), plus de 80%\ndes ménages ont une consommation alimentaire limite ou\nacceptable (97% pour le PDM et 96% pour le mVAM).\n\nVu sous cet angle, les données collectées via téléphone\nportable (mVAM de janvier 2015) sont proches de celles\ncollectées lors des enquêtes face-to-face (PDM de\ndécembre 2014), **_ce qui revient donc à conclure que le_**\n**_mVAM peut être utilisé pour collecter des informations_**\n**_fiables sur le score de consommation alimentaire_** .\n#### **_…la diversité alimentaire…_**\n\nIl s’agit d’un indicateur qui permet d’apprécier la diversité\ndu régime alimentaire des ménages. Les résultats indiquent\nque 6% des ménages ont\n\nMention: \"données du PDM\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Ses relevés décrivent la consommation alimentaire des ménages.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000254:1:2", "occurrence_id": "ab15b2d30d75a117b994e161", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: aux\nde mobilité en dehors des périodes de distribution pourrait donc justifier le fait que certains ménages soient\ninjoignables au moment de l’enquête.\n#### **_Les premiers résultats indiquent que le mVAM permet de collecter des_** **_informations fiables sur la consommation alimentaire…_**\n\n\n\nLes résultats indiquent que seulement 4% des ménages ont\nune consommation alimentaire pauvre (3% pour le PDM),\n24% ont une consommation alimentaire limite et 72% ont une\nconsommation acceptable. Selon les données du PDM\n(décembre 2014) et du mVAM (janvier 2015), plus de 80%\ndes ménages ont une consommation alimentaire limite ou\nacceptable (97% pour le PDM et 96% pour le mVAM).\n\nVu sous cet angle, les données collectées via téléphone\nportable (mVAM de janvier 2015) sont proches de celles\ncollectées lors des enquêtes face-to-face (PDM de\ndécembre 2014), **_ce qui revient donc à conclure que le_**\n**_mVAM peut être utilisé pour collecter des informations_**\n**_fiables sur le score de consommation alimentaire_** .\n#### **_…la diversité alimentaire…_**\n\nIl s’agit d’un indicateur qui permet d’apprécier la diversité\ndu régime alimentaire des ménages. Les résultats indiquent\nque 6% des ménages ont\n\nMention: \"mVAM\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Complète l'enquête par téléphone pour la période suivante.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000261:2:1", "occurrence_id": "944f54af131bd9f93d2e8bb3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  cash available, they only need\nto make one trip to an ATM to successfully withdraw cash\ncompared to up to 10 trips previously, also saving on\ntransportation costs – up to four times, as reported by some.\n\n\n\n**_“My confidence and security have significantly increased when_**\n**_withdrawing assistance from ATMs. I no longer lose money while_**\n\n\n\n**_exchanging from Lebanese Pound to USD for rent payments.”_**\n\n\n\n**_Female participant in a focus group discussion_**\n\n\n\nSome assisted families reported an increase in discriminatory\nstatements after the introduction of dual currency redemption,\nparticularly on social media, but did not face any serious\nincidents in their daily life as a result. Rather, much of the\ntensions already existed within and between communities,\nwith the common misconception that Syrians had always had\naccess to USD and are better covered than the Lebanese, while\nthe Lebanese allegedly do not receive assistance.\n\n\nThe ATM experience survey covering over 3,000 Syrian refugee\nhouseholds confirmed the results of the focus group\ndiscussions. The number of assisted families reporting facing\nprotection risks further reduced in June to 1.3 percent from 2\npercent in May and 12 percent in February. Similarly, the\nnumber of people reporting paying a commission also reduced\nfrom 3 percent in February to 0.1 percent in May and 0.03\npercent in June. People reporting dignified access to the ATMs\nwas 95 percent in June, in line with May values and up from 85\npercent in April and 74 percent in February.\n\n\n\nIn interviews with governors, mayors, and heads of\nmunicipalities, the local authorities also acknowledged the\nbenefits of dual currency redemption in reducing crowding and\nvisibility and improving operational efficiency. At the same\ntime, many raised concerns about the impact on inflation, the\nsocioeconomic situation of Lebanese communities, and\npotential outflows to Syria, noting that most assisted Syrian\nfamilies are in fact not covering their basic needs with the cash\nprovided, making it unlikely that they will be able\n\nMention: \"ATM experience survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"More than three thousand households confirming the focus group results.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000263:1:0", "occurrence_id": "ffbc426befcbbcc71f39f374", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: KEY FIGURES\n\n\n\nREGIONAL BUREAU FOR SOUTHERN AFRICA\n#### **POPULATIONS THAT UNHCR IS MANDATED TO PROTECT AND ASSIST**\n\nAs of 30 June 2023\n\n\nMAP OF POPULATIONS THAT UNHCR IS MANDATED TO PROTECT AND ASSIST\n\n**520,076** **REF**\n\n\n**3,452** **ASY**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n**42,183** **REF**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\ndo not imply official endorsement or acceptance by the United Nations\n\n\n\nMalawi (659,278), The Democratic Republic of the Congo (270,005), Mozambique (41,668) and\nZimbabwe (41,535).\n\n\n\n**Author: UNHCR DIMA - RSA** Contact : rsarbdima@unhcr.org **Source: REF, ASY, OOC, RET** (UNHCR PRIMES, Government); **IDP DRC** (OCHA); IDP Zimbabwe & Mozambique (IOM); **IDP ROC** (Government, Ministry of Social Affairs and Humanitarian Action (MASAH).\n\n\n** REF = Refugee; ASY = Asylum-seeker; OOC = Other person of concern; RET = Returnee              DRC = Democratic Republic of the Congo   ROC = Repulic of the Congo   IDP = Internal Displaced Person  IDP CFI = Conflic-induced IDPs   IDP RET = Returned IDPs\n\n\n\nDate of creation : As of 30 June 2023\n\n\nFor more information visit: UNHCR Data Portal\n\nMention: \"UNHCR PRIMES\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Supplies counts of refugees, asylum seekers and returnees.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000263:1:0", "occurrence_id": "6586061bf365983f1874c201", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: KEY FIGURES\n\n\n\nREGIONAL BUREAU FOR SOUTHERN AFRICA\n#### **POPULATIONS THAT UNHCR IS MANDATED TO PROTECT AND ASSIST**\n\nAs of 30 June 2023\n\n\nMAP OF POPULATIONS THAT UNHCR IS MANDATED TO PROTECT AND ASSIST\n\n**520,076** **REF**\n\n\n**3,452** **ASY**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n**42,183** **REF**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\ndo not imply official endorsement or acceptance by the United Nations\n\n\n\nMalawi (659,278), The Democratic Republic of the Congo (270,005), Mozambique (41,668) and\nZimbabwe (41,535).\n\n\n\n**Author: UNHCR DIMA - RSA** Contact : rsarbdima@unhcr.org **Source: REF, ASY, OOC, RET** (UNHCR PRIMES, Government); **IDP DRC** (OCHA); IDP Zimbabwe & Mozambique (IOM); **IDP ROC** (Government, Ministry of Social Affairs and Humanitarian Action (MASAH).\n\n\n** REF = Refugee; ASY = Asylum-seeker; OOC = Other person of concern; RET = Returnee              DRC = Democratic Republic of the Congo   ROC = Repulic of the Congo   IDP = Internal Displaced Person  IDP CFI = Conflic-induced IDPs   IDP RET = Returned IDPs\n\n\n\nDate of creation : As of 30 June 2023\n\n\nFor more information visit: UNHCR Data Portal\n\nMention: \"UNHCR Data Portal\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Public portal presenting the data as of 30 June.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000264:32:0", "occurrence_id": "3de354744b7e823992b5aa32", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: DTM Mali – Juillet 2020\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n|Régions-Cercles-<br>Communes|Ménages|Enfant ( < 18 ans )|Col4|Col5|Adulte (18 - 59 ans )|Col7|Col8|Personne âgée ( > 59 ans )|Col10|Col11|Total<br>Femme|Total<br>Homme|Total<br>Individus|\n|---|---|---|---|---|---|---|---|---|---|---|---|---|---|\n|**Régions-Cercles-**<br>**Communes**<br>|<br>**Ménages**|**F **|**M **|** Total**<br>**Enfant**|**F **|**M**|**  Total**<br>**Adulte**|**F **|**M **|** Total**<br>**Personne**<br>**âgée**|** Total**<br>**Personne**<br>**âgée**|** Total**<br>**Personne**<br>**âgée**|** Total**<br>**Personne**<br>**âgée**|\n|TOUNA|32|66|61|127|36|17|53|3|1|4\n\nMention: \"DTM Mali – Juillet 2020\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Dénombre les ménages par âge dans les subdivisions administratives.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000266:3:1", "occurrence_id": "faf31636f02a76889b06b09d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: social justice. With more than 25 delegations represented,](https://www.foundationmaxvanderstoel.nl/nieuws/nieuws_item/t/discussing_migration_and_social_justice_in_ouagadougou_burkina_faso)</u>\nof which many from African countries, the seminar was an\nopportunity to generate discussion on migration. During\nthe seminar, speakers emphasized the fact that most African\nmigrants stay within the continent, and mostly within\nthe region. They also highlighted the need to elaborate a\nstrategy to manage and monitor African migration.\n\n\nMonthly Trends Analysis West Africa 4\n\nMention: \"Monthly Trends Analysis West Africa\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Ses analyses soulignent le besoin d'une stratégie de suivi des migrations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000280:9:5", "occurrence_id": "bedd1646921cff82e0c261ba", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  hecha las personas refugiadas y solicitantes\nde asilo en Ciudad de Panamá destacaron un alto nivel de\ndiscriminación que se manifiesta en diferentes contextos\ncomo las escuelas, los centros de salud y los centros de policía,\nentre otros.\n\n\nEn general, los prejuicios relacionados con la nacionalidad\ncolombiana hacen que muchos refugiados sufran una doble\nestigmatización no solo por ser refugiados sino también por ser\ncolombianos y ser tildados como narcotraficantes, guerrilleros,\n\n\n\n20 Diagnóstico Participativo 2014-2015 Diagnóstico Participativo 2014-2015 21\n\nMention: \"Diagnóstico Participativo 2014-2015\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Recoge los testimonios sobre estigmatización de las personas desplazadas.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000246:2:0", "occurrence_id": "4a10900831cfc29084ba1903", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Children and women comprised a significant percentage of arrivals in Italy from Algeria, Libya, and Tunisia in 2023.\nChildren accounted for 17%, consistent with 2022. Most of the children were unaccompanied and separated children\n(UASC), primarily Guinean, Tunisian, Egyptian and Gambian nationals. The share of women among the arrivals increased\nto 10% in 2023 compared to 7% in 2022. This was primarily attributed to changes in the top nationalities, as indicated\nabove, with a notable rise in arrivals from, for example, Guinea and Côte d’Ivoire, where women constituted a large\nshare of arrivals (e.g., of the 16,051 Côte d’Ivoire nationals who arrived in Italy, 5,055 were women, constituting approximately 31.5%, while 7,809 were men, making up about 48.6%). The predominant nationalities of women arriving in Italy\nremained similar to those in 2022, including Ivorian, Guinean, Tunisian, Cameroonian, and Syrian nationals. However,\nmen continued to constitute the majority of arrivals in Italy through the central Mediterranean Sea.\n\nIn 2023, more than 3,105⁷ migrants and refugees are known to have lost their lives or gone missing at sea while\nattempting to cross to Europe along the three Mediterranean routes (eastern, central and western Mediterranean), an\nincrease compared to more than 2,500 deaths recorded in 2022. 61% of the deaths in 2023 took place on the central\nMediterranean Sea. According to IOM’s Missing Migrants Project, more than 700 deaths occurred off the coast of Tunisia, more than 680 off Libya and more than 460 off Italian shores. The increasing number and proportion of deaths off\nthe coast of Tunisia is unprecedented in the last 10 years. Notably, more than 1,900 <sup>8</sup> deaths in the central Mediterranean\nSea in 2023 were linked to mass shipwrecks involving the loss of more than 20 lives at sea. However, the\n\nMention: \"Missing Migrants Project\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its death counts along the Central Mediterranean route.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000284:11:1", "occurrence_id": "480de16822fc828e210e6cd1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: eva which is identifying areas of high restoration](https://unepgrid.ch/en/activity/194E6705)</u>\nand/or protection potential to reduce risk from certain natural hazards (see Figure 4). The spatial\n[data for this project is developed on a global level and can be accessed through a MapX data](https://app.mapx.org/?project=MX-2LD-FBB-58N-ROK-8RH)\n<u>[project. On a national level ACTED, a French NGO, is implementing a project in Rakhine state](https://app.mapx.org/?project=MX-2LD-FBB-58N-ROK-8RH)</u>\nwhich involves planting and conserving mangrove forests.\n\n\n6 OCHA Policy Development and Studies Branch (2016) “Understanding the climate-conflict nexus from a\nhumanitarian perspective: a new quantitative approach”. Available:\n<u>[https://reliefweb.int/sites/reliefweb.int/files/resources/OCHA_OP17_climate%20change_online.p](https://reliefweb.int/sites/reliefweb.int/files/resources/OCHA_OP17_climate%20change_online.pdf)</u>\n<u>[df](https://reliefweb.int/sites/reliefweb.int/files/resources/OCHA_OP17_climate%20change_online.pdf)</u>\n\n\n12\n\nMention: \"MapX data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The global project is accessible through this project platform.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000286:11:1", "occurrence_id": "c83812e37e3d8866fc6796e9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \n**890**\n\n\n\n**63,590** <sup>*</sup>\n\n\n**730**\n\n\n\n**32,610** <sup>*</sup>\n\n\n**200**\n\n\n\n**n/a**\n\n\n**n/a**\n\n\n\n**1,337**\n\n\n**4**\n\n\n\n**2,413**\n\n\n**209**\n\n\n\n**633**\n\n\n**9**\n\n\n\n**n/a**\n\n\n**200** 7\n\n\n\n**762**\n\n\n**11**\n\n\n\n4 *Numbers of departures and numbers of deaths for years 2013-2015 are estimates derived from interviews with refugees who arrived on boats in Indonesia,\nMalaysia and Thailand from 2013-2015. Figures during this timeframe have been rounded to the nearest ten to denote the increased reliance on estimates\nrelative to more recent years.\n5 This includes both Rohingya refugees and Bangladeshi migrants However, as of 2018, the vast majority of those travelling by sea are Rohingya refugees.\n6 \u0007The number of dead and missing are already included in the count of maritime movements that year. For example, 2,413 persons moved in 2020, and 209\nof this number died or went missing.\n7 \u0007In 2017, more than 200 Rohingya fleeing the violence in Rakhine with rafts and boats died between August and October 2017 after their crafts capsized on\nthe river separating Bangladesh and Myanmar. As these are not sea movements, no number is recorded under “maritime movements”.\n\n\n10 **Left Adrift at Sea: Dangerous Journeys of Refugees Across the Bay of Bengal and Andaman Sea**\n\nMention: \"interviews with refugees\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Departures and deaths are estimated from these interviews.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000291:54:0", "occurrence_id": "28d606d5d24c415e0701d178", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Fuente: UDFI-CONADEH, con datos del SI-Quejas, Enero-Diciembre de 2017\n\n\nEn relación con los municipios receptores de personas desplazadas internamente, la disponibilidad de los datos permitió analizar solo el 43.1% de los casos (339) de desplazamiento registrados por el CONADEH en el 2017.\n\n\nComo se muestra en el mapa de flujo de casos de personas desplazadas internamente, los dos principales municipios\nexpulsores, el Distrito Central y San Pedro Sula, son a su vez los de mayor recepción de personas desplazadas.\n\n\nDe acuerdo con los datos, el patrón de desplazamiento que refiere a personas que se movilizan dentro del mismo municipio es menor, en comparación, a las personas que huyen de la violencia hacia otros departamentos y municipios. En\nparticular, hacia los sitios con mayores niveles de desarrollo.\n\n\n**Ilustración No.5 Mapa de Flujo de Casos de Personas Desplazadas Internamente según Municipio**\n\n**Expulsor y Receptor**\n\n\n\n\n\n**54**\n\nMention: \"datos del SI-Quejas\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Los reclamos registrados durante el año en los municipios receptores.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000294:1:0", "occurrence_id": "019133313f1923067adb3608", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ### **FUNDING THE RESPONSE**\n\nFollowing a mid-year review, the 2020 RRRP required US$ 638.7M, and stood **22 per cent funded** by the\n[end of 2020, according to data recorded in the Refugee Funding Tracker. RRRP partners worked to deliver](http://refugee-funding-tracker.org/)\nprotection and basic assistance within the available resources, and many positive outcomes were achieved.\nHowever, underfunding also resulted in certain standards not being met, and scale-down or discontinuation\nof programmes. It also resulted in a lack of investment in long-term and sustainable interventions, particularly\nin the livelihoods sector. In Angola, for example, lack of funds prevented implementation of the Joint\nLivelihood Strategy, setting back progress towards refugees’ self-reliance. In Zambia, only 10 per cent of\nrefugees could be supported with vocational and technical skills training, with many youths falling through\nthe cracks. Funding shortfalls for food assistance resulted in ration reductions and cuts to cash assistance,\nfor example in Burundi, Republic of the Congo, Rwanda and Tanzania. Other sectors were also impacted by\nunderfunding, for example in Republic of the Congo, schools still lacked appropriate furniture, including\nchairs, which negatively impacted quality of education. In Rwanda, limited budget affected referral services\nto secondary and tertiary hospitals and limited access to vaccination. In Tanzania, limited funding meant\nhost communities could not be included in shelter and non-food item (NFI) programming. In Uganda,\nunderfunding compromised capacity to provide mental health and psychosocial support, and hindered work\nin child protection and gender-based violence (GBV) response.\n\n### **COVID-19: ADAPTING AND RESPONDING**\n\n\nIn response to the COVID-19 pandemic, RRRP partners reprioritized and reprogrammed, adapting to reduce\nthe risk of spreading the virus, and to address key needs and impacts.\n\nMention: \"Refugee Funding Tracker\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its records give the percentage of financing reached.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000297:14:0", "occurrence_id": "1378d963845887425651783c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Southeast, the second main source is NGOs, it is municipalities in the Aegean, Social Assistance\n\nand Solidarity Foundations in Marmara and UNHCR for Central Anatolia & Other region.\n\n\n**Minimal differences were identified in terms of access to assistance by female versus male**\n\n**heads of households.** Accordingly, 55% of female head of households receive assistance\n\ncompared to 52% of male head of households. After ESSN and CCTE, female heads of\n\nhouseholds receive the most assistance through UNHCR cash programmes.\n\n\nDifferences in distribution of assistance per geographical location were also identified. Assistance\n\nis mostly provided in the Aegean where 68% confirmed receiving support, followed by Southeast\n\n(54%), Central Anatolia & Other (49%) and Marmara (48%).\n\n### Access to Basic Needs and Hygiene Items\n\n##### Access to Basic Needs\n\n\nSurvey findings indicate that **88% of respondents are not fully able cover their monthly**\n\n**expenses and basic household needs.** Only 12% expressed being able to cover their needs in\n\nfull through existing sources of income. The most socio-economically vulnerable populations\n\ngroups are Afghans and Iraqis, with 98% and 93% respectively stating they were either not or\n\nonly partially able to cover their basic needs.\n\n\nFrom a gender perspective, differences in socio-economic vulnerability were identified.\n\nAccordingly, **53% of female headed households, compared to 43% of male headed**\n\n**households, are unable to cover their monthly expenses.** It is also noted that the expenditure\n\nof male headed households is 15% higher compared to female headed households.\n\n\nNotable differences between geographical areas were also identified in terms of respondents’\n\nability to cover monthly expenses. While in the Central Anatolia & Other region, 16% are fully and\n\n49% are partially able to cover their expenses; only 9% are fully and 35% are partially able to\n\ncover expenses in the Marmara region.\n\nMention: \"Access to Basic Needs\n\n\nSurvey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its results give the share unable to cover monthly expenses.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000314:8:0", "occurrence_id": "c78bcad9c6cb7d1c84ae4fdd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **COLOMBIA** | October 2024\n\n\nAccording to the Index of Probability of Occurrence of Recruitment -IPOR- the high risk of recruitment, use and utilization is\nconcentrated in 232 municipalities in the Pacific region (Nariño, Cauca, Valle del Cauca, Chocó), the Amazon region (Putumayo,\nCaquetá, Meta and Guaviare), border areas with Venezuela (Norte de Santander and Arauca), southern Bolívar and Bajo Cauca\nAntioqueño <sup>xxxiv</sup> . In these geographical areas there are enclaves of illicit economies associated with drug trafficking, micro-trafficking\nand illegal mining in which children and adolescents are used and used and the risks of sexual violence, sexual exploitation and human\ntrafficking are exacerbated. In addition, they face a context of violation of rights due to structural conditions that impede access to\nessential rights and services such as education, food security and health. At the territorial level, gaps persist in the activation of\nemergency funds in the face of imminent threats and in the construction of intersectoral strategies for the construction of protective\nenvironments.\n\nSocial control and constant threats by non-state armed groups in high-risk areas are having an impact on the mental health and\npsychosocial well-being of children and adolescents, families, and communities, creating obstacles to the construction of life projects\nand community processes and encouraging negative coping behaviors such as the use of psychoactive substances. Additionally,\nattacks on schools and threats to teachers, according to the monitoring conducted by COALICO <sup>xxxv</sup>, doubled between 2022 (23 attacks)\nand 2023 (46 attacks), reaching 16,764 children and adolescents affected in the last year, which limits the capacity of education as a\nprotective environment.\n\nRefugee and migrant children and adolescents face protection risks in geographic areas with the presence of non-state armed groups\nand in transit through the country\n\nMention: \"Index of Probability of Occurrence of Recruitment\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Locates high recruitment risk in 232 municipalities.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000324:12:1", "occurrence_id": "ac182820e644c01319cf7030", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: desplazamiento-masivo-y-restricciones-la-movilidad-en-catatumbo-norte-de-santander?_gl=1*10u96yb*_ga*NDIwMjM2MTgyLjE3MjM0NzIyMDc.*_ga_E60ZNX2F68*MTc0MTg3NjEwOS40NC4xLjE3NDE4Nzc0NjcuNTUuMC4w)\n[Santander)](https://reliefweb.int/report/colombia/colombia-reporte-de-situacion-no-02-necesidades-humanitarias-por-desplazamiento-masivo-y-restricciones-la-movilidad-en-catatumbo-norte-de-santander?_gl=1*10u96yb*_ga*NDIwMjM2MTgyLjE3MjM0NzIyMDc.*_ga_E60ZNX2F68*MTc0MTg3NjEwOS40NC4xLjE3NDE4Nzc0NjcuNTUuMC4w)\nviii Ibid.\nix [Governorate of Norte de Santander, PMU Balance Catatumbo, Bulletin N°53 of 17/03/2025.](https://x.com/GoberNorte/status/1901945509239025976/photo/1)\n[x UNODC, Monitoring Report on Territories affected by Coca Cultivation 2022.](https://www.unodc.org/documents/colombia/2023/septiembre-9/INFORME_MONITOREO_DE_TER\n\nMention: \"Monitoring Report on Territories affected by Coca Cultivation 2022\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Son suivi décrit les territoires affectés par les cultures illicites.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000327:1:0", "occurrence_id": "a2f409542c636fc6444624d5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: RECOMMENDATIONS FOR SUSTAINABLE ACCOMMODATION CENTRES IN ROMANIA\n\n## Executive Summary\n\n\n\nFollowing the full-scale invasion of Ukraine by\nthe Russian Federation that began on February\n24, 2022, over 175,000 refugees from Ukraine\nregistered for Temporary Protection in Romania1.\nIn response, the Government of Romania has\nprovided a housing support through various\nprogrammes ensuring the basic needs of refugees\nfrom Ukraine are met, initially through the 50/20\nProgramme (Emergency Ordinance no. 15/2022)\nand then with the transition to established\nprogrammes on 1 May 2023 that differentiated\nlump-sum allowances for holders of temporary\nprotection to cover accommodation and food\ncosts2. On 28 June 2024, a new ordinance3,\nEO No. 96/2024, was adopted, introducing\ntemporary housing support for newly arrived\nrefugees for three to four months and removing\nlegal barriers preventing access to social\nprotection services. Refugees who received\ntemporary protection before July 1, 2024, are\nentitled to social benefits, including child\nallowances and unemployment benefits, while\nthose receiving temporary protection after that\ndate would have access to a one-time,\nfour-month lump-sum allowance before\ntransitioning to the said benefits.\n\n\n\nWhile these programmes aim to support basic\nneeds, findings from the several assessments\nconducted in the past years reveal persistent\nhousing challenges, underscoring the growing\nfinancial strain and increased reliance on\ncollective centres. Thus, the 2023 Multi-Sector\nNeeds Assessment (MSNA)4 and the 2024 Social\nand Economic Insights Survey (SEIS)5 reveal\nsignificant trends and challenges faced by\nUkrainian refugees in Romania regarding the\nhousing situation. In 2023, 4% of refugees lived\nin collective centres, with an additional 8%\nsharing housing or residing in hotels or hostels.\nBy 2024, the percentage of refugees living in\ncollective centres increased to 7%, though the\nmajority (93%) continued to live independently.\nA marked change occurred in refugees' ability to\ncover housing costs. In 2023, 78% could manage\nrent and utilities though many\n\nMention: \"2023 Multi-Sector\nNeeds Assessment\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its results describe growing financial pressure.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000327:1:0", "occurrence_id": "c63b0bf59dbee8f0c3a5f59a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: RECOMMENDATIONS FOR SUSTAINABLE ACCOMMODATION CENTRES IN ROMANIA\n\n## Executive Summary\n\n\n\nFollowing the full-scale invasion of Ukraine by\nthe Russian Federation that began on February\n24, 2022, over 175,000 refugees from Ukraine\nregistered for Temporary Protection in Romania1.\nIn response, the Government of Romania has\nprovided a housing support through various\nprogrammes ensuring the basic needs of refugees\nfrom Ukraine are met, initially through the 50/20\nProgramme (Emergency Ordinance no. 15/2022)\nand then with the transition to established\nprogrammes on 1 May 2023 that differentiated\nlump-sum allowances for holders of temporary\nprotection to cover accommodation and food\ncosts2. On 28 June 2024, a new ordinance3,\nEO No. 96/2024, was adopted, introducing\ntemporary housing support for newly arrived\nrefugees for three to four months and removing\nlegal barriers preventing access to social\nprotection services. Refugees who received\ntemporary protection before July 1, 2024, are\nentitled to social benefits, including child\nallowances and unemployment benefits, while\nthose receiving temporary protection after that\ndate would have access to a one-time,\nfour-month lump-sum allowance before\ntransitioning to the said benefits.\n\n\n\nWhile these programmes aim to support basic\nneeds, findings from the several assessments\nconducted in the past years reveal persistent\nhousing challenges, underscoring the growing\nfinancial strain and increased reliance on\ncollective centres. Thus, the 2023 Multi-Sector\nNeeds Assessment (MSNA)4 and the 2024 Social\nand Economic Insights Survey (SEIS)5 reveal\nsignificant trends and challenges faced by\nUkrainian refugees in Romania regarding the\nhousing situation. In 2023, 4% of refugees lived\nin collective centres, with an additional 8%\nsharing housing or residing in hotels or hostels.\nBy 2024, the percentage of refugees living in\ncollective centres increased to 7%, though the\nmajority (93%) continued to live independently.\nA marked change occurred in refugees' ability to\ncover housing costs. In 2023, 78% could manage\nrent and utilities though many\n\nMention: \"2024 Social\nand Economic Insights Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its trends describe hardships of refugees in collective centres.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000335:5:2", "occurrence_id": "68111f749faa5e8e96014957", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  as a whole. The 2013 Human\nDevelopment Report indicated that 87.7 per cent of the\npopulation are living in poverty country-wide, suggesting\nthat the poverty rate in Orientale province might be even\nhigher today (UNDP, March 2013). A study undertaken\nat the end of 2012 concluded that the majority of the\nprovince’s population was suffering a chronic livelihood\ncrisis and had urgent food and nutritional needs (WFP/\nGovernment of DRC February 2013).\n\n\nDisplacement within Orientale province, and from neighbouring provinces and countries has increased pressure\non already scarce resources. As elsewhere in DRC, a lack\nof infrastructure and government capacity has hampered\nthe provision of security and the delivery of an effective\nhumanitarian response. IDPs in particular face challenges\nin accessing health care, education and land, in part the\nresult of financial constraints but also because of tensions with local communities over resources. Displacement tends to be protracted. More than 55 per cent of\ncurrent IDPs having been displaced since 2008 or 2009\n(IDMC field interview with OCHA, June 2013). Many have\nbeen displaced more than once, but accurate data on\nmultiple displacement is limited.\n\n\nAs such displacement trends in Orientale province appear\nto mirror those elsewhere in DRC, with multiple and protracted displacement a common feature. That said, the\nparticularly vicious nature of LRA violence, and the extreme fear and trauma it has generated, creates specific\nchallenges both for those affected and those responding\nto their needs. The extent to which displacement is protracted is important in this sense. There are IDPs living in\nprotracted displacement in other provinces, such as North\nand South Kivu, but a comparison of the number of new\nIDPs and returnees against cumulative figures suggests\nthat displacement there is more fluid than in areas where\nthe LRA has been active (IDMC, 2008-2013).\n\nMention: \"2013 Human\nDevelopment Report\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Gives the national share of the population living in poverty.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000348:11:1", "occurrence_id": "1186186442643f838eb45ea6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: LAC Rule of Law Report 2023,_ 11\nJanuary 2023, available at: https://ilacnet.org/new-ilac-report-surveys-justice-sector-under-taliban-rule-2/.\n\nxvi Christopher Lehmann, _Justice Matters: A Status Report on Afghanistan Since the Taliban Takeover,_ _ILAC Rule of Law Report 2023,_ 11\nJanuary 2023, available at: https://ilacnet.org/new-ilac-report-surveys-justice-sector-under-taliban-rule-2/.\n\nxvii <u>https://www.care-international.org/sites/default/files/2022-</u>\n<u>08/Afghanistan%20Food%20Security%20Report_advocacy%20brief_August%202022.pdf</u>\n\nxviii https://response.reliefweb.int/afganistan/humanitarian-access-snapshot\n\n\n**Methodology**\n\n\nData from DRC, INTERSOS, IOM, IRC, NRC, Cordaid, AABRAR, DHSA/TKG and UNHCR (in partnership with WAW and ARAA)\nwhich include 39,000 (46% are female respondent) Household-level Surveys, 6,600 Key Informants Interviews (47% female\nKI) and a significant number of Focus Group Discussions in 21 provinces/143 districts—have been used for this report.\nHowever, the protection cluster continues to face challenges in analysing the FGD data. Moreover, data on human rights\nviolations are not made available on a quarterly basis which leads to relying on anecdotal data and anonymized research.\nSensitivity around collecting data and requirement by authorities to accompany assessment limiting partners’ capacity to\ngather quality information. The analysis is guided by the Global Protection Cluster Protection Analysis Framework (PAF).\nOther sources of data that are\n\nMention: \"IOM\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"One of the organisations whose records feed the snapshot.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000348:11:1", "occurrence_id": "a109e42f2d4e26b6469b636e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: LAC Rule of Law Report 2023,_ 11\nJanuary 2023, available at: https://ilacnet.org/new-ilac-report-surveys-justice-sector-under-taliban-rule-2/.\n\nxvi Christopher Lehmann, _Justice Matters: A Status Report on Afghanistan Since the Taliban Takeover,_ _ILAC Rule of Law Report 2023,_ 11\nJanuary 2023, available at: https://ilacnet.org/new-ilac-report-surveys-justice-sector-under-taliban-rule-2/.\n\nxvii <u>https://www.care-international.org/sites/default/files/2022-</u>\n<u>08/Afghanistan%20Food%20Security%20Report_advocacy%20brief_August%202022.pdf</u>\n\nxviii https://response.reliefweb.int/afganistan/humanitarian-access-snapshot\n\n\n**Methodology**\n\n\nData from DRC, INTERSOS, IOM, IRC, NRC, Cordaid, AABRAR, DHSA/TKG and UNHCR (in partnership with WAW and ARAA)\nwhich include 39,000 (46% are female respondent) Household-level Surveys, 6,600 Key Informants Interviews (47% female\nKI) and a significant number of Focus Group Discussions in 21 provinces/143 districts—have been used for this report.\nHowever, the protection cluster continues to face challenges in analysing the FGD data. Moreover, data on human rights\nviolations are not made available on a quarterly basis which leads to relying on anecdotal data and anonymized research.\nSensitivity around collecting data and requirement by authorities to accompany assessment limiting partners’ capacity to\ngather quality information. The analysis is guided by the Global Protection Cluster Protection Analysis Framework (PAF).\nOther sources of data that are\n\nMention: \"UNHCR\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its data are produced in partnership with two national organisations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000348:11:1", "occurrence_id": "254033857be078c0a9828102", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: LAC Rule of Law Report 2023,_ 11\nJanuary 2023, available at: https://ilacnet.org/new-ilac-report-surveys-justice-sector-under-taliban-rule-2/.\n\nxvi Christopher Lehmann, _Justice Matters: A Status Report on Afghanistan Since the Taliban Takeover,_ _ILAC Rule of Law Report 2023,_ 11\nJanuary 2023, available at: https://ilacnet.org/new-ilac-report-surveys-justice-sector-under-taliban-rule-2/.\n\nxvii <u>https://www.care-international.org/sites/default/files/2022-</u>\n<u>08/Afghanistan%20Food%20Security%20Report_advocacy%20brief_August%202022.pdf</u>\n\nxviii https://response.reliefweb.int/afganistan/humanitarian-access-snapshot\n\n\n**Methodology**\n\n\nData from DRC, INTERSOS, IOM, IRC, NRC, Cordaid, AABRAR, DHSA/TKG and UNHCR (in partnership with WAW and ARAA)\nwhich include 39,000 (46% are female respondent) Household-level Surveys, 6,600 Key Informants Interviews (47% female\nKI) and a significant number of Focus Group Discussions in 21 provinces/143 districts—have been used for this report.\nHowever, the protection cluster continues to face challenges in analysing the FGD data. Moreover, data on human rights\nviolations are not made available on a quarterly basis which leads to relying on anecdotal data and anonymized research.\nSensitivity around collecting data and requirement by authorities to accompany assessment limiting partners’ capacity to\ngather quality information. The analysis is guided by the Global Protection Cluster Protection Analysis Framework (PAF).\nOther sources of data that are\n\nMention: \"Household-level Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Thirty-nine thousand household interviews, 46 percent with women.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000348:11:1", "occurrence_id": "26a60c36bae7639c27a4e4f4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: LAC Rule of Law Report 2023,_ 11\nJanuary 2023, available at: https://ilacnet.org/new-ilac-report-surveys-justice-sector-under-taliban-rule-2/.\n\nxvi Christopher Lehmann, _Justice Matters: A Status Report on Afghanistan Since the Taliban Takeover,_ _ILAC Rule of Law Report 2023,_ 11\nJanuary 2023, available at: https://ilacnet.org/new-ilac-report-surveys-justice-sector-under-taliban-rule-2/.\n\nxvii <u>https://www.care-international.org/sites/default/files/2022-</u>\n<u>08/Afghanistan%20Food%20Security%20Report_advocacy%20brief_August%202022.pdf</u>\n\nxviii https://response.reliefweb.int/afganistan/humanitarian-access-snapshot\n\n\n**Methodology**\n\n\nData from DRC, INTERSOS, IOM, IRC, NRC, Cordaid, AABRAR, DHSA/TKG and UNHCR (in partnership with WAW and ARAA)\nwhich include 39,000 (46% are female respondent) Household-level Surveys, 6,600 Key Informants Interviews (47% female\nKI) and a significant number of Focus Group Discussions in 21 provinces/143 districts—have been used for this report.\nHowever, the protection cluster continues to face challenges in analysing the FGD data. Moreover, data on human rights\nviolations are not made available on a quarterly basis which leads to relying on anecdotal data and anonymized research.\nSensitivity around collecting data and requirement by authorities to accompany assessment limiting partners’ capacity to\ngather quality information. The analysis is guided by the Global Protection Cluster Protection Analysis Framework (PAF).\nOther sources of data that are\n\nMention: \"anecdotal data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Lacking quarterly publication, the analysis must rely on these items.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000351:37:3", "occurrence_id": "ca3a599f1add13cc36c52046", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |1969|\n|Republic of**KAZAKHSTAN**|6,121|Unknown|15,844|9|1999|1999|1995|\n|State of**KUWAIT**|398|-|1,519|157|-|-|1969|\n|**KYRGHYZ** Republic|3,292|Unknown|3,753|453|1996|1996|1992|\n\n\n222 United Nations Office of the High Commissioner for Refugees, 17 April 2006, (as of 1 January 2005) at www.unhcr.org.\n223 Internal Displacement Monitoring Centre (as of 24 July 2006) at www.globalidpproject.org.\n224 United Nations Office of the High Commissioner for Refugees, 17 April 2006, (as of 1 January 2005) at www.unhcr.org.\n225 United Nations Office of the High Commissioner for Refugees, 17 April 2006, (as of 1 January 2005) at www.unhcr.org.\n226 United Nations Office of the High Commissioner for Refugees, September 2005.\n227 European Union, 2002.\n228 Norwegian Refugee Council, December 2005.\n229 Norwegian Refugee Council, 2002.\n230 United Nations, March 2006.\n231 United Nations Humanitarian Coordinator, 27 April 2006.\n232 Norwegian Refugee Council, April 2002.\n233 Internal Displacement Monitoring Centre, July 2006.\n234 United Nations Assistance Mission for Iraq, June 2006 based on a 2001 UNHABITAT survey.\n\nMention: \"2001 UNHABITAT survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"The mission's estimates rely on this survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000352:24:0", "occurrence_id": "a49d10dc385966cf7e386b9e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: con algún tipo de discapacidad.\n\n\n**Atencion Diferencial por tipo de poblacion RENEC**\n\n**UDAPV 2009 - 2010**\n\n|TIPO DE POBLACIÓN|Col2|Col3|Col4|Col5|Col6|\n|---|---|---|---|---|---|\n|**AÑO**|**AFRO**|**INDÍGENA**|**CAMPESINO**|**OTROS**|**POBLACIÓN**<br>**ATENDIDA**|\n|2009|47,856|40,072|59,450|22,106|169,484|\n|% Participación|28%|24%|35%|13%|100%|\n|2010|50,013|18,078|42,287|45,834|156,212|\n|% Participación|32%|12%|27%|29%|100%|\n|TOTALES<br>2009 - 2010|97,869|58,150|101,737|67,940|325,696|\n|% Participación|30%|18%|31%|21%|100%|\n\n\n\n**Fuente: Estadísticas UDAPV con corte a diciembre 31 de 2010**\n\n\n5% superan los 60 años.\n\n\n27\n\nMention: \"Estadísticas UDAPV\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Las cifras de participación provienen de este registro con corte a la fecha.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000354:8:1", "occurrence_id": "cdc30bfc8f39cd3ce6f2e70a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: . Restriction of movement for women and girls may limit their\n\n**health services**\n\naccess to health services, while lack of female providers for\nreproductive health services, though improved is also a barrier. HAUS **52% refugees know they**\n2016 <sup>11</sup> [^11: Health Access and Utilization Survey, UNHCR 2016] have also shown that refugees have trouble accessing health **can be assisted through**\nservices when only 30% of those who need health services actively **UNHCR partner clinics if**\nsought services. **they can’t access**\n\n**government health services**\n\nRefugees continue to cite lack of **information on health services** as a\nmajor problem. HAUS Survey demonstrated that only 70% of\nrespondents in urban setting aware of subsidized access to public health services while 52% were\naware of free access through UNHCR supported services. Only 47% of urban refugees were able to\ndetect the nearest clinic.\n\n**Secondary and tertiary care** requires a continued high level of funding to ensure access to essential\ncare such as normal and assisted deliveries, caesarean sections, war injuries, congenital\nabnormalities including cardiac abnormalities and renal failure. Costly complex treatments such as\ncertain types of cancer cannot be supported with available resources necessitating difficult choices\nrelating to resource allocation. In particular, access to critical reproductive health services has been\nimpacted by the withdrawal of free services.\n**The MoH** ’s critical role in providing refugee health services needs to be recognized and supported.\nFacilities in areas hosting large numbers of refugees are often overburdened. HAUS survey\nrevealed an increase in percent of Syrians who sought care at MOH facilities in 2016 (28% in first\nfacility and 57% in second facility) compared to 2015 (23.9 in first facility and 42.9 in second\nfacility). This manifests in shortages of medications – especially those for chronic diseases – and\nbeds, overworked staff and short consultation times.\n\nMention: \"HAUS Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Shows only 70 percent of city dwellers know of subsidised access.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000357:1:0", "occurrence_id": "75969a5f0427957903c2e7a1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: UNHCR established a fixed presence in all three CAMRs\nin 2019. These centers offer the possibility of reaching\nthe highest number of returnees simultaneously,\noptimizing the identification of protection and response\ncases, since it is in these places that all people currently\narrive on their return. Since 2020, due to the COVID-19\npandemic, the attentions were mostly carried out by\ntelephone or with a sporadic presence. Since February\n2022, however, UNHCR re-established a permanent\npresence (Monday to Friday) of UNHCR staff in the three\nCAMRs, ensuring consistency in the identification and\nreferral of protection cases.\n\n\nThere is an effort by state institutions to establish a\ncomprehensive reception system for returnees in the\nCAMRs, which incorporates a perspective of protection\nin the various stages of reception. It involves both\ngovernment actors present in the centers, NGOs, and\nUN agencies. This extensive collaboration between\nstakeholders and partners is essential to ensure a\ncomprehensive and efficient response after\nidentification.\n\n\nSpecifically, the Belén Centre for Attention to Children\nand Migrant Families (CANFM) ensures differential care\nfor children and adolescents, including unaccompanied\nchildren.\n\n\nDynamics of returns\n\n\n#### **_88.555 Honduran returnees_** **_received in the country in 2022._**\n\n2022 is the second year with the highest number of\nreturns in the last eight years, reaching and\nexceeding pre-COVID levels.\n\n\nReturns to Honduras between 2020 and 2022 have\nmore than doubled, according to data both from the\nConsular and Migratory Observatory of Honduras\n(CONMIGHO) and from UNHCR observation based on its\nconstant presence in the CAMRs.\n\n\nBetween 2015 and 2022, Mexico was always the country\nthat returned people to Hondurans the most. However,\nwhile between 2021 and 2022 the number of returnees\nfrom Mexico increased only slightly, returns from the\nUnited States of America quadrupled. 2022 was also\nthe year with the highest number of returns from the\nUnited States of America (42,090)\n\nMention: \"Consular and Migratory Observatory of Honduras\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its data confirm the doubling of flows over the period.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000357:1:0", "occurrence_id": "0196c245aad9d074263ccd13", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: UNHCR established a fixed presence in all three CAMRs\nin 2019. These centers offer the possibility of reaching\nthe highest number of returnees simultaneously,\noptimizing the identification of protection and response\ncases, since it is in these places that all people currently\narrive on their return. Since 2020, due to the COVID-19\npandemic, the attentions were mostly carried out by\ntelephone or with a sporadic presence. Since February\n2022, however, UNHCR re-established a permanent\npresence (Monday to Friday) of UNHCR staff in the three\nCAMRs, ensuring consistency in the identification and\nreferral of protection cases.\n\n\nThere is an effort by state institutions to establish a\ncomprehensive reception system for returnees in the\nCAMRs, which incorporates a perspective of protection\nin the various stages of reception. It involves both\ngovernment actors present in the centers, NGOs, and\nUN agencies. This extensive collaboration between\nstakeholders and partners is essential to ensure a\ncomprehensive and efficient response after\nidentification.\n\n\nSpecifically, the Belén Centre for Attention to Children\nand Migrant Families (CANFM) ensures differential care\nfor children and adolescents, including unaccompanied\nchildren.\n\n\nDynamics of returns\n\n\n#### **_88.555 Honduran returnees_** **_received in the country in 2022._**\n\n2022 is the second year with the highest number of\nreturns in the last eight years, reaching and\nexceeding pre-COVID levels.\n\n\nReturns to Honduras between 2020 and 2022 have\nmore than doubled, according to data both from the\nConsular and Migratory Observatory of Honduras\n(CONMIGHO) and from UNHCR observation based on its\nconstant presence in the CAMRs.\n\n\nBetween 2015 and 2022, Mexico was always the country\nthat returned people to Hondurans the most. However,\nwhile between 2021 and 2022 the number of returnees\nfrom Mexico increased only slightly, returns from the\nUnited States of America quadrupled. 2022 was also\nthe year with the highest number of returns from the\nUnited States of America (42,090)\n\nMention: \"UNHCR observation\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Based on constant presence in reception centres.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000359:38:0", "occurrence_id": "986c04f4518ede95466a1bde", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Institutional Reform and Capacity Building Project Evaluations Unit. “Report on the IRCBP\n2007 National Public Services Survey: Public Services, Governance, Dispute Resolution,\nand Social Dynamics.” Draft for Circulation and Comment. 17 March 2008.\n\nInternational Rescue Committee. “Assessment of Protection Issues, with a Focus on\n\n\n\nAccess to Justice and the Rule of Law: Mae Le Camp, Tak Province and Sites\nOne and Two Camps, Mae Hong Son Province.” 2006.\n\nIRIN News. “Zambia: After the Angolan ‘Guests’ Have Gone.” 12 May 2005.\n\n\n\n<u>http://www.irinnews.org/PrintReport.aspx?ReportID=54370</u>\n\nIRIN News. “Zambia: Anti-corruption drive misses the point.” 30 January 2007.\n\n\n\n<u>http://www.irinnews.org/Report.aspx?ReportId=67716.</u>\n\nIRIN News. “Zambia: Landmark judgment for women in customary marriages.” 21\n\n\n\nDecember 2005. http://www.irinnews.org/PrintReport.aspx?ReportID=57641\n\nJackson, Cecile. “Feminism Spoken Here: Epistemologies for Interdisciplinary Development\n\n\n\nResearch.” _Development and Change_, Vol. 37, No. 3, 2006.\n\nJacobsen, Karen and Loren Landau. “The Dual Imperative in Refugee Research:\n\n\n\nSome Methodological and Ethical Considerations in Social Science Research\non Forced Migration.” University of Witwatersrand, Working Paper No. 19, 2003.\n\nJensen, Erik G. “Justice and the Rule of Law,” in Chales Call and Vanessa Wyeth\n\n\n\n(eds.), _Building States to Build Peace_, Lynne Reinner, 2007.\n\nJessop, B. _State Theory: Putting Capitalist States in Their Place_ . Oxford: Policy Press, 1990\n\nKagwanja, P.M\n\nMention: \"IRCBP\n2007 National Public Services Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its entries cover public services, governance and dispute resolution.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000363:2:1", "occurrence_id": "6a460b80d4cfa8828b647ef0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \nhydroelectric plant and the restriction of use in the other hydroelectric plants upstream of the Dnipro\nRiver may have a considerable impact on energy coverage during winter months. Even with full access\nto functioning heating systems, many families will find it difficult to pay for heating, due to the rising\ncost of living, loss of income and depletion of savings during displacement.\n\nIn addition to addressing heating issues, there is a need for essential household items such as warm\nwinter clothing, thermal underwear, blankets, and quilts to displaced individuals, returnees, and all\nthose affected by the conflict who have lost their belongings, lack access to markets, or cannot afford\nto purchase essential winter items. These provisions can be made either in-kind or through cash\nassistance specifically for winter preparations.\n\nThe 2023-2024 Winterisation plan builds on last year's experiences, lessons learned and stakeholder\nfeedback. These activities and the associated procurement chains will require considerable time and\nfinancial resources. The Government of Ukraine is taking the lead in preparing and implementing\nWinterisation initiatives, while humanitarian agencies will play a complementary role in this aspect as\nwell.\n\n\n\n[1 Ukraine Rapid Damage and Needs Assessment February 2023 (RDNA2)](https://documents1.worldbank.org/curated/en/099184503212328877/pdf/P1801740d1177f03c0ab180057556615497.pdf)\n\n\n\n2\n\nMention: \"Ukraine Rapid Damage and Needs Assessment\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its second edition quantifies damage and needs.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000364:3:1", "occurrence_id": "8434de38adbd0a62642b52d9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  men and boys.” <sup>6</sup> “We are appalled by reports of women and girls being sold at slave\nmarkets in areas controlled by RSF forces and other armed groups, including in North Darfur,” UN experts said. <sup>7</sup>\nThe experts also expressed concern about the increase in child, early and forced marriage, reportedly a result of\nfamily separation, and gender-based violence, including rape and unwanted pregnancies. <sup>8</sup>\n\n\n_3_ <u>[Situation Sudan situation (unhcr.org)](https://data.unhcr.org/en/situations/sudansituation)</u>\n\n_4_ _[Sudan: Disaster before us- GBV situational update as of 1 April 2024](Sudan:%20Disaster%20before%20us-%20GBV%20situational%20update%20as%20of%201%20April%202024)_\n\n_5 The_ Human _Rights Council on 11 October 2023, through resolution A/HRC/RES/54/2 decided to establish an independent international fact-finding mission for_\n_the Sudan,_ <u>[Independent International Fact-Finding Mission for the Sudan | OHCHR](https://www.ohchr.org/en/hr-bodies/hrc/ffm-sudan/index)</u>\n_6_ <u><mark>https://www.ohchr.org/en/hr-bodies/hrc/ffm-sudan/indexInvestigators: Disregard for human rights, law drives crisis in Sudan</mark></u>\n\n_7_ <u>[Sudan: Trafficking for sexual exploitation and recruitment of children\n\nMention: \"Situation Sudan situation\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The situation page assembles the regional figures.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000381:9:1", "occurrence_id": "796955188903db0bca6465b5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: United \nStates of America, France, Germany, Sweden, and Canada), only Sweden features among the top ten \nwhen compared against population size. But all five feature when compared against GDP (PPP) per capita. \nThis may be explained by the relatively large variation in national population size among the industrial-\nized countries. The top five destination countries have a significantly larger population size compared to \nCyprus, Malta, or Liechtenstein. In the case of the GDP (PPP) per capita ratio, the countries included can be \nconsidered as all having highly developed economies and the variation in GDP figures between them is not \nthat significant. With this important difference in mind, the GDP indicator might be more meaningful and \ninteresting when making comparisons among these 44 countries.\n15\tIn order to compare countries, the Purchasing Power Parity (PPP) GDP is considered (Source: International Monetary Fund, World Economic \nOutlook Database, May 2010). Source for national population: United Nations, Population Division, World Population Prospects: The 2008 \nRevision, New York, 2009.\n16\tFor the purposes of this type of analysis, the number of asylum applications registered over a five-year period has been considered to account \nfor some extreme fluctuations in the values.\n17\tAmong the 44 countries included in this report, the number of asylum-seekers per 1 USD GDP (PPP) per capita appears as rather modest when \ncompared to the ‘burden’ developing countries shoulder in receiving asylum-seekers. For instance, Ethiopia and South Africa, two major asylum-\nseeker receiving countries with 22,200 and 222,000 asylum applications respectively registered in 2009, received 26 and 23 asylum-seekers \nrespectively per 1 USD GDP (PPP) per capita during 2009 (Source: 2009 Statistical Yearbook, UNHCR, Geneva).\nCyprus\n24.2\n19.1\n15.2\n14.2\n11.1\n8.8\n7.8\n7.5\n6.8\n5.4\nMalta\nSweden\nLiechtenstein\nNorway\nSwitzerland\nAutria\nGreece\nBelgium\nLuxembourg\n\nMention: \"World Economic \nOutlook Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its purchasing power parity gross domestic product figures.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000381:9:1", "occurrence_id": "98616173fb7c0f0a8340ddb0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: United \nStates of America, France, Germany, Sweden, and Canada), only Sweden features among the top ten \nwhen compared against population size. But all five feature when compared against GDP (PPP) per capita. \nThis may be explained by the relatively large variation in national population size among the industrial-\nized countries. The top five destination countries have a significantly larger population size compared to \nCyprus, Malta, or Liechtenstein. In the case of the GDP (PPP) per capita ratio, the countries included can be \nconsidered as all having highly developed economies and the variation in GDP figures between them is not \nthat significant. With this important difference in mind, the GDP indicator might be more meaningful and \ninteresting when making comparisons among these 44 countries.\n15\tIn order to compare countries, the Purchasing Power Parity (PPP) GDP is considered (Source: International Monetary Fund, World Economic \nOutlook Database, May 2010). Source for national population: United Nations, Population Division, World Population Prospects: The 2008 \nRevision, New York, 2009.\n16\tFor the purposes of this type of analysis, the number of asylum applications registered over a five-year period has been considered to account \nfor some extreme fluctuations in the values.\n17\tAmong the 44 countries included in this report, the number of asylum-seekers per 1 USD GDP (PPP) per capita appears as rather modest when \ncompared to the ‘burden’ developing countries shoulder in receiving asylum-seekers. For instance, Ethiopia and South Africa, two major asylum-\nseeker receiving countries with 22,200 and 222,000 asylum applications respectively registered in 2009, received 26 and 23 asylum-seekers \nrespectively per 1 USD GDP (PPP) per capita during 2009 (Source: 2009 Statistical Yearbook, UNHCR, Geneva).\nCyprus\n24.2\n19.1\n15.2\n14.2\n11.1\n8.8\n7.8\n7.5\n6.8\n5.4\nMalta\nSweden\nLiechtenstein\nNorway\nSwitzerland\nAutria\nGreece\nBelgium\nLuxembourg\n\nMention: \"World Population Prospects\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The revision of the outlook gives the national populations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000385:1:0", "occurrence_id": "83355c2bad281ae02710f966", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The following trends analysis is put together on the basis\nof available secondary data at the time of publication. It is\nrepresentative of the available information and therefore\nindicative of mixed migratory trends in West Africa.\n\n\nThe Mixed Migration Centre (MMC) was established in February\n2018. It brings together various existing regional initiatives –\nhosted or led by the Danish Refugee Council (DRC) – engaged\nin data collection, research, analysis and policy development\non mixed migration issues into a new global network of mixed\nmigration expertise.\n\n\nThe Mixed Migration Centre - West Africa, provides quality\nmixed migration-related information for policy, programming\nand advocacy from a regional perspective. Our core countries of\nfocus are Burkina Faso, Mali and Niger. For updates on North\nAfrica please consult MHub Monthly Trend Bulletins at:\n<u>[http://www.mixedmigrationhub.org/monthly-trend-bulletins/](http://www.mixedmigrationhub.org/monthly-trend-bulletins/)</u>\n\n\nFor more information visit: <u>[www.mixedmigration.org](http://www.mixedmigration.org)</u>\n\n\nYou can contact us by email: <u>[west-africa@mixedmigration.org](mailto:west-africa@mixedmigration.org)</u>\n\n\n\nPhoto credit: DRC\n\nMention: \"MHub Monthly Trend Bulletins\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Monthly source of information on mixed migration movements.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000386:4:0", "occurrence_id": "cb366faa49688a6a46b9e5d0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **SOUTH SUDAN** | March 2025\n\n\n**Threats and Contributing Factors**\n\n\nThe root causes of GBV are deeply embedded in cultural and social norms that\nperpetuate gender inequality and power imbalances. These norms contribute to\npatriarchal structures that often position men as superior, thereby increasing the\nlikelihood of violence. Lasting economic decay perpetuates poverty and insecurity,\nheightening tensions within gender relations and leading to increased instances of\nsexual and physical abuse. Physical violence is the most reported type of GBV accounting\nfor 32% followed by sexual violence (rape and sexual assault) constituting 29% of all\nreported incidents of GBV in South Sudan (2024 GBVIMS).\n\n\nFurthermore, inter/intra-communal violence linked to border disputes, cross-border\nviolence, cyclical and retaliatory attacks, as well as ethnic polarization remained the\nprimary source of subnational violence, accounting for 83% of survivors of gender-based violence. The weak rule of law and\ninadequate access to justice further enable harmful practices, diminishing accountability for perpetrators. Concurrently, seasonal\nfloods as a consequence of climate shocks, contribute to the displacement of communities, further exposing vulnerable populations\nto GBV-related risks.\n\n\nFinally, limited access to education and information perpetuates some of the cultural narratives that undermines women's roles,\nwhile overcrowding inside the IDP camps and settlements further exaggerates the environment for gender-based violence. Longdistance travel for women and girls, driven by essential family responsibilities on fetching firewood and/or water outside of inhabited\nareas, exposes them to heightened risks of attacks, rape, and other physical abuses **,** particularly in areas lacking sufficient resources\nand social infrastructure.\n\n\n**Main Consequences**\nThe main effects of the GBV in South Sudan are profound and far-reaching, impacting individuals and communities alike. Survivors\noften experience severe psychological trauma, which can lead to emotional distress and, in extreme cases, suicide. The physical\nconsequences are equally alarming, with many suffering from injuries\n\nMention: \"GBVIMS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its records give the share of reported incidents.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000391:3:1", "occurrence_id": "0ced7bfe2d5d0282cb6c002c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: . UNHCR will continue to periodically survey refugees to understand\ntheir plans, concerns and needs while ensuring that these inform the planning and programming\nprocess for assistance.\n\n### **POSITION ON RETURN AND PROTECTION CONSIDERATIONS**\n\nAll refugees have the fundamental human right to return in safety and dignity to their country of origin\nat a time of their own choosing. This is codified in several international instruments, including the\n_Universal Declaration of Human Rights_ (Art. 13) and the _International Covenant on Civil and Political_\n_Rights_ (Art. 12). Refugee return should be based on a voluntary, free and informed decision. It should\nnot be coerced either overtly through forced return, or indirectly through changes of policies that\nrestrict refugee rights or by limiting assistance to refugees in the country of asylum which in effect\naffects refugees’ ability to take free decisions. <sup>8</sup> [^8: While this document focuses on an operational framework as related to _voluntary_ return of refugees, also recalled is the\nprinciple of _non-refoulement_, a rule of international customary law. The essence of the principle is that a State may not\noblige a person to return to a territory where s/he may be exposed to persecution.]\n\nAligned with the _CPSS_, this operational framework covers two phases of refugee return:\n\n- _Phase 1 - Self-organized/spontaneous return_ : During Phase 1 (current phase), refugees are\nreturning in relatively small numbers in a self-organized manner or in movements not organized\n\n5 See also: UNHCR, _<u>Comprehensive Protection and Solutions Strategy: Resettlement and Complementary Pathways</u>_ <u>(April</u>\n<u>2018).</u>\n~~6~~ UNHCR, _Fifth Refugee Perceptions and Intentions Survey (Nov. 2018 – Jan. 2019)_ . Publication forthcoming.\n7 _Ibid._\n8 While this document focuses on an operational framework as related to _voluntary_ return of refugees, also recalled is the\nprinciple of _non-refoulement_, a rule\n\nMention: \"Refugee Perceptions and Intentions Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its fifth round covers this field window.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000361:4:0", "occurrence_id": "7999cd7080e05e55e5998154", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <u>Regional Protection Working Group – West and Central Africa</u>\n\n\n**GBV prevention and response** services have been particularly affected across the region. The closure of safe\nspaces, One Stop Centers and mobile clinics, and the interruption of case management services, including\nmedical care and psychosocial support have left over 450,000 people in **Mali** (especially Menaka region) and\nmore than 12,000 people in **Burkina Faso** without support and exposed to negative coping mechanisms. In\n**Niger**, GBVIMS data show a 48% decrease in reported GBV cases during the first quarter of 2025 compared to\n2024 mostly due to the sudden interruption of identification and response services, specifically in the regions\nof Diffa, Maradi, Tillabéri and Tahoua. In **Chad**, where more than 15% of GBV actors had to interrupt or suspend\nGBV programmes due to funding cuts, GBVIMS data show that 59% of identified GBV survivors were unable to\naccess medical services in the first quarter of 2025. In the **Central African Republic**, only 21% of the 65\nlocalities identified as GBV intervention zones retained in the HNRP 2025 are covered by GBV services. In\nrefugee hosting areas like Birao, Sam-Ouandja, Bambouti, where two clinics closed, tens of thousands of\npersons, mainly women and girls, lost access to essential sexual and reproductive health and critical GBV\nsupport, including rape management and psychosocial services.\n\n\nThe impact on children is equally alarming. The disruption in **child protection** **services**, such as **case**\n**management, mental health and psychosocial support** as well as **parenting programs** exposes out-of-school\nchildren, unaccompanied and separated children and children who have experienced human right violations\nto\n\nMention: \"GBVIMS data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its records show a fall in reported cases in the first quarter.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000392:4:1", "occurrence_id": "72a5743277922737891e24a5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  cluster partner activities with transitional\nand cooperation frameworks, ensuring the SNFI cluster acts as a catalyst for reconstruction\nand durable solutions. Aligning humanitarian responses to the large-scale government\n(Resolution 381 <sup>15</sup> [^15: Legislative Update by UNHCR April 2023 - <u>[LINK](https://www.unhcr.org/ua/wp-content/uploads/sites/38/2023/06/MLU_Apr_2023_ENG.pdf)</u>] ) programme \"eRecovery” will be achieved through the Shelter\nInformation Damage Assessment & Response (SIDAR <sup>16</sup> [^16: Shelter Information Damage Assessment and Response Database - LINK] ) database as a key tool to improve\ncoordination, enhance decision-making and foster transparency and accountability and\nadditionally pave the way to let other stakeholders to get engaged <sup>17</sup> [^17: Please note that SIDAR is a tool under development and it is being updates and] . This objective is of\ncritical importance considering the state strategy for regional development outlines that\nthe scale of the destruction of residential buildings exceeds the volume of all new housing\nput into operation in the last seven years.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n11 Themes of Adequate standard of living– <u>[LINK](https://inform-durablesolutions-idp.org/library/adequate-standard-of-living/)</u>\n12 See the activity matrix on shelter cluster website - <u>[LINK](https://sheltercluster.org/response/ukraine)</u>\n13 Please note that Solid fuel and other energy-related activities in CO2 are specifically\ndesigned as acute humanitarian assistance to build resilience for cold winter days and are\nnot considered long-term solution activities. Long-term energy-related solutions fall well\nbeyond the mandate of the Shelter cluster.\n14 Annexed Document To Narrative Summary For: Humanitarian Needs And Response\nPlan 2024 Shelter Cluster Ukraine- <u>[LINK](https://sheltercluster.org/ukraine/documents/annexed-document-narrative-summary-humanitarian-needs-and-response-plan-2024)</u>\n\nMention: \"Shelter\nInformation Damage Assessment & Response\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its database supports the recovery programme.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000393:14:1", "occurrence_id": "6cda1146c60eeabdd65510cd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \nHeads of States and Governments, called for the full implementation of this principle by the\nSecurity Council as well as by the General Assembly. “The EU is ready to contribute and\ncommends other organizations or groups for taking initiatives to advance this principle into\na norm in international relations.” <sup>48</sup> [^48: EU Presidency Statement 14 January 2009; see http://www.europa-eu-un.org/articles/en/article_8416_en.htm] One could argue that PoC has a firm foundation in\ninternational humanitarian law, human rights law and refugee law, and has a “stand-alone”\nquality that does not require buttressing by other concepts such as R2P or the notion of\nhuman security.\n\n\n45 See the OCHA chronology on this theme:\n[http://ochaonline.un.org/HumanitarianIssues/ProtectionofCiviliansinArmedConflict/ChronologyofProtectiono](http://ochaonline.un.org/HumanitarianIssues/ProtectionofCiviliansinArmedConflict/ChronologyofProtectionofCivilians/tabid/1198/language/en-US/Default.aspx)\n[fCivilians/tabid/1198/language/en-US/Default.aspx](http://ochaonline.un.org/HumanitarianIssues/ProtectionofCiviliansinArmedConflict/ChronologyofProtectionofCivilians/tabid/1198/language/en-US/Default.aspx)\n\n46 Culture of Protection, p.6\n\n[47 See DPKO website: http://www.un.org/en/peacekeeping/](http://www.un.org/en/peacekeeping/)\n\n[48 EU Presidency Statement 14 January 2009; see http://www.europa-eu-un.org/articles/en/article_8416_en.htm](http://www.europa-eu-un.org/articles/en/\n\nMention: \"OCHA chronology\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"A thematic chronology on the protection of civilians.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000399:22:1", "occurrence_id": "fc6b28cb5c18b0f3074ec581", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Voice and Agency: Empowering_\n_Women and Girls for Shared Prosperity report,_\n_2014 (http://www.worldbank.org/content/dam/_\n_Worldbank/document/Gender/Voice_and_agency_\n__LOWRES.pdf)_\n\n\n_15. UNESCO Education for All Global Monitoring_\n_Report 2011_\n\n\n_16. UNHCR Population Profiling Verification and_\n_Response Survey, 2011 (http://unhcrpk.org/wp-_\n_content/uploads/2012/11/PPVR-Report.pdf)_\n\n\n_17. UNICEF Child Protection from violence, exploitation_\n_and abuse (http://www.unicef.org/protection/57929_\n__child_labour.html)_\n\n\n_18. World Bank Girls Education Brief, 2014_\n_(http://www.worldbank.org/en/topic/education/_\n_brief/girls-education)_\n\n\n_19. UNESCO Education Transforms Lives report, 2013_\n_(http://unesdoc.unesco.org/images/0022/002231/_\n_223115E.pdf)_\n\n\n_20. UNICEF Progress for Children Report Card, 2005_\n\n\n_21. More information on SSAR and relevant_\n_documentation is available at_\n_http://www.unhcr.org/pages/4f9016576.html_\n\n\n\n**23**\n\nMention: \"UNHCR Population Profiling Verification and_\n_Response Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Ses résultats portent sur le profilage et la vérification des populations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000418:5:0", "occurrence_id": "4a7693d7f94f08b40a7d9a39", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The sea route to Europe: Mediterranean passage in the age of refugees\n\n### 1. Europe’s growing maritime refugee crisis\n\n\nIn the first six months of 2015, 137,000 refugees\n\nand migrants arrived in Europe by sea, in profoundly\n\ndifficult and unsafe conditions. That compares to\n\n75,000 in the same period in 2014, marking an 83\n\nper cent increase over 2014. That number can be\n\nexpected to increase further in the second half of\n\nthe year, especially during the summer months of\n\nJuly, August and September. Arrivals in the second\n\nhalf of 2014, for example, were almost double those\n\nof the first half.\n\n\nIn the first half of 2015, 43,900 Syrians came to\n\nEurope’s shores – the single largest group by a\n\nconsiderable number, accounting for 34 per cent\n\nof all arrivals. This follows a similar trend in 2014,\n\nwhen 69,000 Syrians arrived by sea, 32 per cent of\n\n\n\nall arrivals.\n\n\nMost are likely to qualify as refugees, or receive\n\nsome other form of international protection. In 2014,\n\nthe 28 Member States of the EU gave 95 per cent of\n\nSyrians protection in the first instance, the highest\n\npercentage of any nationality, according to Eurostat.\n\n\nThe second and third highest countries of origin\n\nwere Eritrea and Afghanistan, accounting for\n\n12 per cent and 11 per cent of maritime arrivals\n\nrespectively. In 2014, the 28 EU countries gave 89\n\nper cent of asylum-seekers from Eritrea protection,\n\nand 63 per cent from Afghanistan. Arrivals from\n\nother top countries of origin, including Somalia, Iraq\n\nand Sudan, may also be in need of international\n\nprotection.\n\n\nMany have first fled for safety to neighbouring\n\n\n\nCoffins containing the bodies of refugees and migrants who perished when their boat sank off the coast of Lampedusa. UNHCR/Francesco Maltavolta\n\nMention: \"Eurostat\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its entries give the highest share by nationality.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000426:12:0", "occurrence_id": "02a12d0f1825c243afd0e890", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: POST DISTRIBUTION MONITORING APRIL – 2019\n\n\n**Chart 7:** Quantity of items received versus entitlement\n\n\nFor CRH and LPG packages distributed and monthly refills, refugees reported receiving their entitlement or\nmore. The monthly periodicity of the distribution and refilling may partially explain this result and led to a\npositive perception of quantity being met. However, during the FGDs conducted, LPG was the second most\nitem refugees complained on regarding insufficiency.\n\n\nUse of items\n\n\nAbout 99% of the surveyed refugees reported using all NFI items received while only 1% stored the items,\n0.3% sold them, with very few refugees reporting their items either stolen or exchanged. The overall utility\n\nrate is similar to March and September PDM 2018 results.\n\n\nOnly 1% of refugees reported storing their items for future use. Refugees reported they stored items from\nmost of the packages except the Shelter Kit (from which only sand-bag and toolkit were in some cases not\nkept). The items reported to be kept in other packages were the kitchen set from the CRI Kit, the iron-peg\n\n\nfrom the Pre-Monsoon Kit, the bathing soap from the Female Hygiene Kit and the disposable nappies from\nthe WASH Hygiene Kit (which had the highest number of respondents (17). About 2% of the respondents\nstated that they were storing the WASH Hygiene Kit (Table 1), while only 0.1% of the refugees reported that\nthey were keeping their LPG cylinder for future use.\n\n\nRefugees reported selling items only from the CRI and WASH Hygiene Kits. In the September 2018 PDM\nrefugees reported selling items from all assistance provided, except the WASH Hygiene Kit.\n\n\nThe most sold item was the kitchen set, part of the CRI Kit, reported by only 2 respondents. The rest of items\nreported to be sold were jerrycans and blankets, included in CRI Kit, and jerrycans and gamchas from the\nWASH Hygiene Kits. They were reportedly sold mainly to earn money and to buy\n\nMention: \"FGDs\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"They record complaints about insufficient quantities of a good.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000426:12:0", "occurrence_id": "5a5b6047b694adc195dedeb7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: POST DISTRIBUTION MONITORING APRIL – 2019\n\n\n**Chart 7:** Quantity of items received versus entitlement\n\n\nFor CRH and LPG packages distributed and monthly refills, refugees reported receiving their entitlement or\nmore. The monthly periodicity of the distribution and refilling may partially explain this result and led to a\npositive perception of quantity being met. However, during the FGDs conducted, LPG was the second most\nitem refugees complained on regarding insufficiency.\n\n\nUse of items\n\n\nAbout 99% of the surveyed refugees reported using all NFI items received while only 1% stored the items,\n0.3% sold them, with very few refugees reporting their items either stolen or exchanged. The overall utility\n\nrate is similar to March and September PDM 2018 results.\n\n\nOnly 1% of refugees reported storing their items for future use. Refugees reported they stored items from\nmost of the packages except the Shelter Kit (from which only sand-bag and toolkit were in some cases not\nkept). The items reported to be kept in other packages were the kitchen set from the CRI Kit, the iron-peg\n\n\nfrom the Pre-Monsoon Kit, the bathing soap from the Female Hygiene Kit and the disposable nappies from\nthe WASH Hygiene Kit (which had the highest number of respondents (17). About 2% of the respondents\nstated that they were storing the WASH Hygiene Kit (Table 1), while only 0.1% of the refugees reported that\nthey were keeping their LPG cylinder for future use.\n\n\nRefugees reported selling items only from the CRI and WASH Hygiene Kits. In the September 2018 PDM\nrefugees reported selling items from all assistance provided, except the WASH Hygiene Kit.\n\n\nThe most sold item was the kitchen set, part of the CRI Kit, reported by only 2 respondents. The rest of items\nreported to be sold were jerrycans and blankets, included in CRI Kit, and jerrycans and gamchas from the\nWASH Hygiene Kits. They were reportedly sold mainly to earn money and to buy\n\nMention: \"September 2018 PDM\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its results describe the resale of items received.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000429:29:0", "occurrence_id": "7f38090cb866b3dd8c137e95", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: DTM Mali – Juillet 2021\n\n\nGraphique 10 : Proportion des ménages PDI par types d’abris\n\n\n\n\n\n\n\nEn ce qui concerne la situation des abris des 105 sites\nde déplacement évalués en juillet 2021, plus d’un quart\ndes ménages déplacés (37%) vivaient dans des abris en\npaille ou en tôle. Une autre partie vivaient pendant\nl’évaluation dans des abris en bâche (20%) et des abris\nen dur (21%).\n\n\n\n\n\nGraphique 11 : Types d'abris des ménages déplacés par Cercles\n\n\n\n\n\n\n\n\n\nUne proportion importante de ménages déplacés\n(21%) étaient sans abris pendant l’évaluation du\nmois de juillet 2021. Parmi les ménages sans abris,\n**41 pour cent** vivaient dans les localités situées dans\nles cercles de **Djenne** (région de Mopti) et **39 pour**\n**cent à Goundam** (région de Tombouctou). Une\nautre partie non négligeable des ménages sans abris\navait été évaluée dans les sites de la région de Gao\n(13%).\n\n\nLa plupart des abris des ménages déplacés se\ntrouvaient dans des terrains vagues (40%). Une\nautre proportion des ménages déplacés se\ntrouvaient dans les maisons des familles d’accueil\n(35%). Contre 25 pour cent qui logeaient\nrespectivement dans les centres collectifs, maison\nindividuelle, bâtiment administratif, écoles, etc.\n\n\n\n\n\nPar ailleurs, dans les 105 lieux de déplacement\névalués au\n\nMention: \"DTM Mali\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Ses relevés décrivent les types d'abris des ménages déplacés.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000434:48:1", "occurrence_id": "e90796488eb0a28e75f0431a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  is not a term that has an official definition, but is employed in this report to refer to a\nperson who has been living outside his or her country of origin as a refugee for more than 20 years,\nincluding any children born in exile who hold derivative refugee status. The term is used to provide greater\nclarity on the populations of concern for this study, which focuses on those who have held refugee status\nfor much longer than the five-year period employed by UNHCR for statistical purposes in reporting on\n“protracted refugee situations” (see entry below).\n\n##### Protracted refugee situations\n\n\nThere is no definition of protracted refugee situations in international law. UNHCR’s Executive Committee\n(ExCom) adopted a conclusion in 2009 that defined a protracted refugee situation as a situation where\nrefugees have been in exile “for five or more years after their initial displacement, without immediate\nprospects for implementation of durable solutions”. <sup>107</sup> [^107: ‘Conclusion on Protracted Refugee Situations, No. 109 (LXI) – 2009’, UNHCR Executive Committee, 61st session,\nExtraordinary Meeting, 8 December 2009 available with other resources on protracted refugee situations collected\nat <u>[https://www.unhcr.org/research/eval/4a1d43986/protracted-refugee-situations.html. See also discussion in](https://www.unhcr.org/research/eval/4a1d43986/protracted-refugee-situations.html)</u>\nJames Milner, ‘Protracted Refugee Situations’, in _The Oxford Handbook of Refugee and Forced Migration Studies_,] For statistical purposes, UNHCR reports on\n\n\n104 See definition at: <u>[https://unstats.un.org/legal-identity-agenda/, based on Principles and Recommendations for a](https://unstats.un.org/legal-identity-agenda/)</u>\nVital Statistics System, Revision 3, UN Department of Economic and Social Affairs, 2014, para 279.\n<u>[https:/\n\nMention: \"Vital Statistics System\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its third-revision principles and recommendations are authoritative.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000417:6:1", "occurrence_id": "cae26de7517111fa67482cdd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  marquées par une faiblesse des services sociaux de base, bien que des\nefforts soient en cours par certains acteurs de protection, dont UNICEF, UNHCR, UNFPA, PAM et ONG. Suivant les projections\ndu cadre harmonisé mises à jour en août 2023 sur la situation alimentaire et nutritionnelle, 689147 personnes sont en phase\n3 et 5 d’insécurité alimentaire dans les trois provinces de Ouaddaï, Sila et Wadi Fira. La faiblesse voir le manque des revenus\ndes ménages contraint plusieurs individus à se rabattre sur le peu de ressources naturelles existantes pour combler leurs\nbesoins de base, avec comme corollaire la destruction de l’environnement, notamment la coupure du bois de chauffe à usage\ndomestique. Des cas d’agression physique relevant des cas de violences sur les femmes lors de la collecte de bois. Si des\nressources adéquates ne sont pas disponibles, pour répondre à ces besoins et gaps prioritaires, les risques liés aux\nconflits/tension intercommunautaires, exaspérés par les facteurs ci-dessous sont probants. Ajouté à cela les problèmes\nclassiques fonciers, d’habitat et de transhumance.\n\n\nL’étude sur l’impact de la crise soudanaise conduite par le PNUDviii en juillet 2023 relève un sentiment de discrimination à\nl'égard des communautés d'accueil. En effet, les réponses apportées à la crise par les acteurs humanitaires consistent\nessentiellement à porter secours aux réfugiés alors que les\n\nMention: \"projections\ndu cadre harmonisé\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Ses projections actualisées donnent les effectifs en insécurité alimentaire.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000437:1:0", "occurrence_id": "fe2084496eae78179f420369", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: GBV SAFETY AUDIT REPORT – CARIACO, PEMBA, MOZAMBIQUE\n\n###### Key Message\n\n\nLack of access to safe and inclusive livelihood options for women and girls is increasing\n\nmultiple risks of GBV including sexual exploitation, child marriage, denial of resources, and\n\nintimate partner violence. Displaced women and girls are compelled to resort to the sale\n\nand exchange of sex to meet their basic needs. Livelihoods programmes that engage\n\ndisplaced women and girls at risk in safe, dignified, and sustainable livelihoods options\n\nlinked to GBV response are urgently needed in settings across Cabo Delgado.\n\n\n_The report presents the main findings of the GBV Safety Audit conducted in the_\n\n_neighbourhood of Cariaco, City of Pemba, Cabo Delgado, Mozambique, in June 2022._\n\n\n_The report promotes the UNHCR Policy on The Prevention Of, Risk Mitigation, And_\n\n_Response to Gender-Based Violence of 2020._\n\n\n**Pemba Field Office**\n**Mozambique**\n\n\n**[COVER PHOTOGRAPH:]**\n_Girls discussing their perception of safety and security, Cariaco neighborhood, Pemba, Cabo Delgado. June 2022_\n\n\n2 UNHCR - CUAMM / June 2022\n\nMention: \"GBV Safety Audit\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its findings describe risks in one district of the city.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000446:3:1", "occurrence_id": "7e8791ed3c6a9c08138efcdd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  the Action Against\n\nHunger Bangladesh Surveillance Team conducted the standard expanded nutrition surveys (SENS) in the\n\nRohingya refugee camps in Cox’s Bazar, Bangladesh as a multi-sectoral exercise carried out on behalf of\n\nthe Nutrition Sector. The assessment was authorized by the Government of Bangladesh Ministry of Health\n\nand Family Welfare through National Nutrition Services (NNS), the Institute of Public Health Nutrition\n\n(IPHN), the Cox’s Bazar District Civil Surgeon’s, and the Office of the Refugee Relief and Repatriation\n\nCommissioner (RRRC) and supported by the Assessment and Information Management Technical Working\n\nGroup comprised of UN agencies (mainly UNHCR, WFP, and UNICEF) and implementing partners <sup>1</sup> working\n\nin the camps.\n\n\nThe assessment’s main objective was to assess the health and nutrition status of the Rohingya children\n\nand women living in camps in the Ukhiya and Teknaf Upazilas. It consisted of three population\nrepresentative, cross-sectional surveys using Standardized Expanded Nutrition Survey (SENS)\n\nmethodology. The first nutrition survey was conducted in the Kutupalong Mega Camp comprising of 32\n\n\n1 Working group members include Action against Hunger (ACF), Concern Worldwide (CWW), World Concern/Medair (WCM), Society for Health\nExtension and Development (SHED), Social Assistance and Rehabilitation for the Physically Vulnerable (SARPV), Care Bangladesh, and Save the\nChildren International (SCI).\n\n\n4\n\nMention: \"standard expanded nutrition surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"A multisectoral exercise in the refugee camps.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000449:21:0", "occurrence_id": "da6a4716899d3fa1c263c79c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: return. We use nearest neighbor propensity score matching methods to identify registered returnees in\nthe PPVR who are likely to resemble the registered returnees in the WBPS prior to their return (in 2011).\n\n\nAccess to basic services of good quality such as health care and schools can play an important role in\nreducing vulnerabilities of the poor. While in Pakistan, the vast majority (97 percent) of Afghan returnee\nhouseholds <sup>23</sup> [^23: Registered Afghan households in Pakistan who returned between 2014 and 2017.] lived within one-hour travel time from the nearest health facility and more than half (52%)\nlived within 15 minutes to the nearest health facility. Likewise, 97 percent of Afghan returnee households\nlived within one-hour walking distance from a primary school in Pakistan. The proximity of schools and\nhealth care may be explained by the fact that most Afghans in Pakistan lived in refugee villages (Section 3)\nwhere large groups of people live in small areas and hence facilities such as schools and health care will\ntypically be found within or close to the refugee villages. This is somewhat supported when comparing the\ntype of health facilities Afghan returnees lived closest to prior to and after returning Afghanistan. Prior to\ntheir return, most (45%) Afghan returnees lived closest to Afghan Refugee Health Centers and private\npractitioners and to a lesser extent to public health facilities which tend to be found in urban areas. In\ncontrast, after returning to Afghanistan most returnees lived in proximity to public health clinics and\nhospitals after which is consistent with many Afghans settling in urban areas such as Nangarhar and Kabul.\n\n\nSchool attendance is determined by a range of factors including accessibility of schools, quality of schools,\nwealth of households and social norms. School attendance rates among returnee children in Afghanistan\ndisplay large gender disparities. While 61 percent of households with boys sent all boys to school, this was\nthe case for only 37 percent of households with girls (Figure 9). However, school attendance rates were\nalso lower and displayed larger gender\n\nMention: \"PPVR\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its returnee records are used for matching.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000449:21:0", "occurrence_id": "24bb09427b99650d7a68de7e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: return. We use nearest neighbor propensity score matching methods to identify registered returnees in\nthe PPVR who are likely to resemble the registered returnees in the WBPS prior to their return (in 2011).\n\n\nAccess to basic services of good quality such as health care and schools can play an important role in\nreducing vulnerabilities of the poor. While in Pakistan, the vast majority (97 percent) of Afghan returnee\nhouseholds <sup>23</sup> [^23: Registered Afghan households in Pakistan who returned between 2014 and 2017.] lived within one-hour travel time from the nearest health facility and more than half (52%)\nlived within 15 minutes to the nearest health facility. Likewise, 97 percent of Afghan returnee households\nlived within one-hour walking distance from a primary school in Pakistan. The proximity of schools and\nhealth care may be explained by the fact that most Afghans in Pakistan lived in refugee villages (Section 3)\nwhere large groups of people live in small areas and hence facilities such as schools and health care will\ntypically be found within or close to the refugee villages. This is somewhat supported when comparing the\ntype of health facilities Afghan returnees lived closest to prior to and after returning Afghanistan. Prior to\ntheir return, most (45%) Afghan returnees lived closest to Afghan Refugee Health Centers and private\npractitioners and to a lesser extent to public health facilities which tend to be found in urban areas. In\ncontrast, after returning to Afghanistan most returnees lived in proximity to public health clinics and\nhospitals after which is consistent with many Afghans settling in urban areas such as Nangarhar and Kabul.\n\n\nSchool attendance is determined by a range of factors including accessibility of schools, quality of schools,\nwealth of households and social norms. School attendance rates among returnee children in Afghanistan\ndisplay large gender disparities. While 61 percent of households with boys sent all boys to school, this was\nthe case for only 37 percent of households with girls (Figure 9). However, school attendance rates were\nalso lower and displayed larger gender\n\nMention: \"WBPS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its pre-return respondents serve as the comparison group.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000458:2:2", "occurrence_id": "f9fd1cdb9ff8ea598df7a411", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  care services.\n\n\n**783,796** women and girls accessed\nsexual and reproductive health\nservices.\n\n\n\n**76,258** refugees and other people\nof concern provided with mental\nhealth and psychosocial support\nservices.\n\n\n**11** country operations reported all\ngeographic areas inhabited by people\nof concern were reached by COVID-19\ninformation campaigns.\n\n\n\n**88,028** children (6-59 months)\nadmitted for treatment of moderate\nacute malnutrition.\n\n\n**35,359** children (6-59 months)\nadmitted for treatment of severe acute\nmalnutrition.\n\n\n\n**323,770** children and youth\nsupported with distance/home-based\nlearning.\n\n\n**560,339** refugee children and\nyouth out of school due to mandatory\nschool closures.\n\n\n\n**82** UNHCR GLOBAL REPORT 2020 UNHCR GLOBAL REPORT 2020 **83**\n\nMention: \"UNHCR GLOBAL REPORT 2020\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Ses chiffres décrivent les enfants privés de scolarisation par les fermetures.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000475:63:1", "occurrence_id": "1f7821b76269ec0d18953d6b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: >237</sup> hardly surprising when global action to improve victim assistance makes little reference to\nthe specific needs of child victims, <sup>238</sup> [^238: Both the Maputo Action Plan (A plan from States Parties of that Anti-Personnel Mine Ban Convention to reaffirm their commitment to\nending suffering and casualties caused by anti-personnel mines) and the Bangkok Symposium on Landmine Victim Assistance: Enhancing a\nComprehensive and Sustainable Mine Action, 2015, lack any focus on the specific needs of children.] and the International Mine Action Standards address clearance,\nERW/mine risk education and stockpile destruction, but do not tackle the issue of victim assistance. <sup>239</sup> [^239: UNICEF. 2013. The State of the World’s Children 2013: Children with Disabilities] 2014\nand 2015 have seen the introduction of a number of tools to support strengthening assistance to child victims,\nthese include child-focused guidance for victim assistance from UNICEF, a technical paper submitted by the\ngovernments of Colombia and Austria to the Maputo Review Conference, and a Guide for Comprehensive\nassistance to boys, girls and adolescent landmine victims. <sup>240</sup> [^240: Landmine and Cluster Munition Monitor (2015) The Impact of Mines/ERW on Children, November 2015]\n\n\nAlthough age-sensitive assistance remained among the least measured and reported aspects of victim\nassistance, a small but increasing number of activities were reported by States Parties to the Mine Ban Treaty\nand Convention on Cluster Munitions, who also experience humanitarian crisis:\n\n\n  - In **Colombia**, a local victims’ association held workshops with children of victims to identify the issues\nthey are facing. Where age-sensitive assistance was available, most reported services were for child\nsurvivors rather than the children of people killed. In addition, the government developed a new\nscheme for payment or reimbursement of assistance services to mine/ERW survivors that was\nintroduced to cover children and adolescents.\n\n  - In **Eritrea** an integrated landmine/\n\nMention: \"Landmine and Cluster Munition Monitor\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its report describes the impact of mines on children.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000476:5:1", "occurrence_id": "eccb2ab095f950d867269d27", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  cash assistance and support_\n_for example to cover the costs of prothesis, wheelchairs…”_\n\n\n_Participant of a focus group discussion in Opatovská Nová Ves_ <sup>53</sup> [^53: UNHCR, IOM, WHO, UNICEF (2024), Slovakia: Multi-Sector Needs Assessment 2023 - Annex Report on Persons with Disabilities - June 2024, p.\n4.]\n\n\nThe MSNA indeed confirmed lower levels of employment in refugee households with a member with\ndisability compared to other refugee households. <sup>54</sup> [^54: UNHCR, IOM, WHO, UNICEF (2024), Slovakia: Multi-Sector Needs Assessment 2023 - Annex Report on Persons with Disabilities - June 2024.] Additionally, allowing their access to forms of\nsupport that are currently tied to the possession of a Slovak disability card (which Temporary Protection\nholders cannot obtain) —e.g. discounted transportation fares, access to public services at no or reduced\ncost, or special labor conditions— would also importantly enhance their protection and inclusion.\n\n\nAccording to the Protection Profiling and Monitoring data, 35% of the refugees surveyed between\nJanuary and March 2024 applied for some form of State-provided social protection benefits for\nvulnerable groups. Among those who applied, 90% did not report facing challenges with access. Among\nthe smaller portion (10%) who did, the language barrier was the most reported challenge (49%), followed\nby long waiting times (21%), being considered ineligible (13%), or lack of necessary documents (10%). <sup>55</sup> [^55: UNHCR (2024), Regional Protection Profiling and Monitoring: Protection Risks and Needs of Refugees from Ukraine.]\n\n\nUNHCR, drawing from its partners' counseling and support in numerous individual cases, noted\nchallenges faced by some refugees with disabilities in relation to their access to Disability Allowance for\nRefugees. The Slovak Office of the Commissioner for Persons with Disabilities highlighted several\nsystemic issues, including overly strict eligibility criteria\n\nMention: \"Protection Profiling and Monitoring data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its entries indicate requests for forms of status over the period.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000478:3:0", "occurrence_id": "74ded2646186466255737861", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: In 42 % of interviewed families, two children were attending school in displacement,\nfollowed by families with three kids attending schools (35 %) and four kids attending school\n(29 %). 16 % of families have one child at school, while in 10 % of families five children\nattend the school in displacement.\n\n\nFive families have one elderly in their families, and eight families have two. 11 families were\nhaving at the time of interview a pregnant or lactating women present in their family.\n\n\nMost of the interviewed IDPs currently live in Peshawar district (38 %), whole breakdown of\nlocations is provided below.\n\n\n_Graph 1: Current location of IDPs originating from Shalozan Tangi_\n\n\n**IV.** **Main findings of the Return Intention Survey**\n\n\n**A.** **Displacement timing and trends**\n\n\nOnly two of the interviewed persons came from areas of origin in less than 18 months,\nothers were in displacement for a longer period. For most of the interviewed persons (62 %)\nthe **reason for displacement** was sectarian violence, followed by reason of military\noperations (29 %) and lack of livelihood opportunities due to the conflict (3 %).\n\n\n4\n\nMention: \"Return Intention Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its results describe the timing and trends of displacement.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000441:13:0", "occurrence_id": "f46a8e0764af42465e5753d2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: A countrywide ILO assessment among 400 Syrian refugee households in March 2013 showed that 70% of\nSyrian refugees were employed of the economically active population (33% of total population). The ILO\nassessment shows as well that 11% of all surveyed refugees above 15 years of age were working in Lebanon\nbefore the crisis.\n\nDuring the November Winterisation baseline survey, 22% of 735 households assessed reported that they were\nable to generate an income. Three months later, in February 2014, this proportion increased to 32% of the\npopulation assessed during the post-distribution monitoring assessment. The respondents to this assessment\nwere all beneficiaries of the winterisation programme, and it is therefore likely that the sample is more\nvulnerable compared to the general population.\n\n\n_Mount Lebanon and Beirut_\n\nA small assessment in June by Amel in the southern areas of Beirut showed that over 30% of the 170\nparticipants were employed. This figure was significantly higher among male participants (40%) compared to\nfemale participants (9%). Young people were also found as disadvantaged, with the employment rate among\nparticipants aged 29 or younger at 25%. The assessment’s findings show that the proportion of men and\nwomen who were actively looking for work was similar at 40%, indicating that significantly lower levels of\nemployment among female participants is a result of inability to secure work rather than “inactivity” in the\nemployment market.\n\n**Type of work**\n\n\n<u>Syrian refugees registered, awaiting registration and unregistered</u>\n\nAll assessments available on the topic show that employment is mostly daily and temporary, in the\nconstruction, agriculture or service sectors:\n\n_National_\n\nMercy Corps’ assessment among 1,200 households in April/May 2013 showed that most of the Syrian\nrespondents employed were working as a professional, in daily labour or construction.\n\n\n14\n\nMention: \"November Winterisation baseline survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Of 735 households assessed, 22 percent report being able to generate income.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000441:13:0", "occurrence_id": "4da0442f353337a53ef7517a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: A countrywide ILO assessment among 400 Syrian refugee households in March 2013 showed that 70% of\nSyrian refugees were employed of the economically active population (33% of total population). The ILO\nassessment shows as well that 11% of all surveyed refugees above 15 years of age were working in Lebanon\nbefore the crisis.\n\nDuring the November Winterisation baseline survey, 22% of 735 households assessed reported that they were\nable to generate an income. Three months later, in February 2014, this proportion increased to 32% of the\npopulation assessed during the post-distribution monitoring assessment. The respondents to this assessment\nwere all beneficiaries of the winterisation programme, and it is therefore likely that the sample is more\nvulnerable compared to the general population.\n\n\n_Mount Lebanon and Beirut_\n\nA small assessment in June by Amel in the southern areas of Beirut showed that over 30% of the 170\nparticipants were employed. This figure was significantly higher among male participants (40%) compared to\nfemale participants (9%). Young people were also found as disadvantaged, with the employment rate among\nparticipants aged 29 or younger at 25%. The assessment’s findings show that the proportion of men and\nwomen who were actively looking for work was similar at 40%, indicating that significantly lower levels of\nemployment among female participants is a result of inability to secure work rather than “inactivity” in the\nemployment market.\n\n**Type of work**\n\n\n<u>Syrian refugees registered, awaiting registration and unregistered</u>\n\nAll assessments available on the topic show that employment is mostly daily and temporary, in the\nconstruction, agriculture or service sectors:\n\n_National_\n\nMercy Corps’ assessment among 1,200 households in April/May 2013 showed that most of the Syrian\nrespondents employed were working as a professional, in daily labour or construction.\n\n\n14\n\nMention: \"post-distribution monitoring assessment\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The proportion rises to 32 percent among programme beneficiaries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000455:27:1", "occurrence_id": "0fdb20bc76bc856aca2658bb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: os del Estado responsables\nde programas de protección y asistencia social, vivienda, créditos, empleo,\nformación vocacional, entre otros; deberán en sus respectivas planificaciones\nanuales, presupuestar progresivamente una cuota exclusiva para personas\ndesplazadas forzadamente, en armonía con los principios señalados en el Artículo\n4 de la presente Ley.\n\n\n**ARTÍCULO 26.- UNIDAD DE REGISTRO E INFORMACIÓN.**\nLa Dirección a través de Unidad de Registro e Información, tendrá a su cargo\nla administración del Registro Único de Personas Desplazadas Forzadamente\n**(RUPDEF)**, el cual llevará un registro de las personas, las unidades familiares y las\ncomunidades objeto de atención dentro del Sistema. Este registro permitirá tanto\nla identificación de las personas beneficiarias, como el seguimiento de la respuesta\nofrecida por el Estado y otras entidades que integran el sistema, bajo un irrestricto\nmanejo de confidencialidad.\n\n\n**ARTÍCULO 27.- OBLIGATORIEDAD DE LAS MEDIDAS.**\nLas medidas de asistencia humanitaria y protección que establezca la Secretaría\nde Estado en el Despacho de Derechos Humanos, mediante la Dirección para\nla Protección de Personas Desplazadas Internamente por la Violencia, son de\nobligatorio cumplimiento para órganos y entidades públicas involucradas.\n\n\n**ARTÍCULO 28.-** **UNIDADES MUNICIPALES DE ATENCIÓN Y PROTECCIÓ\n\nMention: \"Registro Único de Personas Desplazadas Forzadamente\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Administra el registro de personas y unidades familiares desplazadas.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000480:35:1", "occurrence_id": "1aa195f84e01ea3771874b4f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \nto the 1951 Convention is maintained.\nResettlement also has a strategic potential\n\n\n\n**34**\n\n\n\n**Resettlement to Western**\n**and Central Europe**\n\n\n**EU Resettlement Programme**\n\n\nUNHCR welcomes a number of resettlement\ninitiatives in Europe, including the\nestablishment of new programmes in four\ncountries, expansion of the existing ad hoc\narrangements in a number of countries and the\nproposal by the European Commission for an\nEU Resettlement Programme. UNHCR prepared\na background paper on the EU Resettlement\nProgramme affirming support for a scheme that\naims to support an increase in resettlement to\nthe EU, ensure a high standard of integration\nsupport for resettled refugees and facilitate\nthe administration of resettlement by UNHCR\nand the relevant state authorities. As the key\nstakeholder having a full overview on the global\nresettlement operations and needs, UNHCR\nstressed the importance of acknowledging and\nformalizing this unique role in the scheme.\n\n\n**Key developments in national**\n**resettlement programmes**\n\n\nSince 2008, UNHCR has advocated for and\nsupported an expansion of the resettlement\nbase in Europe. UNHCR welcomes the\nemergence of new resettlement countries in\nEurope such as the Czech Republic and Romania,\nand more recently in 2010, Spain and Bulgaria\nwho have announced the establishment of\nresettlement programmes from 2010 and 2011\nrespectively. These new resettlement countries\njoin the ten established resettlement countries:\nDenmark, Finland, France, Iceland, Ireland,\nNetherlands, Norway, Portugal, Sweden and\nthe United Kingdom.\n\n\nIn 2009 and 2010, a number of other\ncountries in Europe such as Belgium, Germany,\nLuxembourg and Switzerland initiated _ad hoc_\nresettlement programmes. Regrettably, the\nglobal financial crisis has negatively affected\nthe resettlement programmes in other\ncountries, particularly Iceland and Ireland, and\nmay also have impacted the potential for new\nprogrammes to develop.\n\n\n\nUNHCR Projected Global Resettlement Needs 2011\n\nMention: \"UNHCR Projected Global Resettlement Needs 2011\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its projections shape global resettlement needs.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000483:19:0", "occurrence_id": "e10c80559ac9c31094240486", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: xxxvii <u>[The Six Grave Violations – Office of the Special Representative of the Secretary-General for Children and Armed Conflict](https://childrenandarmedconflict.un.org/six-grave-violations/)</u>\nxxxviii <u>[Protection Cluster: Protection Monitoring Tool Findings 2023, January 2024](https://reliefweb.int/report/ukraine/protection-monitoring-findings-1-january-31-december-2023-enuk?_gl=1*1sshtsv*_gcl_aw*R0NMLjE3MjAwMTgxNzEuRUFJYUlRb2JDaE1JNlAtOV9vNkxod01WSDVSUUJoM2RnQUdvRUFBWUFTQUFFZ0lYZ3ZEX0J3RQ..*_ga*MTkzMzk5NjQ1MC4xNjg2OTM0Nzk4*_ga_E60ZNX2F68*MTcyMDM2NTE1Ni4yODcuMC4xNzIwMzY1MTU2LjYwLjAuMA..)</u>\nxxxix <u>[Guidance on humanitarian evacuations of civilians in Ukraine with special considerations for children, June 2024; IRC Protection Monitoring Report – June 2024](https://reliefweb.int/report/ukraine/guidance-humanitarian-evacuations-civilians-ukraine-special-considerations-children-enuk?_gl=1*1vqku94*_gcl_aw*R0NMLjE3MjE4MDg3OTMuRUFJYU\n\nMention: \"IRC Protection Monitoring Report\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Ses relevés de juin décrivent les situations des civils.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000484:21:1", "occurrence_id": "6027f65b00baca0260cf16ff", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Stateless Persons_ . http://www.unhcr.org/4c11f0be9.html. (Accessed 20.08.2010)\n\nUNRWA (2009) Numbers of Registered Palestinian Refugees. http://www.unrwa.org/userfiles/reg\nref%282%29.pdf. (Accessed 20.08.2010)\n\nUSCRI (2009) World Refugee Survey 2009: U.S. Committee for Refugees and Immigrants.\n\nhttp://www.refugees.org/article.aspx?id=2324&subm=179&area=About%20Refugees&.\n(Accessed 20.08.2010)\n\nWEILER-POLAK, D. (2010) The People who Decide who is a Refugee: 'Difficult on the Outside, Easy\n\non the Inside'. _Haaretz_, July 16. (Hebrew) http://www.haaretz.co.il/hasite/spages/1180034.html.\n\nWILLEN, S. S. (2007) _Transnational Migration to Israel in Global Comparative Context_ . Lanham, MD:\n\nLexington Books.\n\n——— (2010a) ―Citizens, ‗Real‘ Others, and ‗Other‘ Others: Governmentality, Biopolitics, and the\n\nDeportation of Undocumented Migrants from Tel Aviv. In _The Deportation Regime:_\n_Sovereignty, Space, and the Freedom of Movement._ DE GENOVA, N.andPEUTZ N.(eds.)\nDurham: Duke University Press.\n\n——— (2010b) Darfur through a Shoah Lens: Sudanese Asylum Seekers, Unruly Biopolitical Dramas,\n\nand the Politics of Humanitarian Compassion in Israel. In _A Reader in Medical Anthropology:_\n_Theoretical Trajectories, Emergent\n\nMention: \"World Refugee Survey 2009\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"A national compilation of refugee situations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000488:1:0", "occurrence_id": "54be8db8cd71771dc224c027", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ##### WHERE ARE ALL OF THE HUSBANDS?\n\n\n\n\n\nA myriad of reasons surround the absence of men and rise of female-headed\nhouseholds. Of the female-headed households surveyed by protection monitors, **28%**\nreport their husbands to be in military detention, **27%** state that their husbands were\n\nkilled by non-state armed groups and **16%** say they were separated from their husbands\n\nduring an attack by non-state armed groups. These three answers make up 71% of the\n\nreasons given for husbands being absent. For instance, the majority of FHH surveyed in\nBama (53%) report that their husbands had been detained, stating during a focus group\n\ndiscussion: “We arrived together with our husbands to the screening center in Bama.\n\n\n\nAfter being screened, they took our husbands to Giwa Barracks in Maiduguri.” In Dikwa,\nfemales heading their households report that the majority of their husbands had been killed by non-state armed groups\n\n(55%). In Fufore LGA, Adamawa State, most of the women interviewed report to have lost their husbands at the hands\nof violence perpetrated by non-state armed groups.\n\n\nOther reasons stated for the absence of men include that the women are divorced (11%, with highest numbers reported\n\nin Monguno and Ngala), that the husband died for non-violent reasons including due to health issues (8%) and that their\nhusbands were captured/forcefully recruited into non-state armed groups (4%, with the majority of cases reported in\n\nGwoza and Pulka). Smaller numbers of women report that their husbands were staying outside Maiduguri, that their\nhusbands had been killed, or that they are missing due to other reasons (including having claimed asylum in Cameroon\n\nor had been killed in a car accident).\n\n##### PROTECTION ISSUES AFFECTING FEMALE-HEADED HOUSEHOLDS\n\n\nIn patriarchal Northern\n\nMention: \"female-headed households surveyed by protection monitors\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its respondents report military detention of their husband.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000490:2:0", "occurrence_id": "be9b4ccc90e6fa3f2491db5d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: vpl.com.ua 3\n\n\nAccording to a notice issued by the Ministry of Social Policy, the UIDB has been working in test\nmode since 1 August 2016 <sup>5, 6</sup> . In early September 2016, the Minister of Social Policy stated that\nthe UIDB would be fully operational from 01 October 2016 <sup>7</sup> . However, at the end of October 2016,\nthe UIDB was not yet fully operational and still cannot be accessed by NGOs or IDPs.\n\n\nOn 22 September 2016, the Cabinet of Ministers of Ukraine approved Resolution No. 646, which,\namong other things, ensured that the UIDB would be accessible to voluntary, non-governmental\nand other organizations providing assistance to IDPs <sup>8</sup> . The Resolution also calls for the UIDB\nto include information on IDP needs and vulnerabilities (including specific disabilities, medical,\neducational and other needs).\n\n\nOn 05 October 2016, the Ministry of Social Policy recommended that Departments and Divisions\nof Social Protection for the Population (DSPP) employees who had problems with the UIDB use\nthe previous IDP accounting database until defects were removed from the UIDB.\n\n\nRepresentatives of the «Charitable Foundation «Right to Protection» (CF «Right to Protection»)\nrepeatedly received information about problems with many DSPPs’ work due to inadequate UIDB\nfunctionality.\n\n\n**In order to study the UIDB’s effectiveness and identify problems DSPP employees experienced**\n**with the UIDB, CF «Right to Protection» monitored program functionality and usage at DSPPs**\n**in Dnipropetrovsk, Donetsk, Zaporizhzhia, Luhansk and Kharkiv regions on 10-13 October**\n**2016. 76 DSPP employees were interviewed.**\n\n\nDSPP in Volnovaha\n\n\n[7] According to UKRINFORM (http:\n\nMention: \"IDP accounting database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Used while the defects of the new system are corrected.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000500:4:0", "occurrence_id": "7545976b1992108fe40e59ad", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 13. 9540 cas VBG **rapportés** (GBV IMS) soit une augmentation de 44% de personnes\nsurvivantes. La présence de groupes armés radicaux, fragilise d’avantage les\ncommunautés et augmentent les risques de mariage forcé car de plus en plus de famille\nsont obligés d’utiliser leurs filles comme moyens de paiement de la « zakat » imposer par\nles groupes armés.\n\n\n**Lutte anti-mine humanitaire**\n\n\n14. Il a été recensé autour de de 238 incidents causés par engins explosifs au total en 2021\ncontre 181 en 2020 ce qui se traduit par une augmentation des risques d’exposition des\npopulations civiles, des acteurs humanitaires et des forces armées.\n\n\n**Logement, Terre et Propriété**\n\n\n15. 76% des retournés, rapatriés et PDI qui sont sur leurs terres n’ont pas de documents pour\nassurer leur sécurité d’occupation.\n16. 24% des ménages se sentent en insécurité d’occupation des terres qui abritent leurs\nmaisons. La forme la plus précaire d’occupation a été observée dans la région de Menaka\noù 42,02% de ménages sont des squatteurs.\n17. Selon les données du monitoring du cluster de Protection au Mali notamment les\névaluations rapides de protection, les évaluations RRM et les rapports DTM environ\n259006 problèmes liés aux questions logement terre et propriété et ressources naturelles\nont pu être relevés dans les régions de Gao,\n\nMention: \"GBV IMS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Ses cas signalés montrent une augmentation des survivants pris en charge.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000502:2:0", "occurrence_id": "20824d54323ce7c115479c5e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Acronyms and Abbreviations**\n\n\n\n**AIDS**\n\n\n**ART**\n\n\n**ASRH**\n\n\n**ASRHR**\n\n\n**AGD**\n\n**ANC**\n\n**ANM**\n\n**BSC**\n\n**BFS**\n\n**CCSS**\n\n**CHWs**\n\n\n**CMAM**\n\n\n**CMR**\n\n**ECD**\n\n**FAO**\n\n**GAM**\n\n**GBV**\n\n**GCR**\n\n**HFUR**\n\n**HIV**\n\n**ILO**\n\n**IRS**\n\n**IPT**\n\n**IPV**\n\n\n**iRHIS**\n\n\n**ITN**\n\n**IYCF**\n\n\n**LGBTIQ+**\n\n\n**MAM**\n\n\n\nAcquired Immunodeficiency\nSyndrome\n\nAntiretroviral Therapy\n\nAdolescent Sexual and\nReproductive Health\n\nAdolescent Sexual and Reproductive\nHealth and Rights\n\nAge, Gender, and Diversity\n\nAntenatal Care\n\nAnaemia\n\nBalanced Score Card\n\nBaby-Friendly Spaces\n\nCosta Rican Social Protection Fund\n\nCommunity Health Workers\n\nCommunity Management of\nAcute Malnutrition\n\nClinical Management of Rape\n\nEarly Child Development\n\nFood and Agriculture Organization\n\nGlobal Acute Malnutrition\n\nGender Based Violence\n\nGlobal Compact on Refugees\n\nHealth Facility Utilisation Rate\n\nHuman Immunodeficiency Virus\n\nInternational Labour Organization\n\nIndoor Residual Spraying\n\nInterpersonal Therapy\n\nIntimate Partner Violence\n\nIntegrated Refugee Health\nInformation System\n\nInsecticide Treated Net\n\nInfant and Young Child Feeding\n\nLesbian, Gay, Bisexual, Transgender,\nIntersex and Queer\n\nModerate Acute Malnutrition\n\n\n\n**MAMI**\n\n\n**MC**\n\n\n**mhGAP**\n\n\n**MHPSS**\n\n\n**MoH**\n\n**NCDs**\n\n**PEP**\n\n**PHC**\n\n**PLHIV**\n\n**PLW**\n\n\n**PMTCT**\n\n\n**PNC**\n\n**\n\nMention: \"Integrated Refugee Health\nInformation System\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Le système apparaît dans la liste des abréviations du rapport.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000504:8:3", "occurrence_id": "0c9c54f1224ffc60affb0c77", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  de identificación de esquemas de incidencia idóneos para la articulación institucional\nlocal, regional y nacional. El GTP ha liderado sesiones grupales de información para el acceso a ruta de las víctimas de\ndesplazamiento y confinamiento, apoyo en la toma de declaraciones y censos para favorecer inclusión en registro único de\nvíctimas, apoyo en la recepción y caracterización de población desplazada para la entrega de asistencia humanitaria de inmediatez,\nactividades de atención psicosocial o fortalecimiento socioemocional a niños, niñas y adolescentes y actividades de prevención\nante violencias basadas en género, incluido prevención de PSEA. Asimismo, se ha brindado asistencia técnica a la Gobernación de\nNorte de Santander y alcaldías para cualificar los comités de justicia transicional y los subcomités de prevención, protección y\ngarantías de no repetición en el marco de la activación de planes de contingencia como escenarios idóneos para la respuesta a la\nemergencia.\n\n\nPágina 9\n\nMention: \"registro único de\nvíctimas\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Los censos apoyan la inclusión en este registro y la caracterización de la población.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000518:5:0", "occurrence_id": "2f52b31101efccfe26493af7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: REFUGEES AND MIGRANTS FROM SOUTH-WEST ASIA TO EUROPE\n\n## **POPULATION PROFILE**\n\n#### **2. Who are the refugees and migrants moving along the** **South-West Asia Route to Europe?**\n\n\n\nBased on data on arrivals provided by European country\nnational authorities in 2024, Afghan and Iranian\nnationals show a distinct demographic profile, with\nwomen making up 28 per cent of Afghan arrivals and 27\nper cent of Iranian arrivals, compared to 6 per cent of\nBangladeshis and 5 per cent of Pakistanis. <sup>11</sup> [^11: Based on data on irregular arrivals in Europe provided by national authorities of Cyprus, Greece, Italy, Spain, and Malta.] Children\naccount for 41 per cent of Afghan arrivals and 21 per\ncent of Iranians, while the proportions are notably lower\namong Bangladeshis (4 per cent) and Pakistanis (5 per\ncent), potentially highlighting greater protection and\nfamily-related needs among Afghans and Iranians.\n\n\n**<u>DEMOGRAPHIC PROFILE</u>**\n\n\nChildren Women Men\n\n\n\nAfghanistan\n\n\nIran\n\n\nOther\n\n\nPakistan\n\n\nBangladesh\n\n\n\n23%\n\n\n93%\n\n\n97%\n\n\n\n66%\n\n\n69%\n\n\n\n34%\n\n\n\n16%\n\n\n14%\n\n\n7%\n\n\n\n42%\n\n\n18%\n\n\n17%\n\n\n\nProtection monitoring interviews conducted between\nJanuary and December 2024 across South-East\nEuropean countries <sup>12</sup> [^12: UNHCR Regional Protection Monitoring System, Albania, Bosnia and Herzegovina, Bulgaria, Croatia, Kosovo (UNSCR 1244/1999), Montenegro, North Macedonia,\nRomania and Serbia.] reveal significant gender-based\ndisparities in socio-economic profiles among Afghans\nand Iranians on the move. Within the younger cohort\n(18-24 years), all Afghan women interviewed reported\nhaving no formal education, compared to only 10 per\ncent of men in the same age group. Conversely, among\nthose aged 25–59 years, 25 per cent of women had\nattained a university degree, compared to just 7 per\ncent of men. Despite these higher education levels, 75\nper cent of Afghan women reported never having\n\nMention: \"data on arrivals\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Supplied by the national authorities for the calendar year.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000518:5:0", "occurrence_id": "29e0f2c0b064f347365ef69b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: REFUGEES AND MIGRANTS FROM SOUTH-WEST ASIA TO EUROPE\n\n## **POPULATION PROFILE**\n\n#### **2. Who are the refugees and migrants moving along the** **South-West Asia Route to Europe?**\n\n\n\nBased on data on arrivals provided by European country\nnational authorities in 2024, Afghan and Iranian\nnationals show a distinct demographic profile, with\nwomen making up 28 per cent of Afghan arrivals and 27\nper cent of Iranian arrivals, compared to 6 per cent of\nBangladeshis and 5 per cent of Pakistanis. <sup>11</sup> [^11: Based on data on irregular arrivals in Europe provided by national authorities of Cyprus, Greece, Italy, Spain, and Malta.] Children\naccount for 41 per cent of Afghan arrivals and 21 per\ncent of Iranians, while the proportions are notably lower\namong Bangladeshis (4 per cent) and Pakistanis (5 per\ncent), potentially highlighting greater protection and\nfamily-related needs among Afghans and Iranians.\n\n\n**<u>DEMOGRAPHIC PROFILE</u>**\n\n\nChildren Women Men\n\n\n\nAfghanistan\n\n\nIran\n\n\nOther\n\n\nPakistan\n\n\nBangladesh\n\n\n\n23%\n\n\n93%\n\n\n97%\n\n\n\n66%\n\n\n69%\n\n\n\n34%\n\n\n\n16%\n\n\n14%\n\n\n7%\n\n\n\n42%\n\n\n18%\n\n\n17%\n\n\n\nProtection monitoring interviews conducted between\nJanuary and December 2024 across South-East\nEuropean countries <sup>12</sup> [^12: UNHCR Regional Protection Monitoring System, Albania, Bosnia and Herzegovina, Bulgaria, Croatia, Kosovo (UNSCR 1244/1999), Montenegro, North Macedonia,\nRomania and Serbia.] reveal significant gender-based\ndisparities in socio-economic profiles among Afghans\nand Iranians on the move. Within the younger cohort\n(18-24 years), all Afghan women interviewed reported\nhaving no formal education, compared to only 10 per\ncent of men in the same age group. Conversely, among\nthose aged 25–59 years, 25 per cent of women had\nattained a university degree, compared to just 7 per\ncent of men. Despite these higher education levels, 75\nper cent of Afghan women reported never having\n\nMention: \"UNHCR Regional Protection Monitoring System\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its entries cover several south-east European countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000526:13:1", "occurrence_id": "1eab2fb74173d8afcd104ce5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  delivery of humanitarian assistance. Elsewhere across the country intercommunal violence and rising criminality continue to disrupt humanitarian\nactivity. There are also reports that bureaucratic impediments on humanitarian actors are due to interference by local authorities on the ground.\nThere have also been regular demands for jobs by local youth groups in\nBentiu, Renk, Pibor and Melut, which at times have led to protests and\nviolence against humanitarian assets.\n\nIn Afghanistan, data collected by the Humanitarian Access Group (HAG)\nshows that the total number of access related incidents has increased by\n23 %, from 336 incidents in Q4 2020 to at least 436 incidents recorded\nduring Q1 2021. This includes incidents of violence and threats against\nthe humanitarian personnel as well as interference in humanitarian\nprogramming.\n\nWhile in some contexts restrictions on movement, previously in place to\nlimit the spread of Covid-19, have been lifted, in others contexts, movement restrictions remain. For example, in Colombia, certain parts of the\ncountry are only accessible to humanitarian missions considered as\nessential.\nWhile in some contexts restrictions on movement, previously in place to\nlimit the spread of Covid-19, have been lifted, in others contexts, movement restrictions remain. For example in Colombia, certain parts of the\ncountry are only accessible to humanitarian missions considered as\nessential.\n\nMention: \"data collected by the Humanitarian Access Group\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its entries show a rise in access incidents.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000529:1:0", "occurrence_id": "a0c5fe189d22fd4c73d80c3e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Based on the results of the first iteration, the project developed a\nstandardized information product that was then tested to\nunderstand perceptions (sentiment) of refugees and migrants\nfollowing the 2016 terrorist attacks in Nice, France (14 July),\nMunich, Germany (22 July), and Berlin, Germany (18 December).\nThis second iteration measured the volume of Tweets that either\nblamed, or defended refugees and migrants in four languages:\nEnglish, French, Greek and German.\n\n\nOverall, findings revealed a low number of tweets that blamed\nrefugees and migrants for the attacks 6% (Nice), 11% (Munich),\nand 5% (Berlin) respectively. In addition, a large number of tweets\nrelated to the terrorist attack in Berlin (7%) expressed explicit\nsupport for refugees and migrants, condemning racism and\nxenophobia.\n\n\n_Figure 2: Expression of xenophobic sentiment after the terrorist attacks in Nice and_\n_Munich_\n\n\nIt is important to note that while the percentage of posts blaming\nrefugees and migrants for the terrorist incidents in the three cities was\nlow, it is still representative of hundreds of thousands, to several\nmillions tweets: 0.2M (Nice), 6.4M(Munich), and 17.6M(Berlin).\n\n## **CONCLUSIONS**\n\n\nFindings demonstrated that data from social media, in this instance\nTwitter data, can be used to monitor protection issues and the safe\naccess to asylum of migrants and refugees in near real-time. This\ninformation was used and continues to be used by UNHCR to develop\ncommunication strategies to mitigate potential correlations between\nattacks and the arrival of migrants and refugees.\n\n\nWorking with social media requires a dynamic mindset, where a\nproject may need to adapt and change its initial scope, as\ndemonstrated above. The limitations of the first iteration and the\nlessons learned throughout the course of the project helped reshape\nits purpose.\n\n\nIn addition, while these results provide interesting insights into the way\npeople perceive issues\n\nMention: \"Twitter data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Drawn from social media, it allows protection issues to be tracked.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000539:17:2", "occurrence_id": "872cf6bfa0e3a39415e2f847", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Working Group analysis, June 2025.\n\n108 GBV Area of Responsibility trend analysis, December 2024 to May 2025.\n\n109 GBV Area of Responsibility trend analysis, December 2024 to May 2025.\n\n110 ACAPS, Impact of the conflict on mental health and psychosocial needs in Gaza, September 2024.\n\n111 [OPT Protection Cluster Protection Monitoring Snapshot, June 2025. Gaza Protection Cluster Snapshot: Protection Monitoring | 1 April – 10 June](https://reliefweb.int/report/occupied-palestinian-territory/gaza-protection-cluster-snapshot-protection-monitoring-1-april-10-june)\n\n112 Consultation with Gaza Community Mental Health Program, March 2025.\n\n113 [UNICEF, State of Palestine Humanitarian Situation and Needs, June 2024. 2024-HAC-State of Palestine-revised-June.pdf](https://www.unicef.org/media/158391/file/2024-HAC-State%20of%20Palestine-revised-June.pdf)\n\n114 Atfaluna Society for Deaf Children, Gaza Situation Report 2025, March 2025.\n\n115 Key in-field observations shared by HI during structured consultations, June 2025.\n\n116 Key in-field observations shared by HI during structured consultations, June 2025.\n\n117 Key observations shared through surveys with OPDs in Gaza by the Protection Cluster conducted in May and June 2025.\n\n118 [UNRWA, Protection Brief: Situation of Older Persons in Gaza, 24 June, 2025. Protection Brief: Situation of Older Persons in Gaza | UNRWA](https://www.unrwa.org/resources/reports/protection-brief-situation-older-persons-gaza)\n\n119 Key in-field observations shared by HI during structured consultations, June 2025.\n\n120 [HelpAge International, A lifetime of suffering: The challenges faced by older people in Gaza, 20 February 2024. A\n\nMention: \"OPT Protection Cluster Protection Monitoring Snapshot\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its snapshots cover the stated period for protection.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000549:10:0", "occurrence_id": "1dd560fb3daef505e1c1ba9c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **DEPARTAMENTO DE ARAUCA** | Agosto de 2024\n\n\n**Notas**\n\n\n[i Comité Internacional de la Cruz Roja (CICR), Colombia: Retos Humanitarios 2023.](https://www.icrc.org/es/document/colombia-retos-humanitarios-2023)\n\nii Según cifras de Migración Colombia, se tiene registro de más de 270.000 personas provenientes de Venezuela con vocación de\n[permanencia que se han pre-registrado para el Estatuto Temporal de Protección a corte de 30 de junio de 2023.](https://public.tableau.com/app/profile/migraci.n.colombia/viz/EstatutoTemporaldeProteccin-Prerregistros/Pre-registrosPublic)\n\n[iii Para mayor información: Informe MIRA – Comunidades de Suerera, Shubacbarina (Teorama) y Karikachaboquita (Tibú) – Norte de](https://assessments.hpc.tools/assessment/b807ad46-d1e3-415f-bdfc-d47e483a1932)\n<u>[Santander, Colombia, 20 de septiembre 2022 y Alertas Tempranas 050-20, 004-21 y 025-21 de la Defensoría del Pueblo.](https://assessments.hpc.tools/assessment/b807ad46-d1e3-415f-bdfc-d47e483a1932)</u>\n\niv Defensoría del Pueblo, Informe de Seguimiento No. 009-24 Alerta Temprana Estructural No. 011-23 para los muni\n\nMention: \"cifras de Migración Colombia\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Registran más de 270.000 personas con vocación de movilidad.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000559:2:0", "occurrence_id": "a7189b84d7d1c82cc97be7c4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <u>MOUVEMENTS MIXTES ></u> **<u>Tchad</u>** <u>/ JANVIER-JUIN 2024</u>\n\n## Analyse des données | janvier-juin 2024\n\n\nAu cours du premier semestre de l’année 2024, les relais communautaires ont\ncollecté via les téléphones, les données de 6.191 personnes en situation de\nmouvements mixtes sur l’ensemble du territoire national (Fig.1). Ainsi depuis mai\n2018, les données sur les mouvements mixtes collectées et analysées ont permis\nde toucher un cumul de 79.046 personnes.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n#### Profil des personnes interrogées\n\nLa grande majorité des personnes interrogées, sont des hommes adultes (86%) (Fig.2).\n\n\nFigure 2\n\n**Personnes interrogées par sexe et tranche d’âge**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nwww.unhcr.org 3\n\nMention: \"données de 6.191 personnes\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Collectées par téléphone sur l'ensemble du territoire national.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000561:21:1", "occurrence_id": "735d73b7f7c53ea1bd94737f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: tive\ncapacities. Settings such as North Wollo and West Hararghe have experienced famine and deep food\ninsecurity in recent decades, with deep-seated destitution and weakened livelihoods testament to a lack of\nresilience. Assessments of rural livelihoods and change, therefore, must consider structural vulnerability,\nrecurring shocks and persistent stress as defining characteristics of the baseline.\n\nAnalysis of quantitative panel survey data reinforces these findings. Poverty as measured by the US$1.90 a\nday poverty line is high among our sample in both regions and increased significantly over the five-year study\nperiod (see Figure 5.6). High levels of poverty reflect the large PSNP sample, yet we would expect poverty\nrates to have declined over time. That was not the case. Between the endline (January 2021) and follow-up\n(November 2022) the number of households below the poverty line in Oromiya increased, whereas the line\nflattened in Amhara.\n\n**Figure 5.6: Proportion of households in poverty at 2011 USD$1.90 PPP in Amhara and**\n**Oromiya (2018–2022/23)**\n\n\n\n1.0\n\n\n0.8\n\n\n0.6\n\n\n0.4\n\n\n0.2\n\n\n\n\n\n0.91\n\n0.85\n\n\n\n\n\n0.0\n\n|Amhara|Col2|0.83|Col4|Col5|\n|---|---|---|---|---|\n|Oromiya|||||\n|0.52|||||\n|0.33|||||\n||||||\n\n\n2018\n(Baseline)\n\n\n\n2019 2020 2021 2022 2023\n(Endline) (Follow-up)\n\nData Collection Point\n\n\n\n_Source: Authors’ own. Created by Camelia Vasilov using project data._\n_Note: Households with non-missing data: Amhara – baseline: 814; endline: 983; follow-up: 1\n\nMention: \"quantitative panel survey data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Strengthens the qualitative findings on poverty measured at the stated threshold.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000570:5:0", "occurrence_id": "df1aaa3995939e4941caba9b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: It is time to act together:\nDurable solutions for internally displaced persons in Colombia\n\n\nMicro-targeting\n\n\nA methodology has been developed to enable the\nmicro-targeting of intervention territories and the\ncomprehensive planning of service provision, adopting an\narea-based approach.\n\n\nThis methodology takes into account territorial and\nenvironmental factors as well as population aspects of the\ncommunities, including displaced populations and host\ncommunities.\n\n\nCommunity planning\n\n\nA community-based planning methodology has been adapted to\nthe Colombian context and made available to national and\nterritorial entities for implementation in the formulation of\ncommunity solutions plans.\n\n\nThis methodology ensures that both displaced populations and\nhost communities play a central role in the definition,\nprioritization, and implementation of actions.\n\n\n3. Measurement and data\n\n\n- The measurement will include a component for analyzing\npopulation conditions and contextual conditions to ensure\nthe guarantee of rights.\n\n\n- A proposal is being developed in coordination with the\nGovernment to establish a regionalization and prioritization\nmodel for the municipalities to be intervened in Colombia\nunder the durable solutions policy.\n\n\n- A partnership was established with the REACH/IMPACT\norganization to carry out a characterization and analysis\nexercise of the population in the cities of Bogotá and\nQuibdó.\n\n\n- A partnership was consolidated with the World Bank for\npotential funding to set up a working group for measuring\ndurable solutions within the National Statistical System, in\ncollaboration with DANE.\n\n\nThe measurement component is supported by JIPS (Joint\nInternal Displacement Profiling Service). In collaboration with this\norganization, the IDSF has developed a proposal for measuring\ndurable solutions, enabling the country to assess the progress of\ninternally displaced persons in this area.\n\n\n- The proposal is based on international standards and\nadvancements in national measurement instruments.\n\n\n- It has been discussed with public authorities at the national\nand territorial levels, experts in the field, and civil society.\n\n\n- With the approval of the National Government and the\nnecessary data management, the first calculation of this\nmeasurement is expected to take place in the first two\nmonths of 2025.\n\n\n\n4. Costing and financing\n\n\n-\n\nMention: \"National Statistical System\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"A working group on durable solutions would be established there.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000572:11:0", "occurrence_id": "cc1a9929a464af8f3797a564", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 3 ECONOMIC OPPORTUNITIES\n\n\nemployment in the formal sector, despite jobs assessments and surveys having been conducted (Jobs\nand Labour Market Assessment by Relief International in Maban in 2021 and by Danish Refugee Council\nin Jamjang in 2019 and the 2021 World Bank-funded Jobs, Recovery and Peacebuilding in Urban South\nSudan study). These assessments revealed that an overwhelming majority of refugee businesses take place\ninformally across all sectors, largely driven by the lack of a functional county system for registration, a series\nof barriers affecting the establishment of businesses, and a lack of access to start-up capital and technical\nvocational or business skills, attributed in part to the unstable political-socio-economic environment.\n\n\nThe business registration procedures require the business person to register with the Ministry of Justice in\norder to obtain an incorporation, followed by registration with the Chamber of Commerce, the Ministry of\nFinance (National Revenue Authority) and the Ministry of Investment for Operational Licence among other\noffices. The <u>[Registration of Business Names Act, 2008](https://ictpolicyafrica.org/en/document/vgortwpq1yb?page=6)</u> allows anyone to register a business in South Sudan\nwithout any specificity in terms of nationality or status. However, the Registrar is entitled, as per section\n5, subsection 2 of the Act, to obtain all the identity documents of the individuals intending to register a\nbusiness in their name.\n\n\nArticle 12 of the Labour Act on Minimum Working Age states that a child aged 12 years may be engaged to\nperform light work and that a child under the age of 14 years cannot perform any work defined as the worst\nforms of work, which include all forms of slavery or practices similar to slavery, such as sale and trafficking of\nchildren, debt bondage and serfdom, forced or compulsory labour, and forced or compulsory recruitment of\nchildren for use in armed conflict; the use, procurement or offer\n\nMention: \"Jobs\nand Labour Market Assessment\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its assessments did not increase participation in the formal sector.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000576:24:2", "occurrence_id": "ad7cd6cba60d04d892fd40b2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: /mark>\n<mark>implications thereof, as well as the level of information about the situation in Libya which leads to re-routing, require</mark>\n<mark>further investigation.</mark>\n\n\n**Focus 2: Re-routing to Spain from Libya via Morocco**\n\n\n[69 REACH, Tunisia, country of destination and transit for sub-Saharan African migrants, October 2018.](http://bit.ly/2RqxYwV)\n[70 UNHCR, Mediterranean Situation Dashboard, accessed 13 January 2019.](https://data2.unhcr.org/en/situations/mediterranean)\n71 Primary countries of origin of individuals who arrived primarily in Italy in 2017 and who, in 2018, predominantly arrived in Spain include,\namong others, Guinean, Ivorian and Malian nationals. For further information, please consult: UNHCR, <u>[Mediterranean Situation Dashboard,](https://data2.unhcr.org/en/situations/mediterranean)</u>\naccessed 13 January 2019; <u>Brenner, Y., Forin, R., Frouws, B.,</u> <u>[The “Shift” to the Western Mediterranean Migration Route: Myth or Reality?,](http://www.mixedmigration.org/articles/shift-to-the-western-mediterranean-migration-route/)</u>\n22 August 2018, accessed 13 January 2019.\n72 IOM DTM, <u>[Profile and reported vulnerabilities of migrants along the western Mediterranean route, March 2019.](https://reliefweb.int/report/spain/profile-and-reported-vulnerabilities-migrants-along-western-mediterranean-route)</\n\nMention: \"Mediterranean Situation Dashboard\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Dashboard consulted on that date for the Mediterranean situation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000579:9:0", "occurrence_id": "efa5d1e592cb731fec6843b4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ##### **EMPLOYERS AND WORK PERMITS**\n\n###### The survey confirms the reluctance of employers in obtaining work permits for their Syrian refugees’ workers. This reluctance was also indicated in the 2017 periodic analysis.\n\n75% of Syrian refugee workers state that their employers are not willing to proceed in issuing\nthem work permits\n\n\n85% of the respondents made it clear that they would want a WP if the employers were willing\nto do it for them\n\n\n15% do not wish to have a work permit due to fear of commitment to an employer, or due to\ntheir unfounded fear of losing the opportunity to resettle in a third country.\n\n\nHalf of the interviewees seemed to be lacking knowledge of WP benefits and formal sector\nprivileges.\n\nMention: \"2017 periodic analysis\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Indicates employer reluctance to issue work permits.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000568:8:0", "occurrence_id": "2849199e6dd8f7046a8758dc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Les motifs de retour sont principalement l’accès à la terre cultivable, ainsi que la sécurisation de la zone de\nretour. <sup>24</sup> 5% des enfants (dont près de la moitié sont des filles) venant de foyers de retournés analysés par le\nRRM sont des orphelins, dont la plupart ont perdu leurs parents pendant les conflits d’août et décembre 2021. <sup>25</sup>\n\n\n24 IOM, Matrice de suivi des déplacements – République du Cameroun – Région de l’Extrême-Nord – Round 24, février 2022.\n25 IOM, Population Movement Analysis on Logone Birni, Mai 2022.\n\n\n9\n\nMention: \"Matrice de suivi des déplacements\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Son vingt-quatrième tour couvre la région indiquée.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000568:8:0", "occurrence_id": "e3dfed84197640859138ca32", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Les motifs de retour sont principalement l’accès à la terre cultivable, ainsi que la sécurisation de la zone de\nretour. <sup>24</sup> 5% des enfants (dont près de la moitié sont des filles) venant de foyers de retournés analysés par le\nRRM sont des orphelins, dont la plupart ont perdu leurs parents pendant les conflits d’août et décembre 2021. <sup>25</sup>\n\n\n24 IOM, Matrice de suivi des déplacements – République du Cameroun – Région de l’Extrême-Nord – Round 24, février 2022.\n25 IOM, Population Movement Analysis on Logone Birni, Mai 2022.\n\n\n9\n\nMention: \"Population Movement Analysis on Logone Birni\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Analyse locale des mouvements de population.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000593:4:6", "occurrence_id": "1994df6a1d8831289f4bde19", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: )</u>\n[33 Міністерство освіти Словацької Республіки (2022), Najčastejšie otázky a odpovede v súvislosti so situáciou na Ukrajine z pohľadu školstva.](https://ukrajina.minedu.sk/data/att/22984.pdf)\n[34 Державна шкільна інспекція (2023), Stav vzdelávania žiakov – odídencov z Ukrajiny: 1. časť – priebežná správa.](https://www.ssi.sk/wp-content/uploads/2023/05/vzdelavanie_ziakov_odidencov_z_Ukrajiny.pdf)\n35 УВКБ ООН (2023), Analysis of Focus Group Discussions with Refugees in Slovakia: червень 2022 – червень 2023 [Документ на стадії\nпідготовки].\n[36 Державна шкільна інспекція (2023), Stav vzdelávania žiakov – odídencov z Ukrajiny: 1. časť – priebežná správa.](https://www.ssi.sk/wp-content/uploads/2023/05/vzdelavanie_ziakov_odidencov_z_Ukrajiny.pdf)\n37 УВКБ ООН, REACH (2023), <u>[Area Based Assessment Slovakia: February 2023;](https://data.\n\nMention: \"Area Based Assessment Slovakia\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Son évaluation par zone couvre ce mois.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000594:4:0", "occurrence_id": "038833809fbc387cdbafe11a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Context\n\n\nthe sample for the Kakuma 2019 SES. <sup>14</sup> The Kalobeyei SES and Kakuma SES were done via Computer-Assisted\n\n\nPersonal Interviews (CAPI). Due to COVID-19 social distancing measures, the Urban SES was collected via Computer\nAssisted Telephone Interviewing (CATI). The Kalobeyei SES covers 6004 households; the Kakuma SES covers 2,127\nhouseholds; and the Urban SES covers 2,438 households in Nairobi, Nakuru and Mombasa.\n**Questionnaires are aligned with national household survey instruments, while additional modules are added to**\n**explore refugee-specific dynamics.** The SES include modules on demographics, household characteristics, assets,\nemployment, education, consumption, and expenditure, which are aligned with Kenya Integrated Household\nBudget (KIHBS) Survey 2015/16 and the recent Kenya Continuous Household Survey (KCHS) 2019. Additional\nmodules on access to services, vulnerabilities, social cohesion, coping mechanisms to lack of food, displacement\n\ntrajectories and durable solutions are administered to capture refugee-specific challenges. <sup>15</sup>\n\n\n_Box A-2: Limitations_\n\n\n**The mode of data collection limits comparability between urban and camp-based refugees.** As the Urban SES\nwas collected through CATI, the representativeness of the sample and external validity can be limited due to\n\ntelephone coverage, low participation, and response rates. <sup>16</sup> These limitations are a possible source of bias, which\n\n\ncan be partially addressed by adjusting the survey weights using information from the population data. While the\nsampling weights for the Urban SES accounts for differences that might exist between households that have\nphones and all households, it does not account for differences in responses that may arise as a result of collecting\ndata through CATI and CAPI. In addition, the training of enumerators and fieldworks\n\nMention: \"Kakuma 2019 SES\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its sample provides the base for the analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000594:4:0", "occurrence_id": "e342b6f8bb4fc1631709b54e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Context\n\n\nthe sample for the Kakuma 2019 SES. <sup>14</sup> The Kalobeyei SES and Kakuma SES were done via Computer-Assisted\n\n\nPersonal Interviews (CAPI). Due to COVID-19 social distancing measures, the Urban SES was collected via Computer\nAssisted Telephone Interviewing (CATI). The Kalobeyei SES covers 6004 households; the Kakuma SES covers 2,127\nhouseholds; and the Urban SES covers 2,438 households in Nairobi, Nakuru and Mombasa.\n**Questionnaires are aligned with national household survey instruments, while additional modules are added to**\n**explore refugee-specific dynamics.** The SES include modules on demographics, household characteristics, assets,\nemployment, education, consumption, and expenditure, which are aligned with Kenya Integrated Household\nBudget (KIHBS) Survey 2015/16 and the recent Kenya Continuous Household Survey (KCHS) 2019. Additional\nmodules on access to services, vulnerabilities, social cohesion, coping mechanisms to lack of food, displacement\n\ntrajectories and durable solutions are administered to capture refugee-specific challenges. <sup>15</sup>\n\n\n_Box A-2: Limitations_\n\n\n**The mode of data collection limits comparability between urban and camp-based refugees.** As the Urban SES\nwas collected through CATI, the representativeness of the sample and external validity can be limited due to\n\ntelephone coverage, low participation, and response rates. <sup>16</sup> These limitations are a possible source of bias, which\n\n\ncan be partially addressed by adjusting the survey weights using information from the population data. While the\nsampling weights for the Urban SES accounts for differences that might exist between households that have\nphones and all households, it does not account for differences in responses that may arise as a result of collecting\ndata through CATI and CAPI. In addition, the training of enumerators and fieldworks\n\nMention: \"Kalobeyei SES\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Collected through computer-assisted interviews.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000594:4:0", "occurrence_id": "6a559552513016e0fb196036", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Context\n\n\nthe sample for the Kakuma 2019 SES. <sup>14</sup> The Kalobeyei SES and Kakuma SES were done via Computer-Assisted\n\n\nPersonal Interviews (CAPI). Due to COVID-19 social distancing measures, the Urban SES was collected via Computer\nAssisted Telephone Interviewing (CATI). The Kalobeyei SES covers 6004 households; the Kakuma SES covers 2,127\nhouseholds; and the Urban SES covers 2,438 households in Nairobi, Nakuru and Mombasa.\n**Questionnaires are aligned with national household survey instruments, while additional modules are added to**\n**explore refugee-specific dynamics.** The SES include modules on demographics, household characteristics, assets,\nemployment, education, consumption, and expenditure, which are aligned with Kenya Integrated Household\nBudget (KIHBS) Survey 2015/16 and the recent Kenya Continuous Household Survey (KCHS) 2019. Additional\nmodules on access to services, vulnerabilities, social cohesion, coping mechanisms to lack of food, displacement\n\ntrajectories and durable solutions are administered to capture refugee-specific challenges. <sup>15</sup>\n\n\n_Box A-2: Limitations_\n\n\n**The mode of data collection limits comparability between urban and camp-based refugees.** As the Urban SES\nwas collected through CATI, the representativeness of the sample and external validity can be limited due to\n\ntelephone coverage, low participation, and response rates. <sup>16</sup> These limitations are a possible source of bias, which\n\n\ncan be partially addressed by adjusting the survey weights using information from the population data. While the\nsampling weights for the Urban SES accounts for differences that might exist between households that have\nphones and all households, it does not account for differences in responses that may arise as a result of collecting\ndata through CATI and CAPI. In addition, the training of enumerators and fieldworks\n\nMention: \"Urban SES\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Collected by telephone because of distancing measures.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000595:24:1", "occurrence_id": "bf6dd2b18e3be87030727849", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: -Poverty-in-Contexts-of-Forced-Displacement](http://documents.worldbank.org/curated/en/492181635479693932/A-Multi-Country-Analysis-of-Multidimensional-Poverty-in-Contexts-of-Forced-Displacement)</u>_\n\nThis paper **develops a Multidimensional Poverty Index (MPI) to examine patterns**\n\n**of multidimensional poverty among IDPs and refugees, with comparisons to**\n\n**host populations, in five African countries** . The MPI is disaggregated to analyze\n\nvariations in deprivation by displacement status of the household and gender of the\n\nhousehold head. The analysis draws on household survey data from Ethiopia, Nigeria,\n\nSomalia, South Sudan, and Sudan <sup>11</sup> [^11: In Ethiopia, the Skills Profile Survey (2017) sampled refugees in and around camps in the Tigray, Afar, Gambella,\nBenishangul Gumuz, and Somali regions. In Nigeria, the IDP Survey (2018) sampled IDPs and host communities in six\nnortheastern states (Adamawa, Bauchi, Borno, Gombe, Taraba, and Yobe). In Somalia, the High Frequency Survey (2017)\nsampled IDPs and host communities in secure parts of the country. In Sudan, the IDP Profiling Survey (2018) sampled IDPs\nand neighboring host communities in the Abu Shouk and El Salam camps, in Al-Fashir. And in South Sudan, the High\nFrequency Survey Wave 4 (2017) sampled IDPs and host communities in urban areas of seven of the ten pre-war states\n(Western Equatoria, Central Equatoria, Eastern Equatoria, Northern Bahr-El-Ghazl, Western Bahr-El-Ghazal, Warrap, Lakes\nstate).] .\n\n\n[10 This work is part of the program “Building the Evidence on Forced Displacement: A Multi-Stakeholder Partnership”. The](https://www.worldbank.org/en/programs/building-the-evidence-on-forced-displacement)\nprogram is funded by UK aid from the United Kingdom's Foreign, Commonwealth, and Development Office (FCDO). It is\nmanaged by the World Bank Group (WBG) and was established in partnership with the United Nations High Commissioner for\nRefugees (UNHCR). This\n\nMention: \"household survey data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Five countries covered for activity status and household head.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000595:24:1", "occurrence_id": "68f6a97c1bdefe40a305974b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: -Poverty-in-Contexts-of-Forced-Displacement](http://documents.worldbank.org/curated/en/492181635479693932/A-Multi-Country-Analysis-of-Multidimensional-Poverty-in-Contexts-of-Forced-Displacement)</u>_\n\nThis paper **develops a Multidimensional Poverty Index (MPI) to examine patterns**\n\n**of multidimensional poverty among IDPs and refugees, with comparisons to**\n\n**host populations, in five African countries** . The MPI is disaggregated to analyze\n\nvariations in deprivation by displacement status of the household and gender of the\n\nhousehold head. The analysis draws on household survey data from Ethiopia, Nigeria,\n\nSomalia, South Sudan, and Sudan <sup>11</sup> [^11: In Ethiopia, the Skills Profile Survey (2017) sampled refugees in and around camps in the Tigray, Afar, Gambella,\nBenishangul Gumuz, and Somali regions. In Nigeria, the IDP Survey (2018) sampled IDPs and host communities in six\nnortheastern states (Adamawa, Bauchi, Borno, Gombe, Taraba, and Yobe). In Somalia, the High Frequency Survey (2017)\nsampled IDPs and host communities in secure parts of the country. In Sudan, the IDP Profiling Survey (2018) sampled IDPs\nand neighboring host communities in the Abu Shouk and El Salam camps, in Al-Fashir. And in South Sudan, the High\nFrequency Survey Wave 4 (2017) sampled IDPs and host communities in urban areas of seven of the ten pre-war states\n(Western Equatoria, Central Equatoria, Eastern Equatoria, Northern Bahr-El-Ghazl, Western Bahr-El-Ghazal, Warrap, Lakes\nstate).] .\n\n\n[10 This work is part of the program “Building the Evidence on Forced Displacement: A Multi-Stakeholder Partnership”. The](https://www.worldbank.org/en/programs/building-the-evidence-on-forced-displacement)\nprogram is funded by UK aid from the United Kingdom's Foreign, Commonwealth, and Development Office (FCDO). It is\nmanaged by the World Bank Group (WBG) and was established in partnership with the United Nations High Commissioner for\nRefugees (UNHCR). This\n\nMention: \"Skills Profile Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its samples cover the regions around the camps.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000595:24:1", "occurrence_id": "a06d103647204ee117ad9b8e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: -Poverty-in-Contexts-of-Forced-Displacement](http://documents.worldbank.org/curated/en/492181635479693932/A-Multi-Country-Analysis-of-Multidimensional-Poverty-in-Contexts-of-Forced-Displacement)</u>_\n\nThis paper **develops a Multidimensional Poverty Index (MPI) to examine patterns**\n\n**of multidimensional poverty among IDPs and refugees, with comparisons to**\n\n**host populations, in five African countries** . The MPI is disaggregated to analyze\n\nvariations in deprivation by displacement status of the household and gender of the\n\nhousehold head. The analysis draws on household survey data from Ethiopia, Nigeria,\n\nSomalia, South Sudan, and Sudan <sup>11</sup> [^11: In Ethiopia, the Skills Profile Survey (2017) sampled refugees in and around camps in the Tigray, Afar, Gambella,\nBenishangul Gumuz, and Somali regions. In Nigeria, the IDP Survey (2018) sampled IDPs and host communities in six\nnortheastern states (Adamawa, Bauchi, Borno, Gombe, Taraba, and Yobe). In Somalia, the High Frequency Survey (2017)\nsampled IDPs and host communities in secure parts of the country. In Sudan, the IDP Profiling Survey (2018) sampled IDPs\nand neighboring host communities in the Abu Shouk and El Salam camps, in Al-Fashir. And in South Sudan, the High\nFrequency Survey Wave 4 (2017) sampled IDPs and host communities in urban areas of seven of the ten pre-war states\n(Western Equatoria, Central Equatoria, Eastern Equatoria, Northern Bahr-El-Ghazl, Western Bahr-El-Ghazal, Warrap, Lakes\nstate).] .\n\n\n[10 This work is part of the program “Building the Evidence on Forced Displacement: A Multi-Stakeholder Partnership”. The](https://www.worldbank.org/en/programs/building-the-evidence-on-forced-displacement)\nprogram is funded by UK aid from the United Kingdom's Foreign, Commonwealth, and Development Office (FCDO). It is\nmanaged by the World Bank Group (WBG) and was established in partnership with the United Nations High Commissioner for\nRefugees (UNHCR). This\n\nMention: \"IDP Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Samples displaced people and host communities in six states.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000595:24:1", "occurrence_id": "7ce1ecbf2512d78d1f0a218c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: -Poverty-in-Contexts-of-Forced-Displacement](http://documents.worldbank.org/curated/en/492181635479693932/A-Multi-Country-Analysis-of-Multidimensional-Poverty-in-Contexts-of-Forced-Displacement)</u>_\n\nThis paper **develops a Multidimensional Poverty Index (MPI) to examine patterns**\n\n**of multidimensional poverty among IDPs and refugees, with comparisons to**\n\n**host populations, in five African countries** . The MPI is disaggregated to analyze\n\nvariations in deprivation by displacement status of the household and gender of the\n\nhousehold head. The analysis draws on household survey data from Ethiopia, Nigeria,\n\nSomalia, South Sudan, and Sudan <sup>11</sup> [^11: In Ethiopia, the Skills Profile Survey (2017) sampled refugees in and around camps in the Tigray, Afar, Gambella,\nBenishangul Gumuz, and Somali regions. In Nigeria, the IDP Survey (2018) sampled IDPs and host communities in six\nnortheastern states (Adamawa, Bauchi, Borno, Gombe, Taraba, and Yobe). In Somalia, the High Frequency Survey (2017)\nsampled IDPs and host communities in secure parts of the country. In Sudan, the IDP Profiling Survey (2018) sampled IDPs\nand neighboring host communities in the Abu Shouk and El Salam camps, in Al-Fashir. And in South Sudan, the High\nFrequency Survey Wave 4 (2017) sampled IDPs and host communities in urban areas of seven of the ten pre-war states\n(Western Equatoria, Central Equatoria, Eastern Equatoria, Northern Bahr-El-Ghazl, Western Bahr-El-Ghazal, Warrap, Lakes\nstate).] .\n\n\n[10 This work is part of the program “Building the Evidence on Forced Displacement: A Multi-Stakeholder Partnership”. The](https://www.worldbank.org/en/programs/building-the-evidence-on-forced-displacement)\nprogram is funded by UK aid from the United Kingdom's Foreign, Commonwealth, and Development Office (FCDO). It is\nmanaged by the World Bank Group (WBG) and was established in partnership with the United Nations High Commissioner for\nRefugees (UNHCR). This\n\nMention: \"High Frequency Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Samples in the safe parts of the country.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000595:24:1", "occurrence_id": "2ab3ae22230117f39c74f798", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: -Poverty-in-Contexts-of-Forced-Displacement](http://documents.worldbank.org/curated/en/492181635479693932/A-Multi-Country-Analysis-of-Multidimensional-Poverty-in-Contexts-of-Forced-Displacement)</u>_\n\nThis paper **develops a Multidimensional Poverty Index (MPI) to examine patterns**\n\n**of multidimensional poverty among IDPs and refugees, with comparisons to**\n\n**host populations, in five African countries** . The MPI is disaggregated to analyze\n\nvariations in deprivation by displacement status of the household and gender of the\n\nhousehold head. The analysis draws on household survey data from Ethiopia, Nigeria,\n\nSomalia, South Sudan, and Sudan <sup>11</sup> [^11: In Ethiopia, the Skills Profile Survey (2017) sampled refugees in and around camps in the Tigray, Afar, Gambella,\nBenishangul Gumuz, and Somali regions. In Nigeria, the IDP Survey (2018) sampled IDPs and host communities in six\nnortheastern states (Adamawa, Bauchi, Borno, Gombe, Taraba, and Yobe). In Somalia, the High Frequency Survey (2017)\nsampled IDPs and host communities in secure parts of the country. In Sudan, the IDP Profiling Survey (2018) sampled IDPs\nand neighboring host communities in the Abu Shouk and El Salam camps, in Al-Fashir. And in South Sudan, the High\nFrequency Survey Wave 4 (2017) sampled IDPs and host communities in urban areas of seven of the ten pre-war states\n(Western Equatoria, Central Equatoria, Eastern Equatoria, Northern Bahr-El-Ghazl, Western Bahr-El-Ghazal, Warrap, Lakes\nstate).] .\n\n\n[10 This work is part of the program “Building the Evidence on Forced Displacement: A Multi-Stakeholder Partnership”. The](https://www.worldbank.org/en/programs/building-the-evidence-on-forced-displacement)\nprogram is funded by UK aid from the United Kingdom's Foreign, Commonwealth, and Development Office (FCDO). It is\nmanaged by the World Bank Group (WBG) and was established in partnership with the United Nations High Commissioner for\nRefugees (UNHCR). This\n\nMention: \"IDP Profiling Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Samples in two camps of a given town.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000595:24:1", "occurrence_id": "5285192a950d4c23fa31d2cf", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: -Poverty-in-Contexts-of-Forced-Displacement](http://documents.worldbank.org/curated/en/492181635479693932/A-Multi-Country-Analysis-of-Multidimensional-Poverty-in-Contexts-of-Forced-Displacement)</u>_\n\nThis paper **develops a Multidimensional Poverty Index (MPI) to examine patterns**\n\n**of multidimensional poverty among IDPs and refugees, with comparisons to**\n\n**host populations, in five African countries** . The MPI is disaggregated to analyze\n\nvariations in deprivation by displacement status of the household and gender of the\n\nhousehold head. The analysis draws on household survey data from Ethiopia, Nigeria,\n\nSomalia, South Sudan, and Sudan <sup>11</sup> [^11: In Ethiopia, the Skills Profile Survey (2017) sampled refugees in and around camps in the Tigray, Afar, Gambella,\nBenishangul Gumuz, and Somali regions. In Nigeria, the IDP Survey (2018) sampled IDPs and host communities in six\nnortheastern states (Adamawa, Bauchi, Borno, Gombe, Taraba, and Yobe). In Somalia, the High Frequency Survey (2017)\nsampled IDPs and host communities in secure parts of the country. In Sudan, the IDP Profiling Survey (2018) sampled IDPs\nand neighboring host communities in the Abu Shouk and El Salam camps, in Al-Fashir. And in South Sudan, the High\nFrequency Survey Wave 4 (2017) sampled IDPs and host communities in urban areas of seven of the ten pre-war states\n(Western Equatoria, Central Equatoria, Eastern Equatoria, Northern Bahr-El-Ghazl, Western Bahr-El-Ghazal, Warrap, Lakes\nstate).] .\n\n\n[10 This work is part of the program “Building the Evidence on Forced Displacement: A Multi-Stakeholder Partnership”. The](https://www.worldbank.org/en/programs/building-the-evidence-on-forced-displacement)\nprogram is funded by UK aid from the United Kingdom's Foreign, Commonwealth, and Development Office (FCDO). It is\nmanaged by the World Bank Group (WBG) and was established in partnership with the United Nations High Commissioner for\nRefugees (UNHCR). This\n\nMention: \"High\nFrequency Survey Wave 4\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Fourth round covering urban areas of seven regions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000596:4:1", "occurrence_id": "6c677f3483395ec2fc270200", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  and stillbirth audits are particularly important as care often falls short\nbetween different providers and even between different departments or units. However, audits\nalone cannot improve the quality of care and outcomes; unless the recommendations contained\nwithin the audit process are effectively implemented, maternal and neonatal outcomes will not\nimprove. <sup>16,17</sup> [^16: Pattinson R1, Kerber K, et. al. Perinatal mortality audit: counting, accountability, and overcoming challenges\n[in scaling up in low- and middle-income countries. Int J Gynecology Obstet.2009 Oct;107 Suppl 1: S113-21,](http://www.ncbi.nlm.nih.gov/pubmed/19815206)\nS121-2.] [^17: EJ Buchmann. Towards greater effectiveness of perinatal death audit in low- and middle-income countries.\n<u>[BJOG. Volume 121, Issue Supplement s4, pages 134–136, September 2014](http://onlinelibrary.wiley.com/doi/10.1111/bjo.2014.121.issue-s4/issuetoc)</u>]\n\n\n**_Neonatal death and stillbirth audits in Syrian refugee camps_**\n\n\nCollectively, Zaatari and Azraq camps host approximately 113,752 refugees; 76878 live in\nZaatari camp and 36874 live in Azraq camp. Pregnant women receive regular checkups in the\ncamps' clinics throughout their pregnancy. They usually deliver their babies in camp hospitals.\nHowever, complicated cases are referred to other health facilities based on defined criteria\ndepending primarily on the facilities capacities. In line with The United Nations High\nCommissioner for Refugees (UNHCR)’s global strategy, UNHCR Jordan has established a\n\n\n13 Shattnawi KK, Khader YS, Alyahya MS, Al-Sheyab N, Batieha A. Rate, determinants, and causes of\nstillbirth in Jordan: Findings from the Jordan Stillbirth and Neonatal Deaths Surveillance (JSANDS) system.\nBMC Pregnancy Childbirth. 2020 Sep 29;20(1):571\n14 Kerber et al. BMC Pregnancy and Childbirth 2015, 15\n\nMention: \"Jordan Stillbirth and Neonatal Deaths Surveillance\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its data feed the analysis of stillbirth rates and causes.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000602:7:0", "occurrence_id": "4862ef77f7121c1a38a074e3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: #### **References**\n\n1. FAO (Food and Agriculture Organization of the United Nations). 2023. Project brief: _Greening the humanitarian_\n_response in displacement settings – Ecosystem restoration and sustainable forest management for enhanced_\n_energy access and livelihood resilience_ . Rome. <u>[https://doi.org/10.4060/cc5353en](https://doi.org/10.4060/cc5353en)</u>\n2. United Nations High Commissioner for Refugees (UNHCR). 2024. Refugee Data Finder. Cited 2 August 2024.\n<u>[https://www.unhcr.org/refugee-statistics/](https://www.unhcr.org/refugee-statistics/)</u>\n3. UNHCR. 2024. _Focus Area Strategic Plan for Climate Action 2024-2030_ . Geneva, UNHCR. <u>[https://reporting.unhcr.](https://reporting.unhcr.org/climate-action-focus-area-strategic-plan-2024-2030)</u>\n<u>[org/climate-action-focus-area-strategic-plan-2024-2030](https://reporting.unhcr.org/climate-action-focus-area-strategic-plan-2024-2030)</u>\n4. FAO & UNHCR. 2018. _Managing forests in displacement settings: guidance on the use of planted and natural_\n_forests to supply forest products and build resilience in displaced and host communities_ . Rome, FAO. 84 pp.\n<u>[https://openknowledge.fao.org/server/api/core/bitstreams/573bbbad-ff4c\n\nMention: \"Refugee Data Finder\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Consulté à cette date pour les statistiques sur les réfugiés.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000607:5:0", "occurrence_id": "874b65afc8c7d00b34c80bb4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ### LEBANON Beirut\n\n11 December 2012\n\n\nThe most vulnerable cases registered in the UNHCR database that are not able to work and who will need some\n\nkind of support in order to pay their rent. According to UNHCR statistics, this category represents 30% of Syrian\n\nrefugees in Lebanon.\n\nThe caseload of those who have lost their job or who cannot find work due to the saturation of the market will fall\n\nunder the vulnerable families’ category for the purpose of assistance.\n\n\n4.5.2. Overall risks to take into consideration and recommendations\n\n\nThe risk of a price increase in the real estate market as a result of the increase in Syrian refugees’ demand for\n\nhousing, meaning that cash-for-shelter or unconditional cash support may be insufficient for families to find\n\naccommodation.\n\n- Recommendation: Review of the market situation on a regular basis, conduct household economic surveys and\n\nadaptation of the cash amount.\n\n\nThe risk that host families, who were previously hosting refugee families for free, may no longer be willing to do so\n\nif others are receiving rent through a cash-for-shelter program.\n\n- Recommendation: Consider cash support to host families.\n\n\nThe risk of interruption of shelter support payment. It is currently expected that Syrian refugees will stay in\n\nLebanon for a prolonged, albeit temporary, period. Would donors be willing to commit to cash-for-rent programs\n\nfor this length of time?\n\n- Recommendation: Advocate to donors on this matter.\n\n\n4.6. Develop contingency and preparedness in case of massive influx of refugees\n\n\nA massive influx of refugees is highly likely and needs to be taken seriously.\n\n\nThe Lebanese government and MOSA are working on contingency planning and preparedness in case of the crisis\nleads to a massive influx. The identification of a site upon which to set up a camp is underway.\n\n\nIn parallel to government preparation for camp planning, UNHCR and its IP are also looking into contingency\nplanning in case there is a continuous rise in the refugee influx and also in case the planned\n\nMention: \"UNHCR database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The most vulnerable cases are registered there for assistance.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000607:5:0", "occurrence_id": "d19de64a2df2af7f8fc709e1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ### LEBANON Beirut\n\n11 December 2012\n\n\nThe most vulnerable cases registered in the UNHCR database that are not able to work and who will need some\n\nkind of support in order to pay their rent. According to UNHCR statistics, this category represents 30% of Syrian\n\nrefugees in Lebanon.\n\nThe caseload of those who have lost their job or who cannot find work due to the saturation of the market will fall\n\nunder the vulnerable families’ category for the purpose of assistance.\n\n\n4.5.2. Overall risks to take into consideration and recommendations\n\n\nThe risk of a price increase in the real estate market as a result of the increase in Syrian refugees’ demand for\n\nhousing, meaning that cash-for-shelter or unconditional cash support may be insufficient for families to find\n\naccommodation.\n\n- Recommendation: Review of the market situation on a regular basis, conduct household economic surveys and\n\nadaptation of the cash amount.\n\n\nThe risk that host families, who were previously hosting refugee families for free, may no longer be willing to do so\n\nif others are receiving rent through a cash-for-shelter program.\n\n- Recommendation: Consider cash support to host families.\n\n\nThe risk of interruption of shelter support payment. It is currently expected that Syrian refugees will stay in\n\nLebanon for a prolonged, albeit temporary, period. Would donors be willing to commit to cash-for-rent programs\n\nfor this length of time?\n\n- Recommendation: Advocate to donors on this matter.\n\n\n4.6. Develop contingency and preparedness in case of massive influx of refugees\n\n\nA massive influx of refugees is highly likely and needs to be taken seriously.\n\n\nThe Lebanese government and MOSA are working on contingency planning and preparedness in case of the crisis\nleads to a massive influx. The identification of a site upon which to set up a camp is underway.\n\n\nIn parallel to government preparation for camp planning, UNHCR and its IP are also looking into contingency\nplanning in case there is a continuous rise in the refugee influx and also in case the planned\n\nMention: \"household economic surveys\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Recommended for adapting the cash assistance amount.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000615:6:0", "occurrence_id": "8f9e85198485d71e472dc3ee", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Humanitarian coordinators and**\n**humanitarian country teams (HCTs) must:**\n\n\n- Ensure that HCT protection strategies include\nresource mobilisation components and are closely\nlinked and complement the HRPs.\n\n- Ensure that protection requirements in HRPs are\naccurately set and are commensurate with the\nassessed needs in the Humanitarian Needs\nOverview (HNO).\n\n\n**The Global Protection Cluster should:**\n\n\n- Maintain protection funding as a strategic priority in\nits GPC Advocacy Strategy.\n\n- Develop a monitoring framework for tracking\nunderfunding of the protection sector and report\nregularly on this in the GPC Centrality of Protection\nannual report.\n\n- Develop guidelines for a minimum package of\nprotection interventions in HRPs along with the unit\ncosting of different protection interventions.\n\n- Address the issue of protection financing in the\nGPC guidance being developed on the application\nof the nexus approach for field protection clusters.\n\n- Initiate a policy dialogue with development actors\n(e.g. through the OECD Development Assistance\nCommittee (DAC) nexus working group) to ensure\nthat protection financing is addressed within the\nnexus approach.\n\n\nBREAKING THE GLASS CEILING: A SMARTER APPROACH TO PROTECTION FINANCING **7**\n\nMention: \"Humanitarian Needs\nOverview\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its assessed needs are the reference for plan amounts.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000619:23:0", "occurrence_id": "407911dc822c0e3e0aec272d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _only_ marginalized children. For example, differentiated instruction can benefit all children, but the\nimprovements in learning are particularly pronounced in those children who are most left behind\n[by current education systems (Banerjee et al., 2016). In some cases, a combination of interventions](https://www.nber.org/papers/w22746)\ntargeted at specific marginalized groups and more systemic interventions will be required. For\nexample, targeted finance such as cash transfers can help get girls into school, but systemic interventions to improve teaching may be needed to ensure that all children in school (including girls)\n[are learning (DFID, 2018b). It is also important to note that marginalized children are not only in the](https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/733765/Steps-to-Success.pdf)\npoorest countries. Large numbers of poor children live in middle-income countries.\n\n**■ Poor and rural communities**\n\n**The vast majority of children in learning poverty come from poor and/or rural households** . In\ncountries participating in the Program for International Student Assessment, differences in attainment\n[between rural and urban setting were mainly explained by socioeconomic differences (Echazarra &](https://www.oecd-ilibrary.org/docserver/8b1a5cb9-en.pdf?expires=1601578642&id=id&accname=guest&checksum=373959997E15A832CE87CF5CF3EF94C2)\n<u>[Radinger, 2019](https://www.oecd-ilibrary.org/docserver/8b1a5cb9-en.pdf?expires=1601578642&id\n\nMention: \"Program for International Student Assessment\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Participating countries show the achievement gap by setting.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000584:2:0", "occurrence_id": "1b7ed68447af8244a41538b0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 3\n\n\n###### Demographic profile of migrants and refugees in the region\n\n\n\n\n\n\n\n\n\n\n\n**Source:** [ESCWA calculations based on DESA, International Migration Stock, 2020.](https://www.un.org/development/desa/pd/content/international-migrant-stock)\n\n**Note:** Whenever data from DESA is referenced, the term migrants and refugees is used since the data include both populations. However, it is important to note the different\nrealities and migration patterns in the region, with GCC subregion hosting mostly migrant workers, while refugees are mostly concentrated in the Mashreq.\n\n###### Remittances\n\n\n\n\n\n\n\n\n\n\n\n\n\n**Source:** [World Bank, Annual remittances data (updated as of May 2021).](https://www.worldbank.org/en/topic/labormarkets/brief/migration-and-remittances)\n\n###### Sub-regional dynamics\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n**Source:** [ESCWA calculations based on DESA, International Migration Stock, 2020; Internal Displacement Monitoring Centre, Global Internal Displacement Database, 2020.](https://www.un.org/development/desa/pd/content/international-migrant-stock)\n**Note:** Whenever data from DESA is referenced, the term migrants and refugees is used since the data include both populations. However, it is important to note the different\nrealities and migration patterns in the region, with GCC subregion hosting mostly migrant workers, while refugees are mostly concentrated in the Mashreq.\n\nMention: \"International Migration Stock\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its regional calculations give migrant stocks.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000584:2:0", "occurrence_id": "338f20cf1237c2fe08f058d1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 3\n\n\n###### Demographic profile of migrants and refugees in the region\n\n\n\n\n\n\n\n\n\n\n\n**Source:** [ESCWA calculations based on DESA, International Migration Stock, 2020.](https://www.un.org/development/desa/pd/content/international-migrant-stock)\n\n**Note:** Whenever data from DESA is referenced, the term migrants and refugees is used since the data include both populations. However, it is important to note the different\nrealities and migration patterns in the region, with GCC subregion hosting mostly migrant workers, while refugees are mostly concentrated in the Mashreq.\n\n###### Remittances\n\n\n\n\n\n\n\n\n\n\n\n\n\n**Source:** [World Bank, Annual remittances data (updated as of May 2021).](https://www.worldbank.org/en/topic/labormarkets/brief/migration-and-remittances)\n\n###### Sub-regional dynamics\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n**Source:** [ESCWA calculations based on DESA, International Migration Stock, 2020; Internal Displacement Monitoring Centre, Global Internal Displacement Database, 2020.](https://www.un.org/development/desa/pd/content/international-migrant-stock)\n**Note:** Whenever data from DESA is referenced, the term migrants and refugees is used since the data include both populations. However, it is important to note the different\nrealities and migration patterns in the region, with GCC subregion hosting mostly migrant workers, while refugees are mostly concentrated in the Mashreq.\n\nMention: \"Annual remittances data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its annual data feed the remittance analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000584:2:0", "occurrence_id": "5f2d54a8902325f9c702d976", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 3\n\n\n###### Demographic profile of migrants and refugees in the region\n\n\n\n\n\n\n\n\n\n\n\n**Source:** [ESCWA calculations based on DESA, International Migration Stock, 2020.](https://www.un.org/development/desa/pd/content/international-migrant-stock)\n\n**Note:** Whenever data from DESA is referenced, the term migrants and refugees is used since the data include both populations. However, it is important to note the different\nrealities and migration patterns in the region, with GCC subregion hosting mostly migrant workers, while refugees are mostly concentrated in the Mashreq.\n\n###### Remittances\n\n\n\n\n\n\n\n\n\n\n\n\n\n**Source:** [World Bank, Annual remittances data (updated as of May 2021).](https://www.worldbank.org/en/topic/labormarkets/brief/migration-and-remittances)\n\n###### Sub-regional dynamics\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n**Source:** [ESCWA calculations based on DESA, International Migration Stock, 2020; Internal Displacement Monitoring Centre, Global Internal Displacement Database, 2020.](https://www.un.org/development/desa/pd/content/international-migrant-stock)\n**Note:** Whenever data from DESA is referenced, the term migrants and refugees is used since the data include both populations. However, it is important to note the different\nrealities and migration patterns in the region, with GCC subregion hosting mostly migrant workers, while refugees are mostly concentrated in the Mashreq.\n\nMention: \"Global Internal Displacement Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Completes migration stocks with internal displacement.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000627:24:0", "occurrence_id": "82fa986764152bf2b22c4ac7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Darwanaji and Teferi Ber) received theoretically 37% and 12% of all returnees during\nthe period under consideration, but in real terms they received only 27% and 5%.\n\nConversely, Hargeisa and Gabiley received \"officially\" 41% and 9% of the returnees\nwhile \"actually\" the percentages were 55% and 13%. We should however still bear in\nmind that the greatest impact of all was during the spontaneous or self-repatriation\nphase. These differences were not caused - as commonly assumed during the period\nunder consideration (1997-99) - by an \"urbanisation\" of refugees in the camps, i.e.\nrefugees of rural or pastoral origin who got used to easy access to social services in\nthe camps and perceive greater job opportunities in urban areas and as a result decide\nnot to repatriate to their ancestral areas in the bush.\n\nTo be sure, we cannot exclude that this phenomenon played some role in this period\nand maybe a greater role subsequently in other camps, such as the Aware camps <sup>21</sup> [^21: Refugees in the Aware camps (Camaboker, Rabasso and Daror) mainly originate from Burao and the\nrural areas between Burao and Hargeisa, such as Salahley and Odweyne However there was also a\nsubstantial minority from Hargeisa. Repatriation from these camps started only after the period under\nconsideration.] .\nBut the view commonly held by many members of the international community and\nmany \"Hargeisawis\" (Hargeisa dwellers) that this people were illegitimately returning\nto Hargeisa instead of the countryside was not supported by evidence.\n\nFirst we can recall how a 1994 survey conducted in the camps by a Somali\nanthropologist graduated from the LSE concluded that \"many urban poor remain in\nHartasheikh…\" <sup>22</sup> [^22: \"Going Back Home\", _op. cit._, see above for the full quote.] . Second, an unpublished \"Social Assessment of Somali Returnees\nin Awadal and Waqooyi Galbeed Regions of NW Somalia\" conducted in 1998 on\n\nMention: \"1994 survey conducted in the camps\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its conclusions on urban poor are cited and discussed.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000627:24:0", "occurrence_id": "1e9b90b25e84e6337d621d60", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Darwanaji and Teferi Ber) received theoretically 37% and 12% of all returnees during\nthe period under consideration, but in real terms they received only 27% and 5%.\n\nConversely, Hargeisa and Gabiley received \"officially\" 41% and 9% of the returnees\nwhile \"actually\" the percentages were 55% and 13%. We should however still bear in\nmind that the greatest impact of all was during the spontaneous or self-repatriation\nphase. These differences were not caused - as commonly assumed during the period\nunder consideration (1997-99) - by an \"urbanisation\" of refugees in the camps, i.e.\nrefugees of rural or pastoral origin who got used to easy access to social services in\nthe camps and perceive greater job opportunities in urban areas and as a result decide\nnot to repatriate to their ancestral areas in the bush.\n\nTo be sure, we cannot exclude that this phenomenon played some role in this period\nand maybe a greater role subsequently in other camps, such as the Aware camps <sup>21</sup> [^21: Refugees in the Aware camps (Camaboker, Rabasso and Daror) mainly originate from Burao and the\nrural areas between Burao and Hargeisa, such as Salahley and Odweyne However there was also a\nsubstantial minority from Hargeisa. Repatriation from these camps started only after the period under\nconsideration.] .\nBut the view commonly held by many members of the international community and\nmany \"Hargeisawis\" (Hargeisa dwellers) that this people were illegitimately returning\nto Hargeisa instead of the countryside was not supported by evidence.\n\nFirst we can recall how a 1994 survey conducted in the camps by a Somali\nanthropologist graduated from the LSE concluded that \"many urban poor remain in\nHartasheikh…\" <sup>22</sup> [^22: \"Going Back Home\", _op. cit._, see above for the full quote.] . Second, an unpublished \"Social Assessment of Somali Returnees\nin Awadal and Waqooyi Galbeed Regions of NW Somalia\" conducted in 1998 on\n\nMention: \"Social Assessment of Somali Returnees\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"An unpublished assessment covering two north-western regions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000628:3:1", "occurrence_id": "17de0029e70fdd11eb85510e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: /DESA (July 2022), Population Estimates 2022: <u>https://population.un.org/</u>\n<u>wpp/.</u>\n\n6 UNHCR RBWCA (August 2022), Refugee Education Statistics.\n\n7 Project 21 online platform: <u>[https://www.humanitarianresponse.info/en/opera-](https://www.humanitarianresponse.info/en/operations/west-and-central-africa/project-21)</u>\n<u>[tions/west-and-central-africa/project-21.](https://www.humanitarianresponse.info/en/operations/west-and-central-africa/project-21)</u>\n\n8 Project 21, <u>[January-April 2022 Analysis.](https://www.humanitarianresponse.info/sites/www.humanitarianresponse.info/files/projet_21_mise_a_jour_1_janvier-avril_2022.pdf)</u>\n\n\n**4**\n\n\n\n\n\n\n\n\n\nThe reduction in enrolment and attendance is also due to the\nfact that 53 per cent of children reveal that they do not feel\nsafe at school, with almost two-thirds (64 per cent) of children\nreporting having little to no hope for their future <sup>9</sup> [^9: NRC, UNHCR, UNICEF (January 2022). <u>[Central Sahel: Improve children's wellbeing](https://www.humanitarianresponse.info/en/operations/west-and-central-africa/document/central-sahel-improve-childrens-wellbeing-and-learning)</u>\n<u>[and learning - Increasing psychosocial support in schools.](https://www.humanitarianresponse.info/en/operations/west-and-central-africa/document/central-sahel-improve-childrens-wellbeing-and-learning)</u>] .\n\nThe region therefore faces an unprecedented challenge, and\nan important part of its response will be its ability to provide\na better future for its youth. Despite this frightening situation\nand increased awareness of the problem, the education sector\nand more specifically the humanitarian response to education\nneeds are still severely underfunded: for\n\nMention: \"Refugee Education Statistics\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its statistics cover refugee school enrolment.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000639:6:1", "occurrence_id": "decb20f0a7459126049791c8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: hosting districts. Two representatives from REF attend the steering group to represent the\nrefugee voice.\n\n\nRefugees have so far not been included in the national population census in Uganda but there is\ndevelopment in related areas. In 2018, the Uganda Bureau of Statistics (UBOS), with support from the WB\n[and UNHCR, embarked on a Refugee and Host Community Household Survey (World Bank, 2019).](https://documents.worldbank.org/en/publication/documents-reports/documentdetail/571081569598919068/informing-the-refugee-policy-response-in-uganda-results-from-the-uganda-refugee-and-host-communities-2018-household-survey)\nRefugees are also included in the national health management information systems, as well as the\nepidemic preparedness and response plans, including those for Cholera, Ebola and Covid-19. Data on\nrefugee GBV is not systematically collected by the Government, but UNHCR and its humanitarian partners\nare collecting this data and have agreed with the Government that they will feed it into the National GBV\nDatabase once the Government has addressed data protection and confidentiality-related challenges.\n\n\nPriorities for refugee protection and management have been included in the <u>[NDP II and sub-national](http://npa.go.ug/wp-content/uploads/NDPII-Final.pdf)</u>\nlevel development planning processes. During the fourth ministerial level National Partnership Forum,\nheld in February 2018, a commitment was made to improve these priority areas for the NDP III.\n\n\n**2.4** **Access to civil registration and documentation**\n\n\n[The Refugees Act 2006](https://www.refworld.org/docid/4b7baba52.html) [and its 2010 Regulations](https://www.refworld.org/\n\nMention: \"Refugee and Host Community Household Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Conducted with the national statistics office and two agencies.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000639:6:1", "occurrence_id": "a1cb06fe312b8d7b722228cc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: hosting districts. Two representatives from REF attend the steering group to represent the\nrefugee voice.\n\n\nRefugees have so far not been included in the national population census in Uganda but there is\ndevelopment in related areas. In 2018, the Uganda Bureau of Statistics (UBOS), with support from the WB\n[and UNHCR, embarked on a Refugee and Host Community Household Survey (World Bank, 2019).](https://documents.worldbank.org/en/publication/documents-reports/documentdetail/571081569598919068/informing-the-refugee-policy-response-in-uganda-results-from-the-uganda-refugee-and-host-communities-2018-household-survey)\nRefugees are also included in the national health management information systems, as well as the\nepidemic preparedness and response plans, including those for Cholera, Ebola and Covid-19. Data on\nrefugee GBV is not systematically collected by the Government, but UNHCR and its humanitarian partners\nare collecting this data and have agreed with the Government that they will feed it into the National GBV\nDatabase once the Government has addressed data protection and confidentiality-related challenges.\n\n\nPriorities for refugee protection and management have been included in the <u>[NDP II and sub-national](http://npa.go.ug/wp-content/uploads/NDPII-Final.pdf)</u>\nlevel development planning processes. During the fourth ministerial level National Partnership Forum,\nheld in February 2018, a commitment was made to improve these priority areas for the NDP III.\n\n\n**2.4** **Access to civil registration and documentation**\n\n\n[The Refugees Act 2006](https://www.refworld.org/docid/4b7baba52.html) [and its 2010 Regulations](https://www.refworld.org/\n\nMention: \"Data on\nrefugee GBV\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Not systematically collected by the government.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000639:6:1", "occurrence_id": "23e90858d86aa140a9cf3775", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: hosting districts. Two representatives from REF attend the steering group to represent the\nrefugee voice.\n\n\nRefugees have so far not been included in the national population census in Uganda but there is\ndevelopment in related areas. In 2018, the Uganda Bureau of Statistics (UBOS), with support from the WB\n[and UNHCR, embarked on a Refugee and Host Community Household Survey (World Bank, 2019).](https://documents.worldbank.org/en/publication/documents-reports/documentdetail/571081569598919068/informing-the-refugee-policy-response-in-uganda-results-from-the-uganda-refugee-and-host-communities-2018-household-survey)\nRefugees are also included in the national health management information systems, as well as the\nepidemic preparedness and response plans, including those for Cholera, Ebola and Covid-19. Data on\nrefugee GBV is not systematically collected by the Government, but UNHCR and its humanitarian partners\nare collecting this data and have agreed with the Government that they will feed it into the National GBV\nDatabase once the Government has addressed data protection and confidentiality-related challenges.\n\n\nPriorities for refugee protection and management have been included in the <u>[NDP II and sub-national](http://npa.go.ug/wp-content/uploads/NDPII-Final.pdf)</u>\nlevel development planning processes. During the fourth ministerial level National Partnership Forum,\nheld in February 2018, a commitment was made to improve these priority areas for the NDP III.\n\n\n**2.4** **Access to civil registration and documentation**\n\n\n[The Refugees Act 2006](https://www.refworld.org/docid/4b7baba52.html) [and its 2010 Regulations](https://www.refworld.org/\n\nMention: \"National GBV\nDatabase\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"inform\", \"summary\": \"Data collected there would be transferred once protection concerns are resolved.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000643:1:0", "occurrence_id": "0fec363940197ad92b0099ce", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Ce rapport présente les résultats globaux de l’enquête sur les PDIs de mai 2011 réalisée sur deux\naxes (Ndélé Centre/Dar‐el‐Kouti et Bamingui) de la Bamingui Bangoran. Il a été conçu dans le but\nd’améliorer le niveau d’information sur le nombre désagrégé par sexe et âge, et la répartition des\nPDIs ainsi que leurs conditions de vie et leurs besoins de base pour un meilleur ciblage des\ninterventions en leur faveur.\n\n#### **Conclusion générale**\n\n\n**La conclusion générale du profilage est que la situation entre les PDIs et les Non PDIs est**\n**globalement similaire** . Les différences sont nettement moins élevées que dans d’autres\nsituations de crise. Cependant, les PDIs se trouvent dans de situations défavorables en ce qui\nconcerne certains secteurs ou d’aspects à l’intérieur de secteurs, notamment l’eau et\nassainissement.\n\n\nLa similarité générale peut s’expliquer par plusieurs facteurs :\n\n\na) Les PDIs disposent d’une résilience et capacité d’adaptation considérables.\n\n\nb) Un tissu familial favorisant l’absorption des PDIs dans les nouvelles localités est bien en place.\n23,7% de PDIs se considèrent comme des résidents permanents tandis que 22,9% de PDIs sont\naccueillis dans un ménage. En tenant compte de l’assimilation et du fait que 99% des membres\nde ménages proviennent de la même famille et peuvent\n\nMention: \"enquête sur les PDIs\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Ses résultats globaux couvrent deux axes d'une région.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000645:0:1", "occurrence_id": "71218524da07c09ca59ad6f2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  the international community redirecting\nits financial and humanitarian support to the Ukraine\nemergency. Internally Displaced People (IDPs), refugees,\nasylum seekers, returnees, stateless persons and migrants\nare at a heightened risk of food insecurity as many have\nleft behind assets, lost their social capital, and livelihoods.\nA recent UNHCR Standardised Expanded Nutrition Survey\n(SENS) reported “Critical” levels of child malnutrition (wasting,\nstunting and anemia) amongst refugee children, specifically\nin refugee sites in **Ethiopia, Kenya, Sudan, South Sudan,** and\n\n**Uganda** . <sup>2</sup> This situation comes in the wake of humanitarian\n\nfunding shortfalls that have contributed to food ration cuts <sup>3</sup>\nand amplified the cross-sectional protection needs of the\nmost vulnerable.\n\n\n\n<u>1</u> <u>https://reliefweb.int/report/ethiopia/horn-africa-drought-regional-humanitarian-overview-call-action-revised-24-au-</u>\ngust-2022\n2 https://www.icpac.net/documents/572/IGAD_RRFC_2022_ONLINE_4eYMbKh.pdf\n3 https://news.un.org/en/story/2022/06/1120792\n\n\nRegional Protection Working Group\n\n\n##### **01**\n\nMention: \"UNHCR Standardised Expanded Nutrition Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its results report critical levels of child malnutrition.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000648:4:0", "occurrence_id": "c34589dd00ddcfd201d911d7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: At the regional level, UNHCR’s Regional Protection\nMonitoring System (RPMS) was launched in 2021.\nThe system includes disaggregation by age, gender\nand specific needs, including disabilities, using the\nWashington Group Questions.\n\n\nThe Regional Bureau for Europe regularly releases\na range of external information products with\ndisaggregated data on persons of concern. For\nexample, arrival data for the Mediterranean situation\nis disaggregated (or estimated) by age group and\nsex and displayed on UNHCR’s Operational Data\nPortal.⁴ In addition, the joint UNHCR, UNICEF and\nIOM fact sheet on Refugee and Migrant Children\n\n\n\nin Europe provides a detailed breakdown on the\nage and sex of accompanied, unaccompanied and\nseparated children arriving in Europe.⁵\n\n\nTo the extent possible during the COVID-19\npandemic, UNHCR offices in Europe continued to\nconduct participatory assessments with persons of\nconcern and use these assessments to inform and\nadjust programming and advocacy interventions.\nAcross the region, such assessments took place\nin 17 countries in 2021, consulting with refugees,\nasylum-seekers, internally displaced persons\n(IDPs) and stateless persons about key protection\nchallenges and solutions.\n\n\n\n4  UNHCR’s Operational Data Portal for the Mediterranean Situation is available at: <u>[https://data2.unhcr.org/en/situations/mediterranean](https://data2.unhcr.org/en/situations/mediterranean)</u>\n[5  The joint UNHCR, UNICEF and IOM fact sheet on Refugee and Migrant Children in Europe is available at: www.unhcr.org/cy/wp-content/uploads/](https://www.unhcr.org/cy/wp-content/uploads/sites/41/2020/06/UNHCR-UNICEF-and-IOM_Refugee-and-Migrant-children-in-Europe-2019.pdf)\n<u\n\nMention: \"arrival data for the Mediterranean situation\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Disaggregated by age group and sex on the operational dashboard.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000650:51:1", "occurrence_id": "880a0e53da325a0a10c6e447", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Convention as a\nresponse to the post-war European refugee crisis.\nThe compassion shown to Europe’s refugees then\nis barely afforded them now. Those who risk their\nlives attempting the journey to Europe today face\nindifference at best, open hostility at worst. At the\nsame time, countries who do look after refugees\nhave been left to shoulder the burden by their\nsupposed European partners.\n\n\nYet no one watching and reading the stories in\nUNHCR’s annual education report could honestly\ndoubt the desire of these youngsters to see their\nstudies through a full educational cycle. We must\nnot be satisfied with merely helping them to\nsurvive the ordeal of displacement; we must give\nthem the tools to thrive, to be independent,\nself-sufficient and fulfilled individuals.\n\n\nThe time has come for decisive, game-changing\nmeasures to end this neglect. That is why I am\nsupporting UNHCR’s new initiative, the Secondary\nYouth Education Programme, as an effort to rally\nthe international community to this cause. If we are\nto give millions of children, refugees and nonrefugees alike, the chance of an education and the\nopportunity to unlock their potential, innovative\nfinancing projects such this programme,\ncombining the know-how and firepower of the\npublic and private sectors, will play a decisive role.\n\n\n\n52 U N H C R > **S T E P P I N G U P :** **R E F U G E E E D U C AT I O N I N C R I S I S**\n\nMention: \"UNHCR’s annual education report\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its accounts evidence young people's desire to finish school.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000652:3:0", "occurrence_id": "3c37e3093bf768c9d93f000d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ### **II. Funding for Protection programming in Yemen**\n\nTraditionally, there is already low funding for protection interventions in Yemen.\nAs of July 2024, the Financial Tracking Services indicates that Protection and\nAoRs are funded by a meager 37 percent of required funds. This underscores\na broader failure to recognize protection as a life-saving activity that requires\nappropriate funding for stand-alone protection programming as well as for\nintegrated approaches.\n\n### **III. Banking Crisis, Recent Security Incidents, and Escalating** **Protection Risks**\n\nYemen’s humanitarian situation is heavily impacted by ongoing security\nincidents, which hinder operational effectiveness, coupled with a banking crisis\nthat has led to severe disruptions and in some cases, funding withdrawal.\n\nThe ongoing banking crisis is placing additional strain on households and\nheightening social tensions, which in turn exacerbates protection risks. This is\nparticularly evident in increased tension among the affected population and\ncommunity networks, increase in incidents of abuse and exploitation, limited\naccess to essential services by vulnerable groups, as reported by protection\npartners.\n\n### **IV. Exacerbated Protection Risks**\n\nProtection partners have reported a significant rise in unmet basic protection\nneeds, as a direct consequence of halted and/or reduced protection services.\nThe delay or suspension of these services has heightened risks for vulnerable\ngroups, leaving them more exposed to exploitation, abuse and neglect.\n\n### **V. The Impact on Protection Activities**\n\nThe banking crisis, combined with the associated deterioration in the security\nsituation, have impacted the implementation of protection programming\nthrough:\n\n1. Reduction, delay and Suspension of critical protection services\n2. Authorities’ increased interference in Protection Programs\n3. Freedom of movement restrictions, increased risks to staff safety\nincluding risk of detention and increase of tensions with the\ncommunities.\n\n\n4\n\nMention: \"Financial Tracking Services\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its figures indicate the low share of funding for protection.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000654:4:1", "occurrence_id": "2312dd68a514a72567841226", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  often not disaggregated by refugee status and difficult to keep\ntrack of. At the same time, most common national surveys or censuses that include education-related\nquestions do not include specific information on refugee status. As a result, refugee children are often\nabsent or not fully represented in the statistical picture when looking at common national education\ndata sources.\n\n\nUNHCR collects information on education from different sources, which include the formal registration\nprocess, administrative data and household surveys. Such activities require adequate financial\nresources as well. In line with the global thinking and with its Data Transformation Strategy 2020-25,\nUNHCR is continuously strengthening efforts to generate and use refugee education-related data.\n\n\n5 UNHCR / October 2020\n\nMention: \"UNHCR\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Gathers education information from several sources.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000654:4:1", "occurrence_id": "4f5a869ebf17134bac224867", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  often not disaggregated by refugee status and difficult to keep\ntrack of. At the same time, most common national surveys or censuses that include education-related\nquestions do not include specific information on refugee status. As a result, refugee children are often\nabsent or not fully represented in the statistical picture when looking at common national education\ndata sources.\n\n\nUNHCR collects information on education from different sources, which include the formal registration\nprocess, administrative data and household surveys. Such activities require adequate financial\nresources as well. In line with the global thinking and with its Data Transformation Strategy 2020-25,\nUNHCR is continuously strengthening efforts to generate and use refugee education-related data.\n\n\n5 UNHCR / October 2020\n\nMention: \"administrative data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"One of the information sources on education.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000654:4:1", "occurrence_id": "99e730743e36e4a1777de623", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  often not disaggregated by refugee status and difficult to keep\ntrack of. At the same time, most common national surveys or censuses that include education-related\nquestions do not include specific information on refugee status. As a result, refugee children are often\nabsent or not fully represented in the statistical picture when looking at common national education\ndata sources.\n\n\nUNHCR collects information on education from different sources, which include the formal registration\nprocess, administrative data and household surveys. Such activities require adequate financial\nresources as well. In line with the global thinking and with its Data Transformation Strategy 2020-25,\nUNHCR is continuously strengthening efforts to generate and use refugee education-related data.\n\n\n5 UNHCR / October 2020\n\nMention: \"household surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Another information source, requiring financial resources.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000661:0:3", "occurrence_id": "074d6ef2cd0ddc4a1c4d4093", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  droit de l’homme, perpétrés contre<br>des civils qui ont poussé des milliers de personnes à se déplacer. Au 30 juin 2019, on compte 50 643 PDIs (49% d’hommes et 51% de femmes) dont 76% vivent avec<br>les communautés hôtes et 24% sur les sites spontanés, ainsi que 41 108 PDIs retournés (48% d’hommes et 52% de femmes), et 4 374 rapatriés du Burkina Faso, de<br>la Mauritanie et du Niger. L’insécurité alimentaire touche 8% de la population (soit 217 000 personnes). Près de 154 00 élèves n’ont pas accès à l’école dans la région<br>de Mopti en raison du nombre élevé d’écoles non fonctionnelles. Par ailleurs, 12,6% des localités disposent de CSCom (dans la localité) tandis que 24,2% des localités<br>sont à plus de 15 kms des CSCom. Les principaux défis humanitaires de la région sont : (i) le contexte sécuritaire qui s’est énormément dégradé en raison de la<br>recrudescence des conflits intercommunautaires; (ii) les mouvements forcés de population et les difficultés d’accès aux services sociaux de base ; (iii) une<br>augmentation des besoins humanitaires (Source :Profil humanitaire de la région de Mopti, février 2019).|La**région de Mopti** a été mar\n\nMention: \"Profil humanitaire de la région de Mopti\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Ses constats décrivent l'augmentation des besoins humanitaires de la région.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000662:4:0", "occurrence_id": "952481c4d50410258d568707", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: frequency include seeking alternative/additional jobs (37%),\nlimiting portion size at meals (32%), removing children from\nschool (16%), skipping meals for an entire day (11%) and selling\ndomestic assets (5%).\n\n\n**Best practices by IP staff:**\n\n\nWFP, with its cooperating partners, has adopted a number of\nmeasures to strengthen gender mainstreaming across all levels\nof its NWA response. Efforts are focused on creating an ena‐\nbling environment for promoting gender equality and women’s\nempowerment in food and nutrition security. These measures\ninclude the provision of female‐exclusive entrances and waiting\nareas for female beneficiaries and also the presence of female\nstaff at distribution points, all of which were clearly noted by\nfemale beneficiaries during the course of this assessment.\n\n\n\non a one‐week recall by the respondents on the food con‐\nsumed within the household, and using standard thresholds,\nhouseholds are classified into one of the three groups, namely,\n“acceptable” food consumption, “borderline” food consump‐\ntion, and “poor” food consumption. From the analysis of the\nfood consumption data from the household survey (Figure 6),\noverall, 41.6 percent of the households were found to have\n“acceptable” food consumption, while 6.6 percent fell in the\n“poor” consumption, while the majority (51.8%) were shown to\nhave “borderline” levels of consumption.\n\n\n**Figure 7: Negative coping strategies**\n\n\nOne fourth of the respondents have engaged in negative coping\nstrategies to manage their daily food needs. Among those, the\nmost common ones (Figure 7) include skipping meals (63%),\nrelying on less preferred foods (58%), and restricting consump‐\ntion by adults to ensure their children eat properly (42%). Oth‐\ner negative coping strategies, though seen with somewhat less\n\n\n5\n\n\n\n\n\n**Overall satisfaction:**\n\n\nFemales were\n\nMention: \"food consumption data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Drawn from the household survey, they rank consumption in three categories.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000667:6:0", "occurrence_id": "2474120d6a9f06ba472b17ac", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: HCT-coordinated response: Protection Analysis Update\n\nJune 2022\n\n\nOn 30 March, a new Law No. 16 of 2022 was issued criminalizing torture. The law stipulates imprisonment for anyone that deliberately perpetrates\ntorture. While the law criminalizes torture and assigns a penalty of at least three years’ imprisonment, it was widely criticized given how pervasive\nthe reported widespread and systematic use of torture is by parties of conflict as well as the failure to mention any measures that could be taken\nto prevent torture or offer redress to past victims of torture.\n\n\nRestrictions on freedom of speech and the press have been a major characteristic of all parties to the conflict Prior to the start of the conflict, a\n2010 Press Freedom Index ranked Syria 173 <sup>rd</sup> out of 178 nations. The issuing of a new Law No. 15 of 2022 amended a number of articles of the\nPenal Code that restricted freedom of speech and criminalized “whoever announces news that aims to prejudice the Syrian state by improving the\nimage of a hostile country”, casting further doubts about positive changes.\n\n###### **Almost complete erosion of civilian population resources and opportunities**\n\n\nSyria’s GDP plummeted from 252.52 billion in 2010 to 22.78 billion in 2019 <sup>5</sup> [^5: World Bank, 2022.], transitioning uncontrollably from a middle-income to a low-income\ncountry. In 2021, both macro-economic conditions, and the affordability of basic commodities to the population, have worsened and continue\ndeteriorate\n\n\nThe Syrian pound is steadily depreciating and lost close to 80 per cent of its value in 2021 alone. The reference food basket price was marked at\nthe highest level for six consecutive months (since 2013), reaching 231,004 SYP (around +200 per cent compared to 2021), mainly due to fuel\nshortage, increase in global food prices and inflation. Approximately 90 per cent of the population now lives below the poverty line <sup>\n\nMention: \"2010 Press Freedom Index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its ranking placed this country at the stated position before the conflict.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000672:7:1", "occurrence_id": "c894dee892bba8b558a02cee", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \nsaberes dinámicas que estimularon la\n\nreflexión y la identificación de factores\n\nde riesgo, como los elementos identi\nficadores de este delito, solicitado por\n\nlos funcionarios en un diagnóstico de\n\n\n\nnecesidades formativas llevado a cabo\n\nprevio a la organización del evento.\n\n\n**La Organización Internacional para**\n\n**las Migraciones (OIM),** entre mayo y\n\njulio de 2022, en materia de prevención\n\nde la trata de personas, sensibilizó a\n\n\n#### **Boletín trata de personas**\n##### Venezuela\n\n54,124 personas en comunidades, y se\n\ncapacitaron a 734 personas integrantes\n\nde ONG y 1,283 funcionarios y funcio\nnarias de entes gubernamentales sobre\n\ndetección, identificación, derivación y\n\nasistencia a víctimas de trata de per\nsonas, en los distintos proyectos que\n\nejecuta la OIM.\n\n\nEn el marco del Dia Mundial de la Trata\n\nde Personas, la OIM efectuó 38 activi\ndades de sensibilización, orientaciones\n\ninformativas, jornadas de capacitación,\n\ncine foros, distribución de materiales\n\npreventivos en los estados Amazonas,\n\nApure, Bolívar, Falcon, Táchira, Sucre y\n\nZulia, del 25 al 30 de julio de 2022, en\n\nlos ámbitos comunitario, educativo, insti\ntucional y social.\n\n\nOtro logro significativo para la OIM fue\n\nel lanzamiento de una campaña de ori\nentación y prevención de la TdP en el\n\nMention: \"Boletín trata de personas\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Rinde cuentas de las actividades de prevención y sensibilización.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000678:11:1", "occurrence_id": "27f02630fccc868e63d171d5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  pour renforcer la redevabilité de la réponse\nhumanitaire aux communautés affectées.\n\n- L’appui et le renforcement des capacités des structures nationales et locales et le travail avec les\ncommunautés locales pour réduire les risques de protection\n\n- Le soutien financier et humain en faveur du secteur protection afin de mettre en œuvre les projets approuvés\ndans le plan de réponses humanitaires.\n\n- La réponse adéquate aux recommandations émanant des matrices régionales bimensuelles de protection par\nles autorités civiles et militaires et la MINUSMA\n\nLes activités prévues sont les suivantes :\n\n- Formations des membres du cluster sur les principales stratégies de plaidoyer et de gestion efficace de\nl’information\n\n- Plaidoyer pour la sécurisation des zones nécessitant le renforcement de l’environnement protecteur\nauprès des autorités, de la coordination humanitaire et de la MINUSMA\n\n- Analyses et interventions régulière dans des réunions de l’Inter-Cluster, CIMCOORD, l’EHP, Protection\ndes Civils de la MINUSMA et autres instances de coordination\n\n- Réunions, négociations et activités de plaidoyer avec des autorités militaires et civiles et avec d’autres\ngroupes cibles\n\n- Emission des documents publics de plaidoyer\n\n- Ateliers de formation pour les acteurs humanitaires sur la protection transversale et autres aspects de\nprotection\n\n- Contribution à l’élaboration de la stratégie de la MINUSMA pour la Protection des Civils etc.\n\n## **VIII. Mécanisme de coordination du cluster, suivi\n\nMention: \"matrices régionales bimensuelles de protection\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Leurs recommandations appellent une réponse des autorités et de la mission.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000681:2:15", "occurrence_id": "903e6aca8e3b22ae34d26214", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .6<br>20.0<br>20.8|<br>20.4<br>24.6<br>23.6<br>20.0<br>20.8|<br>20.4<br>24.6<br>23.6<br>20.0<br>20.8|<br>20.4<br>24.6<br>23.6<br>20.0<br>20.8|<br>20.4<br>24.6<br>23.6<br>20.0<br>20.8|||||\n||||||||||||||||||||||||||||||\n||||||||||||||||||||||||||||||\n\n\nAfghanistan Tajikistan Kyrgyzstan India Turkmenistan Bhutan Sri Lanka\nPakistan Nepal Bangladesh Uzbekistan Maldives Kazakhstan Iran\n\n\nSources: ALCS 2013–14 and UNDESA (2015)\n\n\n\n3\n\nMention: \"ALCS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Source des effectifs de population pour les pays listés.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000681:2:15", "occurrence_id": "702e99b667c754410b644a57", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .6<br>20.0<br>20.8|<br>20.4<br>24.6<br>23.6<br>20.0<br>20.8|<br>20.4<br>24.6<br>23.6<br>20.0<br>20.8|<br>20.4<br>24.6<br>23.6<br>20.0<br>20.8|<br>20.4<br>24.6<br>23.6<br>20.0<br>20.8|||||\n||||||||||||||||||||||||||||||\n||||||||||||||||||||||||||||||\n\n\nAfghanistan Tajikistan Kyrgyzstan India Turkmenistan Bhutan Sri Lanka\nPakistan Nepal Bangladesh Uzbekistan Maldives Kazakhstan Iran\n\n\nSources: ALCS 2013–14 and UNDESA (2015)\n\n\n\n3\n\nMention: \"UNDESA\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Complète les effectifs pour les mêmes pays.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000689:2:1", "occurrence_id": "32269a80a5af4cc2a6bf8e68", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  commonly reported protection issue affecting**\n**women and girls**, cited by 41% of respondents overall. The level of reporting for this issue did not evolve significantly\nthroughout the entire monitoring period and remained higher in camps (49%) than in out-of-camp locations (40%). In\naddition, **lack of access to sexual and reproductive health services is reported as the fourth most common protection**\n**issue affecting women and girls**, cited by 23% of respondents overall. Of particular concern, the proportion of respondents\nin camps reporting this issue increased from 7% to 34% between Round 1 and Round 6.\n\n**The lack of safe space and privacy remains the third main protection issue affecting women and girls**, cited by 38% of\nrespondents overall. This proportion is similar in camp and out-of-camps locations and did not significantly evolve\nthroughout the monitoring period.\n\n**Violence and abuse within the household is the fifth most commonly reported issue**, cited by 18% of respondents overall\nand at similar levels in camp and out-of-camp locations. It is worth nothing that the level of reporting for Round 6 was\nlower than for Round 1, where it was cited by 27% of respondents. However, given complexity and sensitivities of domestic\nviolence, as well as various methodological limitations, this trend should not be interpreted as definitive indication that\ndomestic violence has actually reduced. <sup>2</sup> [^2: Protection monitoring at the community level only collects information about the perceptions and experiences of Key Informants and, unlike other\ndata collection mechanism such as the GBVIMS, does not collect reliable actual incidents of gender-based violence. With only 19% of KIs consisting\nof women, there is a significant limit to the effective representation of women’s experiences in the results of the monitoring.]\n\n2.3. <u>Child Protection</u>\n\n\n**Similarly, the majority of respondents reported that protection issues affecting children have significantly\n\nMention: \"GBVIMS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Unlike the mechanism described, it collects real incidents.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000690:4:1", "occurrence_id": "846247c2b5abc7b9ac03bc4c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  civilian infrastructure and basic\nstructures (such as health care and education), lack of civil documentation, loss of livelihood\nopportunities and property, and human rights and international humanitarian law violations which in\nturn further exacerbate underlying structural vulnerabilities.\n\n\nIn this regard, the Government presented the Annual Contingency Plan and the interventions\nforeseen to provide protection and assistance to affected populations, including (but not limited to)\nfood assistance, delivery of non-food items and dignity kits, and provision of shelters and\nrehabilitation of damaged infrastructures.\n\n\n**2.** D. Eckstein, V. Künzel, L. Schäfer, Global Climate Risk Index 2021, Germanwatch e.V., January 2021. Available at: <u>[Eckstein, V. Künzel, L. Schäfer,](https://www.germanwatch.org/sites/germanwatch.org/files/Global%20Climate%20Risk%20Index%202021_1.pdf)</u>\n<u>Global</u> <u>Climate</u> <u>Risk</u> <u>Index</u> <u>2021,</u> <u>[Germanwatch](https://www.germanwatch.org/sites/germanwatch.org/files/Global%20Climate%20Risk%20Index%202021_1.pdf)</u> <u>e.V.,</u> <u>January</u> <u>2021.</u> <u>Available</u> <u>at:</u>\n<u>[https://www.germanwatch.org/sites/germanwatch.org/files/Global%20Climate%20Risk%20Index%202021_1.pdf. .](https\n\nMention: \"Global Climate Risk Index 2021\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its index feeds the thinking on risk reduction.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000691:4:3", "occurrence_id": "a7f7f84055765a72e885c4c8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  protection services in 2024 in points of entry and adjacent reception and transit centres, and\nrefugee camps across the country.\n\n\n**1.** **Points of Entry**\n\n\nAs of the end of February 2024, South Sudan has recorded almost 588,000 arrivals at its points of entry, with the majority\n(79%) being South Sudanese returnees. However, reports indicate an increase in Sudanese refugees entering South Sudan\ncompared to the conflict's onset in April. Notably, 97% of all refugees and asylum-seekers in South Sudan originate from Sudan\n(UNHCR population update, 3 March 2024). <mark>Furthermore, by the end of 2023, an estimated 70,000 South Sudanese from</mark>\n<mark>Ethiopia have returned to South Sudan due to insecurity in Ethiopia.</mark>\n\n\nPage 5\n\nMention: \"UNHCR population update\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its updates give the origin of refugees and asylum seekers.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000692:2:0", "occurrence_id": "ad367a0b881d36909d146e0b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **GBV IMS Steering Committee and GBV IMS Task Force Lebanon**\n\n\nviolence, including rape and sexual assault present the third most reported type of GBV incident with 17\npercent of all incidents reported.\n\n\n**_Intimate Partner Violence (IPV)_**\n\n\n**62 percent** of all GBV incidents were perpetrated by the intimate partner of the survivor which is an\nincrease of IPV by 5 percent compared to Q2 2022. Data extracted from the GBVIMS indicate that both\nadults and children are subjected to IPV, with **11 percent** of female children reporting being subjected to\nviolence by their partner, including married children/ adolescent girls exposed to IPV, marking a 4 percent\nincrease compared to the second quarter of 2022.\n\n\n**_Forced/ Early Marriage_**\n\n\nEarly marriage is the most prominent type of GBV perpetrated against children, constituting **84 percent**\nof the forced marriage incidents reported in this quarter. According to the gender findings of UNICEF 's\n<mark>recent report</mark> <sup>8</sup> [^8: United Nations Children’s Fund (2022). “Deprived Childhood”. Accessed at:\n<u>[https://www.unicef.org/lebanon/reports/deprived-childhoods](https://www.unicef.org/lebanon/reports/deprived-childhoods)</u>] <mark>on multidimensional child poverty, adolescent girls, regardless of their nationality, shared</mark>\n<mark>experiences of harassment in public spaces and restriction of their mobility preventing them from accessing</mark>\n<mark>services. Additionally, 22 percent of Syrian girls aged 15-18 who do not attend school report marriage as the</mark>\n<mark>main reason behind this. The same report reveals that girls’ right to play is further restricted due to harassment</mark>\n<mark>and gender norms, as they are more likely to be confined to the home\n\nMention: \"GBVIMS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its extracts give the share of affected adults and children.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000695:3:0", "occurrence_id": "144d9b0b7ca00cc96c1fa946", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Access to decent work:** The key barrier to accessing work, according\nto respondents who are unemployed, is a lack of knowledge of the Czech\nlanguage (59%). This finding was also reflected in focus group\ndiscussions conducted by UNHCR with Ukrainian refugee women and\nmen in Brno, Ostrava in Prague in August and September, where\nrefugees reported challenges in securing jobs that matched their\nqualifications due to language barriers. All participants in the focus group\ndiscussions emphasized the need for better language proficiency to\nimprove their employment prospects.\n\n\n\nof Protection\nMonitoring\nrespondents of\nworking age are\nemployed in host\ncountry\n\n\n\nNevertheless, 67% of respondents reported being\nemployed compared to 60% in Q1. . <sup>4</sup> Moreover,\nrespondents’ access to earnings also increased\nfrom 87% to 93% during the same time period.\nWorking excessively long hours also decreased\nfrom 24% to 16%. These findings ostensibly\nindicate an improvement in access to employment\nand in working conditions amongst Ukrainian\nTemporary Protection holders.\n\nEmployment continued to be mentioned as one of\nthe top two priority needs (32%) of households\ninterviewed.\n\nMoreover, in terms of information, 29% of respondents indicated a need for more information\nabout job opportunities; this continued to be the highest reported information need.\n\n\n4 As per the 2024 Socio-Economic Insights Survey conducted by UNHCR, this figure could be as high as 79% amongst working-age\nUkrainian refugees. The data was collected in the period 15 May-16 July.\n\n\nUNHCR / September 2024 4\n\nMention: \"2024 Socio-Economic Insights Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its results put the affected share up to 79 percent of working age.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000706:4:1", "occurrence_id": "795385117f97dfe3cc0815a6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \nproper living premises for IDPs who protected the independence, sovereignty and territorial\nintegrity of Ukraine. The subvention is directed towards the payment of monetary compensation\nfor the acquisition of living premises in residential buildings on the primary and secondary real\nestate markets or for investment in residential buildings. Under paragraph 2 of the Procedure,\nthe right to receive monetary compensation is given to IDPs who protected the independence,\nsovereignty and territorial integrity of Ukraine and took direct part in the ATO, in ensuring its\nimplementation, being directly in the ATO areas when it took place, and are recognized as\npersons with disabilities as a result of the war of the 3rd category under paragraphs 11–14 of\npart 2 of Article 7 or as combatants under paragraphs 19 and 20 of part 1 of Article 6 of the\nLaw of Ukraine “On the Status of War Veterans, Guarantees of Their Social Protection” and\nwho are registered as citizens requiring the improvement of living conditions (apartment register). The monetary compensation of IDPs, who have the status of ATO participant, is paid in\nfull in the order of registration in the apartment register. To address the relevant issue, rayons,\ncity-rayonses (except for the city of Kyiv), and city councils form committees that verify facts\nand documents, as well as decide the allocation or refusal of monetary compensation and\nthe amount. The list of documents to be submitted for the application for the allocation of\nmonetary compensation is provided for in paragraphs 6 and 7 of the Procedure.\n\n\n6 CMU Decree No. 280 as of 18.04.2018 - “On Provision of Housing for IDPs that Protected the Independence, Sovereignty and Territorial\nIntegrity of Ukraine” http://zakon3.rada.gov.ua/laws/show/280–2018-%D0%BF\n\n\n#### **_3.1. Donetsk oblast_**\n\n\n\n|Number of IDPs<br>registered in the oblast<br>\n\nMention: \"apartment register\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Registers citizens needing improved living conditions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000707:0:0", "occurrence_id": "309a9bcbb644a310098c3787", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Based on observations and feedback from assisted families and communities, dual currency disbursement in May 2023 was\nmuch more efficient than LBP-only disbursement in the first 4 months of the year. It has also allowed refugees receiving cash\nassistance to access a more appropriate transfer value.\n\n\nFamilies were able to access their assistance in a quicker, safer, and overall, more dignified manner. Significant reductions in\ncrowding, waiting times, and tensions at redemption points also had positive impacts on the wider community. Access to dual\ncurrency redemption also benefitted retailers and minimal impact is expected on the wider Lebanese economy.\n\n\nFeedback for this report was gathered from various sources¹ that all point to the significant improvements as a result of the\nre-introduction of dual currency disbursement.\n\n\n\n**I.** **Background: Increasing Operational**\n**Challenges**\n\n\nIn the beginning of 2023, operational challenges were\nincreasingly threatening the sustainability of delivering cash\nassistance to refugees in local currency. The required\nincrease of transfer values in February and March to\nmaintain the level of assistance in a context of high inflation\nand currency depreciation, led to widespread occurrences of\nredemption points running out of cash. As a result,\ncrowding, waiting times, and cost to redeem increased\nsignificantly for assisted families. These challenges in turn\nled to more visibility, tensions, and harassment at\nredemption points.\n\n\nThe transfer value increases in February and March were not\nenough to offset the additional inflation caused by recordhigh currency depreciation and price increases in local\ncurrency between January and April 2023. <sup>2</sup> As a result,\ntransfer values only covered 41 and 22 percent of food and\nnon-food needs in April 2023.\n\n\nLoss of purchasing power had an impact on households’\ncapacity to meet food and other basic needs. WFP-UNHCR\nBasic Needs outcome monitoring conducted in March 2023\nshowed that 68 percent of assisted households did not have\nthe economic capacity to meet their basic needs, up from 62\npercent in June 2022\n\nMention: \"WFP-UNHCR\nBasic Needs outcome monitoring\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its March monitoring gives the share of households with no basic need covered.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000716:45:0", "occurrence_id": "3bfbcaf1da8f5b4a1812fe5b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: DTM Mali – Décembre 2020\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n|Région-Cercle-Commune|Ménage|Enfant ( < 18 ans )|Col4|Col5|Adulte (18 - 59 ans )|Col7|Col8|Personne âgée ( > 59 ans )|Col10|Col11|Femme|Homme|Individus|\n|---|---|---|---|---|---|---|---|---|---|---|---|---|---|\n|<br>**Région-Cercle-Commune**|<br>**Ménage**|**F **|**M **|**Total Enfant**<br>**( < 18 ans )**|**F **|**M **|**Total**<br>**(18 - 59 ans )**|**F **|**M **|**Total**<br>**( > 59 ans )**|**Total**<br>**( > 59 ans )**|**Total**<br>**( > 59 ans )**|**Total**<br>**( > 59 ans )**|\n|Tessit|988|1761|1350|3111|1350|1233|2583|118|58|176|3229|2641|5870|\n|Tin Hama|541|954|732|1686|732|668|1400|64|31\n\nMention: \"DTM Mali – Décembre 2020\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Dénombre les ménages par âge dans les subdivisions administratives.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000720:2:1", "occurrence_id": "780f932506a33eb3dea0c89c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  reports, providing strategic directions to SGBV programmes\nbased on identified gaps and trends.\n\n\nIt is important to highlight that the data and trends noted in this report are\nnot representative of the prevalence of SGBV in Jordan (or among refugee\npopulations) as these trends are based solely on incidents reported by\nsurvivors to the Data Gathering Organisations (DGOs) engaged in SGBV\nresponse and using the GBVIMS in 2019. It is accordingly not advisable to\nuse these findings as a proxy for the prevalence of SGBV in any settings or\nto use it in isolation to monitor the quality of programmatic interventions.\nDespite the above limitations, the GBVIMS is considered the highest\nquality SGBV incident data currently available to the humanitarian\nactors, which can be used effectively for trend analysis and improving\ncoordination of SGBV prevention and response.\n\n\n\nAlthough we registered an increase in the percentage of Non-Syrian\nrefugees assisted, it still remains low, albeit not necessarily indicating\nSGBV does not happen within these communities but rather a need to\nincrease outreach to share information about services available as well as\ntheir inclusion in participating within SGBV programmes.\n\n\n**It is important to underline that the majority of**\n**survivors reached services more than one month**\n\n**after the incident (69% in 2019 compared to**\n**71.5% in 2018). This trend has been a constant**\n**trend over the last three years.**\n\n\nFinally, it is important to underline that the majority of survivors reached\nservices more than one month after the incident (69% in 2019 compared\nto 71.5% in 2018). This trend has been a constant trend over the last\nthree years and this indicates the need to explore innovative approaches\nfor community-based outreach efforts to inform refugees about services\navailable for survivors and the importance of seeking timely assistance in\nparticular for survivors of sexual violence.\n\n\nTIME BETWEEN INCIDENT AND DISCOLURE\n\n\n2018 2019\n\n\n\n**It\n\nMention: \"GBVIMS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Its data are not recommended as a substitute for findings.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000733:4:2", "occurrence_id": "9e73a1300b96c9cbca662999", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \nNSAG. As of September 2022, a total of **2,160 events of violence**\nhave been reported in Somalia and Somaliland during the course of\nthe year through the Armed Conflict Location & Event Data Project\n\n\n7 Somalia National Bureau of Statistics, 2022\n8 Somalia Registered Refugees and Asylum Seekers, UNHCR Somalia\n\n5\n\nMention: \"Armed Conflict Location & Event Data Project\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Ses relevés documentent les incidents signalés au cours de l'année.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000752:1:2", "occurrence_id": "5efb9087a3635ae8e02e3782", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: * de los niños, niñas y adolescentes (NNA) que integraban los grupos\nentrevistados no estaban acompañados o estaba separados de su padre, madre o cuidador legal,\nespecíficamente: **7%** niños no acompañados y **2%** niños separados. Cabe señalar que, según los datos\noficiales de INM, del total de 12,574 personas registradas, 2,978 fueron NNA y 100 de estos fueron registrados\ncomo NNA no acompañados (3% del total del NNA). El equipo de monitoreo tuvo la oportunidad de realizar\nentrevistas focalizadas a 21 de esos 100 NNA no acompañados registrados por las autoridades, lo que explica\nque la proporción de los mismos en la muestra sea superior a la registrada por el INM.\n\n\nwww.acnur.org/mexico 2\n\nMention: \"datos\noficiales de INM\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Registran el total de personas y la cifra de niños no acompañados.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000754:5:0", "occurrence_id": "f5db7e61f49d84cdd1f68b53", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: MOVEMENT RESTRICTIONS AND CORRELATION WITH ACCESS\nTO EDUCATION\n\n\n\nMovement restrictions in the context of violence in the Central\nSahel go beyond the common association where movements\nare limited / restricted towards displaced and refugee\npopulations. Considering that the majority of respondents\nparticipating in P21 surveys consist of local and host\ncommunities (64%), followed by IDPs (33%), the restriction of\nmovement sheds light on the plight of local populations who\nface limitations in their mobility within and outside their homes\nor communities for a week or more due to armed conflict and\nwidespread violence.\n\n\n### **70%**\n\n**OF CHILDREN**\n\n**FACE MOVEMENT RESTRICTIONS**\n**TO GO TO SCHOOL**\n\n\n\nThis dire situation deprives them of accessing essential goods and services, including markets, food, education,\nhealthcare, water, sanitation, and livelihood opportunities. Consequently, entire communities are subjected to fear,\nanxiety, as they are unable to freely move around due to the constant threat of armed confrontations, explosive devices,\nand the presence of hostile actors engaged in armed conflict.\n\n\nWithin the Central Sahel region, P21's findings demonstrate that movement restrictions and confinement have a direct\nimpact on access to education. School enrollment rates decrease proportionally when these limitations are imposed.\nChildren are forced to discontinue their education, unable to avail themselves of child services and support. Driven by\nthe fear of abduction, rape, or even death at the hands of warring parties, these children and their families distance\nthemselves from normal life. The situation is exacerbated by the considerable distances they must traverse to reach the\nnearest school.\n\n\nIn 2022, a staggering 60% of respondents confirmed that movement restrictions and the consequences of armed conflict\nwere the primary obstacles preventing children from attending school in the Central Sahel. For formal primary enrolment,\nrates are 40% lower among children who face movement restrictions and they are only half as likely to be able to attend\nKoranic schools.\n\n\n\nReasons for school dropout\n\nMention: \"P21 surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"The sample composition limits the reach of the conclusions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000713:25:0", "occurrence_id": "5ca6dc3a2ff013c884487d47", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: borrowing and lending.\n\n\nOn the negative side, some groups reported that the financial assistance made their\nJordanian neighbors feel jealous, since Jordanians cannot get such assistance when\nthey are poor or out of work. Some of the refugees met during focus group\ndiscussions were unable to focus on the topic at hand, often because the presence of\nUNHCR staff made them focus on issues of financial assistance and resettlement.\n\n\nIn both these areas, some refugees perceive unfairness in the distribution of these\nvalued resources. Nevertheless, the evaluation of the cash assistance program showed\nthat the current beneficiaries are highly satisfied with this form of assistance. <sup>1</sup> It is\nnon-recipients and former recipients who appear to have concerns about the\nprogram’s fairness.\n\n\nThis perception of unfairness is also evident in the increased number of complaints\nUNHCR received after re-assessing cases receiving financial assistance. The number\nof written complaints increased over 100% with the introduction of re-assessments of\neligibility for financial assistance, from an average of 97 letters per month to 200\nletters per month.\n\n\nVarious letters and interviews reveal complaints that other persons receiving\nassistance are not deserving of assistance, while they are. In some cases, people\ndenounce others by name. Other studies suggest that means-testing, such as UNHCR\ndoes in establishing eligibility for financial assistance, undermines trust in institutions\nand contributes to weaker overall social trust. <sup>2</sup>\n\n\nSuggested activities:\n\n\n    - Consider financial and social impact of establishing eligibility for\nfinancial assistance using open criteria, so that persons can be\ndeemed eligible or ineligible in a more transparent way.\n\n\n_g) Focus on building social ties between non-Iraqis and Jordanians_\n\n\nNon-Iraqis have weak ties to Jordanian society. Some speak poor Arabic and have\nfew opportunities to interact socially with Jordanians. Racial discrimination\nmarginalizes the non-Iraqis.\n\n1 See “Cash Assistance Impact Surveys in Jordan and Syria,” UNHCR Jordan,\n\nMention: \"Cash Assistance Impact Surveys in Jordan and Syria\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its results concern the effects of cash assistance.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000762:9:2", "occurrence_id": "f28af9e2fdbcacc6f0f9a23a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  . . . . . . . . . . . . . . . . . .\n\n1 Los valores equivalentes para Argentina eran, respectivamente, 120.000 en 2001 y 123.000 en 2010 (Martínez\nPizarro y Orrego Rivera, 2016: 20).\n\n2 Ver https://r4v.info/es/situations/platform.\n\n3 Ver https://data2.unhcr.org/es/documents/download/68962\n\n4 Ver http://argentina.iom.int/co/news/el-n%C3%BAmero-de-refugiados-y-migrantes-de-venezuela-alcanza-los-3-millones#overlay=node/384/edit\n\n5 El ACNUR realizó relevamientos denominados de “monitoreo de protección” en Argentina, Ecuador, Chile y Perú.\nPor su parte, la OIM realizó varias rondas de recolección de información en base a una Matriz de seguimiento del\ndesplazamiento (DTM, por sus siglas en inglés). Los detalles y resultados de estos relevamientos pueden consultarse\nen https://r4v.info/es/situations/platform\n\n\n\n**Capítulo 1** . Venezolanos/as en Argentina. Una aproximación a las cifras\n\n\ncas de la población venezolana en desplazamiento. Los informes indican\nque quienes han partido de Venezuela eran en su gran mayoría personas\njóvenes (menores de 44 años) que contaban con empleo en su país al\nmomento de partir. En términos generales, la proporción entre varones\ny mujeres tie\n\nMention: \"Matriz de seguimiento del\ndesplazamiento\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Sus rondas de recolección producen resultados publicados por separado.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000763:72:0", "occurrence_id": "076b230e7fdde939a303a2f7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Syrian refugees living outside camps in Jordan - Home visit data findings - 2013**\n\n\nAmman had a higher relative distribution of aggregate income earned through work (+10%) than\nwould be expected based on the number of refugees living there, while the relative distribution of\naggregate income in Irbid was much lower (-8%) relative to the total refugee population (Figure\n59). This is partly due to the fact that a larger proportion of refugee households in Amman reported\nearning money from work (31%) compared to households in Irbid (21%). Higher salaries in Amman\ncould also contribute to greater aggregate income.\n\n\n\n\n\n_Source: UNHCR /IRD Home Visits 2013_\n\n\n**2.5** **Income from pension**\n\n\n0.1% of all visited refugee cases reported income from a pension (Figure 61).\n\n\n\n\n\n\n\n\n\n_Source: UNHCR /IRD Home Visits 2013_\n\n\n\n78\n\nMention: \"UNHCR /IRD Home Visits\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its home visits give incomes of the refugee cases monitored.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000760:9:1", "occurrence_id": "12dd9f3f8df2cf36c4e89cbf", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: * Fan gasifier/\nfan jet\n\n**•** Char stoves\n\n\n\n**Renewable-**\n**fuel stoves****\n\nEnergy from\nrenewable sources\n\n\n**•** Biogas\n\n**•** Ethanol\n\n**•** Methanol\n\n**•** Solar ovens\n\n**•** High-efficiency\nbiomass stoves\nwith managed\nfuel supply\n\n**•** Renewably\nsourced\nelectricity\n\n\n\nUse fossil fuels or\nelectricity; high\nfuel efficiency;\nvery low CO and\nPM emissions\n\n\n**•** LPG and\ndimethyl\nether\n(DME)\n\n**•** Electric/\ninduction\n\n**•** Natural gas\n\n\n\n**Typical**\n**technologies/**\n**fuels**\n\n\n\n\n**•** Legacy\nbiomass\nand coal\nchimney\n\n**•** Basic\nefficiency\ncharcoal\nstove\n\n**•** Basic\nefficiency\nwoodstove\n\n\n\n\n- A rocket stove burns small-diameter logs in a combustion chamber that contains a vertical chimney.\n** The term ’renewable’ is context-specific; it can apply to solid biomass if the source is sustainably managed.\nSource: Definitions based on wording as published in ESMAP and GACC (2015), _The state of the global clean and improved cooking sector: Technical_\n_Report 007/15_, World Bank, p. 13.\n\n\nBased on data from the GACC stove catalogue, <sup>14</sup> [^14: catalog.cleancookstoves.org.] HEDON stove database, <sup>15</sup> [^15: www.hedon.info/Stoves+Database.] UNHCR reports, the\nDeutsche Gesellschaft für Internationale Zusammenarbeit (GIZ), USAID, Chatham House and\nmanufacturer websites, 47 cookstoves were reviewed against this typology. The resulting classification\nis shown in Table 3. Devices known to have been used in refugee operations are indicated in red text.\n\n\n13 ESMAP and GACC (2015), _The state of the global clean and improved cooking sector: Technical Report 007/\n\nMention: \"GACC stove catalogue\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its data feed the clean cooking sector analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000760:9:1", "occurrence_id": "afdcf0494ad10d26da80ac58", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: * Fan gasifier/\nfan jet\n\n**•** Char stoves\n\n\n\n**Renewable-**\n**fuel stoves****\n\nEnergy from\nrenewable sources\n\n\n**•** Biogas\n\n**•** Ethanol\n\n**•** Methanol\n\n**•** Solar ovens\n\n**•** High-efficiency\nbiomass stoves\nwith managed\nfuel supply\n\n**•** Renewably\nsourced\nelectricity\n\n\n\nUse fossil fuels or\nelectricity; high\nfuel efficiency;\nvery low CO and\nPM emissions\n\n\n**•** LPG and\ndimethyl\nether\n(DME)\n\n**•** Electric/\ninduction\n\n**•** Natural gas\n\n\n\n**Typical**\n**technologies/**\n**fuels**\n\n\n\n\n**•** Legacy\nbiomass\nand coal\nchimney\n\n**•** Basic\nefficiency\ncharcoal\nstove\n\n**•** Basic\nefficiency\nwoodstove\n\n\n\n\n- A rocket stove burns small-diameter logs in a combustion chamber that contains a vertical chimney.\n** The term ’renewable’ is context-specific; it can apply to solid biomass if the source is sustainably managed.\nSource: Definitions based on wording as published in ESMAP and GACC (2015), _The state of the global clean and improved cooking sector: Technical_\n_Report 007/15_, World Bank, p. 13.\n\n\nBased on data from the GACC stove catalogue, <sup>14</sup> [^14: catalog.cleancookstoves.org.] HEDON stove database, <sup>15</sup> [^15: www.hedon.info/Stoves+Database.] UNHCR reports, the\nDeutsche Gesellschaft für Internationale Zusammenarbeit (GIZ), USAID, Chatham House and\nmanufacturer websites, 47 cookstoves were reviewed against this typology. The resulting classification\nis shown in Table 3. Devices known to have been used in refugee operations are indicated in red text.\n\n\n13 ESMAP and GACC (2015), _The state of the global clean and improved cooking sector: Technical Report 007/\n\nMention: \"HEDON stove database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Completes the catalogue on cookstove characteristics.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000765:8:0", "occurrence_id": "0e575ef9054a227e08dd10cd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ### **Violations contre les droits des enfants**\n\nLa portée et la complexité des conflits, les déplacements et des catastrophes naturelles continuent d’avoir un\n\nimpact alarmant sur les enfants au Mali. Malgré les contraintes de financement et d’accès, 159,599 enfants vul\nnérables (49 % de filles) ont bénéficié d’interventions en santé mentale et psychosociale grâce à des partenariats\n\nefficaces et à une bonne coordination. Ce résultat est en ligne avec le résultat de l’année précédente, reflétant des\n\nniveaux de financement stables mais insuffisants et une population relativement constante d’enfants touchés sur\n\nla période de deux ans <sup>3</sup> [^3: Selon la Matrice de suivi des déplacements (DTM), 206,194 enfants ont été enregistrés comme déplacés en décembre 2023, contre 255,434 en décembre\n2022.] .\n\n\nLe domaine de responsabilité a continué à soutenir les enfants affectés par le conflit et le déplacement avec la\n\nprise en charge et/ou réunification familiale de 2,938 enfants non-accompagnés et séparés (39 % de filles) ainsi\n\nque des soins temporaires et/ou réintégration socio-économique à 596 enfants associés aux forces et groupes\n\narmés (18 % de filles). L’impact sur les enfants de la résurgence du conflit dans le nord depuis août 2023 est évi\ndent, entraînant des déplacements et des séparations familiales. Plus de 40% des 2,938 enfants affectés ont été\n\nidentifiés et assistés dans les régions de Kidal, Ménaka\n\nMention: \"Matrice de suivi des déplacements\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Ses enregistrements comparent les effectifs d'enfants déplacés sur deux ans.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000770:3:0", "occurrence_id": "f9ad9a864c295f0ca182c803", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ### Poland – Protection Sector & Information Protection Sector & Information Poland –\n##### ManM a nagement WG gement WG\n\n**Mental health and psychosocial support needs:** High levels of psychological distress associated\nwith the conflict and family separation are reported by the community. UNHCR reported that\n14% of queries at the blue dots in Warsaw and Krakow from 23 March to 6 May were related\nto psychosocial concerns. Caregivers were often seeking advice related to children changing\nbehaviours like frequent aggression and withdrawal. Emotional reactions for adults included\nanxiety attacks and sleeping disturbances. In general, refugees were affected by family separation, loss, fear and worry about the future. Older persons expressed a fear to die outside of\ntheir home country. According to IMC MHPSS assessment <sup>1</sup> [^1: <u>[https://data2.unhcr.org/en/documents/details/92210](https://eur02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fdata2.unhcr.org%2Fen%2Fdocuments%2Fdetails%2F92210&data=05%7C01%7Ckernin%40unhcr.org%7C0f8b504808b744554e6108da2f6ca752%7Ce5c37981666441348a0c6543d2af80be%7C0%7C0%7C637874440945190941%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=EG8ScWS36mhNh2V9gRsGpac%2BhCRmfSH%2FKfLXYeZnrdw%3D&reserved=0)</u>], stress factors included inability to\nmeet basic needs, lack of clear information, and lack of access to long-term accommodation.\nIdentified coping responses included the human interactions and new friendships, empathy and\nmutual support, religious practices, keeping active, and limiting the amount of time searching\nfor information and watching the news. Also, 9% of health issues identified by the medical team\n(INTERSOS) in Korczowa were related to mental health concerns. Of these, 58% concerned\nfemale patients aged 18-64 years old. Access to treatment for adults and children with preexisting severe mental health conditions and psychosocial\n\nMention: \"IMC MHPSS assessment\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its findings describe fears expressed by older people.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000774:10:0", "occurrence_id": "724adb6553755634b7077e96", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 4. Access to National Public Services\n\n\n### **5 Cross Sectors**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n**5.1** **Gender**\n\n\nDespite improvements in recent years, the normative framework does not yet fully ensure gender equity\nand access to basic services regardless of gender, including for refugees and asylum-seekers. Mauritania\nranks 141 out of 153 countries in the <u>[World Economic Forum’s 2020 Global Gender Gap Report. Economic](https://www.weforum.org/reports/gender-gap-2020-report-100-years-pay-equality)</u>\nparticipation and opportunities for women remain limited, with around 30 per cent of Mauritanian women\nparticipating in the labour force and a 43 per cent literacy rate. As mentioned above, the most consequential\npolicy sub-dimensions in terms of socioeconomic development are as follows:\n\n\ni. **Justice and Security**, the challenges faced to prevent and address gender-based violence;\n\n\nii. **Rights to Work and Rights at Work**, improving women’s access to decent, well-paid work;\n\n\niii. **Education**, improving female enrolment in primary and secondary education, particularly in Mbera\n\ncamp; and\n\n\niv. **Protection for Vulnerable Groups**, improving access to services for survivors of gender-based violence.\n\n\n**5.2** **Social inclusion**\n\n\nThe current implementation of policies in some policy sub-dimensions often does not enable the effective\nand full inclusion of refugees in national systems. While authorities have expressed a willingness to include\nrefugees in national systems, be they the labour market (GRF pledge) or the education system, some\npolicies and their implementation would require further amendments to operationalize these pledges. As\nmentioned above, the most consequential differences or restriction in terms of socioeconomic development\naffecting refugees with particular characteristics are as follows:\n\n\ni. **Education\n\nMention: \"2020 Global Gender Gap Report\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its ranking places this country at the stated position among 153 countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000775:12:0", "occurrence_id": "d12d46edffb8343c33758efd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ### **II. RESPUESTA (VII)**\n\n\n#### **Boletín trata de personas**\n##### Venezuela septiembre-noviembre 2023\n\n\n\ny Apure) donde se proyectó la película\n\n“Las Elegidas”. Además, aproximada\nmente 100 funcionarios/as del Estado,\n\ntanto de instituciones del sistema de\n\nprotección como del sistema educativo,\n\nfueron capacitados para la prevención,\n\nmitigación y atención de la trata de per\nsonas en varios estados del país (Sucre,\n\nDelta Amacuro y Amazonas durante el\n\ntrimestre septiembre-noviembre).\n\n\nEn cuanto a la asistencia, desde Tinta\n\nVioleta para ejemplificar la situación que\n\nviven las múltiples víctimas de trata, com\npartimos las historias de tres de ellas. En\n\nel Distrito Capital se ha proporcionado\n\nasistencia directa y diversos servicios\n\nmultisectoriales a una sobreviviente de\n\nTrata de Personas remitida por el centro\n\nde DDHH de la UCAB. Esta persona fue\n\nvíctima de explotación sexual en Perú,\n\nreclutada por una persona de su entorno\n\nfamiliar en mayo del presente año para\n\n\n\ntrata de personas en las comunidades\n\nen donde la organización interviene. De\n\nesta manera, se logró capacitar a 12\n\npersonas parte del equipo técnico de\n\nFUNDANA.\n\n\n_©CESVI.Capacitaciones sobre trata de_\n\n_personas a FUNDANA.(2023)_\n\n\n\n**Tinta Violeta** **_._** Como parte de la\n\nrespuesta, Tinta Violeta y las orga\nnizaciones aliadas en la Red\n\nMention: \"Boletín trata de personas\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Rinde cuentas de las actividades de respuesta en el periodo indicado.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000779:3:1", "occurrence_id": "fa53cf875341cc9b7a21b759", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |4%<br>3%<br>0%|\n|**Region**<br> <br>|<br> <br>|\n|Metropolitan<br>40% <br>69%|72% <br>61% <br>61%|\n|Rest of country<br>60% <br>31%|28% <br>39% <br>39%|\n|**Sex of head of household**<br> <br>|<br> <br>|\n|Male<br>51%<br>59%|59%<br>49%<br>76%|\n|Female<br>49%<br>41%|41%<br>51%<br>24%|\n|**Children in the household**<br> <br>|<br> <br>|\n|HHs with at least one child<br>55%<br>66%|67%<br>60%<br>60%|\n|<br> No children<br>45%<br>34%|33%<br>40%<br>40%|\n\n\n\n_Source: CASEN 2022._\n\nOverall, there are several differences between the refugee and migrant population and nationals in terms of\ntheir sociodemographic profile. Notable distinctions include age distribution, geographic concentration, and\nhousehold composition. Refugees and migrants are predominantly younger, with 67% between the ages of\n18 and 44, compared to just 39% of nationals—highlighting the need for youth-focused employment,\neducation, and integration policies. In terms of location, the majority reside in the Metropolitan Region,\nindicating a strong urban concentration that places additional demands on local infrastructure and public\nservices. Additionally, refugee and migrant households are more likely to include children: only 34% are\nchild-free\n\nMention: \"CASEN 2022\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Its results distinguish refugee and migrant populations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000781:1:0", "occurrence_id": "ea64d03ca431480d6839ade9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: programme withstanding the state hosts 82% of all IDP population across BAY States, with 214 IDP camps (62 formal\nand 152 informal sites) and 470 host community sites, with an estimated 436,323 HHs (2,084,063 individuals).\n\nThis year's rainy season continues to significantly affect people in camps and host communities, impeding\nhumanitarian aid access while increasing demand across identified flood-prone areas. According to Flood\nVulnerability Mapping analysis, an estimated 299,008 households (representing 623,007 individuals) reside in 148\nareas highly susceptible to flooding, necessitating urgent humanitarian assistance across Borno (162,571 HH),\nAdamawa (2,608 HH), and Yobe (2,027 HH). Additionally, 1,228 locations hosting 1,070,552 internally displaced\npersons (IDPs) and IDP returnees within host communities are also at risk.\n\n\nMap of flood and windstorm affected sites per LGA\n\n\nIn north-east Nigeria, there are approximately four million displaced persons (22 per cent IDPs in camps, 50 per cent\nreturnees, and28 per cent IDPs living within host communities) across the BAY states in need of shelter and NFI\nassistance. Of them, 65 per cent are classified as having severe needs, 32 per cent with extreme needs and four per\ncent have catastrophic levels of needs for Shelter and NFI assistance. Majority of these caseload have limited access\nto critical services.\n\nLimited access to site management responses, shelter and settlement solutions, sanitation facilities, and safe water,\nhas elevated protection concerns, with women, girls, and children constituting most of the displaced people caseload\n\nMention: \"Flood\nVulnerability Mapping analysis\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Estimates households residing in the identified flood zones.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000782:31:0", "occurrence_id": "c70a29529d1f40a4e77979a6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: livelihood strategies. Given that repatriates no longer need international protection\nagainst persecution, their need for continued mobility reflects their states’ continued\ninability to provide meaningful rights, particularly in socio-economic terms.\n\n\n171.   If repatriation is in any sense a viable option, such refugees are no longer\nconceptually distinct from other similarly deprived citizens of fragile states (apart\nfrom the fact they have already crossed an international border). This solution may\ntherefore be seen as a ”fairer” option in that it does not unnecessarily privilege\nrefugees’ access to mobile livelihoods over the access of other deprived and poor\ncitizens.\n\n\n172.   Secondly, a general liberalization of regional migration would better reflect\nboth the historical and future needs for mobility to be used as a means of facilitating\naccess to rights and resources. As Bakewell’s work in Zambia and Angola\n(Bakewell (2000)) and Monsutti’s work in Afghanistan (Monsutti (2006, 2008)) has\ndemonstrated, transnational approaches to livelihoods frequently pre-date the\ncreation of fixed nation-state boundaries.\n\n\n173.   This is particularly true in the global South where such borders were artificially\nimposed across far more persuasive tribal or ethnic “imagined communities”. This the\ncase, for example with the Pushtun on the Afghan-Pakistani border who despite being\n‘citizens of two distinct and often antagonistic states’ have retained close ties\n(International Crisis Group (2009): 17).\n\n\n174.   If effective durable solutions for refugees are to be implemented, these\nsolutions need to take account of existing cultural and socio-economic norms, and not\ntie repatriation to sedentary understandings of communities. Given the continuing\nmassive inequality between states and regions (for example, Afghanistan’s GDP in\n2008 was $10.2 billion, compared to Iran’s GDP of $385 billion (World Bank figures)),\nthere also needs to be a recognition that “sedentary” solutions are not likely to be\npossible in the short-\n\nMention: \"World Bank figures\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its figures compare the gross domestic product of the two countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000789:14:0", "occurrence_id": "e8b2a176e564e014be6a0f3c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Refugee-hosting countries_ refers to all countries hosting\nrefugees that appear on the DAC List of ODA-recipients\n[(http://www.oecd.org/dac/financing-sustainable-de-](http://www.oecd.org/dac/financing-sustainable-development/development-finance-standards/DAC_List_ODA_Recipients2018to2020_flows_En.pdf)\n<u>[velopment/development-finance-standards/DAC_](http://www.oecd.org/dac/financing-sustainable-development/development-finance-standards/DAC_List_ODA_Recipients2018to2020_flows_En.pdf)</u>\n<u>[List_ODA_Recipients2018to2020_flows_En.pdf).](http://www.oecd.org/dac/financing-sustainable-development/development-finance-standards/DAC_List_ODA_Recipients2018to2020_flows_En.pdf)</u>\n\n###### Methodology\n\n\n**Computation Method:**\n\n\n**Disaggregation:** This indicator can be disaggregated by type of aid, donor, recipient country and by\ngeographical region and sub-region.\n\n\n**Methods and Guidance:** OECD Survey on Financing\n[for Refugee-Hosting Contexts: http://www.oecd.org/](http://www.oecd.org/dac/conflict-fragility-resilience/humanitarian-financing/financing-forced-displacement.htm)\n<u>[dac/confict-fragility-resilience/humanitarian-fl](http://www\n\nMention: \"DAC List of ODA-recipients\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Définit la liste des pays d'accueil éligibles à l'analyse.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000789:14:0", "occurrence_id": "f712417a1b1449d8fd0491b1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: _Refugee-hosting countries_ refers to all countries hosting\nrefugees that appear on the DAC List of ODA-recipients\n[(http://www.oecd.org/dac/financing-sustainable-de-](http://www.oecd.org/dac/financing-sustainable-development/development-finance-standards/DAC_List_ODA_Recipients2018to2020_flows_En.pdf)\n<u>[velopment/development-finance-standards/DAC_](http://www.oecd.org/dac/financing-sustainable-development/development-finance-standards/DAC_List_ODA_Recipients2018to2020_flows_En.pdf)</u>\n<u>[List_ODA_Recipients2018to2020_flows_En.pdf).](http://www.oecd.org/dac/financing-sustainable-development/development-finance-standards/DAC_List_ODA_Recipients2018to2020_flows_En.pdf)</u>\n\n###### Methodology\n\n\n**Computation Method:**\n\n\n**Disaggregation:** This indicator can be disaggregated by type of aid, donor, recipient country and by\ngeographical region and sub-region.\n\n\n**Methods and Guidance:** OECD Survey on Financing\n[for Refugee-Hosting Contexts: http://www.oecd.org/](http://www.oecd.org/dac/conflict-fragility-resilience/humanitarian-financing/financing-forced-displacement.htm)\n<u>[dac/confict-fragility-resilience/humanitarian-fl](http://www\n\nMention: \"OECD Survey on Financing\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Ses données structurent l'analyse par pays et par région.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000791:5:0", "occurrence_id": "f7a8c0d9c87b4534f126f1ab", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Endnotes**\n\n1. These figures represent the total number of migrants and refugees who arrived in Italy and Malta through the Central Mediterranean Sea after departing\nfrom Algeria, Libya, and Tunisia in 2022. The numbers may include migrants and refugees who made more than one attempt to cross the sea from these\ncountries. These figures also include those who were intercepted or rescued by the authorities of Algeria, Libya and Tunisia and disembarked in these\ncountries in 2022.\n\n\nData on disembarkation in the three North African countries is sourced from a variety of channels, including official reports shared by national authorities,\nand media monitoring, and data gathered at the disembarkation points where UNHCR and IOM partners were present.\n\n\nData on arrivals in Italy and Malta is received from the Ministry of Interior of both countries.\n\n\n2. Central Mediterranean Route refers to sea departures from Algeria, Libya, and Tunisia, and to sea arrivals to Italy and Malta through the Mediterranean\nSea.\n\n\n3. Sources: UNHCR, IOM, UNDSS, Italian MoI, Official website of the Tunisian Ministry of Interior and the social media pages of the Tunisian Ministry of\nInterior and Tunisian National Guard, media and social media monitoring.\n\n\n4. Arrivals from Algeria were also recorded in Spain, but these fall outside the scope of this factsheet.\n\n\n5. See IOM <u>[Missing Migrants Project and UNHCR](https://missingmigrants.iom.int/)</u> <u>[Dead and Missing at Sea dashboard to explore data. The numbers of dead and missing in this factsheet](https://data2.unhcr.org/en/dataviz/95?sv=0&geo=0&_gl=1*1ju2cm2*_rup_ga*MTQ0ODMwODQ4Ny4xNjgzNjQxNzc2*_rup_ga_EVDQT\n\nMention: \"data gathered at the disembarkation points\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Collected where the partners were present.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000791:5:0", "occurrence_id": "961ae29ce0cd7fcf66471a88", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Endnotes**\n\n1. These figures represent the total number of migrants and refugees who arrived in Italy and Malta through the Central Mediterranean Sea after departing\nfrom Algeria, Libya, and Tunisia in 2022. The numbers may include migrants and refugees who made more than one attempt to cross the sea from these\ncountries. These figures also include those who were intercepted or rescued by the authorities of Algeria, Libya and Tunisia and disembarked in these\ncountries in 2022.\n\n\nData on disembarkation in the three North African countries is sourced from a variety of channels, including official reports shared by national authorities,\nand media monitoring, and data gathered at the disembarkation points where UNHCR and IOM partners were present.\n\n\nData on arrivals in Italy and Malta is received from the Ministry of Interior of both countries.\n\n\n2. Central Mediterranean Route refers to sea departures from Algeria, Libya, and Tunisia, and to sea arrivals to Italy and Malta through the Mediterranean\nSea.\n\n\n3. Sources: UNHCR, IOM, UNDSS, Italian MoI, Official website of the Tunisian Ministry of Interior and the social media pages of the Tunisian Ministry of\nInterior and Tunisian National Guard, media and social media monitoring.\n\n\n4. Arrivals from Algeria were also recorded in Spain, but these fall outside the scope of this factsheet.\n\n\n5. See IOM <u>[Missing Migrants Project and UNHCR](https://missingmigrants.iom.int/)</u> <u>[Dead and Missing at Sea dashboard to explore data. The numbers of dead and missing in this factsheet](https://data2.unhcr.org/en/dataviz/95?sv=0&geo=0&_gl=1*1ju2cm2*_rup_ga*MTQ0ODMwODQ4Ny4xNjgzNjQxNzc2*_rup_ga_EVDQT\n\nMention: \"Data on arrivals in Italy and Malta\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Received from the two interior ministries concerned.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000791:5:0", "occurrence_id": "0ca45982004c06cbece44bda", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Endnotes**\n\n1. These figures represent the total number of migrants and refugees who arrived in Italy and Malta through the Central Mediterranean Sea after departing\nfrom Algeria, Libya, and Tunisia in 2022. The numbers may include migrants and refugees who made more than one attempt to cross the sea from these\ncountries. These figures also include those who were intercepted or rescued by the authorities of Algeria, Libya and Tunisia and disembarked in these\ncountries in 2022.\n\n\nData on disembarkation in the three North African countries is sourced from a variety of channels, including official reports shared by national authorities,\nand media monitoring, and data gathered at the disembarkation points where UNHCR and IOM partners were present.\n\n\nData on arrivals in Italy and Malta is received from the Ministry of Interior of both countries.\n\n\n2. Central Mediterranean Route refers to sea departures from Algeria, Libya, and Tunisia, and to sea arrivals to Italy and Malta through the Mediterranean\nSea.\n\n\n3. Sources: UNHCR, IOM, UNDSS, Italian MoI, Official website of the Tunisian Ministry of Interior and the social media pages of the Tunisian Ministry of\nInterior and Tunisian National Guard, media and social media monitoring.\n\n\n4. Arrivals from Algeria were also recorded in Spain, but these fall outside the scope of this factsheet.\n\n\n5. See IOM <u>[Missing Migrants Project and UNHCR](https://missingmigrants.iom.int/)</u> <u>[Dead and Missing at Sea dashboard to explore data. The numbers of dead and missing in this factsheet](https://data2.unhcr.org/en/dataviz/95?sv=0&geo=0&_gl=1*1ju2cm2*_rup_ga*MTQ0ODMwODQ4Ny4xNjgzNjQxNzc2*_rup_ga_EVDQT\n\nMention: \"Dead and Missing at Sea dashboard\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Allows exploration of sea death and missing figures.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000794:6:1", "occurrence_id": "850ad45b5e93adc9cbada8cf", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: placement Tracking Matrix) permet\nau gouvernement du Mali d’obtenir une meilleure compréhension des caractéristiques des populations\ndéplacées et retournées, ainsi que de mettre à jour leurs besoins et accès aux services de base. Ces\ninformations sont collectées puis disséminées auprès de la communauté humanitaire et des pouvoirs publics\nafin de leur fournir une analyse complète des caractéristiques, mouvements et besoins des populations\ndéplacées dans le pays et d’orienter les programmes d’aide humanitaire.\n\n### **II. METHODOLOGIE**\n\n\nLes activités de la DTM sont basées sur la méthodologie approuvée par la Commission Mouvement de Populations et\nsont menées par des équipes mixtes composées de membres, de la Direction Nationale du Développement Social\n(DNDS), de l’OIM et de l’UNHCR.\n\n\n**Enregistrement :**\n\n\nA. PDI et PDI retournées :\n\n\nLes opérations d’enregistrement consistent à faire le profilage des populations déplacées en collectant des données\nrelatives à leur profil démographique, l’historique de déplacement, les besoins, l’assistance reçue ainsi que les\nvulnérabilités identifiées.\n\n\n     - Formation : La formation des équipes en charge de la collecte de données a lieu par groupe de 20 à 30 personnes.\nLes formations sont menées par les agents de la DNDS. Les superviseurs et enquêteurs de la DNDS sont les cibles\nprivilégiées de ces sessions qui couvrent notamment\n\nMention: \"placement Tracking Matrix\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Permet au gouvernement d'améliorer sa compréhension de la situation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000802:20:1", "occurrence_id": "96540c22abbfc89936986e34", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  found within mixed migratory\nmovements who require a differentiated and focused response from states,\ninternational organizations and other actors. Others requiring special protection and\nassistance include victims of gender-based violence and other types of violence, the\nsick and elderly and people with disabilities.\n\nThe extent and nature of child migration from the Horn of Africa and Great Lakes\nregion through East and Southern Africa is difficult to quantify or analyse as little\nempirical evidence exists in relation to this issue. It seems likely that most refugee and\nmigrant children in Southern Africa originate from within the SADC region itself.\n\nUNICEF, for example, estimates that there are around 20,000 child migrants in South\nAfrica, the majority of whom are from Zimbabwe. The TMTF survey of migrants\ndetained in Tanzania found 144 Somali detainees under 18 years of age (some 37 per\ncent of the total number of Somalis surveyed), 64 of whom were under 15 (16 per\ncent).\n\nSuch figures (although they are not fully consistent) suggest that the movement of\nchildren across international borders in Southern Africa is a significant issue. If it is to\nbe addressed in an effective and equitable manner, adequate training and resources\nwill be required so as to ensure that the authorities can undertake a thorough Best\n\n\n19\n\nMention: \"TMTF survey of migrants\ndetained in Tanzania\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its entries count detained minors by nationality.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000812:1:0", "occurrence_id": "e276ffa7d69c08e3a1b78f09", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The following trends analysis is put together on the basis\nof available secondary data at the time of publication. It is\nrepresentative of the available information and therefore\nindicative of mixed migratory trends in West Africa.\n\n\nThe Mixed Migration Centre (MMC) was established in February\n2018. It brings together various existing regional initiatives –\nhosted or led by the Danish Refugee Council (DRC) – engaged\nin data collection, research, analysis and policy development\non mixed migration issues into a new global network of mixed\nmigration expertise.\n\n\nThe Mixed Migration Centre - West Africa, provides quality\nmixed migration-related information for policy, programming\nand advocacy from a regional perspective. Our core countries of\nfocus are Burkina Faso, Mali and Niger. For updates on North\nAfrica please consult MHub Monthly Trend Bulletins at:\n<u>http://www.mixedmigrationhub.org/monthly-trend-bulletins/</u>\n\n\nFor more information visit: <u>www.mixedmigration.org</u>\n\n\nYou can contact us by email: <u>west-africa@mixedmigration.org</u>\n\n\nPhoto credit: iStockphoto.com/yoh4nn\n\nMention: \"MHub Monthly Trend Bulletins\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Monthly information source on North Africa.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000817:20:1", "occurrence_id": "639f4fd99f96a6e9a7fd4b6b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \nUp to 90 governmental staff (30 enumerators in\neach statistics office) were trained and they were\nprovided with on-line data collection tool.\n\n\n**Household survey and sampling strategy:**\nA sample of 3,600 households (HH) were selected\nfrom Kurdistan Region of Iraq (1200 households\nfrom each governorate that cover 400 IDP HH, 400\nRefugees HH and 400 Host Community HH). The\nsamples are stratified by population group and\nurban typology: District Canters, Periphery and\ntowns.\n**Outcomes:** 3 profiling studies will be launched:\n1. Erbil in June.\n2. Duhok and 3. Sulaymaniyah by end July 2016.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nSource: IOM DTM and UNHCR registration (proGres)\nAs of 30April2016\n\n**PROFILING STEERING COMMITTEE MEMBERS (8**\n**members):**\n\n- 3 Government agencies Erbil Refugee Council\n(ERC), Joint Crisis Coordination Centre (JCC),\nKurdistan Region Statistics Office (KRSO)\n\n- 5 UN agencies (IOM, UNFPA, UN OCHA, UN\nHabitat and UNHCR)\n\n- The project has been funded by DFID.\n\n\n**PROFILING TECHNICAL WORKING GROUP:**\nStatistics and governor offices of Erbil, Duhok and\nSulaymaniyah, UNHCR. The group is supported by\nJoint IDP Profiling Service (JIPS).\n\n\n20\n\nMention: \"IOM DTM and UNHCR registration\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Their combined records give the population figures on that date.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000819:12:1", "occurrence_id": "423a2d0f9e6207e21d2515db", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Evaluation of the introduction of post-exposure prophylaxis in the clinical management of rape survivors in Kibondo refugee camps, Tanzania. Geneva: UNHCR, 2005.\n\n5. UNHCR, WFP. Integration of HIV/AIDS activities with food and nutrition support in refugee settings: specific programme strategies. Geneva: UNHCR and WFP, 2004.\n\n6. WHO. HIV drug resistance; http://www.who.int/hiv/drugresistance/en/. Geneva, 2006.\n\n7. WHO. Antiretroviral drugs for treating pregnant women and preventing HIV infection in infants in resource-limited settings:\ntowards universal access; http://www.who.int/hiv/pub/guidelines/pmtct/en/index.html. Geneva, 2006.\n\n8. WHO, UNHCR. Clinical management of rape survivors. Geneva, 2004.\n\n9. United Nations High Commissioner for Refugees, 2006. Global refugee trends; <u>http://www.unhcr.org/statistics/STATIS-</u>\n<u>TICS/4486ceb12.pdf. Geneva, 2005.</u>\n\n10. United Nations High Commissioner for Refugees. Protracted Refugee Situations, Standing Committee 30th meeting.\nEC/54/SC/CRP.14. Geneva, 10 June 2004.\n\n11. United Nations General Assembly. Declaration of commitment on HIV/AIDS. Geneva: United Nations and UNAIDS, 2001.\n\n12. Spiegel PB. HIV/AIDS among conflict-affected and displaced populations: dispelling myths and taking action. _Disasters_ .\n2004; **28** (3):322-339.\n\n13. Inter-Agency Standing Committee (IASC). Guidelines for HIV/AIDS interventions in emergency settings. Geneva: IASC reference\ngroup, 2004.\n\n14. Spiegel P, Harroff-Tavel H. HIV/AIDS in internally displaced persons in 8 priority countries.\n\nMention: \"Global refugee trends\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"A compilation of global refugee trends.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000824:9:4", "occurrence_id": "5dca9b2253ee8395da35ba1f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  of 95% and a margin of error\nof 10% (unless stated otherwise). Data for the mantika of Sirte is indicative only and should not be\ninterpreted with statistical significance.\n10 NB: As outlined, all findings presented on refugees and migrants are based on a qualitative\nmethodology, based on purposive sampling. Hence, any comparison with findings on the Libyan\npopulation originating from the REACH MSNA should be only treated as indicative and cannot be\ngeneralised to represent overall trends between refugees and migrants and Libyans in the country.\n11 REACH/UNHCR, <u>[Access to cash and the impact of the liquidity crisis on refugees and migrants](http://www.reachresourcecentre.info/system/files/resource-documents/reach_lby_brief_refugees_and_migrants_access_to_cash_june2018.pdf)</u>\n<u>[in Libya, June 2018 and REACH/UNHCR,](http://www.reachresourcecentre.info/system/files/resource-documents/reach_lby_brief_refugees_and_migrants_access_to_cash_june2018.pdf)</u> <u>[Refugees and migrants’ access to resources, healthcare](http://www.reachresourcecentre.info/system/files/resource-documents/reach_lby_report_merf_december_2017_0.pdf)</u>\n<u>[and housing in Libya: Key challenges and coping mechanisms, December 2017.](http://www.reachresourcecentre.info/system/files/resource-documents/reach_lby_report_merf_december_2017_0.pdf)</u>\n[12 See REACH/UNHCR, Access to cash and the impact\n\nMention: \"REACH MSNA\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Any comparison with its results remains indicative and not generalisable.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000828:1:0", "occurrence_id": "9a773a69282aed6676d26369", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: timely start of activities is crucial to ensure the preventive activities can be planned and\nimplemented.\n\n4.6 million people are identified to need shelter and NFI assistance, with slightly more than 88%\nof households reported living in a makeshift shelter according to REACH assessment or partially\ndamaged, or fully destroyed shelter, due to previous conflicts and disasters. Considering the high\ncost of shelter repairs, and the already negative net income of households, reconstruction comes\nat a high cost that may not be affordable for most. Failure to address these issues would keep\nhundreds of vulnerable households in inadequate, unsecured, and often overcrowded shelters,\nwith severe implications for their health, protection, socio- economic situation and personal\nsecurity - especially for children, the elderly, disabled people, women, and girls.\n\n\nSomalia is gradually recovering from its prolonged conflict, which has led to widespread\ndisplacement. Internally displaced persons (IDPs) face significant challenges, particularly in\nsecuring Housing, Land, and Property (HLP) rights. With weak governance structures, Somalia’s\nlegal systems remain fragmented, complicating access to justice for vulnerable populations.\nMany IDPs live in urban centers, where land tenure insecurity is rampant, leading to frequent\nforced evictions, which further displace already vulnerable communities. Eviction data gathered\nby the HLP AoR since 2015, indicates that over 1.7 million IDPs experienced forced evictions and\nby the end of September 2024 alone, over 98,000 people have been forcefully evicted, often in\nviolation of both national and international guidelines. Up to 60% of all evictions reported are in\noccurring in IDPs located in cities, in particular Mogadishu, Baidoa, Kismayo, Bosaso, Garowe\namong others <sup>2</sup>\n\nThe justice system in Somalia is complex, combining formal institutions, clan-based customary\npractices (xeer), and Shari’a courts. This legal plurality has made it difficult for IDPs and\nmarginalized groups to\n\nMention: \"REACH assessment\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"More than 88 percent of households report a makeshift shelter.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000828:1:0", "occurrence_id": "0a67672fc74f5f71423fe01b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: timely start of activities is crucial to ensure the preventive activities can be planned and\nimplemented.\n\n4.6 million people are identified to need shelter and NFI assistance, with slightly more than 88%\nof households reported living in a makeshift shelter according to REACH assessment or partially\ndamaged, or fully destroyed shelter, due to previous conflicts and disasters. Considering the high\ncost of shelter repairs, and the already negative net income of households, reconstruction comes\nat a high cost that may not be affordable for most. Failure to address these issues would keep\nhundreds of vulnerable households in inadequate, unsecured, and often overcrowded shelters,\nwith severe implications for their health, protection, socio- economic situation and personal\nsecurity - especially for children, the elderly, disabled people, women, and girls.\n\n\nSomalia is gradually recovering from its prolonged conflict, which has led to widespread\ndisplacement. Internally displaced persons (IDPs) face significant challenges, particularly in\nsecuring Housing, Land, and Property (HLP) rights. With weak governance structures, Somalia’s\nlegal systems remain fragmented, complicating access to justice for vulnerable populations.\nMany IDPs live in urban centers, where land tenure insecurity is rampant, leading to frequent\nforced evictions, which further displace already vulnerable communities. Eviction data gathered\nby the HLP AoR since 2015, indicates that over 1.7 million IDPs experienced forced evictions and\nby the end of September 2024 alone, over 98,000 people have been forcefully evicted, often in\nviolation of both national and international guidelines. Up to 60% of all evictions reported are in\noccurring in IDPs located in cities, in particular Mogadishu, Baidoa, Kismayo, Bosaso, Garowe\namong others <sup>2</sup>\n\nThe justice system in Somalia is complex, combining formal institutions, clan-based customary\npractices (xeer), and Shari’a courts. This legal plurality has made it difficult for IDPs and\nmarginalized groups to\n\nMention: \"Eviction data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its entries since that year count displacements by evictions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000834:5:0", "occurrence_id": "ea19bc53ffd49db3cef90a14", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: aftermath of the 2010 earthquake. The monumental catastrophe ravaged Port-au-Prince,\nHaiti’s capital city, causing millions of Haitians to flee their homes and thousands to leave\nthe country. Many of those have chosen to take to the sea, paying smugglers, known as\n“coyotes”, up to 4,000USD to transport them on a treacherous voyage through the\nDominican Republic to Panama and then on to Ecuador and either passing through Bolivia,\nPeru or Colombia before reaching Brazilian cities such as Brasiléia and Tabatinga in the\nstates of Acre and Amazonas.\n\n\nFor a temporary solution to their migratory status, Haitians have applied for asylum when\nentering the country. Even though the Government of Brazil does not officially recognise\nHaitian migrants as refugees because they do not fit the definition of the _Brazilian Law_\n_9.474/1997_, the Government of Brazil does acknowledge their need for humanitarian\nsupport. As a result, more than 10,000 humanitarian visas have been granted to Haitian\nnationals. According to data from the Federal Police, 24,515 Haitian nationals have entered\nBrazilian territory in 2014, while 16,982 have applied for asylum in the same period. The total\nnumber of Haitians currently in Brazilian territory is 39,762.\n\n\nBy mid-2015, the total number of recognized refugees in Brazil was 8,400 (25 per cent of\nwhom are women). Recognized refugees come from 81 different countries, mostly from\nSyria, Colombia, Angola and the Democratic Republic of Congo (DRC). There are 1,093\nfrom Colombia; 1,480 from Angola and 844 from the DRC. <sup>3</sup> [^3: CONARE figures, August 2015.] For the first time in years, the\nlargest population of refugees is from Syria instead of Angola or Colombia.\n\n\nWith Brazil’s expanded economy and its compassionate approach to extending visas and\nresidency permits to vulnerable populations, it is not surprising that so many people have\nsought this state’s protection. Moreover\n\nMention: \"data from the Federal Police\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its registers give entries into national territory by year.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000834:5:0", "occurrence_id": "3d5d3f0b377f87bd6bff257a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: aftermath of the 2010 earthquake. The monumental catastrophe ravaged Port-au-Prince,\nHaiti’s capital city, causing millions of Haitians to flee their homes and thousands to leave\nthe country. Many of those have chosen to take to the sea, paying smugglers, known as\n“coyotes”, up to 4,000USD to transport them on a treacherous voyage through the\nDominican Republic to Panama and then on to Ecuador and either passing through Bolivia,\nPeru or Colombia before reaching Brazilian cities such as Brasiléia and Tabatinga in the\nstates of Acre and Amazonas.\n\n\nFor a temporary solution to their migratory status, Haitians have applied for asylum when\nentering the country. Even though the Government of Brazil does not officially recognise\nHaitian migrants as refugees because they do not fit the definition of the _Brazilian Law_\n_9.474/1997_, the Government of Brazil does acknowledge their need for humanitarian\nsupport. As a result, more than 10,000 humanitarian visas have been granted to Haitian\nnationals. According to data from the Federal Police, 24,515 Haitian nationals have entered\nBrazilian territory in 2014, while 16,982 have applied for asylum in the same period. The total\nnumber of Haitians currently in Brazilian territory is 39,762.\n\n\nBy mid-2015, the total number of recognized refugees in Brazil was 8,400 (25 per cent of\nwhom are women). Recognized refugees come from 81 different countries, mostly from\nSyria, Colombia, Angola and the Democratic Republic of Congo (DRC). There are 1,093\nfrom Colombia; 1,480 from Angola and 844 from the DRC. <sup>3</sup> [^3: CONARE figures, August 2015.] For the first time in years, the\nlargest population of refugees is from Syria instead of Angola or Colombia.\n\n\nWith Brazil’s expanded economy and its compassionate approach to extending visas and\nresidency permits to vulnerable populations, it is not surprising that so many people have\nsought this state’s protection. Moreover\n\nMention: \"CONARE figures\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its figures describe the origin composition of refugee populations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000841:4:1", "occurrence_id": "1e2b805cdb3ca1a31d2b8bd6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: . Through case management, progress monitoring, particularly in areas\nof relocation, was prioritized to ensure comprehensive services were provided by protection staff\nfrom the receiving refugee camps.\n\n\n**Additional assistance**\n\n\nDue to limited resources, case-by-case referrals were made through Individual Protection\nAssistance, where those with heightened risks and increased vulnerabilities were provided with\ncash to mitigate protection risks. Out of the total 19, 559 identifies during the first half of the\nyear, **10,920 (56 percent)** received targeted assistance in transit centers. Similarly, partners\nJesuit Refugee Service (JRS) and Chagai Foundation distributed assistive devices, including\nwheelchairs and crutches, to **133** individuals with mobility challenges, enabling ease their\nmovement from one service provider to another.\n\n\n**Monitoring and tracking**\n\n\nThe identification, case management, and ProGres database continued to support the tracking of\nindividuals with specific needs. Between January and June, **514 individuals** (41 percent female)\nregistered for relocation to Ajuong Thuok and Maban refugee camps. To facilitate further\nassistance, UNHCR shared information from the ProGres database from Renk to these locations,\nmaintaining confidentiality. Most of the identified cases during relocation included Women at Risk,\n\n\n5\n\nMention: \"ProGres database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Supports monitoring of individuals with specific needs.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000850:9:0", "occurrence_id": "6dac0538aadf649cfd0e0e62", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Refugee Enrolment: What the data tells us\n\n\nThis report serves not only as a reminder of the\nbarriers young refugees face in fulfilling their\ndreams and ambitions; it highlights the strong\npartnerships that are needed to break down those\nbarriers and open the doors to the classroom.\n\n\nData methodology\n\n\n\nStatistics on refugee enrolment and population\nnumbers have been drawn from UNHCR’s\npopulation database, as well as education datacollecting tools (such as school administrative\ndata, registration data, and household surveys) in\n12 countries (Chad, Ethiopia, Iraq, Jordan, Kenya,\nLebanon, Pakistan, Rwanda, South Sudan,\nTanzania, Turkey, Uganda). Data refers to 2019.\n\n\nThe total refugee population for the selected\ncountries is 10,539,446; more than half the 20.4\nmillion refugees under UNHCR’s mandate.\n\n\nThis report uses Gross Enrolment Ratios (GER)\nrather than net figures as in previous years. This\nmeans that we include all children enrolled in\nschool, regardless of whether they are in the right\nclass for their age. We have not included children\nin non-formal education.\n\n\nCalculating how many children are in or out of\nschool accurately requires age-specific enrolment\ndata at all levels, which is challenging in many\nparts of the world. In addition, some countries\ninclude displaced children in their national\neducation systems but do not disaggregate based\non international protection status, further clouding\nthe picture.\n\n\n\nSince refugee populations and education systems\nin these 12 countries cannot be assumed to be\nrepresentative of the situation of the global\nrefugee population, we cannot infer universal\nconclusions from this education data. However, it\nhelps provide indications of challenges and trends\nin refugee education.\n\n\nFinally, the 3 per cent figure in tertiary education\nrepresents the approximate portion of refugee\nyouth known to be enrolled in tertiary education.\nBecause tertiary enrolment data is not centrally\nrecorded in most countries and is rarely\ndisaggregated to reflect refugee status, this report\nhas to rely on a variety of self-reported, public or\nother easily accessible sources.\n\nMention: \"UNHCR’s\npopulation database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Source of refugee counts and registrations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000850:9:0", "occurrence_id": "1b6487900b13bf61d6241875", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Refugee Enrolment: What the data tells us\n\n\nThis report serves not only as a reminder of the\nbarriers young refugees face in fulfilling their\ndreams and ambitions; it highlights the strong\npartnerships that are needed to break down those\nbarriers and open the doors to the classroom.\n\n\nData methodology\n\n\n\nStatistics on refugee enrolment and population\nnumbers have been drawn from UNHCR’s\npopulation database, as well as education datacollecting tools (such as school administrative\ndata, registration data, and household surveys) in\n12 countries (Chad, Ethiopia, Iraq, Jordan, Kenya,\nLebanon, Pakistan, Rwanda, South Sudan,\nTanzania, Turkey, Uganda). Data refers to 2019.\n\n\nThe total refugee population for the selected\ncountries is 10,539,446; more than half the 20.4\nmillion refugees under UNHCR’s mandate.\n\n\nThis report uses Gross Enrolment Ratios (GER)\nrather than net figures as in previous years. This\nmeans that we include all children enrolled in\nschool, regardless of whether they are in the right\nclass for their age. We have not included children\nin non-formal education.\n\n\nCalculating how many children are in or out of\nschool accurately requires age-specific enrolment\ndata at all levels, which is challenging in many\nparts of the world. In addition, some countries\ninclude displaced children in their national\neducation systems but do not disaggregate based\non international protection status, further clouding\nthe picture.\n\n\n\nSince refugee populations and education systems\nin these 12 countries cannot be assumed to be\nrepresentative of the situation of the global\nrefugee population, we cannot infer universal\nconclusions from this education data. However, it\nhelps provide indications of challenges and trends\nin refugee education.\n\n\nFinally, the 3 per cent figure in tertiary education\nrepresents the approximate portion of refugee\nyouth known to be enrolled in tertiary education.\nBecause tertiary enrolment data is not centrally\nrecorded in most countries and is rarely\ndisaggregated to reflect refugee status, this report\nhas to rely on a variety of self-reported, public or\nother easily accessible sources.\n\nMention: \"school administrative\ndata\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"One of the collection tools in the twelve countries covered.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000850:9:0", "occurrence_id": "9c2d15e55bbc3b2f10315fb5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Refugee Enrolment: What the data tells us\n\n\nThis report serves not only as a reminder of the\nbarriers young refugees face in fulfilling their\ndreams and ambitions; it highlights the strong\npartnerships that are needed to break down those\nbarriers and open the doors to the classroom.\n\n\nData methodology\n\n\n\nStatistics on refugee enrolment and population\nnumbers have been drawn from UNHCR’s\npopulation database, as well as education datacollecting tools (such as school administrative\ndata, registration data, and household surveys) in\n12 countries (Chad, Ethiopia, Iraq, Jordan, Kenya,\nLebanon, Pakistan, Rwanda, South Sudan,\nTanzania, Turkey, Uganda). Data refers to 2019.\n\n\nThe total refugee population for the selected\ncountries is 10,539,446; more than half the 20.4\nmillion refugees under UNHCR’s mandate.\n\n\nThis report uses Gross Enrolment Ratios (GER)\nrather than net figures as in previous years. This\nmeans that we include all children enrolled in\nschool, regardless of whether they are in the right\nclass for their age. We have not included children\nin non-formal education.\n\n\nCalculating how many children are in or out of\nschool accurately requires age-specific enrolment\ndata at all levels, which is challenging in many\nparts of the world. In addition, some countries\ninclude displaced children in their national\neducation systems but do not disaggregate based\non international protection status, further clouding\nthe picture.\n\n\n\nSince refugee populations and education systems\nin these 12 countries cannot be assumed to be\nrepresentative of the situation of the global\nrefugee population, we cannot infer universal\nconclusions from this education data. However, it\nhelps provide indications of challenges and trends\nin refugee education.\n\n\nFinally, the 3 per cent figure in tertiary education\nrepresents the approximate portion of refugee\nyouth known to be enrolled in tertiary education.\nBecause tertiary enrolment data is not centrally\nrecorded in most countries and is rarely\ndisaggregated to reflect refugee status, this report\nhas to rely on a variety of self-reported, public or\nother easily accessible sources.\n\nMention: \"registration data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The second collection tool mentioned.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000850:9:0", "occurrence_id": "462ba76de9f6fcc1ba52818b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Refugee Enrolment: What the data tells us\n\n\nThis report serves not only as a reminder of the\nbarriers young refugees face in fulfilling their\ndreams and ambitions; it highlights the strong\npartnerships that are needed to break down those\nbarriers and open the doors to the classroom.\n\n\nData methodology\n\n\n\nStatistics on refugee enrolment and population\nnumbers have been drawn from UNHCR’s\npopulation database, as well as education datacollecting tools (such as school administrative\ndata, registration data, and household surveys) in\n12 countries (Chad, Ethiopia, Iraq, Jordan, Kenya,\nLebanon, Pakistan, Rwanda, South Sudan,\nTanzania, Turkey, Uganda). Data refers to 2019.\n\n\nThe total refugee population for the selected\ncountries is 10,539,446; more than half the 20.4\nmillion refugees under UNHCR’s mandate.\n\n\nThis report uses Gross Enrolment Ratios (GER)\nrather than net figures as in previous years. This\nmeans that we include all children enrolled in\nschool, regardless of whether they are in the right\nclass for their age. We have not included children\nin non-formal education.\n\n\nCalculating how many children are in or out of\nschool accurately requires age-specific enrolment\ndata at all levels, which is challenging in many\nparts of the world. In addition, some countries\ninclude displaced children in their national\neducation systems but do not disaggregate based\non international protection status, further clouding\nthe picture.\n\n\n\nSince refugee populations and education systems\nin these 12 countries cannot be assumed to be\nrepresentative of the situation of the global\nrefugee population, we cannot infer universal\nconclusions from this education data. However, it\nhelps provide indications of challenges and trends\nin refugee education.\n\n\nFinally, the 3 per cent figure in tertiary education\nrepresents the approximate portion of refugee\nyouth known to be enrolled in tertiary education.\nBecause tertiary enrolment data is not centrally\nrecorded in most countries and is rarely\ndisaggregated to reflect refugee status, this report\nhas to rely on a variety of self-reported, public or\nother easily accessible sources.\n\nMention: \"household surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The third collection tool mentioned.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000850:9:0", "occurrence_id": "a975efadc930e54410e86ba5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Refugee Enrolment: What the data tells us\n\n\nThis report serves not only as a reminder of the\nbarriers young refugees face in fulfilling their\ndreams and ambitions; it highlights the strong\npartnerships that are needed to break down those\nbarriers and open the doors to the classroom.\n\n\nData methodology\n\n\n\nStatistics on refugee enrolment and population\nnumbers have been drawn from UNHCR’s\npopulation database, as well as education datacollecting tools (such as school administrative\ndata, registration data, and household surveys) in\n12 countries (Chad, Ethiopia, Iraq, Jordan, Kenya,\nLebanon, Pakistan, Rwanda, South Sudan,\nTanzania, Turkey, Uganda). Data refers to 2019.\n\n\nThe total refugee population for the selected\ncountries is 10,539,446; more than half the 20.4\nmillion refugees under UNHCR’s mandate.\n\n\nThis report uses Gross Enrolment Ratios (GER)\nrather than net figures as in previous years. This\nmeans that we include all children enrolled in\nschool, regardless of whether they are in the right\nclass for their age. We have not included children\nin non-formal education.\n\n\nCalculating how many children are in or out of\nschool accurately requires age-specific enrolment\ndata at all levels, which is challenging in many\nparts of the world. In addition, some countries\ninclude displaced children in their national\neducation systems but do not disaggregate based\non international protection status, further clouding\nthe picture.\n\n\n\nSince refugee populations and education systems\nin these 12 countries cannot be assumed to be\nrepresentative of the situation of the global\nrefugee population, we cannot infer universal\nconclusions from this education data. However, it\nhelps provide indications of challenges and trends\nin refugee education.\n\n\nFinally, the 3 per cent figure in tertiary education\nrepresents the approximate portion of refugee\nyouth known to be enrolled in tertiary education.\nBecause tertiary enrolment data is not centrally\nrecorded in most countries and is rarely\ndisaggregated to reflect refugee status, this report\nhas to rely on a variety of self-reported, public or\nother easily accessible sources.\n\nMention: \"education data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Its coverage does not allow universal conclusions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000850:9:0", "occurrence_id": "7ad4a7d4bf2c40535362511e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Refugee Enrolment: What the data tells us\n\n\nThis report serves not only as a reminder of the\nbarriers young refugees face in fulfilling their\ndreams and ambitions; it highlights the strong\npartnerships that are needed to break down those\nbarriers and open the doors to the classroom.\n\n\nData methodology\n\n\n\nStatistics on refugee enrolment and population\nnumbers have been drawn from UNHCR’s\npopulation database, as well as education datacollecting tools (such as school administrative\ndata, registration data, and household surveys) in\n12 countries (Chad, Ethiopia, Iraq, Jordan, Kenya,\nLebanon, Pakistan, Rwanda, South Sudan,\nTanzania, Turkey, Uganda). Data refers to 2019.\n\n\nThe total refugee population for the selected\ncountries is 10,539,446; more than half the 20.4\nmillion refugees under UNHCR’s mandate.\n\n\nThis report uses Gross Enrolment Ratios (GER)\nrather than net figures as in previous years. This\nmeans that we include all children enrolled in\nschool, regardless of whether they are in the right\nclass for their age. We have not included children\nin non-formal education.\n\n\nCalculating how many children are in or out of\nschool accurately requires age-specific enrolment\ndata at all levels, which is challenging in many\nparts of the world. In addition, some countries\ninclude displaced children in their national\neducation systems but do not disaggregate based\non international protection status, further clouding\nthe picture.\n\n\n\nSince refugee populations and education systems\nin these 12 countries cannot be assumed to be\nrepresentative of the situation of the global\nrefugee population, we cannot infer universal\nconclusions from this education data. However, it\nhelps provide indications of challenges and trends\nin refugee education.\n\n\nFinally, the 3 per cent figure in tertiary education\nrepresents the approximate portion of refugee\nyouth known to be enrolled in tertiary education.\nBecause tertiary enrolment data is not centrally\nrecorded in most countries and is rarely\ndisaggregated to reflect refugee status, this report\nhas to rely on a variety of self-reported, public or\nother easily accessible sources.\n\nMention: \"tertiary enrolment data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Is not centrally registered in most countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000853:37:0", "occurrence_id": "d649a741d0fb7942757da6f2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |GOAL 2: Ensure that alternatives to detention are available in law and implemented in practice|Col2|Col3|Col4|Col5|Col6|Col7|Col8|Col9|Col10|Col11|Col12|Col13|Col14|\n|---|---|---|---|---|---|---|---|---|---|---|---|---|---|\n|2013 detention baseline – UNHCR Global Strategy<br>– Beyond Detention 2014-2019|2013 detention baseline – UNHCR Global Strategy<br>– Beyond Detention 2014-2019|Canada|Hungary|Indonesia|Israel|Lithuania|Malaysia|Malta|Mexico|Thailand|United Kingdom|United States|Zambia|\n|**SUB-GOAL 1:** \u0007Legal and policy frameworks include alternative(s) to immigration detention.|**SUB-GOAL 1:** \u0007Legal and policy frameworks include alternative(s) to immigration detention.|**SUB-GOAL 1:** \u0007Legal and policy frameworks include alternative(s) to immigration detention.|**SUB-GOAL 1:** \u0007Legal and policy frameworks include alternative(s) to immigration detention.|**SUB-GOAL 1:** \u0007Legal and policy frameworks include alternative(s) to immigration detention.|**SUB-GOAL 1:** \u0007Legal and policy frameworks include alternative(s) to immigration detention.|**SUB-GOAL 1:** \u0007Legal and policy frameworks include alternative(s) to immigration detention.|**SUB-GOAL 1:** \u0007Legal and policy frameworks include alternative(s) to\n\nMention: \"2013 detention baseline\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Sert de référence à la stratégie mondiale sur la détention.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000854:16:0", "occurrence_id": "1162713dbfe5b6db677c52ae", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Kevin, 11, and his brother Jason, 14, Guatemalan refugees, fled to Mexico to escape violence. In their new school in Saltillo, they are able to pursue their \npassion of robotics and science. ©UNHCR/ENCARNI PINDADO\nCASE STUDY\nMACHINE  \nLEARNING\nRobot-mad brothers pursue  \nlove of science in new school\nJason and Kevin González* are robot crazy. At \nschool in Guatemala, the brothers signed up \nfor robotics workshops and were encouraged \nto take part in science competitions. \nThe two of them even managed to win first prize in \ntheir category in a national robotics contest. For \nthe science-mad duo, life seemed great. “We had \neverything a kid could want,” says Jason, 14.\n“And all of the sudden, we had \nto leave behind the life we \nknew.”\nLike thousands of refugees from the North of \nCentral America region, Jason, his 11-year-old \nsibling and their father fled the gang-related \nviolence in Guatemala and headed north for \nMexico. \nAccording to UNHCR’s Global Trends report, by \nthe end of 2017 the number of asylum-seekers \nand refugees from the region had reached more \nthan 294,000 – up by 58 per cent from the year \nbefore and 16 times more people than at the end \nof 2011.\nOnce over the border to Tapachula, in southern \nMexico’s Chiapas state, the Gonzalez’s  applied for \nrefugee status and the boys set about trying to \nrestart their education. \nBut that was not as easy as abc. “The first school \nwe tried did not accept my kids,” says Andres, the \nboys’ father. “They said they had no spaces, but \nwe felt it had more to do with us being foreigners, \nCentral Americans.”\nEventually, with the support of UNHCR partner \nRET International, a humanitarian organization \nfocused on education, the children were enrolled \nin a public boarding school, where they stayed \nduring the week and went back to their father\n\nMention: \"Global Trends report\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its figures describe the evolution of the region's asylum seekers.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000861:15:1", "occurrence_id": "24f71861228ade32962c7849", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  while there is significant variation on a governorate\nbasis, on average, the number of unregistered Syrian refugees was about 15 per cent of the number\nof registered ones. <sup>20</sup> Combining UNHCR registration data and CFSME figures for the unregistered,\nit is estimated that there were a total of about 727,600 Syrian refugees residing in Jordan in August\n2015. This suggests a Syrian refugee density of about 9.8 per cent for Jordan as a whole with\nthe highest density in the Mafraq governorate (excluding the refugee camps)—at 20.2 per cent. <sup>21</sup>\n\nExcluding the Za’atari and Azraq refugee camps, whose residents’ movement outside the camp\nare restricted, the number of registered and unregistered Syrian refugees living in Jordanian host\ncommunities is estimated at about 624,870 in August 2015 (see Table 1).\n\n\n17 These include UNHCR registered refugees, non-registered refugees and pre-existing Syrian migrants previously resident in Jordan.\nMeeting with Omar Nuseir, Senior Coordinator for Humanitarian Affairs, Ministry of Planning and International Cooperation, 20\nNovember 2014.\n18 UNHCR data is taken from http://data.unhcr.org/syrianrefugees/country.\n19 WFP and REACH Comprehensive Food Security Monitoring Exercise; Syrian Refugees in Jordan, July 2014. This large and\nimportant survey, conducted in December 2013 and January 2014, covered a representative sample of some 7,100 Syrian refugee\nhouseholds. The total number of individuals assessed across all households surveyed was 35,780. The survey is referred to\nhereafter as the CFSME.\n20 WFP confirmed that unregistered household members in their survey were almost exclusively Syrian refugees, November 2014.\n21 Refugee density is calculated as the ratio: total refugees/(Jordanians + total refugees).\n\n\n**17**\n\nMention: \"UNHCR registration data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Combined with the survey figures they give the estimated total.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000861:15:1", "occurrence_id": "9f8884c282a408679fd0dba5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  while there is significant variation on a governorate\nbasis, on average, the number of unregistered Syrian refugees was about 15 per cent of the number\nof registered ones. <sup>20</sup> Combining UNHCR registration data and CFSME figures for the unregistered,\nit is estimated that there were a total of about 727,600 Syrian refugees residing in Jordan in August\n2015. This suggests a Syrian refugee density of about 9.8 per cent for Jordan as a whole with\nthe highest density in the Mafraq governorate (excluding the refugee camps)—at 20.2 per cent. <sup>21</sup>\n\nExcluding the Za’atari and Azraq refugee camps, whose residents’ movement outside the camp\nare restricted, the number of registered and unregistered Syrian refugees living in Jordanian host\ncommunities is estimated at about 624,870 in August 2015 (see Table 1).\n\n\n17 These include UNHCR registered refugees, non-registered refugees and pre-existing Syrian migrants previously resident in Jordan.\nMeeting with Omar Nuseir, Senior Coordinator for Humanitarian Affairs, Ministry of Planning and International Cooperation, 20\nNovember 2014.\n18 UNHCR data is taken from http://data.unhcr.org/syrianrefugees/country.\n19 WFP and REACH Comprehensive Food Security Monitoring Exercise; Syrian Refugees in Jordan, July 2014. This large and\nimportant survey, conducted in December 2013 and January 2014, covered a representative sample of some 7,100 Syrian refugee\nhouseholds. The total number of individuals assessed across all households surveyed was 35,780. The survey is referred to\nhereafter as the CFSME.\n20 WFP confirmed that unregistered household members in their survey were almost exclusively Syrian refugees, November 2014.\n21 Refugee density is calculated as the ratio: total refugees/(Jordanians + total refugees).\n\n\n**17**\n\nMention: \"CFSME\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its unregistered households complement the registration data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000866:5:0", "occurrence_id": "e581c1641a7d018cdd050f92", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **102 K** people\nbenefited with\n\n**NFI**\n\n\n\n**227 K** people\n\n**Energy**\n\n\nbenefited with **Shelter**\n\n\n\n\n\n\n\n**935 K** people reached\nby shelter, energy and\nNFI activities during\n2021.\n\n\n\n**94** **health centers** were benefited from repairs,\nenergy improvements and NFIs, providing an\nimportant support to the COVID-19 pandemic\nprevention, including the support for triage and\nvaccination areas. **35** **temporary**\n**collective** centers in border areas (including\nPASIs) were benefited by the response, focusing\non supporting vulnerable people on the\nmove. **78** **community** **centers** and **21**\n**schools** providing protection and essential\nservices to population have been benefited\nthrough shelter, energy and NFI response.\n\n\n\n**US$6.7** millions, funding\nreceived during 2021\naccording to Financial\nTracking Service (FTS).\n\n\n**1106 street solar lamps** were installed in public\nplaces, including the surroundings of\ninstitutions benefited by the response, to\nprovide illumination also preventing associated\nprotection risks. **42 electric systems**, including\nphotovoltaic systems for institutions,\norganizations, community centers and\ncollective shelters, were installed allowing\naccess to basic services for the most vulnerable\npopulation.\n\n\n\nBasic NFI distribution was critical in the\n\n**222** **refugee** **housing** **units** **(RHUs)** were assistance of vulnerable people specially those\ninstalled: 104 RHUs in health centres, 24 in affected by flooding and by armed\ncollective temporary shelters and PASIs, 21 in conflicts. NFIs delivered included **14,037**\nmilitary and civil coordination authority spaces, **portable** **solar** **lamps,**\n\nMention: \"Financial\nTracking Service\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its records give the amount received during the year.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000874:0:1", "occurrence_id": "fff88cdeed04c5483d7ca374", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: PMS Background</mark>**\nThe purpose of the Somalia Protection Monitoring System (SPMS) is the systematic and regular collection\nand analysis of information over an extended period to identify trends and patterns of violations of rights\nand protection risks for populations of concern to inform effective programming and advocacy. Data\ncollection in South West State started in March 2019.\n\n\nThe Joint Analysis Workshops have as aim to reflect on findings and agree on validity. As best practice,\nthe JAW is preceded by a community preparation meeting, with the same aim. The inputs of community\nmembers are considered in the JAW, organized with key stakeholders active in the relevant state. The\nseparation of stakeholders and community reflection meetings is to ensure community members feel\nuninhibited in sharing opinions and recommendations.\n\n\n\n**<mark>Headline Protection Trends</mark>**\nThe Danish Refugee Council (DRC), in partnership\nwith the Protection Cluster and facilitation support\nfrom AMARD, conducted a two-hour online\nworkshop to discuss findings from data collected\nthrough the SPMS in South West State in the\nperiod January to March 2022.\n\n\n<mark>The participants also engaged in group work and</mark>\n<mark>were divided into three groups. Each group</mark>\n<mark>discussed one protection concern. Below are the</mark>\n<mark>key outcomes from the discussions.</mark>\n\n\nThe discussions focused on the following guiding\nquestions:\n\nMention: \"data collected\nthrough the SPMS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its regional findings were discussed in an online workshop.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000875:6:1", "occurrence_id": "98d2bbd3a0f6f41827eeccb2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  which are placed under the leadership of the Governorates and which cover the overall\nresponse to refugees and IDPs in areas of mixed displacements. They are made up of Regional Technical\nDirectors, Prefects, the President of the Regional Council and humanitarian actors. These committees\nmeet monthly under the leadership of the Governors and with two co-leads: the Regional Directorate of\nCivil Status/Migration and Refugees, and the Representative of the Ministry of Humanitarian Action. These\ncommittees are supported by the United Nations Office for the Coordination of Humanitarian Affairs\n(OCHA) and UNHCR.\n\n\nSince September 2017, a government-led decentralization process has resulted in increased responsibility\nand autonomy for Regional Councils and municipalities in four key sectors: education, health, environment\nand water/sanitation. Regional sectoral working groups are led by the concerned regional technical\ndirectorates with the support of humanitarian actors.\n\n\nThe Government largely relies on UNHCR refugee data for its data collection and management process,\nhaving notably access to UNHCR electronic registration system for asylum-seekers. Overall refugee data\nis not part of the national database; however, refugees are increasingly becoming part of Government\nsurveys conducted in Niger. In 2018, refugees were included in the national poverty exercise led by the\nGovernment’s National Institute of Statistics and the World Bank. The resulting data, which is still being\nprocessed for publication, offers a rich comparative analysis of refugees and their hosts. Refugee data is\nalso taken into account through the national health information system (Système National d’information\nSanitaire – SNIS), which aligns with the Health Ministry’s national policy on health information management\nand provides reliable quality information to decision-makers, health officials and partners. Refugee\npopulations are also fully integrated into the Government’s COVID-19 pandemic response plan.\n\n\nAt the various levels of governance, there is no formal process through which refugees are consulted\nregarding decisions that affect them. While there are refugee committees in Niger, including in\n\nMention: \"UNHCR refugee data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The government relies heavily on them for its data management.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000885:1:1", "occurrence_id": "d8fc8d63c537815492ea81d1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  such as women, children, adolescents, persons with disability, returnees, internally\ndisplaced persons (IDPs), ethnic and religious minorities, nomadic and stateless persons. The lack\nof a Tazkira has a spillover effect, as it is a requirement to acquire other documentation e.g. a\npassport. <sup>i</sup> Consequently, accessing legal identity, efficient identity management and removing\nbarriers is crucial as a fundamental human right that duty-bearers must diligently safeguard to\nmitigate protection risks and achieve lasting solutions. Information on civil documentation and\nregistration rates are currently still limited. The latest World Bank report (2015) indicates that\nmerely 42% of the total births are registered in Afghanistan. Data from the Protection Cluster\nProtection Monitoring tool (December 2023) highlights that 55% of assessed households reported\nat least one family member lacking documentation, with the majority having never managed to\nobtain it. This disproportionately affects women and girls. Also, the UNHCR Protection monitoring\ntool (CPBM 2023) indicates that rural and culturally conservative areas are more affected, such as\nin Uruzgan and Hilmand Provinces, where 77% of the families reported lacking civil documentation,\nthe majority of those being women and girls.\n\n\nMAY 2024 1\n\nMention: \"Protection Cluster\nProtection Monitoring tool\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its data give the share of households reporting at least one case.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000885:1:1", "occurrence_id": "1dcf187acef7272f041222e8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  such as women, children, adolescents, persons with disability, returnees, internally\ndisplaced persons (IDPs), ethnic and religious minorities, nomadic and stateless persons. The lack\nof a Tazkira has a spillover effect, as it is a requirement to acquire other documentation e.g. a\npassport. <sup>i</sup> Consequently, accessing legal identity, efficient identity management and removing\nbarriers is crucial as a fundamental human right that duty-bearers must diligently safeguard to\nmitigate protection risks and achieve lasting solutions. Information on civil documentation and\nregistration rates are currently still limited. The latest World Bank report (2015) indicates that\nmerely 42% of the total births are registered in Afghanistan. Data from the Protection Cluster\nProtection Monitoring tool (December 2023) highlights that 55% of assessed households reported\nat least one family member lacking documentation, with the majority having never managed to\nobtain it. This disproportionately affects women and girls. Also, the UNHCR Protection monitoring\ntool (CPBM 2023) indicates that rural and culturally conservative areas are more affected, such as\nin Uruzgan and Hilmand Provinces, where 77% of the families reported lacking civil documentation,\nthe majority of those being women and girls.\n\n\nMAY 2024 1\n\nMention: \"UNHCR Protection monitoring\ntool\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its results show conservative rural areas are more affected.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000888:3:1", "occurrence_id": "02af440bb6f86bb5c4f97efc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  in 2018_** <sup>**_1_**</sup> (Figure 1\nand Table 1). The United States remained the main G20 destination country with around 2.2 million new\nmigrants arriving in 2018. After a peak above 2.3 million in 2016, migration flows to the United States\ndecreased slightly for the second consecutive year. At 1.4 million, new permanent and temporary migration\nflows to Germany were stable in 2018 and remained higher than in any year prior to 2015. About 60% of\nthese new migrants were citizens of another EU country. Labour migration from Asia to Saudi Arabia is\none of the main migration channels in the G20 an account for a large share of the 1.8 million residence\npermits issued annually.\n\n\n**Figure 1. Recent changes in migration flows to selected G20 countries, 2016-2018**\n\n\n2018 or latest year available % change 2017/2016 % change 2018/2017\n\n\n\n**Millions** **%**\n\n\n\n4\n\n\n\n40\n\n\n\n3\n\n\n2\n\n\n1\n\n\n0\n\n\n- 1\n\n\n- 2\n\n\n\n30\n\n\n20\n\n\n10\n\n\n0\n\n\n- 10\n\n\n- 20\n\n\n\nSource: see Table 1 below.\n\n\n1 Data on international migration flows are not available for all G20 countries and are not systematically\ncomparable. The data presented here are compiled from many different sources, notably based on OECD regional\nmonitoring systems, completed with national sources. They include all types of migration ranging from settlement to\ntemporary migration, including students, but exclude visitors.\n\n\n4\n\nMention: \"OECD regional\nmonitoring systems\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Completed by national sources to cover all migration types.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000907:43:1", "occurrence_id": "7ba4b04ba39778f48444c0e6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  cash transfers.** However,\ncumbersome banking practices have delayed\npayments in some countries. UNHCR\nMacedonia cites that as a problem and UNHCR\nIndia notes that it has not yet been able to use\nATMs for cash transfers.\n\n\n#### **MATERIAL SUPPORT** **FOR URBAN REFUGEES:** **WHAT WORKS WELL?**\n\n\n**• Linking material support to self-reliance**\n**programs has worked well** in Lebanon,\nCosta Rica, and Ecuador. This includes\nmicro-credit, income generation and\nentrepreneurship activities, vocational\ntechnical and skill training.\n\n\n**• Including refugees in state social programs**\n**is underway in some countries.** In Ukraine,\nTurkey and Macedonia, refugees with legal\nstatus who meet the criteria can access\nexisting services and assistance including\nmaterial support.\n\n\n**• A rigorous system to identify the most**\n**vulnerable**  - including outreach, systematic\nprioritization, SOPs, regular home visits,\ncase-by-case reviews and continuous needs\nassessments – helps to target material\nassistance effi ciently.\n\n\n**• Vouchers and ATMs have worked well** in\nEgypt, Ecuador, Jordan and Syria to distribute\ncash transfers in a manner that respects the\ndignity of recipients and enables them to use\nassistance in a manner consistent with their\npriorities.\n\n\nA Global Survey – 2012 45\n\nMention: \"Global Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its results concern how assistance is used against beneficiary priorities.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000886:11:3", "occurrence_id": "a4773d3a8f98e767d3cedee3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Senior Protection\n\n\nAssociate, UNHCR Kenya), and Jedediah Rooney Fix (Senior Economist, UNHCR Regional Bureau for EHAGL).\n\n\n - The document benefited from helpful review and comments from several colleagues at UNHCR, World Bank, and the Immigration Policy Lab\n\n\n(IPL) at Stanford University, including Theresa Beltramo, John Wagacha Burton, Sonia Gomez, Anne Laako, Dania Khan, Philip Kinara, Sigrid\n\n\nWeber, and Precious Zikhali. Craig Loschmann, Florence Nana Pokuaah Nimoh, Ivy Wambui Njuguna, Kizitos Charloz Okisai, and Monika Sandvik\n\n\nprovided additional guidance and support during the preparation of this activity. REMIT Kenya conducted the household surveys for this activity\n\n\nwith support from the UNHCR Division of International Protection (DIP). The authors gratefully acknowledge the contribution of the team\n\n\nbehind the 2019 Shona Socioeconomic Survey <sup>[8]</sup> and the COVID-19 Rapid Response Phone Surveys <sup>[9]</sup> which allowed for a comparison of current\n\n\nsurvey data with the status before obtaining citizenship and the condition of the Shona population during the COVID-19 outbreak.\n\n\n - Recommended citation: Rahman, Masud; Okute, Samwel Ochieng; Fix, Jedediah Rooney (October 2024). _From Stateless to Citizens: the journey_\n\n\n_of the Shona community in Kenya_ (English). Nairobi: UNHCR.\n\n\n**[www.unhcr.org](http://www.unhcr.org/)** 12\n\nMention: \"2019 Shona Socioeconomic Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its data are the reference for later comparisons.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000886:11:3", "occurrence_id": "6e77084d3ef8918b25fc4e76", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Senior Protection\n\n\nAssociate, UNHCR Kenya), and Jedediah Rooney Fix (Senior Economist, UNHCR Regional Bureau for EHAGL).\n\n\n - The document benefited from helpful review and comments from several colleagues at UNHCR, World Bank, and the Immigration Policy Lab\n\n\n(IPL) at Stanford University, including Theresa Beltramo, John Wagacha Burton, Sonia Gomez, Anne Laako, Dania Khan, Philip Kinara, Sigrid\n\n\nWeber, and Precious Zikhali. Craig Loschmann, Florence Nana Pokuaah Nimoh, Ivy Wambui Njuguna, Kizitos Charloz Okisai, and Monika Sandvik\n\n\nprovided additional guidance and support during the preparation of this activity. REMIT Kenya conducted the household surveys for this activity\n\n\nwith support from the UNHCR Division of International Protection (DIP). The authors gratefully acknowledge the contribution of the team\n\n\nbehind the 2019 Shona Socioeconomic Survey <sup>[8]</sup> and the COVID-19 Rapid Response Phone Surveys <sup>[9]</sup> which allowed for a comparison of current\n\n\nsurvey data with the status before obtaining citizenship and the condition of the Shona population during the COVID-19 outbreak.\n\n\n - Recommended citation: Rahman, Masud; Okute, Samwel Ochieng; Fix, Jedediah Rooney (October 2024). _From Stateless to Citizens: the journey_\n\n\n_of the Shona community in Kenya_ (English). Nairobi: UNHCR.\n\n\n**[www.unhcr.org](http://www.unhcr.org/)** 12\n\nMention: \"COVID-19 Rapid Response Phone Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Allow current data to be compared with the baseline situation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000908:2:1", "occurrence_id": "94658adb4674982c8ffe4261", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: sôb so zdravotným znevýhodnením, osôb s vážnymi zdravotnými ťažkosťami alebo starších\nosôb, ktoré často čelia zvýšeným rizikám pre ich ochranu, ako aj zvýšeným zdravotným potrebám a\nnákladom. Od jeho spustenia 24. apríla do konca júna 2023 túto finančnú podporu získalo 6 618 osôb a\nprogram pokračoval ďalej.\n\n#### Prístup k zamestnaniu\n\n\nZabezpečenie spravodlivého a rovnocenného prístupu k dôstojnej práci je nevyhnutné pre úspešné\nsociálno-ekonomické začlenenie utečeniek a utečencov do hostiteľskej spoločnosti. Je chvályhodné, že\nutečenky a utečenci získajú okamžité právo zamestnať sa, keď získajú štatút dočasného útočiska na\nSlovensku. <sup>14</sup> [^14: Utečenci so dočasným útočiskom na Slovensku v súčasnosti nemôžu vykonávať samostatnú zárobkovú činnosť.] Podľa respondentov prieskumu Protection Profiling a Monitoring na Slovensku, 42 %\nuviedlo, že boli ekonomicky aktívni, <sup>15</sup> zatiaľ čo 30 % bolo v čase rozhovoru nezamestnaných. Medzi\nhlavné dôvody nezam\n\nMention: \"Protection Profiling a Monitoring\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Ses répondants déclarent leur activité économique.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000909:3:2", "occurrence_id": "5708fe858563e82e4292330d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  total), lack of shelter (59%), hunger\n(52%), robbery (43%), health problems (33%, mostly leg\ninjuries, fever, stomach infections) and documents stolen\n(22%). Among their most pressing needs, young migrants\nand refugees reported cash assistance, accommodation,\nlegal assistance, clothing and the possibility to continue the\njourney.\n\nOverall, 45% of all young migrant and refugee respondents\nanswered 'yes' to at least one of the six direct indicators of\nabuse, exploitation and violence based on their experience. <sup>13</sup> .\nOf the total number of young migrant and refugee who\nresponded positively to at least one of these indicators, 6%\nare female. Greece and Türkiye were the countries where\nmost of the reported violence and abuses occurred, followed\nby other countries of transit before the respondents reached\nBiH.\n\n\n_[Source: IOM’s DTM Europe — Flow Monitoring Surveys in Spain and Bosnia and Herzegovina](https://migration.iom.int/europe)_ _(2021)_\n\n\n\n_4_\n\nMention: \"DTM Europe\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its flow monitoring covers two transit countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000909:3:2", "occurrence_id": "b5416165df7f7acd25a16a16", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  total), lack of shelter (59%), hunger\n(52%), robbery (43%), health problems (33%, mostly leg\ninjuries, fever, stomach infections) and documents stolen\n(22%). Among their most pressing needs, young migrants\nand refugees reported cash assistance, accommodation,\nlegal assistance, clothing and the possibility to continue the\njourney.\n\nOverall, 45% of all young migrant and refugee respondents\nanswered 'yes' to at least one of the six direct indicators of\nabuse, exploitation and violence based on their experience. <sup>13</sup> .\nOf the total number of young migrant and refugee who\nresponded positively to at least one of these indicators, 6%\nare female. Greece and Türkiye were the countries where\nmost of the reported violence and abuses occurred, followed\nby other countries of transit before the respondents reached\nBiH.\n\n\n_[Source: IOM’s DTM Europe — Flow Monitoring Surveys in Spain and Bosnia and Herzegovina](https://migration.iom.int/europe)_ _(2021)_\n\n\n\n_4_\n\nMention: \"Flow Monitoring Surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Describe the transit stages before arrival in the destination country.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000912:11:0", "occurrence_id": "4903bdb1121e97fe51a21432", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: the main country of origin of asylum-seekers to the\nUnited States of America with 8,700 applications,\nfollowed by Guatemala (7,700), Mexico (7,500),\nHonduras (7,400), and China (7,300).\n\nHungary ranked fourth with 65,400 new asylum\napplications registered during the first half of 2015.\nThis constituted a 13-fold increase compared to\nthe first semester of 2014 (4,800 new applications)\nand a 79 per cent increase compared to the second\n(36,500). A large number of those applying for asylum in Hungary move onward to other countries\nin the European Union, where they are typically\nregistered again as asylum-seekers.\n\nA sharp shift in the composition and dynamics of registered asylum applications in Hungary\nwas observed during the reporting period. The\nfirst quarter of 2015 was dominated by asylum\napplicants originating from Serbia and Kosovo\n(S/RES/1244 (1999)), with more than 22,800 persons lodging such a claim, yet this figure dropped\nto less than 500 during the second quarter.\n\nIn contrast, the second quarter in Hungary witnessed a soaring number of asylum applicants from\nAfghanistan (13,600 claims, up from 4,000 during\nthe first quarter) and the Syrian Arab Republic\n(8,400, up from 2,400). These two countries together accounted for two-thirds of all asylum applications in Hungary during the second quarter.\nOverall, Serbia and Kosovo (S/RES/1244 (1999))\nwas the main place of origin for asylum applicants\nin Hungary during the reporting period (23,300\n\n\n**12** UNHCR Mid-Year Trends 2015\n\n\n\napplications), followed by Afghanistan (17,600), the\nSyrian Arab Republic (10,800), and Iraq (3,200).\n\nAs of December 2015, close to 2.2 million registered Syrian refugees in Turkey benefitted from the\nGovernment’s Temporary Protection Regime. In\n\nMention: \"UNHCR Mid-Year Trends 2015\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its figures give asylum seeker origins by period.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000913:4:0", "occurrence_id": "ad9c33a51d800b2b9b4628d1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Global Health Development/Eastern Mediterranean Public Health Network (GHD/EMPHNET)\n\n\nservices in these countries. The majority of stillbirths occurring annually are a result of poor\n\nmaternal care or inadequate management of pregnancy-related care. <sup>11</sup>\n\n\nThe stillbirth rate is a sensitive marker of the quality of care in pregnancy and childbirth, and\n\n\nthe strength of the health system. There is a paucity of quality information on the causes of\n\n\nstillbirth globally. Despite growing research projects on maternal-child health, still, there is\n\n\nlittle effort has been made in developing countries to explore the causes of stillbirths. At the\n\n\ncountry level, accurate data on stillbirths are urgently needed to enable tracking of the quality\n\n\nof antenatal and intrapartum care, understand the causes of these deaths, and thus identify\n\nareas for prevention. <sup>12</sup> Many countries in the region including Jordan do not include stillbirth\n\n\nin their vital statistics reporting system.\n\n\nThe rate of stillbirth in Jordan is 9.9 per 1000 total births. The main contributing fetal\n\n\nconditions of antepartum stillbirths were antepartum death of unspecified cause (33.7%), acute\n\n\nantepartum event (hypoxia) (33.7%), congenital malformations and chromosomal\n\n\nabnormalities (13.3%), and disorders related to the length of gestation and fetal growth\n\n(10.8%). <sup>13</sup> The main contributing maternal conditions of antepartum stillbirths included\n\n\ncomplications of the placental cord and membranes (48.7%), maternal complications of\n\n\npregnancy (23.1%), and maternal medical and surgical conditions (23.1%).\n\n\n**_Neonatal death and stillbirth audits_**\n\n\nNeonatal death and stillbirth audits are the processes of systematically capturing information\n\n\non the number and causes of all neonatal deaths and stillbirths and the potentially avoidable\n\n\nfactors linked to deaths, to provide data for\n\nMention: \"Neonatal death and stillbirth audits\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Systematically capture the number and causes of these deaths.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000920:7:1", "occurrence_id": "d06f5962c8ba460f4deeee6a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: </mark>** <mark>The humanitarian</mark>\n<mark>crisis in Afghanistan has led to widespread mental health and psychosocial problems. The ongoing</mark>\n<mark>emergencies like displacement, natural disasters, returns added more to the experience of Afghan</mark>\n<mark>communities emphasizing the urgent need for improved mental health and psychosocial support</mark>\n<mark>(MHPSS) services.</mark>\n\n<mark>[The 2024 Afghanistan Humanitarian Needs Response Plan quotes “over half of Afghans are distressed,](https://www.unocha.org/publications/report/afghanistan/afghanistan-humanitarian-needs-and-response-plan-2024-summary)</mark>\n<mark>and one in five individuals is suffering from severe forms of mental health problems.” Data collected</mark>\n<mark>from Community-Based Protection Monitoring (CBPM) reveals that 27 per cent of respondents have</mark>\n<mark>noticed negative changes in their family members' behaviors. Additionally, 45 per cent of respondents</mark>\n<mark>note experiencing stress, affecting their sleep and daily activities. In addition, over 50 per cent of refugee</mark>\n<mark>returnees have increased MHPSS needs, with noticeable changes in behavior such as stress, mood,</mark>\n<mark>and appetite changes, leading to aggression or self-harm. Death by suicide is  reported across various</mark>\n<mark>provinces, reflecting the profound mental health challenges faced by individuals. While specific data on</mark>\n<mark>suicide rates is not available, anecdotal evidence suggests higher rates in some regions. Potential risk</mark>\n<mark>factors for this disparity include cultural practices like child marriage, family violence, and</mark>\n<mark\n\nMention: \"Community-Based Protection Monitoring\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its data give the share of respondents reporting negative changes.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000921:24:0", "occurrence_id": "240b2bd18bfc392dfc028c9a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Governorate of Origin**\n\n\nThose from _Baghdad_ represent 84% in the first quarter of 2007 and 62% in the fourth quarter 2008. Those from\n\n_Ninewa_ increases from 3% in the first quarter 2007 to 18% the last quarter of 2008. Those from _Anbar_ are on\n\naverage 3%, _Basrah_ 4%, _Diyala_ 3% throughout the period.\n\nAll other governorates account for 1% or less of the active caseload.\n\n\nTwo Iraqi refugees during an English learning session at the UNHCR community centre in the Southern suburbs of Beirut.\n\n\n. Iraq Refugees Registration Data Comparative Analysis, 2007-2008. - 24\n\nMention: \"Iraq Refugees Registration Data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Its comparative analysis covers the two stated years.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000927:10:0", "occurrence_id": "0265822a92dd7a7a7ff6306f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **4 - Refugee community practices and cultural patterns for household and**\n**care arrangements**\n\nThe lack of available space has an impact on the mental health of the family. The informants\nindicated that according to culture, it is not accepted for children above 8 years old to share same\nbedroom with their parents but in a significant number of cases that is the only option they have\n(for example in caravans). They also indicate that mixing boys and girls above 8 years old is not\nculturally and religiously accepted, however there is no other possibility for refugee families.\n\n**5- Shelter maintenance**\n\nAccording to the Winterization Rapid Assessment conducted by REACH in collaboration with\nUNHCR in Zatari camp from 15-22 Aug 2016 <sup>2</sup> [^2: <u>http://www.reachresourcecentre.info/system/files/resource-</u>\n<u>[documents/reach_jor_factsheet_zaataricamp_winterizationassessment_shelternfis_aug2016_1.pdf](http://www.reachresourcecentre.info/system/files/resource-documents/reach_jor_factsheet_zaataricamp_winterizationassessment_shelternfis_aug2016_1.pdf)</u>\n##### 10], most of the respondents indicated they were living\nin caravan structures, with only 4% indicating they had mixed caravan and tent structures. Overall,\n57% of respondents did not consider their family’s shelter to be suitable for the 2016 winter.\nAmongst respondents who consider the shelter to be unsuitable, the most frequently cited reason\nis a leaking roof (54%).\nCapacity to maintain shelters: The majority of respondents (60%) considered their family’s ability\nto make repairs and conduct shelter maintenance themselves to be weak or very weak. 32% of\nthe respondents who perceived their family’s ability to perform repairs as very weak were female\nheaded households (FHH), compared to 17% of male headed households with the same\nperception. The most frequently cited reason for why respondents consider their families’ capacity\nto perform shelter maintenance as weak or very weak\n\nMention: \"Winterization Rapid Assessment\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Carried out in collaboration with the agency for this camp.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000932:0:3", "occurrence_id": "60dcfde1dde10ea94d488eed", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: g. IBCs, the Global Network on Migration, the Global Refugee Forum and other developmentrelated platforms etc.) Since 2020, the IBC also assumes the functions of a regional UN Network on Migration and will support actions to assist Member States in implementing the GCM at the\n[country and regional levels in alignment with the UN Development System. More information can be found here.](https://www.unece.org/runcwelcome/issue-based-coalitions/large-movements-of-people.html)\n2 Source: <u>[UN DESA, Trends in International Migrant Stock: The 2019 Revision Migrants by Age and Sex. United Nations, New York, 2019.](https://www.un.org/en/development/desa/population/migration/data/estimates2/estimates19.asp)</u> For more data, please visit\n<u>[https://migrationdataportal.org/?i=stock_abs_&t=2019](https://migrationdataportal.org/?i=stock_abs_&t=2019)</u> [and https://www.unhcr.org/data.html.](https://www.unhcr.org/data.html)\n3 A new normal: UN lays out roadmap to lift economies and save jobs after COVID-19, United Nations Press Release, 27 April 2020.\n\nMention: \"Trends in International Migrant Stock\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its revision classifies migrants by age and sex.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000936:72:4", "occurrence_id": "f86277d7fcf1f13e8aba004e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: a Digital del permiso laboral: La DGME en\ncoordinación con Ministerio de Trabajo y Seguridad Social y con\nel apoyo del ACNUR, creará un mecanismo digital de consulta\nque permita a las personas refugiadas certificar ante posibles\nempleadores la validez de su permiso laboral.\n\n- Programa intermediación laboral para personas refugiadas:\n(a) El Ministerio de Trabajo y Seguridad Social brindará acceso\nefectivo a los programas de intermediación laboral como www.\nbuscoempleo.go.cr; Programa EMPLEATE, Mi Primer Empleo\ny PRONAE 4x4, además de los programas de apoyo a la\nmicroempresa. (b) La Inspección Nacional del Trabajo contemplará\nen sus procedimientos a las personas refugiadas con el objetivo\nde asegurar que sus derechos laborales y las obligaciones\npatronales sean reconocidos en igualdad de condiciones que los\nnacionales. (c) También se llevarán a cabo acciones informativas\nsobre la legislación vigente para mejorar el reconocimiento de\nlos documentos de identidad de las solicitantes de refugio y\nrefugiadas y reducir la discriminación y xenofobia durante los\nprocesos de reclutamiento y selección de personal.\n\n- Proyecto fomento del emprendedurismo: El Ministerio de\nEconomía, Industria y Comercio incluirá los emprendimientos\nliderados por personas refugiadas en el Registro de\nEmprendedores y fomentará su participación en encuentros\nempresariales y otras actividades organizadas por el Ministerio\npara fortalecer la micro y peque\n\nMention: \"Registro de\nEmprendedores\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"inform\", \"summary\": \"Incluirá los emprendimientos liderados por personas refugiadas.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000937:4:0", "occurrence_id": "d6fcb1721ffc1ee3ce0201af", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: DTM Mali – Avril 2020\n\n##### **PERSONNES DÉPLACÉES INTERNES**\n\n\nLes opérations de collecte et de mise à jour des données menées dans le cadre du programme DTM\nmontrent une augmentation du nombre de personnes déplacées au Mali entre janvier et avril 2020. Le\nnombre de PDIs est en effet passé de 207 751 personnes (rapport CMP de décembre 2019) à **250 998** en\navril 2020 soit une augmentation de **43 247** individus. Cette augmentation est due à des violences variées\ndans les Régions de Mopti, Ségou, Tombouctou, Gao et Ménaka et la bande frontalière Mali-Burkina FasoNiger. En effet, ces violences ont provoqué le déplacement de populations de leurs villages et hameaux\npour trouver refuge dans les localités où la situation sécuritaire semble plus calme.\n\n\nTableau 1 : Déplacés Internes au Mali par région\n\n|Région|Ménage|Femme|Homme|Total Individus|Taille moyenne|\n|---|---|---|---|---|---|\n|Bamako|663|1 570|1 121|2 691|4|\n|Gao|11 002|34 625|28 008|62 633|6|\n|Kayes|381|861|763|1 624|4|\n|Kidal|222|486|404|890|4|\n|Koulikoro|413|1 053|932|1 985|5|\n|Ménaka|3 486|11 719|9 272|20 991|6|\n|Mopti|18 608|53 772|48\n\nMention: \"DTM Mali\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Ses opérations de collecte dénombrent les personnes déplacées internes.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000944:6:5", "occurrence_id": "bf7123387ca023011beae87a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: b418e56fd08dc014cb52d%7Ce5c37981666441348a0c6543d2af80be%7C0%7C0%7C638386676157333620%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=pjnZexqiC%2BgWxbLpt2pAil8AH4xGMEcdHYumk3%2BtOhY%3D&reserved=0)</u>\n\n\nWhile the government recognized the need to provide additional services and logistical support to remote\nlocations, including in refugee-hosting areas, the limited resources and capacity of NIRA have led to\nsignificant delays in the issuance of birth certificates for refugees and hosts. To improve the timeliness\nof future birth registration services, the Government is working to establish links between the medical\ndatabases of hospitals registering births and the national NIRA database system. Uganda is also developing\na Civil Registration and Vital Statistics (CRVS) strategy with a view to providing guidelines for the delivery\nof civil registration services and the production of vital statistics, including for refugees. It is noted that\nlimited birth registration and certification for refugees, including adults born in Uganda, places them at risk\nof statelessness.\n\n\n**2.5** **Justice and security**\n\n\nRefugees are entitled to the same rights as nationals regarding access to justice, including access to legal\nassistance under applicable laws in Uganda, in line with the Refugees Act and its Regulations. The level of\nsecurity enjoyed by refugees has been comparable to that enjoyed by\n\nMention: \"medical\ndatabases of hospitals registering births\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"inform\", \"summary\": \"The government seeks to link them to the national registration system.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000944:6:5", "occurrence_id": "e3dde907f0d8d48b428d0c8b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: b418e56fd08dc014cb52d%7Ce5c37981666441348a0c6543d2af80be%7C0%7C0%7C638386676157333620%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=pjnZexqiC%2BgWxbLpt2pAil8AH4xGMEcdHYumk3%2BtOhY%3D&reserved=0)</u>\n\n\nWhile the government recognized the need to provide additional services and logistical support to remote\nlocations, including in refugee-hosting areas, the limited resources and capacity of NIRA have led to\nsignificant delays in the issuance of birth certificates for refugees and hosts. To improve the timeliness\nof future birth registration services, the Government is working to establish links between the medical\ndatabases of hospitals registering births and the national NIRA database system. Uganda is also developing\na Civil Registration and Vital Statistics (CRVS) strategy with a view to providing guidelines for the delivery\nof civil registration services and the production of vital statistics, including for refugees. It is noted that\nlimited birth registration and certification for refugees, including adults born in Uganda, places them at risk\nof statelessness.\n\n\n**2.5** **Justice and security**\n\n\nRefugees are entitled to the same rights as nationals regarding access to justice, including access to legal\nassistance under applicable laws in Uganda, in line with the Refugees Act and its Regulations. The level of\nsecurity enjoyed by refugees has been comparable to that enjoyed by\n\nMention: \"NIRA database system\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The national system to which hospital databases would be linked.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000946:5:1", "occurrence_id": "a4737be03498de1c86f724f6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  going through the second layer of information, which\nconsists of supplementary information and verification achieved through contact to protection cluster partners in the field. Finally,\nthe data is regularly triangulated with other humanitarian databases.\n\nThe data presented in this report consist of a combination of unverified open source data, cross referenced open source data, eye\nwitness accounts and incidents verified by protection cluster partners, and thus the data as a whole should not be treated as\nindependently verified. CIMP works continuously on updating and verifying data, and can be contacted anytime with further\ninformation on incidents as well as enquiries on data and level of verification of specific incidents.\n\nCIMP monitors civilian impact that occurs after an incident of armed violence have taken place, thus CIMP numbers on\ndisplacement, loss of livelihood and restriction of movements/obstruction to flight only covers households that have experienced\na direct impact from armed violence, e.g. a house destroyed or a vehicle hit. Therefore, CIMP data does not include full numbers of\npeople being displaced, loosing livelihood or experiencing restricted freedom of movement/obstruction to flight, where numbers\nare naturally much higher than what is captured by CIMP.\n\nCivilian impact incidents recorded by CIMP are divided into direct and indirect impact, with associated direct and indirect\nprotection implications. Direct impact includes incidents in which individuals or households are directly affected by the incident,\ne.g. damage to houses and farms, damage to markets and local businesses, impact on vehicles or as well as exposure to UXOs and\narmed conflict generating casualties. Indirect impact can broadly be defined as incidents of armed violence impacting on\ninfrastructure and basic services and in turn restricting access of civilians to various vital services, infrastructure and goods, e.g.\nhealthcare, education, food and water and infrastructure. Due to the nature of the indirect impact, the number of households\n\n\n###### **6**\n\nMention: \"CIMP data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Covers only direct impact, not displacement or livelihood losses.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000948:10:1", "occurrence_id": "68e5370f8e0a465d860d7b76", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: un taux de\nprévalence de l’anémie à 48,1% et de mortalité de 0,62 décès/10000 personnes/jour. Sur la même cible, l’enquête\nSMART 2011 a indiqué un taux de MAG de 16 % dans ces\ndeux regions, ce qui dépasse le seuil « critique » (15 %) définit\npar l’OMS.\n\n\nLa mission a observé des signes de malnutrition (insuffisance\npondérale, retard de croissance) dans certains villages visités.\nLe PAM, l’UNICEF et le HCR mettent en œuvre des programmes complémentaires de lutte contre la malnutrition à\ntravers des partenaires opérationnels tels que la DRSP\n(Délégation Régionale de la Santé Publique), ASAD\n(Association pour l’Appui au Développement) AHA (African\nHumanitarian Action), FICR (Fédération Internationale de la\nCroix Rouge), IMC (International Medical Corps). Ces ONG et\nAssociations partenaires mènent le dépistage de la malnutrition en stratégie avancée dans leurs zones respectives d’intervention. Elles animent des séances de sensibilisation pour la\nlutte contre la malnutrition dans les communautés et organisent des séances de démonstrations culinaires. Elles appuient\ndes formations sanitaires sur le suivi des personnes dépistées\nmalnutries, supervisent le respect du protocole national de\nnutrition.\n\n\nDe 2008 à 2012, le suivi de la nutrition dans les deux régions\nà travers les enquêtes SMART et EDS-MICS n’a pas toujours\nciblé à la fois les populations hôtes et réfugiées\n\nMention: \"enquête\nSMART 2011\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Ses taux de malnutrition aigüe modérée dépassent le seuil critique.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000948:10:1", "occurrence_id": "889c8e0a024bcecfa7c2512f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: un taux de\nprévalence de l’anémie à 48,1% et de mortalité de 0,62 décès/10000 personnes/jour. Sur la même cible, l’enquête\nSMART 2011 a indiqué un taux de MAG de 16 % dans ces\ndeux regions, ce qui dépasse le seuil « critique » (15 %) définit\npar l’OMS.\n\n\nLa mission a observé des signes de malnutrition (insuffisance\npondérale, retard de croissance) dans certains villages visités.\nLe PAM, l’UNICEF et le HCR mettent en œuvre des programmes complémentaires de lutte contre la malnutrition à\ntravers des partenaires opérationnels tels que la DRSP\n(Délégation Régionale de la Santé Publique), ASAD\n(Association pour l’Appui au Développement) AHA (African\nHumanitarian Action), FICR (Fédération Internationale de la\nCroix Rouge), IMC (International Medical Corps). Ces ONG et\nAssociations partenaires mènent le dépistage de la malnutrition en stratégie avancée dans leurs zones respectives d’intervention. Elles animent des séances de sensibilisation pour la\nlutte contre la malnutrition dans les communautés et organisent des séances de démonstrations culinaires. Elles appuient\ndes formations sanitaires sur le suivi des personnes dépistées\nmalnutries, supervisent le respect du protocole national de\nnutrition.\n\n\nDe 2008 à 2012, le suivi de la nutrition dans les deux régions\nà travers les enquêtes SMART et EDS-MICS n’a pas toujours\nciblé à la fois les populations hôtes et réfugiées\n\nMention: \"EDS-MICS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Son suivi n'a pas toujours ciblé à la fois hôtes et réfugiés.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000950:10:0", "occurrence_id": "f8c4e4045d8b8a2ddb2795ea", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: UNHCR’S MOST UNDERFUNDED SITUATIONS IN 2021\n\n## The Iraq and Syria situations\n\n\n\nUNHCR estimates that 3.3 million people are in\n\nneed of critical assistance to help them prepare\n\nfor the forthcoming winter. This will be the 11 <sup>th</sup>\n\nconsecutive winter in displacement for some, and\n\nmany continue to face increased hardships due to\n\nthe economic situation and the COVID-19\n\npandemic. Most of the planned interventions will\n\nbe in the form of cash assistance, with critical\n\nfunding gaps affecting Syrian and Iraqi refugees,\n\nIDPs and returnees in Egypt, Iraq, Jordan,\n\nLebanon and for cross-border activities from\n\nTurkey to northwest Syria. $88 million is still\n\nrequired for winter needs across the region.\n\n\nHundreds of thousands of refugees, IDPs and\n\nreturnees still need cash assistance to meet their\n\nbasic needs. UNHCR is seeking $40 million,\n\nincluding $12 million in Iraq, $8 million in Jordan\n\nand $20 million in Lebanon. Without this vital help,\n\nvulnerable families may turn to additional negative\n\ncoping mechanisms or be plunged into deeper\n\n\n_Destroyed homes are noticeable in Bardiya village and its outskirts, Iraq,_\n_as many people lack sufficient financial means to rebuild their homes._\n_© UNHCR/Firas Al-Khateeb_\n\n\n\nvulnerability. In Lebanon, socioeconomic\n\ndeterioration accompanied by hyperinflation,\n\ndeclining subsidies and rising insecurity has left\n\npeople angry and disillusioned. Thefts and\n\nclashes over goods – particularly food, medicine\n\nand fuel – are increasing. The situation is\n\nextremely challenging for everyone as rising\n\nnumbers of Lebanese are falling into poverty, but\n\nit is even worse for Syrian refugees.\n\n\nThe preliminary results of the 2021 Vulnerability\n\nAssessment of Syrian Refugees indicate that 88\n\nper cent of Syrian refugees are in extreme\n\npoverty. Cash assistance could reach 170,000\n\nrefugee families and prove critical for the survival\n\nof many.\n\n\nInside Iraq, 6,100 Syrian refugee families, 1,750\n\nrefugee and asylum-seeker families of other\n\nnationalities,\n\nMention: \"2021 Vulnerability\n\nAssessment of Syrian Refugees\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its preliminary results give the share in extreme poverty.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000957:4:0", "occurrence_id": "de065fc3df039367403ca41e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **RÉSUMÉ**\n\n\n**Contexte**\nLe Lac est une province du Tchad, pays de l’Afrique centrale, qui partage ses frontières avec\nle Niger, le Nigeria et le Cameroun. Depuis 2014, la province est constamment en proie à une\ninsécurité déclenchée par de violentes attaques des groupes armés non étatiques et des\nopérations militaires. En 2020, la situation sécuritaire s'est dégradée, provoquant une\nrecrudescence de nouveaux mouvements de population. En octobre 2021, l'OIM estime que\nplus de **457 000** personnes en situation de déplacement se trouvent actuellement dans la\nprovince du Lac, dont environ 400 000 sont des personnes déplacées internes (PDI) <sup>1</sup> [^1: Organisation Internationale pour la Migration (OIM), Octobre 2021, Displacement Tracking Matrix (DTM).\n<u>[Tchad — Dashboard de Déplacement 16 (Août-Septembre 2021)](https://displacement.iom.int/reports/tchad-%E2%80%94-dashboard-de-d%C3%A9placement-16-aout-septembre-2021%C2%A0?close=true)</u> <u>| Displacement (iom.int)</u>] .\n\n\n**Situation**\nLe climat hostile, le niveau de pauvreté extrême et le manque d'infrastructures dans toute la\nprovince du Lac, rendent la plupart des communautés d'accueil incapables d’apporter une\nassistance de base aux personnes déplacées qui arrivent. Ce contexte socio-économique\nprécaire rend les personnes déplacées particulièrement vulnérables, les exposant aux\naffrontements intercommunautaires dont l’une des raisons est l’amenuisement des\nressources naturelles. Ces populations dépendent presque entièrement de l’assistance\nfournie par les acteurs humanitaires, en abris, nourriture et eau.\n\n\nLes partenaires du Cluster Abris/AME ont été en mesure de fournir plus\n\nMention: \"Displacement Tracking Matrix\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Son tableau de bord couvre la période indiquée pour ce pays.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000958:3:0", "occurrence_id": "ea84b7441a3dadab4fa51969", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: delayed disclosures from GBV survivors who might need more time to accept and process the abuse\nthey were subjected to before reporting to specialized service providers. Other reports justify this\npercentage with the heavy caseload GBV service providers are currently facing.\n\n**Security protection services** and **legal assistance services** constitute the highest percentage of most\ndeclined referrals by GBV survivors, accounting to 44 percent and 38 percent respectively. Data\ntriangulated with Referrals Information Management System report (RIMS) on women and girls’ barriers\nto access humanitarian services, <sup>5</sup> traditional **gender norms and cultural barriers** and **access to**\n**information and services** were identified as the two main barriers identified by female key informants.\nAccording to interviewed women, gendered cultural norms often prevent women and girls from\naccessing educational, professional, or psychosocial support services including access to legal and justice\nservices. Furthermore, many intersectional dimensions further hamper vulnerable women and girls in\nneed of humanitarian assistance from accessing assistance or services, notably due to their legal or\nimmigration status, their nationality or the absence of a nationality, and/or sexual orientation as well.\n\nAdding to this, data triangulated with a study by UN Women and partner organisations on **Access to**\n**Justice for Sexual and Gender Based Violence case of Syrian Refugees in Lebanon** <sup>**6**</sup> **,** It was commonly\nreported by both interviewed men and women that they tend to resolve the GBV reports/ incidents\ninternally in case the GBV is perpetrated within the household, before bringing the case to court. This is\nreflective of social pressures, fears associated with a lack of documentation, financial constraints to\nengaging in the legal system, and the fact that informal justice processes take less time to reach an\noutcome, even if the outcome is not in the favor of women and/ or\n\nMention: \"Referrals Information Management System report\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its triangulated data describe access barriers for women and girls.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000967:9:2", "occurrence_id": "6c4b5d8645507a4539e208c4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: sous de 1 200 mm/an.\n\nD. Aires protégées\n[La Base de données mondiale des Aires Protégées (lien en page 11)](https://protectedplanet.net/)\ncite 52 aires protégées dans le pays. Les aires protégées sont des\npoints névralgiques de la biodiversité, en plus d’abriter les moyens de\nsubsistance et les ressources naturelles utilisées par les communautés\nlocales pour construire leurs logements.\nParmi ces aires, il y a 5 sites du patrimoine mondial naturel ou mixte\n(voir lien en page 11), 4 sites Ramsar (zones humide d’importance\ninternationale) et 3 Réserves de la Biosphère de l’UNESCO-MAB.\n\n\nE. Statut administratif\nLes structures administratives comprennent des entités territoriales\nautonomes, semi-autonomes, décentralisées et déconcentrées régulées\npar différentes lois. La loi du 28 février 2015 détermine le découpage\nadministratif actuel du pays. Le territoire congolais est divisé désormais\nen vingt-six provinces, dont l’entité urbaine de Kinshasa.\n\n\n\n\n\n\n\n\n\ncommons\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n10 / 60 Fiche de réponse abris détaillée : RDC (SUD-EST) | Cultures constructives locales pour des habitats durables et résilients\n\nMention: \"Base de données mondiale des Aires Protégées\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Elle cite 52 aires protégées dans le pays considéré.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000968:8:1", "occurrence_id": "5b5b8c802960b2b55b68708a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \nrefugee enrolment is growing. But the Government wants to further expand access, having\ncommitted to increase enrolment in primary,\nsecondary and tertiary education at the Leaders’\nSummit on Refugees in New York in 2016.\n\n**•** Ethiopia hosts more than 905,000 refugees\nand asylum seekers\n\n**•** The number of refugee children in the\ncountry’s primary schools has increased\nfrom 118,275 in the 2016/17 academic year to\n132,563 in 2017/2018\n\n**•** The gross primary school enrolment rate\nin 2017/2018 was 72%. The government’s\npledged target is 75%.\n\n\n_(UNHCR, December 2018)_\n\n\n\nHow to replicate\nthis practice\n\nIncluding refugees in national education statistics can be replicated in other countries,\nhelping the world move towards reaching the\nSustainable Development Goal of providing\naccess to education for every child. Here are\nthe four steps towards replicating Ethiopia’s\nRefugee EMIS in other countries:\n\n\n**~~STEP 1:~~** Advocate for government support.\nIn Ethiopia, ARRA championed the cause and\ndrove the process, which was then jointly supported by UNHCR and UNICEF.\n\n\n**~~STEP 2:~~** Use tools that already exist. The MoE\nalready owned the original EMIS tool in Ethiopia\nand simply adapted it to include refugees.\n\n\n**~~STEP 3:~~** Replicate the Refugee EMIS feature\nthat allows UNHCR to track progress every academic year, showing the proportion of children\nleft with no access to education.\n\n\n**~~STEP 4:~~** If there is accurate data, then key\ninformation can be linked to planning for other\nrelated activities, such as livelihoods.\n\n\nIt is important for any country looking to replicate the Ethiopian Refugee EMIS to bear in mind\nthat it suits a context where refugees are still living in camps or specific districts, where it is easier\nto track which schools refugees attend. It does\nnot\n\nMention: \"national education statistics\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Including refugees there can be replicated in other countries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000968:8:1", "occurrence_id": "3f7032a0189c836a2c031b46", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \nrefugee enrolment is growing. But the Government wants to further expand access, having\ncommitted to increase enrolment in primary,\nsecondary and tertiary education at the Leaders’\nSummit on Refugees in New York in 2016.\n\n**•** Ethiopia hosts more than 905,000 refugees\nand asylum seekers\n\n**•** The number of refugee children in the\ncountry’s primary schools has increased\nfrom 118,275 in the 2016/17 academic year to\n132,563 in 2017/2018\n\n**•** The gross primary school enrolment rate\nin 2017/2018 was 72%. The government’s\npledged target is 75%.\n\n\n_(UNHCR, December 2018)_\n\n\n\nHow to replicate\nthis practice\n\nIncluding refugees in national education statistics can be replicated in other countries,\nhelping the world move towards reaching the\nSustainable Development Goal of providing\naccess to education for every child. Here are\nthe four steps towards replicating Ethiopia’s\nRefugee EMIS in other countries:\n\n\n**~~STEP 1:~~** Advocate for government support.\nIn Ethiopia, ARRA championed the cause and\ndrove the process, which was then jointly supported by UNHCR and UNICEF.\n\n\n**~~STEP 2:~~** Use tools that already exist. The MoE\nalready owned the original EMIS tool in Ethiopia\nand simply adapted it to include refugees.\n\n\n**~~STEP 3:~~** Replicate the Refugee EMIS feature\nthat allows UNHCR to track progress every academic year, showing the proportion of children\nleft with no access to education.\n\n\n**~~STEP 4:~~** If there is accurate data, then key\ninformation can be linked to planning for other\nrelated activities, such as livelihoods.\n\n\nIt is important for any country looking to replicate the Ethiopian Refugee EMIS to bear in mind\nthat it suits a context where refugees are still living in camps or specific districts, where it is easier\nto track which schools refugees attend. It does\nnot\n\nMention: \"Refugee EMIS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"The national model is presented as replicable elsewhere.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000970:1:0", "occurrence_id": "313cc3d7d37b39b97b01be29", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Independence:** Activities must be autonomous from the political, economic, military or other\nobjectives that any actor may hold with regard to areas where humanitarian action is being\nimplemented\n\n**Respect:** Actions and responses of all actors will be guided by respect for the choices, wishes, rights\nand the dignity of the persons of concern.\n\n**Confidentiality:** All information regarding persons of concern with protection risks shall be kept\nconfidential. No identifying information on specific cases discussed during Working Group will be\nrevealed in public sources.\n\n\nThe activities of members of the PWG shall be guided by the protection principles, namely:\n\n       - Do no harm principles through avoiding exposing people to further harm as a result of your\nactions\n\n       - Access to assistance through ensuring people’s access to assistance – in proportion to need\nand without discrimination\n\n       - Physical and psychological protection to persons of concern against violence and coercion\n\n       - Rights-based approach through assisting persons of concern to claim their rights, access\navailable remedies and recover from the effects of abuse.\n\n**Activities and tasks**\n\nWithin this framework, the PWG tasks will include:\n\n**1.** <u>Improving coordination and information sharing to inform humanitarian response and strategic</u>\n<u>decision making:</u>\n\na. Ensure information sharing among members of PWG;\nb. Ensure service delivery is coordinated among Protection actors as reflected in the 3RP and\n\nother strategic planning documents;\nc. Maintain sector information including by regularly updating 4W, dashboards and other\n\ntools (ActivityInfo);\nd. Identify and analyze trends and emerging issues in coordination with other sectors;\ne. Identify and analyse gaps (across and within sectors) as well as find coordinated solutions\n\nwhile avoiding duplication of efforts\nf. Develop a national PWG strategy and work plan to address protection priorities at the\nnational level. The work plan will include a set of prioritized activities and outputs and\nspecific timeframes and responsible and collaborating\n\nMention: \"ActivityInfo\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Tool for updating sector information and dashboards.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000972:35:0", "occurrence_id": "471d99349db51aabf415c9b0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ## ANNEX: METHODOLOGY\n\n\n#### INTRODUCTION\n\nIDMC is working in several ways to increase understanding of disaster related displacement. Since 2008,\nIDMC has tracked disaster related displacement by\nhand-screening disaster event data to create high-quality displacement estimates. In 2013, work commenced on\nmodelling future displacement risk. In 2014, modelled\nhistoric displacement estimates were introduced to estimate historic risk for a much wider range of years, 1970\nto 2012. All of these approaches help to better inform\nthe state of disaster related displacement risk throughout the world. Each approach has particular strengths\nand weaknesses that together help paint a more complete picture of disaster-related displacement.\n\n\nThe historic modelled displacement and displacement\nrisk figures have been broken down into the following\nunits of analysis:\n\n     - global\n\n     - regional (Latin America and the Caribbean, Southeast Asia, South Asia and the Horn of Africa)\n\n     - national\n\n\nWith the exception of the Horn of Africa, the global, regional and national displacement risk estimates\nare based upon a probabilistic risk model similar to\nthe one used throughout the 2015 GAR and its predecessors. Due to the complex causality and time scale,\ndrought-related displacement risk in the Horn of Africa\nwas calculated using a system dynamics model.\n\n\n38 REPORT\n\n\n#### DISPLACEMENT RISK ASSOCIATED WITH SUDDEN-ONSET HAZARDS: USING HISTORIC DISPLACEMENT ESTIMATES\n\nIDMC’s DiDD, annual disaster displacement estimate\ndatabase for 2008-2013 has been augmented with historic modelled displacement estimates created by using\nexisting disaster loss data from 1970 to 2012. These\nmodelled displacement figures are provided on a country/year/event type basis, and, where possible, also at a\nsub-national resolution. They have been constructed as\nanalogs to the actuarial concept of AAL (Average Annual Loss, or Average Annual Displacement in this case).\nThese figures are also carried forward as risk estimates\nfor 2015-2020.\n\n\nLoss data for this analysis was obtained from EM-DAT\nand Des\n\nMention: \"disaster event data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Screened manually since that year to produce displacement estimates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000972:35:0", "occurrence_id": "fde785aaa0bafd057a41452c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ## ANNEX: METHODOLOGY\n\n\n#### INTRODUCTION\n\nIDMC is working in several ways to increase understanding of disaster related displacement. Since 2008,\nIDMC has tracked disaster related displacement by\nhand-screening disaster event data to create high-quality displacement estimates. In 2013, work commenced on\nmodelling future displacement risk. In 2014, modelled\nhistoric displacement estimates were introduced to estimate historic risk for a much wider range of years, 1970\nto 2012. All of these approaches help to better inform\nthe state of disaster related displacement risk throughout the world. Each approach has particular strengths\nand weaknesses that together help paint a more complete picture of disaster-related displacement.\n\n\nThe historic modelled displacement and displacement\nrisk figures have been broken down into the following\nunits of analysis:\n\n     - global\n\n     - regional (Latin America and the Caribbean, Southeast Asia, South Asia and the Horn of Africa)\n\n     - national\n\n\nWith the exception of the Horn of Africa, the global, regional and national displacement risk estimates\nare based upon a probabilistic risk model similar to\nthe one used throughout the 2015 GAR and its predecessors. Due to the complex causality and time scale,\ndrought-related displacement risk in the Horn of Africa\nwas calculated using a system dynamics model.\n\n\n38 REPORT\n\n\n#### DISPLACEMENT RISK ASSOCIATED WITH SUDDEN-ONSET HAZARDS: USING HISTORIC DISPLACEMENT ESTIMATES\n\nIDMC’s DiDD, annual disaster displacement estimate\ndatabase for 2008-2013 has been augmented with historic modelled displacement estimates created by using\nexisting disaster loss data from 1970 to 2012. These\nmodelled displacement figures are provided on a country/year/event type basis, and, where possible, also at a\nsub-national resolution. They have been constructed as\nanalogs to the actuarial concept of AAL (Average Annual Loss, or Average Annual Displacement in this case).\nThese figures are also carried forward as risk estimates\nfor 2015-2020.\n\n\nLoss data for this analysis was obtained from EM-DAT\nand Des\n\nMention: \"EM-DAT\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Source of the analysis's loss data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000978:11:1", "occurrence_id": "553cfb0c0369887e1b066c57", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  situation creates serious humanitarian challenges, especially on the Dodecanese islands, where new arrivals are detained for any amount of time in police cells, port facilities or\nin other temporary locations. Newcomers could be detained for anytime ranging from several hours\nto several days, depending on the location and the police capacity. Furthermore, the authorities are\noften unable to provide sufficient food and other assistance. <sup>43</sup> [^43: UNHCR, Press Release, 8 October 2014, <u>http://goo.gl/w0ZkoU.</u>] Some limited supplementary assistance\n(mainly Non Food Items - NFIs) is provided by UNHCR. In practice, in order to release the pressure, the\npolice hastily conduct first reception procedures and persons concerned are released without proper\ndetermination of nationality or assessment of medical or specific needs. After release, these persons\nare effectively left to look after themselves, without reception, support, or access to emergency shelter. <sup>44</sup> [^44: For more details on second reception asylum procedures, see relevant section on page 19.]\n\n\n38 According to FRS official data, 575 Syrians were referred to the police (58.2 per cent of all referrals) during the first nine\nmonths of 2014, while 800 have been referred from the Mobile Units of Samos and Lesvos to the police (27 per cent of all\nreferrals).\n\n\n39 UNHCR has issued advisories which remain valid at present for Syria, Iraq, northern Mali, South Sudan, the Kivus and\nadjacent areas of the Democratic Republic of Congo, Central African Republic, Somalia (under certain conditions) and northeastern Nigeria.\n\n\n40 In 2013 Greece assured the ECtHR that no forced returns take place to Syria, Somalia, the State of Palestine, Eritrea,\nMauritania and Myanmar and that individuals from these countries are released after identification and registration:\n<u>http://goo.gl/eOtaYI.</u>\n\n\n41 Rhodes, Kos, Tilos\n\nMention: \"FRS official data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its figures give police referrals by origin.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000981:0:2", "occurrence_id": "ffc6fd696b7322670e44b458", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: :** **18 m2**(3m x 6m)|\n|**SHELTER DENSITY**<br>**3.6 m2**per person|**SHELTER DENSITY**<br>**3.6 m2**per person|\n\n\n**DIRECT COST** **USD** **700**\n\n\n**PROJECT COST** **USD** **1,000**\n\n\n<u>[*IOM Nigeria Displacement Report, Round 41, Baseline Assessment in Northeast Nigeria](https://dtm.iom.int/reports/nigeria-north-east-displacement-report-41-june-2022)</u>\n\n[**Humanitarian Response Plan, Nigeria, 2022 (February 2022)](https://reliefweb.int/report/nigeria/nigeria-humanitarian-response-plan-2022-february-2022)\n\n\n\nthrough cash-for-work programs.\n\n\n\n<mark>PROJECT</mark>\n\n\n<mark>CONTEXT</mark>\n\n\n\nPLANNING\n\n\n\n<u>CONFLICT</u>\n<mark>TIMELINE</mark> <u><mark>JUL</mark></u> <u><mark>JAN</mark></u> <u><mark>MAR</mark></u> <u><mark>APR</mark></u> <u><mark>MAY</mark></u> <u><mark>JUN</mark></u> <u><mark>JUL</mark></u>\n\n\n\n<u><mark>DEC</mark></u>\n\n\n\n<u><mark\n\nMention: \"IOM Nigeria Displacement Report\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Son évaluation de référence dans le nord-est donne les coûts par personne.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000982:4:0", "occurrence_id": "997ebe0eb9f3144e8baa0eb5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: including pathways to the workstreams and groups to elevate issues to Senior Official and Ministerial\nlevel. The existing workstreams and groups could go through a streamlining exercise re-establishing terms\nof reference, expected level of participation from Member States and links to other groups as well as the\nCo-Chairs and the agenda setting process. The expectations on the Co-Chairs of the working groups could\nalso be clarified. The Regional Support Office role could be strengthened and resourced to further support\nthe workstreams and the groups to communicate with the Co-Chairs and support in other secretariat\nfunctions providing a sense of continuity to the process and agenda setting. This would support improved\nconnections from working level considerations that are aligned with Member State priorities to the\nhigher-level agenda set by the Co-Chairs.\n\n\n**Recommendation 6: Harness Private Sector Partnerships to deliver on the Bali Process commitments**\n\n\nThe latest Global Estimates on Modern Slavery indicate that 50 million people were living in modern\nslavery in 2021. Of these people, 28 million were in forced labour globally, over 15 million of which were\nin the Asia-Pacific region. Migrant workers are three times more likely to become victims of forced labour.\nUnfortunately, the number of people in modern slavery has risen significantly in the last five years, with\nbusiness being responsible for 86 per cent of forced labour cases.\n\n\nThe Bali Declaration stresses the importance of engaging with the private sector to “expand legal and\nle _gitimate opportunities for labour migration and to combat human trafficking and related exploitation,_\n_including by promoting and implementing humane and non-abusive labour practices throughout the_\n_international supply chains.”_\n\n\nTo further deliver on the Bali Process commitments, the following suggestions could be considered. Given\nthe role of the private sector in modern slavery, harnessing private sector partnerships is key. This can be\ndone through strengthening private sector capacities to address some of the root causes of trafficking,\npeople smuggling and\n\nMention: \"Global Estimates on Modern Slavery\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its estimates give the number of people in modern slavery.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000984:2:0", "occurrence_id": "9c6a384d6a49de74f389f620", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **2018 PROGRESS REPORT**\nRegional Refugee & Resilience Plan (3RP) 2018-2019\n###### SELECTED REGIONAL SECTOR ACHIEVEMENTS\n\n\nFull inter-agency 3RP dashboards as of 30 June are available here: <u>http://www.3rpsyriacrisis.org/wp-</u>\n<u>content/uploads/2018/09/3RP-Regional-Quarterly-Dashboards-June</u> ~~<u>-2018.pdf</u>~~\n\n\n\n**<mark>PROTECTION</mark>**\n\n**33,739**\n\n\n**72,077**\n\n\n**<mark>FOOD SECURITY</mark>**\n\n**2,332,690**\n\n\n**5,145**\n\n\n**<mark>EDUCATION</mark>**\n\n**1,202,905**\n\n\n\ngirls and boys who are receiving specialized child\nprotection services\n\npersons receiving Sexual and Gender-Based Violence\n(SGBV) response services\n\n\nindividuals who receive food assistance (cash, voucher\nor in-kind)\n\nindividuals receiving food & agricultural livelihoods\nsupport\n\n\nchildren (5-17 years, girls and boys) enrolled in formal\ngeneral education\n\n\n\n**1,345** <sup>classrooms constructed, established or rehabilitated</sup>\n\n\n\n\n\n\n\n\n\n\n\n**<mark>BASIC NEEDS</mark>**\n\n**2,217,095**\n\n**38,628**\n\n\n\nindividuals receiving core relief items in-kind\n\n\nindividuals receiving unconditional, sector-specific or\nemergency cash assistance\n\n\n\n\n\n**<mark>WASH</mark>**\n\n\n\n**364,769**\n\n\n\npeople with access to adequate quantity of safe water\nthrough temporary provision\n\n\n\n**168,956** people with access to appropriate sanitation facilities\nand services\n\n\n\n**<mark>LIVELIHOODS & SOCIAL\n\nMention: \"inter-agency 3RP dashboards\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Publient les réalisations sectorielles régionales du plan de résilience.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:000991:30:0", "occurrence_id": "467d35a7e6d62448590c19d0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **29 /** Synthesis of key findings from Inter-Agency Humanitarian Evaluations (IAHEs) of the international responses to crises in\nthe Philippines (Typhoon Haiyan), South Sudan and the Central African Republic\n\n\npolicy documents. Since planning energies seemed to be directed more towards resource\nmobilization than operational planning, the logic of the HPC (that plans should be based on\nevidence etc.) was not followed through. Similarly, the guidance on locating decision-making in\nthe field, monitoring the impact of humanitarian action and adjusting response programmes was\nnot followed in any systematic way.\n\n\n2.5.3 \u0007Leadership by the Humanitarian Coordinator, Humanitarian Country Team and\nbeyond\n\n\n101. L3 status confers empowered leadership on the designated HC in order to enable timely\ndecisions on setting overall priorities, allocating resources, monitoring performance and dealing\nwith under-performance. <sup>10</sup> [^10: \u0007Inter-Agency Standing Committee, ‘Humanitarian System-Wide Emergency Activation: definition and procedures’, reference\ndocument PR/1204/4078/7 (IASC, 13 April 2013), section 1, paragraph 3.] In CAR, the IAHE found that the quick deployment of a Senior HC\n(SHC) from the L3 HC Roster greatly enhanced the coordination of a response that had up to\nthat point “lumbered in reactive mode”. Having an empowered leader clearly helped with making\ndecisions instead of relying on HCT consensus, but the SHC’s leadership was nonetheless\nundermined by a “poorly functioning HCT” and weak inter-cluster coordination and information\nmanagement. The IAHE concludes that in a complex emergency and protracted crisis like that\nin CAR, “it would be more appropriate to have a SHC for a sustained period instead of deployed\nquickly for a short time”.\n\n\nBox 3: Extract from the Inter-Agency Standing Committee concept paper on empowered\nleadership (revised March 2014)\n\n\nWhile the existing tools an HC has may be adequate for many situations, especially if used\nby a skilled HC working together with an effective HCT and strong national partners, for L3\nemergencies,\n\nMention: \"Inter-Agency Humanitarian Evaluations\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Their synthesis brings together key findings from crisis responses.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001021:11:1", "occurrence_id": "fe0ad9b95dad6f72a0630a9e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: نیازمند کمک می باشند ، )تن 114,000 عودت کننده گان( فیصد این تخمین مینماید که20 .ریزی نموده استIOM\n\n\"مشخص شده است .2021-2019های سال پاسخ انساندوستانه افغانستان برای\"\n\n### **شتریمنابع ب**\n\n\n**افغانستان در ادارهIOM**\n\n\nIOM Afghanistan Website .1\n\n[Weekly Situation Update on Return of Undocumented Afghans from Iran and Pakistan](http://afghanistan.iom.int/pakistan-returns) .a\n\n[IOM Humanitarian Compendium](https://humanitariancompendium.iom.int/) .2\n\n[IOM Afghanistan Displacement Tracking Matrix Website](http://www.globaldtm.info/afghanistan/) .3\n\n[IOM Afghanistan Facebook](https://www.facebook.com/iomafghanistan/) .4\n\n[IOM Afghanistan Twitter](https://twitter.com/IOMAfghanistan) .5\n\n[IOM Afghanistan Instagram](https://www.instagram.com/iomafghanistan/) .6\n\n\n**در افغانستان** **ادارهUNHCR**\n\n\n[UNHCR Afghanistan data portal](https://data2.unhcr.org/en/country/afg) .1\n\n[Humanitarian Data Exchange: UNHCR Afghanistan data](http://data.humdata.org/organization/unhcr-afghanistan?sort=metadata_modified+desc) . 2\n\n[UNHCR Global Focus – Afghanistan](http:/\n\nMention: \"IOM Afghanistan Displacement Tracking Matrix Website\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Le site national publie les données de suivi des déplacements.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001021:11:1", "occurrence_id": "4619ae87b8b0e7bdb58c8b40", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: نیازمند کمک می باشند ، )تن 114,000 عودت کننده گان( فیصد این تخمین مینماید که20 .ریزی نموده استIOM\n\n\"مشخص شده است .2021-2019های سال پاسخ انساندوستانه افغانستان برای\"\n\n### **شتریمنابع ب**\n\n\n**افغانستان در ادارهIOM**\n\n\nIOM Afghanistan Website .1\n\n[Weekly Situation Update on Return of Undocumented Afghans from Iran and Pakistan](http://afghanistan.iom.int/pakistan-returns) .a\n\n[IOM Humanitarian Compendium](https://humanitariancompendium.iom.int/) .2\n\n[IOM Afghanistan Displacement Tracking Matrix Website](http://www.globaldtm.info/afghanistan/) .3\n\n[IOM Afghanistan Facebook](https://www.facebook.com/iomafghanistan/) .4\n\n[IOM Afghanistan Twitter](https://twitter.com/IOMAfghanistan) .5\n\n[IOM Afghanistan Instagram](https://www.instagram.com/iomafghanistan/) .6\n\n\n**در افغانستان** **ادارهUNHCR**\n\n\n[UNHCR Afghanistan data portal](https://data2.unhcr.org/en/country/afg) .1\n\n[Humanitarian Data Exchange: UNHCR Afghanistan data](http://data.humdata.org/organization/unhcr-afghanistan?sort=metadata_modified+desc) . 2\n\n[UNHCR Global Focus – Afghanistan](http:/\n\nMention: \"UNHCR Afghanistan data portal\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Le portail national rassemble les données de référence.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001022:3:0", "occurrence_id": "f19192fa805f2b47afbf76f5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 3\n\n\n\nINTRODUCTION\n\n\nThis Annual Report on the Centrality of Protection in Humanitarian Action 2022 intends to provide an analysis of\nhow the Centrality of Protection (CoP) has been implemented across crisis contexts where the humanitarian cluster\nsystem is activated. The report, prepared by UNHCR and the Global Protection Cluster (GPC) for the Inter-Agency\nStanding Committee (IASC), examines measures taken by the Humanitarian Coordinator (HC) / Humanitarian\nCountry Team (HCT) to address critical protection risks, with the support from the Protection Clusters/Sectors,\nOCHA and humanitarian partners.\n\n\nThe findings of the report are based on information from the Global Protection Cluster, Humanitarian Programme\nCycle (HPC) Quality Review conducted by OCHA, as well as consultations with ProCap advisors, Protection Cluster\nCoordination teams (including AoRs), OCHA focal points, and cluster coordinators from other sectors in a set of\nselected operations (e.g. Afghanistan, Ukraine, Colombia, Venezuela, and Nigeria). Data collection also involved\ntwo surveys undertaken by the GPC and its field Protection Clusters – including the annual Cluster Coordination\nPerformance Monitoring (CCPM) and a dedicated perception survey on the Centrality of Protection¹ which\ncomplemented the HPC Quality Review indicators on how Centrality of Protection is integrated and addressed in\nthe Humanitarian Needs Overviews (HNOs) and Humanitarian Response Plans (HRPs). The report also benefited\nfrom a dedicated session on Centrality of Protection held during the Global Protection Conference in May 2023.\n\n\nThe report considers some of the challenges and gaps that operations face with the implementation of CoP, as\nrevealed through consultations and surveys. Case studies from selected operations are used to illustrate both best\npractices and challenges that operations are encountering. The analysis is structured around five key elements\nwhere humanitarian actors, under the leadership of HC/HCTs, are contributing to achieving collective protection\noutcomes: (i) Protection Risk Analysis, (ii) HCT Protection Strategies\n\nMention: \"Cluster Coordination\nPerformance Monitoring\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its annual exercise assesses cluster coordination.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001022:3:0", "occurrence_id": "8e7989d871d5d20783e29724", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 3\n\n\n\nINTRODUCTION\n\n\nThis Annual Report on the Centrality of Protection in Humanitarian Action 2022 intends to provide an analysis of\nhow the Centrality of Protection (CoP) has been implemented across crisis contexts where the humanitarian cluster\nsystem is activated. The report, prepared by UNHCR and the Global Protection Cluster (GPC) for the Inter-Agency\nStanding Committee (IASC), examines measures taken by the Humanitarian Coordinator (HC) / Humanitarian\nCountry Team (HCT) to address critical protection risks, with the support from the Protection Clusters/Sectors,\nOCHA and humanitarian partners.\n\n\nThe findings of the report are based on information from the Global Protection Cluster, Humanitarian Programme\nCycle (HPC) Quality Review conducted by OCHA, as well as consultations with ProCap advisors, Protection Cluster\nCoordination teams (including AoRs), OCHA focal points, and cluster coordinators from other sectors in a set of\nselected operations (e.g. Afghanistan, Ukraine, Colombia, Venezuela, and Nigeria). Data collection also involved\ntwo surveys undertaken by the GPC and its field Protection Clusters – including the annual Cluster Coordination\nPerformance Monitoring (CCPM) and a dedicated perception survey on the Centrality of Protection¹ which\ncomplemented the HPC Quality Review indicators on how Centrality of Protection is integrated and addressed in\nthe Humanitarian Needs Overviews (HNOs) and Humanitarian Response Plans (HRPs). The report also benefited\nfrom a dedicated session on Centrality of Protection held during the Global Protection Conference in May 2023.\n\n\nThe report considers some of the challenges and gaps that operations face with the implementation of CoP, as\nrevealed through consultations and surveys. Case studies from selected operations are used to illustrate both best\npractices and challenges that operations are encountering. The analysis is structured around five key elements\nwhere humanitarian actors, under the leadership of HC/HCTs, are contributing to achieving collective protection\noutcomes: (i) Protection Risk Analysis, (ii) HCT Protection Strategies\n\nMention: \"perception survey on the Centrality of Protection¹\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Completes the quality indicators on protection centrality.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001030:12:0", "occurrence_id": "cde742fb763f81d541dd8641", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **2017 ANNUAL REPORT** Regional Refugee & Resilience Plan 2017-2018\n\n\nenhancing the transparency and competitiveness of pricing, reducing barriers to accessing the best deals and making the most\nof every dollar injected into the Lebanese economy.\n\n\nFood for Training and Food for Asset activities targeting and benefiting both Syrian refugees and host communities fostered social\ncohesion and led to a significant improvement in terms of acceptance, integration and cohabitation as well as social solidarity.\nTo monitor the social cohesion effects of projects in Lebanon and Egypt, a Social Cohesion Score (SCS) was developed through\nthe responses of 477 Syrian, Egyptian and Lebanese individuals engaged in livelihood initiatives. Through answering questions\non inter-community relations, competition over resources and trustworthiness in institutions, the index provided a quantifiable\nassessment of the social cohesion effects of programmes. In 2017, SCS was piloted before the start of livelihood activities and\nfollow up measurements will continue in 2018 to identify areas of underlying tension to be addressed.\n\n\n\n\n\n2 JFO Vulnerability assessment data, December 2017\n3 Including PRS and Syrian refugees at King Abdullah Park who were supported through cash assistance.\n\n\n\n<u>13</u>\n\nMention: \"JFO Vulnerability assessment data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its data identify underlying tensions to address.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001031:15:1", "occurrence_id": "b95c41b7041ba40c5adb3d83", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  change. The dramatic\n\n\ndecrease of water availability in some regions and recurrent flooding in others will impact\n\n\nupon economic opportunities and conditions of life will deteriorate in affected areas. Such\n\n\ndeterioration may not necessarily cause displacement, but it may prompt people to consider\n\n\nmigration as a way to adapt to the changing environment, and explain why people move to\n\n\nregions with better living conditions and income opportunities. However, if areas become\n\n\nuninhabitable over time because of further deterioration, finally leading to complete\n\n\ndesertification, permanent flooding of coastal zones or similar situations, population\n\n\nmovements will amount to forced displacement and become permanent. Many factors will\n\n\ncontribute to population movements in such situations including political, economic and\n\n\nsocial elements that may push people to move to other locations inside their own country or\n\n\nabroad. Thus, for example, droughts in the Sahel region forced large numbers of people to\n\n\nleave their habitual places of residence, but in Mali the drought contributed to a reduction of\n\n\ncross-border movements because families no longer had sufficient resources to pay for travel\n\n\n38 See NRC-IDMC global estimates for 2008, 2009, 2010, above note 35. Piguet, above note 31, reports that\nbetween 2000 and 2005, 106 million people, on average, were affected by flooding.\n\n39 Piguet, above note 31, 5-6.\n\n40 Piguet, above note 31, 6, referring to C. E. Haque, _Hazards in a fickle environment: Bangladesh_ (Kluwer Academic\nPublisher, Dordrecht 1997).\n\n\n14\n\nMention: \"NRC-IDMC global estimates\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Its global estimates cover the stated years.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001047:2:2", "occurrence_id": "0eef14929b8933dd2478d401", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  cuatro.**\n**8** **Fuentes: estimaciones de las estadísticas demográficas sobre las personas desplazadas internas (PDI) (Observatorio de Desplazamiento Interno [IDMC]);**\n**personas palestinas refugiadas bajo el mandato de la UNRWA (UNRWA); las personas refugiadas, personas en situación similar a la de los refugiados,**\n**solicitantes de asilo y otras personas que necesitan protección internacional se basan en los datos disponibles (ACNUR) y en las estimaciones sobre la**\n**población mundial (Departamento de Asuntos Económicos y Sociales de las Naciones Unidas).**\n\n\nACNUR > **TENDENCIAS GLOBALES 2022** 3\n\nMention: \"estadísticas demográficas sobre las personas desplazadas internas\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Sus estimaciones sustentan el cálculo de la población desplazada.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001047:2:2", "occurrence_id": "fd13da8d822c2978d967532e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  cuatro.**\n**8** **Fuentes: estimaciones de las estadísticas demográficas sobre las personas desplazadas internas (PDI) (Observatorio de Desplazamiento Interno [IDMC]);**\n**personas palestinas refugiadas bajo el mandato de la UNRWA (UNRWA); las personas refugiadas, personas en situación similar a la de los refugiados,**\n**solicitantes de asilo y otras personas que necesitan protección internacional se basan en los datos disponibles (ACNUR) y en las estimaciones sobre la**\n**población mundial (Departamento de Asuntos Económicos y Sociales de las Naciones Unidas).**\n\n\nACNUR > **TENDENCIAS GLOBALES 2022** 3\n\nMention: \"ACNUR\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Sus datos disponibles sustentan las cifras de necesidades de protección.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001050:9:0", "occurrence_id": "36bf8560bde4432eb66ffb83", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Análisis de protección - Chocó, Diciembre 2021\n\n\n**RIESGO 3: Homicidios y amenazas**\n\n\nDesde la firma del acuerdo de paz, 40 líderes o defensores\nde derechos humanos han sido asesinados, incluyendo 6\ndurante 2021. Se ha evidenciado un aumento progresivo en\nel número de homicidios al que están vinculados jóvenes.\nSegún los datos del Instituto Nacional de Medicina Legal,\ndurante 2021 se han presentado **188 homicidios** en el\ndepartamento del Chocó, de los cuales el 68% (128 casos)\ntenían como factor de vulnerabilidad pertenecer a grupos\nétnicos, el 97% eran hombres y el 47% jóvenes entre 18 y 28\naños; 2 eran de nacionalidad venezolana En el 55% el\npresunto agresor fue la delincuencia común, los grupos\nalzados al margen de la ley, la delincuencia organizada y/o\nmiembros de las fuerzas armadas y de policía y en el 83% de\nlos casos, la circunstancia del hecho estuvo relacionada con\najuste de cuentas y/o con violencia sociopolítica <sup>8</sup> [^8: Este porcentaje está calculado sobre los casos con información disponible, 30% de los\nhomicidios.] .\n\n\nLas amenazas realizadas a líderes sociales y defensores de\nderechos humanos, funcionarios públicos y profesores\ncuando se movilizan para trabajar con los niños, y en su\nintento para educarlos y alejarlos de acciones delictivas,\ntambién hacen parte de las dinámicas que se\n\nMention: \"datos del Instituto Nacional de Medicina Legal\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Registran los homicidios vinculados a jóvenes en el departamento.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001055:26:1", "occurrence_id": "5f422038a6debc2d65dd1908", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ، التي تبلغ من العمر\nالجئ سوري. تبقى معظم األيام داخل 700 لبنان يقيم به أكثر من\n\n\n\nJuly 2013 survey of 1,800 Jordanian nationals, ‘Current Issues in Jordan,’ by the Centre for.Strategic Studies at the University of Jordan. The survey was led by Dr. Walid AlkhatibUNHCR / E. Dorfman .الغرفة الوحيدة التي تتقاسمها عائلتها 22\n\n\n\nJuly 2013 survey of 1,800 Jordanian nationals, ‘Current Issues in Jordan,’ by the Centre for.Strategic Studies at the University of Jordan. The survey was led by Dr. Walid AlkhatibUNHCR / E. Dorfman .الغرفة الوحيدة التي تتقاسمها عائلتها 22\n\nMention: \"Current Issues in Jordan\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Enquête auprès de 1 800 nationaux sur les enjeux du pays.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001061:2:2", "occurrence_id": "963bd36d9a0a35de109bfa14", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  refugee self-reliance in food security and\nnutrition. This Joint Strategy represents the fulfilment\nof this recommendation.\n\n\n**1.** UNHCR 2015, UNHCR Global Trends 2014.\n**2.** ‘Durable solutions’ refer to the point when refugees no longer need\ninternational protection, generally after voluntary repatriation, local integration, or resettlement to a third country. Complementary pathways to\nprotection and solutions are being expanded, always with consideration\ngiven to durability.\n**3.** The Solutions Alliance aims to improve the lives of displaced persons and\nthe communities that host them by responding more collaboratively to\ndisplacement and contributing to durable solutions. More information is\navailable at www.solutionsalliance.org\n\n\n\n**3**\n\nMention: \"UNHCR Global Trends 2014\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its global trends underpin the strategy described.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001063:51:8", "occurrence_id": "a92600d9b1b0f15a4c8322a4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \n[obtaining-asylum-united-states. Individuals apply for asylum defensively](http://www.uscis.gov/humanitarian/refugees-asylum/asylum/obtaining-asylum-united-states)\nwhen he or she requests asylum as a defense against removal from the U.S.\nIndividuals are generally placed into defensive asylum procedures in one of\ntwo ways: (1) “they are referred to an Immigration Judge by USCIS after they\nhave been determined to be ineligible for asylum at the end of the affirmative\nasylum process,” or (2) they are placed in removal proceedings, as discussed\nabove. Id. Immigration judges hear defensive asylum cases and decide\nwhether the individual is eligible for asylum.\n\n**16** _See, e.g.,_ Decree No. 97-96, 24 October 1996, Law to Prevent, Punish,\nand Eradicate Intrafamily Violence (Guatemala); Decree No. 132-97, 29\nSeptember 1997, the Law on Domestic Violence (Honduras); Decree No.\n902, 5 December 1996, the Law on Intrafamily Violence (El Salvador).\n\n**17** All names of women interviewed for this report have been changed to\npreserve their confidentiality.\n\n**18** UNHCR, Population Statistics Database, “Asylum-Seekers (Monthly\nData)”, _available at_ [http://popstats.unhcr.org/en/asylum_seekers](http://popstats.unhcr.org/en/asylum_seekers)\n\n**19** The word “transgender” is “an an umbrella term for people\nwhose gender identity and/or expression is different from cultural\nexpectations based on the sex they were assigned at birth. Being\ntransgender does not imply any specific sexual orientation. Therefore,\ntransgender people\n\nMention: \"Population Statistics Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its monthly series cover asylum seekers.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001064:24:0", "occurrence_id": "d3a020f1d6c75790b23b0aef", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: and greater appreciation of UNHCR’s work within Nairobi’s refugee community (a\nloose concept used in this context to encompass areas beyond the city limits and to\ninclude registered and non-registered refugees, people with pending and rejected\nasylum applications, as well as irregular migrants from refugee-producing\ncountries).\n\n\n**National staff and outreach workers**\n\n\n87.   UNHCR has appointed a number of national officers to its evolving urban\nrefugee programme, university graduates whose local origins, knowledge of Nairobi\nand interpersonal skills provide the organization with an invaluable means of\ncommunication with refugees and other stakeholders. They deserve significant credit\nfor the increased visibility and credibility that the organization has been able to\nestablish in the city.\n\n\n88.   At the same time, UNHCR has worked with an implementing partner, GTZ, to\nestablish a group of 15 community outreach workers, men and women who are\ndrawn from the different national and ethnic groups to be found amongst Nairobi’s\nrefugee population and who are responsible for outreach activities in different parts\nof the city. This initiative was an outcome of the 2006 Participatory Assessment,\nwhen refugees complained that they had to travel to the Branch Office in order to\naccess the information they needed.\n\n\n89.   The outreach workers normally have a degree or diploma, are recruited on a\ncompetitive basis and in accordance with AGDM principles. They are paid a modest\nincentive (around 8,000 shillings a month) which does not require them to have a\nwork permit. GTZ convenes monthly meetings to train the outreach workers on a\nvariety of issues, including how to respond to refugee arrests, how to refer survivors\nof SGBV to appropriate services, and how to identify vulnerable individuals who\nneed special support. They are well known to local police stations and play an\nimportant role in negotiating the release of refugees from detention.\n\n\n90.   In the course of this review, it became clear that the community outreach\nworkers provide an important link between UNHCR, partner agencies, civil society\nand local officials.\n\nMention: \"2006 Participatory Assessment\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"An initiative follows from its findings on displacement to the office.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001066:80:1", "occurrence_id": "b831ee6e840d381b17a7e0ef", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 10 de 11 de 2019). Requisitos para residencia de inversionista.\n\n<u>Obtenido de Dirección General de Migración: www.migracion.gob.do/Menu/SubList/44</u>\n\n\nX **X** Oficina Nacional de Estadística- ONE (2017) Segunda Encuesta Nacional De Inmigrantes –ENI-2017–\n\n<u>República Dominicana</u>\n\n\nX **X** Oficina Nacional de Estadística- ONE (2019) Panorama Estadístico Migración Laobral hacia República\n\n<u>Dominicana: evolución ocupacional del origen al destino</u>\n\n\nX **X** Organización Internacional del Trabajo (2019). Formalización de trabajadores independientes en\n\n<u>Paraguay.</u>\n\n\nX **X** Organización Internacional del Trabajo (2018) Cabrera Donna. Estudio sobre un marco comparativo\n\nde experiencias y casos de buenas prácticas regionales y extra regionales sobre el vínculo entre\n<u>necesidades del mercado laboral y política migratoria.</u>\n\n\nX **X** Organización Internacional del Trabajo. (2011). Desarrollo de Cadenas de Valor para trabajo decente.\n\n<u>Ginebra, Suiza: Organización Internacional del Trabajo. Recuperado el 2019</u>\n\nMention: \"Segunda Encuesta Nacional De Inmigrantes\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Son deuxième tour couvre la population immigrée du pays.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001071:4:0", "occurrence_id": "4fe593d307a7ef8d8e40081b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Egypt\nSudan\nEritrea\nEthiopia\nSomalia\nKenya\nUganda\nRwanda\nBurundi\nMauritius\nComoros\nUnited\nRep. of\nTanzania\nSeychelles\nMaldives\nSri Lanka\nIndia\nPakistan\nTajikistan\nKyrgyzstan\nKazakhstan\nRussian Federation\nUzbekistan\nTurkmenistan\nKuwait\nIraq\nTurkey\nRussian Federation\nSyrian\nArab Rep.\nJordan\nSaudi Arabia\nBahrain\nQatar\nUnited \nArab \nEmirates\nOman\nAfghanistan\nIslamic Rep. \nof Iran\nBangladesh\nMyanmar\nBhutan\nChina\nNepal\nMongolia\nJapan\nThailand\nViet Nam\nLao People’s Dem. Rep.\nDem. People’s Rep. of Korea\nRep. of Korea\nPhilippines\nPalau\nPapua New Guinea\nTimor-Leste\nAustralia\nNew Zealand\nSolomon Islands\nCambodia\nBrunei Darussalam\nMalaysia\nSingapore\nIndonesia\nMalawi\nMozambique\nZimbabwe\nBotswana\nAfrica\nLesotho\nSwaziland\nMadagascar\nZambia\nDjibouti\nYemen\nSpain\nFrance\nPortugal\nIreland\nUnited Kingdom of \nGreat Britain\nand Northern Ireland\nBelgium\nNetherlands\nDenmark\nGermany\nNorway\nSweden\nLatvia\nCzech Rep.\nFinland\nLithuania\nAustria\nSwitzerland\nLiechtenstein\nItaly\nEstonia\nPoland\nSlovakia\nSloveniaCroatia\nAlbania\nBosnia and \nHerzegovina\nThe former \nYugoslav Rep. of \nMacedonia\nSerbia 3\nMalta\nMontenegro\nGreece\nBulgaria\nRomania\nHungary\nRep.\nof Moldova\nUkraine\nBelarus\nLuxembourg\nAndorra\nHoly See\nSan Marino\nMonaco\nAfrican \nublic\nocratic \nthe Congo\nSouth \nSudan\n5\nUNHCR Mid-Year Trends 2013\n\nMention: \"UNHCR Mid-Year Trends 2013\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Sa publication de milieu d'année accompagne la cartographie des populations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001072:2:0", "occurrence_id": "3d256ffd2a510092eaa0b370", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: # **Shelter Thematic Report**\n\n### **Objective**\n\nThe report aims to analyze the secondary impacts of shelter issues in Sudan and discuss possible solutions\nleveraging existing capacities. Given the large-scale displacement and pre-existing poor shelter conditions, it is\ncritical to understand how these conditions impact populations amidst evolving conflict dynamics and how best\nto address them.\n\n### **Contextual Background**\n\nIn Sudan, humanitarian needs were at record levels even before the outbreak of the conflict in April 2023. The\ncountry has long faced challenges with internal displacement, stemming from various conflicts and\nenvironmental factors. Internally Displaced Persons (IDPs) have consistently faced poor shelter conditions,\nmarked by overcrowding and inferior housing quality, along with widespread issues like limited privacy,\ninadequate lighting, and constrained access to vital household items. Humanitarian organizations' efforts to\nimprove these conditions have been significantly hampered by funding constraints, with the shelter cluster only\n[43% funded on the revised 2023 HRP (FTS](https://fts.unocha.org/plans/1123/summary) 24/01/2023). Findings from the 2023 HNO identified 2.9 million people\nin need of shelter and NFI assistance prior to the conflict, with three in four Sudanese living in damaged shelters\nand many reporting severe damages that compromised the structural safety of their houses or total structural\ncollapse. Even families with more durable housing options struggled with poor living conditions. Inflation and\nsharp increases in the prices of goods have also reduced families’ purchasing power, limiting their ability to invest\nin shelter repair and maintenance. The most commonly reported shelter issues included leaks, cracked roofs, and\npartial roof collapses. Furthermore, findings from the 2022 MSNA showed that only a quarter of assessed\nhouseholds in the country had the security of tenure which protects tenants from forced eviction. For those living\nin camps and informal settlements, 73% of households reported having only collective\n\nMention: \"2023 HRP\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its revised amounts and financing are tracked by the monitoring service.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001072:2:0", "occurrence_id": "46d6a779ad1bd3913e8078e5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: # **Shelter Thematic Report**\n\n### **Objective**\n\nThe report aims to analyze the secondary impacts of shelter issues in Sudan and discuss possible solutions\nleveraging existing capacities. Given the large-scale displacement and pre-existing poor shelter conditions, it is\ncritical to understand how these conditions impact populations amidst evolving conflict dynamics and how best\nto address them.\n\n### **Contextual Background**\n\nIn Sudan, humanitarian needs were at record levels even before the outbreak of the conflict in April 2023. The\ncountry has long faced challenges with internal displacement, stemming from various conflicts and\nenvironmental factors. Internally Displaced Persons (IDPs) have consistently faced poor shelter conditions,\nmarked by overcrowding and inferior housing quality, along with widespread issues like limited privacy,\ninadequate lighting, and constrained access to vital household items. Humanitarian organizations' efforts to\nimprove these conditions have been significantly hampered by funding constraints, with the shelter cluster only\n[43% funded on the revised 2023 HRP (FTS](https://fts.unocha.org/plans/1123/summary) 24/01/2023). Findings from the 2023 HNO identified 2.9 million people\nin need of shelter and NFI assistance prior to the conflict, with three in four Sudanese living in damaged shelters\nand many reporting severe damages that compromised the structural safety of their houses or total structural\ncollapse. Even families with more durable housing options struggled with poor living conditions. Inflation and\nsharp increases in the prices of goods have also reduced families’ purchasing power, limiting their ability to invest\nin shelter repair and maintenance. The most commonly reported shelter issues included leaks, cracked roofs, and\npartial roof collapses. Furthermore, findings from the 2022 MSNA showed that only a quarter of assessed\nhouseholds in the country had the security of tenure which protects tenants from forced eviction. For those living\nin camps and informal settlements, 73% of households reported having only collective\n\nMention: \"2023 HNO\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Its findings identify shelter needs before the conflict.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001108:8:0", "occurrence_id": "f985af0703df4ce295a60587", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Data Source: IOM- Displacement Tracking matrix\n\nMention: \"IOM- Displacement Tracking matrix\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"La source déclarée des données de la page.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001138:31:0", "occurrence_id": "191806ee665d956b993ad002", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 150 000 000 Capital\n\n\nInternational boundary\n\n\n### **South-West Asia: Afghan Refugees,** **Returnees and National Population** January 2012\n\n\n\nand the designations used on this\nmap do not imply official endorsement\n\n\n\nAdministrative boundary\n\n\nLine of control as promulgated\nin the 1972 SIMLA agreement\n\n\nAfghan Refugee Population\n\n\n\nSources:\nUNHCR, Global Insight digital mapping\n\n\n\n75 000 000\n\n\n15 000 000\n\n\n\n© 1998 Europa Technologies Ltd.\n\n\n\nUNJLC - UNSDI-T\n\n\n\nUrban\nRural and Others\n\n\n\nThe boundaries and names shown\n\n\n\n25,000\nindividuals\n\n\n\nor acceptance by the United Nations.\n\n\n\nNational population \"World Population Prospects: The 2010 Revision\", New York, 2011Source: UNHCR, National population: United Nations, Population Division, Source: Solutions Strategy for Afghan Refugees, to Support Voluntary Repatriation, Sustainable Reintegration and Assistance to Host Countrie\n\n\n\n66\n\n\n\n67\n\nMention: \"Global Insight digital mapping\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Ses relevés numériques cartographient la population réfugiée.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001138:31:0", "occurrence_id": "bf692d16574533e8ee9ece02", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 150 000 000 Capital\n\n\nInternational boundary\n\n\n### **South-West Asia: Afghan Refugees,** **Returnees and National Population** January 2012\n\n\n\nand the designations used on this\nmap do not imply official endorsement\n\n\n\nAdministrative boundary\n\n\nLine of control as promulgated\nin the 1972 SIMLA agreement\n\n\nAfghan Refugee Population\n\n\n\nSources:\nUNHCR, Global Insight digital mapping\n\n\n\n75 000 000\n\n\n15 000 000\n\n\n\n© 1998 Europa Technologies Ltd.\n\n\n\nUNJLC - UNSDI-T\n\n\n\nUrban\nRural and Others\n\n\n\nThe boundaries and names shown\n\n\n\n25,000\nindividuals\n\n\n\nor acceptance by the United Nations.\n\n\n\nNational population \"World Population Prospects: The 2010 Revision\", New York, 2011Source: UNHCR, National population: United Nations, Population Division, Source: Solutions Strategy for Afghan Refugees, to Support Voluntary Repatriation, Sustainable Reintegration and Assistance to Host Countrie\n\n\n\n66\n\n\n\n67\n\nMention: \"World Population Prospects\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Ses perspectives donnent les populations nationales de référence.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001174:73:0", "occurrence_id": "d1eac165c8b903a5b8fced89", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Jordan Home Visits Report 2014 - Living in the shadows**\n\n\n###### _“The economy of Syrian refugee households is being pushed_\n\n\n###### _to the brink of disaster. In a context of dwindling support_\n\n\n###### _and a lack of alternatives, more funding will be required_\n\n\n###### _in order to ensure refugee protection and assistance.”_\n\n\n\n\n- Andrew Harper, UNHCR Representative, Jordan\n\n\n\n74\n\nMention: \"Jordan Home Visits Report 2014\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Ses visites domiciliaires décrivent l'économie des ménages réfugiés.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001183:25:0", "occurrence_id": "fb9d46479e109dd7f15d8a6c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **خطة الستجابة لالجئين وتعزيز القدرة على الصمود في مصر**\n\n26 2025 ديسمبر-يناير\n\n\n\nكمــا تبــذل جهــود كبيــرة لتعزيــز المحليــات، ال ـــيما مــن\nخالل تحديــد ودعــم المنظمــات التــي يقودهــا الالجئــون.\nوـــيـري تكثيــف هــذه الـهــود لتمكيــن مـتمعــات\nالالجئيــن، وتعزيــز قــدرات المنظمــات المحليــة، ودعــم\nالســلطات الوطنيــة والمحليــة، لتعزيــز نهــج أكثــر شــموال <sup>ي</sup>\n\nوتشــاركية.\n\n\n\nوقركاء **ـ** ن الش **ـ** ات بي **ـ** ادل البيان **ـ** ات وتب **ـ** ع المعلوم **ـ** م جم **ـ** د ت **ـ**\n“، ActivityInfo” مــن خالل أداة إلدارة المعلومــات تســمى\nة كل\n\nMention: \"ActivityInfo\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Un outil de gestion des informations sectorielles.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001192:9:6", "occurrence_id": "e07285a3ea81cd03e6e2b6b1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: //reliefweb.int/report/syrian-arab-republic/unprecedented-threat-explosive-ordnance-across-syria-ma-aor-situation-update-no1-december-2024-january-2025)</u>\n<u>[Across Syria, January 2025](https://reliefweb.int/report/syrian-arab-republic/unprecedented-threat-explosive-ordnance-across-syria-ma-aor-situation-update-no1-december-2024-january-2025)</u>\n\n- <u>[Syria governorates IDPs and IDP returnee's overview 26 February 2025](https://data.unhcr.org/en/documents/details/114787)</u>\n\n- <u>[Syria governorates of return overview- 06 March 2025](https://data.unhcr.org/en/documents/details/114994)</u>\n\n- <u>[UNICEF Syria Humanitarian Situation Report No. 3, 29 December 2024](https://reliefweb.int/report/syrian-arab-republic/unicef-syria-humanitarian-situation-report-no-3-29-december-2024)</u>\n\n- <u>[IOM, Responding to Syria at a Time of Need; Statement by IOM Chief, 20 December 2024](https://www.iom.int/news/responding-syria-time-need-statement-iom-chief)</u>\n\n- <u>[Older Persons Protection Barriers Report, North-West Syria (July 2024) [EN/AR] - Syrian Arab Republic |](https\n\nMention: \"Syria governorates of return overview\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Ses aperçus par gouvernorat suivent les retours à ces dates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001201:1:119", "occurrence_id": "74699d38aa69d5221a119df3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <br>!<br>~~!~~<br>~~!~~<br>~~!~~<br>!<br>!<br>!<br>!<br>~~!~~<br>~~!~~<br>~~!~~<br>~~!!~~<br>!<br>~~!~~<br>!<br>!<br>!<br>!<br>!<br>!<br>!<br>!<br>!<br>!<br>!!<br>!!<br>!<br>!<br>!<br>**!**<br>!<br>!<br>!<br>!!<br>!<br>! !<br>!<br>!<br>!<br>!<br>!<br>!<br>!<br>!<br>!!<br>!<br>!<br>!<br>**!**<br>!<br>!<br>!<br>!<br>!<br>!<br>!<br>**Aakar**<br>Mhamm|||||||||||||||||||||||||||\n|needs. This finding is consistent~~ with data from VASyR~~<br>percentage of Syrian refugee ho~~useholds (49%) that did~~<br>Interventions with municipalities through the Mappin<br>municipalities.<br>Coordination among different a~~ctor\n\nMention: \"data from VASyR\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Ses résultats sur les ménages correspondent aux constats rapportés.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001216:2:0", "occurrence_id": "406a2b21e247177131bfe9c4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: REGIONAL SUMMARIES I **SOUTHERN AFRICA** REGIONAL SUMMARIES I **SOUTHERN AFRICA**\n\n### KEY RESULTS AND TRENDS IN 2020\n\n\n\nUNHCR’S **PROGRAMMATIC** RESPONSE\n\n\n\n\n\n**CHILD PROTECTION**\n\n\n\n**WASH**\n\n\n\n**1,651** unaccompanied or separated\nchildren had a best interests assessment\n\n\n**GENDER-BASED VIOLENCE**\n\n\n\n\n\n**TRENDS** IN RESPONSE\n\n**Cash assistance by sector in Africa | 2016-2020**\n\nBasic needs Seasonal grants Life-saving support Solutions\n\n\n2020\n\n\n2019\n\n\n\n\n\n\n\n\n\n**3,390** gender-based violence incidents\nreported for which survivors received\npsychosocial counselling.\n\n\n**CORE RELIEF ITEMS**\n\n**124,164** households provided with core\nrelief items.\n**68,020** women received sanitary\nmaterials.\n\n\n**HEALTH**\n\n**1.05** under-5 mortality rate (per 1,000\nunder-5s per month) in refugee camps.\n\n\n\n\n\n2018\n\n\n2017\n\n\n2016\n\n\n\nUSD millions\n***Including **$8.8 million** in the **Southern Africa** region **.**\n\n\n\n**Individual registration records | 2016-2020**\n\n\nIndividual registration records in PRIMES proGres v4\nindividuals (5 years and above) with biometric records in PRIMES\n\n\nIn millions\n\n\n\n**Resettlement departures | 2016-2020**\n\n\nIndividuals\n\n\n\n4,000\n\n3,500\n\n3,000\n\n\n2,500\n\n2,000\n\n1,500\n\n1,000\n\n\n\n\n\n2016 2017 2018 2019 2020\n\n\n\n2016 2017 2018 2019 2020\n\n\n\n\n\n\n\n**SHELTER**\n\n**76,547** people of concern received\nemergency shelter.\n**62%** of households lived in adequate\ndwellings.\n\n\n\n\n\n\n\n**19** litres of potable water available on\naverage per person per day in refugee\ncamps.\n**62%\n\nMention: \"PRIMES proGres v4\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Contient les enregistrements individuels et biométriques sur cette période.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001225:3:0", "occurrence_id": "0508c605004f82952ce457e6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **3RP** / Regional Refugee and Resilience Plan\n\n\n\n**Annual Report 2023**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n_[Data Source: UNHCR data portal,](https://data.unhcr.org/en/situations/syria)_\n_and Government of Türkiye_\n\n\n\n_1_ _The host community is defined by looking at the administrative areas where_\n\n_refugees are present. Host communities can be targeted not only individuals_\n_but also as communities and institutional (municipalities and government_\n_institutions) levels._\n\n\n**4**\n\n\n\n_2_\n\nMention: \"UNHCR data portal\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Combiné aux sources gouvernementales pour le rapport annuel.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001231:25:0", "occurrence_id": "7f4e5d673310c59d5f4ced02", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: DTM Mali – Septembre 2019\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n|Régions-Cercles-Communes|Ménages|Enfant ( < 18 ans )|Col4|Col5|Adulte (18 - 59 ans )|Col7|Col8|Personne âgé e ( > 59 ans )|Col10|Col11|Total<br>Femme|Total<br>Homme|Total<br>Individus|\n|---|---|---|---|---|---|---|---|---|---|---|---|---|---|\n|**Régions-Cercles-Communes**|**Ménages**|**F **|**M **|**Total Enfant**<br>**( < 18 ans )**|**F **|**M **|**Total Adulte**<br>**(18 - 59 ans )**|**F **|**M **|**Total Prs. âgée**<br>**( > 59 ans )**|**Total Prs. âgée**<br>**( > 59 ans )**|**Total Prs. âgée**<br>**( > 59 ans )**|**Total Prs. âgée**<br>**( > 59 ans )**|\n|SOCOURA|1273|2261|1851|4112|1152|1109|2261|160|107|267\n\nMention: \"DTM Mali – Septembre 2019\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Dénombre les ménages par âge dans les subdivisions administratives.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001254:1:3", "occurrence_id": "6c776d1b5c18fd6199cb5d8a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Central Administration of Statistics (CAS) year 2002 dataset,   Poverty data: CAS, UNDP and MoSA Living Conditions and Household Budget Survey 2004-5,\nSyrian Refugee Population - UNHCR as of 30/05/2015, Humanitarian Intervention Data - Activity Info, Palestinian Refugee Population- UNRWA\n\nMention: \"Humanitarian Intervention Data - Activity Info\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"L'une des sources assemblées pour le profil de la situation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001266:22:0", "occurrence_id": "40398545734a4ce29424bffb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: UNHCR Map of Tanzania:\n\n\nSource: UNHCR Global Appeal 2006-UNHCR in the United Republic of Tanzania. Online.\nhttp://www.unhcr.org\n\n\n21\n\nMention: \"UNHCR Map of Tanzania\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Sa carte provient de l'appel mondial de l'agence pour ce pays.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001290:9:0", "occurrence_id": "cd4074787c7717c40926648f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **RÉPUBLIQUE DEMOCRATIQUE DU CONGO** | Mars 2024\n\n\n<mark>2023. Ces hommes leur reprochaient le fait d’avoir fourni des renseignements aux forces de défense et de sécurité (FARDC)</mark>\n<mark>sur leur localisation. Il en est de même de la restriction faite par le même groupe contre les femmes déplacées internes venues</mark>\n<mark>de Shabunda et qui vivent à Ngando, Kisogo, Nyabalume (Territoire de Mwenga) depuis mai 2023. Il était reproché à ces</mark>\n<mark>femmes d’avoir fourni des renseignements à la justice congolaise sur des cas de viol, enlèvement et autres abus commis par</mark>\nle groupe armé Raiya Mutomboki Ndarumanga en territoire de Shabunda.\n\n\n<mark>La liberté de mouvement est aussi entravée par les différentes barrières illégales érigées sur des routes par des groupes armés</mark>\n<mark>ou par des Forces nationales de défense et de sécurité. Sur ces barrières s’enregistrent plusieurs cas d’extorsion des biens et</mark>\nde taxation illégale.\n\n\n\n<mark>Plusieurs cas d’enlèvements ont aussi été</mark>\n<mark>rapportés durant l’année 2023. Selon les</mark>\n<mark>données du SAR/HCR, 1,859 cas d’enlèvements</mark>\n<mark>ou disparitions forcées ont été répertoriés au</mark>\n\nMention: \"données du SAR/HCR\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Ses répertoriations décrivent les enlèvements et disparitions forcées.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001296:18:0", "occurrence_id": "67596954b7f01b8dee028789", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <mark>Conflict Scan| Juillet 2018 - Diffa</mark>\n\n#### **Evolution des conflits dans la région de Diffa**\n\n\nEn octobre 2017 Search a mené un Conflict Scan dans la région de Diffa à Kablewa, Garin Wanzam,\nNguigmi, et Kindjandi. <sup>15</sup> [^15: SFCG. _Niger conflict scan._ October 2017.] Bien que le Conflit Scan qui a été fait en 2018 aie ciblé des sites différentes de\nla région de à Diffa, une juxtaposition des deux Conflict Scans permet tout de même de montrer les\ntendances en générale dans les deux études pour offrir une image globale de l’évolution des conflits au\nniveau de la région. Les tensions et conflits identifiés dans le rapport de 2017 qui ont plus d’importance\ndans les sites ciblés ont été : les conflits liés aux ressources ; les conflits liés à la délinquance et au\nbanditisme ; les conflits entre éleveurs et agriculteurs ; et la violence basée sur le genre. Une comparaison\ndes résultats de Conflict Scan de 2017 et celui de 2018, montre que les mêmes conflits se sont relevé\ncomme étant plus importants dans les différents sites.\n\n\nDans l’étude de 2017, les conflits liés aux ressources ont été identifiés comme le plus importants. Ces\nconflits comprenaient la distribution de l’aide humanitaire et l’accès à l’eau. La mauvaise gestion des\ndistributions a été identifiée en tant que tension aux sites, mais les conflits autour des distributions de\nl’aide humanitaire a rarement été mention\n\nMention: \"Conflict Scan\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Ses relevés mensuels décrivent l'évolution des conflits dans la région.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001297:2:2", "occurrence_id": "26eb4e9bb0c2807991f95a9f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Country and Lending Groups.](https://datahelpdesk.worldbank.org/knowledgebase/articles/906519-world-bank-country-and-lending-groups)\n**6** The proportion of children is based on the availability of disaggregated data by age and sex for all people who are forcibly displaced,\nstateless, returnees or others of concern to UNHCR (77 per cent of 14.3 million).\n**7** Data Source: World Population Prospects 2022 (https://population.un.org/wpp/).\n\n\n2022 Asia & the Pacific Regional Population Trends Analysis: Forced Displacement 3\n\nMention: \"World Population Prospects 2022\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Source des populations mondiales et des personnes relevant de l'agence.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001300:12:2", "occurrence_id": "4baa8b46ddf3405346782748", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ladesh/brief/cox-s-bazar-panel-survey-briefs)</u> <sup>10</sup>\n\nFinally, refugees are included in samples and surveys conducted by national statistics\n\n\n[10 Two panel surveys conducted in Jordan are described in Krafft and Assad (2021) and in Stillman et al (2022).](https://sciendo.com/it/article/10.2478/izajodm-2021-0008)\n\n\n12\n\nMention: \"panel surveys conducted in Jordan\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Deux panneaux incluant les réfugiés dans les échantillons nationaux.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001305:2:4", "occurrence_id": "88e01394322af531e76588d0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: for Worldview,](https://bit.ly/2LKJSOY)</u> _Why Libyan Elections Probably Won't Happen this Year_, 4 June 2018,\n<u>[https://bit.ly/2K3ZvoI; ECFR,](https://bit.ly/2K3ZvoI)</u> _Libya Elections 2018: The Missing Ingredient_ [, 1 June 2018, https://bit.ly/2Hx7vrS; Human Rights Watch](https://bit.ly/2Hx7vrS)\n(HRW), _Libya: No Free Elections in Current Climate_, 21 March 2018, <u>[http://www.refworld.org/docid/5b6422ba4.html.](http://www.refworld.org/docid/5b6422ba4.html)</u>\n21 By May 2017, the Presidency Council reportedly consolidated its control over the capital Tripoli, with support from powerful Tripoli-based\narmed groups, including the Tripoli Revolutionaries Brigade (TRB), the Central Security/Abu Salim armed group (CS/AS) and the Special\nDeterrence Force (SDF); OHCHR, _Arbitrary_ _and_ _Unlawful_ _Detention_ _in_ _Libya_, April 2018,\n<u>[http://www.refworld.org/docid/5b5590154.html, p. 9. See also, Small Arms Survey,](http://www.refworld.org/docid/5b5590154.html)</u> _Capital of Militias_  _Tripoli’s Armed Groups\n\nMention: \"Small Arms Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Ses travaux documentent les dynamiques armées dans la capitale concernée.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001316:3:2", "occurrence_id": "5f984fa33f990dcbd50f9cfc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: راطية40،900( )، السودان744،100( السودان\n. ويف هذه 377،700( )، والعراق383،900( )، وأريترييا458،400( )، وميانامر470،600( <sup>(((</sup> )\n\n.البلدان أغلبية سكانية مسلمة، فاملسلمون هم جزء كبري من أعداد الالجئني\n\nتعاين هيئات اإلغاثة من حاجة مزمنة إىل مصادر متويل إضافية لتلبية احتياجات الالجئني. وألنّ\nاملسلمني هم اجلزء األكرب من الالجئني يف العامل، فمن املعقول التطلّع إىل املجتمع املسلم باعتباره\nمليار دوالر 200-20 ، قُدِّر ما منحه املسلمون من األموال بني2008 مصدرًا ممكنًا للتمويل. ويف عام\n\n؛ ويشمل هذا التقدير الصدقة والزكاة.(((سنويًّا\n\n\n\n(اجتاهات منتصف سنة2015 Mid-Year Trends ،املفوضية السامية لألمم املتحدة لشؤون الالجئني 1.\n6. اطُّلِع عىل -4 .) ص2015 ،) (جن\n\nMention: \"Mid-Year Trends\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Ses tendances de milieu d'année fondent les estimations citées.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001336:2:2", "occurrence_id": "5079022e4852ca8fa6ce9869", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  او نورو هاون ګ او پاکستان لهران ی ونه ګ افغانان د اهر کال زر\n\nړنیو غانانو لومدونکو افیرته ی راستن. د بانمن ړی کیتی نور هم زې ولن ټ د جذب ظرف کوربه د ځایي ر ډې د رې شم ګ رات والو ډله کبله صورت مومي. د ک\n\nاودونکو هیا ېځ ېک. د بیی شامل دې خدماتو ته السرسی،ا می تعل او حقوقي مرستیاځ، روغت پناهی، اوو یتړا ېک ړ خوا،ه ی مع،شت ې مک ځ ته السرس\n\nپالنیی ای مخت ندوي او هم په ملي پرېن ېښ ګرڅدونکو افغانانو اندیې د راستن ېک ځ من ته راغل په۲۰۱۶ې DiREC) چوي ی ه یټ کم (دونکو اجراېراستن\n\n.کوي ون ګډ د تنه ښ غو ۍ رامونو ېک د هغوګې ون ړ جو او پرو\n\n\nچارو وزارت سره په فعالهدونکو ېوالو او راستنګړې هګ ت\n\nMention: \"DiREC\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Ses estimations portent sur les retours des Afghans enregistrés.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001345:3:8", "occurrence_id": "cb8ec5c33cb3e2b1f1fb4f58", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  ي (م **ـ** لجتماع **ا** �ي **ـ** النف\nات **ـ** ام إدارة معلوم **ـ** ي بنظ� <sup>ن</sup> **ـ** ل المع **ـ** ق العم **ـ** ي فرياء � <sup>ف</sup> **ـ** ات أعض **ـ** بع منظم **ـ** دى س **ـ** ل\n)؛ ويعتــرب فريــقُ العمــل GBV IMS( العنــف المبــ�لجتماعــي **ا** ي عــى النــوع <sup>ن</sup>\nةَ **ـ** الهيئ <sup>1</sup> ي **ـ** لجتماع **ا** وع **ـ** ى الن **ـ** ي ع� <sup>ن</sup> **ـ** ف المب **ـ** ات العن **ـ** ام إدارة معلوم **ـ** ي بنظ� <sup>ن</sup> **ـ** المع\nا **ـ** ة عليه **ـ** ف، والمحافظ **ـ** ذا العن **ـ** ة به **ـ** ات ذات العاق **ـ** ع البيان **ـ** ن\n\nMention: \"GBV IMS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Le groupe de travail sur la violence fondée sur le genre s'appuie sur ce système.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001352:41:0", "occurrence_id": "dc6b84b0e9ea9baaa846e58c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: DTM Mali – Avril 2021\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n|Régions-Cercles<br>-Communes|Enfant ( < 18 ans )|Col3|Col4|Col5|Adulte (18 - 59 ans )|Col7|Col8|Personne âgée ( > 59 ans )|Col10|Col11|Total<br>Femme|Total<br>Homme|Total<br>Individus|\n|---|---|---|---|---|---|---|---|---|---|---|---|---|---|\n|**Régions-Cercles**<br>**-Communes**<br>|**Ménages**|**F **|**M **|**Total**<br>**Enfant**|**F **|**M **|**Total**<br>**Adulte**|**F **|**M **|**Total**<br>**Personne**<br>**âgée**|**Total**<br>**Personne**<br>**âgée**|**Total**<br>**Personne**<br>**âgée**|**Total**<br>**Personne**<br>**âgée**|\n|OUELESSEBOUGOU|17|119|11|130|43|11|54|6|4|10|168\n\nMention: \"DTM Mali – Avril 2021\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Dénombre les enfants et adultes par subdivision administrative.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001355:16:1", "occurrence_id": "226a6a36e2f204fb6605c76a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: t-earthquake-response-plan-october-2023-march-2024-november-2023-endarips)_\n_[iv GiHA, Gender Update #1: Forced returns from Pakistan, 2023](https://reliefweb.int/report/pakistan/gender-update-1-forced-returns-pakistan)_\n_[v USIP: Tracking the Taliban's (Mis)Treatment of Women,2023](https://www.usip.org/tracking-talibans-mistreatment-women)_\n_[vi USIP: The Challenges Facing Afghans with Disabilities, 2024](https://www.usip.org/publications/2024/02/challenges-facing-afghans-disabilities#:~:text=Attempts%20to%20Account%20for%20Disabled%20Individuals%20in%20Afghanistan&text=In%20one%20report%20released%20by,with%2041%20percent%20being%20women.)_\n_vii This percentage is debated, OCHA Humanitarian Overview 2023 indicates 10%, while REACH, in the WoAA 2022 indicates 23%. However,_\n_this latter high percentage is debated by experts of the International Crisis Group: Taliban Restrictions on Women's Rights Deepen_\n_[Afghanistan's Crisis, available at: https://icg-prod.s3.amazonaws.com/s3fs-public/2023-02/329-afghanistan-womens-rights.pdf)](https://icg-prod.s3.amazonaws.com/s3fs-public/2023-02/329-afghanistan-womens-rights.pdf)_\n_\n\nMention: \"OCHA Humanitarian Overview 2023\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Son indication diffère des autres estimations sur le même pourcentage.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001355:16:1", "occurrence_id": "a65559a542dd67a715d3461e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: t-earthquake-response-plan-october-2023-march-2024-november-2023-endarips)_\n_[iv GiHA, Gender Update #1: Forced returns from Pakistan, 2023](https://reliefweb.int/report/pakistan/gender-update-1-forced-returns-pakistan)_\n_[v USIP: Tracking the Taliban's (Mis)Treatment of Women,2023](https://www.usip.org/tracking-talibans-mistreatment-women)_\n_[vi USIP: The Challenges Facing Afghans with Disabilities, 2024](https://www.usip.org/publications/2024/02/challenges-facing-afghans-disabilities#:~:text=Attempts%20to%20Account%20for%20Disabled%20Individuals%20in%20Afghanistan&text=In%20one%20report%20released%20by,with%2041%20percent%20being%20women.)_\n_vii This percentage is debated, OCHA Humanitarian Overview 2023 indicates 10%, while REACH, in the WoAA 2022 indicates 23%. However,_\n_this latter high percentage is debated by experts of the International Crisis Group: Taliban Restrictions on Women's Rights Deepen_\n_[Afghanistan's Crisis, available at: https://icg-prod.s3.amazonaws.com/s3fs-public/2023-02/329-afghanistan-womens-rights.pdf)](https://icg-prod.s3.amazonaws.com/s3fs-public/2023-02/329-afghanistan-womens-rights.pdf)_\n_\n\nMention: \"WoAA 2022\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Son pourcentage élevé est contesté par les experts.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001384:31:2", "occurrence_id": "b57802c63e075bcd6838df11", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: \n|**Territoire**|**Collectivité**|KARUBA<br> <br> <br> <br> <br>|KABAYA|4|0,14%|\n|**Territoire**|**Collectivité**|KARUBA<br> <br> <br> <br> <br>|KALOBA|5|0,18%|\n|**Territoire**|**Collectivité**|KARUBA<br> <br> <br> <br> <br>|KANIRO|14|0,50%|\n\n\n\nData Center for IDP Population 32\n\nMention: \"Data Center for IDP Population\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Publie les décomptes de population déplacée par territoire et collectivité.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001417:37:2", "occurrence_id": "20958ac8162bf6f9c2092141", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ضية الخاصة بالعنف الجنسي والعنف الجندري. وقد أصدرت ال� <sup>ت</sup>\nيركّز **مرص** و **لبنان** و **ردنال** <sup>**أ**</sup> ي رس � <sup>ف</sup> ي هن ربات ال <sup>أ</sup> المفوضية تقريراً بعنوان «نساء بمفردهن»، حول الالجئات السوريات اللوا� <sup>ت</sup>\n\nدوار، والعزلة، والعنف الجنسي والعنف الجندري.مان المالي ، وتغ�ي ال <sup>أ</sup> عى مسائل الإسكان، والغذاء، والصحة، والعمل وال <sup>أ</sup>\n\n\n**نظـام إدارة المعلومات حول العنف الجندري**\n\n)، لضمان جمع وإدارة وتشارك بيانات آمنة GBVIMS( دعمت المفوضية ن� نظام إدارة المعلومات حول العنف الجندري <sup>ش</sup>\nي العمليات المختلفة. فنظام إدارة المعلومات حول العنف الجندري وأخالقية ورسية حول العنف الجنسي والعنف الج\n\nMention: \"GBVIMS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"L'agence a appuyé ce système pour la collecte et le partage sûrs des données.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001432:23:0", "occurrence_id": "a4db51e4bf3bce24c195d814", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <mark>Conflict Scan| Septembre 2018 – Tillabéri (</mark> Ayorou, Ouallam, Abala)\n\n\nImpact des tensions interethniques\n\n\nLes tensions interethniques ont été mentionnées par les répondants comme source potentiel de\nconflits violents et mortels au niveau des localités visées par ce conflict scan, ces tensions augmentent\nla méfiance et les préjugés entre les membres de la communauté. Compte tenu de la situation\nsécuritaire qui sujette déjà à des attaques groupes armés prônant la violence, le sentiment d’insécurité\naugmente dans l’esprit dans la population. Les répondants déclarent que si les altercations entre\nethnies continuent de se produire on ne saurait plus faire de différence entre ces dernières et les\nattaques des groupes armé parce que cela constitue une perte pour la diversité culturelle de la région.\nCes tensions accentuent les accusations mutuelles d’appartenance d’une ethnie ou une autre à des\ngroupes armés, cette situation constitue un obstacle à la fréquentation, l’interaction positive et la\nprotection des communautés. Un leader des communautés hôtes de Abala affirment : _«_ **_Nous n’avons_**\n**_pas de problème avec les Peulhs mais leur présence dans notre communauté nous expose à des_**\n**_attaques et des tueries qui peuvent augmenter l’insécurité_** _»_ .\n\n\nRisque des tensions interethniques\n\n\n\n\n\nLe code de couleur des tensions entre les\nethnies est **violet** . Ces conf\n\nMention: \"Conflict Scan\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Ses relevés décrivent l'impact des tensions dans les localités indiquées.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001434:38:1", "occurrence_id": "d4a682df663f13c4e8fbda64", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: , Misión\na Centroamérica: Viaje hacia EE.UU. de menores no acompañados, (noviembre de 2013). También se puede\nconsultar Center for Gender and Refugee Studies, Kids in Need of Defense (KIND), A treacherous Journey: Child\nMigrants Navigating the U.S. Immigration System (February, 2014).\n\n67. CoIDH, Derechos y garantías de niñas y niños en el contexto de la migración y/o en necesidad de protección, párr.\n250.\n\n68. Según datos de CEPAL la tasa de analfabetismo en Guatemala es la más elevada de Centroamérica y llega al\n25.2% de la población mayor de 15 años. Comisión Económica para América Latina, Panorama Social de América\nLatina (CEPAL, 2012). http://www.cepal.org/publicaciones/xml/5/48455/PanoramaSocial2012.pdf (Fecha de\nconsulta: 17 de julio de 2014).\n\n69. Es la referencia que se hace a la “novia” dentro del lenguaje de las pandillas.\n\n70. Sandra Guillot Cuéllar, “Poder y violencia en la zona gris. Un análisis de la situación de los niños salvadoreños\nmigrantes en los espacios sociales transnacionales desde la antropología de las emociones” (Tesis para obtener el\ngrado de Licenciada en Antropología social, México, D.F. 2013).\n\n71. Rodolfo Casillas, “Las Rutas de los Centroamericanos por México, un ejercicio de caracterización, actores principales\ny complejidades” en\n\nMention: \"datos de CEPAL\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Su tasa de analfabetismo es la más elevada de la región.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001437:0:7", "occurrence_id": "fab2697b34f1b452f1ce88ad", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  : (i) des difficultés d’accès et un besoin de sécurisation des axes routiers afin de<br>faciliter la libre circulation des personnes et de leurs biens et d’améliorer l’accès humanitaire, particulièrement Tombouctou-Bambara Maoudé ;<br>Tombouctou-Goundam-Tonka-Niafunké ; Tombouctou-Rharous-Gossi ; Léré-Niafunké; (ii) une augmentation des besoins humanitaires et un faible niveau de<br>financement pour des réponses durables aux crises (Source :Profil humanitaire de la région de Tombouctou, mars 2019).|\n|**Risques de Protection**|**Données existantes**|**Menaces**|**Vulnérabilités**|**Capacités**|**Evaluation**<br>**du risque**|\n|<br>**Violences psychologique et**<br>**émotionnelle**<br> <br> <br> <br>||La situation de crise aggrave les<br>cas de violence psychologique et<br>émotionnelle.<br>L’insécurité<br>grandissante avec des tirs à<br>l’arme légère ou les fortes<br>détonations créent la psychose<br>chez les enfants et les femmes.<br> <br>Le stress, l’insécurité  et la<br>précarité de condition de vie les<br>rendent certains hommes plus\n\nMention: \"Profil humanitaire de la région de Tombouctou\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Ses constats décrivent le faible financement des réponses durables.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001446:16:0", "occurrence_id": "82fef5b533b52c417c2e9d32", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: MONITOREO DE PROTECCIÓN\n\n#### 12. Cohesión Social e Integración en la Comunidad\n\n\nEl 84% de las personas encuestadas dijeron estar totalmente de acuerdo o de acuerdo de\nsentirse seguros en el área donde viven y sólo el 13% se mostró en desacuerdo o totalmente en\ndesacuerdo. Es importante mencionar que el tipo de delitos que las personas encuestadas\nmencionaron haber sido víctimas en Panamá aumentaron en severidad con respecto a la\ninformación recopilada durante el de Monitoreo de Protección realizado en 2019, ya que\nseñalaron que han sido sujetos a robos a mano armada, mientras que algunos reportaron haber\nsido testigos de actos de violencia como asesinatos y ejecuciones, narcotráfico, actividades\ndelictivas de pandillas y grupos criminales. Las personas de nacionalidad colombiana,\nnicaragüense y salvadoreña presentaron el mayor nivel de percepción de riesgo e inseguridad\nen sus comunidades o áreas de residencia.\n\n\nVarias personas mencionaron que sus hábitos diarios no\nincluyen actividades sociales o de esparcimiento en los\nespacios públicos, lo que se puede traducir en estilos de vida\nde aislamiento y únicamente enfocados en el grupo familiar,\nespecialmente en actividades que generen ingresos para cubrir\nnecesidades y servicios básicos del hogar, con poca\ninteracción fuera del hogar.\n\nAdemás, más de un tercio de las personas encuestadas dijo haber sido discriminado por ser\nextranjero, in\n\nMention: \"Monitoreo de Protección\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Su recopilación anterior sirve de comparación para la severidad.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001463:4:0", "occurrence_id": "62f585f00beeff5f32b27753", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **RESETTLEMENT KEY FIGURES**\n\n\n**Submitted Cases**\n\n\n\n**Departure Cases**\n\n\n\nREGIONAL BUREAU FOR SOUTHERN AFRICA\n**REFUGEES AND ASYLUM-SEEKERS INVOLVED IN RESETTLEMENT IN SOUTHERN AFRICA**\n\n\nAs of 30 November 2023\n\n\n**MOVEMENTS OF GROUPS OF 15 OR MORE RESETTLEMENT CASE MEMBERS**\n\n\nContry of Origin Country of Asylum Country of Resettlement\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n**MAP OF THE DEPARTURES BY COUNTRY OF SUBMISSION**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nData sources: UNHCR PRIMES, UNHCR Resettlement Statistics Report.\nFor more information or to contribute, please contact UNHCR RBSA DIMA (rsarbdima@unhcr.org)\n\nMention: \"UNHCR PRIMES\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Source des données par pays de soumission des dossiers.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001463:4:0", "occurrence_id": "69c29a8f0133f0d2a54daccc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **RESETTLEMENT KEY FIGURES**\n\n\n**Submitted Cases**\n\n\n\n**Departure Cases**\n\n\n\nREGIONAL BUREAU FOR SOUTHERN AFRICA\n**REFUGEES AND ASYLUM-SEEKERS INVOLVED IN RESETTLEMENT IN SOUTHERN AFRICA**\n\n\nAs of 30 November 2023\n\n\n**MOVEMENTS OF GROUPS OF 15 OR MORE RESETTLEMENT CASE MEMBERS**\n\n\nContry of Origin Country of Asylum Country of Resettlement\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n**MAP OF THE DEPARTURES BY COUNTRY OF SUBMISSION**\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nData sources: UNHCR PRIMES, UNHCR Resettlement Statistics Report.\nFor more information or to contribute, please contact UNHCR RBSA DIMA (rsarbdima@unhcr.org)\n\nMention: \"UNHCR Resettlement Statistics Report\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Seconde source des statistiques de réinstallation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001470:8:1", "occurrence_id": "c0a4802fd584506f5c55032f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  año pasado, representa un cambio\ntransformacional para miles de jóvenes y sus\ncomunidades.\n\n\n\n[4 UNESCO-UIS. (2021). Estadísticas del UIS. http://data.uis.unesco.org/](http://data.uis.unesco.org/) (disponible en inglés).\n\n[5 UNESCO-UIS. (2021). Estadísticas del UIS. http://data.uis.unesco.org/](http://data.uis.unesco.org/) (disponible en inglés).\n\n6 UNESCO-UIS. (2021).\n\n7 En función de todos los países con datos disponibles. La innovación, la acción conjunta de los socios y la mejor accesibilidad\na datos contribuyeron a alcanzar una tasa de inscripción más alta.\n\n\nA C N U R > **I N F O R M E D E E D U C A C I Ó N 2 0 2 1 >** **M A N T E N I E N D O E L R U M B O** **9**\n\nMention: \"Estadísticas del UIS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Su base de datos de indicadores educativos está disponible en línea.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001474:32:0", "occurrence_id": "e21ebcbde1eea3509cadd21b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Engagements pour l’Afrique australe en 2019\n\n\nScalabrini Centre (Le Cape)\n\n\n\n32\n\n\n\n\n- Plaider pour la ratification par l’Afrique du Sud des instruments internationaux relatifs à l’apatridie; collecte de\ndonnées et législation conforme aux conventions des Nations Unies.\n\n- Continuer à plaider contre l’apatridie en Afrique, en particulier en Afrique du Sud, par la recherche et le plaidoyer,\net aider les individus.\n\n\nSouthern African Nationality Network\n(Southern Africa)\n\n\n- Mener des campagnes de collecte de fonds pour soutenir les activités visant à protéger les apatrides et à mettre\nfin à l’apatridie\n\n- Plaider et soutenir la SADC dans la création d’un plan d’action régional pour mettre fin à l’apatridie dans la région\nde la SADC\n\n- Plaider pour l’adoption et la mise en œuvre rapides du projet de protocole sur l’élimination de l’apatridie en\nAfrique\n\n- Encourager les membres du réseau à intégrer l’apatridie dans le travail de leur organisation\n\n- Fournir une formation sur l’apatridie et la nationalité aux gouvernements de la SADC\n\n- Recherche sur les apatrides en Afrique australe et création d’une base de données SADC\n\n\nSolutions d’inclusion financière des réfugiés\n\n\n- Faire bénéficier à 1 000 000 de réfugiés dans la région australe et en Ouganda du « Modèle d’inclusion financière\net d’avancement des réfugiés » au cours des 3 prochaines années. Le modèle facilite l‘affranchissement des\n\nMention: \"base de données SADC\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Créée pour la recherche sur les apatrides en Afrique australe.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001480:3:1", "occurrence_id": "9b7ec046622238a7f35187c9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: --|---|\n|**0**<br>**601**<br>**592**<br>**242**<br>**224**<br>**184**<br>**COD**<br>**UGA**<br>**ZMB**<br>**MWI**<br>**MOZ**|**392**<br>**311**<br>**244**<br>**197**<br>**COD**<br>**ZMB**<br>**UGA**<br>**ZAF**<br>**MWI**||\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nData sources: UNHCR PRIMES.\nFor more information or to contribute, please contact\nUNHCR RBSA DIMA (rsarbdima@unhcr.org)\n\n\n|CUMULATIVE CROSS BORDER MOVEMENTS TREND PER YEAR AND PR|OPORTIONS SINCE 2019<br>Inward<br>movements<br>Intra region<br>25% movements<br>48%<br>Outward<br>movements<br>28%<br>shown are restricted only to movements of a minimum group pe<br>sylum, due to the change of nationality of some PoCs, country of o<br>the Congo , COG = Republic of the Congo, SSD- South Sudan, KE|\n|---|---|\n|926<br>3,276<br>1,777<br>2,051<br>4,755<br>2,554<br>Before<br>2019<br>2019<br>2020<br>\n\nMention: \"UNHCR PRIMES\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Source unique des données présentées pour ce pays.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001493:4:6", "occurrence_id": "40fdb419884f65ad42e01286", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  نبغو لامنيمأ مراجعأ وتوتتراتاج تذا **د** اع ات **د** )1 <sup>(</sup> اأ **د** اا الردام **د** من التوص عدع ا2022عام\n\nؤول الثجئ ن وإ ش **د** اأ ش **د** تراتاجاأ رإلوعفا األعاءا والتزمت ماوت **د** اعاتذا و فيا المجال واتااك مإلا بأ أك.ر ات **د**\n\nةتإلدرا ععا أك.ر عالاأ لادول و الا اتدتجابأ اعا تذا لذيا التإل ااا ويشدمل كلط وتدع اتدتراتاجاأ لممداأ\n\nبشكل تااعل المنيمأ و فيا المجال لثل **دددددد** او الوال  مع الدول تش **دددددد** أل تنماأ قد ا الاجوء بالتش **دددددد**\n\nالمنوا الإلاعمأا\n\n\n\n__________\n\nمتاح على UNHCR, “UNHCR Asylum Capacity Development Evaluation” ) 1(\n\n[.https://www.unhcr.org/media/es/2022/01-unhcr-asylum-capacity-development-evaluation](https://www.unhcr\n\nMention: \"UNHCR Asylum Capacity Development Evaluation\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Son évaluation sur le développement des capacités d'asile est citée.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001506:4:1", "occurrence_id": "a8306167d661629b8f95f007", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: sup>2</sup> .\n\nDe las cerca de 141 mil hectáreas con suelos productivos del\ndepartamento de Norte de Santander, el 28% se encuentran\nutilizadas por cultivos de coca, encontrándose el 48% en el\nmunicipio de Tibú, seguido por El Tarra (15%), Sardinata (11%)\ny Teorama (11%).\n\n\n2 SIDCO: Sistema de información de Drogas de Colombia. Observatorio de Drogas\nde Colombia (O.D.C).\n\n5\n\nMention: \"Sistema de información de Drogas de Colombia\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Sus porcentajes por municipio provienen del observatorio nacional.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001526:49:1", "occurrence_id": "da3478e5606c5cc93d22d9d3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  food, cash or other resources 4% 3% 9% 0% 0% 0%\n\nLivelihood coping strategy to meet essential\nneeds\n\nHH not adopting coping strategies 55% 57% 59% 51% 48% 43%\n\nStress coping strategies 6% 9% 4% 5% 13% 5%\n\nCrisis coping strategy 10% 11% 11% 8% 13% 10%\n\nEmergency coping strategies 29% 22% 27% 36% 25% 43%\n\n\nSeptember 2023| UNHCR & WFP Joint Post Distribution Monitoring: South Sudan 50\n\nMention: \"Joint Post Distribution Monitoring\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Ses résultats décrivent les stratégies de coping d'urgence.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001542:12:0", "occurrence_id": "4077074d33d8a6f47fac9925", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: DTM Mali – Décembre 2021\n\n#### **Démographie**\n\n\nProfil démographique des PDIs\n\n\nTableau II : Répartition par tranches d’âges et par sexe des personnes déplacées internes au Mali\n\n\n\n\n\n|Enfants (0-17 ans)|Col2|Col3|Adultes (18 -59as)|Col5|Col6|Personne âgée (59ans et plus)|Col8|Col9|TOTAL|\n|---|---|---|---|---|---|---|---|---|---|\n|**Filles**|**Garçons**|**Total**|**Femmes**|**Hommes**|**Total**|**Femmes**|**Hommes**|**Total**|**Total**|\n|116 715|106 718|223 433|71 638|47 294|118 932|4 409|3 336|7 745|350 110|\n|33%|31%|64%|20%|14%|34%|1%|1%|2%|**100%**|\n\n\nLa population déplacée est constituée de 55 pour cent de femmes et 45 pour cent d’hommes. Par ailleurs,\n64% des personnes déplacées sont des enfants (âgés de 0 à 17 ans), la population active représente 34%\nde la population déplacée et les 60 ans et plus représentent 2%.\n\nTableau III : Répartition des déplacés climatique (\n\nMention: \"DTM Mali\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Son profil démographique couvre les personnes déplacées internes.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001566:41:0", "occurrence_id": "991b50052b78c9af7c345645", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: hôtes. **Pourtant, par rapport à l'ensemble de la Mauritanie, où la part des ménages possédant un bien de**\n**base est supérieure à 89%,** <sup>**29**</sup> [^29: Enquête EDSM 2021. Même dans les zones rurales, la part est supérieure à 83% au niveau national.] **même les communautés hôtes sont relativement pauvres en biens.**\n\n\n**Si l'on se concentre sur les réfugiés, le clivage selon l'ethnie et le sexe est le même que pour la sécurité**\n**alimentaire.** Toutefois, le schéma géographique est moins extrême et quelque peu différent.\n\n\nSi l'on considère les biens de base, les ménages dirigés\npar des hommes sont beaucoup plus nombreux à les\nposséder (46,8%) que ceux dirigés par des femmes\n(32,3%). La différence entre les ethnies est beaucoup plus\nfaible. Alors que 37,3% des ménages de langue\ntouarègue possèdent des biens de base, c'est le cas de\n40,8% des ménages de langue arabe et 42,9% des\nménages parlant d’autres langues. Selon l'âge du chef de\nménage, il n'y a guère de différences, les ménages dirigés\npar un adulte en âge de travailler étant légèrement plus\nsusceptibles de posséder des biens de base (39,1 %) que\nceux dirigés par une personne âgée (36,4 %). Les tendances sont les même\n\nMention: \"Enquête EDSM 2021\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Ses résultats portent la part des ménages possédant un bien de base au-delà du seuil indiqué.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001634:18:6", "occurrence_id": "b3ef164d08465e59ebcdbff4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ghanistan)_\n\n_[xxvi IOM, Story: Every Step of Resilience: The Unwavering Spirit of Afghans Navigating Life Through Crises, December 2024](https://roasiapacific.iom.int/stories/every-step-resilience-unwavering-spirit-afghans-navigating-life-through-crises)_\n\n_[xxvii UNHCR, Afghanistan Border Monitoring Report January-September 2024](https://data.unhcr.org/en/documents/details/112585#:~:text=UNHCR%20Afghanistan%20Border%20Monitoring%20Report%20January%2DSeptember%202024%3A%20UNHCR's%20border,faced%20across%20official%20crossing%20points.)_\n\n_[xxviii UNOCHA, Multi-Sectoral Rapid Needs Assessment of Afghan returnees, April 2024 ; ADSP, Samuel Hall, Briefing Note: Solutions for Afghan nationals ordered](https://www.unocha.org/publications/report/afghanistan/multi-sectoral-rapid-needs-assessment-mrat-afghan-returnees-april-2024)_\n_<u>[to return from Pakistan, May 2024](https://static1.squarespace.com/static/5cfe2c8927234e0001688343/t/6666955555ed6e514bb2b991/1717998937637/Briefing-Note_Solutions-for-Afghan-nationals-from-Pakistan-1.pdf)</u>_\n\n_[xxix GiHA, Gender Alert: Returns from Iran and Pakistan, July 2024](https://reliefweb.int/report/afghanistan/gender-alert-return\n\nMention: \"Afghanistan Border Monitoring Report\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Son suivi frontalier couvre cette période.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001654:2:1", "occurrence_id": "2064a870385eeab616fd2f33", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ية](https://www.worldbank.org/en/topic/labormarkets/brief/migration-and-remittances) **:المصدر**\n#### الديناميات على المستوى دون اإلقليمي\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n[.Internal Displacement Monitoring Centre, Global Internal Displacement Database, 2020 ؛DESA, International Migration Stock, 2020 حسابات اإلسكوا استناداً إلى](https://www.internal-displacement.org/database/displacement-data.%20Accessed%20on%203%20June%202021) **:المصدر**\n\nيستخدم مصطلح “المهاجرون والالجئون” في بيانات إدارة الشؤون لجتماعية ألن البيانات تشمل الفئتين. ومع ذلك، من المهم اإلشارة إلى اختالف الوقائع وأنماطلقتصادية و **ا** **ا** **:مالحظة**\n\n\n\n**ا** **ا**\n\n.الهجرة في المنطقة، حيث تستضيف مجموعة بلدان مجلس التعاون الخليجي في الغالب العمال المهاجرين، بينما يتركز الالجئون في الغالب في المشرق\n\nMention: \"Global Internal Displacement Database\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Source des calculs régionaux sur le déplacement interne.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001654:2:1", "occurrence_id": "3f263a383bbc2c118a3a49ad", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: ية](https://www.worldbank.org/en/topic/labormarkets/brief/migration-and-remittances) **:المصدر**\n#### الديناميات على المستوى دون اإلقليمي\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n[.Internal Displacement Monitoring Centre, Global Internal Displacement Database, 2020 ؛DESA, International Migration Stock, 2020 حسابات اإلسكوا استناداً إلى](https://www.internal-displacement.org/database/displacement-data.%20Accessed%20on%203%20June%202021) **:المصدر**\n\nيستخدم مصطلح “المهاجرون والالجئون” في بيانات إدارة الشؤون لجتماعية ألن البيانات تشمل الفئتين. ومع ذلك، من المهم اإلشارة إلى اختالف الوقائع وأنماطلقتصادية و **ا** **ا** **:مالحظة**\n\n\n\n**ا** **ا**\n\n.الهجرة في المنطقة، حيث تستضيف مجموعة بلدان مجلس التعاون الخليجي في الغالب العمال المهاجرين، بينما يتركز الالجئون في الغالب في المشرق\n\nMention: \"International Migration Stock\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Seconde source des calculs régionaux sur les stocks migratoires.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001688:0:2", "occurrence_id": "9e28bbab8650576e44de77cf", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .</u> **<u>2</u>**\n\n\n\n\n|~0.05 arc<br>degrees|1984-2018|https://developers.google.com/earth-engine/datasets/|\n|---|---|---|\n|~0.05 arc<br>degrees|1984-2018|catalog/UCSB-CHG_CHIRPS_DAILY|\n|500 m|2016|https://developers.google.com/earth-engine/datasets/|\n|500 m|2016|catalog/MODIS_006_MCD12Q1|\n|~30 m|-|https://lpdaac.usgs.gov/products/nasadem_hgtv001/|\n|~30 m|-||\n|90 m|-|https://code.earthengine.google.com/<br>f7174cd11d23d78baca4021f2645ef33|\n\n\n\n**Yemen Flood Susceptibility - Analysis Methodology, March 2020**\n\nMention: \"UCSB-CHG_CHIRPS_DAILY\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Couche journalière à résolution fine couvrant cette période.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "reliefweb", "source_key": "reliefweb:001688:0:2", "occurrence_id": "ea2a87a8a49e766af0365a56", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .</u> **<u>2</u>**\n\n\n\n\n|~0.05 arc<br>degrees|1984-2018|https://developers.google.com/earth-engine/datasets/|\n|---|---|---|\n|~0.05 arc<br>degrees|1984-2018|catalog/UCSB-CHG_CHIRPS_DAILY|\n|500 m|2016|https://developers.google.com/earth-engine/datasets/|\n|500 m|2016|catalog/MODIS_006_MCD12Q1|\n|~30 m|-|https://lpdaac.usgs.gov/products/nasadem_hgtv001/|\n|~30 m|-||\n|90 m|-|https://code.earthengine.google.com/<br>f7174cd11d23d78baca4021f2645ef33|\n\n\n\n**Yemen Flood Susceptibility - Analysis Methodology, March 2020**\n\nMention: \"MODIS_006_MCD12Q1\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Couche à 500 m couvrant l'occupation des sols.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:004717:41:2", "occurrence_id": "a19d63c4f881e8aa68b952a8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Foundational Urban Infrastructure and Climate Resilience|Col2|\n|---|---|\n|**Person-days of jobs created (Days) **|**Person-days of jobs created (Days) **|\n|Description|Measures short-term employment generated through labor-intensive works under drainage, solid waste management,<br>urban upgrading, and housing reconstruction activities.|\n|Frequency|Quarterly|\n|Data source|Contractor payrolls; site supervision reports|\n\n\n\n32 Andrianirina, Ratsialonana Rivo; Raserijaona, Gaël; Anand, Aanchal; Teyssier, André. Towards an Urban Land Reform in Madagascar? : Support to Urban\nLand Policy (English). Washington, D.C.: World Bank Group. http://documents.worldbank.org/curated/en/517211592452939643.\n\n\nPage 28\n\n\nOfficial Use Only\n\nMention: \"Contractor payrolls; site supervision reports\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Source trimestrielle pour l'indicateur d'emploi et de reconstruction.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000214:17:0", "occurrence_id": "39ba3259890f9fa8f994f5a3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nLao PDR Civil Registration and Vital Statistics Project (P167601)\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n|No.|Province|Region|Districts<br>(number)|Population<br>(number)|CMIS<br>Phase|Health<br>Facilities<br>Ready for<br>DHIS2a<br>(number)|NT2 with<br>Health<br>Database|Intervention<br>versus<br>Control|\n|---|---|---|---|---|---|---|---|---|\n|17|Attapeu|Southern|5|139,628|IV|—|—|Control|\n|16|Saravane|Southern|8|396,942|IV|46|—|Control|\n|18|Sekong|Southern|4|113,048|IV|18||Control|\n\n\n_Note:_ a. Number of health care facilities that can directly access DHIS2 (that have computers, Internet connection,\nand staff with basic computer skills), based on the MOH 2017 data.\nb. Provinces for the nutrition convergence approach activities.\n\n\nPage 13 of 54\n\nMention: \"MOH 2017 data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Décrit l'équipement et les compétences informatiques du personnel par province.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:001096:33:2", "occurrence_id": "9adc1fd1d4daa4bac0966da2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: records newly registered<br>and/or updated in LRMIS<br>(Number)||0.00|0.00|574,000.00|2,147,000.00<br>|3,910,000.00|4,640,000.00|\n|of which, newly registered<br>and/or updated land<br>records jointly filed or<br>individually filed by women<br>(Number)||0.00|0.00|174,000.00|661,000.00<br>|1,228,000.00|1,485,000.00|\n|Area covered by cadastral<br>surveying (Hectare(Ha))||0.00|1,800.00|199,000.00|424,000.00<br>|690,000.00|853,000.00|\n\n\nPage 24 of 51\n\nMention: \"LRMIS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Enregistre les titres nouvellement immatriculés ou mis à jour.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000975:17:0", "occurrence_id": "b94c3f1cb5e87d7ffe14513c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nHealth System Performance Strengthening Project (P172504)\n\n\n**Figure 2. Under-5 Mortality Rate per Province (per 1,000 Live Births)**\n\n\n250\n\n\n200\n\n\n150\n\n\n100\n\n\n50\n\n\n0\n\n\n_Source:_ DHS MICS 2014–2015.\n\n|Col1|Table 2. Adult Mortality Rate per Age Group for Men|Col3|and Women in 2004 and 2014–2015|Col5|\n|---|---|---|---|---|\n||**Women Ages 15–49 Mortality Rate (per 1,000)**|**Women Ages 15–49 Mortality Rate (per 1,000)**|**Men Ages 15–49 Mortality Rate (per 1,000)**|**Men Ages 15–49 Mortality Rate (per 1,000)**|\n||**DHS-II 2004**|**DHS MICs 2014–2015**|**DHS-II 2004**|**DHS MICs 2014–2015**|\n|15–19|4.340|3.560|5.720|4.160|\n|20–24|3.288|4.080|5.828|4.450|\n|25–29|5.363|4.670|4.931|4.490|\n|30–34|6.565|5.860|7.229|5.720|\n|35–39|4.925|5.450|5.219|6.320|\n|40–44|8.437|6.790|6.018|8.240|\n|45–49|6.950|5.220|8.740|8.390|\n|**15–49**|**5\n\nMention: \"DHS MICS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Source des taux de mortalité par province et par groupe d'âge.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000975:17:0", "occurrence_id": "9c69c0018e3d050e110b8456", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nHealth System Performance Strengthening Project (P172504)\n\n\n**Figure 2. Under-5 Mortality Rate per Province (per 1,000 Live Births)**\n\n\n250\n\n\n200\n\n\n150\n\n\n100\n\n\n50\n\n\n0\n\n\n_Source:_ DHS MICS 2014–2015.\n\n|Col1|Table 2. Adult Mortality Rate per Age Group for Men|Col3|and Women in 2004 and 2014–2015|Col5|\n|---|---|---|---|---|\n||**Women Ages 15–49 Mortality Rate (per 1,000)**|**Women Ages 15–49 Mortality Rate (per 1,000)**|**Men Ages 15–49 Mortality Rate (per 1,000)**|**Men Ages 15–49 Mortality Rate (per 1,000)**|\n||**DHS-II 2004**|**DHS MICs 2014–2015**|**DHS-II 2004**|**DHS MICs 2014–2015**|\n|15–19|4.340|3.560|5.720|4.160|\n|20–24|3.288|4.080|5.828|4.450|\n|25–29|5.363|4.670|4.931|4.490|\n|30–34|6.565|5.860|7.229|5.720|\n|35–39|4.925|5.450|5.219|6.320|\n|40–44|8.437|6.790|6.018|8.240|\n|45–49|6.950|5.220|8.740|8.390|\n|**15–49**|**5\n\nMention: \"DHS-II 2004\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Sert de référence antérieure pour les taux de mortalité adulte.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000975:17:0", "occurrence_id": "7e61063f8e7ce90e756cc596", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nHealth System Performance Strengthening Project (P172504)\n\n\n**Figure 2. Under-5 Mortality Rate per Province (per 1,000 Live Births)**\n\n\n250\n\n\n200\n\n\n150\n\n\n100\n\n\n50\n\n\n0\n\n\n_Source:_ DHS MICS 2014–2015.\n\n|Col1|Table 2. Adult Mortality Rate per Age Group for Men|Col3|and Women in 2004 and 2014–2015|Col5|\n|---|---|---|---|---|\n||**Women Ages 15–49 Mortality Rate (per 1,000)**|**Women Ages 15–49 Mortality Rate (per 1,000)**|**Men Ages 15–49 Mortality Rate (per 1,000)**|**Men Ages 15–49 Mortality Rate (per 1,000)**|\n||**DHS-II 2004**|**DHS MICs 2014–2015**|**DHS-II 2004**|**DHS MICs 2014–2015**|\n|15–19|4.340|3.560|5.720|4.160|\n|20–24|3.288|4.080|5.828|4.450|\n|25–29|5.363|4.670|4.931|4.490|\n|30–34|6.565|5.860|7.229|5.720|\n|35–39|4.925|5.450|5.219|6.320|\n|40–44|8.437|6.790|6.018|8.240|\n|45–49|6.950|5.220|8.740|8.390|\n|**15–49**|**5\n\nMention: \"DHS MICs 2014–2015\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Source comparative des taux de mortalité adulte.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:001757:13:3", "occurrence_id": "1df1ad7c9ae376cd0df4b7d3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .aspx)_\n_[lieu-thong-ke-ve-tinh-trang-dinh-duong-tre-em-qua-cac-nam.aspx](http://viendinhduong.vn/news/vi/106/61/0/a/so-lieu-thong-ke-ve-tinh-trang-dinh-duong-tre-em-qua-cac-nam.aspx)_ .\n11 General Statistics Office and UNICEF. 2015. Vietnam Multiple Indicator Cluster Survey 2014, Final Report. Ha Noi, Vietnam\n12 General Statistics Office and UNICEF. 2015. Vietnam Multiple Indicator Cluster Survey 2014, Final Report. Ha Noi, Vietnam\n13 General Statistics Office and UNICEF. 2015. Vietnam Multiple Indicator Cluster Survey 2014, Final Report. Ha Noi, Vietnam.\n\n\nPage 9 of 31\n\nMention: \"Vietnam Multiple Indicator Cluster Survey 2014\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Rapport final sur les indicateurs multiples des enfants.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:001897:14:2", "occurrence_id": "f7bf3efe5362dce97c73781c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: . <sup>23</sup> Climate change\n\n\n19 Source: TURKSTAT Water and Wastewater Statistics, 2020 Bulletin dated 16 December 2021.\n_<u>[https://data.tuik.gov.tr/Bulten/Index?p=Water-and-Wastewater-Statistics-2020-37197](https://data.tuik.gov.tr/Bulten/Index?p=Water-and-Wastewater-Statistics-2020-37197)</u>_\n20 International Water Association data, 2019\n21 Source: TURKSTAT Water and Wastewater Statistics, 2020 Bulletin dated 16 December 2021.\n_<u>[https://data.tuik.gov.tr/Bulten/Index?p=Water-and-Wastewater-Statistics-2020-37197](https://data.tuik.gov.tr/Bulten/Index?p=Water-and-Wastewater-Statistics-2020-37197)</u>_\n22 22 Source: Bilgehan Nas, Sinan Uyanik, Ahmet Aygün, Selim Doğan, Gürsel Erul, K. Batuhan Nas, Sefa Turgut, Mustafa Cop, Taylan Dolu;\nWastewater reuse in Türkiye: from present status to future potential. Water Supply 1 February 2020; 20 (1): 73–82. doi:\n_[https://doi.org/10.2166/ws.2019.136](https://doi.org/10.2166/ws.2019.136)_\n23 Data source: World Bank DataBank, https://data.worldbank.org/indicator/SL.AGR.EMPL.ZS?locations=TR.\n\n\nPage 11 of 85\n\nMention: \"TURKSTAT Water and Wastewater Statistics\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Bulletin statistique sur l'eau et les eaux usées.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:001897:14:2", "occurrence_id": "f253ec214fedc39de18481f3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: . <sup>23</sup> Climate change\n\n\n19 Source: TURKSTAT Water and Wastewater Statistics, 2020 Bulletin dated 16 December 2021.\n_<u>[https://data.tuik.gov.tr/Bulten/Index?p=Water-and-Wastewater-Statistics-2020-37197](https://data.tuik.gov.tr/Bulten/Index?p=Water-and-Wastewater-Statistics-2020-37197)</u>_\n20 International Water Association data, 2019\n21 Source: TURKSTAT Water and Wastewater Statistics, 2020 Bulletin dated 16 December 2021.\n_<u>[https://data.tuik.gov.tr/Bulten/Index?p=Water-and-Wastewater-Statistics-2020-37197](https://data.tuik.gov.tr/Bulten/Index?p=Water-and-Wastewater-Statistics-2020-37197)</u>_\n22 22 Source: Bilgehan Nas, Sinan Uyanik, Ahmet Aygün, Selim Doğan, Gürsel Erul, K. Batuhan Nas, Sefa Turgut, Mustafa Cop, Taylan Dolu;\nWastewater reuse in Türkiye: from present status to future potential. Water Supply 1 February 2020; 20 (1): 73–82. doi:\n_[https://doi.org/10.2166/ws.2019.136](https://doi.org/10.2166/ws.2019.136)_\n23 Data source: World Bank DataBank, https://data.worldbank.org/indicator/SL.AGR.EMPL.ZS?locations=TR.\n\n\nPage 11 of 85\n\nMention: \"International Water Association data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Source associative complémentaire sur l'eau.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:001897:14:2", "occurrence_id": "fb7299f8b8506697e9e815f2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: . <sup>23</sup> Climate change\n\n\n19 Source: TURKSTAT Water and Wastewater Statistics, 2020 Bulletin dated 16 December 2021.\n_<u>[https://data.tuik.gov.tr/Bulten/Index?p=Water-and-Wastewater-Statistics-2020-37197](https://data.tuik.gov.tr/Bulten/Index?p=Water-and-Wastewater-Statistics-2020-37197)</u>_\n20 International Water Association data, 2019\n21 Source: TURKSTAT Water and Wastewater Statistics, 2020 Bulletin dated 16 December 2021.\n_<u>[https://data.tuik.gov.tr/Bulten/Index?p=Water-and-Wastewater-Statistics-2020-37197](https://data.tuik.gov.tr/Bulten/Index?p=Water-and-Wastewater-Statistics-2020-37197)</u>_\n22 22 Source: Bilgehan Nas, Sinan Uyanik, Ahmet Aygün, Selim Doğan, Gürsel Erul, K. Batuhan Nas, Sefa Turgut, Mustafa Cop, Taylan Dolu;\nWastewater reuse in Türkiye: from present status to future potential. Water Supply 1 February 2020; 20 (1): 73–82. doi:\n_[https://doi.org/10.2166/ws.2019.136](https://doi.org/10.2166/ws.2019.136)_\n23 Data source: World Bank DataBank, https://data.worldbank.org/indicator/SL.AGR.EMPL.ZS?locations=TR.\n\n\nPage 11 of 85\n\nMention: \"World Bank DataBank\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Fournit l'indicateur d'emploi agricole du pays.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:002248:52:1", "occurrence_id": "7bda715fab0a5e8b7567dadb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: security and digital<br>government foundations<br>designed to strengthen<br>cybersecurity<br>(disaggregated by gender)|Annual<br>|MICT/DTO<br>|MICT/DTO progress<br>reports, outlining<br>attendance at<br>cybersecurity training<br>activities<br>|MICT/DTO<br>|\n|Of which female|<br>||<br>|||\n|Government Enterprise Architecture<br>Developed|Government Enterprise<br>Architecture has been<br>developed and is<br>operational|Annual<br>|MICT/DTO<br>progress<br>report /<br>project data<br>|MICT/DTO to report on<br>progress<br>|MICT/DTO<br>|\n\n\n\nPage 48\n\nMention: \"project data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Rapportees annuellement par la division du developpement des technologies.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:002356:2:2", "occurrence_id": "a263c83d8fe1c0903f6610d3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  and 1.18 million in Tamil Nadu live with some form of disability (Census**\n**2011). However, some recent studies** <sup>**1**</sup> **in Tamil Nadu estimate that these figures may just have doubled.** As per\n\n\n1 Rakhi Dandona, Anamika Pandey, Sibin George, G. Anil Kumar, Lalit Dandona, India’s disability estimates: Limitations and way forward, 2019\n\n\nSeptember 29, 2021 Page 3 of 10\n\nMention: \"Census**\n**2011\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Estime le nombre de personnes vivant avec une forme de handicap.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:003479:11:2", "occurrence_id": "829e10f6831c77b90490acda", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  severely food insecure, compared to 17.9 percent\nin urban areas. <sup>6</sup> This is likely to worsen and put further pressure on vulnerable households.\n\n\n1 Country Environmental Analysis (CEA, 2022)\n2 Comoros Poverty Assessment Report, World Bank, 2021.\n3Ibid.\n4 Ibid.\n5 Ibid.\n6 Ibid.\n\n\nPage 7 of 59\n\nMention: \"Comoros Poverty Assessment Report\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Evalue les pressions sur les menages vulnerables.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:003576:41:2", "occurrence_id": "eaca7b554ad4a4ed62660781", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: >|Quarterly<br>|\n|Data source<br>|Contractor completion certificates; SEEG operational records<br>|\n|Methodology for Data<br>Collection|Calculate the total of production facility plant capacity additions or restored capacity; verify with<br>as-built data.|\n|Responsibility for Data<br>Collection<br>|PMU M&E Specialist; MAUEE M&E Unit; SEEG M&E Unit; SEEG Works Department<br>|\n|**New piped household water connections (Number)**<br> <br>|**New piped household water connections (Number)**<br> <br>|\n|Description<br>|Cumulative number of new household water connections installed under the Project.<br>|\n|Frequency<br>|Quarterly<br>|\n|Data source|SEEG connection registry|\n|Methodology for Data<br>Collection|SEEG extracts quarterly connection data and validates against Project completion reports.|\n|Responsibility for Data<br>Collection<br>|PMU M&E Specialist; SEEG M&E Unit; SEEG Metering Division<br>|\n|**Existing piped household water connections benefiting from rehabilitation works undertaken (Number)**|**Existing piped household water connections benefiting from rehabilitation works undertaken (Number)**|\n\n\n\nPage 31\n\nMention: \"SEEG operational records\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Registres trimestriels des capacites de production restaurees.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:003576:41:2", "occurrence_id": "402a5d64a2e58eb5b3c045bc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: >|Quarterly<br>|\n|Data source<br>|Contractor completion certificates; SEEG operational records<br>|\n|Methodology for Data<br>Collection|Calculate the total of production facility plant capacity additions or restored capacity; verify with<br>as-built data.|\n|Responsibility for Data<br>Collection<br>|PMU M&E Specialist; MAUEE M&E Unit; SEEG M&E Unit; SEEG Works Department<br>|\n|**New piped household water connections (Number)**<br> <br>|**New piped household water connections (Number)**<br> <br>|\n|Description<br>|Cumulative number of new household water connections installed under the Project.<br>|\n|Frequency<br>|Quarterly<br>|\n|Data source|SEEG connection registry|\n|Methodology for Data<br>Collection|SEEG extracts quarterly connection data and validates against Project completion reports.|\n|Responsibility for Data<br>Collection<br>|PMU M&E Specialist; SEEG M&E Unit; SEEG Metering Division<br>|\n|**Existing piped household water connections benefiting from rehabilitation works undertaken (Number)**|**Existing piped household water connections benefiting from rehabilitation works undertaken (Number)**|\n\n\n\nPage 31\n\nMention: \"as-built data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Verifient les capacites de production ajoutees ou restaurees.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:003576:41:2", "occurrence_id": "2f8200645ac7bbec1a8156e9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: >|Quarterly<br>|\n|Data source<br>|Contractor completion certificates; SEEG operational records<br>|\n|Methodology for Data<br>Collection|Calculate the total of production facility plant capacity additions or restored capacity; verify with<br>as-built data.|\n|Responsibility for Data<br>Collection<br>|PMU M&E Specialist; MAUEE M&E Unit; SEEG M&E Unit; SEEG Works Department<br>|\n|**New piped household water connections (Number)**<br> <br>|**New piped household water connections (Number)**<br> <br>|\n|Description<br>|Cumulative number of new household water connections installed under the Project.<br>|\n|Frequency<br>|Quarterly<br>|\n|Data source|SEEG connection registry|\n|Methodology for Data<br>Collection|SEEG extracts quarterly connection data and validates against Project completion reports.|\n|Responsibility for Data<br>Collection<br>|PMU M&E Specialist; SEEG M&E Unit; SEEG Metering Division<br>|\n|**Existing piped household water connections benefiting from rehabilitation works undertaken (Number)**|**Existing piped household water connections benefiting from rehabilitation works undertaken (Number)**|\n\n\n\nPage 31\n\nMention: \"SEEG connection registry\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Registre trimestriel des raccordements installes.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:003638:39:0", "occurrence_id": "c1c207cf0f5bb19687fd0311", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **<mark>Banco Mundial</mark>**\n<mark>Mejoramiento de los Servicios de Información del MIDIS dentro del Sistema Nacional de</mark>\nFocalización (SINAFO) a Nivel Nacional (P179923)\n\n\n**Plan de Monitoreo y Evaluación: Indicadores de ODP por Resultados de ODP**\n\n|Hacer que el Sistema Nacional de Focalización de Perú sea más inclusivo para su uso por programas sociales focalizados|Col2|\n|---|---|\n|**Porcentaje de hogares que viven en áreas seleccionadas de Perú con información actualizada en el PGH**|**Porcentaje de hogares que viven en áreas seleccionadas de Perú con información actualizada en el PGH**|\n|Descripción||\n|Frecuencia|**Semestral**|\n|Fuente de datos|**Informe PGH**|\n|Metodología para la<br>recolección de datos|**Cifras actuales**<br> <br>**Numerador: Número de hogares que viven en áreas seleccionadas de Perú con información**<br>**actualizada en el PGH.**<br>**Denominador: Número de hogares que viven en áreas seleccionadas de Perú (estimado**<br>**siguiendo la metodología descrita en la Nota Técnica de MIDIS No. por definir)**|\n|Responsabilidad de la<br>recolección de datos|**DGFIS/MI\n\nMention: \"Informe PGH\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Source semestrielle des chiffres actuels du programme.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:003638:51:1", "occurrence_id": "2ec0142505b6b8f7987f10a3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  PMI para crear un índice de focalización de\nhogares (IFH). La primera fase de implementación del PGH se centró en áreas urbanas y, para 2007, incluía\ndatos de 1.3 millones de hogares que vivían en las 25 ciudades más grandes del país. En 2008, el PGH se\namplió con la información de los hogares rurales recopilada por el programa JUNTOS como parte de su\nproceso de focalización. Entre 2010 y 2011, el MEF continuó la implementación del PGH a través de una\nactualización del algoritmo IFH utilizando los datos de la Encuesta Nacional de Hogares (ENAHO) de 2009\ny el formato y las directrices de la FSU para la expansión del PGH utilizando un enfoque censal que se\naplicaría centralmente o por las unidades locales de focalización de los municipios (ULE).\n\n\n3. La creación del MIDIS en 2011 creó un nuevo marco institucional y de políticas para el PGH. En\n2012, la administración del PGH pasó al MIDIS y el uso del PGH se volvió obligatorio para todos los\nprogramas sociales con clasificación socioeconómica individual o de hogares.\n\n\n4. En 2013, se llevó a cabo la recolección de datos de hogares a mayor escala en la historia del PGH,\nalcanzando a más de 3 millones de hogares con un enfoque censal. En el mismo año, el MIDIS inició una\nserie de ajustes a la metodología y operatividad del PGH para alinearlo con el nuevo marco de políticas y\nhacerlo un instrumento\n\nMention: \"datos de 1.3 millones de hogares\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Donnees des menages des grandes villes incluses dans le programme.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:003638:51:1", "occurrence_id": "f7807570c110a52750eb9762", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  PMI para crear un índice de focalización de\nhogares (IFH). La primera fase de implementación del PGH se centró en áreas urbanas y, para 2007, incluía\ndatos de 1.3 millones de hogares que vivían en las 25 ciudades más grandes del país. En 2008, el PGH se\namplió con la información de los hogares rurales recopilada por el programa JUNTOS como parte de su\nproceso de focalización. Entre 2010 y 2011, el MEF continuó la implementación del PGH a través de una\nactualización del algoritmo IFH utilizando los datos de la Encuesta Nacional de Hogares (ENAHO) de 2009\ny el formato y las directrices de la FSU para la expansión del PGH utilizando un enfoque censal que se\naplicaría centralmente o por las unidades locales de focalización de los municipios (ULE).\n\n\n3. La creación del MIDIS en 2011 creó un nuevo marco institucional y de políticas para el PGH. En\n2012, la administración del PGH pasó al MIDIS y el uso del PGH se volvió obligatorio para todos los\nprogramas sociales con clasificación socioeconómica individual o de hogares.\n\n\n4. En 2013, se llevó a cabo la recolección de datos de hogares a mayor escala en la historia del PGH,\nalcanzando a más de 3 millones de hogares con un enfoque censal. En el mismo año, el MIDIS inició una\nserie de ajustes a la metodología y operatividad del PGH para alinearlo con el nuevo marco de políticas y\nhacerlo un instrumento\n\nMention: \"información de los hogares rurales\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Recueillies par le programme pour le ciblage rural.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:003638:51:1", "occurrence_id": "fafbc310632ec0f9bd6c7f47", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  PMI para crear un índice de focalización de\nhogares (IFH). La primera fase de implementación del PGH se centró en áreas urbanas y, para 2007, incluía\ndatos de 1.3 millones de hogares que vivían en las 25 ciudades más grandes del país. En 2008, el PGH se\namplió con la información de los hogares rurales recopilada por el programa JUNTOS como parte de su\nproceso de focalización. Entre 2010 y 2011, el MEF continuó la implementación del PGH a través de una\nactualización del algoritmo IFH utilizando los datos de la Encuesta Nacional de Hogares (ENAHO) de 2009\ny el formato y las directrices de la FSU para la expansión del PGH utilizando un enfoque censal que se\naplicaría centralmente o por las unidades locales de focalización de los municipios (ULE).\n\n\n3. La creación del MIDIS en 2011 creó un nuevo marco institucional y de políticas para el PGH. En\n2012, la administración del PGH pasó al MIDIS y el uso del PGH se volvió obligatorio para todos los\nprogramas sociales con clasificación socioeconómica individual o de hogares.\n\n\n4. En 2013, se llevó a cabo la recolección de datos de hogares a mayor escala en la historia del PGH,\nalcanzando a más de 3 millones de hogares con un enfoque censal. En el mismo año, el MIDIS inició una\nserie de ajustes a la metodología y operatividad del PGH para alinearlo con el nuevo marco de políticas y\nhacerlo un instrumento\n\nMention: \"Encuesta Nacional de Hogares\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Donnees utilisant l'algorithme de focalisation mis a jour.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:003638:39:2", "occurrence_id": "689f29d087143f738b3ef868", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <br>**Denominador: Número de hogares que actualizaron su clasificación socioeconómica (a**<br>**demanda) en el año del informe.**<br>**DGFIS/MIDIS**|\n|**Hacer que el Sistema Nacional de Focalización de Perú sea más integrado para su uso por programas sociales focalizados**|**Hacer que el Sistema Nacional de Focalización de Perú sea más integrado para su uso por programas sociales focalizados**|\n|**Número de nuevas bases de datos administrativas que intercambian información oportuna en SINAFO**|**Número de nuevas bases de datos administrativas que intercambian información oportuna en SINAFO**|\n|Descripción<br>Frecuencia<br>Fuente de datos<br>Metodología para la<br>recolección de datos<br>Responsabilidad de la<br>recolección de datos|<br>**Anual**<br>**Informe DGFIS semestral del Proyecto**<br>**Cifras actuales**<br>**Bases de datos objetivo: SIGRID, SINPAD, CONADIS, MIGRACIONES, RENIEC, SUSALUD,**<br>**OSINERGMIN, SUNAT, SUNARP, SINADEF. Intercambiar información oportuna significa que no**<br>**se necesita ningún acto administrativo antes del intercambio de datos.**<br>**DGFIS/MIDIS**\n\nMention: \"SIGRID\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Base de donnees cible pour l'echange semestriel d'information.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:003638:39:2", "occurrence_id": "a4723d6a23c5df675ee11640", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <br>**Denominador: Número de hogares que actualizaron su clasificación socioeconómica (a**<br>**demanda) en el año del informe.**<br>**DGFIS/MIDIS**|\n|**Hacer que el Sistema Nacional de Focalización de Perú sea más integrado para su uso por programas sociales focalizados**|**Hacer que el Sistema Nacional de Focalización de Perú sea más integrado para su uso por programas sociales focalizados**|\n|**Número de nuevas bases de datos administrativas que intercambian información oportuna en SINAFO**|**Número de nuevas bases de datos administrativas que intercambian información oportuna en SINAFO**|\n|Descripción<br>Frecuencia<br>Fuente de datos<br>Metodología para la<br>recolección de datos<br>Responsabilidad de la<br>recolección de datos|<br>**Anual**<br>**Informe DGFIS semestral del Proyecto**<br>**Cifras actuales**<br>**Bases de datos objetivo: SIGRID, SINPAD, CONADIS, MIGRACIONES, RENIEC, SUSALUD,**<br>**OSINERGMIN, SUNAT, SUNARP, SINADEF. Intercambiar información oportuna significa que no**<br>**se necesita ningún acto administrativo antes del intercambio de datos.**<br>**DGFIS/MIDIS**\n\nMention: \"SINPAD\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Base de donnees cible pour l'echange d'information.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:003638:39:2", "occurrence_id": "41d82ff60f5b31649092a963", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <br>**Denominador: Número de hogares que actualizaron su clasificación socioeconómica (a**<br>**demanda) en el año del informe.**<br>**DGFIS/MIDIS**|\n|**Hacer que el Sistema Nacional de Focalización de Perú sea más integrado para su uso por programas sociales focalizados**|**Hacer que el Sistema Nacional de Focalización de Perú sea más integrado para su uso por programas sociales focalizados**|\n|**Número de nuevas bases de datos administrativas que intercambian información oportuna en SINAFO**|**Número de nuevas bases de datos administrativas que intercambian información oportuna en SINAFO**|\n|Descripción<br>Frecuencia<br>Fuente de datos<br>Metodología para la<br>recolección de datos<br>Responsabilidad de la<br>recolección de datos|<br>**Anual**<br>**Informe DGFIS semestral del Proyecto**<br>**Cifras actuales**<br>**Bases de datos objetivo: SIGRID, SINPAD, CONADIS, MIGRACIONES, RENIEC, SUSALUD,**<br>**OSINERGMIN, SUNAT, SUNARP, SINADEF. Intercambiar información oportuna significa que no**<br>**se necesita ningún acto administrativo antes del intercambio de datos.**<br>**DGFIS/MIDIS**\n\nMention: \"CONADIS\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Base de donnees cible pour l'echange d'information.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:003638:39:2", "occurrence_id": "e45b440fb9360db5041d3b47", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <br>**Denominador: Número de hogares que actualizaron su clasificación socioeconómica (a**<br>**demanda) en el año del informe.**<br>**DGFIS/MIDIS**|\n|**Hacer que el Sistema Nacional de Focalización de Perú sea más integrado para su uso por programas sociales focalizados**|**Hacer que el Sistema Nacional de Focalización de Perú sea más integrado para su uso por programas sociales focalizados**|\n|**Número de nuevas bases de datos administrativas que intercambian información oportuna en SINAFO**|**Número de nuevas bases de datos administrativas que intercambian información oportuna en SINAFO**|\n|Descripción<br>Frecuencia<br>Fuente de datos<br>Metodología para la<br>recolección de datos<br>Responsabilidad de la<br>recolección de datos|<br>**Anual**<br>**Informe DGFIS semestral del Proyecto**<br>**Cifras actuales**<br>**Bases de datos objetivo: SIGRID, SINPAD, CONADIS, MIGRACIONES, RENIEC, SUSALUD,**<br>**OSINERGMIN, SUNAT, SUNARP, SINADEF. Intercambiar información oportuna significa que no**<br>**se necesita ningún acto administrativo antes del intercambio de datos.**<br>**DGFIS/MIDIS**\n\nMention: \"MIGRACIONES\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Base de donnees migratoires cible pour l'echange.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:003638:39:2", "occurrence_id": "e1f2d122b569f8bd4364a6bd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <br>**Denominador: Número de hogares que actualizaron su clasificación socioeconómica (a**<br>**demanda) en el año del informe.**<br>**DGFIS/MIDIS**|\n|**Hacer que el Sistema Nacional de Focalización de Perú sea más integrado para su uso por programas sociales focalizados**|**Hacer que el Sistema Nacional de Focalización de Perú sea más integrado para su uso por programas sociales focalizados**|\n|**Número de nuevas bases de datos administrativas que intercambian información oportuna en SINAFO**|**Número de nuevas bases de datos administrativas que intercambian información oportuna en SINAFO**|\n|Descripción<br>Frecuencia<br>Fuente de datos<br>Metodología para la<br>recolección de datos<br>Responsabilidad de la<br>recolección de datos|<br>**Anual**<br>**Informe DGFIS semestral del Proyecto**<br>**Cifras actuales**<br>**Bases de datos objetivo: SIGRID, SINPAD, CONADIS, MIGRACIONES, RENIEC, SUSALUD,**<br>**OSINERGMIN, SUNAT, SUNARP, SINADEF. Intercambiar información oportuna significa que no**<br>**se necesita ningún acto administrativo antes del intercambio de datos.**<br>**DGFIS/MIDIS**\n\nMention: \"RENIEC\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Base d'identite cible pour l'echange d'information.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:003638:39:2", "occurrence_id": "63a23bf353a876ded613c478", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <br>**Denominador: Número de hogares que actualizaron su clasificación socioeconómica (a**<br>**demanda) en el año del informe.**<br>**DGFIS/MIDIS**|\n|**Hacer que el Sistema Nacional de Focalización de Perú sea más integrado para su uso por programas sociales focalizados**|**Hacer que el Sistema Nacional de Focalización de Perú sea más integrado para su uso por programas sociales focalizados**|\n|**Número de nuevas bases de datos administrativas que intercambian información oportuna en SINAFO**|**Número de nuevas bases de datos administrativas que intercambian información oportuna en SINAFO**|\n|Descripción<br>Frecuencia<br>Fuente de datos<br>Metodología para la<br>recolección de datos<br>Responsabilidad de la<br>recolección de datos|<br>**Anual**<br>**Informe DGFIS semestral del Proyecto**<br>**Cifras actuales**<br>**Bases de datos objetivo: SIGRID, SINPAD, CONADIS, MIGRACIONES, RENIEC, SUSALUD,**<br>**OSINERGMIN, SUNAT, SUNARP, SINADEF. Intercambiar información oportuna significa que no**<br>**se necesita ningún acto administrativo antes del intercambio de datos.**<br>**DGFIS/MIDIS**\n\nMention: \"SUSALUD\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Base de sante cible pour l'echange d'information.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:003638:39:2", "occurrence_id": "dff71e6994c05f9f064b8812", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <br>**Denominador: Número de hogares que actualizaron su clasificación socioeconómica (a**<br>**demanda) en el año del informe.**<br>**DGFIS/MIDIS**|\n|**Hacer que el Sistema Nacional de Focalización de Perú sea más integrado para su uso por programas sociales focalizados**|**Hacer que el Sistema Nacional de Focalización de Perú sea más integrado para su uso por programas sociales focalizados**|\n|**Número de nuevas bases de datos administrativas que intercambian información oportuna en SINAFO**|**Número de nuevas bases de datos administrativas que intercambian información oportuna en SINAFO**|\n|Descripción<br>Frecuencia<br>Fuente de datos<br>Metodología para la<br>recolección de datos<br>Responsabilidad de la<br>recolección de datos|<br>**Anual**<br>**Informe DGFIS semestral del Proyecto**<br>**Cifras actuales**<br>**Bases de datos objetivo: SIGRID, SINPAD, CONADIS, MIGRACIONES, RENIEC, SUSALUD,**<br>**OSINERGMIN, SUNAT, SUNARP, SINADEF. Intercambiar información oportuna significa que no**<br>**se necesita ningún acto administrativo antes del intercambio de datos.**<br>**DGFIS/MIDIS**\n\nMention: \"OSINERGMIN\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Base de donnees cible pour l'echange d'information.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:003638:39:2", "occurrence_id": "7ac87513428d45bbcee4bf3c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <br>**Denominador: Número de hogares que actualizaron su clasificación socioeconómica (a**<br>**demanda) en el año del informe.**<br>**DGFIS/MIDIS**|\n|**Hacer que el Sistema Nacional de Focalización de Perú sea más integrado para su uso por programas sociales focalizados**|**Hacer que el Sistema Nacional de Focalización de Perú sea más integrado para su uso por programas sociales focalizados**|\n|**Número de nuevas bases de datos administrativas que intercambian información oportuna en SINAFO**|**Número de nuevas bases de datos administrativas que intercambian información oportuna en SINAFO**|\n|Descripción<br>Frecuencia<br>Fuente de datos<br>Metodología para la<br>recolección de datos<br>Responsabilidad de la<br>recolección de datos|<br>**Anual**<br>**Informe DGFIS semestral del Proyecto**<br>**Cifras actuales**<br>**Bases de datos objetivo: SIGRID, SINPAD, CONADIS, MIGRACIONES, RENIEC, SUSALUD,**<br>**OSINERGMIN, SUNAT, SUNARP, SINADEF. Intercambiar información oportuna significa que no**<br>**se necesita ningún acto administrativo antes del intercambio de datos.**<br>**DGFIS/MIDIS**\n\nMention: \"SUNAT\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Base fiscale cible pour l'echange d'information.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:003638:39:2", "occurrence_id": "1121c43d2936f339b9095b66", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <br>**Denominador: Número de hogares que actualizaron su clasificación socioeconómica (a**<br>**demanda) en el año del informe.**<br>**DGFIS/MIDIS**|\n|**Hacer que el Sistema Nacional de Focalización de Perú sea más integrado para su uso por programas sociales focalizados**|**Hacer que el Sistema Nacional de Focalización de Perú sea más integrado para su uso por programas sociales focalizados**|\n|**Número de nuevas bases de datos administrativas que intercambian información oportuna en SINAFO**|**Número de nuevas bases de datos administrativas que intercambian información oportuna en SINAFO**|\n|Descripción<br>Frecuencia<br>Fuente de datos<br>Metodología para la<br>recolección de datos<br>Responsabilidad de la<br>recolección de datos|<br>**Anual**<br>**Informe DGFIS semestral del Proyecto**<br>**Cifras actuales**<br>**Bases de datos objetivo: SIGRID, SINPAD, CONADIS, MIGRACIONES, RENIEC, SUSALUD,**<br>**OSINERGMIN, SUNAT, SUNARP, SINADEF. Intercambiar información oportuna significa que no**<br>**se necesita ningún acto administrativo antes del intercambio de datos.**<br>**DGFIS/MIDIS**\n\nMention: \"SUNARP\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Base cadastrale cible pour l'echange d'information.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:003638:39:2", "occurrence_id": "415d37fe554beab36f6051a9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: <br>**Denominador: Número de hogares que actualizaron su clasificación socioeconómica (a**<br>**demanda) en el año del informe.**<br>**DGFIS/MIDIS**|\n|**Hacer que el Sistema Nacional de Focalización de Perú sea más integrado para su uso por programas sociales focalizados**|**Hacer que el Sistema Nacional de Focalización de Perú sea más integrado para su uso por programas sociales focalizados**|\n|**Número de nuevas bases de datos administrativas que intercambian información oportuna en SINAFO**|**Número de nuevas bases de datos administrativas que intercambian información oportuna en SINAFO**|\n|Descripción<br>Frecuencia<br>Fuente de datos<br>Metodología para la<br>recolección de datos<br>Responsabilidad de la<br>recolección de datos|<br>**Anual**<br>**Informe DGFIS semestral del Proyecto**<br>**Cifras actuales**<br>**Bases de datos objetivo: SIGRID, SINPAD, CONADIS, MIGRACIONES, RENIEC, SUSALUD,**<br>**OSINERGMIN, SUNAT, SUNARP, SINADEF. Intercambiar información oportuna significa que no**<br>**se necesita ningún acto administrativo antes del intercambio de datos.**<br>**DGFIS/MIDIS**\n\nMention: \"SINADEF\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Base de personnes disparues cible pour l'echange.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:003638:59:2", "occurrence_id": "41ce687e87674c63d1870a9d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: zar el ejercicio de recolección<br>de datos.<br> <br> <br>Además, en la plataforma de<br>capacitación, se capacitará a los<br>recursos humanos con los<br>protocolos de emergencia (por<br>ejemplo, choques relacionados con<br>el clima), el uso de mapas de<br>peligros climáticos y datos de<br>ubicación de hogares y la<br>implementación de la evaluación de<br>necesidades pos-desastre. MIDIS<br>invertirá US$164K en cursos de<br>capacitación específicos<br>relacionados con el clima.<br>|El equipo tecnológico<br>energéticamente eficiente<br>cumplirá con los más altos<br>estándares internacionales<br>(para computadoras y<br>tabletas). Además, el Proyecto<br>hará la transición de<br>computadoras de escritorio a<br>laptops más eficientes<br>energéticamente. Cumplir con<br>los estándares internacionales<br>y adoptar laptops para mejoras<br>tecnológicas a nivel de ULE se<br>espera que genere ahorros de<br>energía de al menos el 20%.<br>|\n|**Subcomponente 1.2. Mejorar la experiencia de los hogares con el PGH**|**Subcomponent\n\nMention: \"ubicación de hogares\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Utilisees avec les cartes de dangers climatiques.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:003955:52:3", "occurrence_id": "18e2ff7ff52b90d542028522", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  (MtCO2/year) (Text)  ** <br> <br>|<br>**Net GHG emissions reductions per year (MtCO2/year) (Text)  ** <br> <br>|\n|<br>Description <br>|<br>This indicator will measure the tons of relevant greenhouse gas reduced compared to alternatives. <br>|\n|<br>Frequency|<br>Annual <br>|\n|<br>Data source <br>|<br>Updated project economic analysis to estimate the emissions from combination of fossil-fuel based power generations<br>and from the Rogun HPP as per supply mix simulations under “without Rogun project” scenario on least-cost generation<br>expansion plans and simulation of power system electricity generation/dispatch. Climate Watch Historical GHG<br>Emissions, World Development Indicators|\n|Methodology for Data<br>Collection <br>|<br>The annual Rogun HPP annual generation data. <br>|\n|<br>Responsibility for Data|Rogun PMG|\n\n\n\nPage 32\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Citees avec les emissions historiques pour la production d'electricite.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:003955:52:3", "occurrence_id": "110eec24ac7eb92f28948526", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  (MtCO2/year) (Text)  ** <br> <br>|<br>**Net GHG emissions reductions per year (MtCO2/year) (Text)  ** <br> <br>|\n|<br>Description <br>|<br>This indicator will measure the tons of relevant greenhouse gas reduced compared to alternatives. <br>|\n|<br>Frequency|<br>Annual <br>|\n|<br>Data source <br>|<br>Updated project economic analysis to estimate the emissions from combination of fossil-fuel based power generations<br>and from the Rogun HPP as per supply mix simulations under “without Rogun project” scenario on least-cost generation<br>expansion plans and simulation of power system electricity generation/dispatch. Climate Watch Historical GHG<br>Emissions, World Development Indicators|\n|Methodology for Data<br>Collection <br>|<br>The annual Rogun HPP annual generation data. <br>|\n|<br>Responsibility for Data|Rogun PMG|\n\n\n\nPage 32\n\nMention: \"Rogun HPP annual generation data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Donnees annuelles de production de la centrale.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:004507:48:3", "occurrence_id": "53904ae746ed139cdb28070a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  the sum of all woman-months worked divided by sum of all person-months worked<br>|\n|Frequency<br>|Annual<br>|\n|Data source <br>|Civil works contractor timesheets|\n|<br>Methodology for<br>Data Collection|<br>DSC will provide spreadsheet-based analysis of timesheets|\n\n\n\nPage 36\n\nMention: \"Civil works contractor timesheets\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Somme des personnes-mois travaillées pour l'indicateur annuel.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:004547:40:3", "occurrence_id": "b0c57828482bbceed74a6f45", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  using digitally enabled services (new services) - Climate-Vulnerable (Number of people)  ** <br>|**People using digitally enabled services (new services) - Climate-Vulnerable (Number of people)  ** <br>|\n|<br>Description<br>|<br>This indicator measures the total number of climate-vulnerable people using digitally enabled (new) services supported<br>by the project.<br>|\n|Frequency|<br>Annually<br>|\n|<br>Data source<br>|<br>Sectoral ministries, national climate vulnerability and risk maps, geospatial data, multilateral and development partner<br>assessments<br>|\n|Methodology for Data<br>Collection|<br>Climate-vulnerable people will be identified by linking registrants’ residential information to national vulnerability maps or<br>climate risk profiles.|\n\n\n\nPage 31\n\nMention: \"national climate vulnerability and risk maps\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Sources annuelles des ministères sectoriels pour le projet.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:004547:40:3", "occurrence_id": "cb25d7a62a570155ee73b1a0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  using digitally enabled services (new services) - Climate-Vulnerable (Number of people)  ** <br>|**People using digitally enabled services (new services) - Climate-Vulnerable (Number of people)  ** <br>|\n|<br>Description<br>|<br>This indicator measures the total number of climate-vulnerable people using digitally enabled (new) services supported<br>by the project.<br>|\n|Frequency|<br>Annually<br>|\n|<br>Data source<br>|<br>Sectoral ministries, national climate vulnerability and risk maps, geospatial data, multilateral and development partner<br>assessments<br>|\n|Methodology for Data<br>Collection|<br>Climate-vulnerable people will be identified by linking registrants’ residential information to national vulnerability maps or<br>climate risk profiles.|\n\n\n\nPage 31\n\nMention: \"geospatial data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Données géospatiales citées avec les évaluations des partenaires.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:004547:40:3", "occurrence_id": "d1b42bad7d335839c20539a4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  using digitally enabled services (new services) - Climate-Vulnerable (Number of people)  ** <br>|**People using digitally enabled services (new services) - Climate-Vulnerable (Number of people)  ** <br>|\n|<br>Description<br>|<br>This indicator measures the total number of climate-vulnerable people using digitally enabled (new) services supported<br>by the project.<br>|\n|Frequency|<br>Annually<br>|\n|<br>Data source<br>|<br>Sectoral ministries, national climate vulnerability and risk maps, geospatial data, multilateral and development partner<br>assessments<br>|\n|Methodology for Data<br>Collection|<br>Climate-vulnerable people will be identified by linking registrants’ residential information to national vulnerability maps or<br>climate risk profiles.|\n\n\n\nPage 31\n\nMention: \"climate risk profiles\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Reliés aux informations résidentielles des inscrits.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:005326:44:2", "occurrence_id": "b4e700a8c3233a891b49cf73", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  the national crediting program operated, covering both public-sector and private-sector<br>subprojects supported by the project.|\n|Frequency|Annual|\n|Data source|KTB MRV and issuance records; third-party verification reports; registry accounts.|\n|Methodology for Data|Aggregate the volume of all VERs formally issued under the national carbon crediting mechanism during the project period; ensure|\n\n\nPage 32\n\nMention: \"KTB MRV and issuance records\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Agregent le volume de crédits formellement émis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:005326:44:2", "occurrence_id": "7ea08799688f4ad1dac566c1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  the national crediting program operated, covering both public-sector and private-sector<br>subprojects supported by the project.|\n|Frequency|Annual|\n|Data source|KTB MRV and issuance records; third-party verification reports; registry accounts.|\n|Methodology for Data|Aggregate the volume of all VERs formally issued under the national carbon crediting mechanism during the project period; ensure|\n\n\nPage 32\n\nMention: \"third-party verification reports\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"validate\", \"summary\": \"Vérifient indépendamment les crédits émis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:005326:44:2", "occurrence_id": "9089508384e19bceee5ea7aa", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  the national crediting program operated, covering both public-sector and private-sector<br>subprojects supported by the project.|\n|Frequency|Annual|\n|Data source|KTB MRV and issuance records; third-party verification reports; registry accounts.|\n|Methodology for Data|Aggregate the volume of all VERs formally issued under the national carbon crediting mechanism during the project period; ensure|\n\n\nPage 32\n\nMention: \"registry accounts\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Comptes de registre consignant les crédits émis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006914:7:1", "occurrence_id": "f57ebc404b866284cacf337e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  2 shows the breakdown of monthly electricity</mark>\n<mark>consumption by slab for each decile of household per capita expenditure. While there is a</mark>\n<mark>clear difference in consumption across deciles, the graph shows that the vast majority of</mark>\n<mark>households except in the top decile consume less than 300 kWh/month, and therefore</mark>\n<mark>continue to benefit from subsidies even after the 2013 reforms. However, since the total</mark>\n<mark>subsidy provided to households with moderate consumption (between around 150 and 300</mark>\n<mark>kWh/month) is greater than that for households below 150 kWh/month, the bulk of</mark>\n<mark>subsidies still goes to better‐off households.</mark> One potential shortcoming of the above analysis\nis that the survey asks about total electricity consumption over the preceding month. Self‐\nreports may be inaccurate, or may include expenditure on other forms of electrical power\nsuch as batteries or uninterrupted power supplies. In order to corroborate these findings,\nwe now examine actual billing data.\n\n\nData on monthly electricity consumption, along with other household details, were obtained\nfor 325,926 households from three electricity distribution companies (DISCOs): Gujranwala\nElectric Power Company (GEPCO), Faisalabad Electric Supply Company (FESCO) and\nIslamabad Electric Supply Company (IESCO). <sup>4</sup> [^4: Other DISCOs were approached for inclusion in the analysis, but data could not be sourced with the right\nspecifications in time for this study.] These DISCOs together serve almost 38\npercent of Pakistan’s residential electricity consumers. <sup>5</sup> [^5: For a breakdown of billing data please refer to the Appendix.] The three DISCOs also cover a\nsignificant share of northern Pakistan and include a mix of semi‐urban, rural and urban\nareas. For this analysis IESCO\n\nMention: \"actual billing data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Household electricity billing records for 325,926 households corroborate survey findings on consumption.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:006921:63:0", "occurrence_id": "07cd98ed7c851794010711bb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: Table A16: Industry measure of Technology Gap across EU (2013‐14); neck‐and‐neck sectors\n\n\n**NACE** **Type**\n**Sector description** **Gap**\n**2‐digit** **sector**\n\n\n12 Tobacco products 0.089 m\n39 Remediation activities and other waste management 0.111 s\n37 Sewerage 0.124 s\n78 Employment activities 0.125 s\n75 Veterinary activities 0.129 s\n72 Scientific research and development 0.130 s\n21 Basic pharmaceuticals products and preparations 0.136 m\n80 Security and investigation activities 0.138 s\n30 Other transport equipment 0.139 m\n32 Other manufacturing 0.142 m\n51 Air transport 0.143 s\n62 Computer programming, consultancy and related 0.144 s\n24 Basic metals 0.146 m\n26 Computer, electronic and optical products 0.147 m\n19 Coke and refined petroleum products 0.148 m\n79 Travel agency, tour operator reservation activities 0.149 s\n13 Textiles 0.149 m\n27 Electrical equipment 0.150 m\n17 Paper and paper products 0.152 m\n53 Postal and courier services 0.155 s\n63 Information services activities 0.158 s\n38 Waste collection, treatment and disposal activities 0.162 s\n18 Printing and reproduction of recorded media 0.162 m\n15 Leather and related products 0.162 m\n33 Repair and installation of machinery and equipment 0.162 m\n36 Waste collection, treatment and supply 0.166 s\n28 Machinery and equipment nec 0.166 m\n31 Furniture 0.170 m\n<u>22</u> <u>Rubber and plastic products</u> <u>0.171</u> <u>m</u>\nSource: Amadeus; authors’ calculations. The Gap stands for Technology gap as defined in Aghion _et al_ . (2005) and\nthe distinction between Neck‐and‐neck industries and Un\n\nMention: \"Amadeus\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Firm-level database behind the technology-gap and industry productivity estimates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007472:39:0", "occurrence_id": "25e119e6ed73736ee02ff717", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Figure 4: Unconditional differences (Muslim – Christian) in average nutritional status, by**\n**marital status**\n\na) Rural log BMI b) Urban log BMI\n\n\n\n\n\n\n\n\n\n\n\n\n\na) Rural share of underweight b) Urban share of underweight\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nNote: The figure gives average predicted log BMI (upper panels) and underweight (bottom panels) differences\nbetween Muslims and Christians by marital status based on a regression of log BMI on a dummy variable for\nMuslim, dummies for marital status, and the interactions between Muslim and each marital status. These differences\nare evaluated at the sample mean of all the independent variables included in the regression. The zero line identifies\nno difference between Muslim and Christian women. 95% confidence intervals are shown around the difference in\nmean nutritional status for each marital status.\nSource: Authors calculations using the Nigeria DHSs of 2008 and 2013.\n\n\n38\n\nMention: \"Nigeria DHSs of 2008 and 2013\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Two survey rounds used to estimate nutritional status differences by marital status.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007478:18:1", "occurrence_id": "d26b9b4a4ecc416f696da399", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .\nWhat we do observe in Afghanistan is a seasonal gap between food CPI and NPI. Thus, vegetables\nseem to be driving the difference between NPI and the food CPI, since the price of vegetables is\nseasonal but has not increased over time. <sup>26</sup> [^26: We also caution that the Afghanistan results are not directly comparable to the Pakistan and India results, because, due to data\nlimitations, the Afghanistan NPI is a hybrid between the NPI and the food CPI.] In India, prices have increased the fastest for vegetables,\nfruits, seafood, and meat products, and slowest for beverages, sugar, and oils and fats. Not only\nhave vegetable and pulse prices been increasing the fastest, these prices have also been the most\nseasonal. Thus, the more subdued gap between NPI and food CPI in India is not so much due to a\nweak link between vegetable prices and seasonality, but rather due to the smaller difference in the\nnutritional and expenditure weights for vegetables and pulses in India, relative to those for Pakistan\nand Afghanistan.\n\n\n**5** **Extension to CoRD: Using household survey data in Sri Lanka and Pakistan**\n\nThis main proof of concept in this paper is to demonstrate that data from existing price monitoring\nsystems can easily be used to measure and understand the cost of nutritious food across time and\ngeographic locations. In calculating CoRD we had used food prices collected by the NSO in\nselected markets (cities) to calculate the CPI. The main concern, however, is that these prices might\nnot be a good representation of the actual prices experienced by certain consumers, especially\nthose living in remote rural areas. An alternative to CPI data would be to use food prices from\nhousehold surveys (deduced from the unit values in a consumption module). Household surveys\nare less ideal for real time price monitoring as they are administered less frequently (for example,\nevery five years, as opposed to monthly for the CPI). But they are\n\nMention: \"food prices collected by the NSO\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Market food prices used to compute the CPI and the cost of a nutritious diet.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007478:18:1", "occurrence_id": "3d6882c8c3930b5249a1f96a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: .\nWhat we do observe in Afghanistan is a seasonal gap between food CPI and NPI. Thus, vegetables\nseem to be driving the difference between NPI and the food CPI, since the price of vegetables is\nseasonal but has not increased over time. <sup>26</sup> [^26: We also caution that the Afghanistan results are not directly comparable to the Pakistan and India results, because, due to data\nlimitations, the Afghanistan NPI is a hybrid between the NPI and the food CPI.] In India, prices have increased the fastest for vegetables,\nfruits, seafood, and meat products, and slowest for beverages, sugar, and oils and fats. Not only\nhave vegetable and pulse prices been increasing the fastest, these prices have also been the most\nseasonal. Thus, the more subdued gap between NPI and food CPI in India is not so much due to a\nweak link between vegetable prices and seasonality, but rather due to the smaller difference in the\nnutritional and expenditure weights for vegetables and pulses in India, relative to those for Pakistan\nand Afghanistan.\n\n\n**5** **Extension to CoRD: Using household survey data in Sri Lanka and Pakistan**\n\nThis main proof of concept in this paper is to demonstrate that data from existing price monitoring\nsystems can easily be used to measure and understand the cost of nutritious food across time and\ngeographic locations. In calculating CoRD we had used food prices collected by the NSO in\nselected markets (cities) to calculate the CPI. The main concern, however, is that these prices might\nnot be a good representation of the actual prices experienced by certain consumers, especially\nthose living in remote rural areas. An alternative to CPI data would be to use food prices from\nhousehold surveys (deduced from the unit values in a consumption module). Household surveys\nare less ideal for real time price monitoring as they are administered less frequently (for example,\nevery five years, as opposed to monthly for the CPI). But they are\n\nMention: \"CPI data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Consumer price index discussed as a possibly unrepresentative price source for rural consumers.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000009:6:0", "occurrence_id": "391fdea7703710ed5ccb2343", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: success of the grant to be independently monitored from the ERPA.\n\nThe M&E system would be implemented by ORCU as the OFLP coordination unit, which is hosted\nby OFWE. The ORCU coordination unit would be staffed with an M&E specialist (and support\nstaff). The OFLP M&E system would operate at regional, zone, woreda and kebele levels using\ndedicated OFLP staff working closely with existing Government staff at each level in bureaus,\nzones, woreda offices, and Development Agents (DAs) working at kebele level.\n\nData will be gathered on a semi-annual and annual basis, drawing as much as possible from existing\ndata sources and information management tools to be put in place. This includes the M&E system,\nthe Forest MIS as well as the regional MRV system. Specific templates would be designed for data\ncollection and detailed training provided to relevant staff on how these would be used.\n\nThere would be periodic implementation support missions with an M&E focus over the lifetime of\nboth the grant and the ERPA period of the Program. OFLP would also include a Mid-Term Review\napproximately 24 months after grant effectiveness to assess progress and identify areas for course\ncorrection where needed. An Implementation Completion and Results Report (ICR) will be\nconducted at the end of the grant period as per WB procedures.\n\nMonitoring, Reporting and Verification (MRV) and Reference Level (RL) for the ERPA\n\nThe results in terms of reducing deforestation and net GHG would be based on the monitoring of\nchanges in forest cover (activity data) and associated GHG emissions (based on emissions factors\nfor the different types of forests within the regional state) compared to a business-as-usual baseline\nor Reference Level (see Annex 7 for details).\n\nThe Reference Level for OFLP and monitoring would rely on and be linked to the national\nReference Level and MRV system that is currently being developed as part\n\nMention: \"activity data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Monitored changes in forest cover underpin the ERPA deforestation and GHG results.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000012:11:2", "occurrence_id": "9fb4e109ef82088c63261a79", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  security and rural livelihoods. <sup>6</sup> For instance,\nagricultural growth rates have fluctuated with changes in rainfall due to climate events like El Niño and La Niña, causing\na 21.2 percent contraction in 2018. By 2050, temperatures in Zambia are projected to rise to 2.2°C, with the greatest\nincreases in the southern and eastern crop production areas. <sup>7</sup> [^7: CIAT; World Bank. 2017. Climate Smart Agriculture in Zambia. CSA Country Profiles for Africa Series] Rainfall is expected to increase in the northern parts\n\n\n1 The country protected areas include: 20 national parks, 39 game management areas, 432 forest reserves, 59 botanical reserves, 42 important bird\nareas, and 2 bird sanctuaries.\n2 ZamStats 2023. 2022 Census of population and housing. Preliminary report.\n3 ZamStats 2022. Annual labor force survey report.\n4 The World Bank 2021. COVID-19, Poverty, and Social Safety Net Response in Zambia. Policy Research Working Paper No. 9571. World Bank,\n[Washington, DC. http://hdl.handle.net/10986/35249](http://hdl.handle.net/10986/35249)\n5 Official external creditors and the authorities agreed in October 2023 on a Memorandum of Understanding outlining treatment covering around\nUS$6.3 billion of official debt. Zambia is seeking comparable treatment with its Eurobond holders (US$3.5 billion) and other external commercial\nlenders (US$3.2 billion).\n6 https://www.adaptationundp.org/explore/africa/zambia#:~:text=Zambia%20has%20been%20experiencing%20adverse,changes%20in%20the%20growing%20season. Also,\nThe World Bank. April 2017. Project Appraisal Document for A Zambia Integrated Forest Landscape Project.\n7 CIAT; World Bank. 2017. Climate Smart Agriculture in Zambia. CSA Country Profiles\n\nMention: \"2022 Census of population and housing\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"Cited preliminary census report behind population and settlement figures.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000012:11:2", "occurrence_id": "cc8464022bde2b1800ca34fc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  security and rural livelihoods. <sup>6</sup> For instance,\nagricultural growth rates have fluctuated with changes in rainfall due to climate events like El Niño and La Niña, causing\na 21.2 percent contraction in 2018. By 2050, temperatures in Zambia are projected to rise to 2.2°C, with the greatest\nincreases in the southern and eastern crop production areas. <sup>7</sup> [^7: CIAT; World Bank. 2017. Climate Smart Agriculture in Zambia. CSA Country Profiles for Africa Series] Rainfall is expected to increase in the northern parts\n\n\n1 The country protected areas include: 20 national parks, 39 game management areas, 432 forest reserves, 59 botanical reserves, 42 important bird\nareas, and 2 bird sanctuaries.\n2 ZamStats 2023. 2022 Census of population and housing. Preliminary report.\n3 ZamStats 2022. Annual labor force survey report.\n4 The World Bank 2021. COVID-19, Poverty, and Social Safety Net Response in Zambia. Policy Research Working Paper No. 9571. World Bank,\n[Washington, DC. http://hdl.handle.net/10986/35249](http://hdl.handle.net/10986/35249)\n5 Official external creditors and the authorities agreed in October 2023 on a Memorandum of Understanding outlining treatment covering around\nUS$6.3 billion of official debt. Zambia is seeking comparable treatment with its Eurobond holders (US$3.5 billion) and other external commercial\nlenders (US$3.2 billion).\n6 https://www.adaptationundp.org/explore/africa/zambia#:~:text=Zambia%20has%20been%20experiencing%20adverse,changes%20in%20the%20growing%20season. Also,\nThe World Bank. April 2017. Project Appraisal Document for A Zambia Integrated Forest Landscape Project.\n7 CIAT; World Bank. 2017. Climate Smart Agriculture in Zambia. CSA Country Profiles\n\nMention: \"Annual labor force survey report\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"Cited annual labour force survey behind employment figures.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000012:13:0", "occurrence_id": "a0a87a9397b51a2eb3282706", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: hectares of forest, or roughly six percent of its total tree cover, resulting in the release of an estimated 252 MtCO2e\ninto the atmosphere. The land use change and forestry sector alone accounted for 61 percent of Zambia's national\nGHG emissions in 2011.\n\n7. **The Eastern Province, a region plagued by land degradation and poverty, has been selected to pilot the**\n**implementation of climate change mitigation and low carbon development.** The Eastern Province includes the\nresource-rich and biodiverse Luangwa Valley and some of Zambia's most renowned national parks, such as the\nlucrative South Luangwa National Park, offering vast potential for the development of sustainable, natural resourcebased livelihoods, particularly in tourism. The forested landscapes in Eastern Province play a crucial role in driving\neconomic growth, creating jobs, providing clean water and essential building materials, and supplying energy to\nvarious sectors. However, the forests are facing numerous challenges, including illegal encroachment into protected\nareas, habitat fragmentation, and the effects of deforestation and uncontrolled charcoal production. Together, these\nfactors contribute to biodiversity loss and elevated GHG emissions. The Eastern Province is one of the poorest regions\nin Zambia, with 70 percent of its 2.4 million residents living in poverty and 48 percent lacking adequate food access\nthroughout the year. <sup>12,13</sup> [^12: Indaba Agricultural Policy Research Institute, IAPRI (2019). Rural Agricultural Livelihoods Survey Report.] [^13: ZamStats 2022. Zambia 2022 Census of Population and Housing] These rural communities are heavily reliant on agriculture and natural resources for their\nlivelihoods, perpetuating a cycle of poverty and resource degradation. To break this cycle, it is imperative to address\nthe underlying causes of forest loss and degradation and promote sustainable land management practices. The\nGovernment of the Republic of Zambia (GRZ) is committed to transforming Eastern Province into a comprehensive\nrural development model that combines forest protection and the promotion of sustainable agricultural practices. To\nachieve this goal, the Eastern Province\n\nMention: \"Rural Agricultural Livelihoods Survey Report\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"Footnote source for rural food access and livelihood shares.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000012:13:0", "occurrence_id": "2c1f9680a1e7e2860b9c167f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: hectares of forest, or roughly six percent of its total tree cover, resulting in the release of an estimated 252 MtCO2e\ninto the atmosphere. The land use change and forestry sector alone accounted for 61 percent of Zambia's national\nGHG emissions in 2011.\n\n7. **The Eastern Province, a region plagued by land degradation and poverty, has been selected to pilot the**\n**implementation of climate change mitigation and low carbon development.** The Eastern Province includes the\nresource-rich and biodiverse Luangwa Valley and some of Zambia's most renowned national parks, such as the\nlucrative South Luangwa National Park, offering vast potential for the development of sustainable, natural resourcebased livelihoods, particularly in tourism. The forested landscapes in Eastern Province play a crucial role in driving\neconomic growth, creating jobs, providing clean water and essential building materials, and supplying energy to\nvarious sectors. However, the forests are facing numerous challenges, including illegal encroachment into protected\nareas, habitat fragmentation, and the effects of deforestation and uncontrolled charcoal production. Together, these\nfactors contribute to biodiversity loss and elevated GHG emissions. The Eastern Province is one of the poorest regions\nin Zambia, with 70 percent of its 2.4 million residents living in poverty and 48 percent lacking adequate food access\nthroughout the year. <sup>12,13</sup> [^12: Indaba Agricultural Policy Research Institute, IAPRI (2019). Rural Agricultural Livelihoods Survey Report.] [^13: ZamStats 2022. Zambia 2022 Census of Population and Housing] These rural communities are heavily reliant on agriculture and natural resources for their\nlivelihoods, perpetuating a cycle of poverty and resource degradation. To break this cycle, it is imperative to address\nthe underlying causes of forest loss and degradation and promote sustainable land management practices. The\nGovernment of the Republic of Zambia (GRZ) is committed to transforming Eastern Province into a comprehensive\nrural development model that combines forest protection and the promotion of sustainable agricultural practices. To\nachieve this goal, the Eastern Province\n\nMention: \"ZamStats 2022\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"Producer and year citation for the national census reference.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000012:13:0", "occurrence_id": "71f501a8217ba569e12f7f91", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: hectares of forest, or roughly six percent of its total tree cover, resulting in the release of an estimated 252 MtCO2e\ninto the atmosphere. The land use change and forestry sector alone accounted for 61 percent of Zambia's national\nGHG emissions in 2011.\n\n7. **The Eastern Province, a region plagued by land degradation and poverty, has been selected to pilot the**\n**implementation of climate change mitigation and low carbon development.** The Eastern Province includes the\nresource-rich and biodiverse Luangwa Valley and some of Zambia's most renowned national parks, such as the\nlucrative South Luangwa National Park, offering vast potential for the development of sustainable, natural resourcebased livelihoods, particularly in tourism. The forested landscapes in Eastern Province play a crucial role in driving\neconomic growth, creating jobs, providing clean water and essential building materials, and supplying energy to\nvarious sectors. However, the forests are facing numerous challenges, including illegal encroachment into protected\nareas, habitat fragmentation, and the effects of deforestation and uncontrolled charcoal production. Together, these\nfactors contribute to biodiversity loss and elevated GHG emissions. The Eastern Province is one of the poorest regions\nin Zambia, with 70 percent of its 2.4 million residents living in poverty and 48 percent lacking adequate food access\nthroughout the year. <sup>12,13</sup> [^12: Indaba Agricultural Policy Research Institute, IAPRI (2019). Rural Agricultural Livelihoods Survey Report.] [^13: ZamStats 2022. Zambia 2022 Census of Population and Housing] These rural communities are heavily reliant on agriculture and natural resources for their\nlivelihoods, perpetuating a cycle of poverty and resource degradation. To break this cycle, it is imperative to address\nthe underlying causes of forest loss and degradation and promote sustainable land management practices. The\nGovernment of the Republic of Zambia (GRZ) is committed to transforming Eastern Province into a comprehensive\nrural development model that combines forest protection and the promotion of sustainable agricultural practices. To\nachieve this goal, the Eastern Province\n\nMention: \"Zambia 2022 Census of Population and Housing\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"National census cited for rural population and livelihood context.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000012:38:0", "occurrence_id": "bc984fa2850197b38360b9c6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nEASTERN PROVINCE JURISDICTIONAL SUSTAINABLE LANDSCAPE PROGRAM – EMISSIONS REDUCTION PROJECT\n(P155827)\n\n|Volume of not-for-profit|finance (public or private) leveraged to contribute to ISFL objectives (Amount(USD))|\n|---|---|\n|Description|The indicator measures the dollar amount not-for profit financing (public or private) leveraged to contribute to ISFL<br>objectives|\n|Frequency|Reporting periods as set in Emissions Reduction Purchase Agreement|\n|Data source|Project Progress Report|\n|Methodology for Data<br>Collection|Review of Project Progress Report|\n|Responsibility for Data<br>Collection|PIU|\n\n\n\n28\n\nMention: \"Project Progress Report\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Listed data source reviewed for the leveraged finance indicator.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000013:11:2", "occurrence_id": "d18d7165babd99afd61e331b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: s high levels of education and labor market participation. Both the Global Gender Gap Index and the\nGender Equity Index show that Uzbekistan is close to attaining gender equality in education (enrollment, educational\nattainment, and literacy) and health (sex ratio at birth and healthy life expectancy). According to the 2019 L2CU, 95\npercent of citizens believed that women have equal access to education, employment, and healthcare. Nevertheless,\ngender disparities persist with one manifestation being that women tend to be employed in the social sector or in parttime, seasonal, low-paying, or unskilled jobs in the formal and informal sectors <sup>10</sup> .\n\n\n7 WB report. 2019. _China/Russia 2030 – Opportunities for Central Asia’s Agriculture_ presents challenges and opportunities for realizing horticulture\nexport potential in the countries of Central Asia, including Uzbekistan.\n\n8 Sutton, W. _et al_ . 2013. _Reducing the Vulnerability of Uzbekistan’s Agricultural Systems to Climate Change: Impact Assessment and Adaptation_\n_Options_ . Washington, DC: World Bank.\n\n9 WB background paper. 2019. _Enhancing the Livelihoods of Rural Women in Uzbekistan_ .\n\n10 Asian Development Bank. 2018. _Uzbekistan Country Gender Assessment Update_ .\n\n\nPage 8 of 70\n\nMention: \"2019 L2CU\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Living conditions survey reporting perceived equal access to education and health services.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000014:7:0", "occurrence_id": "b509343e6bc56fbdc09ffb9d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nAgriculture Modernization and Competitiveness Project (P158372)\n\n\nidentify and promote high-productivity fodder crops suitable for production on both irrigated and dry lands to inform\ndecision of farmers on selecting appropriate varieties and techniques suitable for various agro-ecological zones.\n\n\n24 **Good agricultural practices** : Uzbekistan does not have a code of Good Agricultural Practices (GAP) for crop and\nlivestock production, which is common in other countries to promote climate-smart agricultural practices. In many\ncountries the adoption of GAP is a precondition for farming or for receiving subsidies. It is also used as a precondition for\nadopting stricter Global GAPs and other standards such as organic and halal. There are many locally-tested best farming\npractices in Uzbekistan, but they are not put together in one coherent place and they are not widely disseminated and\nadopted. The project could support the preparation of the GAP code and guidelines using the best practice local and global\nknowledge; introduce pilot training courses and adjust draft code and guidelines based on experience/feedback; and\npublish final version of the code and guidelines and promote it widely through training programs for extension staff and\nfarmers.\n\n\n25 The project could also support investments to update soil fertility maps, which is needed for GAP adoption. Such\ninvestments include upgrading soil testing laboratory infrastructure, digitalize map information with open-date access,\ndeveloping best practice guidelines on soil monitoring and fertilizer use adjusted to different soil types and agro-ecological\nzones, making this information available to farmers through field-based advisers and extension services, and procuring\nportable field-based soil testing equipment to allow extension services and other advisers to provide immediate feedback\nto farmers on soil quality and proposed adjustments needed to enhance soil fertility.\n\n\n26 **Institutional strengthening** : Engagement will be required with the Ministry of Innovations, Ministry of Agriculture,\nother agencies/institutes and private sector (farmers and agribusinesses) to develop mechanisms to bring research and\nprivate sector together for better defining\n\nMention: \"soil fertility maps\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Mapped soil fertility needed for good agriculture practice adoption.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000016:11:0", "occurrence_id": "70f50a56cb2f2f2c70fa14a6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nBalochistan Livelihoods and Entrepreneurship Project (P159292)\n\n\n\n\n\n\n\n10. **A ‘double burden’ on resources has affected development outcomes for refugees and host communities in**\n**Balochistan.** Even-though the country has witnessed average growth rates of four percent in the past decade, these have\nnot led to commensurate growth in the province for four main reasons. First, population density in the province is low,\nat 38 persons per square kilometer which falls to 15 in the rural areas. Given the scattered nature of settlements in\nBalochistan, this implies negligible and cost-ineffective service delivery. Second, resource mobilization at the grassroots\nlevel is particularly constrained and this is exacerbated by, amongst others, lack of skills and training, low awareness\nregarding market opportunities, limited access to finance and high cost of doing business <sup>6</sup> [^6: According to the Sub National Doing Business (SNDB) Report 2010, the capital of Balochistan, Quetta is ranked the lowest across 13\nsurveyed districts and cities.] . Resource mobilization capacity\nof the province is restricted given the low tax base. Third, like most traditional and rural societies, even in Balochistan\nthere is a dominance of consumption expenditure and the emphasis on savings for investment purposes is limited.\nFourth, low capacity and a dearth of qualified human resources in Balochistan hinders service delivery to the poor.\n\n\n11. **Agriculture is the backbone of the economy of Pakistan and Balochistan.** A fifth of the national economy is\nagriculture and accounts for nearly half of the jobs. In recent years more jobs have been created in non-farm sectors like\nmanufacturing and service **sectors** than in agriculture. Agriculture remains the main source of income for 64 percent of\nthe households in the province with 43.4 percent employed in the sector in 2017. Female labor force participation rate\nin Balochistan of 11.3 percent in 2014-2015, is less than half of the national average. Climate change exacerbates extreme\nweather events\n\nMention: \"Sub National Doing Business (SNDB) Report 2010\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"Sub-national ranking cited for the cost of doing business across districts.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000016:22:1", "occurrence_id": "a4a156a502357d19a68d72e7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Plans (RAP) will be developed, as sub-project locations become known, in accordance with the TORs approved by\nthe World Bank.\n\n\n**D. Gender**\n\n\n49. **A gender analysis has been completed and gender action plan has been incorporated in Project design.** The\nresults framework for the Project is fully disaggregated by gender. The Project involves supporting entrepreneurs at the\ncommunity level and data reveals that women are engaged in livelihoods activities (such as handicrafts, kitchen gardening\netc.). Through the LIPs in particular and through training and enterprise development opportunities, the Project will seek\nto elicit this participation and support investment and skills development initiatives for women, thereby improving their\nearnings. And as mentioned above, women voices will be made a part of the overall citizens engagement agenda.\n\n\n**E. Climate**\n\n\n50. **Climate Considerations:** The Project has been screened for climate risk using the WBG’s Climate and Disaster Risk\nScreening Project Level Tool <sup>16</sup> [^16: The tool is available here: _[http://climatescreeningtools.worldbank.org/](http://climatescreeningtools.worldbank.org/)_] . Based on the review of historic and projected data at Project location, risks and potential\nimpacts of exposure to climatic trends and geophysical hazards are rated high for extreme precipitation, flash flooding,\ndrought and earthquake. Climate and geophysical hazards are assessed as Moderate for each operational component for\nthe Project lifetime and beyond. Risk mitigation measures have been incorporated in the project design and include\namong others the promotion of resilience-enhancing technologies (e.g. use of green technologies such as solar, in energy\nand water management; use of climate-resilient materials and design in the construction of transport and storage\ninfrastructures and value-adding facilities. It is expected that increased income of beneficiaries through project\nintervention will increase their resilience to climate-related impacts.\n\n\n51. **GHG Accounting:** The Project used the EX-ACT tool measure\n\nMention: \"historic and projected data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Historic and projected climate data screened at the project location for hazard ratings.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000017:2:1", "occurrence_id": "fa8c3c11ad654903510f7885", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  warrants, tracking of completion certificates against\nwarrants for capital projects, processing requisitions and quotations, printing of payment vouchers\nand other supporting documents, processing payment adjustments, bank reconciliation, cash flow\nand procurement plans.\nc. There is a general lack of reporting capability and where they exist, they are in the form of\nprinted reports which are difficult to use for further analysis.\nd. End users lack the facility to review information online or to generate their own reports.\ne. There are many instances of batch processing of data as opposed to online real time\nprocessing common in most modern systems.\nf. There is a lack of a clear path or ability to upgrade current applications for missing\nfunctionalities.\ng. System documentation is limited and scanty.\nh. The current system has a text based user interface as opposed to a graphical interface,\nwhich affects end user productivity.\ni. Chart of accounts needed to be reviewed to facilitate other management, monitoring and\nreporting requirements such as MTEF, performance budgeting and reporting.\nj. Routine processes such as bank reconciliation are not automated leading to creation of huge\nbacklogs for unreconciled accounts.\nk. Current system uses batch processing as opposed on online real-time processing used by\nmost modern system. If one transaction is not complete in a batch then the all transactions are\nrejected.\nl. Management of budget data is centralized, with bulk of the work being done by Budget\nDepartment at Ministry of Finance.\n\n\nPage 3 of 8\n\nMention: \"budget data\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"Budget data management described as centralized in the finance ministry.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000017:5:0", "occurrence_id": "ede2b8168a0a5047b251c1cc", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: account for nearly half of the total number of civil servants, and there are also in the process of\nundertaking their own reform initiatives linked to improving frontline service delivery.\n\n\n**II.** **Project Development Objective(s)**\n\n**Proposed Development Objective(s)**\n15. The development objective is to improve financial control, and the quality and timeliness of\npublic financial management information.\n\n\n**Key Results**\n16. Achievement of the development objective will be measured through the – (i) the\nimprovement in commitment control over the budget; (ii) the provision of timely and reliable inyear budget execution reports; and (iii) the provision of timely and reliable annual financial\nstatements. In this regard an updated PEFA PFM Performance Measurement Report will be used as\na source of verification. A results framework will be developed in the project document supporting\nthe financing agreement with the Swaziland Government.\n\n\n**III.** **Preliminary Description**\n\n**Concept Description**\n17. The project will support the procurement, configuration and implementation of an off-theshelf financial management system for all budget votes funded through the Government’s\nConsolidated Fund. In broad terms, it has been agreed that the two sources of funding will be used\nas follows:\na. Government has confirmed in September 2015 that it will provide the necessary funding for\nthe purchase of hardware and software in the 2016/17 MTEF. This is estimated to be USD5.00\nmillion.\nb. The EC funding of USD3.64m (EUR3.34m) would be channeled through a “Recipient\nExecuted Trust Fund” at the World Bank, and fund the technical assistance required for – (i) the\ndesign and implementation of the new PFM business processes, controls and information and\ntechnical architectures; (ii) direct costs associated with the delivery of the technical assistance; and\n(iii) project management.\n\n18\n\nMention: \"PEFA PFM Performance Measurement Report\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Updated assessment used as a source of verification for PFM reforms.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000020:14:1", "occurrence_id": "f624ac3988c146f2908978b2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  people annually, which puts\npressure on the economy, natural resources, infrastructure including digital, urban planning, and job creation.\n\n\n8. **Tanzania’s cost of adaptation to current climate impacts is estimated at approximately US$500 million annually**\n**and, by 2030, could increase to US$1 billion a year if no action is taken.** <sup>2</sup> Individual annual events have economic costs\nin excess of 1 percent of GDP. Recognizing its vulnerability to climate impacts, the GoT adopted the National Climate\nChange Strategy in 2012, a plan that identifies climate change risks for 18 sectors and 12 cross-cutting areas, and it\nproposes over 200 strategic interventions to mitigate risks. However, the implementation of climate-resilient\napproaches and its integration in projects has been insufficient. Currently, there is a lack of action plans to address\nclimate change for most sectors, and the country is already experiencing the problems related to inaction. As an\nexample, important historical paper archives of birth registration are located close to the coast in Dar-es-Salaam and are\nat risk from rising sea levels. Digitization of these paper archives will help to mitigate this risk.\n\n\n9. **The Tanzania social context in rural settings is less favorable to women.** It is estimated that the losses in human\ncapital wealth that can be attributed to gender inequality in Tanzania are up to US$111 billion. <sup>3</sup> Agriculture employs\n56 percent of the total women labor force (76 percent in rural areas), but approximately 64 percent of them do not\nreceive any kind of payment <sup>4</sup> . Furthermore, data show the prevalence of different forms of violence against women in\nTanzania. The 2015–2016 Tanzania Demographic and Health Survey (DHS) and Malaria Indicator Survey (2015–2016)\nreveals that 39.5 percent\n\nMention: \"historical paper archives of birth registration\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Coastal paper civil-registration archives exposed to rising sea levels.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000020:14:1", "occurrence_id": "410683ce294a141c350520bd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  people annually, which puts\npressure on the economy, natural resources, infrastructure including digital, urban planning, and job creation.\n\n\n8. **Tanzania’s cost of adaptation to current climate impacts is estimated at approximately US$500 million annually**\n**and, by 2030, could increase to US$1 billion a year if no action is taken.** <sup>2</sup> Individual annual events have economic costs\nin excess of 1 percent of GDP. Recognizing its vulnerability to climate impacts, the GoT adopted the National Climate\nChange Strategy in 2012, a plan that identifies climate change risks for 18 sectors and 12 cross-cutting areas, and it\nproposes over 200 strategic interventions to mitigate risks. However, the implementation of climate-resilient\napproaches and its integration in projects has been insufficient. Currently, there is a lack of action plans to address\nclimate change for most sectors, and the country is already experiencing the problems related to inaction. As an\nexample, important historical paper archives of birth registration are located close to the coast in Dar-es-Salaam and are\nat risk from rising sea levels. Digitization of these paper archives will help to mitigate this risk.\n\n\n9. **The Tanzania social context in rural settings is less favorable to women.** It is estimated that the losses in human\ncapital wealth that can be attributed to gender inequality in Tanzania are up to US$111 billion. <sup>3</sup> Agriculture employs\n56 percent of the total women labor force (76 percent in rural areas), but approximately 64 percent of them do not\nreceive any kind of payment <sup>4</sup> . Furthermore, data show the prevalence of different forms of violence against women in\nTanzania. The 2015–2016 Tanzania Demographic and Health Survey (DHS) and Malaria Indicator Survey (2015–2016)\nreveals that 39.5 percent\n\nMention: \"Tanzania Demographic and Health Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Survey estimates on violence against women prevalence.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000020:14:1", "occurrence_id": "354e31921e46ee8d37c81d64", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  people annually, which puts\npressure on the economy, natural resources, infrastructure including digital, urban planning, and job creation.\n\n\n8. **Tanzania’s cost of adaptation to current climate impacts is estimated at approximately US$500 million annually**\n**and, by 2030, could increase to US$1 billion a year if no action is taken.** <sup>2</sup> Individual annual events have economic costs\nin excess of 1 percent of GDP. Recognizing its vulnerability to climate impacts, the GoT adopted the National Climate\nChange Strategy in 2012, a plan that identifies climate change risks for 18 sectors and 12 cross-cutting areas, and it\nproposes over 200 strategic interventions to mitigate risks. However, the implementation of climate-resilient\napproaches and its integration in projects has been insufficient. Currently, there is a lack of action plans to address\nclimate change for most sectors, and the country is already experiencing the problems related to inaction. As an\nexample, important historical paper archives of birth registration are located close to the coast in Dar-es-Salaam and are\nat risk from rising sea levels. Digitization of these paper archives will help to mitigate this risk.\n\n\n9. **The Tanzania social context in rural settings is less favorable to women.** It is estimated that the losses in human\ncapital wealth that can be attributed to gender inequality in Tanzania are up to US$111 billion. <sup>3</sup> Agriculture employs\n56 percent of the total women labor force (76 percent in rural areas), but approximately 64 percent of them do not\nreceive any kind of payment <sup>4</sup> . Furthermore, data show the prevalence of different forms of violence against women in\nTanzania. The 2015–2016 Tanzania Demographic and Health Survey (DHS) and Malaria Indicator Survey (2015–2016)\nreveals that 39.5 percent\n\nMention: \"Malaria Indicator Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Companion survey round cited with the DHS for the same estimate.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000020:21:0", "occurrence_id": "55c3b392360e0b556c353ba2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nDigital Tanzania Project (P160766)\n\n\nto the implementation of other components, notably sub-component 2.2. on rural connectivity. Activities to be funded\nunder this sub-component will include consultants and the costs associated with stakeholder consultations.\n\n\n**_1.2._** **_Infrastructure to support National ICT Development and e-Commerce (US$15.3 million equivalent)_**\n\n_a)_ _Enhancing the national addressing and postcode system (US$12.0 million equivalent)_\n\n28. This is an ongoing activity which will scale up the pilot projects for the National Spatial Data Infrastructure (NSDI),\nalready started and managed by the TCRA, which currently covers 66 wards <sup>21</sup> . This initiative is expected to cover an\nadditional 711 wards from 37 councils, which includes capital cities, strategic areas, central business districts, and other\nhigh-revenue-generating areas. The activities under this initiative involve development of digital maps, naming of\nroads/streets and installation of signage and house number plates; data collection; enhancement/update of the national\nAddress Database, preparation/review of policies and regulations, ICT works, and awareness and capacity building. The\nNSDI, or digitized map with multiple different layers, will provide a platform on which information can be layered to\nsupport key activities such as the national digital identification (ID) and spatial maps for national development, taking\ninto account requirements for data protection. The fieldwork for the mapping would be carried out under the guidance\nof the MCIT, TCRA, and Tanzania Posts Corporation (TPC). Ground mapping and photography would be supplemented\nby commercial firms recruited competitively to assist with digital mapping using remote sensing data from satellites and\naerial photography. Activities to be funded would include supply of goods, consultancy services, and coverage of\noperational costs.\n\n\n_b)_ _Development\n\nMention: \"remote sensing data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Satellite and aerial imagery used by commercial firms for digital mapping.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000020:32:0", "occurrence_id": "c2e11b6fd51670fc095d74d2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nDigital Tanzania Project (P160766)\n\n\nthe GoT is acting as an anchor tenant in guaranteeing an income for investors that commit funds to building\nout their network in urban and rural areas. Again, the incremental additional investment to which the project\nwill contribute is difficult to estimate, as it will be the outcome of a competitive bidding process, but it is\nlikely to be substantial as the project is committing US$33m of IDA financing to the sub-component.\n\nc) **PCM could also arise under sub-component 1.1,** where the funding of the soft-centers and FabLabs are likely\nto stimulate investment from those private companies whose growth is nurtured through these initiatives.\nPrivate capital may also be invested alongside public funds in the planned National Center for ICT Professional\nDevelopment and Innovation. Finally, public investment in a national addressing and postcode system under\nsub-component 1.2 should stimulate private sector investment in eCommerce.\n\n\n**B. Results Monitoring and Evaluation Arrangements**\n\n64. A national ICT data collection survey has been undertaken in preparation of the project. This baseline survey,\nfinanced through the Digital Development Platform and using the Survey of Well-being via Instant and Frequent Tracking\n(SWIFT) methodology, was carried out by the WB in 2018-19. Subsequent surveys are expected to be carried out every\ntwo to three years (that is, twice during the life of the project and once following completion), covering over 1,400\nhouseholds in rural and urban locations and including Zanzibar. The SWIFT data diagnostic focused on key indicators\nrelated to ICT adoption, usage, and impact and was disaggregated by income group, gender, type of mobile\nphone/service, and so on. Areas targeted as part of the rural connectivity program will receive specific attention to help\ninform project design and course corrections. In addition, a _Digital Sources of Growth_ study was commissioned by the\nWB in 2019 and formed\n\nMention: \"national ICT data collection survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Baseline national ICT survey undertaken in project preparation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000020:32:0", "occurrence_id": "b1187eb19b258c8478e8d9e2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nDigital Tanzania Project (P160766)\n\n\nthe GoT is acting as an anchor tenant in guaranteeing an income for investors that commit funds to building\nout their network in urban and rural areas. Again, the incremental additional investment to which the project\nwill contribute is difficult to estimate, as it will be the outcome of a competitive bidding process, but it is\nlikely to be substantial as the project is committing US$33m of IDA financing to the sub-component.\n\nc) **PCM could also arise under sub-component 1.1,** where the funding of the soft-centers and FabLabs are likely\nto stimulate investment from those private companies whose growth is nurtured through these initiatives.\nPrivate capital may also be invested alongside public funds in the planned National Center for ICT Professional\nDevelopment and Innovation. Finally, public investment in a national addressing and postcode system under\nsub-component 1.2 should stimulate private sector investment in eCommerce.\n\n\n**B. Results Monitoring and Evaluation Arrangements**\n\n64. A national ICT data collection survey has been undertaken in preparation of the project. This baseline survey,\nfinanced through the Digital Development Platform and using the Survey of Well-being via Instant and Frequent Tracking\n(SWIFT) methodology, was carried out by the WB in 2018-19. Subsequent surveys are expected to be carried out every\ntwo to three years (that is, twice during the life of the project and once following completion), covering over 1,400\nhouseholds in rural and urban locations and including Zanzibar. The SWIFT data diagnostic focused on key indicators\nrelated to ICT adoption, usage, and impact and was disaggregated by income group, gender, type of mobile\nphone/service, and so on. Areas targeted as part of the rural connectivity program will receive specific attention to help\ninform project design and course corrections. In addition, a _Digital Sources of Growth_ study was commissioned by the\nWB in 2019 and formed\n\nMention: \"Survey of Well-being\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Baseline methodology carrying out the ICT well-being survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000020:32:0", "occurrence_id": "4988196b71dae7805567d22c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nDigital Tanzania Project (P160766)\n\n\nthe GoT is acting as an anchor tenant in guaranteeing an income for investors that commit funds to building\nout their network in urban and rural areas. Again, the incremental additional investment to which the project\nwill contribute is difficult to estimate, as it will be the outcome of a competitive bidding process, but it is\nlikely to be substantial as the project is committing US$33m of IDA financing to the sub-component.\n\nc) **PCM could also arise under sub-component 1.1,** where the funding of the soft-centers and FabLabs are likely\nto stimulate investment from those private companies whose growth is nurtured through these initiatives.\nPrivate capital may also be invested alongside public funds in the planned National Center for ICT Professional\nDevelopment and Innovation. Finally, public investment in a national addressing and postcode system under\nsub-component 1.2 should stimulate private sector investment in eCommerce.\n\n\n**B. Results Monitoring and Evaluation Arrangements**\n\n64. A national ICT data collection survey has been undertaken in preparation of the project. This baseline survey,\nfinanced through the Digital Development Platform and using the Survey of Well-being via Instant and Frequent Tracking\n(SWIFT) methodology, was carried out by the WB in 2018-19. Subsequent surveys are expected to be carried out every\ntwo to three years (that is, twice during the life of the project and once following completion), covering over 1,400\nhouseholds in rural and urban locations and including Zanzibar. The SWIFT data diagnostic focused on key indicators\nrelated to ICT adoption, usage, and impact and was disaggregated by income group, gender, type of mobile\nphone/service, and so on. Areas targeted as part of the rural connectivity program will receive specific attention to help\ninform project design and course corrections. In addition, a _Digital Sources of Growth_ study was commissioned by the\nWB in 2019 and formed\n\nMention: \"SWIFT data diagnostic\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Diagnostic covering 1,400 households disaggregated by income and location.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000021:19:0", "occurrence_id": "75f7a2ffe9009694f6c6f091", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nTanzania Digital Foundations Project (P160766)\n\n\nc) **Other areas where PCM may be expected include sub-component 1.1,** where the funding of the\n\nsoft-centers and fab labs is likely to stimulate investment from those private companies whose\ngrowth is nurtured through these initiatives. Private capital may also be invested alongside public\nfunds in the planned National Center for ICT Professional Development and Innovation. Finally,\nthe public investment in a national addressing and postcode system under sub-component 1.2\nshould stimulate private sector investment in eCommerce.\n\n\n**B. Results Monitoring and Evaluation Arrangements**\n\n47. A national ICT data collection survey has been undertaken in preparation for the project. This baseline\nsurvey, financed through the Digital Development Platform and using the Survey of Well-being via Instant and\nFrequent Tracking (SWIFT) methodology, was carried out by the WB in 2018-19. Subsequent surveys are\nexpected to be carried out every two to three years (i.e., twice during the life of the project and once following\ncompletion), covering over 1,400 households in rural and urban locations and including Pemba and Zanzibar.\nThe study focused on key indicators related to ICT adoption, usage, and impact and was disaggregated by income\ngroup, gender, type of mobile phone/service, etc. Areas targeted as part of the rural connectivity program will\nreceive specific attention to help inform project design and course corrections. In addition, a Digital Sources of\nGrowth study was commissioned by the Bank in 2019 and formed the basis for a special focus chapter on the\ndevelopment of the Digital Economy in Tanzania in the 14 <sup>th</sup> edition of the Tanzania Economic Update (May 2020).\nUnder the project, a National ICT Statistical Management Information System will be developed (Subcomponent\n1.2b), which will allow for improved monitoring of key performance indicators (KPIs) in the\n\nMention: \"Survey of Well-being\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Baseline well-being survey methodology carried out by the WB.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000022:17:1", "occurrence_id": "f6de3c55dd0670577d065e36", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  to participate in full and benefit from the development process\n\n      - **Sustainability.** A sustainable development pathway that is resilient to disaster and climate\nchange, entails sustainable use of natural resources, and successfully manages the inevitable\nurbanization transition.\n\n\n16. **The Project will also contribute to the GOB’s Climate Change Strategy and Action Plan.** By\ndeliberately contributing to the country’s efforts on climate resilience and emission reduction, the Project\nis directly aligned with two of the Plan’s six pillars: (i) comprehensive disaster management and (ii)\nmitigation and low carbon development. The Project’s investments in the digital government platform\nwill facilitate important redundancy in the network, thereby protecting against outages caused by climate\nevents. They will also increase availability and use of public services by citizens and strengthen the GOB’s\nability to share data and information in the wake of disasters. Providing digital public services through the\ngovernment platform will also generate significant energy savings by reducing transportation, electricity,\nand paper usage across government and for citizens and businesses.\n\n\n17. **The EDGE Project is fully aligned with the World Bank Group’s Country Partnership Framework**\n**(CPF) for FY16 to FY20.** <sup>**20**</sup> [^20: World Bank. Country Partnership Framework for Bangladesh for the Period FY16 to 20 (Report No. 103713-BD; March 8, 2016\ndiscussed at the Board on April 5, 2016). Available at:] This Project will contribute to the CPF’s first two focus areas: Growth and\n\n\nToolkit.” Available at: _https://www.oecd.org/internet/broadband-policies-for-latin-america-and-the-caribbean-9789264251823-_\n_en.htm_ .\n17 Daffodil International University. 2018. “Survey on ICT Job Market in 2018.” Available at:\n_[https://cdc.daffodilvarsity.edu.bd/images/download/Survey-on-ICT-Job-Market19.pdf](https\n\nMention: \"Survey on ICT Job Market in 2018\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"University survey cited on the ICT labour market.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000022:33:0", "occurrence_id": "76f0799e7b42d4d9c8a11332", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nBangladesh: Enhancing Digital Government and Economy (EDGE) Project (P161086)\n\n\nfinancial or audit reports. It will be responsible for all Project-related FM activities as under previous\nprojects. The GOB’s general financial rules and treasury rules will be used as the basis for exercising\nappropriate controls over Project transactions wherever applicable. These regulations will be\nsupplemented by the Project Operations Manual (POM), which will be prepared to the WB’s acceptance\nas a legal covenant in the Finance Agreement. The POM includes a detailed description of the Project’s\ncomponents, funds flow arrangement, accounting, internal control, reporting, and oversight.\n\n\n73. IDA will disburse funds to the Project’s Designated Account (DA) based on quarterly interim\nunaudited financial reports (IUFRs) in agreed format that will include a six-month fund requirement\nforecast. A separate DA in the form of Convertible Taka Special Account will be opened in a commercial\nbank acceptable to the Association. There will not be any direct fund transfers to any other entities and\nall the eligible payments will be made using the Designated Account. The IUFRs should be submitted\nwithin 45 days from the end of the reporting period (fiscal quarter). However, the costs relating to fuel,\nsalaries, allowances, and taxes over 15 percent of total credit amount will be financed by the GOB as\nparallel financing and tracked in the IUFR like other IDA-funded projects. Details are in Annex 1.\n\n\n**(ii) Procurement**\n\n\n74. Procurement for the project will be carried out in accordance with the WB’s Procurement\nRegulations for IPF Borrowers for Goods, Works, Non-Consulting and Consulting Services, dated July 1,\n2016 (revised in November 2017 and August 2018). The Project will be subject to the WB’s Anticorruption\nGuidelines, dated October 15, 2006, revised in January 2011, and as of July 1,\n\nMention: \"quarterly interim\nunaudited financial reports\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Reports on which disbursements to the designated account are based.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000022:54:1", "occurrence_id": "17bf2e80bb53c848c08f9577", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: sup> . The gender gap in economic activity is wide in Bangladesh, and even\nmore so in the IT sector. Data from the 2018 job market survey shows that women account for only 16\npercent of ICT professionals. Women’s lower employment in this sector is partly because they are less\nlikely than men to have the digital skills required for DE jobs. For example, in tertiary education, women\naccount for 34 percent of graduates in Arts and Humanities but only 20 percent of graduates in STEM\nfields (World Bank Education Statistics Database). In addition, women who do acquire the skills do not\nalways transition into employment. For example, while 25 percent of female students of all students\nstudied ICT or Computer Science related subjects in over 90 universities and institutions of the country in\n2005-2016 time period, but only 13 percent joined the ICT industry after completion of their study. Other\nissues that affect women’s entry in the ICT sector and their movement up in such jobs include gender\ndiscrimination. For example, networking opportunities and promotions go to men in tech careers at a\nhigher rate than to women <sup>33</sup> [^33: Fintech Magazine. 2018. Bangladesh witnesses increased participation of women in tech. Available at:\nwww.fintechbd.com/bangladesh-witnesses-increased-participation-of-women-in-tech/] . Other studies of the labor market in Bangladesh show that once in a job,\nwomen are only half as likely as men to receive on-the-job training and many employers hold\ndiscriminatory attitudes toward female employees <sup>34</sup> [^34: World Economic Forum. 2019. Digital technology can cut global emissions by 15%. Here’s how. Available at:\n_[https://www.weforum.org/agenda/2019/01/why-digitalization-is-the-key-to-exponential-climate-action/](https://www.weforum.org/agenda/2019/01/why-digitalization-is-the-key-to-exponential-climate-action/)_] . This component will target female participation\nratio of 30 percent in its relevant activities to increase women’s\n\nMention: \"2018 job market survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Survey share of women among ICT professionals.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000022:54:1", "occurrence_id": "99738e9f2158f92bbbc45490", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: sup> . The gender gap in economic activity is wide in Bangladesh, and even\nmore so in the IT sector. Data from the 2018 job market survey shows that women account for only 16\npercent of ICT professionals. Women’s lower employment in this sector is partly because they are less\nlikely than men to have the digital skills required for DE jobs. For example, in tertiary education, women\naccount for 34 percent of graduates in Arts and Humanities but only 20 percent of graduates in STEM\nfields (World Bank Education Statistics Database). In addition, women who do acquire the skills do not\nalways transition into employment. For example, while 25 percent of female students of all students\nstudied ICT or Computer Science related subjects in over 90 universities and institutions of the country in\n2005-2016 time period, but only 13 percent joined the ICT industry after completion of their study. Other\nissues that affect women’s entry in the ICT sector and their movement up in such jobs include gender\ndiscrimination. For example, networking opportunities and promotions go to men in tech careers at a\nhigher rate than to women <sup>33</sup> [^33: Fintech Magazine. 2018. Bangladesh witnesses increased participation of women in tech. Available at:\nwww.fintechbd.com/bangladesh-witnesses-increased-participation-of-women-in-tech/] . Other studies of the labor market in Bangladesh show that once in a job,\nwomen are only half as likely as men to receive on-the-job training and many employers hold\ndiscriminatory attitudes toward female employees <sup>34</sup> [^34: World Economic Forum. 2019. Digital technology can cut global emissions by 15%. Here’s how. Available at:\n_[https://www.weforum.org/agenda/2019/01/why-digitalization-is-the-key-to-exponential-climate-action/](https://www.weforum.org/agenda/2019/01/why-digitalization-is-the-key-to-exponential-climate-action/)_] . This component will target female participation\nratio of 30 percent in its relevant activities to increase women’s\n\nMention: \"World Bank Education Statistics Database\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Graduate shares by field of study drawn from the education statistics database.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000026:25:1", "occurrence_id": "f45129c4692ecb436491f3fb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Saint**\n**Lucia is slightly below that of men, particularly in traditionally male-dominated sectors such as energy.**\nAccording to national estimates, in 2019 the total labor force participation was estimated to be 56 percent; the\nlabor force participation of females was 54 percent and males was 58 percent. <sup>40</sup> Women are systematically\nunderrepresented in specific—more profitable—sectors of work and even more so in leadership positions within\nthese sectors. International studies show that these gender employment gaps are mainly due to stereotypes and\nbiases, which result in lower confidence and lower enrollment in male-dominated fields such as STEM, so that\nwomen are discouraged from applying for jobs in these fields and leave them at higher rates. <sup>41</sup> Other factors\ninclude discrimination in the workplace and a disproportionate amount of family and household responsibilities\nfor women.\n\n44. **Globally, gaps in female participation in the labor market tend to be more pronounced in the energy**\n**sector, as it has a higher percentage of technical positions.** A 2017 study by the World Petroleum Council and\nBoston Consulting Group put the share of women in the worldwide oil and gas workforce at 22 percent.\nFurthermore, while women account for 27 percent of entry-level jobs in these two sectors that require a college\ndegree and 25 percent of midcareer-level jobs, their share is only 17 percent in senior and executive roles. <sup>42</sup>\nSimilarly, in 2015, the electricity, gas and water supply sectors were found to have women in 22 percent of senior\n\n\n39 University of West Indies Statistical Review 2012-2013, 2016-2017.\n_[https://www.mona.uwi.edu/principal/sites/default/files/principal/statistics/2016-2017/C.P6d%20-](https://www.\n\nMention: \"national estimates\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"National labour force participation estimates disaggregated by sex.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000026:40:2", "occurrence_id": "74dcd379734f29392e0122a1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  the GoSL prior to\nthe commencement of civil works.\n\n\n104. **Gender-based violence (GBV) appears to be prevalent across the Caribbean.** <sup>**60**</sup> According to the GBV\nCountry Profile -Saint Lucia by the World Bank, the rate of partner and ex-partner violence reported by women to\nsocial agencies was 419 per 100,000 in 2015. Based on data from the Division of Human Services and Family Affairs,\nthe rate of violence against girls in 2015 was 1,268 per 100,000 girls. <sup>61</sup> The UN in 2020 listed the country as having\none of the highest rates of femicide in Latin America and the Caribbean (LAC) at 4.4 per 100,000 women. <sup>62</sup> In\nterms of decision-making, voice and agency, women remain significantly under-represented in the country’s\npolitical and decision-making domain, with only 17 percent of women in the national Parliament. <sup>63</sup> [^63: World Bank (2014); “Proportion of Women in National Parliament”; https://data.worldbank.org/indicator/SG.GEN.PARL.ZS (accessed on\nJuly 17, 2018)]\n\n\n60 Pan American Health Organization and the Centers for Disease Control and Prevention (2012); “Violence Against Women in Latin America\nand the Caribbean”\n61 This covers sexual assault and rape.\n[62 http://unlirec.screativa.com/wp-content/uploads/2018/05/Caribbean-Legal-Study-on-Gender-SALW-Sept.-2020.pdf](http://unlirec.screativa.com/wp-content/uploads/2018/05/Caribbean-Legal-Study-on-Gender-SALW-Sept.-2020.pdf)\n[63 World Bank (2014); “Proportion of Women in National Parliament”; https://data.worldbank.org/indicator/SG.GEN\n\nMention: \"GBV\nCountry Profile -Saint Lucia\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Country profile reporting partner-violence rates per 100,000.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000026:40:2", "occurrence_id": "a51751481e69efa06132d321", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  the GoSL prior to\nthe commencement of civil works.\n\n\n104. **Gender-based violence (GBV) appears to be prevalent across the Caribbean.** <sup>**60**</sup> According to the GBV\nCountry Profile -Saint Lucia by the World Bank, the rate of partner and ex-partner violence reported by women to\nsocial agencies was 419 per 100,000 in 2015. Based on data from the Division of Human Services and Family Affairs,\nthe rate of violence against girls in 2015 was 1,268 per 100,000 girls. <sup>61</sup> The UN in 2020 listed the country as having\none of the highest rates of femicide in Latin America and the Caribbean (LAC) at 4.4 per 100,000 women. <sup>62</sup> In\nterms of decision-making, voice and agency, women remain significantly under-represented in the country’s\npolitical and decision-making domain, with only 17 percent of women in the national Parliament. <sup>63</sup> [^63: World Bank (2014); “Proportion of Women in National Parliament”; https://data.worldbank.org/indicator/SG.GEN.PARL.ZS (accessed on\nJuly 17, 2018)]\n\n\n60 Pan American Health Organization and the Centers for Disease Control and Prevention (2012); “Violence Against Women in Latin America\nand the Caribbean”\n61 This covers sexual assault and rape.\n[62 http://unlirec.screativa.com/wp-content/uploads/2018/05/Caribbean-Legal-Study-on-Gender-SALW-Sept.-2020.pdf](http://unlirec.screativa.com/wp-content/uploads/2018/05/Caribbean-Legal-Study-on-Gender-SALW-Sept.-2020.pdf)\n[63 World Bank (2014); “Proportion of Women in National Parliament”; https://data.worldbank.org/indicator/SG.GEN\n\nMention: \"data from the Division of Human Services and Family Affairs\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Agency data behind the violence-against-girls rate.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000026:61:1", "occurrence_id": "f061d88ec166848f9278aeb9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  FM\nsystems. Disbursement of project funds will be processed in accordance with World Bank procedures as stipulated\nin the Disbursement and Financial Information Letter (DFIL), and the Disbursement Guidelines for Investment\nProject Financing. The following disbursement methods will be available: advance, reimbursement, direct\npayment and special commitments as defined in the DFIL.\n\n\n12. **Advances will be processed into a pooled Designated Account (DA) denominated in US$, to be opened**\n**and maintained at the Bank of Saint Lucia.** Advances to the DA will be made based on the forecast of eligible\nexpenditures for a six months period, based on IFRs. As eligible expenditures are incurred, the PIU will withdraw\nthe amount to be financed by the Bank from the DA in accordance with the financing agreement. The PIU will\noperate a local currency account to finance project expenditures in local currency, to which funds from the US$DA\nwill be periodically transferred. These accounts will be operated in accordance with the Bank’s Disbursement\nGuidelines. The project will report on the use of the advances through quarterly IFRs showing the actual\nexpenditures incurred by category of expenditures, component, and subcomponent.\n\n13. **Retroactive financing for the IDA Credit will be allowed up to an amount of US$766,000** equivalent to\ncover payments made for project eligible expenditures under Category 2 of the IDA Financing Agreement, up to\n12 months before the date of signing of the Financing Agreement.\n\n14. **The PIU will be responsible for preparing IFRs on a quarterly basis to be submitted to the World Bank.**\nThese reports will provide the required monitoring information and will be used for disbursement purposes. The\nformat and content of the quarterly IFRs will be provided in the related project Disbursement Letter and POM.\nThe IFRs should cover: (i)\n\nMention: \"IFRs\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Financial reports supporting advance forecasts of eligible expenditures.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000027:26:1", "occurrence_id": "c629b8438241aa3235a317d7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  pay for sanitation services. The project\nwill also support the operationalization of specialized platforms for communications and receiving\nfeedback on DWASA’s services. These strong communication and community mobilization measures,\ntogether with rehabilitated and new sewers and house service connections, will ensure utilization and\nsustainability of assets.\n\n\n**IV.** **PROJECT APPRAISAL SUMMARY**\n\n\n**A. Technical, Economic and Financial Analysis (if applicable)**\n\n51. **Technical analysis.** The combined analysis of data and assessments in past studies, up-to-date\ndata collected from topographical surveys, geotechnical data, closed-circuit television (CCTV) camera\ninspection of existing sewers, wastewater quality analysis, GIS data, and on-site infrastructure\nassessments, together with the outcome of hydraulic sewer network calculation and updated SPS design,\nall contributed to a conclusion on the best way forward. The technical design also evaluated if and where\nrehabilitation of existing works and installations would be feasible and advantageous. The results\nconvincingly showed that not much of the existing collection, pumping, and treatment infrastructure is\nsuited for rehabilitation (see annex 2). All results of the technical analysis and justification of the suggested\nway forward have been described in conceptual design reports (CDRs) for (a) the Pagla sewer system,\nincluding laterals and house service connections; (b) trunk mains (Eastern Trunk Main and Western Trunk\nMain); and (c) sewage and septage treatment. The CDRs were used as supplementary information to the\nbidding documents, which have also been elaborated. To optimize the use of funds, the sewage collection\nsystem and the trunk mains have been designed for a long-term horizon up to 2050, whereas the Pagla\nSTP is designed for a short-term horizon up to 2025. This approach safeguards the sewer lines, for which\ncapacity upgrade is difficult and ensures that these sewer lines can remain in place and function properly\nin the\n\nMention: \"topographical surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Survey data feeding the technical design analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000027:26:1", "occurrence_id": "99c2571c04be3b75b321f9ab", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  pay for sanitation services. The project\nwill also support the operationalization of specialized platforms for communications and receiving\nfeedback on DWASA’s services. These strong communication and community mobilization measures,\ntogether with rehabilitated and new sewers and house service connections, will ensure utilization and\nsustainability of assets.\n\n\n**IV.** **PROJECT APPRAISAL SUMMARY**\n\n\n**A. Technical, Economic and Financial Analysis (if applicable)**\n\n51. **Technical analysis.** The combined analysis of data and assessments in past studies, up-to-date\ndata collected from topographical surveys, geotechnical data, closed-circuit television (CCTV) camera\ninspection of existing sewers, wastewater quality analysis, GIS data, and on-site infrastructure\nassessments, together with the outcome of hydraulic sewer network calculation and updated SPS design,\nall contributed to a conclusion on the best way forward. The technical design also evaluated if and where\nrehabilitation of existing works and installations would be feasible and advantageous. The results\nconvincingly showed that not much of the existing collection, pumping, and treatment infrastructure is\nsuited for rehabilitation (see annex 2). All results of the technical analysis and justification of the suggested\nway forward have been described in conceptual design reports (CDRs) for (a) the Pagla sewer system,\nincluding laterals and house service connections; (b) trunk mains (Eastern Trunk Main and Western Trunk\nMain); and (c) sewage and septage treatment. The CDRs were used as supplementary information to the\nbidding documents, which have also been elaborated. To optimize the use of funds, the sewage collection\nsystem and the trunk mains have been designed for a long-term horizon up to 2050, whereas the Pagla\nSTP is designed for a short-term horizon up to 2025. This approach safeguards the sewer lines, for which\ncapacity upgrade is difficult and ensures that these sewer lines can remain in place and function properly\nin the\n\nMention: \"geotechnical data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Ground conditions data combined into the technical analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000027:26:1", "occurrence_id": "c686a07b14e00bed8b98f551", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  pay for sanitation services. The project\nwill also support the operationalization of specialized platforms for communications and receiving\nfeedback on DWASA’s services. These strong communication and community mobilization measures,\ntogether with rehabilitated and new sewers and house service connections, will ensure utilization and\nsustainability of assets.\n\n\n**IV.** **PROJECT APPRAISAL SUMMARY**\n\n\n**A. Technical, Economic and Financial Analysis (if applicable)**\n\n51. **Technical analysis.** The combined analysis of data and assessments in past studies, up-to-date\ndata collected from topographical surveys, geotechnical data, closed-circuit television (CCTV) camera\ninspection of existing sewers, wastewater quality analysis, GIS data, and on-site infrastructure\nassessments, together with the outcome of hydraulic sewer network calculation and updated SPS design,\nall contributed to a conclusion on the best way forward. The technical design also evaluated if and where\nrehabilitation of existing works and installations would be feasible and advantageous. The results\nconvincingly showed that not much of the existing collection, pumping, and treatment infrastructure is\nsuited for rehabilitation (see annex 2). All results of the technical analysis and justification of the suggested\nway forward have been described in conceptual design reports (CDRs) for (a) the Pagla sewer system,\nincluding laterals and house service connections; (b) trunk mains (Eastern Trunk Main and Western Trunk\nMain); and (c) sewage and septage treatment. The CDRs were used as supplementary information to the\nbidding documents, which have also been elaborated. To optimize the use of funds, the sewage collection\nsystem and the trunk mains have been designed for a long-term horizon up to 2050, whereas the Pagla\nSTP is designed for a short-term horizon up to 2025. This approach safeguards the sewer lines, for which\ncapacity upgrade is difficult and ensures that these sewer lines can remain in place and function properly\nin the\n\nMention: \"GIS data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Geographic information layers used in the sewer network analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000027:45:1", "occurrence_id": "26b6e8f39d4b2932a897a9d8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: >career progression through getting<br>relevant placement after receiving<br>technical, management, leadership and<br>orientation training|This indicator will measure<br>female staff who are able to<br>progress from their current<br>grades and/or get<br>placement as a result of the<br>training provided under the<br>project.|Once a<br>year,<br>starting<br>Year 2<br>|Data from<br>DWASA<br>Administrativ<br>e Division<br>|Administrative/HR data<br>from DWASA<br>|PMU<br>|\n|Experience STP operator contracted for a<br>minimum of 3 years|This indicator will track the<br>contracting and the|Once, in<br>Year 3|Administrativ<br>e data from|Administrative and<br>procurement data|PMU<br>|\n\n\n\nPage 41 of 78\n\nMention: \"Administrative/HR data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Utility HR records tracking training outcomes and staffing indicators.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000027:45:1", "occurrence_id": "fce9a8a91847823e6c752357", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: >career progression through getting<br>relevant placement after receiving<br>technical, management, leadership and<br>orientation training|This indicator will measure<br>female staff who are able to<br>progress from their current<br>grades and/or get<br>placement as a result of the<br>training provided under the<br>project.|Once a<br>year,<br>starting<br>Year 2<br>|Data from<br>DWASA<br>Administrativ<br>e Division<br>|Administrative/HR data<br>from DWASA<br>|PMU<br>|\n|Experience STP operator contracted for a<br>minimum of 3 years|This indicator will track the<br>contracting and the|Once, in<br>Year 3|Administrativ<br>e data from|Administrative and<br>procurement data|PMU<br>|\n\n\n\nPage 41 of 78\n\nMention: \"procurement data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Administrative and procurement records tracking contracting milestones.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000027:56:1", "occurrence_id": "4e47f65b47dd52eaba8ddb81", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  effluents discharged\ninto the very last gravity sewer sections upstream of that facility. Further, the current Pagla STP is not able\nto meet the legally prevailing discharge standards set by the GoB’s DoE. Consequently, the project will\nsupport investments in the reactivation/new construction of sewer connections,\nrehabilitation/reconstruction of sewerage, and a new STP. This is the largest component of the DSIP.\n\n\n10. The technical details of the DSIP are based on the combined recommendations according to the\nDhaka Sewerage Master Plan <sup>29</sup> [^29: DWASA. 2013. _Dhaka Sewerage Master Plan_ .] and an update of input data and its analysis by the DB/DBO consultant\nhired under this project. The Dhaka Sewerage Master Plan had recommended refurbishment and new\nconstruction of secondary/tertiary sewers, the Eastern Trunk Main (using open-cut\nconstruction/microtunnelling) and associated SPSs, as well as new construction of the Pagla STP using the\ntrickling filter technology. The update under the DSIP included topographical surveys, geotechnical data,\nCCTV camera inspection of existing sewers, wastewater quality analysis, GIS data, on-site infrastructure\nassessments, hydraulic sewer network calculation, updated SPS design, cost comparisons, and several\nmore details, all of which have been summarized in CDRs, separately for (a) Pagla sewer system, including\nlaterals and house service connections; (b) trunk mains (Eastern Trunk Main and Western Trunk Main);\nand (c) sewage and septage treatment. The CDRs will be used as supplementary information to the bidding\ndocuments. Together with the ‘employer’s specifications’, these CDRs will provide the necessary guidance\nfor the detailed design that has to be done by the contractors. To optimize the use of funds, the sewage\ncollection system and the trunk mains have been designed for a long-term horizon up to 2050, whereas\nthe Pagla STP is designed for a short-term horizon up to\n\nMention: \"topographical surveys\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Survey data feeding the technical design analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000027:56:1", "occurrence_id": "2c565eb251952d10433ec388", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  effluents discharged\ninto the very last gravity sewer sections upstream of that facility. Further, the current Pagla STP is not able\nto meet the legally prevailing discharge standards set by the GoB’s DoE. Consequently, the project will\nsupport investments in the reactivation/new construction of sewer connections,\nrehabilitation/reconstruction of sewerage, and a new STP. This is the largest component of the DSIP.\n\n\n10. The technical details of the DSIP are based on the combined recommendations according to the\nDhaka Sewerage Master Plan <sup>29</sup> [^29: DWASA. 2013. _Dhaka Sewerage Master Plan_ .] and an update of input data and its analysis by the DB/DBO consultant\nhired under this project. The Dhaka Sewerage Master Plan had recommended refurbishment and new\nconstruction of secondary/tertiary sewers, the Eastern Trunk Main (using open-cut\nconstruction/microtunnelling) and associated SPSs, as well as new construction of the Pagla STP using the\ntrickling filter technology. The update under the DSIP included topographical surveys, geotechnical data,\nCCTV camera inspection of existing sewers, wastewater quality analysis, GIS data, on-site infrastructure\nassessments, hydraulic sewer network calculation, updated SPS design, cost comparisons, and several\nmore details, all of which have been summarized in CDRs, separately for (a) Pagla sewer system, including\nlaterals and house service connections; (b) trunk mains (Eastern Trunk Main and Western Trunk Main);\nand (c) sewage and septage treatment. The CDRs will be used as supplementary information to the bidding\ndocuments. Together with the ‘employer’s specifications’, these CDRs will provide the necessary guidance\nfor the detailed design that has to be done by the contractors. To optimize the use of funds, the sewage\ncollection system and the trunk mains have been designed for a long-term horizon up to 2050, whereas\nthe Pagla STP is designed for a short-term horizon up to\n\nMention: \"geotechnical data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Ground conditions data combined into the technical analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000027:56:1", "occurrence_id": "068d9a430285cf7182f123f7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  effluents discharged\ninto the very last gravity sewer sections upstream of that facility. Further, the current Pagla STP is not able\nto meet the legally prevailing discharge standards set by the GoB’s DoE. Consequently, the project will\nsupport investments in the reactivation/new construction of sewer connections,\nrehabilitation/reconstruction of sewerage, and a new STP. This is the largest component of the DSIP.\n\n\n10. The technical details of the DSIP are based on the combined recommendations according to the\nDhaka Sewerage Master Plan <sup>29</sup> [^29: DWASA. 2013. _Dhaka Sewerage Master Plan_ .] and an update of input data and its analysis by the DB/DBO consultant\nhired under this project. The Dhaka Sewerage Master Plan had recommended refurbishment and new\nconstruction of secondary/tertiary sewers, the Eastern Trunk Main (using open-cut\nconstruction/microtunnelling) and associated SPSs, as well as new construction of the Pagla STP using the\ntrickling filter technology. The update under the DSIP included topographical surveys, geotechnical data,\nCCTV camera inspection of existing sewers, wastewater quality analysis, GIS data, on-site infrastructure\nassessments, hydraulic sewer network calculation, updated SPS design, cost comparisons, and several\nmore details, all of which have been summarized in CDRs, separately for (a) Pagla sewer system, including\nlaterals and house service connections; (b) trunk mains (Eastern Trunk Main and Western Trunk Main);\nand (c) sewage and septage treatment. The CDRs will be used as supplementary information to the bidding\ndocuments. Together with the ‘employer’s specifications’, these CDRs will provide the necessary guidance\nfor the detailed design that has to be done by the contractors. To optimize the use of funds, the sewage\ncollection system and the trunk mains have been designed for a long-term horizon up to 2050, whereas\nthe Pagla STP is designed for a short-term horizon up to\n\nMention: \"GIS data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Geographic information layers used in the sewer network analysis.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000032:21:0", "occurrence_id": "1a7738a7b62e6526346a278d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nDRC Transport and Connectivity Support Project (P161877)\n\n\n39. This subcomponent will finance the operational costs of the _Cellule Infrastructures_ (CI) for\ncoordination and management, including staff costs. It will finance project management and coordination\nactivities, including: (i) salary of key CI personnel proportionately to their involvement in project\nimplementation; (ii) external financial audits; and (iii) acquisition of office furniture, equipment, and\nlogistics. The subcomponent will also support the CI for the monitoring and evaluation of the project,\nincluding use of geo-referencing instruments for monitoring planned and rehabilitated roads, and impact\nevaluations.\n\n40. This subcomponent will also finance innovative supervision tools (for example, drones and social\nmedia monitoring <sup>51</sup> [^51: The objective of social media monitoring is to harness social media analytics and remote‐based geospatial analysis to identify\nand monitor for potentially project‐associated incidents and other social impacts. The real‐time and remote-based social media\nmonitoring has the potential to provide early warning of harm.] ), communication activities, and the development and use of internet-based tools for\nremote supervision (such as through geolocating software like Mapillary and GEMS <sup>52</sup> [^52: GEMS: Geo-Enabling Initiative for Monitoring and Supervision] ) particularly in hardto-reach areas, and zones affected by conflict or unrest. Finally, it will finance in-time analysis and\ninformation on conflict and insecurity dynamics in the project targeted areas to be aware and sensitive of\npossible security challenges and potentially harmful actions to quickly adapt, respond, and ensure staff,\ncommunity beneficiary safety and well-being. This will include accessing information from a variety of\nsources, including existing mechanisms such as Kivu Security Tracker that provide continual data and\ninformation on conflict and insecurity dynamics in Eastern DRC. Should similar arrangements be required\nfor the Kasais (bearing in mind the contextual differences between the provinces), this will be considered\nand resourced accordingly.\n\n**<u>Component 2:\n\nMention: \"Kivu Security Tracker\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Continual conflict and insecurity data for eastern DRC safety monitoring.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000033:13:0", "occurrence_id": "1169c52fe8bb35ef06a202f5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nPower System Efficiency and Resilience Project (P162151)\n\n\ninvestments in more carbon‐intensive coal generation, high‐cost emergency power, and LNG‐fueled generation.\nThe net benefits of increasing high‐efficiency thermal generation capacity by 300 MW in Yangon have been\nestimated at US$408 million (between 2019 and 2030) in net present value (NPV) terms, after accounting for\ninvestment costs.\n\n\n**Figure 1. Forecasted Supply/Demand Gap** **Figure 2. Indicative Least‐Cost Generation Expansion Plan** <sup>**11**</sup>\n\n\n\n\n\n50\n\n\n40\n\n\n30\n\n\n20\n\n\n10\n\n\n    \n2019 2020 2021 2022 2023 2024\n\n\nSupply deficit Generation under construction\n\n\nExisting supply Consumption\n\n\n_Source:_ World Bank estimates based on data from\nMOEE.\n\n\n\n50\n\n\n40\n\n\n30\n\n\n20\n\n\n10\n\n\n \n\n\n\n\n\n\n\n\n\n\n2020 2021 2022 2023 2024 2025 2026\n\n\nGen Exist+Cons Cap. New Hydro New Gas\nNew LNG New VRE (incl. solar) Imports\nNew Gen Rentals Consumption\n\n\n_Source:_ World Bank staff.\n\n\n\n**(ii) Remove Bottlenecks to Provide Access to Modern Electricity Services to the Poorest**\n\n\n9. **Access to electricity from the public electric grid is unequal, with the poorest and ethnic minorities less**\n**likely to be connected.** As noted earlier, the successful implementation of electrification efforts through the NEP\nhas increased the rate of access to the public grid to around half of Myanmar’s population. However, the poorest\nhouseholds are yet to benefit from access to modern electricity services. According to the latest Myanmar Living\nConditions Survey (MLCS 2017), <sup>12</sup> 80 percent of the bottom 20 percent of households were not connected to the\npublic grid. According to the World Bank Myanmar Systematic Country Diagnostic, <sup>13</sup> ethnic minority households\nliving predominantly in areas away from the public grid were about 20\n\nMention: \"data from\nMOEE\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Ministry supply and consumption data behind the generation estimates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000033:13:0", "occurrence_id": "d9669c68a9dd220d87b60296", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nPower System Efficiency and Resilience Project (P162151)\n\n\ninvestments in more carbon‐intensive coal generation, high‐cost emergency power, and LNG‐fueled generation.\nThe net benefits of increasing high‐efficiency thermal generation capacity by 300 MW in Yangon have been\nestimated at US$408 million (between 2019 and 2030) in net present value (NPV) terms, after accounting for\ninvestment costs.\n\n\n**Figure 1. Forecasted Supply/Demand Gap** **Figure 2. Indicative Least‐Cost Generation Expansion Plan** <sup>**11**</sup>\n\n\n\n\n\n50\n\n\n40\n\n\n30\n\n\n20\n\n\n10\n\n\n    \n2019 2020 2021 2022 2023 2024\n\n\nSupply deficit Generation under construction\n\n\nExisting supply Consumption\n\n\n_Source:_ World Bank estimates based on data from\nMOEE.\n\n\n\n50\n\n\n40\n\n\n30\n\n\n20\n\n\n10\n\n\n \n\n\n\n\n\n\n\n\n\n\n2020 2021 2022 2023 2024 2025 2026\n\n\nGen Exist+Cons Cap. New Hydro New Gas\nNew LNG New VRE (incl. solar) Imports\nNew Gen Rentals Consumption\n\n\n_Source:_ World Bank staff.\n\n\n\n**(ii) Remove Bottlenecks to Provide Access to Modern Electricity Services to the Poorest**\n\n\n9. **Access to electricity from the public electric grid is unequal, with the poorest and ethnic minorities less**\n**likely to be connected.** As noted earlier, the successful implementation of electrification efforts through the NEP\nhas increased the rate of access to the public grid to around half of Myanmar’s population. However, the poorest\nhouseholds are yet to benefit from access to modern electricity services. According to the latest Myanmar Living\nConditions Survey (MLCS 2017), <sup>12</sup> 80 percent of the bottom 20 percent of households were not connected to the\npublic grid. According to the World Bank Myanmar Systematic Country Diagnostic, <sup>13</sup> ethnic minority households\nliving predominantly in areas away from the public grid were about 20\n\nMention: \"Myanmar Living\nConditions Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Household survey showing grid access gaps among the poorest households.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000033:13:0", "occurrence_id": "e179a5c8d19cd525bf610879", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nPower System Efficiency and Resilience Project (P162151)\n\n\ninvestments in more carbon‐intensive coal generation, high‐cost emergency power, and LNG‐fueled generation.\nThe net benefits of increasing high‐efficiency thermal generation capacity by 300 MW in Yangon have been\nestimated at US$408 million (between 2019 and 2030) in net present value (NPV) terms, after accounting for\ninvestment costs.\n\n\n**Figure 1. Forecasted Supply/Demand Gap** **Figure 2. Indicative Least‐Cost Generation Expansion Plan** <sup>**11**</sup>\n\n\n\n\n\n50\n\n\n40\n\n\n30\n\n\n20\n\n\n10\n\n\n    \n2019 2020 2021 2022 2023 2024\n\n\nSupply deficit Generation under construction\n\n\nExisting supply Consumption\n\n\n_Source:_ World Bank estimates based on data from\nMOEE.\n\n\n\n50\n\n\n40\n\n\n30\n\n\n20\n\n\n10\n\n\n \n\n\n\n\n\n\n\n\n\n\n2020 2021 2022 2023 2024 2025 2026\n\n\nGen Exist+Cons Cap. New Hydro New Gas\nNew LNG New VRE (incl. solar) Imports\nNew Gen Rentals Consumption\n\n\n_Source:_ World Bank staff.\n\n\n\n**(ii) Remove Bottlenecks to Provide Access to Modern Electricity Services to the Poorest**\n\n\n9. **Access to electricity from the public electric grid is unequal, with the poorest and ethnic minorities less**\n**likely to be connected.** As noted earlier, the successful implementation of electrification efforts through the NEP\nhas increased the rate of access to the public grid to around half of Myanmar’s population. However, the poorest\nhouseholds are yet to benefit from access to modern electricity services. According to the latest Myanmar Living\nConditions Survey (MLCS 2017), <sup>12</sup> 80 percent of the bottom 20 percent of households were not connected to the\npublic grid. According to the World Bank Myanmar Systematic Country Diagnostic, <sup>13</sup> ethnic minority households\nliving predominantly in areas away from the public grid were about 20\n\nMention: \"World Bank Myanmar Systematic Country Diagnostic\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Diagnostic reporting grid access gaps for ethnic minority households.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "prwp", "source_key": "prwp:007479:14:3", "occurrence_id": "6c2e249cb6a9cf7c4c56b2b9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: >_8 _<br>_9 _<br>_10_|\n|Complete pass-through<br>Lower-bound elasticity<br>Medium elasticity<br>Upper-bound elasticity|−1.81<br>−2.09<br>−2.18<br>−2.22<br>−2.23<br>−2.14<br>−1.94<br>−1.70<br>−1.33<br>−0.67<br>−0.81<br>−1.08<br>−1.22<br>−1.29<br>−1.34<br>−1.33<br>−1.23<br>−1.10<br>−0.88<br>−0.46<br>−0.36<br>−0.56<br>−0.67<br>−0.74<br>−0.79<br>−0.80<br>−0.75<br>−0.67<br>−0.55<br>−0.29<br>0.09<br>−0.04<br>−0.13<br>−0.18<br>−0.23<br>−0.26<br>−0.26<br>−0.25<br>−0.21<br>−0.12|\n\n\n\n_Source:_ Estimates based on Indonesia National Socioeconomic Survey 2016.\n_Note:_ The table shows the share of total consumption for each decile. Complete pass-through refers to elasticity equal to zero; consumers\npay all the increased prices; and this does not affect the quantity purchased.\n\n\n10 The analysis assumes the prices of both clove and white cigarettes increase by 25 percent. However, given the complex tobacco\n\nMention: \"Indonesia National Socioeconomic Survey 2016\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Consumption share estimates by decile are computed from this household survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000001:1:0", "occurrence_id": "45fe0be730864e8fb8e696c7", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Sectoral and Institutional Context**\nIndia has the largest energy access deficit of any single country. The problem is compounded by\nunreliability of supply and low consumption levels where access is available. Rural consumers,\ntypically the poorest, constitute the bulk of India’s 394 million people without electricity (93\npercent of India’s un-electrified population or roughly 289 million people). More than two-thirds of\nun-electrified households belong to the lowest 40 percent of the income ladder and are largely\nconcentrated in five states – Uttar Pradesh, Bihar, West Bengal, Rajasthan, and Orissa (NSS, 2010).\nWhere electricity is available, supply is intermittent. Industries and commercial enterprises suffer\ndue to unreliable supply and are forced to invest in expensive back-up generation.\nAdditionally, India faces energy security and environmental concerns around its generation mix. 80\npercent of electricity generated and 57 percent of installed power capacity is coal fired. The power\nsector accounts for about one half of the total CO2 emissions in the country, as against a global\naverage of one-third. . According to a recent World Bank study ‘Energy Intensive Sectors on the\nIndian Economy- Path to Low Carbon Development’, even under the lowest emissions scenario,\nelectricity will remain the highest contributor to the growth of carbon emissions until 2031, and\ncoal-based plants will continue to dominate India’s energy mix with a 51 percent share in 2031. The\nprimacy of thermal fuels also means that the sector remains vulnerable to endemic coal shortages\nand price fluctuations\n\nIn order to achieve its vision of universal access, India has decided to adopt a two pronged\napproach, focusing on both grid and off-grid solutions. Progress towards achieving this goal gained\nmomentum with the launch of the flagship Rajiv Gandhi Grameen Vidyutikaran Yojana (RGGVY)\nin 2005, consolidating all ongoing rural electrification programs at the time. The Government had\nalso promoted decentralized distributed generation (DDG)\n\nMention: \"NSS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Household survey cited for the concentration of the bottom 40 percent across five states.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000003:2:0", "occurrence_id": "1dcaf8b7b57f9149aa1fd608", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nNorth Macedonia Public Sector Energy Efficiency Project (P149990)\n\n\n\n\n\n\n\n\n\nOther Decision (as needed)\n\n\n**B. Introduction and Context**\n\nCountry Context\n\n\n**The Republic of North Macedonia is a land-locked country at the heart of the Balkans, characterized by its**\n**mountainous terrain, intersected by valleys and lowlands.** It is a transit region, which mainly has continental climate in\nthe mountains, and Mediterranean climate along the river valleys and towards the south. In terms of connectivity, the\ncountry sits on two Pan-European corridors, namely the Corridor VIII and Corridor X, and its proximity with the EU\nprovides access to a large export market with 650 million customers. According to the last census of 2002, the\npopulation is about 2 million, of which 25 percent live in the capital Skopje, while 40 percent live in rural areas. Ethnic\nMacedonians make up 64 percent of the total population, ethnic Albanians 25 percent, other ethnicities, such as Turks,\nabout 4 percent, and Roma about 3 percent. It is not possible to determine the current ethnic distribution in the\nabsence of a more recent census. Ethnic relations have been a cause of tensions in the past.\n\n\n**Following the resolution of the extended political and electoral crisis of 2015–2017, the Government committed to an**\n**ambitious reform agenda and acceleration of EU and NATO accession.** North Macedonia’s extended political crisis in\n2015–17 highlighted institutional weaknesses and affected the citizens’ confidence in state institutions <sup>1</sup> . Political\nuncertainty continued after the December 11, 2016 parliamentary elections, due to the difficult process of formation of\na parliamentary majority. A new government was inaugurated on May 2017, with a program articulating an ambitious\nreform agenda. The government’s program focuses on economic growth, job creation, fair taxation, support to SMEs\nand reform of the social protection for the\n\nMention: \"last census of 2002\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Census population shares for the capital and rural areas.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000004:4:0", "occurrence_id": "547d203e7f144ca36f1d6a7a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nNorth Macedonia Public Sector Energy Efficiency Project (P149990)\n\n\n**Country context**\n\n\n1. **North Macedonia is a landlocked country at the heart of the Balkans, characterized by its**\n**mountainous terrain that is intersected by valleys and lowlands.** Its proximity to the European Union (EU)\npotentially provides the country with access to a large export market of 650 million customers. According to the\nlast census of 2002, the population is about two million people of which 25 percent live in the capital Skopje; 40\npercent reside in rural areas; and the remaining share lives in smaller urban centers.\n\n\n2. **The resolution of the decades-long dispute with Greece over the country’s name marks a turning**\n**point in North Macedonia’s history as an independent nation.** On June 12, 2018, the Governments of North\nMacedonia and Greece signed the Prespa Agreement <sup>1</sup> aimed at resolving the prolonged name issue <sup>2</sup> . The\nparliament in Skopje endorsed the necessary constitutional changes introducing the new name of “Republic of\nNorth Macedonia” on January 11, 2019. The use of the new name entered into force in February 2019 after\nratification of the Prespa Agreement by the Greek Parliament. In parallel, North Macedonia signed the NATO\naccession protocol, a process that had stalled for years due to the dispute. In April 2018, the European\nCommission (EC) recommended the opening of negotiations with North Macedonia, but on October 17, 2019,\nthe Council of the EU failed to reach the decision on opening negotiations with North Macedonia. The Council\nwill revert to the issue before the EU-Western Balkans summit in Zagreb in May 2020 <sup>3</sup> . Following the European\nCouncil's decision, the Prime Minister announced early elections which all political parties agreed to hold on\n\nMention: \"last census of 2002\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Census population shares for the capital and rural areas.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000008:24:0", "occurrence_id": "5a916eeb468c6c487245ad67", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nOromia Forested Landscape Program – Emission Reduction Project (P151294)\n\n\nof institutions, the definition of sound policies in the various sectors, and technical capacity\nbuilding to enhance impacts. It also includes all the actions needed to be able to generate\nERCs: definition of the methodologies, establishment and update of the baselines, the\nprocesses for the measurement and reporting of the emissions, the verification and issuance\nof ERCs, marketing of ERs to identify buyers, and so on. Finally, the enabling environment\nincludes the program management and operating costs for the BSP and the safeguards\nmanagement system.\n\n\n39. The underlying activities are implemented by different nongovernmental organizations (NGOs),\nsector bureaus, and private sectors and are needed to generate the ERs. However, they are not financed\nby this operation, which is exclusively limited to the purchase of ERCs.\n\n\n**C. Description of Project Components**\n\n\n**Principles of Operation**\n\n\n40. **The project focuses on purchasing ERCs, financing enabling activities for the carbon**\n**measurement (MRV), and the operating costs of the program.** The project consists of two types of\ntransactions: (a) a transaction for the purchase of ERCs through an ERPA and (b) investments for operating\nthe project and improving the MRV system through grants. ERCs are generated through an MRV system\ninvolving independent verification of monitoring reports. The quantity of ERCs created each year is\ndetermined after comparing the actual emissions from specific sectors against a previously determined\nbaseline for those sectors. For that purpose, and following the national system for GHG inventory, each\nsector is divided into GHG emission reporting subcategories.\n\n\n41. **The ISFL methodology states that the project will monitor the subcategories that are the main**\n**absolute contributors to the country GHG inventory and has set clear rules to select the ‘eligible**\n*\n\nMention: \"country GHG inventory\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Inventory defines which subcategories the ISFL methodology must monitor.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000008:39:1", "occurrence_id": "e75bb482e0510a699e1523ae", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The federal MRV task force monitors the<br>process and reports to the federal REDD+<br>Steering Committee.<br>• <br>The MRV Unit under the Forest Resources<br>Assessment and Monitoring Directorate of<br>EFD manages workflows and day-to-day<br>coordination.|• <br>OEPA/ORCU, supported by the OFLP<br>Steering Committee and Technical<br>Working Group, monitors<br>implementation of MRV within the<br>region.|\n|Measurement|• <br>The EGSII collects LULC data.<br>• <br>The EFD MRV Unit produces the map.<br>• <br>The MRV Unit in coordination with NRS<br>regularly collects, analyzes, and aggregates<br>primary data.|• <br>The ORCU MRV team collects primary<br>and secondary data on program<br>interventions (that is, geographical<br>information on A/R activities, program-<br>level biomass survey data, and so on).|\n|Reporting|• <br>The EFD MRV Unit in coordination with<br>NRS calculates GHG emissions at the<br>regional level.<br>• <br>The EFD MRV Unit delivers official GHG<br>emissions estimates.|• <br>The ORCU MRV Unit will compile results<br>of the EFD MRV Unit for the region and<br>submit a report in the form of a<br>Program Document (for example,<br>following the BioCF-ISFL methodology<br>and Verified Carbon Standard).|\n|Verification|• <br>It is conducted by national or international<br>entities.<br>• <\n\nMention: \"LULC data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Land use and land cover data collected for MRV map production.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000008:39:1", "occurrence_id": "bff79b6fc9d62190ea53c27e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The federal MRV task force monitors the<br>process and reports to the federal REDD+<br>Steering Committee.<br>• <br>The MRV Unit under the Forest Resources<br>Assessment and Monitoring Directorate of<br>EFD manages workflows and day-to-day<br>coordination.|• <br>OEPA/ORCU, supported by the OFLP<br>Steering Committee and Technical<br>Working Group, monitors<br>implementation of MRV within the<br>region.|\n|Measurement|• <br>The EGSII collects LULC data.<br>• <br>The EFD MRV Unit produces the map.<br>• <br>The MRV Unit in coordination with NRS<br>regularly collects, analyzes, and aggregates<br>primary data.|• <br>The ORCU MRV team collects primary<br>and secondary data on program<br>interventions (that is, geographical<br>information on A/R activities, program-<br>level biomass survey data, and so on).|\n|Reporting|• <br>The EFD MRV Unit in coordination with<br>NRS calculates GHG emissions at the<br>regional level.<br>• <br>The EFD MRV Unit delivers official GHG<br>emissions estimates.|• <br>The ORCU MRV Unit will compile results<br>of the EFD MRV Unit for the region and<br>submit a report in the form of a<br>Program Document (for example,<br>following the BioCF-ISFL methodology<br>and Verified Carbon Standard).|\n|Verification|• <br>It is conducted by national or international<br>entities.<br>• <\n\nMention: \"primary data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"Primary data collected, analysed and aggregated by the MRV unit.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000008:39:1", "occurrence_id": "bab94e8a696724aca8bb0ac2", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: The federal MRV task force monitors the<br>process and reports to the federal REDD+<br>Steering Committee.<br>• <br>The MRV Unit under the Forest Resources<br>Assessment and Monitoring Directorate of<br>EFD manages workflows and day-to-day<br>coordination.|• <br>OEPA/ORCU, supported by the OFLP<br>Steering Committee and Technical<br>Working Group, monitors<br>implementation of MRV within the<br>region.|\n|Measurement|• <br>The EGSII collects LULC data.<br>• <br>The EFD MRV Unit produces the map.<br>• <br>The MRV Unit in coordination with NRS<br>regularly collects, analyzes, and aggregates<br>primary data.|• <br>The ORCU MRV team collects primary<br>and secondary data on program<br>interventions (that is, geographical<br>information on A/R activities, program-<br>level biomass survey data, and so on).|\n|Reporting|• <br>The EFD MRV Unit in coordination with<br>NRS calculates GHG emissions at the<br>regional level.<br>• <br>The EFD MRV Unit delivers official GHG<br>emissions estimates.|• <br>The ORCU MRV Unit will compile results<br>of the EFD MRV Unit for the region and<br>submit a report in the form of a<br>Program Document (for example,<br>following the BioCF-ISFL methodology<br>and Verified Carbon Standard).|\n|Verification|• <br>It is conducted by national or international<br>entities.<br>• <\n\nMention: \"biomass survey data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Programme-level biomass survey data used for carbon accounting.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000009:2:0", "occurrence_id": "fbbabed607bb92b2532a5f5a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: intense in zones in the west (West Wollega, Qeleme Wollega, Ilubabor) and east (Bale and Guji).\nPreliminary data have shown that throughout Oromia, nearly 157,000 ha of forest was lost between\n2000 and 2013, or around 12,000 ha/year. This has resulted in over 46 million tons of CO2\nequivalent emitted into the atmosphere over this period, or around 3.5 million tons annually\n(calculated based on Hansen et al, 2013 and Det Norske Veritas, 2015). At the same time, the\nhistoric afforestation/reforestation (A/R) rate is almost 3,100 ha/year, leading to a gain of over\n330,000tons of CO2e annually.\n\nDeforestation and forest degradation in Oromia are driven primarily by small-scale conversions for\nagricultural expansion as well as wood extraction for firewood and charcoal purposes. Subsistence\nagriculture is the main economic activity throughout Oromia, with farmers cultivating diverse crops\nsuch as barley, wheat, beans, potatoes, and cabbage in highlands and bananas, maize and teff in\nlowlands. Extraction of fuel wood is a driver of degradation throughout Ethiopia. Firewood is the\nprimary source of energy for 94 percent of Ethiopia’s population and the most important forest\nproduct consumed in Ethiopia, with total 2013 consumption exceeding 116 million m³. The\nmajority of firewood is produced from natural forest, including woodlands and shrub lands, and\ncurrent firewood demand is estimated to significantly exceed the sustainable yield potential of\nremaining forest areas. Indirect drivers include inadequate development and implementation of\nland-use plans, weak cross-sectoral policy and investment coordination, population growth and\nmigration into forested areas, as well as road expansion.\n\nForests in Oromia are managed by, affected by, or used by a range of government institutions and\ncitizens. Coordination of investments, institutions, information, and incentives that impact or are\nimpacted by forest resources is\n\nMention: \"Preliminary data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Preliminary figures on forest loss area and rate across Oromia.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000033:17:0", "occurrence_id": "62e4a2a5fec9c934bc55a35b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nPower System Efficiency and Resilience Project (P162151)\n\n\n**Figure 3. Impact of Unreliable Electricity Supply to Firms**\n\n\n\n100\n\n\n80\n\n\n60\n\n\n40\n\n\n20\n\n\n0\n\n\n\n\n\n\n\n\n\n\n\nEAP average Myanmar Lower‐middle income average\n\n\n_Source:_ World Bank staff based on the Myanmar 2016 Enterprise\nSurvey.\n\n\n\nAs shown in figure 3, according to the latest\navailable Enterprise Survey for Myanmar\n(2016), the share of small enterprises reporting\nelectricity supply as being the biggest obstacle\n(15 percent) to doing business is double that of\nlarge firms (8 percent) and over three times the\naverage in East Asia and Pacific region (6\npercent). Low‐quality services also\ndisproportionately affect low‐income\nhouseholds. According to the MLCS 2017, 40\npercent of Yangon residents reported voltage\nfluctuations as their main concern, compared to\njust 20 percent at the national level. Unable to\nafford backup power generators and voltage\nregulators, lower‐income households bear the\ncosts of long outages and damages to electrical\nequipment.\n\n\n\n17. **The required development of power infrastructure represents an opportunity to improve the resilience**\n**of electricity services to natural hazards and climate change** . Due to its location, Myanmar is among the world's\nmost hazard‐prone countries. <sup>19</sup> [^19: Eckstein, D., M. L. Hutfils, and M. Winges. 2019. _Global Climate Risk Index 2019_ . Germanwatch.\n<u>https://germanwatch.org/sites/germanwatch.org/files/Global%20Climate%20Risk%20Index%202019_2.pdf</u>] Climate change is exacerbating the country's vulnerability to periodic floods,\nearthquakes, fires, storms, cyclones, and droughts. However, climate hazards are not considered at the planning\nstage and there is no emergency preparedness to enable quick and effective recovery of electricity supply to\naffected areas. For example, a rapid assessment of climate hazards revealed that 80 percent of the\n\nMention: \"Myanmar 2016 Enterprise\nSurvey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Enterprise survey behind firm-level electricity supply complaints.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000033:17:0", "occurrence_id": "791af8c281e043a9228c10f6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nPower System Efficiency and Resilience Project (P162151)\n\n\n**Figure 3. Impact of Unreliable Electricity Supply to Firms**\n\n\n\n100\n\n\n80\n\n\n60\n\n\n40\n\n\n20\n\n\n0\n\n\n\n\n\n\n\n\n\n\n\nEAP average Myanmar Lower‐middle income average\n\n\n_Source:_ World Bank staff based on the Myanmar 2016 Enterprise\nSurvey.\n\n\n\nAs shown in figure 3, according to the latest\navailable Enterprise Survey for Myanmar\n(2016), the share of small enterprises reporting\nelectricity supply as being the biggest obstacle\n(15 percent) to doing business is double that of\nlarge firms (8 percent) and over three times the\naverage in East Asia and Pacific region (6\npercent). Low‐quality services also\ndisproportionately affect low‐income\nhouseholds. According to the MLCS 2017, 40\npercent of Yangon residents reported voltage\nfluctuations as their main concern, compared to\njust 20 percent at the national level. Unable to\nafford backup power generators and voltage\nregulators, lower‐income households bear the\ncosts of long outages and damages to electrical\nequipment.\n\n\n\n17. **The required development of power infrastructure represents an opportunity to improve the resilience**\n**of electricity services to natural hazards and climate change** . Due to its location, Myanmar is among the world's\nmost hazard‐prone countries. <sup>19</sup> [^19: Eckstein, D., M. L. Hutfils, and M. Winges. 2019. _Global Climate Risk Index 2019_ . Germanwatch.\n<u>https://germanwatch.org/sites/germanwatch.org/files/Global%20Climate%20Risk%20Index%202019_2.pdf</u>] Climate change is exacerbating the country's vulnerability to periodic floods,\nearthquakes, fires, storms, cyclones, and droughts. However, climate hazards are not considered at the planning\nstage and there is no emergency preparedness to enable quick and effective recovery of electricity supply to\naffected areas. For example, a rapid assessment of climate hazards revealed that 80 percent of the\n\nMention: \"MLCS 2017\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Household survey reporting voltage fluctuation concerns by location.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000039:14:2", "occurrence_id": "7fd105729d6895fbeba2b8bb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  a collaborative process approach that is managed by a Forest Management Council\n(FMC). The FMC is tasked to develop sustainable forest management involving the different Forest Management Entities\n(FME) and communities that live within and around the boundaries of the FME (these include Special Use Forests - protected\nareas in Vietnam, Protection Forest Management Boards and State Forest Companies). The ACMA is set up to specifically\ninvolve poor forest dependent communities, baselines and information on the status of the communities is collected through:\nA REDD Needs Assessment (RNA), A Social Screening Report (SSR), and an updated management plan. ACMA is described in\nthe ER-PD and ESMF.\n\n\nPage 9 of 66\n\nMention: \"REDD Needs Assessment\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Assessment collecting baseline information on forest dependent communities.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000039:14:2", "occurrence_id": "ee91858aa35f5d0d8e3e82ff", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  a collaborative process approach that is managed by a Forest Management Council\n(FMC). The FMC is tasked to develop sustainable forest management involving the different Forest Management Entities\n(FME) and communities that live within and around the boundaries of the FME (these include Special Use Forests - protected\nareas in Vietnam, Protection Forest Management Boards and State Forest Companies). The ACMA is set up to specifically\ninvolve poor forest dependent communities, baselines and information on the status of the communities is collected through:\nA REDD Needs Assessment (RNA), A Social Screening Report (SSR), and an updated management plan. ACMA is described in\nthe ER-PD and ESMF.\n\n\nPage 9 of 66\n\nMention: \"Social Screening Report\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Screening report collecting community status information.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000039:45:0", "occurrence_id": "9728ff95313011413b38b3ea", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nNorth Central Region Emission Reductions Program (P162605)\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n|Topic|Risks and/or challenges|Potential solutions/ mitigations|\n|---|---|---|\n|||Binh, Quang Tri and Thua Thien Hue and the Hmong<br>language do not readily lend themselves to written<br>translation.;|\n||Women’s access to information less<br>than men’s;|More attention to targeting women by facilitating if<br>deemed necessary separate consultations with village<br>women facilitated by a female facilitator in the language<br>of women’s choice;|\n||Women’s active involvement in<br>consultations less than men’s;|More attention to times of meetings|\n||Poor persons (women and men) less<br>likely to receive adequate information;<br>|More attention to targeting and focusing on involvement<br>of poor households using the data supplied by DOLISA<br>and the survey findings of the SESA;|\n|Institutional<br>Framework|Extension services for forestry and<br>upland agriculture very weak;<br>|Long-term program to overhaul extension systems to<br>reach uplands areas that are evidence driven and based<br>on the specific needs of different localities;|\n||Limited exposure to participatory<br>approaches;|Overseas Development Assistance (ODA) required to<br>impart participatory approaches/techniques and/or<br>work with local applied social research institutes that<br>either have a demonstrated track record in participatory<br>approaches of have signaled their ability and willingness<br>to be involved with such approaches;|\n||Limited staffing and budgets;<br>|Staffing and budgeting another intractable issue (re\n\nMention: \"data supplied by DOLISA\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Agency data used to target poor households.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000039:45:0", "occurrence_id": "fa380ab8914750e84f29ddd8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nNorth Central Region Emission Reductions Program (P162605)\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n|Topic|Risks and/or challenges|Potential solutions/ mitigations|\n|---|---|---|\n|||Binh, Quang Tri and Thua Thien Hue and the Hmong<br>language do not readily lend themselves to written<br>translation.;|\n||Women’s access to information less<br>than men’s;|More attention to targeting women by facilitating if<br>deemed necessary separate consultations with village<br>women facilitated by a female facilitator in the language<br>of women’s choice;|\n||Women’s active involvement in<br>consultations less than men’s;|More attention to times of meetings|\n||Poor persons (women and men) less<br>likely to receive adequate information;<br>|More attention to targeting and focusing on involvement<br>of poor households using the data supplied by DOLISA<br>and the survey findings of the SESA;|\n|Institutional<br>Framework|Extension services for forestry and<br>upland agriculture very weak;<br>|Long-term program to overhaul extension systems to<br>reach uplands areas that are evidence driven and based<br>on the specific needs of different localities;|\n||Limited exposure to participatory<br>approaches;|Overseas Development Assistance (ODA) required to<br>impart participatory approaches/techniques and/or<br>work with local applied social research institutes that<br>either have a demonstrated track record in participatory<br>approaches of have signaled their ability and willingness<br>to be involved with such approaches;|\n||Limited staffing and budgets;<br>|Staffing and budgeting another intractable issue (re\n\nMention: \"survey findings of the SESA\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Survey findings used alongside agency data for targeting.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000040:49:0", "occurrence_id": "9adfe5c66f26d35bf70d689a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nEnhancing Community Driven Development Project (P162647)\n\n\nthe NCDDP, which indicates that GAD has primarily served to facilitate the implementation of the\nproject. Indeed, in some instances, the township team of DRD and the township technical advisor have\nwork with GAD to update government lists of unregistered villages that have been identified through\nthe initial NCDDP assessment. We expect similar support in implementing the ECDDP. However, there\nis a risk that where the ECDDP is operating in areas where EAOs have not signed the National Ceasefire\nAgreement (Shan and Rakhine States), the Ministry of Home Affairs may assign armed police as security\nto Bank missions that include foreign members, which could negatively affect our efforts at supervision.\nThe suitability of this and potential alternative options available to the Bank should be discussed\n<u>portfolio wide.</u>\n\n8. **Monitoring and evaluation** . Does the results framework include indicators to adequately\nmonitor social inclusion? Is technology being used to overcome challenges to monitoring inclusion?\nWhat other aspects of social inclusion are monitored by the project through other mechanisms\n(surveys, studies, consultations, etc.)?\n\n\nThe results framework includes an indicator for women’s participation rates and other data\ndisaggregated by gender, but not by ethnicity. However, the project MIS includes detailed information\nof the ethnic and religious make-up of all villages where the project operates. As such, the task team\nhas access to very precise information on the ethnicity of project beneficiaries.\nThe project’s management information system is a dynamic web-based publicly accessible tool that\nuses smart-phone technology to capture and upload project data in real-time and allows rapid\nmonitoring and follow-up on both process and results. The ECDDP will support the refinement of the\nMIS, with a view to expanding efforts to harness ICT technologies to improve the consistency and\nquality of community engagement. This may include the development of customized\n\nMention: \"government lists of unregistered villages\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Village lists updated jointly with the general administration department.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000041:53:1", "occurrence_id": "940acc2f727efe471ee14f80", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: . For example, despite, in\nprinciple, owning similar amounts of land, male-headed households own, on average, a higher amount of\nlivestock or farms. <sup>14</sup> [^14: Author’s own calculations using Comoros DHS data (2012). https://dhsprogram.com/what-we-do/survey/survey-display-\n[443.cfm.](https://dhsprogram.com/what-we-do/survey/survey-display-443.cfm)]\n\n\n8. **Agency/voice and decision-making.** According to DHS data regarding economic autonomy,\nwomen who were currently in union and who had been paid in money for their work were asked who\nmainly decides how to use this money. In only 40 percent of the cases, women said that they mainly\ndecide how to use their earnings. A quarter of the women said that the decision was made jointly with\nthe spouse and in a third of the cases it is mainly the spouse who decides on the use of the woman's\nearnings. Regarding the role and involvement of women in decision-making at the household level, the\nfinal say in certain decision-making such as women’s health care and major household purchases was\nstudied. In all cases, the data show that the involvement of the spouse in making each decision is\nimportant. In 48 percent of cases, the spouse mainly decides on the women’s health care, the latter\ndeciding independently only in 21 percent of cases. The decisions concerning important purchases for the\n\n\n11 World Bank. 2019. _Little Data Book on Gender_ [. https://data.worldbank.org/products/data-books/little-data-book-on-gender.](https://data.worldbank.org/products/data-books/little-data-book-on-gender)\nAfrican Development Bank. 2016. “Country Strategy Paper 2016–2020.”\n[https://www.afdb.org/fileadmin/uploads/afdb/Documents/Project-and-Operations/COMOROS_-_2016-](https://www.afdb.\n\nMention: \"Comoros DHS data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Survey calculations behind household land and livestock ownership.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000041:53:1", "occurrence_id": "159220cdc074785983ceafc8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: . For example, despite, in\nprinciple, owning similar amounts of land, male-headed households own, on average, a higher amount of\nlivestock or farms. <sup>14</sup> [^14: Author’s own calculations using Comoros DHS data (2012). https://dhsprogram.com/what-we-do/survey/survey-display-\n[443.cfm.](https://dhsprogram.com/what-we-do/survey/survey-display-443.cfm)]\n\n\n8. **Agency/voice and decision-making.** According to DHS data regarding economic autonomy,\nwomen who were currently in union and who had been paid in money for their work were asked who\nmainly decides how to use this money. In only 40 percent of the cases, women said that they mainly\ndecide how to use their earnings. A quarter of the women said that the decision was made jointly with\nthe spouse and in a third of the cases it is mainly the spouse who decides on the use of the woman's\nearnings. Regarding the role and involvement of women in decision-making at the household level, the\nfinal say in certain decision-making such as women’s health care and major household purchases was\nstudied. In all cases, the data show that the involvement of the spouse in making each decision is\nimportant. In 48 percent of cases, the spouse mainly decides on the women’s health care, the latter\ndeciding independently only in 21 percent of cases. The decisions concerning important purchases for the\n\n\n11 World Bank. 2019. _Little Data Book on Gender_ [. https://data.worldbank.org/products/data-books/little-data-book-on-gender.](https://data.worldbank.org/products/data-books/little-data-book-on-gender)\nAfrican Development Bank. 2016. “Country Strategy Paper 2016–2020.”\n[https://www.afdb.org/fileadmin/uploads/afdb/Documents/Project-and-Operations/COMOROS_-_2016-](https://www.afdb.\n\nMention: \"DHS data regarding economic autonomy\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Survey module on who decides how women's earnings are used.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000041:53:3", "occurrence_id": "b04bb488d87191faad15d7b6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: -Systematic-Country-Diagnostic)\n[14 Author’s own calculations using Comoros DHS data (2012). https://dhsprogram.com/what-we-do/survey/survey-display-](https://dhsprogram.com/what-we-do/survey/survey-display-443.cfm)\n[443.cfm.](https://dhsprogram.com/what-we-do/survey/survey-display-443.cfm)\n\n\nPage <mark>51</mark> of <mark>54</mark>\n\nMention: \"Comoros DHS data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Footnote source for household asset ownership calculations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000041:54:0", "occurrence_id": "871f4a4ff4dc1f84ca8f11df", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nComoros Solar Energy Integration Platform (P162783)\n\n\nhousehold are taken mainly by the spouse in 40 percent of the cases and independently by women only\nin 27 percent of cases.\n\n\n9. **Paid/unpaid labor, income and economic opportunities.** Women participate in the labor force\nmuch less than men do, with participation rates of 33 percent and 57 percent, respectively. The\nparticipation rate of women is one of the lowest in the region. <sup>15</sup> [^15: ILO stats. https://ilostat.ilo.org/data/.] Women’s unemployment rate is higher\nat 10.4 percent against 6.7 percent for men. <sup>16</sup> [^16: Ibid.] Even if women are employed, they are more likely to have\nan employment that qualifies as ‘vulnerable’, meaning that they lack formal arrangements and tend to be\nself-employed (70 percent versus 55 percent, respectively). <sup>17</sup> [^17: https://data.worldbank.org/indicator/SL.EMP.VULN.MA.ZS.] Employment in the Comoros is characterized\nby predominantly own-account workers, more than 60 percent of the employed work for themselves.\nHowever, women are disproportionally affected by this trend, as more than 64 percent of women are selfemployed workers compared to about 46 percent of men. <sup>18</sup> [^18: World Bank. 2019. _Systematic Country Diagnostic_ .\n[http://documents.worldbank.org/curated/en/354101559590231457/Comoros-Towards-a-More-United-and-Prosperous-Union-](http://documents.worldbank.org/curated/en/354101559590231457/Comoros-Towards-a-More-United-and-Prosperous-Union-of-Comoros-Systematic-Country-Diagnostic)\n[of-Comoros-Systematic-Country-Diagnostic.](http://documents.worldbank.org/curated/en/354101559590231457/Comoros-Towards-a-More-United-and-Prosperous-Union-of-Comoros-Systematic-Country-Diagnostic)]\n\n\n10. Labor market data for the Comoros not only show a lower women’s\n\nMention: \"ILO stats\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"Labour statistics cited for participation and unemployment rates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000041:54:0", "occurrence_id": "4e0e380a1d169b28e891d269", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nComoros Solar Energy Integration Platform (P162783)\n\n\nhousehold are taken mainly by the spouse in 40 percent of the cases and independently by women only\nin 27 percent of cases.\n\n\n9. **Paid/unpaid labor, income and economic opportunities.** Women participate in the labor force\nmuch less than men do, with participation rates of 33 percent and 57 percent, respectively. The\nparticipation rate of women is one of the lowest in the region. <sup>15</sup> [^15: ILO stats. https://ilostat.ilo.org/data/.] Women’s unemployment rate is higher\nat 10.4 percent against 6.7 percent for men. <sup>16</sup> [^16: Ibid.] Even if women are employed, they are more likely to have\nan employment that qualifies as ‘vulnerable’, meaning that they lack formal arrangements and tend to be\nself-employed (70 percent versus 55 percent, respectively). <sup>17</sup> [^17: https://data.worldbank.org/indicator/SL.EMP.VULN.MA.ZS.] Employment in the Comoros is characterized\nby predominantly own-account workers, more than 60 percent of the employed work for themselves.\nHowever, women are disproportionally affected by this trend, as more than 64 percent of women are selfemployed workers compared to about 46 percent of men. <sup>18</sup> [^18: World Bank. 2019. _Systematic Country Diagnostic_ .\n[http://documents.worldbank.org/curated/en/354101559590231457/Comoros-Towards-a-More-United-and-Prosperous-Union-](http://documents.worldbank.org/curated/en/354101559590231457/Comoros-Towards-a-More-United-and-Prosperous-Union-of-Comoros-Systematic-Country-Diagnostic)\n[of-Comoros-Systematic-Country-Diagnostic.](http://documents.worldbank.org/curated/en/354101559590231457/Comoros-Towards-a-More-United-and-Prosperous-Union-of-Comoros-Systematic-Country-Diagnostic)]\n\n\n10. Labor market data for the Comoros not only show a lower women’s\n\nMention: \"Labor market data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"National labour market data read for gender gaps in participation.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000042:27:1", "occurrence_id": "220c11ac6003b52fabbffd2f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |31|14|19|8|17|63|152|59|211|\n|Percent of vulnerable youth 15-24|43.3|31.9|32.9|19.7|33.4|28.4|30.5|33.0|31.2|\n\n\n_Source: Team’s calculation based on the 2017 Employment Survey_\n\n\n58. **In order to efficiently identify potential beneficiaries, several modalities will be used.** These may\ninclude: (i) mining the databases of the Social Registry to identify youth meeting the required profile; (ii)\nmining the existing ANE and other programs’ databases to identify potentially eligible youth who were\nplaced on waiting lists in various other programs; and (iii) purposefully targeting community-based\ncommunication and identification campaigns.\n\n59. **The selection of beneficiaries will rely on both self-selection in response to calls for youth**\n**identified through either of the three channels to register for the program, and on screening done by**\n**employment counsellors.** The final geographical allocation of beneficiaries will depend on the geographical\ndistribution of the vulnerable youth, the demand for skilled labor from employers, and the requirement to\nensure that at least 40 percent of participants are women. The detailed procedures for outreach,\nidentification, selection, and enrollment in program interventions will be defined in the POM.\n\n**D. Results Chain**\n\n60. **The project supports interventions to increase economic inclusion for vulnerable youth in several**\n**ways.** It does so: (i) by improving access to services to improve their job readiness and ability to link up\nwith existing labor demand, or to engage in further employability-enhancing activities (Component 1); (ii)\nby improving access to job-specific training and internships to respond to the demand of local employers\n(Component 2); and (iii) by promoting micro\n\nMention: \"2017 Employment Survey_\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Survey underlying the vulnerable youth shares in the table.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000042:27:1", "occurrence_id": "5241633e681cc122495c7583", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |31|14|19|8|17|63|152|59|211|\n|Percent of vulnerable youth 15-24|43.3|31.9|32.9|19.7|33.4|28.4|30.5|33.0|31.2|\n\n\n_Source: Team’s calculation based on the 2017 Employment Survey_\n\n\n58. **In order to efficiently identify potential beneficiaries, several modalities will be used.** These may\ninclude: (i) mining the databases of the Social Registry to identify youth meeting the required profile; (ii)\nmining the existing ANE and other programs’ databases to identify potentially eligible youth who were\nplaced on waiting lists in various other programs; and (iii) purposefully targeting community-based\ncommunication and identification campaigns.\n\n59. **The selection of beneficiaries will rely on both self-selection in response to calls for youth**\n**identified through either of the three channels to register for the program, and on screening done by**\n**employment counsellors.** The final geographical allocation of beneficiaries will depend on the geographical\ndistribution of the vulnerable youth, the demand for skilled labor from employers, and the requirement to\nensure that at least 40 percent of participants are women. The detailed procedures for outreach,\nidentification, selection, and enrollment in program interventions will be defined in the POM.\n\n**D. Results Chain**\n\n60. **The project supports interventions to increase economic inclusion for vulnerable youth in several**\n**ways.** It does so: (i) by improving access to services to improve their job readiness and ability to link up\nwith existing labor demand, or to engage in further employability-enhancing activities (Component 1); (ii)\nby improving access to job-specific training and internships to respond to the demand of local employers\n(Component 2); and (iii) by promoting micro\n\nMention: \"ANE and other programs’ databases\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"curate\", \"summary\": \"Existing programme databases mined to identify eligible youth.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000042:46:1", "occurrence_id": "f555d2ff7bd126ac03a23f04", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: br>labor force survey data. The<br>endline will assess beneficiaries<br>employment status six months<br>after completing the program.<br> <br>The definition of youth<br>employment is: a person aged 15-<br>24 who has over the past 7 days<br>for at least one hour either: (i)<br>performed work for wage or salary|Annual<br>|Labor force<br>survey data;<br>Beneficiary<br>survey<br>|Existing labor force survey<br>data; Survey firm covering a<br>representative sample of<br>beneficiaries<br>|PTCU; Ministry of<br>Employment, Youth and<br>Sports<br>|\n\n\n42\n\nMention: \"labor force survey data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Survey data measuring beneficiary employment at endline.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000042:46:1", "occurrence_id": "7141d04f83e27da547373fe8", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: br>labor force survey data. The<br>endline will assess beneficiaries<br>employment status six months<br>after completing the program.<br> <br>The definition of youth<br>employment is: a person aged 15-<br>24 who has over the past 7 days<br>for at least one hour either: (i)<br>performed work for wage or salary|Annual<br>|Labor force<br>survey data;<br>Beneficiary<br>survey<br>|Existing labor force survey<br>data; Survey firm covering a<br>representative sample of<br>beneficiaries<br>|PTCU; Ministry of<br>Employment, Youth and<br>Sports<br>|\n\n\n42\n\nMention: \"labor force survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Existing national survey paired with a beneficiary survey for the indicator.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000042:64:0", "occurrence_id": "47be7b0313c6e2d7f49a9cd5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nMauritania Youth Employability Project (P162916)\n\n\n**<mark>ANNEX 2: Economic and Financial Analysis</mark>**\n\n\n1. **The Mauritania Youth Employability Project has multiple components which are expected to**\n**impact beneficiaries’ outcomes by** : (i) enhancing their life, management, and technical skills and\ntherefore employability; (ii) providing them with cash grants, which may be used to kickstart income\ngenerating activities; and (iii) linking them up with suitable employers through internship and placement\nprograms.\n\n2. **For the purpose of this analysis, the benefits are defined as the average increase in the annual**\n**earnings as a result of program participation**, although these activities could reasonably expect to\ngenerate other monetary benefits (such as accumulation of assets for microenterprises, for example),\nhuman capital benefits, and socioeconomic benefits (such as women’s empowerment). While some of\nthese benefits will be discussed in the narrative, only expected increases in earnings will be included\nexplicitly in the cost-benefit analysis.\n\n3. **The cost-benefit analysis rests on the following core assumptions:**\n\n\n   - The benefits are expressed as the anticipated annual average increase in earnings of beneficiaries\nin US dollars.\n\n   - The expected baseline earnings of participants are estimated using 2014 Living Standards Survey\n( _<u>Enquête Permanente sur les Conditions de Vie des ménages</u>_ <u>, EPCV) data and based on the mean</u>\nincome for the cohort of Mauritanian youth aged 15 – 24.\n\n   - The gains are estimated over the duration of a working life, which is estimated at 40 years, taking\ninto account that the mean beneficiary age should be 20 years old and the official retirement age\nis 60 years old.\n\n   - All calculations assume a discount rate of 10 percent.\n\n   - The gains are estimated separately\n\nMention: \"2014 Living Standards Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Survey estimating baseline earnings for the youth beneficiary cohort.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000042:64:0", "occurrence_id": "e508eb17cc5dd9149202bf70", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nMauritania Youth Employability Project (P162916)\n\n\n**<mark>ANNEX 2: Economic and Financial Analysis</mark>**\n\n\n1. **The Mauritania Youth Employability Project has multiple components which are expected to**\n**impact beneficiaries’ outcomes by** : (i) enhancing their life, management, and technical skills and\ntherefore employability; (ii) providing them with cash grants, which may be used to kickstart income\ngenerating activities; and (iii) linking them up with suitable employers through internship and placement\nprograms.\n\n2. **For the purpose of this analysis, the benefits are defined as the average increase in the annual**\n**earnings as a result of program participation**, although these activities could reasonably expect to\ngenerate other monetary benefits (such as accumulation of assets for microenterprises, for example),\nhuman capital benefits, and socioeconomic benefits (such as women’s empowerment). While some of\nthese benefits will be discussed in the narrative, only expected increases in earnings will be included\nexplicitly in the cost-benefit analysis.\n\n3. **The cost-benefit analysis rests on the following core assumptions:**\n\n\n   - The benefits are expressed as the anticipated annual average increase in earnings of beneficiaries\nin US dollars.\n\n   - The expected baseline earnings of participants are estimated using 2014 Living Standards Survey\n( _<u>Enquête Permanente sur les Conditions de Vie des ménages</u>_ <u>, EPCV) data and based on the mean</u>\nincome for the cohort of Mauritanian youth aged 15 – 24.\n\n   - The gains are estimated over the duration of a working life, which is estimated at 40 years, taking\ninto account that the mean beneficiary age should be 20 years old and the official retirement age\nis 60 years old.\n\n   - All calculations assume a discount rate of 10 percent.\n\n   - The gains are estimated separately\n\nMention: \"Enquête Permanente sur les Conditions de Vie des ménages\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"National living conditions survey providing the income baseline.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000046:28:2", "occurrence_id": "63a38cc1246c89e8e3c90518", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  by the World Bank and other development agencies, the\ntariff is not designed to recoup capital expenditures, which will be covered by the Treasury per its agreement with\nONAS. With consistent application of the tariff increases recommended by the inter-ministerial council of\nDecember 6, 2021 (30 percent in 2022, 11 percent in 2023-26, and 6 percent in 2027-31) <sup>28</sup>, ONAS is expected to\ncover its operating costs and the concession fee by 2026 (refer to Annex 2 for more details on cost coverage).\nFailure to implement the approved tariff increases would result in a greater fiscal burden for the government,\nwhich would be obligated to subsidize ONAS’ operating costs, as it has been doing for the last several years. The\ngovernment’s SOE reform agenda includes plans to reduce operational subsidies for SOEs providing commercial\nservices.\n\n\n28 _Tariff increase percentage considers actual impacts on total revenues using prior data i.e., a nominal increase of 30 percent in 2022 has_\n_a positive impact of 25 percent on 2023 revenues – the difference accounts for low-income consumers with no tariff impact and changing_\n_consumption patterns_\n\n\n19\n\nMention: \"prior data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Earlier revenue data used to convert nominal tariff increases into realised impacts.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000046:59:6", "occurrence_id": "68669ec7d04e11742ec67521", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: 5._ _Cost of concession contracts has been inflated at 5.5% from the offer price quoted in 2020 (resulting figure per the_\n_contractual formula using a CPI of 6.2%)_\n_6._ _Tariff increase percentage considers actual impacts on total revenues using prior data i.e., a nominal increase of 30% in 2022_\n_has a positive impact of 24.8% on 2023 revenues – the difference accounts for low-income consumers with no tariff impact_\n_and changing consumption patterns_\n\n\n6. **Higher inflation scenario.** The inflation rate of 6.2 percent used in the model is based on the 5-year\naverage CPI rate for the period 2017 to 2021. However, persistently higher inflation over the medium to long term\nwould result in a greater financial burden assuming future tariff increases occur in line with the inter-ministerial\nrecommendation. Table A2.2 shows the financial impact in a scenario with a 25 percent higher rate of average\nconsumer price inflation (7.8 percent over the next 10 years). Under this scenario, ONAS would have a\nconsiderably higher deficit over the life of the project and would not cover its operating costs during the life of\nthe contracts.\n\n\n50\n\nMention: \"prior data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Earlier revenue data used to convert nominal tariff increases into realised impacts.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000047:13:0", "occurrence_id": "ee84ecdb44f7e7417bd0b107", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nProjet d’appui au PPP en matière d’assainissement en Tunisie (P162957)\n\n\nprévoit une réduction de 10 à 30 pour cent des précipitations annuelles. On s'attend également à une\naugmentation de la fréquence et de l'intensité des sécheresses et des inondations.\n\n\n**B. Contexte sectoriel et institutionnel**\n\n\n8. **Malgré les progrès réalisés dans l'amélioration de l'accès aux services d'approvisionnement en eau**\n**et d'assainissement (AEA), 19 pour cent des eaux usées et des boues fécales générées par les ménages**\n**tunisiens ne sont pas traitées ou éliminées de manière sûre** <sup>**9**</sup> [^9: Organisation mondiale de la santé et UNICEF Programme commun de surveillance, estimations de 2021.] **.** Au cours des dernières décennies, la Tunisie a\nfait des progrès substantiels dans l'amélioration de l'accès aux services d'AEA, et aujourd’hui un accès aux\nservices d'eau de base et à l'assainissement de base s’élève respectivement à 97 pour cent et 98 pour cent. <sup>10</sup> [^10: Idem]\nMalgré ces progrès, environ 360 000 personnes utilisent encore des services d'assainissement non améliorés\net, alors que la majorité de la population n'ayant pas accès aux égouts vit dans les zones rurales du pays, les\neaux usées produites par plus de 1,7 million de citadins ne sont toujours pas traitées\n\nMention: \"UNICEF Programme commun de surveillance\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"Cited joint surveillance programme estimates behind the waste safety claim.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000047:14:0", "occurrence_id": "45cc14d8820cdbd8b0eb8160", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nProjet d’appui au PPP en matière d’assainissement en Tunisie (P162957)\n\n\n10. **L'ONAS fait face à des difficultés croissantes pour fournir des services d'assainissement de qualité,**\n**en partie dues à une situation financière fragile qui s'est aggravée après la pandémie de COVID 19** . L'ONAS\nfournit actuellement des services de collecte et de traitement des eaux usées à près de 63 pour cent de la\npopulation du pays et à plus de 76 pour cent de la population vivant dans des zones urbaines ou en milieu\ncommunal. Avec des infrastructures vieillissantes et surchargées, seuls 62 pour cent des coûts d'exploitation\nde l’ONAS sont couverts par les redevances des usagers <sup>12</sup> [^12: Idem] . Les investissements en capital sont\nprincipalement financés par l'État. En outre, une partie des factures ne sont pas payées et les recouvrements\nde l'ONAS pour 2020 étaient de 78 pour cent du montant facturé, contre 86 pour cent en 2019 <sup>13</sup> [^13: Données de 2019 et 2020, comptes audités de l’ONAS.] . Comme\nles autres EP en Tunisie, l'ONAS reçoit régulièrement des subventions gouvernementales pour couvrir les\ndépenses d’investissement. Parallèlement, l'ONAS est assujetti partiellement à la taxe sur la valeur ajoutée\n(TVA), alors qu'il n'y a pas de TVA sur les services qu'il fournit <sup\n\nMention: \"Données de 2019 et 2020\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Audited utility data behind the collection rate comparison.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000047:14:0", "occurrence_id": "b47085f5220f89b7a5ee1df0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nProjet d’appui au PPP en matière d’assainissement en Tunisie (P162957)\n\n\n10. **L'ONAS fait face à des difficultés croissantes pour fournir des services d'assainissement de qualité,**\n**en partie dues à une situation financière fragile qui s'est aggravée après la pandémie de COVID 19** . L'ONAS\nfournit actuellement des services de collecte et de traitement des eaux usées à près de 63 pour cent de la\npopulation du pays et à plus de 76 pour cent de la population vivant dans des zones urbaines ou en milieu\ncommunal. Avec des infrastructures vieillissantes et surchargées, seuls 62 pour cent des coûts d'exploitation\nde l’ONAS sont couverts par les redevances des usagers <sup>12</sup> [^12: Idem] . Les investissements en capital sont\nprincipalement financés par l'État. En outre, une partie des factures ne sont pas payées et les recouvrements\nde l'ONAS pour 2020 étaient de 78 pour cent du montant facturé, contre 86 pour cent en 2019 <sup>13</sup> [^13: Données de 2019 et 2020, comptes audités de l’ONAS.] . Comme\nles autres EP en Tunisie, l'ONAS reçoit régulièrement des subventions gouvernementales pour couvrir les\ndépenses d’investissement. Parallèlement, l'ONAS est assujetti partiellement à la taxe sur la valeur ajoutée\n(TVA), alors qu'il n'y a pas de TVA sur les services qu'il fournit <sup\n\nMention: \"comptes audités de l’ONAS\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Audited accounts verifying the utility's billed and collected amounts.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000050:6:1", "occurrence_id": "4ac77cea2e94d9f53c1f4349", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: . To address the specific challenges, MoTIHUD has designed the Kenya Urban Support Program (KUSP,\nP156777) to strengthen urban governance, prepare participatory strategic and spatial planning, and finance\nurban infrastructure and services. Under the KUSP, as of January 2019, all counties have prepared their\nCounty Urban Institutional Development Strategies, established urban areas with approved charters,\nestablished boards, appointed urban managers, and enacted a budget vote as addressed by the Urban Areas\nand Cities Act (amended in 2016). Each county government will receive urban development grants for eligible\nactivities. Despite counties’ growing importance, urban development grant readiness assessments\nundertaken in 2018 revealed that only seven counties were ready to implement the grants.\n\n\n**The proposed pilot project will complement the World Bank’s ongoing and pipeline transport operations in**\n**Kenya by emphasizing holistic and inclusive transport and urban development planning.** The Meru town\nWestern and Eastern Bypasses was constructed with partial financing by the World Bank to alleviate traffic\ncongestion in Meru town center by providing alternative routes for through traffic. While the bypasses are\nanticipated to support better access to economic opportunities for the entire population in the town, the\nintended beneficiaries could be increased by improving the access roads connecting the underserved\ncommunities in the vicinity to these bypass roads and other towns’ main road networks. Access roads in\nthese communities are in poor condition, are inadequately maintained, and become impassable during the\nrainy season, hindering residents’ access to economic opportunities. Given existing financial capacity\nconstraints, the Meru County government is currently unable to fully respond to the needs of these\nunderserved communities. Empowering underserved communities and involving them in basic service\ndelivery planning and delivery through community road maintenance will fill this gap while the county builds\nits urban management capacity. Piloting access road improvement using Do-nou also provides an\n\n\nJun 29, 2020 Page 7 of 18\n\nMention: \"urban development grant readiness assessments\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"County readiness assessments showing only seven counties able to implement grants.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000052:18:1", "occurrence_id": "89cea36df8ce4168e7ad731e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  market clusters and the terminal markets needs to be\ndeveloped.\n\n\n8. **Women in Himachal Pradesh are primarily employed in agriculture and agro-based activities and have**\n**fluctuating incomes resulting from climate variations.** The female labor force participation for Himachal Pradesh\nis about 25 percent of which the percentage of women agriculture workers is as high as 82.7 percent (Agriculture\nCensus 2011). Women contribute to the majority of the activities in farm-based livelihoods and barring ploughing,\nwomen contribute substantially in all other farm related activities, including sowing, irrigation, using fertilizers,\nreaping and post-harvest management of the produce <sup>6</sup> [^6: The Approach Paper to the Eleventh Five Year Plan of Himachal Pradesh] . However, there is clear occupational segregation as\nwomen are relatively less engaged in off-farm, market-related activities, logistics and technical STEM-related job\nroles <sup>7</sup> [^7: Based on extensive stakeholder consultations] . Changing climatic conditions, an expected decrease in horticulture production and associated impact on\nagriculture-based livelihoods are likely to affect women cultivators and their farm-based sources of income more\nprominently as compared to men <sup>8</sup> [^8: The State Action Plan on Climate Change] . The proposed gender actions under the project will, therefore, focus on\nbuilding the skills sets of women producers’ groups/self-help groups and women employed in institutions to\nenable income diversification and career advancement.\n\n9. **The incidence of road crashes along HP roads is among the highest in India.** In 2016, as reported by the\nRoad Accident Data Management System (RADMS) 7,036 persons were injured in road accidents of which 1,272\nsuffered fatal injuries. The fatality rate in the State is 18.17 people per 100,000 population, which is comparatively\nhigher to the national level fatality rate of 11.53 people per 100,000 population. 85 percent of the fatalities\ninvolved motorized transport, mainly caused by over-speeding\n\nMention: \"Agriculture\nCensus 2011\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Census share of women among agricultural workers.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000053:3:0", "occurrence_id": "09d48386f9192de1a0a731f1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nHimachal Pradesh State Roads Transformation Program (P163328)\n\n\nspecialized services, including medical tourism, IT and education, as well as renewable energy generation, including hydro\npower. However, ensuring development effectiveness is a challenge in the absence of synergy among development\ninitiatives and institutional efficiency. The World Bank has been supporting agriculture, forestry, public finance, green\ngrowth, sustainable development and transport projects, but implementation has been sluggish due to institutional\ncapacity limitations.\n\n\nSectoral and Institutional Context\n\n4. **HP is endowed with abundant fruit belts and natural tourism potential, which could help the State to be a lead**\n**in green growth, if its transport infrastructure and logistics system are well developed.** The economic growth of the state\nis mainly attributed to agriculture/ horticulture and related activities as it provides employment to about 62 percent of\nthe total workers in the state. However, the roads connecting farm areas to the wholesale markets are substandard\n(narrow mountainous roads) and horticultural products are transported by informal cartels of short-bed truck operators.\nMoreover, the Himalayan mountains and valleys covered with the state record (SoER, HP) of 3,295 species of plants and\n5,721 species of fauna are attracting tourists to HP and make tourism account for seven percent of the State GDP. The\nnatural scenery in the Himalayan mountains, including the Rotana Pass is attracting enormous tourists. The tourism\nindustry is growing fast and the number of tourists in HP has increased from 7 million in 2005 to 13 million in 2010 and 18\nmillion in 2016. Students from all over India join the universities in Solan, which are providing high quality education in\nagriculture, IT and medicine, which makes HP a center for specialized services. Himachal’s hydropower plants generate\n7,000-Megawatt electricity. Recently, manufacturing plants, including pharmaceutical, cement, small steel mills and wool\nfactories are emerging. The absence of physical and digital logistics infrastructure for the\n\nMention: \"state record\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"inform\", \"summary\": \"State of environment record cited for species counts attracting tourism.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000058:14:1", "occurrence_id": "02a1cbc56f4a6d99e02b2cf3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  high vulnerability to climate change adds mounting pressure to its already strained urban**\n**infrastructure, which highlights the importance of focusing on strengthening urban resilience to climate change.**\nNepal ranks as the fourth most vulnerable country to climate change impacts (Climate Change Vulnerability Index, 2011)\ndue to its complex topography and variability in climate within short distances. The number and frequency of landslides,\nflash floods, and droughts has increased in recent years, and erratic weather patterns projected in climate models may\nexacerbate the impact of these disasters in the coming years. <sup>8</sup> Mean annual temperatures throughout Nepal are\nprojected to increase between 0.5°C and 2.0°C by the 2030s, <sup>9</sup> while projections show a three-fold increase in monsoon\nrainfall and more frequent summer floods. <sup>10</sup> The risk of flooding in the river basins of the non-Himalayan region due to\nhigher monsoon precipitation is projected to increase by as much as 14 to 40 percent by the 2030s. <sup>11</sup> The absence of\nguidelines and technical standards for good-quality infrastructure decreases municipalities’ resilience to climate\nhazards. <sup>12</sup> In a context of rapid urbanization, building resilience to climate change is crucial for the prosperity of Nepal’s\nurban areas. The GoN’s National Urban Development Strategy demonstrates a commitment to increasing the resilience,\nmitigation, and adaptation of urban areas to climate change effects.\n\n14. **Nepal has a long history of citizen engagement and participation, and the new decentralized governance**\n**model provides a unique opportunity to support the development of more participatory and accountable systems in**\n**local governments.** A plethora of local institutions, including UCs, cooperative organizations, traditional authorities,\nand local management committees have long existed in Nepal and have\n\nMention: \"Climate Change Vulnerability Index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Index ranking Nepal fourth most vulnerable to climate impacts.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000058:21:2", "occurrence_id": "f2955bf8479e57bc9da31c96", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  (according to the 2011 census) <sup>28</sup> through increased access to improved strategic\n\n\n26 Example of eligible activities include (i) regular maintenance of municipal infrastructure, including maintenance and cleaning of streets,\nroads and drainages, public spaces, parks, and community facilities; and (ii) construction of some basic infrastructure, including community\nwater facilities such as water distribution schemes, water kiosks, wells, and public washing facilities.\n27 The Project does not propose to support provincial governments currently as their role is still evolving within the new federal set-up. However, this will be\nreconsidered at the mid-term review and the project design will be adjusted accordingly to provide any technical support that provincial governments may require\nin the urban sector aligned with the PDO.\n28 Composed of approximately 1,568,500 residents in the 17 target municipalities under Component 1, and 20,000 beneficiaries under the COVID related activities\n(Component 3 specifically). Reference is made to the CBS data available [https://cbs.gov.np/wpcontent/upLoads/2018/12/Population_Ward_Level_753_Local_Unit.pdf].\n\n\nPage 11 of 55\n\nMention: \"2011 census\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Census population used to size beneficiary coverage.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000058:21:2", "occurrence_id": "1fae1f142926441cc10d6dbf", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  (according to the 2011 census) <sup>28</sup> through increased access to improved strategic\n\n\n26 Example of eligible activities include (i) regular maintenance of municipal infrastructure, including maintenance and cleaning of streets,\nroads and drainages, public spaces, parks, and community facilities; and (ii) construction of some basic infrastructure, including community\nwater facilities such as water distribution schemes, water kiosks, wells, and public washing facilities.\n27 The Project does not propose to support provincial governments currently as their role is still evolving within the new federal set-up. However, this will be\nreconsidered at the mid-term review and the project design will be adjusted accordingly to provide any technical support that provincial governments may require\nin the urban sector aligned with the PDO.\n28 Composed of approximately 1,568,500 residents in the 17 target municipalities under Component 1, and 20,000 beneficiaries under the COVID related activities\n(Component 3 specifically). Reference is made to the CBS data available [https://cbs.gov.np/wpcontent/upLoads/2018/12/Population_Ward_Level_753_Local_Unit.pdf].\n\n\nPage 11 of 55\n\nMention: \"CBS data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Ward-level population data used to count project beneficiaries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000059:12:0", "occurrence_id": "09d7256a2ba0c2b407d73384", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nPhilippines Customs Modernization Project (P163428)\n\n\n**Figure 2. The share of exports in GDP has been worsening**\n\n\nSource: World Development Indicators (2020)\n\n9. The country’s export competitiveness is impeded by high trade costs. The Doing Business 2020 report highlights\nthat trade costs in the Philippines are some of the highest within the Association of Southeast Asian Nations (ASEAN)\n(Figure 3). Investors in the Philippines face the highest import costs and the second highest export costs, just behind\nMyanmar. Results from the 2018 Logistics Performance Index highlight the low efficiency of the clearance process by\nborder agencies (2.53/5), including Customs, together with the low quality of trade and transport infrastructure (2.73/5).\nThe resulting high trade costs hinder the country’s export competitiveness and ability to create high quality jobs. This is\nespecially important as the Philippines handles the 26 <sup>th</sup> largest volume of container traffic in the world.\n\n\n**Figure 3. Investors in the Philippines pay some of the highest trade costs in ASEAN**\n\n\n\n\n\n\n\nSource: Doing Business 2020, World Bank Group\n\n\n10. The call for higher rates of export-led economic growth, improved revenue collection performance and\nsignificantly higher levels of integrity will continue to put Customs in the spotlight. Today, the trading community uses\njust-in-time supply chains to maximize competitive advantage, and it demands that border management agencies do\nnot disrupt those chains by employing outdated practices and procedures and imposing excessive red tape.\n\n\nPage 8 of 43\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Exports as share of GDP plotted from the world development indicator series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000059:43:0", "occurrence_id": "2580ea1b514c2305050fc7d6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nPhilippines Customs Modernization Project (P163428)\n\n\nprocedures and inspections from plant and veterinary agencies; (b) chemicals and pharmaceutical\nproducts (HS 28-38), which are subjected to fewer health and safety regulations; (c) raw materials and\ninputs such as mineral, wood and metal products, (HS 25-27, 39-49, and 68-83), which are assumed to\nhave fewer regulations; and (d) consumer and industrial products (HS 50-67 and 84-97), which are\nsubjected to some testing and technical requirements (such as labelling);\nb) The projected growth rate is calculated for all trade volumes and is based on the median growth rate for\n\n2012-2017 at around 3.6 percent to remain conservative and to consider the potential trade slow down\ndue to Covid-19. The Philippines has experienced substantial economic and therefore average trade\ngrowth rates are substantially higher at 9.7 percent.\nc) Cost savings from simplification and automation would be 1.6 percent of trade costs, which is half of the\n\n3.2 percent assumed savings from automation in ASEAN by the OECD study; <sup>14</sup> [^14: Organisation for Economic Co-Operation and Development (2015) “OECD Trade Facilitation Indicators: Moving ASEAN Forward”\nhttp://www.oecd.org/trade/facilitation/oecd-tfi-asean-july-2015.pdf.]\nd) The cost savings will be different for each product category as they are subjected to different levels of\n\nscrutiny: agricultural, fishery and food products will receive the full cost savings; chemicals and\npharmaceutical products will receive 60 percent of the cost savings; consumer and industrial products will\nreceive 30 percent of the cost savings; and raw materials and inputs will receive 10 percent of the cost\nsavings;\ne) The cost savings are also different for export and import flows. Imports are subjected to the full gamut of\n\ncustoms processes and procedures so they will receive the full cost savings, while exports are subjected\nto less inspections, so the cost savings are\n\nMention: \"OECD Trade Facilitation Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Study supplying the assumed automation savings percentages.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000061:20:1", "occurrence_id": "6f235c025fb81b7dca016b77", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: *the ERPA period.** This estimate was calculated after deducting a series of buffers to the ex-ante ERs\nestimates to account for uncertainty and reversal risk. <sup>11</sup> Gross ex ante ERs for the implementation period\nwere estimated at 3.5 million tCO2e, which represents a 43 percent reduction in emissions and enhances\nremovals by sinks from the business-as-usual estimates of the FRL. Table 2 details the ex-ante GHG\nemission reductions and removals of the ER Program.\n\n\n34. **Measuring and monitoring of emissions and/or removals against Fiji’s reference level will result**\n**in reporting emissions/removals every two years.** <sup>**12**</sup> Monitoring of activity data (for example, harvest\nvolumes and land use change) will occur annually. Landsat medium (10 m–30 m) resolution and remote\nsensing imagery will be used to identify the potential forest change areas, and ground surveys and/or\nhigh-resolution remote sensing imagery will be used for annual activity monitoring. For the monitoring of\nground data sources, emission factors are estimated from a combination of national data (for example,\nfrom National Forest Inventory [NFI] and permanent sample plots [PSPs] using the Intergovernmental\nPanel on Climate Change [IPCC] default values). The ERPA payments will take place according to a BSP\nagreed with beneficiaries before the ERPA signature.\n\n\n11 According to the buffer guidelines, a conservativeness factor of 8 percent deforestation and reforestation, a 15 percent\nuncertainty buffer for forest degradation (which relies on so-called proxy methods), and a 26 percent reversal risk buffer were\napplied to the estimated total expected emissions of the country under the program.\n12 This may change based on the results of the ERPA negotiation between the GoF and the FCPF.\n\n\nPage 16 of 92\n\nMention: \"activity data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"Harvest volumes and land use change monitored annually for the reference level.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000061:20:1", "occurrence_id": "2c19b4209b4b94364f0e788e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: *the ERPA period.** This estimate was calculated after deducting a series of buffers to the ex-ante ERs\nestimates to account for uncertainty and reversal risk. <sup>11</sup> Gross ex ante ERs for the implementation period\nwere estimated at 3.5 million tCO2e, which represents a 43 percent reduction in emissions and enhances\nremovals by sinks from the business-as-usual estimates of the FRL. Table 2 details the ex-ante GHG\nemission reductions and removals of the ER Program.\n\n\n34. **Measuring and monitoring of emissions and/or removals against Fiji’s reference level will result**\n**in reporting emissions/removals every two years.** <sup>**12**</sup> Monitoring of activity data (for example, harvest\nvolumes and land use change) will occur annually. Landsat medium (10 m–30 m) resolution and remote\nsensing imagery will be used to identify the potential forest change areas, and ground surveys and/or\nhigh-resolution remote sensing imagery will be used for annual activity monitoring. For the monitoring of\nground data sources, emission factors are estimated from a combination of national data (for example,\nfrom National Forest Inventory [NFI] and permanent sample plots [PSPs] using the Intergovernmental\nPanel on Climate Change [IPCC] default values). The ERPA payments will take place according to a BSP\nagreed with beneficiaries before the ERPA signature.\n\n\n11 According to the buffer guidelines, a conservativeness factor of 8 percent deforestation and reforestation, a 15 percent\nuncertainty buffer for forest degradation (which relies on so-called proxy methods), and a 26 percent reversal risk buffer were\napplied to the estimated total expected emissions of the country under the program.\n12 This may change based on the results of the ERPA negotiation between the GoF and the FCPF.\n\n\nPage 16 of 92\n\nMention: \"National Forest Inventory\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"National inventory and plots supplying emission factors.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000061:20:1", "occurrence_id": "c44560e1aef3880f09a9c808", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: *the ERPA period.** This estimate was calculated after deducting a series of buffers to the ex-ante ERs\nestimates to account for uncertainty and reversal risk. <sup>11</sup> Gross ex ante ERs for the implementation period\nwere estimated at 3.5 million tCO2e, which represents a 43 percent reduction in emissions and enhances\nremovals by sinks from the business-as-usual estimates of the FRL. Table 2 details the ex-ante GHG\nemission reductions and removals of the ER Program.\n\n\n34. **Measuring and monitoring of emissions and/or removals against Fiji’s reference level will result**\n**in reporting emissions/removals every two years.** <sup>**12**</sup> Monitoring of activity data (for example, harvest\nvolumes and land use change) will occur annually. Landsat medium (10 m–30 m) resolution and remote\nsensing imagery will be used to identify the potential forest change areas, and ground surveys and/or\nhigh-resolution remote sensing imagery will be used for annual activity monitoring. For the monitoring of\nground data sources, emission factors are estimated from a combination of national data (for example,\nfrom National Forest Inventory [NFI] and permanent sample plots [PSPs] using the Intergovernmental\nPanel on Climate Change [IPCC] default values). The ERPA payments will take place according to a BSP\nagreed with beneficiaries before the ERPA signature.\n\n\n11 According to the buffer guidelines, a conservativeness factor of 8 percent deforestation and reforestation, a 15 percent\nuncertainty buffer for forest degradation (which relies on so-called proxy methods), and a 26 percent reversal risk buffer were\napplied to the estimated total expected emissions of the country under the program.\n12 This may change based on the results of the ERPA negotiation between the GoF and the FCPF.\n\n\nPage 16 of 92\n\nMention: \"permanent sample plots\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Sample plots combined with inventory data for emission factors.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000061:49:4", "occurrence_id": "96e0e5c230d26ca074daac25", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: br>**land area**<br>**does not**<br>**include**<br>**group of**<br>**islands**|**1,636,557**|**89%**|**89%**|**884,887.00**|**97%**|**97%**|\n|**TOTAL FIJI**<br>**WIDE**|**1,837,600**||||||\n\n\n\n_Source:_ MoF 2016 Key Statistics total population of Fiji is 884,887.\n\n\n\nP\n\nMention: \"MoF 2016 Key Statistics\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Finance ministry statistics behind the population totals in the table.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000061:62:2", "occurrence_id": "8bcf6bcd7b11c57608f6fb68", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: F**<br>Collaborators:<br>Ministry of iTaukei<br>Affairs<br>District REDD+ WG<br>iTaukei Lands Trust<br>Board<br>Provincial Council<br>NGO, CSOs<br>Sawmillers Association|Govt.<br>Funding<br>|\n|**Subcomponent 1.3 Forest information system**|**Subcomponent 1.3 Forest information system**|**Subcomponent 1.3 Forest information system**|**Subcomponent 1.3 Forest information system**|\n|1.3.1. Upgrade Forest<br>information & data<br>base systems<br>|• System Upgrade in YR 1<br>• One training & refresher course per year<br>for the MoF staff on FIS processing|**Lead Agency: MoF**<br>Collaborators:<br>Ministry of iTaukei<br>Affairs<br>District REDD+ WG<br>iTaukei Lands Trust<br>Board<br>NGO, CSOs<br>Sawmillers Association|Govt.<br>Funding<br>|\n|1.3.2 improved<br>monitoring and<br>reporting to feed<br>forest information<br>system<br>|<br>• One report per annum on compliance to<br>Environment and social safeguards<br>• Divisional REDD+ WG Quarterly<br>Monitoring reports|**Lead Agency: MoF**<br>Collaborators:<br>Ministry of iTaukei<br>Affairs<br>District REDD+ WG|Govt.<br>Funding<br>|\n\n\n\nP\n\nMention: \"Monitoring reports\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"Quarterly divisional reports tracking safeguards compliance.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000068:10:0", "occurrence_id": "084a515c7e2016fd30ddd4a5", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nMaldives Urban Development and Resilience Project (P163957)\n\n\nthe country. Large-scale migration to the Malé region <sup>1</sup> [^1: Approximately 53 percent of the resident population of Malé is made up of migrants from other atolls, and Malé has seen an increase in\nits total Maldivian population share between the 2006 and 2014 Census (from 34.7 percent to 38.8 percent)] - while most atolls has experienced a shrinkage due to\noutmigration <sup>2</sup> [^2: Spellman, L.H, 2015. _Empirical Analyses of Migration in Small Islands: The Role of Environment and Social Factors._] - has led to severe overcrowding that is aggravating social and environmental vulnerabilities of\ninhabitants. Malé is among the densest cities in the world, with 177,500 residents housed on a 1.95 km <sup>2</sup> island <sup>3</sup> [^3: Government of Maldives National Bureau of Statistics, 2019 Population Projection] .\nIn the outer atolls, small size of population and distance to markets are significant barriers to sustainable service\ndelivery. Many people living in the outer atolls do not have piped water supply or sewer networks, while\nemployment and educational opportunities are limited.\n\n6. **To address demographic pressures in Greater Malé, the Government of Maldives (GoM) has pursued a**\n**two-pronged approach:**\n\n\n\na. **A focus on the sustainable development of the Greater Malé region considering existing migration**\n\n**trends.** In parallel to strengthening urban services in Malé, the creation of a reclaimed island, Hulhumalé,\nwas planned with the objective of reducing congestion in Malé and meeting development demands.\nHulhumalé is the first planned city in the country. Its planning and design incorporate climate adaptation\nconsiderations and the 400-hectare landmass sits approximately two meters above mean sea level.\nReclamation of Hulhumalé Phase 1 (160 hectares) was\n\nMention: \"2006 and 2014 Census\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Two census rounds showing Malé's rising share of the national population.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000068:10:0", "occurrence_id": "725dd204ba9be8db4ccdf5e6", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nMaldives Urban Development and Resilience Project (P163957)\n\n\nthe country. Large-scale migration to the Malé region <sup>1</sup> [^1: Approximately 53 percent of the resident population of Malé is made up of migrants from other atolls, and Malé has seen an increase in\nits total Maldivian population share between the 2006 and 2014 Census (from 34.7 percent to 38.8 percent)] - while most atolls has experienced a shrinkage due to\noutmigration <sup>2</sup> [^2: Spellman, L.H, 2015. _Empirical Analyses of Migration in Small Islands: The Role of Environment and Social Factors._] - has led to severe overcrowding that is aggravating social and environmental vulnerabilities of\ninhabitants. Malé is among the densest cities in the world, with 177,500 residents housed on a 1.95 km <sup>2</sup> island <sup>3</sup> [^3: Government of Maldives National Bureau of Statistics, 2019 Population Projection] .\nIn the outer atolls, small size of population and distance to markets are significant barriers to sustainable service\ndelivery. Many people living in the outer atolls do not have piped water supply or sewer networks, while\nemployment and educational opportunities are limited.\n\n6. **To address demographic pressures in Greater Malé, the Government of Maldives (GoM) has pursued a**\n**two-pronged approach:**\n\n\n\na. **A focus on the sustainable development of the Greater Malé region considering existing migration**\n\n**trends.** In parallel to strengthening urban services in Malé, the creation of a reclaimed island, Hulhumalé,\nwas planned with the objective of reducing congestion in Malé and meeting development demands.\nHulhumalé is the first planned city in the country. Its planning and design incorporate climate adaptation\nconsiderations and the 400-hectare landmass sits approximately two meters above mean sea level.\nReclamation of Hulhumalé Phase 1 (160 hectares) was\n\nMention: \"2019 Population Projection\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Projection cited for resident population density.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000068:17:0", "occurrence_id": "8c662180ffc39ef8aad9a2ad", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nMaldives Urban Development and Resilience Project (P163957)\n\n\nthe effectiveness of the EOCC and building permitting system.\n\n31. **Climate Co-Benefits.** The Project will contribute to both adaptation and mitigation Co-Benefits. Upon\nscreening for climate and disaster risks, the climate Co-Benefits are estimated at 52%. Project investments will\nmitigate present and future climate change-related risks and impacts—particularly heavy rainfall and monsoon\nfloods, coastal erosion, and sea level rise—by introducing adaptation measures in infrastructure investments\n(rainwater harvesting and storage, STP and emergency management systems under Component 1) and provide\ntechnical inputs to strengthen resilience of future investments in selected cities (analytical studies related to\nstormwater drainage and sustainable urban planning under Component 2). Climate risks will also be addressed\nduring the preparation phase by factoring precipitation estimates into facility design.\n\n\n**C. Project Beneficiaries**\n\n32. **The primary beneficiaries of the proposed operation will be the residents of Malé and Hulhumalé Phase**\n**1, estimated to be 176,749** <sup>**14**</sup> [^14: Calculated based on HDC population figures for Hulhumalé and Bureau of Statistics 2019 projection for resident Maldivian population.\nResident foreign population is excluded as data are not as accurate.] **.** Residents and tourists will benefit from access to improved, more resilient urban\nservices including emergency response and urban infrastructure. Improved urban planning will benefit the\npopulations of the regions covered by feasibility studies and stormwater masterplans (Greater Malé) and regional\ndevelopment plans (indicatively Addu, Thiladhunmathi, and Faadhippolhu). Strengthened emergency response\nsystems will benefit the entire country. The Project will support institutional strengthening and capacity\ndevelopment of GoM agencies including MNPI, Ministry of Finance (MoF), Ministry of Housing and Urban\nDevelopment (MHUD), HDC, MWSC, Maldives Fire and Rescue Service, and NDMA.\n\n\n14 Calculated based on HDC population figures for Hulhumal\n\nMention: \"HDC population figures for Hulhumalé\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Housing development corporation figures used to estimate project residents.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000069:49:0", "occurrence_id": "a22b16966a56ca028ae40376", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nMarine Fisheries and Socio-Economic Development Project (P163980)\n\n\n\n\n\n|Col1|delinquents).|Col3|Col4|Col5|Col6|\n|---|---|---|---|---|---|\n|Proportion of<br>nearshore waters<br>under improved<br>management|The indicator is measured as the area (km2) within<br>the JCMAs developed and supported by the<br>project, as a % of all nearshore waters.<br> <br>Total area under nearshore waters is (including<br>territorial seas to 12nm) is 10,860 Km2.<br> <br>‘Improved management’ to be assessed against<br>three criteria where all three must be met: 1) JCMA<br>plan prepared and approved by national and<br>county authorities; 2) JCMA plan includes, at a<br>minimum: (i) defined boundaries agreed by<br>communities; (ii) at least one no-take area agreed<br>by communities; and (iii) management measures<br>for at least two priority fisheries that explicitly<br>control fishing effort (spatial or seasonal closure;<br>gear restrictions, gear or fisher numbers etc.); and<br>3) BMU surveillance reports indicate a stable or<br>improving trend in non-compliance per patrol<br>effort.<br> <br>Cumulative targets.|Annual<br>|New and revised<br>JCMA<br>management<br>plans and BMU<br>surveillance<br>reports<br>|Review of new and<br>revised JCMA<br>management plans<br\n\nMention: \"BMU surveillance reports\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Surveillance reports indicating compliance trends per patrol effort.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000069:54:0", "occurrence_id": "aacfa1a002d2e15fa1b15b9b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nMarine Fisheries and Socio-Economic Development Project (P163980)\n\n\n\n\n\n\n\n\n\n\n\n\n\n|Col1|Cumulative targets|Col3|Col4|Col5|Col6|\n|---|---|---|---|---|---|\n|BMUs achieving a<br>minimum of 4<br>performance targets|The indicator measures the number of BMUs, out<br>of approx. 84 along the KE coast, achieving at least<br>4 annual performance targets: 1) at least 6 out of<br>12 monthly meetings of BMU Executive Committee<br>are conducted and minutes submitted to, and<br>available at, county fisheries department; 2) at<br>least 2 BMU members’ assembly meetings<br>conducted and minutes submitted to, and available<br>at, county fisheries department; 3) BMU has an<br>annually updated register of fishers and vessels,<br>with at least 75% of fishers having licenses; 4)<br>members of BMU monitoring, control and<br>surveillance (MCS) sub-committee participate in at<br>least one patrol, in at least 6 out of 12 months; 5)<br>at least 65% of annual required budget is received<br>in revenue for the previous calendar year; and 6)<br>BMU has collected and submitted at least 80% of<br>the target number of catch records for catch<br>monitoring in at least 10 out of 12 months.<br> <br>Cumulative targets.|Annual<br>|Field survey<br>|Field survey<br>|NPCU M&E function<br>|\n|Operational<br>surveillance activities<br>in Kenyan marine<br\n\nMention: \"register of fishers and vessels\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"Annually updated register used to verify licensing coverage.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000070:56:0", "occurrence_id": "f7d442f438652d571147ac71", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nMozambique National Urban Development and Decentralization Project (P163989)\n\n\n**ANNEX** **3:** **Performance Grants**\n\n\n**COUNTRY: Mozambique**\n**National Urban Development and Decentralization Project**\n\n\n1. This activity will support eligible municipalities through performance-based grants that will\nfinance urban infrastructure and basic services investments and institutional development. The grants\nprovided in this subcomponent are designed to complement the basic design of the GoM’s system of\nintergovernmental transfers to municipalities (general purpose grants [FCA] and capital grants). GoM\nacknowledges the need to reform its current transfer system, which suffers from various shortcomings. <sup>14</sup> [^14: _Horizontal Funds Distribution to Subnational Government in Mozambique_, Report No: AUS0000186.]\nDue to the shortcomings of the existing transfer system, whose reform, supported by this project, is only\nstarting, the performance grant will not employ the allocation formula of existing GoM grants, and will be\ntreated as a project grant with the expectation that its performance focus will produce a demonstration\neffect that could later on be scaled up at the national level.\n\n\n**Per Capita Formula**\n\n\n2. Several options have been analyzed in determining the horizontal structure of the grant. The first\noption was to construct a basic allocation formula based on urban infrastructure needs. However, it was\ndetermined that establishing the need for infrastructure would require data on access to services that is\ncurrently not available. This option would also require a complex formula, thus possibly making it less\ntransparent and more complex to calculate at this point. The second option is to base the allocation\nexclusively on the size of the population of the local entities. The exact size of the per capita allocation\nwould then depend on the size of the annual grant pool and the total population of each local entity\neligible for the performance grant. Official census data from 2017 have been used to calculate grant\namounts for the participating municipalities.\n\n\n**Geographic Scope\n\nMention: \"census data from 2017\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Official population figures used to allocate performance grants.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000070:71:1", "occurrence_id": "a6def4c832d37bb16e1e3329", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  are responsible for the maintenance of unclassified roads (trunk and other).\nHowever, this data is currently unavailable, so data from classified roads under the responsibility of the\nNational Roads Agency will be used as a proxy. The cost-benefit analysis assumes that average travel time\nbefore the rehabilitation of classified roads under the responsibility of National Roads Administration\n( _Administração Nacional de Estradas_, ANE) for Xai-Xai reduced by half, from 60 min to 30 min for each\n40km traveled.\n\n\n13. The benefits were related to the savings of travel time, which bring savings on the value of time\nand reduction of vehicle operation. To estimate the benefits, the following information was used: a) daily\ntraffic count before and after the project; b) average travel time before and after the project; c) value of\ntime, and d) vehicle operating cost.\n\n\n_Daily Traffic Count_\n\n\n14. For the _without_ project scenario, the information of daily traffic was based on figures used at\nappraisal per alternate route used for diverting traffic for the recently closed Bank-financed MMDP\n(US$50 million). Given that Xai-Xai’s population is 132,884 (2017 census) or approximately 10 percent of\nMaputo’s population, the analysis uses ten percent of the traffic count for Maputo and applies it to XaiXai. The traffic was counted on March 2010 according to type of vehicle. For the _with_ project scenario,\nthe figure was obtained from Maputo’s Master Plan of Urban Transport conducted by Nikon Koei in March\n\n\n20 All numbers use 2010 prices, using an exchange rate of MZN 34 to US$1.\n\n\nPage 68 of 79\n\nMention: \"daily\ntraffic count\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Before and after traffic counts used to estimate time savings.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000070:71:1", "occurrence_id": "537f27fb7b15dea54b6a5f94", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  are responsible for the maintenance of unclassified roads (trunk and other).\nHowever, this data is currently unavailable, so data from classified roads under the responsibility of the\nNational Roads Agency will be used as a proxy. The cost-benefit analysis assumes that average travel time\nbefore the rehabilitation of classified roads under the responsibility of National Roads Administration\n( _Administração Nacional de Estradas_, ANE) for Xai-Xai reduced by half, from 60 min to 30 min for each\n40km traveled.\n\n\n13. The benefits were related to the savings of travel time, which bring savings on the value of time\nand reduction of vehicle operation. To estimate the benefits, the following information was used: a) daily\ntraffic count before and after the project; b) average travel time before and after the project; c) value of\ntime, and d) vehicle operating cost.\n\n\n_Daily Traffic Count_\n\n\n14. For the _without_ project scenario, the information of daily traffic was based on figures used at\nappraisal per alternate route used for diverting traffic for the recently closed Bank-financed MMDP\n(US$50 million). Given that Xai-Xai’s population is 132,884 (2017 census) or approximately 10 percent of\nMaputo’s population, the analysis uses ten percent of the traffic count for Maputo and applies it to XaiXai. The traffic was counted on March 2010 according to type of vehicle. For the _with_ project scenario,\nthe figure was obtained from Maputo’s Master Plan of Urban Transport conducted by Nikon Koei in March\n\n\n20 All numbers use 2010 prices, using an exchange rate of MZN 34 to US$1.\n\n\nPage 68 of 79\n\nMention: \"2017 census\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Census population used to scale traffic counts to Xai-Xai.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000070:72:0", "occurrence_id": "53d0edcfe80d3b1d1ad4e20c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nMozambique National Urban Development and Decentralization Project (P163989)\n\n\n2014 and corroborated with figures from the Direction of Transportation and Traffic of the Maputo\nMunicipal Council in August 2016 conducted at time that the implementation of traffic lights was being\nstudied.\n\n\n**Table 4.2. Daily Traffic before and After the Rehabilitation of Roads in Xai-Xai**\n\n|Col1|Category of Vehicles|Col3|Col4|Col5|Col6|\n|---|---|---|---|---|---|\n||**Buses**|**Semi-collective**|**Light Vehicle**|**Trucks**|**Total**|\n|After Intervention|54|702|964|113|1,833|\n|Before Intervention (discounting by 3% annually)|47|618|848|99|1,612|\n\n\n\n15. The traffic count at time of appraisal was measured in specific points of the alternate routes. It\nshowed that the number of vehicles per day using these routes was in the range of 1,612 and 1,833. The\nvalue of time is based on figures presented in the Master Plan of Urban Transport for Maputo, which was\nestimated based on the per capita GDP in Maputo city. It uses the growth rate of the per capita GDP to\nproject the value of time up to 2035. This evaluation uses the value of time for 2012 transformed to 2010\nprices. No increase was included in the evaluation, even though it will increase with the GDP growth rate.\nHowever, given the uncertainties of growth rate on the economy and the much smaller size of Xai-Xai\ncompared to Maputo, this evaluation kept the value constant to work on the conservative side.\n\n\n**Table 4.3. Value of Traffic Time**\n\n|Col1\n\nMention: \"figures from the Direction of Transportation and Traffic\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"validate\", \"summary\": \"Municipal traffic figures corroborating the 2014 counts.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000074:3:1", "occurrence_id": "9e10034921b17dfcb0bb38e0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  these priorities of the PNDH by directly supporting the poorest in the\nDry Corridor, including women and youth.\n\n\n**Nicaragua is highly vulnerable to extreme weather events including hurricanes, tropical storms, droughts,**\n**and seismic activity** . The country ranks 4 out of 177 countries most affected by extreme weather events\n(Global Climate Risk Index, 2014). The Caribbean coast is most vulnerable to hydro-meteorological events,\nand the Pacific and Central regions are typically more exposed to floods and landslides. Drought risk is\nconcentrated in the Dry Corridor, which includes many departments of the Central region, and covers about\n28 percent of the territory. Climate variability will continue to have a large impact on vulnerable rural poor in\nthese areas concentrated in agricultural labor, due to impacts on productivity through to crop loss, loss of\nassets, impacts on household and child nutrition, child development and human capital accumulation.\n\n\n<u>Sectoral and Institutional Context</u>\n\n**Agriculture in Nicaragua remains the main engine of economic growth and poverty reduction.** The sector\nrepresents 17 percent of GDP, in contrast to the LAC average of 5.5 percent of the GDP, and 70 percent of total\nexports are of primary products (Nicaragua SCD, 2017). Per a recent study of the Nicaragua agricultural sector,\nthe contribution of agriculture to GDP reaches 27 percent if forward and backward linkages are\nconsidered (Agriculture in Nicaragua: Performance, Challenges, and Options, November 2015, World Bank). The\nsector accounts for 30 percent of total labor force employment, double the regional average of 15 percent, and\nis the main source of livelihoods for 80 percent of rural households. Agriculture also accounts for roughly 80\npercent of poverty reduction in rural areas. A sectoral decomposition of the contribution to poverty reduction\nin the 2017 Nicaragua SCD showed that the increase in agriculture earnings explained most of the decline in\n<u>rural poverty between 2005 and 2014. Subsistence and\n\nMention: \"Nicaragua SCD\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Diagnostic cited for agriculture's share of GDP and exports.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000074:3:2", "occurrence_id": "481aa1eeace06ab458b4c77b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  roughly 80\npercent of poverty reduction in rural areas. A sectoral decomposition of the contribution to poverty reduction\nin the 2017 Nicaragua SCD showed that the increase in agriculture earnings explained most of the decline in\n<u>rural poverty between 2005 and 2014. Subsistence and transitional family farms make up 81 percent of total</u>\n\n\nJun 24, 2020 Page 4 of 19\n\nMention: \"2017 Nicaragua SCD\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Sectoral decomposition of rural poverty reduction contributions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000074:5:0", "occurrence_id": "66dc77400c5e228bc128b704", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nNicaragua Dry Corridor Nutrition-Sensitive Agriculture Project\n\n\nsix months. Climate variability has devastating consequences on the cultivation of basic grain crops, which are\npart of the region’s subsistence agriculture, as well as on other smallholder semi-commercially oriented\nagriculture and cattle-farming activities.\n\n\n**The Dry Corridor is increasingly vulnerable to climate change and climate variability.** Historical climate data\nanalyzed by the International Center for Tropical Agriculture (CIAT by its acronym in Spanish) found that, for\ninstance, there is a strong warming trend across the country, manifested through diurnal temperature\nincreases (~0.40 °C per decade) in deforested areas. These rates are more than five percent higher than average\ntemperature change rates in tropical areas. Rising temperatures and more frequent droughts and floods will\npresent a major challenge for the country’s production systems by 2030[1]. This broader trend in climate\naffects climate variability, which significantly reduces the availability of water resources for crop and livestock\nproduction resulting in substantial economic losses to agricultural production and productivity [2].\n\n\n**Water resources are abundant throughout Nicaragua, however weak water management and inadequate**\n**infrastructure for water regulation and irrigation are negatively impacting agricultural productivity.** Overall,\nthe eastern slopes flowing towards the Caribbean coast are the wettest areas in the country, compared to the\ndryer western slopes (Dry Corridor) flowing towards the Pacific coast where most population live. Agriculture is\nthe largest user of water in Nicaragua, consuming 79 percent of the total. At present, only five percent of\npermanent cropland, about 99,400 hectares (ha) out of two million hectares is irrigated (Barzev, Radoslav:\nNicaragua: _Uso, manejo y valoración del agua en el sector agropecuario nacional_ ; 2013).\n\n\n**Women and female youth in Nicaragua are often disadvantaged in terms of access to inputs, control of**\n**output\n\nMention: \"Historical climate data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Historical record analysed for warming trends in the Dry Corridor.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000075:13:3", "occurrence_id": "e693ce54b8ef1f5cccab71b4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  rainy season would increase flooding, harming not</mark>\n<mark>only crops but roads (which are mainly unpaved), which would reduce the mobility of people and goods,</mark>\nincrease travel times, and raise transport costs.\n\n\n<mark>9.</mark> **<mark>Growth in the agricultural sector has been modest, driven mostly by an expansion of cultivated area</mark>**\n**<mark>rather than intensification</mark>** <mark>. Long-term annual agricultural growth is estimated at 3 percent, consistent with the</mark>\n\n\n9 Groupement SHER Ingénieurs - Conseils-TROPIC Environnement (2018), “PASAG- Schéma Directeur de Développement de l’Irrigation” (Namur,\nAugust).\n10 République de Guinée – Revue du Secteur Agricole. MINAGRI-Banque Mondiale, Janvier 2018.\n11 WBG (2019), “Guinea: The Economic Benefits of a Gender Inclusive Society” (Washington, DC, http://documents.worldbank.org/\ncurated/en/607191569339066369/Guinea-The-Economic-Benefits-of-a-Gender-Inclusive-Society).\n12 République de Guinée (2014), Recensement Général de la Population\n13 Maiga M (2020), Etude sur le Gender Tag pour le Projet de Développement de l’Agriculture Commerciale de Guinée\n14 OECD (2014), “Guinea Social Institutions and Gender Index 2014 Survey” (Paris, Organization for Economic Co-operation and Development).\n15 WBG (2019), “Guinea: The Economic Benefits of a Gender Inclusive Society” (Washington, DC, http://documents.worldbank.org/\ncurated\n\nMention: \"Recensement Général de la Population\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"General population census cited in the footnotes.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000075:13:3", "occurrence_id": "630457cbd15f6024eb564c98", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  rainy season would increase flooding, harming not</mark>\n<mark>only crops but roads (which are mainly unpaved), which would reduce the mobility of people and goods,</mark>\nincrease travel times, and raise transport costs.\n\n\n<mark>9.</mark> **<mark>Growth in the agricultural sector has been modest, driven mostly by an expansion of cultivated area</mark>**\n**<mark>rather than intensification</mark>** <mark>. Long-term annual agricultural growth is estimated at 3 percent, consistent with the</mark>\n\n\n9 Groupement SHER Ingénieurs - Conseils-TROPIC Environnement (2018), “PASAG- Schéma Directeur de Développement de l’Irrigation” (Namur,\nAugust).\n10 République de Guinée – Revue du Secteur Agricole. MINAGRI-Banque Mondiale, Janvier 2018.\n11 WBG (2019), “Guinea: The Economic Benefits of a Gender Inclusive Society” (Washington, DC, http://documents.worldbank.org/\ncurated/en/607191569339066369/Guinea-The-Economic-Benefits-of-a-Gender-Inclusive-Society).\n12 République de Guinée (2014), Recensement Général de la Population\n13 Maiga M (2020), Etude sur le Gender Tag pour le Projet de Développement de l’Agriculture Commerciale de Guinée\n14 OECD (2014), “Guinea Social Institutions and Gender Index 2014 Survey” (Paris, Organization for Economic Co-operation and Development).\n15 WBG (2019), “Guinea: The Economic Benefits of a Gender Inclusive Society” (Washington, DC, http://documents.worldbank.org/\ncurated\n\nMention: \"Guinea Social Institutions and Gender Index 2014 Survey\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"OECD survey cited on social institutions in Guinea.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000075:54:0", "occurrence_id": "b271a912db892988649d7e78", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nGuinea Commercial Agriculture Development Project (P164184)\n\n\n\n\n\n|Col1|component beneficiaries)|Col3|Collection<br>System|Col5|Col6|\n|---|---|---|---|---|---|\n|Increase in volume of agri-food products<br>commercialized by project supported<br>beneficiaries|Quantity of marketed<br>agricultural and agri-food<br>products (for export or<br>domestic markets) by<br>project supported<br>beneficiaries in targeted<br>value-chains|bi-annually<br>|M&E data<br>system<br>|registry from MINAGRI<br>and Ministry of<br>Commerce<br>|PCIU<br>|\n|of which volumes exported|volumes of agri-food<br>products exported|bi-annually<br>|registry of<br>exports<br>|registry from customs<br>|PCIU<br>|\n|Firms benefiting from private sector<br>initiatives||bi-annually<br>|M&E system,<br>DFF monitorin<br>g systems on<br>Windows A &<br>B <br>|Registry of firms<br>benefiting from Project<br>support in data<br>collection system<br>|PCIU<br>|\n|Amount of private investments in<br>commercial agriculture mobilized through<br>Program's support (US$)|Amount of private<br>investments in commercial<br>agriculture mobilized<br>through Program's support<br>(US$)|bi-annually<br>|Monitoring<br>and\n\nMention: \"registry from MINAGRI\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"Agriculture ministry registry feeding the sales volumes indicator.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000075:54:0", "occurrence_id": "13d66655462b4adb8d583111", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nGuinea Commercial Agriculture Development Project (P164184)\n\n\n\n\n\n|Col1|component beneficiaries)|Col3|Collection<br>System|Col5|Col6|\n|---|---|---|---|---|---|\n|Increase in volume of agri-food products<br>commercialized by project supported<br>beneficiaries|Quantity of marketed<br>agricultural and agri-food<br>products (for export or<br>domestic markets) by<br>project supported<br>beneficiaries in targeted<br>value-chains|bi-annually<br>|M&E data<br>system<br>|registry from MINAGRI<br>and Ministry of<br>Commerce<br>|PCIU<br>|\n|of which volumes exported|volumes of agri-food<br>products exported|bi-annually<br>|registry of<br>exports<br>|registry from customs<br>|PCIU<br>|\n|Firms benefiting from private sector<br>initiatives||bi-annually<br>|M&E system,<br>DFF monitorin<br>g systems on<br>Windows A &<br>B <br>|Registry of firms<br>benefiting from Project<br>support in data<br>collection system<br>|PCIU<br>|\n|Amount of private investments in<br>commercial agriculture mobilized through<br>Program's support (US$)|Amount of private<br>investments in commercial<br>agriculture mobilized<br>through Program's support<br>(US$)|bi-annually<br>|Monitoring<br>and\n\nMention: \"registry from customs\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"Customs registry of exports behind exported volumes.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000076:10:1", "occurrence_id": "0e105bcebcf52efe7626caaa", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: sup>\n\n\n6. **Limited structural economic transformation continues to make the country vulnerable to**\n**external shocks.** The country’s growth continues to be driven by two main sectors – mining and\nagriculture. Agriculture is the largest single contributor to GDP, accounting for more than 50 percent of\nGDP over the last 4 years, while employing more than 58 percent of Sierra Leone’s labor force, compared\nto 35.4 percent in services and 5.7 percent in industry (which includes mining) <sup>6</sup> [^6: Kiendrebeogo,Youssouf; Mansaray,Kemoh. 2019. Sierra Leone - Economic Diversification Study. Washington, D.C :\n[WBG.https://hubs.worldbank.org/docs/imagebank/pages/docprofile.aspx?nodeid=31681723](https://hubs.worldbank.org/docs/imagebank/pages/docprofile.aspx?nodeid=31681723)] . Labor productivity in the\n\n\n3 World Bank national accounts data, and OECD National Accounts data files.\n_[https://data.worldbank.org/indicator/NY.GNP.PCAP.CD?locations=SL](https://data.worldbank.org/indicator/NY.GNP.PCAP.CD?locations=SL)_ .\n4 World Bank Data (2019) Labor force participation rate, total (percent of total population ages 15-64) (modeled ILO estimate) Fragile and conflict affected situations.\n_[https://data.worldbank.org/indicator/SL.TLF.ACTI.ZS?locations=F1](https://data.worldbank.org/indicator/SL.TLF.ACTI.ZS?locations=F1)_ .\n5 The Sierra Leone Integrated Surveys of 2011 and 2018 report poverty rates for the country as 52.9 percent and 56.8 percent\nrespectively. However, a change in methodology and setting of a\n\nMention: \"World Bank national accounts data\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"National accounts series behind income and productivity figures.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000076:10:1", "occurrence_id": "e2b6c2c91d487e43e3303385", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: sup>\n\n\n6. **Limited structural economic transformation continues to make the country vulnerable to**\n**external shocks.** The country’s growth continues to be driven by two main sectors – mining and\nagriculture. Agriculture is the largest single contributor to GDP, accounting for more than 50 percent of\nGDP over the last 4 years, while employing more than 58 percent of Sierra Leone’s labor force, compared\nto 35.4 percent in services and 5.7 percent in industry (which includes mining) <sup>6</sup> [^6: Kiendrebeogo,Youssouf; Mansaray,Kemoh. 2019. Sierra Leone - Economic Diversification Study. Washington, D.C :\n[WBG.https://hubs.worldbank.org/docs/imagebank/pages/docprofile.aspx?nodeid=31681723](https://hubs.worldbank.org/docs/imagebank/pages/docprofile.aspx?nodeid=31681723)] . Labor productivity in the\n\n\n3 World Bank national accounts data, and OECD National Accounts data files.\n_[https://data.worldbank.org/indicator/NY.GNP.PCAP.CD?locations=SL](https://data.worldbank.org/indicator/NY.GNP.PCAP.CD?locations=SL)_ .\n4 World Bank Data (2019) Labor force participation rate, total (percent of total population ages 15-64) (modeled ILO estimate) Fragile and conflict affected situations.\n_[https://data.worldbank.org/indicator/SL.TLF.ACTI.ZS?locations=F1](https://data.worldbank.org/indicator/SL.TLF.ACTI.ZS?locations=F1)_ .\n5 The Sierra Leone Integrated Surveys of 2011 and 2018 report poverty rates for the country as 52.9 percent and 56.8 percent\nrespectively. However, a change in methodology and setting of a\n\nMention: \"OECD National Accounts data files\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"Companion national accounts file cited with the World Bank series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000076:10:1", "occurrence_id": "a470979eb7e97c721c545a5d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: sup>\n\n\n6. **Limited structural economic transformation continues to make the country vulnerable to**\n**external shocks.** The country’s growth continues to be driven by two main sectors – mining and\nagriculture. Agriculture is the largest single contributor to GDP, accounting for more than 50 percent of\nGDP over the last 4 years, while employing more than 58 percent of Sierra Leone’s labor force, compared\nto 35.4 percent in services and 5.7 percent in industry (which includes mining) <sup>6</sup> [^6: Kiendrebeogo,Youssouf; Mansaray,Kemoh. 2019. Sierra Leone - Economic Diversification Study. Washington, D.C :\n[WBG.https://hubs.worldbank.org/docs/imagebank/pages/docprofile.aspx?nodeid=31681723](https://hubs.worldbank.org/docs/imagebank/pages/docprofile.aspx?nodeid=31681723)] . Labor productivity in the\n\n\n3 World Bank national accounts data, and OECD National Accounts data files.\n_[https://data.worldbank.org/indicator/NY.GNP.PCAP.CD?locations=SL](https://data.worldbank.org/indicator/NY.GNP.PCAP.CD?locations=SL)_ .\n4 World Bank Data (2019) Labor force participation rate, total (percent of total population ages 15-64) (modeled ILO estimate) Fragile and conflict affected situations.\n_[https://data.worldbank.org/indicator/SL.TLF.ACTI.ZS?locations=F1](https://data.worldbank.org/indicator/SL.TLF.ACTI.ZS?locations=F1)_ .\n5 The Sierra Leone Integrated Surveys of 2011 and 2018 report poverty rates for the country as 52.9 percent and 56.8 percent\nrespectively. However, a change in methodology and setting of a\n\nMention: \"Sierra Leone Integrated Surveys\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Two survey rounds reporting national poverty rates despite methodology change.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000076:14:0", "occurrence_id": "bd842da22b1ef0429c1b9996", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nSierra Leone Economic Diversification Project (P164212)\n\n\n\nFigure 5: Investment Climate Constraints:\n_% of firms identifying as major or very severe_\n\n\n\nFigure 6: Investment Climate Constraints in Sierra Leone:\n\n_% of firms identifying as major or very severe by gender_\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nSource: World Bank Enterprise Survey 2017. Source: WeFi Tourism Enterprise Survey (2019).\n\n15. **Sierra Leone’s private sector is comprised of MSMEs which are mostly informal, with few formal**\n**companies** . While the Corporate Affairs Commission and the Office of Administrator and Registrar\nGeneral (OARG) report 6,000 registered companies, databases across various Ministries, Department and\nAgencies (MDAs) as well as private sector industry associations point to a much larger number of mainly\nmicro and small-scale businesses of 1-5 employees in the informal sector <sup>16</sup> [^16: SME Database 2017, Ministry of Trade and Industry.], who tend to operate at the\nsurvival level. The Credit Registry of the Central Bank of Sierra Leone reports a total of only 26,000\naccounts for the entire country, including accounts owned by sole proprietors <sup>17</sup> [^17: Sierra Leone, Getting Credit indicator, Doing Business 2020.] . Large scale FDI, while\nbeginning to take an interest in Sierra Leone again after the country was declared Ebola free in March\n2016, remains low (see Figure 2).\n\n\n16. **Sierra Leone’s performance in key global indexes reflects the challenging business environment**\n**in the country** . Sierra Leone ranks 134th out of 140 countries in the World Economic Forum (WEF) Global\nCompetitiveness rankings of 2018 (Figure 5). The country is underperforming the SSA average on every\ncomponent of the index. In terms of the World Bank’s Doing Business ranking (Figure 6), Sierra Leone is\nranked 163 out of 190 economies in 2020, which is equal to its\n\nMention: \"World Bank Enterprise Survey 2017\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Firm survey behind constraints reported by gender.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000076:14:0", "occurrence_id": "98d9da79cf0e17df7f6f375d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nSierra Leone Economic Diversification Project (P164212)\n\n\n\nFigure 5: Investment Climate Constraints:\n_% of firms identifying as major or very severe_\n\n\n\nFigure 6: Investment Climate Constraints in Sierra Leone:\n\n_% of firms identifying as major or very severe by gender_\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nSource: World Bank Enterprise Survey 2017. Source: WeFi Tourism Enterprise Survey (2019).\n\n15. **Sierra Leone’s private sector is comprised of MSMEs which are mostly informal, with few formal**\n**companies** . While the Corporate Affairs Commission and the Office of Administrator and Registrar\nGeneral (OARG) report 6,000 registered companies, databases across various Ministries, Department and\nAgencies (MDAs) as well as private sector industry associations point to a much larger number of mainly\nmicro and small-scale businesses of 1-5 employees in the informal sector <sup>16</sup> [^16: SME Database 2017, Ministry of Trade and Industry.], who tend to operate at the\nsurvival level. The Credit Registry of the Central Bank of Sierra Leone reports a total of only 26,000\naccounts for the entire country, including accounts owned by sole proprietors <sup>17</sup> [^17: Sierra Leone, Getting Credit indicator, Doing Business 2020.] . Large scale FDI, while\nbeginning to take an interest in Sierra Leone again after the country was declared Ebola free in March\n2016, remains low (see Figure 2).\n\n\n16. **Sierra Leone’s performance in key global indexes reflects the challenging business environment**\n**in the country** . Sierra Leone ranks 134th out of 140 countries in the World Economic Forum (WEF) Global\nCompetitiveness rankings of 2018 (Figure 5). The country is underperforming the SSA average on every\ncomponent of the index. In terms of the World Bank’s Doing Business ranking (Figure 6), Sierra Leone is\nranked 163 out of 190 economies in 2020, which is equal to its\n\nMention: \"WeFi Tourism Enterprise Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Tourism firm survey cited alongside the enterprise survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000076:14:0", "occurrence_id": "9bc578d5eb1ca5cf60337230", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nSierra Leone Economic Diversification Project (P164212)\n\n\n\nFigure 5: Investment Climate Constraints:\n_% of firms identifying as major or very severe_\n\n\n\nFigure 6: Investment Climate Constraints in Sierra Leone:\n\n_% of firms identifying as major or very severe by gender_\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nSource: World Bank Enterprise Survey 2017. Source: WeFi Tourism Enterprise Survey (2019).\n\n15. **Sierra Leone’s private sector is comprised of MSMEs which are mostly informal, with few formal**\n**companies** . While the Corporate Affairs Commission and the Office of Administrator and Registrar\nGeneral (OARG) report 6,000 registered companies, databases across various Ministries, Department and\nAgencies (MDAs) as well as private sector industry associations point to a much larger number of mainly\nmicro and small-scale businesses of 1-5 employees in the informal sector <sup>16</sup> [^16: SME Database 2017, Ministry of Trade and Industry.], who tend to operate at the\nsurvival level. The Credit Registry of the Central Bank of Sierra Leone reports a total of only 26,000\naccounts for the entire country, including accounts owned by sole proprietors <sup>17</sup> [^17: Sierra Leone, Getting Credit indicator, Doing Business 2020.] . Large scale FDI, while\nbeginning to take an interest in Sierra Leone again after the country was declared Ebola free in March\n2016, remains low (see Figure 2).\n\n\n16. **Sierra Leone’s performance in key global indexes reflects the challenging business environment**\n**in the country** . Sierra Leone ranks 134th out of 140 countries in the World Economic Forum (WEF) Global\nCompetitiveness rankings of 2018 (Figure 5). The country is underperforming the SSA average on every\ncomponent of the index. In terms of the World Bank’s Doing Business ranking (Figure 6), Sierra Leone is\nranked 163 out of 190 economies in 2020, which is equal to its\n\nMention: \"SME Database 2017\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Trade ministry database cited on micro and small business counts.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000076:14:0", "occurrence_id": "f58181ec6ec50a66ad222f5c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nSierra Leone Economic Diversification Project (P164212)\n\n\n\nFigure 5: Investment Climate Constraints:\n_% of firms identifying as major or very severe_\n\n\n\nFigure 6: Investment Climate Constraints in Sierra Leone:\n\n_% of firms identifying as major or very severe by gender_\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nSource: World Bank Enterprise Survey 2017. Source: WeFi Tourism Enterprise Survey (2019).\n\n15. **Sierra Leone’s private sector is comprised of MSMEs which are mostly informal, with few formal**\n**companies** . While the Corporate Affairs Commission and the Office of Administrator and Registrar\nGeneral (OARG) report 6,000 registered companies, databases across various Ministries, Department and\nAgencies (MDAs) as well as private sector industry associations point to a much larger number of mainly\nmicro and small-scale businesses of 1-5 employees in the informal sector <sup>16</sup> [^16: SME Database 2017, Ministry of Trade and Industry.], who tend to operate at the\nsurvival level. The Credit Registry of the Central Bank of Sierra Leone reports a total of only 26,000\naccounts for the entire country, including accounts owned by sole proprietors <sup>17</sup> [^17: Sierra Leone, Getting Credit indicator, Doing Business 2020.] . Large scale FDI, while\nbeginning to take an interest in Sierra Leone again after the country was declared Ebola free in March\n2016, remains low (see Figure 2).\n\n\n16. **Sierra Leone’s performance in key global indexes reflects the challenging business environment**\n**in the country** . Sierra Leone ranks 134th out of 140 countries in the World Economic Forum (WEF) Global\nCompetitiveness rankings of 2018 (Figure 5). The country is underperforming the SSA average on every\ncomponent of the index. In terms of the World Bank’s Doing Business ranking (Figure 6), Sierra Leone is\nranked 163 out of 190 economies in 2020, which is equal to its\n\nMention: \"Credit Registry of the Central Bank of Sierra Leone\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Central bank registry reporting the small number of credit accounts.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000076:14:0", "occurrence_id": "fd777573e2e1a4a7696054ef", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nSierra Leone Economic Diversification Project (P164212)\n\n\n\nFigure 5: Investment Climate Constraints:\n_% of firms identifying as major or very severe_\n\n\n\nFigure 6: Investment Climate Constraints in Sierra Leone:\n\n_% of firms identifying as major or very severe by gender_\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nSource: World Bank Enterprise Survey 2017. Source: WeFi Tourism Enterprise Survey (2019).\n\n15. **Sierra Leone’s private sector is comprised of MSMEs which are mostly informal, with few formal**\n**companies** . While the Corporate Affairs Commission and the Office of Administrator and Registrar\nGeneral (OARG) report 6,000 registered companies, databases across various Ministries, Department and\nAgencies (MDAs) as well as private sector industry associations point to a much larger number of mainly\nmicro and small-scale businesses of 1-5 employees in the informal sector <sup>16</sup> [^16: SME Database 2017, Ministry of Trade and Industry.], who tend to operate at the\nsurvival level. The Credit Registry of the Central Bank of Sierra Leone reports a total of only 26,000\naccounts for the entire country, including accounts owned by sole proprietors <sup>17</sup> [^17: Sierra Leone, Getting Credit indicator, Doing Business 2020.] . Large scale FDI, while\nbeginning to take an interest in Sierra Leone again after the country was declared Ebola free in March\n2016, remains low (see Figure 2).\n\n\n16. **Sierra Leone’s performance in key global indexes reflects the challenging business environment**\n**in the country** . Sierra Leone ranks 134th out of 140 countries in the World Economic Forum (WEF) Global\nCompetitiveness rankings of 2018 (Figure 5). The country is underperforming the SSA average on every\ncomponent of the index. In terms of the World Bank’s Doing Business ranking (Figure 6), Sierra Leone is\nranked 163 out of 190 economies in 2020, which is equal to its\n\nMention: \"Global\nCompetitiveness rankings\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Ranking cited for the challenging business environment.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000076:38:0", "occurrence_id": "a01d6b891324b288ac6d694a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nSierra Leone Sustainable Development Tourism Project (P164212)\n\n\n**E. Rationale for World Bank Involvement and Role of Partners**\n\n\n84. The proposed project will contribute to achievement of the WBG’s twin goals of ending extreme\npoverty and boosting shared prosperity by stimulating investment and the growth of the private sector in\nSierra Leone. The World Bank’s support will help the GoSL address market, policy, coordination,\ninformation, and critical infrastructure failures as well as weak business and regulatory environment.\n\n\n85. The World Bank’s support will draw from its extensive experience, expertise, and best practices\nin private sector development, entrepreneurship development, women’s economic empowerment,\ninvestment, and SME financing across different global regions. In doing so, the World Bank’s support will\nprovide integrated solutions to the key market failures and constraints holding back SME competitiveness\nin the country. On business environment specifically, a number of surveys and studies, including the\nEnterprise Survey, Doing Business report, We-Fi Tourism Enterprise Survey and the recent Doing Business\nReform Road Map Assessment have suggested specific interventions to address key market failures. The\nWorld Bank’s intervention is further justified by its strong convening power in the country which can help\nachieve donor and line ministry coordination around the Government diversification agenda.\n\n\n86. The project preparation included consultations with the IFC, and the World Bank team will seek\nto leverage IFC’s strategy in the tourism/hotel space which includes (i) targeting projects with strong\ncommercial viability, development impact and additionality; (ii) exploring new regions in the country; (iii)\nfocusing on markets with clear supply/demand gap supporting different formats (business hotels, budget\nchains, resorts, mixed use developments, serviced apartments); (iv) supporting resort development and\nthe promotion of heritage destinations in line with government tourism plans; (v) encouraging green\nbuildings and leveraging Excellence in Design for Greater Efficiencies (EDGE)\n\nMention: \"Enterprise Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Survey among those identifying SME constraints in the business environment.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000076:38:0", "occurrence_id": "eaff0b18d4fbb80c7bbdc08b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nSierra Leone Sustainable Development Tourism Project (P164212)\n\n\n**E. Rationale for World Bank Involvement and Role of Partners**\n\n\n84. The proposed project will contribute to achievement of the WBG’s twin goals of ending extreme\npoverty and boosting shared prosperity by stimulating investment and the growth of the private sector in\nSierra Leone. The World Bank’s support will help the GoSL address market, policy, coordination,\ninformation, and critical infrastructure failures as well as weak business and regulatory environment.\n\n\n85. The World Bank’s support will draw from its extensive experience, expertise, and best practices\nin private sector development, entrepreneurship development, women’s economic empowerment,\ninvestment, and SME financing across different global regions. In doing so, the World Bank’s support will\nprovide integrated solutions to the key market failures and constraints holding back SME competitiveness\nin the country. On business environment specifically, a number of surveys and studies, including the\nEnterprise Survey, Doing Business report, We-Fi Tourism Enterprise Survey and the recent Doing Business\nReform Road Map Assessment have suggested specific interventions to address key market failures. The\nWorld Bank’s intervention is further justified by its strong convening power in the country which can help\nachieve donor and line ministry coordination around the Government diversification agenda.\n\n\n86. The project preparation included consultations with the IFC, and the World Bank team will seek\nto leverage IFC’s strategy in the tourism/hotel space which includes (i) targeting projects with strong\ncommercial viability, development impact and additionality; (ii) exploring new regions in the country; (iii)\nfocusing on markets with clear supply/demand gap supporting different formats (business hotels, budget\nchains, resorts, mixed use developments, serviced apartments); (iv) supporting resort development and\nthe promotion of heritage destinations in line with government tourism plans; (v) encouraging green\nbuildings and leveraging Excellence in Design for Greater Efficiencies (EDGE)\n\nMention: \"Doing Business report\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Report cited among evidence on business environment constraints.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000076:38:0", "occurrence_id": "5d1715ba9b6872ce7662d47f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nSierra Leone Sustainable Development Tourism Project (P164212)\n\n\n**E. Rationale for World Bank Involvement and Role of Partners**\n\n\n84. The proposed project will contribute to achievement of the WBG’s twin goals of ending extreme\npoverty and boosting shared prosperity by stimulating investment and the growth of the private sector in\nSierra Leone. The World Bank’s support will help the GoSL address market, policy, coordination,\ninformation, and critical infrastructure failures as well as weak business and regulatory environment.\n\n\n85. The World Bank’s support will draw from its extensive experience, expertise, and best practices\nin private sector development, entrepreneurship development, women’s economic empowerment,\ninvestment, and SME financing across different global regions. In doing so, the World Bank’s support will\nprovide integrated solutions to the key market failures and constraints holding back SME competitiveness\nin the country. On business environment specifically, a number of surveys and studies, including the\nEnterprise Survey, Doing Business report, We-Fi Tourism Enterprise Survey and the recent Doing Business\nReform Road Map Assessment have suggested specific interventions to address key market failures. The\nWorld Bank’s intervention is further justified by its strong convening power in the country which can help\nachieve donor and line ministry coordination around the Government diversification agenda.\n\n\n86. The project preparation included consultations with the IFC, and the World Bank team will seek\nto leverage IFC’s strategy in the tourism/hotel space which includes (i) targeting projects with strong\ncommercial viability, development impact and additionality; (ii) exploring new regions in the country; (iii)\nfocusing on markets with clear supply/demand gap supporting different formats (business hotels, budget\nchains, resorts, mixed use developments, serviced apartments); (iv) supporting resort development and\nthe promotion of heritage destinations in line with government tourism plans; (v) encouraging green\nbuildings and leveraging Excellence in Design for Greater Efficiencies (EDGE)\n\nMention: \"We-Fi Tourism Enterprise Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Tourism survey cited among business environment evidence.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000076:38:0", "occurrence_id": "5d7312110f19d11a081a722a", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nSierra Leone Sustainable Development Tourism Project (P164212)\n\n\n**E. Rationale for World Bank Involvement and Role of Partners**\n\n\n84. The proposed project will contribute to achievement of the WBG’s twin goals of ending extreme\npoverty and boosting shared prosperity by stimulating investment and the growth of the private sector in\nSierra Leone. The World Bank’s support will help the GoSL address market, policy, coordination,\ninformation, and critical infrastructure failures as well as weak business and regulatory environment.\n\n\n85. The World Bank’s support will draw from its extensive experience, expertise, and best practices\nin private sector development, entrepreneurship development, women’s economic empowerment,\ninvestment, and SME financing across different global regions. In doing so, the World Bank’s support will\nprovide integrated solutions to the key market failures and constraints holding back SME competitiveness\nin the country. On business environment specifically, a number of surveys and studies, including the\nEnterprise Survey, Doing Business report, We-Fi Tourism Enterprise Survey and the recent Doing Business\nReform Road Map Assessment have suggested specific interventions to address key market failures. The\nWorld Bank’s intervention is further justified by its strong convening power in the country which can help\nachieve donor and line ministry coordination around the Government diversification agenda.\n\n\n86. The project preparation included consultations with the IFC, and the World Bank team will seek\nto leverage IFC’s strategy in the tourism/hotel space which includes (i) targeting projects with strong\ncommercial viability, development impact and additionality; (ii) exploring new regions in the country; (iii)\nfocusing on markets with clear supply/demand gap supporting different formats (business hotels, budget\nchains, resorts, mixed use developments, serviced apartments); (iv) supporting resort development and\nthe promotion of heritage destinations in line with government tourism plans; (v) encouraging green\nbuildings and leveraging Excellence in Design for Greater Efficiencies (EDGE)\n\nMention: \"Doing Business\nReform Road Map Assessment\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Assessment cited for proposed reform interventions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000078:2:0", "occurrence_id": "4356a9442d24f13898dd80ff", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nGovernance of Extractives for Local Development Project (P164271)\n\n\nOther Decision (as needed)\n\n\n**B. Introduction and Context**\n\nCountry Context\n1. **Niger is exposed to multiple challenges adversely impacting its development:** (i) regional insecurity and violent\nextremist groups threaten stability, fuel pre-existing tensions and cause population displacement including refugees; (ii)\nhigh population growth and a large population of youth with few economic opportunities; (iii) an improved macro-fiscal\nsituation which nonetheless remains exposed to multiple external shocks due to dependency on agriculture and\nextractives; (iv) increased rural poverty and in the number of poor; (v) deep inequalities, particularly gender inequalities.\n\n\n2. **There is significant scope to improve basic service delivery** . Niger ranked 189 out of 189 on the 2018 Human\nDevelopment Index, reflecting low education and health outcomes. The net enrollment rate in primary education was 63\npercent (58 percent for women) in 2016. The child malnutrition rate was 42.2 percent (2017) and life expectancy was 60\nyears (2016). Service quality is the main challenge. A third of primary school teachers are not in the classroom and 41\npercent have the minimum knowledge required to teach. About a third of health workers are absent and two thirds do\nnot correctly diagnose more than two out of five tracer conditions. Half of the health facilities do not have functional basic\nequipment, and less than a quarter of the facilities have water, sanitation and power. <sup>1</sup> [^1: Service Delivery Indicators (SDI 2016).] High population growth, the size\nof the country, increased insecurity, limited state presence and capability as well as insufficient resources make it\nchallenging to respond to increasing demand for basic services.\n\n\n3. **The macro-fiscal situation has improved but Niger remains highly exposed to external shocks.** Real annual\ngrowth rates averaged 6.4 percent between 2012 and 2017\n\nMention: \"Service Delivery Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Facility survey behind equipment and service access gaps.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000078:4:1", "occurrence_id": "21b48f21bd0a3bee357183b1", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Artisanal and Small-Scale Miners (ASM/SSM)\nlack the means and skills (i.e. in technical, business, finance and environmental management) required for well-organized\nand managed ASM/SSMs. Key areas to develop include all aspects of modern mining management, including tax and\nauditing, and mineral extraction and processing technologies.\n\n\n11. **Limited availability of reliable geoscientific data is a constraint to promote and develop the sector** . Available\ndata on geological resources covers less than 15 percent of Niger’s territory. More detailed and targeted analytical and\ngeodata interpretation work is necessary to assess and market the mineral potential of the country. This is particularly\nrelevant as Niger does show untapped potential for several strategic or ‘battery’ minerals (i.e. lithium, copper, cobalt).\nInvesting in geodata will ultimately result in greater interest by exploration firms due to reduced costs and risks and will\ncontribute to strengthen the GoN’s capacity to attract and secure more favorable terms with investors.\n\n\n12. **Niger has recently re-committed to revenue transparency and good governance in extractive industries.** Niger\nwas an active member of the Extractive Industries Transparency Initiative (EITI) since 2005. In October 2017 Niger was\nsuspended from EITI due to inadequate progress in implementing the 2016 Standard and insufficient civil society\nengagement. The GoN since confirmed its intention to re-commit to EITI and is currently working on submitting their\nrenewed application. <sup>12</sup> [^12: GoN has established a high-level Ad Hoc Committee to prepare a new EITI strategy; appointed a new Permanent Secretary; and\ndrafted a new EITI Decree currently under review by a multi-stakeholder group.] Re-joining EITI will be an important signal towards the investment community.\n\n\n13. **There is room for boosting extractive sector upstream and downstream economic linkages** [. When done right,](https://twitter.com/intent/tweet?text=When+done+right%2C+developing+local+content+\n\nMention: \"geoscientific data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"inform\", \"summary\": \"Reliable geoscientific coverage discussed as a constraint to mining.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000078:4:1", "occurrence_id": "84497e1d1ebb1793c09d6557", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Artisanal and Small-Scale Miners (ASM/SSM)\nlack the means and skills (i.e. in technical, business, finance and environmental management) required for well-organized\nand managed ASM/SSMs. Key areas to develop include all aspects of modern mining management, including tax and\nauditing, and mineral extraction and processing technologies.\n\n\n11. **Limited availability of reliable geoscientific data is a constraint to promote and develop the sector** . Available\ndata on geological resources covers less than 15 percent of Niger’s territory. More detailed and targeted analytical and\ngeodata interpretation work is necessary to assess and market the mineral potential of the country. This is particularly\nrelevant as Niger does show untapped potential for several strategic or ‘battery’ minerals (i.e. lithium, copper, cobalt).\nInvesting in geodata will ultimately result in greater interest by exploration firms due to reduced costs and risks and will\ncontribute to strengthen the GoN’s capacity to attract and secure more favorable terms with investors.\n\n\n12. **Niger has recently re-committed to revenue transparency and good governance in extractive industries.** Niger\nwas an active member of the Extractive Industries Transparency Initiative (EITI) since 2005. In October 2017 Niger was\nsuspended from EITI due to inadequate progress in implementing the 2016 Standard and insufficient civil society\nengagement. The GoN since confirmed its intention to re-commit to EITI and is currently working on submitting their\nrenewed application. <sup>12</sup> [^12: GoN has established a high-level Ad Hoc Committee to prepare a new EITI strategy; appointed a new Permanent Secretary; and\ndrafted a new EITI Decree currently under review by a multi-stakeholder group.] Re-joining EITI will be an important signal towards the investment community.\n\n\n13. **There is room for boosting extractive sector upstream and downstream economic linkages** [. When done right,](https://twitter.com/intent/tweet?text=When+done+right%2C+developing+local+content+\n\nMention: \"geodata\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Interpretation work needed to assess mineral potential.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000079:4:2", "occurrence_id": "b1f8c85995389a580119bbda", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  The Agadem-Cotonou oil pipeline is expected to boost production from 20,000 barrels/day in 2019 to 110,000 barrels in 2022.\n15 Available data on geological resources cover less than 15 percent of Niger’s overall territory.\n\n\nJan 27, 2020 Page 5 of 8\n\nMention: \"data on geological resources\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Footnote noting geological coverage of less than 15 percent of territory.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000081:46:0", "occurrence_id": "16686bec77b5c56d6e8e1e68", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Banco Mundial**\nProyecto de Seguridad Hídrica y Resiliencia del Valle de México (PROSEGHIR) (P164389)\n\n\n**Subcomponente C2.1: Diseño e implementación de un observatorio de acuíferos para el Valle de México**\n\n\n6. Mediante este subcomponente se proporcionarán las herramientas para el monitoreo y la gestión eficaz de los\nrecursos de aguas subterráneas en la ZMVM, respaldando el diseño y la puesta en marcha de un observatorio de acuíferos\npara el Valle de México. Se financiarán, entre otras cosas:\n\n\na. La recopilación de datos de referencia y la preparación de un diagnóstico sobre la calidad y la cantidad\n\nde agua en los acuíferos (véase el esquema de acuíferos en el anexo 5, mapa 1), incluidas las pruebas\nde bombeo de aproximadamente 20 pozos en toda la ZMVM.\n\n\nb. La preparación de un inventario de pozos dentro del Valle de México, que incluirá datos sobre los\n\nusuarios de agua, y en el que se actualizará el inventario anterior publicado en 2007 y se brindará\ninformación esencial para una gestión eficaz de los acuíferos de la ZMVM.\n\n\nc. La expansión y automatización de la red piezométrica del Valle de México para monitorear las\n\nvariaciones mensuales de las aguas subterráneas, incluido: i) un diagnóstico de los pozos existentes\nque conform\n\nMention: \"datos de referencia\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"Recopilación de datos base para el diagnóstico de calidad y cantidad de agua.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000081:46:0", "occurrence_id": "fb346a0dd231cc598a05da8f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Banco Mundial**\nProyecto de Seguridad Hídrica y Resiliencia del Valle de México (PROSEGHIR) (P164389)\n\n\n**Subcomponente C2.1: Diseño e implementación de un observatorio de acuíferos para el Valle de México**\n\n\n6. Mediante este subcomponente se proporcionarán las herramientas para el monitoreo y la gestión eficaz de los\nrecursos de aguas subterráneas en la ZMVM, respaldando el diseño y la puesta en marcha de un observatorio de acuíferos\npara el Valle de México. Se financiarán, entre otras cosas:\n\n\na. La recopilación de datos de referencia y la preparación de un diagnóstico sobre la calidad y la cantidad\n\nde agua en los acuíferos (véase el esquema de acuíferos en el anexo 5, mapa 1), incluidas las pruebas\nde bombeo de aproximadamente 20 pozos en toda la ZMVM.\n\n\nb. La preparación de un inventario de pozos dentro del Valle de México, que incluirá datos sobre los\n\nusuarios de agua, y en el que se actualizará el inventario anterior publicado en 2007 y se brindará\ninformación esencial para una gestión eficaz de los acuíferos de la ZMVM.\n\n\nc. La expansión y automatización de la red piezométrica del Valle de México para monitorear las\n\nvariaciones mensuales de las aguas subterráneas, incluido: i) un diagnóstico de los pozos existentes\nque conform\n\nMention: \"inventario de pozos\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"Inventario de pozos que actualiza el publicado en 2007.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000081:46:0", "occurrence_id": "e4d176aeb9e64d0aacf1b2b9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **Banco Mundial**\nProyecto de Seguridad Hídrica y Resiliencia del Valle de México (PROSEGHIR) (P164389)\n\n\n**Subcomponente C2.1: Diseño e implementación de un observatorio de acuíferos para el Valle de México**\n\n\n6. Mediante este subcomponente se proporcionarán las herramientas para el monitoreo y la gestión eficaz de los\nrecursos de aguas subterráneas en la ZMVM, respaldando el diseño y la puesta en marcha de un observatorio de acuíferos\npara el Valle de México. Se financiarán, entre otras cosas:\n\n\na. La recopilación de datos de referencia y la preparación de un diagnóstico sobre la calidad y la cantidad\n\nde agua en los acuíferos (véase el esquema de acuíferos en el anexo 5, mapa 1), incluidas las pruebas\nde bombeo de aproximadamente 20 pozos en toda la ZMVM.\n\n\nb. La preparación de un inventario de pozos dentro del Valle de México, que incluirá datos sobre los\n\nusuarios de agua, y en el que se actualizará el inventario anterior publicado en 2007 y se brindará\ninformación esencial para una gestión eficaz de los acuíferos de la ZMVM.\n\n\nc. La expansión y automatización de la red piezométrica del Valle de México para monitorear las\n\nvariaciones mensuales de las aguas subterráneas, incluido: i) un diagnóstico de los pozos existentes\nque conform\n\nMention: \"datos sobre los\n\nusuarios de agua\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"Datos de usuarios incluidos en el inventario de pozos.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000083:2:2", "occurrence_id": "90172828561d2da59a4dc139", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  work. The lack of skills and quality jobs for women has repercussions for women’s\nfertility decisions, and, ultimately, for the development of Burundi.\n\n\n1 Youth Labour Statistics, International Labour Organization, https://ilostat.ilo.org/topics/youth/.\n2 Data.worldbank.org.\n3 Enquêtes sur les Conditions de Vie des Ménages au Burundi (ECVMB), 2013-2014.\n\n\nJan 20, 2021 Page 3 of 12\n\nMention: \"Youth Labour Statistics\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"Footnote source of youth labour statistics behind the skills discussion.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000084:15:1", "occurrence_id": "7276834f1f678f5e88b398fa", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  there are few sites suitable for wind\npower generation in Burundi, but some locations such as the shores of Lake Tanganyika (wind speed is 4 to 5 m/s) could\nprove to have favorable conditions for the exploitation of such energy. To-date, no feasibility studies on wind power have\nbeen carried out in Burundi. There are also several available sources of waste biomass and agricultural residue in different\nforms in Burundi, which may be harvested and transformed into energy sources, although a resource assessment is\nneeded to evaluate competing uses.\n\n\n8. **Third, insufficient energy supply is undermining economic growth opportunities** . A 2014 World Bank Enterprise\nSurvey, the latest so far, found that 22 percent of Burundian firms identified poor electricity access and reliability as major\nbarriers to investment, compared to an average 15 percent across Sub-Saharan Africa. <sup>9</sup> [^9: Enterprise Surveys (http://www.enterprisesurveys.org). World Bank, Burundi 2014.] Outages are frequent due to\nvarious reasons: insufficient supply related to lack of investment in new generation capacity along with inadequate\nmaintenance of the existing plants, excessive reliance on hydropower (affecting supply during droughts), and high\ntechnical and commercial losses and operational failures. The imbalance between supply and demand in Burundi resulted\nin rolling blackouts in 2015 and 2016 and service interruptions. <sup>10</sup> [^10: Tractebel. (2018). _Plan Directeur de Production et de Transport de l’Energie Électrique au Burundi. Rapport final - Tome 2: Volume 1_\n_Prévision de la demande_ . Page 21.] The situation will not improve in the short term, as\ndemand is expected to continue to grow <sup>11</sup> [^11: Peak demand is expected to grow to 89 MW in 2020 and 249 MW in 2030.] while new generations projects in the pipeline will not come online before\n2021, if not later.\n\n\n4 Reliable data on energy access are notnages\n\nMention: \"2014 World Bank Enterprise\nSurvey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Firm survey reporting the share citing poor electricity as an investment barrier.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000084:15:2", "occurrence_id": "b053d73ecca5efe7e73c55aa", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: nages Burundais,_ ECVMB 2017) reports that 7.5 percent of the population has access to electricity.\nTracking SDG7: The Energy Progress Report 2018 estimated that nationwide access stands at 9 percent.\n5 https://trackingsdg7.esmap.org/results\n6World Bank, June 2018. Subjective Poverty and Recent Changes in Living Conditions: Findings from the Rapid Household Survey (2017).\n7 World Bank, June 2016. Implementation Completion and Results Report on a Global Environment Facility Grant to the Republic of Burundi.\n8 World Development Indicators. 2016. Electric power consumption (kWh per capita).\nhttps://data.worldbank.org/indicator/EG.USE.ELEC.KH.PC\n9 Enterprise Surveys (http://www.enterprisesurveys.org). World Bank, Burundi 2014.\n10 Tractebel. (2018). _Plan Directeur de Production et de Transport de l’Energie Électrique au Burundi. Rapport final - Tome 2: Volume 1_\n_Prévision de la demande_ . Page 21.\n11 Peak demand is expected to grow to 89 MW in 2020 and 249 MW in 2030.\n\n\nPage 9 of 86\n\nMention: \"Tracking SDG7\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Energy progress report estimating nationwide electricity access.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000084:15:2", "occurrence_id": "f18b91e2e35819a531b97acd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: nages Burundais,_ ECVMB 2017) reports that 7.5 percent of the population has access to electricity.\nTracking SDG7: The Energy Progress Report 2018 estimated that nationwide access stands at 9 percent.\n5 https://trackingsdg7.esmap.org/results\n6World Bank, June 2018. Subjective Poverty and Recent Changes in Living Conditions: Findings from the Rapid Household Survey (2017).\n7 World Bank, June 2016. Implementation Completion and Results Report on a Global Environment Facility Grant to the Republic of Burundi.\n8 World Development Indicators. 2016. Electric power consumption (kWh per capita).\nhttps://data.worldbank.org/indicator/EG.USE.ELEC.KH.PC\n9 Enterprise Surveys (http://www.enterprisesurveys.org). World Bank, Burundi 2014.\n10 Tractebel. (2018). _Plan Directeur de Production et de Transport de l’Energie Électrique au Burundi. Rapport final - Tome 2: Volume 1_\n_Prévision de la demande_ . Page 21.\n11 Peak demand is expected to grow to 89 MW in 2020 and 249 MW in 2030.\n\n\nPage 9 of 86\n\nMention: \"Rapid Household Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Rapid household survey reporting subjective poverty and living conditions.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000084:15:2", "occurrence_id": "846773ad6ff8af7843f08fef", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: nages Burundais,_ ECVMB 2017) reports that 7.5 percent of the population has access to electricity.\nTracking SDG7: The Energy Progress Report 2018 estimated that nationwide access stands at 9 percent.\n5 https://trackingsdg7.esmap.org/results\n6World Bank, June 2018. Subjective Poverty and Recent Changes in Living Conditions: Findings from the Rapid Household Survey (2017).\n7 World Bank, June 2016. Implementation Completion and Results Report on a Global Environment Facility Grant to the Republic of Burundi.\n8 World Development Indicators. 2016. Electric power consumption (kWh per capita).\nhttps://data.worldbank.org/indicator/EG.USE.ELEC.KH.PC\n9 Enterprise Surveys (http://www.enterprisesurveys.org). World Bank, Burundi 2014.\n10 Tractebel. (2018). _Plan Directeur de Production et de Transport de l’Energie Électrique au Burundi. Rapport final - Tome 2: Volume 1_\n_Prévision de la demande_ . Page 21.\n11 Peak demand is expected to grow to 89 MW in 2020 and 249 MW in 2030.\n\n\nPage 9 of 86\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Indicator series cited for electric power consumption per capita.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000084:15:2", "occurrence_id": "922596fb027cbe250d28eb96", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: nages Burundais,_ ECVMB 2017) reports that 7.5 percent of the population has access to electricity.\nTracking SDG7: The Energy Progress Report 2018 estimated that nationwide access stands at 9 percent.\n5 https://trackingsdg7.esmap.org/results\n6World Bank, June 2018. Subjective Poverty and Recent Changes in Living Conditions: Findings from the Rapid Household Survey (2017).\n7 World Bank, June 2016. Implementation Completion and Results Report on a Global Environment Facility Grant to the Republic of Burundi.\n8 World Development Indicators. 2016. Electric power consumption (kWh per capita).\nhttps://data.worldbank.org/indicator/EG.USE.ELEC.KH.PC\n9 Enterprise Surveys (http://www.enterprisesurveys.org). World Bank, Burundi 2014.\n10 Tractebel. (2018). _Plan Directeur de Production et de Transport de l’Energie Électrique au Burundi. Rapport final - Tome 2: Volume 1_\n_Prévision de la demande_ . Page 21.\n11 Peak demand is expected to grow to 89 MW in 2020 and 249 MW in 2030.\n\n\nPage 9 of 86\n\nMention: \"Enterprise Surveys\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Enterprise survey round cited for firm-level constraints.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000084:16:1", "occurrence_id": "62db188e888e270946a929df", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: |3|49|1|1|\n|||Bujumbura<br>Mairie|68|68|68|68|\n|||Rumonge|10|41|5|5|\n\n\n_<mark>Source:</mark>_ <mark>Demographic and Health Survey, 2013.</mark>\n\n9. **Fourth, the dependence on wooded biomass is overwhelming with significant implications for deforestation**\n**and environment.** Biomass represents 95 percent of primary energy consumption, while electricity only accounts for 1.3\npercent and oil 2.5 percent. An estimated 98 percent of the population <sup>12</sup> [^12: ESDB-III : Troisième Enquête Démographique et de Santé au Burundi 2016 – 2017.], in urban and rural areas uses wood or woodfuel as the main source for heating and cooking. Cooking happens indoor for 92 percent of the population (96 percent for\nrural and 60 percent for urban). These poor cooking solutions (primarily with wood and residues) combined with the high\ndensity and demographic growth exert a high pressure on the woodstock. It is estimated that the fraction of nonrenewable biomass is at 79 percent, meaning that only one-fifth of the wood used in a year is regrown that year. This\nsituation results in (a) significant natural capital depletion, estimated at 29 percent of GDP per capita in 2010 and largely\ndue to a net depletion of forests, <sup>13</sup> [^13: World Bank 2016. Burundi Poverty Assessment, page 153 - 154.] and (b) dire health costs for families (especially women and the youngest members of\nfamilies) due to respiratory illness related to indoor air pollution and the physical strain of an ever-expanding radius of\nwood collection.\n\n\n_Off-grid service delivery mechanisms are essential to the GoB vision of 30 percent electricity access by 2030_\n\n\n10. **Alternative service delivery mechanisms are needed to provide services to customers in off-grid areas, either**\n\nMention: \"Demographic and Health Survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Survey source of the household table on energy access by commune.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000085:2:1", "occurrence_id": "14bc3e3576eabc69fbab3c40", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  (“mass poverty”).\nBurundi showed initial signs of economic recovery from 2010 to 2014 when GDP grew by an average of 4.2\npercent per year. The renewed outbreak of political disturbance in April 2015 resulted in economic\ncontraction, with a negative GDP growth of 3.9 percent in 2015 and 0.6 percent in 2016. <sup>2</sup> However, 2017\nagain showed signs of recovery as annual GDP growth was positive, at 0.5 percent. This pick up was\nsupported by rebounds in coffee production and other industries, as chronic power shortages were reduced.\nWorld Bank projections also show positive real GDP growth rates, up to an estimated 2.5 percent in 2020. <sup>3</sup>\nSecurity has improved since 2015 but remains fragile.\n\n3. **Burundi is a fragile and conflict-affected country with considerable development challenges.** Despite\nstrong efforts and improvement in key health and education indicators, Burundi ranks 165th out of 189\ncountries on the 2018 Statistical Update of the Human Development Index. After more than 10 years of\nopen conflict, almost three years of economic embargo, and a four-year post-war emergency period,\nBurundi has one of the lowest per capita Gross National Income (GNIs) in the world (US$280 in 2016).\nAccording to the Poverty Assessment of 2016 and the most recent national household survey ( _Enquête sur_\n\n\n1 Food and Agriculture Organization and World Bank population estimates. https://data.worldbank.org/indicator/EN.POP.DNST\n2 The political unrest and violence that erupted in April 2015 was in the wake of the contested reelection of the incumbent President for a third term\nin office in open violation of the Arusha accords\n3 World Bank Group, Burundi Interim Economic Update, August 2018. Macroeconomics, Trade and Investment Global Practice.\n\n\nApril 2019 Page 3 of 14\n\nMention: \"Poverty Assessment of 2016\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Assessment cited alongside the household survey for poverty figures.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000085:3:0", "occurrence_id": "977ffdc20dcfb1e99514b37b", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nBurundi Offgrid Access Project (P164435)\n\n\n_les Conditions de Vie des Ménages Burundais,_ ECVMB, 2017) <sup>4,</sup> [^4: ECVMB, _Enquête sur les Conditions de Vie des Ménages Burundais_ (2017).] around 65 percent of the Burundian\npopulation is classified as poor (living under the national poverty line of US$2 a day) and over 70 percent\nlives on less than US$1.90 per capita and per day.\n\n4. **Electricity deficiency is one of the principal barriers to social and economic development in Burundi** . Not\nmore than 7.6 percent of the population of Burundi has access to electricity <sup>5</sup> [^5: Global Tracking Framework, 2016.] —one of the lowest in the world.\nAccording to the ECVMB 2017, the electricity rate in rural Burundi is as little as 1.8 percent <sup>6</sup> [^6: 5 percent in the Country Partnership Framework (CPF) for the Republic of Burundi For the Period of FY18-FY22 (Draft for Discussion). The World\nBank (2018) versus 7 percent in the Burundi 2016 RISE country data versus 7.5 percent in the Subjective Poverty and recent Changes in Living\nConditions: Findings from the Rapid Household Survey (ECVMB 2017), World Bank, June 2018.], which makes it\nimpossible to provide key social services such as education and health to the population. A 2014 World Bank\nEnterprise Survey, the latest so far, found that 22 percent of Burundian firms identified poor electricity\naccess and reliability as major barriers to investment, compared to an average 15 percent across Sub\nSaharan Africa. <sup>7</sup> [^7: Enterprise Surveys (http://www.enterprisesurveys.org). The World Bank. Burundi (2014).] Fewer power shortages took place in 2017, thanks to the extra 30MW\n\nMention: \"Global Tracking Framework\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Framework cited for the electricity access share.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000085:3:0", "occurrence_id": "b2790b25a5c51de84c8053cf", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nBurundi Offgrid Access Project (P164435)\n\n\n_les Conditions de Vie des Ménages Burundais,_ ECVMB, 2017) <sup>4,</sup> [^4: ECVMB, _Enquête sur les Conditions de Vie des Ménages Burundais_ (2017).] around 65 percent of the Burundian\npopulation is classified as poor (living under the national poverty line of US$2 a day) and over 70 percent\nlives on less than US$1.90 per capita and per day.\n\n4. **Electricity deficiency is one of the principal barriers to social and economic development in Burundi** . Not\nmore than 7.6 percent of the population of Burundi has access to electricity <sup>5</sup> [^5: Global Tracking Framework, 2016.] —one of the lowest in the world.\nAccording to the ECVMB 2017, the electricity rate in rural Burundi is as little as 1.8 percent <sup>6</sup> [^6: 5 percent in the Country Partnership Framework (CPF) for the Republic of Burundi For the Period of FY18-FY22 (Draft for Discussion). The World\nBank (2018) versus 7 percent in the Burundi 2016 RISE country data versus 7.5 percent in the Subjective Poverty and recent Changes in Living\nConditions: Findings from the Rapid Household Survey (ECVMB 2017), World Bank, June 2018.], which makes it\nimpossible to provide key social services such as education and health to the population. A 2014 World Bank\nEnterprise Survey, the latest so far, found that 22 percent of Burundian firms identified poor electricity\naccess and reliability as major barriers to investment, compared to an average 15 percent across Sub\nSaharan Africa. <sup>7</sup> [^7: Enterprise Surveys (http://www.enterprisesurveys.org). The World Bank. Burundi (2014).] Fewer power shortages took place in 2017, thanks to the extra 30MW\n\nMention: \"Burundi 2016 RISE country data\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Country data set compared against other access estimates.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000085:3:0", "occurrence_id": "da268278178bfa7974a25374", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nBurundi Offgrid Access Project (P164435)\n\n\n_les Conditions de Vie des Ménages Burundais,_ ECVMB, 2017) <sup>4,</sup> [^4: ECVMB, _Enquête sur les Conditions de Vie des Ménages Burundais_ (2017).] around 65 percent of the Burundian\npopulation is classified as poor (living under the national poverty line of US$2 a day) and over 70 percent\nlives on less than US$1.90 per capita and per day.\n\n4. **Electricity deficiency is one of the principal barriers to social and economic development in Burundi** . Not\nmore than 7.6 percent of the population of Burundi has access to electricity <sup>5</sup> [^5: Global Tracking Framework, 2016.] —one of the lowest in the world.\nAccording to the ECVMB 2017, the electricity rate in rural Burundi is as little as 1.8 percent <sup>6</sup> [^6: 5 percent in the Country Partnership Framework (CPF) for the Republic of Burundi For the Period of FY18-FY22 (Draft for Discussion). The World\nBank (2018) versus 7 percent in the Burundi 2016 RISE country data versus 7.5 percent in the Subjective Poverty and recent Changes in Living\nConditions: Findings from the Rapid Household Survey (ECVMB 2017), World Bank, June 2018.], which makes it\nimpossible to provide key social services such as education and health to the population. A 2014 World Bank\nEnterprise Survey, the latest so far, found that 22 percent of Burundian firms identified poor electricity\naccess and reliability as major barriers to investment, compared to an average 15 percent across Sub\nSaharan Africa. <sup>7</sup> [^7: Enterprise Surveys (http://www.enterprisesurveys.org). The World Bank. Burundi (2014).] Fewer power shortages took place in 2017, thanks to the extra 30MW\n\nMention: \"Rapid Household Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Survey reporting the household electricity rate used in the comparison.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000085:3:0", "occurrence_id": "35138467034180b97576386f", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nBurundi Offgrid Access Project (P164435)\n\n\n_les Conditions de Vie des Ménages Burundais,_ ECVMB, 2017) <sup>4,</sup> [^4: ECVMB, _Enquête sur les Conditions de Vie des Ménages Burundais_ (2017).] around 65 percent of the Burundian\npopulation is classified as poor (living under the national poverty line of US$2 a day) and over 70 percent\nlives on less than US$1.90 per capita and per day.\n\n4. **Electricity deficiency is one of the principal barriers to social and economic development in Burundi** . Not\nmore than 7.6 percent of the population of Burundi has access to electricity <sup>5</sup> [^5: Global Tracking Framework, 2016.] —one of the lowest in the world.\nAccording to the ECVMB 2017, the electricity rate in rural Burundi is as little as 1.8 percent <sup>6</sup> [^6: 5 percent in the Country Partnership Framework (CPF) for the Republic of Burundi For the Period of FY18-FY22 (Draft for Discussion). The World\nBank (2018) versus 7 percent in the Burundi 2016 RISE country data versus 7.5 percent in the Subjective Poverty and recent Changes in Living\nConditions: Findings from the Rapid Household Survey (ECVMB 2017), World Bank, June 2018.], which makes it\nimpossible to provide key social services such as education and health to the population. A 2014 World Bank\nEnterprise Survey, the latest so far, found that 22 percent of Burundian firms identified poor electricity\naccess and reliability as major barriers to investment, compared to an average 15 percent across Sub\nSaharan Africa. <sup>7</sup> [^7: Enterprise Surveys (http://www.enterprisesurveys.org). The World Bank. Burundi (2014).] Fewer power shortages took place in 2017, thanks to the extra 30MW\n\nMention: \"2014 World Bank\nEnterprise Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Firm survey on electricity barriers to investment.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000086:4:2", "occurrence_id": "fdee747f1bdb3de3f36d521c", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Bank,\nJune 2018.\n5 Implementation Completion and Results Report on a Global Environment Facility Grant to the Republic of Burundi. World Bank,\nJune 2016.\n6 World Development Indicators. Electric power consumption (kWh per capita), 2016,\nhttps://data.worldbank.org/indicator/EG.USE.ELEC.KH.PC\n7 Enterprise Surveys (http://www.enterprisesurveys.org). The World Bank. Burundi (2014).\n8 Plan Directeur de Production et de Transport de l’Energie Électrique au Burundi. Tractebel. (2018). Rapport final - Tome 2:\nVolume 1 - Prévision de la demande. Page 21.\n9 Peak demand is expected to grow to 89 MW in 2020 and 249 MW in 2030.9\n10 ESDB-III : Troisième Enquête Démographique et de Santé au Burundi 2016-2017\n11 Burundi Poverty Assessment, World Bank 2016 – page 153-154\n\n\nJan 03, 2020 Page 5 of 13\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Indicator series for per capita electricity consumption.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000086:4:2", "occurrence_id": "2956df725ff07b86f0b71743", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Bank,\nJune 2018.\n5 Implementation Completion and Results Report on a Global Environment Facility Grant to the Republic of Burundi. World Bank,\nJune 2016.\n6 World Development Indicators. Electric power consumption (kWh per capita), 2016,\nhttps://data.worldbank.org/indicator/EG.USE.ELEC.KH.PC\n7 Enterprise Surveys (http://www.enterprisesurveys.org). The World Bank. Burundi (2014).\n8 Plan Directeur de Production et de Transport de l’Energie Électrique au Burundi. Tractebel. (2018). Rapport final - Tome 2:\nVolume 1 - Prévision de la demande. Page 21.\n9 Peak demand is expected to grow to 89 MW in 2020 and 249 MW in 2030.9\n10 ESDB-III : Troisième Enquête Démographique et de Santé au Burundi 2016-2017\n11 Burundi Poverty Assessment, World Bank 2016 – page 153-154\n\n\nJan 03, 2020 Page 5 of 13\n\nMention: \"Enterprise Surveys\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Enterprise round cited for firm constraints.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000094:3:0", "occurrence_id": "9802d705c6febfdad248f526", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nLand Tenure Improvement Project (P164906)\n\n\nand the improvement of processes for investors to acquire land. Currently, Tanzania ranks 142 of 187 economies in\nterms of ease of registering property in the 2018 Doing Business Report.\n\n\nOne of the key institutions that buttress a strong business environment is the institution of property rights and a\nsystem for land registration. Clear, enforceable and tradeable property rights provide security of tenure for\ncommunities, families and individuals to invest and reduce the potential for conflicts over land and other natural\nresources. Property rights through secure tenure are, therefore, an important part of what is needed to achieve\ninclusive growth.\n\n\nTanzania has one of the lowest records of land registration (<10%) in Africa. The current land registration process is\nnot systematic but sporadic, which is expensive and does not provide a complete database for planning, taxation and\nother key uses. While community norms may provide a perceived tenure security, as pressure on land resources\nincreases, documentation of land rights may help property holders to be more secure and more likely to invest in\nimprovements to the property, as well as reducing potential for land conflicts. The lack of property ownership/use\ndata also negatively affects tax revenue and effective planning for growth and development.\n\n\nLand disputes are a common feature of both rural and urban areas. Conflicts are tied to increasing population\npressure; conflicting land uses, such as grazing versus cultivation; proliferating peri-urban development, overlapping\nland concessions, and “land grabbing”. Women’s land rights are relatively well-supported in Tanzania’s formal legal\nframework, but women’s land rights are often undermined as customary laws, which favor male inheritance, are\nwidely practiced.\n\n\nThe 1999 Village Land Act determines that village lands may be demarcated, and land use plans created to provide for\nCertificates of Village Land (CVL) as well as Certificates of Customary Rights of Occupancy (CCROs) for people living\nwithin the villages. The\n\nMention: \"2018 Doing Business Report\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Report ranking Tanzania on the ease of registering property.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000095:2:0", "occurrence_id": "5f32295f065a0ed438e53a19", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nLand Tenure Improvement Project (P164906)\n\n\n|International Development Association (IDA)|150.00|\n|---|---|\n|IDA Credit|150.00<br>|\n\n\n\n\n\n\n\nOther Decision (as needed)\n\n\n**B. Introduction and Context**\n\n1. **Tanzania has been one of the stronger performers in Sub-Saharan Africa over the last decade**\n**but more needs to be done to achieve its development vision** . While the last decade saw a reduction in\noverall inequality and an increase in the rate of consumption growth of the bottom 40 percent, large\nincome and welfare differences exist between urban and rural citizens. Tanzania’s stable political and\nmacroeconomic environment and diverse natural resources provide many opportunities for inclusive\ngrowth. For Tanzania to realize the potential presented by these strategic advantages, it will need to\nstrengthen institutional capacity, create an open and competitive policy environment, promote publicprivate partnerships to raise investment levels to fill the large infrastructure gaps, raise human capital\nlevels and coordinate across government to manage Tanzania’s rich natural resources for equitable\ngrowth.\n\n2. **Tanzania continues to achieve strong economic growth but the reduction in poverty is fragile.**\nAccording to official data, real GDP grew by 7 percent in 2018 and 6.6 percent in Q1 2019. Growth is\nprojected near 6 percent over the medium-term subject to implementation of business environment and\nfiscal management reforms and a benign external environment. According to the Tanzania Poverty\nAssessment <sup>1</sup> [^1: Tanzania Mainland Poverty Assessment,\n<u>[http://documents.worldbank.org/curated/en/431111575939381087/Executive-Summary](http://documents.worldbank.org/curated/en/431111575939381087/Executive-Summary)</u>] between 2007 and 2018 the poverty rate fell from 34.4 to 26.4 percent, and extreme poverty\nfeel from 12 percent to 8 percent. Importantly poverty\n\nMention: \"Tanzania Poverty\nAssessment\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Assessment cited for the medium-term growth and poverty outlook.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000095:2:1", "occurrence_id": "fe9b0923a91649a40549ec98", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  from 12 percent to 8 percent. Importantly poverty rates fell for both urban and rural areas – from 20\nto 15.8 percent and from 39 to 31 percent respectively. However, vulnerability remains high and poverty\nreduction has not kept pace with population growth: for every four Tanzanians who moved out of poverty,\nthree fell into it. This vulnerability is likely to be exacerbated by crises such as the coronavirus outbreak\nwhich will further impact the poor and vulnerable.\n\n3. **Land rights are key for the economic recovery and building resilience to limit Tanzania’s**\n**vulnerability to Covid-19 and future disasters** . Once the immediate COVID-19 crisis passes, the proposed\ninvestments under the project position the country well, particularly to protect the poor’s most valuable\nassets (land and property) and because access to geo-spatial data sets that could be overlaid with various\n\n\n1 Tanzania Mainland Poverty Assessment,\n<u>[http://documents.worldbank.org/curated/en/431111575939381087/Executive-Summary](http://documents.worldbank.org/curated/en/431111575939381087/Executive-Summary)</u>\n\n\nDec 05, 2019 Page 3 of 14\n\nMention: \"geo-spatial data sets\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"inform\", \"summary\": \"Spatial layers discussed as overlayable with property and asset data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000095:2:1", "occurrence_id": "3a19fb58c2594f8444f576be", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  from 12 percent to 8 percent. Importantly poverty rates fell for both urban and rural areas – from 20\nto 15.8 percent and from 39 to 31 percent respectively. However, vulnerability remains high and poverty\nreduction has not kept pace with population growth: for every four Tanzanians who moved out of poverty,\nthree fell into it. This vulnerability is likely to be exacerbated by crises such as the coronavirus outbreak\nwhich will further impact the poor and vulnerable.\n\n3. **Land rights are key for the economic recovery and building resilience to limit Tanzania’s**\n**vulnerability to Covid-19 and future disasters** . Once the immediate COVID-19 crisis passes, the proposed\ninvestments under the project position the country well, particularly to protect the poor’s most valuable\nassets (land and property) and because access to geo-spatial data sets that could be overlaid with various\n\n\n1 Tanzania Mainland Poverty Assessment,\n<u>[http://documents.worldbank.org/curated/en/431111575939381087/Executive-Summary](http://documents.worldbank.org/curated/en/431111575939381087/Executive-Summary)</u>\n\n\nDec 05, 2019 Page 3 of 14\n\nMention: \"Tanzania Mainland Poverty Assessment\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"Footnoted assessment behind the poverty discussion.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000095:3:1", "occurrence_id": "9d35f3fc7d64ea6306be12a9", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  poverty, up from\n12.3 million in 2012 and 13.2 million in 2007. Agriculture contributes about 31.5 percent to the GDP and\nabout 73 percent of Tanzanian households depend on agriculture for their livelihoods.\n\n\nSectoral and Institutional Context\n\n6. **The Constitution of Tanzania provides that every person has the right to own property** . All land\nin Tanzania is public land vested in the President, as a trustee for and on behalf of all the citizens of\nTanzania. The 1995 National Land Policy (NLP) led to the development and passage of the 1999 Village\nLand Act and 1999 Land Act, which established three categories of land: General, Reserved and Village\nLand. About 68% of all land is Village Land; 30% is Reserved Land and only 2% is General Land. In 2005 the\nGovernment prepared the Strategic Plan for the Implementation of Land Laws (SPILL), which was revised\nin 2013 but has only been partly implemented. If adequately implemented and enforced, the legal\nframework provides a basis for equitably recognizing and protecting women’s land rights and pastoral and\ncommunal rights to land and natural resources.\n\n7. **Customary land rights are protected** . The 1999 Village Land Act recognizes the rights of village\nresidents to collectively hold land (both communal land and land that has been transferred to individuals)\nthat they have traditionally used under customary law. The Village Land Act determines that village lands\nmay be demarcated, and land use plans created to provide for Certificates of Village Land (CVL) and\nCertificates of Customary Rights of Occupancy (CCROs) for people living within the villages. As of now, less\nthan 10 percent of approximately 6 million households have received CCROs. Obtaining any form of land\n\n\n2 IMF DataMapper, October 2019.\n3 Tanzania Economic Update. Issue 12. July 2019.\n4 World Bank Open Data. https://data.worldbank.org/\n\n\nDec 05, 2019 Page 4\n\nMention: \"IMF DataMapper\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"Fund data tool cited in the footnote series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000095:3:1", "occurrence_id": "4f1dc4cc887384aa946cf4ef", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  poverty, up from\n12.3 million in 2012 and 13.2 million in 2007. Agriculture contributes about 31.5 percent to the GDP and\nabout 73 percent of Tanzanian households depend on agriculture for their livelihoods.\n\n\nSectoral and Institutional Context\n\n6. **The Constitution of Tanzania provides that every person has the right to own property** . All land\nin Tanzania is public land vested in the President, as a trustee for and on behalf of all the citizens of\nTanzania. The 1995 National Land Policy (NLP) led to the development and passage of the 1999 Village\nLand Act and 1999 Land Act, which established three categories of land: General, Reserved and Village\nLand. About 68% of all land is Village Land; 30% is Reserved Land and only 2% is General Land. In 2005 the\nGovernment prepared the Strategic Plan for the Implementation of Land Laws (SPILL), which was revised\nin 2013 but has only been partly implemented. If adequately implemented and enforced, the legal\nframework provides a basis for equitably recognizing and protecting women’s land rights and pastoral and\ncommunal rights to land and natural resources.\n\n7. **Customary land rights are protected** . The 1999 Village Land Act recognizes the rights of village\nresidents to collectively hold land (both communal land and land that has been transferred to individuals)\nthat they have traditionally used under customary law. The Village Land Act determines that village lands\nmay be demarcated, and land use plans created to provide for Certificates of Village Land (CVL) and\nCertificates of Customary Rights of Occupancy (CCROs) for people living within the villages. As of now, less\nthan 10 percent of approximately 6 million households have received CCROs. Obtaining any form of land\n\n\n2 IMF DataMapper, October 2019.\n3 Tanzania Economic Update. Issue 12. July 2019.\n4 World Bank Open Data. https://data.worldbank.org/\n\n\nDec 05, 2019 Page 4\n\nMention: \"World Bank Open Data\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"Open data portal cited for the underlying statistics.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000098:5:0", "occurrence_id": "41c22e6c0a05463d93912316", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nTanzania Roads to Inclusion and Socioeconomic Opportunities (RISE) Program (P164920)\n\n\nactivities; for instance, women still transport cobs from the field to the house head-loaded, while men use\nwheel barrels or motorbikes. <sup>3</sup> [^3: Fox L. Wiggins S., Ludi E., Mdee A. 2018. “The Lives of Rural Women and Girls. What does an Inclusive Rural Transformation that Empowers\nWomen Look Like?” London: ODI, page 132.] Furthermore, data show the prevalence of different forms of violence\nagainst women in Tanzania: the National Demographic and Health Survey (DHS) reveals concerning\nstatistics regarding women experiencing physical violence with greater incidence in rural areas where\nalmost 42 percent of women ages 15–49 have experienced physical violence, compared to 35.5 percent\nin urban areas. While Tanzania has legislation that addresses sexual harassment (SH) in the workplace,\nthere are no laws on domestic violence. <sup>4</sup> [^4: Systematic Country Diagnostic (SCD)/National DHS, 2010.]\n\n\n11. **A faster path for economic growth will be largely dependent on the swift actions and policies**\n**decided in the short and medium term.** Efforts linked to job creation and support to rural population with\naccess to health and educational services and enhancement of the sectoral conditions for agriculture to\nimprove food security will be critical. Labor-intensive interventions, such as those in the construction and\nmaintenance of roads will smooth or complement incomes for the population in the rural areas. Resilient\nroad infrastructure and transport services becomes essential to ensure continuation of lifeline mobility\nand connectivity and avoid disruptions in trade, especially in agricultural value chains that are core to\nensure food security of the country, and to ensure access to essential services.\n\n\nSectoral and Institutional Context\n\n\n12. **Road accessibility is one of the necessary conditions for inclusion and socioeconomic**\n**opportunities.** Safe\n\nMention: \"National Demographic and Health Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Survey reporting physical violence prevalence by rural residence.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000098:6:1", "occurrence_id": "5962fba2888918da5a909844", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  15,000 km are passable only by four-wheel drive vehicles. <sup>7</sup> [^7: Source: Draft National Transport Policy 2017.] Besides, the\nrural access index (RAI) for Tanzania, an indicator that measures the proportion of rural people who have\naccess to an all-season road within a distance of 2 km, is 24.6 percent. <sup>8</sup> [^8: Transport and ICT. 2016. Measuring Rural Access: Using New Technologies. Washington DC: World Bank, License: Creative Commons\nAttribution CC BY 3.0.] This means that approximately 33\nmillion rural people lack access to all-season roads.\n\n\n17. **Maintenance of the network is a major challenge for Tanzania’s government and its**\n**communities.** There is a substantial gap between the demand and the supply of maintenance funds. The\navailable funds provided by the Road Fund Board (RFB) <sup>9</sup> (responsible agency for the maintenance funding)\nonly support 72 percent and 36 percent of the routine and periodic maintenance demands respectively.\nFor FY18/19 the district road network received only 25.1 percent of the budget needed for maintenance.\nGiven the financial constraints for maintenance funding, improved road asset management planning and\nengineering practices will be needed both at the national level with regards to the policy and planning\nframework, and at the districts level. Furthermore, maintenance budget limitations and planning, in the\ncontext of climate change vulnerabilities (even more acute in the case of unpaved regional and district\nnetwork segments), are significantly hindering Tanzania’s capacity to provide all-season road access to its\nrural communities.\n\n\n18. **Tanzania is among the countries with the highest road traffic fatality rates, with 33.4 deaths per**\n**100,000 people, which has a significant impact on its GDP.** In 2013, there were more than 4,000 reported\n\n\n5 World Bank.\n\nMention: \"rural access index\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Indicator measuring rural access to all-season roads.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000100:4:0", "occurrence_id": "0f3fb0d716839c7e6c36db05", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nTanzania Roads to Inclusion and Socioeconomic Opportunities (RISE) Project (P164920) – Additional Financing\nRequest 3\n\n\n\n\n\n<mark>Other Decision (as needed)</mark>\n\n\n**B. Introduction and Context**\n\nCountry Context\n\n\n1. **Tantania’s economy has remained resilient, growing by about 5.8 percent in the first half of 2025 with growth**\n**projected at 6.2 percent in 2026, but poverty reduction continues to lag** . The absolute poverty rate is projected to\nfall only marginally, from 42.9 percent in 2024 to 41.7 percent by 2026, so that the number of Tanzanians living in\npoverty continues to rise alongside a population that has grown from 61.7 million at the 2022 Census to over 70 million\ntoday economy.\n\n2. **Growth is increasingly urban, and urbanization is outpacing the capacity to manage it.** The urban population\nis growing at 4.8 percent annually, among the fastest rates in Sub-Saharan Africa, and now accounts for 36.9 percent\nof the total. Cities generate more than half of national GDP, but their expansion has been largely informal and spatially\nunmanaged, pushing settlement onto low-lying, hazard-exposed land and concentrating disaster risk in the country's\nmost productive places.\n\n\n3. **Tanzania is the most flood-affected country in East Africa and ranks among the most climate-vulnerable**\n**globally, with floods and droughts.** Between late 2023 and early 2024, Tanzania experienced above-average rains for\ntwo consecutive rainy seasons: October–December 2023 (Vuli rains) and March–May 2024 (Masika rains). The heavy\nrains, related to the El-Niño phenomenon, caused a series of floods, landslides, and debris flows across all 26 regions\nof Tanzania, claiming lives and affecting livelihoods, settlements, farmlands\n\nMention: \"2022 Census\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Census population baseline behind the population growth statement.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000102:5:1", "occurrence_id": "b33e6516e860e09f08dbd718", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  PINs). International\nDiagnostic Coding Standards (ICD-10) is used from the most providers, but it is not a clear level of encoding\nquality because as stated before the audit process is not methodical. In addition, the problem is the\nprocessing and the control of primary care benefit (PCB) from rural health units (RHU) on PHIC side. One of\nthe reasons is that due to the technical difficulties and lack of effective project management, implementation\nof Electronic Claims Submission (eClaims) system is in delays on provider’s side. Hospitals complain that PHIC\ndoes not provide enough up-to-date information on the rights of the PHIC’s members (Online Claims\nEligibility Verification), so it may happen that the patient uses the same benefits in more places than he/she\nis eligible for. Analysis of how much unjustifiable reimbursement annually PHIC has paid to the health care\nprovider (HCP) because of incurrence of data is not known during this review.\n\n\n**Philhealth’s monitoring and oversight functions**\n\n\nPopulation based data is very necessary to monitor progress towards UHC: On financial protection - family\nincome-expenditure survey (FIES) is currently led by the National Bureau of Statistics and conducted every 3years. On service coverage - demographic health survey (DHS) is currently conducted every 5 years. For\nsampling, the population is represented at the regional but not province (or district level). There is ongoing\ndiscussion if the population representation can be at the province level. As administrative data remains\nweak, and to compliment, these surveys have been key to monitoring financial protection and health outputs\nand outcomes in the country.\n\n\n**Weak analytical capacity of PhilHealth** is one of the key obstacles. This limits DOH and Philhealth’s ability to\nimprove their monitoring and oversight functions (a) to prevent potential manipulations, and (b) to adjust\n\n\nFeb 14, 2018 Page 6\n\nMention: \"family\nincome-expenditure survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Survey led by the statistics bureau monitoring financial protection.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000102:5:1", "occurrence_id": "e6770fbc5bac797b80406d41", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  PINs). International\nDiagnostic Coding Standards (ICD-10) is used from the most providers, but it is not a clear level of encoding\nquality because as stated before the audit process is not methodical. In addition, the problem is the\nprocessing and the control of primary care benefit (PCB) from rural health units (RHU) on PHIC side. One of\nthe reasons is that due to the technical difficulties and lack of effective project management, implementation\nof Electronic Claims Submission (eClaims) system is in delays on provider’s side. Hospitals complain that PHIC\ndoes not provide enough up-to-date information on the rights of the PHIC’s members (Online Claims\nEligibility Verification), so it may happen that the patient uses the same benefits in more places than he/she\nis eligible for. Analysis of how much unjustifiable reimbursement annually PHIC has paid to the health care\nprovider (HCP) because of incurrence of data is not known during this review.\n\n\n**Philhealth’s monitoring and oversight functions**\n\n\nPopulation based data is very necessary to monitor progress towards UHC: On financial protection - family\nincome-expenditure survey (FIES) is currently led by the National Bureau of Statistics and conducted every 3years. On service coverage - demographic health survey (DHS) is currently conducted every 5 years. For\nsampling, the population is represented at the regional but not province (or district level). There is ongoing\ndiscussion if the population representation can be at the province level. As administrative data remains\nweak, and to compliment, these surveys have been key to monitoring financial protection and health outputs\nand outcomes in the country.\n\n\n**Weak analytical capacity of PhilHealth** is one of the key obstacles. This limits DOH and Philhealth’s ability to\nimprove their monitoring and oversight functions (a) to prevent potential manipulations, and (b) to adjust\n\n\nFeb 14, 2018 Page 6\n\nMention: \"demographic health survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Survey measuring service coverage for universal health coverage.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000105:3:0", "occurrence_id": "e55200356d08cdaf407f1726", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nGroundnut Competitiveness and Agriculture Diversification Project (P164967)\n\n\nnascent structural transformation. In urban areas, labor income among the poor has been boosted by growth in laborintensive sectors, such as construction.\n\n\n5. **However persistent spatial disparities are evidenced, with rural areas- particularly the Groundnut Basin (GNB)-**\n**lagging behind their urban counterparts on a wide range of non-monetary indicators** . In 2011, nearly 70 percent of the\npoor and 84 percent of the extreme poor lived in rural areas, accounting for 57 percent of the total population and\nsuggesting rural poverty to be deeper and more severe. Female-headed households were less likely to be poor, as women\ntend to head smaller households, and are more likely to receive cash and in-kind transfer <sup>1</sup> [^1: Moreover, due to self-selection women, heading their households, may represent a sub-sample of women with different preferences, norms, and\ncharacteristics.] . The poorest communes are\nlocated in the center of the country known as the Groundnut Basin (particularly in the region of Diourbel, Fatick, Kaolack\nand Kaffrine), and in the Kolda (south) and the eastern part of Tambacounda (south-east), which is also part of the\nextended groundnut basin. The poverty profile constructed using the DHS wealth indicators for 2015 suggests that the\ncharacteristics of the poor have largely remained unchanged since 2011 in terms of geographic distribution, households’\nsize and education levels. The extended Groundnut Basin is the country’s bread basket, supplying key staples (mostly\nmillet, sorghum, and maize) and fodder from groundnuts and cowpeas. Those areas have the highest population density\n(100-120 people/km2) after the capital city area.\n\n\n6. **Agriculture employs the largest share of the poor, particularly in the Groundnut Basin. Indeed, 62 percent of**\n**the rural population depend on agriculture**, which employs more than 36 percent\n\nMention: \"DHS wealth indicators\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Wealth indicators used to construct the poverty profile.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000105:3:1", "occurrence_id": "69db3ff675d3146e542ebfdd", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  after the capital city area.\n\n\n6. **Agriculture employs the largest share of the poor, particularly in the Groundnut Basin. Indeed, 62 percent of**\n**the rural population depend on agriculture**, which employs more than 36 percent of the total labor force of the country,\naccording to the 2015 Labor Force Survey, and poverty incidence among farmers is the highest. In rural areas, nearly 80\npercent of paid workers in the bottom quintile of the distribution of consumption worked in agriculture, while the more\nprosperous are engaged in largely informal services. Among the poorest are the farmers growing rain-fed crops particularly\nin the Groundnut Basin. Households having animal production as a main source of livelihood are also among the poorest.\nOn the other hand, poverty is at its lowest rates among horticulture farmers, mainly based in the Niayes, on the coast\nbetween Saint-Louis and Dakar, or rice producers along the Senegal River valley. Groundnut farming is a primary source\nof income for a vast majority of family farms, accounting for about half of the total cropped area in the country and\nemploying two-thirds of the rural population, mostly below the official poverty line. In fact, groundnuts are grown by 27\npercent of all households in Senegal, and by 52 percent of households in extreme poverty (ANSD, 2011). Moreover,\nsupporting labor-intensive activities down the value chain, such as transportation, storage, and processing, remains an\nimportant source of non-farm employment in both formal and informal sectors.\n\n\n7. **While the economic prospects are good with a projected real GDP growth at 7.0% in 2019, much more needs to**\n**be done in implementing the Government’s longer-term strategy (PSE) to transform the rural economy and reduce**\n**poverty and spatial imbalances.** Supported by robust macroeconomic fundamentals, Senegal is expected to remain one\nof West Africa’s top\n\nMention: \"2015 Labor Force Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Survey share of the labour force employed in agriculture.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000106:84:2", "occurrence_id": "76daac959a9754d0f3403780", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Women’s quality of work lower\nthan men (SCD 2018) <sup>47</sup>\n\n\n\n\n- Women’s land rights, particularly\nfemale-headed household <sup>49</sup>\n\n\n\n\n- Right to sell or use land as\ncollateral by urban/rural (Annex 2)\n\n\n\n\n- N/A\n\n\n\n\n- N/A - When prioritizing investments\nfor the upgrading pilot in\nAgatare under RUDP,\npreferences and women and\nmen differed\n\n\n\n**Voice and**\n**agency**\n\n\n\n\n- Women’s participation in\nleadership at the sub-national\nlevel (SCD 2018)\n\n- Participation in the government\nplanning and implementation\n<u>process</u> <sup>50</sup>\n\n\n\n3. **Although Rwanda’s commitment to gender equality is widely recognized, serious gaps remain and progress on**\n**these areas can be instrumental in helping Rwanda meet its development objectives** . Areas where further attention is\nneeded include: stalled progress reducing fertility after prior impressive reductions; translating gains in human capital into\n\n\n47 Only 2.5 percent of women in Rwanda work as managers (0.1 percent), professionals (2.2 percent) and technical and associate professionals (0.2\npercent) according to the Fifth Integrated Household Living Survey (EICV5) for 2016/17.\n48 Hitayezu et al. (2018) The dynamics of unplanned settlements in the City of Kigali. International Growth Centre.\n49 Ann-Sofie Isaksson (2015) Unequal Property Rights: A Study of Land Right Inequalities in Rwanda, Oxford Development Studies, 43:1, 60-83, DOI:\n10.1080/13600818.2014.955466\n50 Never Again Rwanda (2018) Local Government Imihigo Process: Understanding the factors contributing to low participation.\n\n\nPage 80 of 95\n\nMention: \"Fifth Integrated Household Living Survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Survey shares of women in managerial and professional occupations.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000106:84:2", "occurrence_id": "4fe051e4e85f74b50ad9e8ee", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  Women’s quality of work lower\nthan men (SCD 2018) <sup>47</sup>\n\n\n\n\n- Women’s land rights, particularly\nfemale-headed household <sup>49</sup>\n\n\n\n\n- Right to sell or use land as\ncollateral by urban/rural (Annex 2)\n\n\n\n\n- N/A\n\n\n\n\n- N/A - When prioritizing investments\nfor the upgrading pilot in\nAgatare under RUDP,\npreferences and women and\nmen differed\n\n\n\n**Voice and**\n**agency**\n\n\n\n\n- Women’s participation in\nleadership at the sub-national\nlevel (SCD 2018)\n\n- Participation in the government\nplanning and implementation\n<u>process</u> <sup>50</sup>\n\n\n\n3. **Although Rwanda’s commitment to gender equality is widely recognized, serious gaps remain and progress on**\n**these areas can be instrumental in helping Rwanda meet its development objectives** . Areas where further attention is\nneeded include: stalled progress reducing fertility after prior impressive reductions; translating gains in human capital into\n\n\n47 Only 2.5 percent of women in Rwanda work as managers (0.1 percent), professionals (2.2 percent) and technical and associate professionals (0.2\npercent) according to the Fifth Integrated Household Living Survey (EICV5) for 2016/17.\n48 Hitayezu et al. (2018) The dynamics of unplanned settlements in the City of Kigali. International Growth Centre.\n49 Ann-Sofie Isaksson (2015) Unequal Property Rights: A Study of Land Right Inequalities in Rwanda, Oxford Development Studies, 43:1, 60-83, DOI:\n10.1080/13600818.2014.955466\n50 Never Again Rwanda (2018) Local Government Imihigo Process: Understanding the factors contributing to low participation.\n\n\nPage 80 of 95\n\nMention: \"EICV5\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Same survey round cited for occupational and income shares.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000106:85:2", "occurrence_id": "3119d5bf3383b6c77a17ce93", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  better** Women’s ownership of a - Create a joint-bank account for wife and husband in case of compensation for\n**jobs/access to** bank account is lower land acquisition. This is the continuation of the current practice under RUDP\n**finance** than men (both nationally whereby the City of Kigali compensated people affected by the upgrading pilot\n\n\n\nWomen’s ownership of a\nbank account is lower\nthan men (both nationally\n<u>and in urban areas)</u>\n\n\n\n\n- Create a joint-bank account for wife and husband in case of compensation for\nland acquisition. This is the continuation of the current practice under RUDP\nwhereby the City of Kigali compensated people affected by the upgrading pilot\n<u>in Agatare through cash transfer to a joint account.</u>\n\n\n\n51 This refers to the average monthly cash income from employment at main job by sex, age group, level of education, occupation group and\nurban/rural area, RLFS 2018\n52 This refers to the percentage of individuals (18+) with bank account by sex, and Area of residence and consumption quintile (EICV5, EICV4)\n53 This refers to the percentage of households’ head with the right to sell or use land as collateral by urban/rural, province and consumption quintile\n(EICV5). In comparison, percentage of households with the loan from formal sources of credit using houses as collateral is similar between male and\nfemale headed households (17.6 percent and 16.7 percent) nationally and even higher for female-headed households in urban areas (31.6 percent\nversus 40.6 percent).\n54 Hitayezu, P., Rajashekar, A., & Stoelinga, D. (2018). The dynamics of unplanned settlements in the City of Kigali. International Growth Centre.\n\n\nPage 81 of 95\n\nMention: \"RLFS 2018\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Labour force survey behind monthly cash income by sex and area.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000107:3:1", "occurrence_id": "96dbd047fc76815a0f1936ce", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  line declining. Since 2014,\nhowever, the rate of poverty reduction has decreased due to the impacts of the 2016 drought, coupled with a slow down\nin structural transition. Unusually high food prices following the drought resulted in households significantly reducing their\nfood consumption. <sup>4</sup> The combination of low farm productivity and a high concentration of the population in rural areas\nhas also contributed to the low elasticity of poverty reduction in recent years. <sup>5</sup>\n\n\n2 Diao, Randriamamonjy, and Thurlow, 2017.\n3 The contribution of urbanization to growth depends on the productivity gap between similar sectors in rural and urban areas, which occurs\nbecause (a) urban workers are generally better educated and higher skilled and (b) capital-labor ratios tend to be higher in urban sectors where\ncapital investments are more likely to be concentrated.\n4 World Bank and National Institute of Statistics Rwanda (NISR) household surveys. The impact of the heavy rainfalls in early 2018 is yet to be seen\nas it is not included in the latest household survey released in late 2018.\n5 Rwanda Systematic Country Diagnostic (SCD), Draft Concept Note, March 21, 2019.\n\n\nApr 14, 2019 Page 4 of 17\n\nMention: \"household surveys\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"National statistics household surveys behind the welfare discussion.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000107:3:1", "occurrence_id": "f3e7fb29cf05d33f708e705d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  line declining. Since 2014,\nhowever, the rate of poverty reduction has decreased due to the impacts of the 2016 drought, coupled with a slow down\nin structural transition. Unusually high food prices following the drought resulted in households significantly reducing their\nfood consumption. <sup>4</sup> The combination of low farm productivity and a high concentration of the population in rural areas\nhas also contributed to the low elasticity of poverty reduction in recent years. <sup>5</sup>\n\n\n2 Diao, Randriamamonjy, and Thurlow, 2017.\n3 The contribution of urbanization to growth depends on the productivity gap between similar sectors in rural and urban areas, which occurs\nbecause (a) urban workers are generally better educated and higher skilled and (b) capital-labor ratios tend to be higher in urban sectors where\ncapital investments are more likely to be concentrated.\n4 World Bank and National Institute of Statistics Rwanda (NISR) household surveys. The impact of the heavy rainfalls in early 2018 is yet to be seen\nas it is not included in the latest household survey released in late 2018.\n5 Rwanda Systematic Country Diagnostic (SCD), Draft Concept Note, March 21, 2019.\n\n\nApr 14, 2019 Page 4 of 17\n\nMention: \"household survey\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Most recent survey round noted as pre-dating the 2018 rainfall impact.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000107:4:0", "occurrence_id": "c71fb444b75c4f66762fa1ce", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nRwanda Urban Development Project II (P165017)\n\n\n_Figure_ 2 _. Poverty trends in Rwanda (left) and Rural vs. Urban poverty trends (right)_\n\n\n_Source: NISR (2015, 2018)_\n\n\n5. **As evidenced by the 2016 drought, Rwanda is vulnerable to the impacts of climate change and efforts to**\n**improve adaptive capacity against climate risks and hazards are needed in order to sustain its growth trajectory.** Since\n1970, Rwanda’s average temperature has increased by 1.4˚C and under some scenarios is projected to increase by 2˚C by\nthe 2030s. <sup>**6**</sup> [^6: Rwanda Strategic Program for Climate Resilience. 2017. http://www.fonerwa.org/sites/default/files/SPCR.pdg] Analysis of rainfall trends shows an increasing occurrence of extreme weather events over time with rainy\nseasons becoming shorter and more intense. This greatly increases both flood hazards and erosion risks in mountainous\nparts of the country. Seasonal droughts are recurring and expected to be prolonged going forward. Droughts result in\nsharp reductions in food consumption and can lead to famine, loss of animals, depletion of water resources and disease\noutbreaks. Rwanda’s exposure to weather risks not only undermines its growth prospects but also slows down poverty\nreduction as a result of inflation and reduced food consumption among the poor.\n\n6. **Going forward, Rwanda faces the challenge of sustaining historically high growth rates and ensuring it is coupled**\n**with poverty reduction and increases in shared prosperity** . While Rwanda’s medium term macroeconomic outlook is\nfavorable, with GDP growth projected at the rate of 7 to 8 percent per annum, sustained by improved agriculture; strong\nexports, both traditional and nontraditional; and large infrastructure projects, such as the current construction of the new\nairport. <sup>7</sup> There are, however, several\n\nMention: \"Poverty trends in Rwanda\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Figure series of national and rural-urban poverty trends.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000107:4:0", "occurrence_id": "42caa3049df956441cf769fe", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nRwanda Urban Development Project II (P165017)\n\n\n_Figure_ 2 _. Poverty trends in Rwanda (left) and Rural vs. Urban poverty trends (right)_\n\n\n_Source: NISR (2015, 2018)_\n\n\n5. **As evidenced by the 2016 drought, Rwanda is vulnerable to the impacts of climate change and efforts to**\n**improve adaptive capacity against climate risks and hazards are needed in order to sustain its growth trajectory.** Since\n1970, Rwanda’s average temperature has increased by 1.4˚C and under some scenarios is projected to increase by 2˚C by\nthe 2030s. <sup>**6**</sup> [^6: Rwanda Strategic Program for Climate Resilience. 2017. http://www.fonerwa.org/sites/default/files/SPCR.pdg] Analysis of rainfall trends shows an increasing occurrence of extreme weather events over time with rainy\nseasons becoming shorter and more intense. This greatly increases both flood hazards and erosion risks in mountainous\nparts of the country. Seasonal droughts are recurring and expected to be prolonged going forward. Droughts result in\nsharp reductions in food consumption and can lead to famine, loss of animals, depletion of water resources and disease\noutbreaks. Rwanda’s exposure to weather risks not only undermines its growth prospects but also slows down poverty\nreduction as a result of inflation and reduced food consumption among the poor.\n\n6. **Going forward, Rwanda faces the challenge of sustaining historically high growth rates and ensuring it is coupled**\n**with poverty reduction and increases in shared prosperity** . While Rwanda’s medium term macroeconomic outlook is\nfavorable, with GDP growth projected at the rate of 7 to 8 percent per annum, sustained by improved agriculture; strong\nexports, both traditional and nontraditional; and large infrastructure projects, such as the current construction of the new\nairport. <sup>7</sup> There are, however, several\n\nMention: \"rainfall trends\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Rainfall trend analysis showing more frequent extreme weather events.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000107:5:0", "occurrence_id": "32fee4605a3afb767f5f8d72", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nRwanda Urban Development Project II (P165017)\n\n\npopulation. With population growth of 6.4 percent per year from 2002 to 2012, Kigali is one of the fastest-growing cities\nin Africa and the most favored destination for rural migrants. <sup>9</sup> [^9: In 2014, 57 percent of Kigali’s population had migrated to Kigali at some point in their lives, and 14 percent had migrated from rural areas in the\npast three years (World Bank 2017).] Half of the urban population outside Kigali is found along\ntwo urban corridors: the Musanze-Nyabihu-Rubavu corridor (one-third) and the Muhanga-Huye corridor including Nyanza\nand Ruhango districts (18 percent). Much of Rwanda’s remaining urban population is spread between the roads\nconnecting Kigali to Bugesera, Kayonza, the Burundian border, and the more isolated settlements of Nyagatare and\nRusizi. <sup>10</sup> [^10: Diao, Randriamamonjy, and Thurlow, 2017.] The evolution of urban settlements and emerging urban corridors are evidenced in Figure 3.\n\n_Figure 3. Evolution of urban settlements between 2002 (left) and 2015 (right), using WorldPop datasets_ <sup>_11_</sup> [^11: The 2015 WorldPop datasets for Rwanda was used as the population map.\n<u>[http://www.worldpop.org.uk/data/summary/?doi=10.5258/SOTON/WP00223](http://www.worldpop.org.uk/data/summary/?doi=10.5258/SOTON/WP00223)</u>]\n\n\n_Source: Reshaping urbanization in Rwanda, Synthesis Note, World Bank, 2017._\n\n8. **Rwanda’s urban system has been shaped within the context of an emerging policy framework that has**\n**developed since the early 2000s.** In 2000, Rwanda Vision 2020 acknowledged the need to develop urban infrastructure\nin view of\n\nMention: \"WorldPop datasets_\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Gridded population maps used to show settlement evolution.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000107:5:0", "occurrence_id": "abfa915971ac7e00a8e0ac85", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nRwanda Urban Development Project II (P165017)\n\n\npopulation. With population growth of 6.4 percent per year from 2002 to 2012, Kigali is one of the fastest-growing cities\nin Africa and the most favored destination for rural migrants. <sup>9</sup> [^9: In 2014, 57 percent of Kigali’s population had migrated to Kigali at some point in their lives, and 14 percent had migrated from rural areas in the\npast three years (World Bank 2017).] Half of the urban population outside Kigali is found along\ntwo urban corridors: the Musanze-Nyabihu-Rubavu corridor (one-third) and the Muhanga-Huye corridor including Nyanza\nand Ruhango districts (18 percent). Much of Rwanda’s remaining urban population is spread between the roads\nconnecting Kigali to Bugesera, Kayonza, the Burundian border, and the more isolated settlements of Nyagatare and\nRusizi. <sup>10</sup> [^10: Diao, Randriamamonjy, and Thurlow, 2017.] The evolution of urban settlements and emerging urban corridors are evidenced in Figure 3.\n\n_Figure 3. Evolution of urban settlements between 2002 (left) and 2015 (right), using WorldPop datasets_ <sup>_11_</sup> [^11: The 2015 WorldPop datasets for Rwanda was used as the population map.\n<u>[http://www.worldpop.org.uk/data/summary/?doi=10.5258/SOTON/WP00223](http://www.worldpop.org.uk/data/summary/?doi=10.5258/SOTON/WP00223)</u>]\n\n\n_Source: Reshaping urbanization in Rwanda, Synthesis Note, World Bank, 2017._\n\n8. **Rwanda’s urban system has been shaped within the context of an emerging policy framework that has**\n**developed since the early 2000s.** In 2000, Rwanda Vision 2020 acknowledged the need to develop urban infrastructure\nin view of\n\nMention: \"WorldPop datasets\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"Population map layer underlying the urban settlement figure.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000108:13:0", "occurrence_id": "a5dba9f4580c6d5e946ccf1d", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nSecond Rwanda Urban Development Project (P165017)\n\n\n**Component 3: Institutional Capacity Development and Project Management (US$ 10.67 million)**\n\n32. This component will support institutional and capacity development at the national level, overall project\ncoordination by MININFRA and project management costs across implementation agencies <sup>25</sup> [^25: Implementation agencies include MININFRA PCU, REMA SPIU, LODA SPIU, CoK KUUT, Huye PIU, Muhanga PIU, Musanze PIU, Nyagatare\nPIU, Rubavu PIU and Rusizi PIU.] at the national level,\nCoK and secondary cities.\n\n**_Subcomponent 3a: Institutional capacity development at national level (US$ 2.95 million)_**\n\n33. This subcomponent will support the ICD in the CoK and secondary cities through a set of TAs led by MININFRA\nin coordination with other relevant ministries including MINALOC and RHA, so as to effectively meet the\ninstitutional and capacity building needs in the CoK and secondary cities, while ensuring that all TAs are well\naligned with the mandates and policies/strategies of the national government. Technical support will be provided\nto develop a national urbanization strategy to monitor the performance and growth of urban areas in Rwanda,\nwhich includes setting up an urbanization monitoring information system. Another set of TAs at the national level\nwill be provided under the broad umbrella of the housing policy to build on and expand investments in urban\nupgrading and sites and services schemes under RUDP II. Finally, TA for the implementation of a national solid\nwaste management strategy will be supported through the development of feasibility studies of disposal facilities\nin the secondary cities. Under this subcomponent, the project will seek to disseminate knowledge generated from\ncity to city in Rwanda.\n\n\n**_Subcomponent 3b: Project management (US$ 7.72 million)_**\n\n34. Project management activities to be supported include fiduciary (financial management and procurement),\nenvironmental and social management, implementation supervision, contract management, monitoring and\nevaluation (M&E\n\nMention: \"urbanization monitoring information system\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"System to be set up to monitor urban performance and growth.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000108:3:0", "occurrence_id": "781857c7964c309bb1dc5655", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nSecond Rwanda Urban Development Project (P165017)\n\n\nWhile still a low-income country, Rwanda today is ahead of more than 20 countries in the region in terms of GDP\nper capita.\n\n2. **While sustained economic growth has led to significant poverty reduction, it has been stagnant in recent**\n**years.** Poverty declined from 60.4 to 38.2 percent of the population between 2000 and 2017 <sup>2</sup> [^2: Based on the Integrated Household Living Conditions Survey 5 (EICV5) - Poverty Panel Report] with an\nimprovement in social indicators. Poverty has also become less severe with the distance between poor\nhouseholds’ consumption from the poverty line declining. However, the rate of poverty reduction has decreased\nfollowing the 2016 drought, with unusual increase in food prices and reduction in household food consumption,\ncoupled with a slowdown in structural transition <sup>3</sup> [^3: World Bank and National Institute of Statistics Rwanda (NISR) household surveys. The impact of the heavy rainfalls in early 2018 is yet to\nbe seen as it is not included in the latest household survey released in late 2018.] The combination of low farm productivity and a high\nconcentration of the population in rural areas has also reduced the elasticity of poverty reduction in recent years. <sup>4</sup> [^4: Rwanda Systematic Country Diagnostic (SCD), Concept Note, May 29, 2019.]\nGoing forward, Rwanda faces the challenge of sustaining historically high growth rates and accelerating poverty\nreduction. Growth is estimated to have reached 8.5 percent in 2019 <sup>5</sup> [^5: Rwanda Economic Update 15th Edition, January 2020.] due to public investments and fiscal stimulus.\nAs the fiscal expansion for the NST1 subsides in the medium term, attention must turn to improving allocation of\neconomic resources through better market functioning and increased private sector investment.\n\n3.\n\nMention: \"Integrated Household Living Conditions Survey 5\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Poverty panel report behind the 2000-2017 decline.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000108:3:0", "occurrence_id": "fc34bb8c3dd60f6af26b6012", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nSecond Rwanda Urban Development Project (P165017)\n\n\nWhile still a low-income country, Rwanda today is ahead of more than 20 countries in the region in terms of GDP\nper capita.\n\n2. **While sustained economic growth has led to significant poverty reduction, it has been stagnant in recent**\n**years.** Poverty declined from 60.4 to 38.2 percent of the population between 2000 and 2017 <sup>2</sup> [^2: Based on the Integrated Household Living Conditions Survey 5 (EICV5) - Poverty Panel Report] with an\nimprovement in social indicators. Poverty has also become less severe with the distance between poor\nhouseholds’ consumption from the poverty line declining. However, the rate of poverty reduction has decreased\nfollowing the 2016 drought, with unusual increase in food prices and reduction in household food consumption,\ncoupled with a slowdown in structural transition <sup>3</sup> [^3: World Bank and National Institute of Statistics Rwanda (NISR) household surveys. The impact of the heavy rainfalls in early 2018 is yet to\nbe seen as it is not included in the latest household survey released in late 2018.] The combination of low farm productivity and a high\nconcentration of the population in rural areas has also reduced the elasticity of poverty reduction in recent years. <sup>4</sup> [^4: Rwanda Systematic Country Diagnostic (SCD), Concept Note, May 29, 2019.]\nGoing forward, Rwanda faces the challenge of sustaining historically high growth rates and accelerating poverty\nreduction. Growth is estimated to have reached 8.5 percent in 2019 <sup>5</sup> [^5: Rwanda Economic Update 15th Edition, January 2020.] due to public investments and fiscal stimulus.\nAs the fiscal expansion for the NST1 subsides in the medium term, attention must turn to improving allocation of\neconomic resources through better market functioning and increased private sector investment.\n\n3.\n\nMention: \"household surveys\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Joint surveys cited on consumption and structural transition.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000109:11:0", "occurrence_id": "20d23b2e8d33ed55980824e3", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nSecond Rwanda Urban Development Project (P165017)\n\n\nsector investments in areas such as hospitality, tourism and recreation, to accelerate the implementation of urban\nplanning solutions that center biodiversity, ecosystem services, and climate adaptation and mitigation. TA will be\nprovided to explore opportunities for partnerships with the private sector and investments in wetlands and\nwetland buffer zones. The intention is that this investment will serve to accelerate business development and\nsustainable use of wetlands by local communities and tourists to the City of Kigali.\n\n\n26. To support integrated urban planning, a city-wide highly detailed topographic model will be produced using\nan aerial LiDAR survey. This will be a critical knowledge product that serves multiple purposes and sectors and will\nbecome an invaluable tool for integrated urban planning. A technical assistance will also support CoK to develop\na framework for GHG accounting and framework, that will take into account the transport, construction, energy,\nenvironment, and solid waste management sectors. This framework will allow the estimation of GHG emission\nreductions from all the investments in the project, including low-carbon upgrading to wetland and solid waste\nmanagement activities. The framework will pave the way for significant GHG emission reductions across multiple\nsectors. The outcomes of these activities are key knowledge products which can be disseminated nationally and\nbeyond.\n\n\n27. The project will also support advocacy, knowledge exchange and partnerships on sustainable urbanization,\nbuilding upon the global SCIP as international knowledge is a key component of the GEF financing. The global\nprogram will enable Rwanda to scale up engagement from its secondary cities and the national level to regional\nand global levels. Kigali will both learn from the SCIP and create knowledge that will be shared through the\nplatform internationally. Knowledge generated from the project will be documented and products will be\ndisseminated to benefit cities across the country and internationally as well as to support the sustainability of\nproject outcomes. Project resources will be specifically allocated to allow urban policy experts to participate in\nglobal activities and forums. The project will\n\nMention: \"aerial LiDAR survey\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"source\", \"summary\": \"City-wide topographic model produced from the aerial survey.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000110:4:0", "occurrence_id": "e8cbb2831a612c39d4c9cb64", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nSecond Rwanda Urban Development Project (P165017) – Additional Financing Request 5\n\n\n\n\n\n<mark>Other Decision (as needed)</mark>\nTogether with the Second Additional Financing of US$10 million, approval is sought for the extension of the Second Rwanda\nUrban Development Project (RUDP II, Project) closing date from December 31, 2025, to December 31, 2026.\n\n\n**B. Introduction and Context**\n\nCountry Context\n\n1. **Rwanda has experienced strong, uninterrupted economic growth over the past two decades and is among the**\n**fastest growing economies in Africa.** Rwanda’s per capita income increased from US$201 in 2000 to US$801 in 2019 <sup>1</sup> [^1: Based on the World Development Indicators/World Bank] .\nEconomic growth averaged 7.7 percent per year during the same period compared to the regional average of 4.4 percent.\nThis was the second highest rate of economic growth in Sub-Saharan Africa. The annual percentage growth rate of Gross\nDomestic Product (GDP) reached 9.4 percent in 2019, accelerating from 8.6 percent in 2018 and 6.1 percent in 2017.\nAccelerated growth is attributed to robust performance across key sectors, as well as increased public investments for\nimplementation of the National Strategy for Transformation (NST1). Major economic sectors such as industry (propelled by\nconstruction) and services (driven by trade and transport) have grown at 8-10 percent annually since 2006, and agriculture\nat 5.4 percent. Public investments increased from 4.9 percent in 2003 to about 9.4 percent in 2018. While still a low-income\ncountry, Rwanda today is ahead of more than 20 countries in the region in terms of GDP per capita.\n\n2. **Sustainable urbanization has been one of the country’s key development strategies and has contributed**\n**significantly to economic growth and structural change in Rwanda.** The\n\nMention: \"World Development Indicators\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Indicator series behind the per capita income series.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000110:4:1", "occurrence_id": "d1a77d9dadd0d5db589b44e4", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  20 countries in the region in terms of GDP per capita.\n\n2. **Sustainable urbanization has been one of the country’s key development strategies and has contributed**\n**significantly to economic growth and structural change in Rwanda.** The Government has identified sustainable\nurbanization and off-farm job creation as critical for achieving its vision of becoming a middle-income country. Structural\ntransformation or the movement of labor out of low-productivity agriculture to industry and services, and from farms to\ncities, has indeed accompanied Rwanda’s growth in the 2000s. Urbanization has accounted for 37 percent of national\nstructural change (GDP growth through labor reallocation across sectors), and urban areas have accounted for 48 percent\nof national labor productivity growth over the past 15 years. <sup>2</sup> [^2: Diao, Randriamamonjy, and Thurlow. 2017. “Urbanization and the Future of Economic Development in Rwanda.” International Food Policy Research\nInstitute, Washington, DC. Background paper for Future Drivers of Growth in Rwanda.] Urbanization in Rwanda has also been accompanied by\npoverty reduction primarily in areas with high density and good connectivity although this effect tapers off as one moves\naway from city centers. <sup>3</sup> [^3: World Bank. 2017. Reshaping Urbanization in Rwanda: Economic and Spatial Trends and Proposals; Note 3: Urbanization, Job Creation, and Poverty\nReduction in Rwanda.] According to the National Census in 2012, nearly a third of the poor live in the periphery of\nsecondary cities (10-20km away from the center), where opportunities for poverty reduction exist by better connecting the\nperiphery with urban centers.\n\n\nSectoral and Institutional Context\n\n3. **Rapid and well-managed urbanization is essential for Rwanda to achieve its twin aspirations for growth and**\n**improved quality of life** . With faster urbanization, Rwanda’s economy could be 20 percent larger by 2050, if cities are well\nmanaged to efficiently connect jobs and homes and generate the kind of productivity\n\nMention: \"National Census\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Census share of the poor living in secondary city peripheries.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000113:3:2", "occurrence_id": "e6c0c357b6d618a1cab53e7e", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: farm jobs for rural women remain limited. New opportunities in agribusiness are\nimportant to reverse the decline in women’s participation in agriculture.\n\n5. **HP has a particular role to play in contributing to India’s INDC commitments.** Its topography and available water\nresources are well suited to hydro generation with an installed capacity of 3,421 MW controlled jointly by the GoHP (8\npercent) and GoI (92 percent), against an estimated potential of over 27,000 MW. Only 14 percent of India’s total\ngeneration is from hydro, of which 8 percent is from HP. <sup>3</sup> [^3: Data from Wikipedia sources.] Meeting the INDC commitments on reducing the carbon intensity\nof India’s energy generation will depend on further exploitation of hydro in states like HP. Improved watershed and\ncatchment area management upstream of hydro reservoirs can reduce operational costs from siltation and water turbidity\n. Similarly, HP enjoys around 1.126 million ha of forest cover, representing 25 percent of the land area in the state and 1.5\npercent of India’s total forest cover of 70.827 million ha. <sup>4</sup> [^4: 2017 State of Forest Report.] Although currently a small share of the overall area, HP has\nsubstantial potential to further expand forest cover (including through increased density) and therefore will play a major\nrole in contributing to the national commitments on carbon sequestration.\n\n[1 World Resources Institute. See https://www.wri.org/blog/2015/02/3-maps-explain-india-s-growing-water-risks](https://www.wri.org/blog/2015/02/3-maps-explain-india-s-growing-water-risks)\n2 <u>[http://documents.worldbank.org/curated/pt/375181504083376244/pdf/119178-BRI-\n\nMention: \"2017 State of Forest Report\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"source\", \"summary\": \"Report behind state forest cover area and share.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000113:3:3", "occurrence_id": "4d677f8558026fba4e177f71", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: -maps-explain-india-s-growing-water-risks)\n2 <u>[http://documents.worldbank.org/curated/pt/375181504083376244/pdf/119178-BRI-P157572-Himachal-Pradesh-Gender.pdf](http://documents.worldbank.org/curated/pt/375181504083376244/pdf/119178-BRI-P157572-Himachal-Pradesh-Gender.pdf)</u>\n3 Data from Wikipedia sources.\n4 2017 State of Forest Report.\n\n\nJun 04, 2019 Page 4 of 16\n\nMention: \"Data from Wikipedia sources\""}, {"role": "assistant", "content": "{\"role\": \"background\", \"usage_action\": \"source\", \"summary\": \"Open encyclopedia source acknowledged for background data.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000116:3:2", "occurrence_id": "73b41502062d3922568144ec", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  of the (pretransfers) poor received the transfers in 2015. Evidence (ILCS 2015) also shows that many poor households in\nArmenia do not apply to the flagship safety net program - the Family Benefit cash transfer program (FBP) either because they think they won’t be eligible based on the current scoring formula or they are not aware\n\n\nApr 12, 2018 Page 4 of 12\n\nMention: \"ILCS 2015\""}, {"role": "assistant", "content": "{\"role\": \"supporting\", \"usage_action\": \"analyze\", \"summary\": \"Household survey evidence on non-take-up of the safety net.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000116:7:0", "occurrence_id": "a0f69fc92d85021c31b14561", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text: **The World Bank**\nPromoting Social Inclusion and Self-Reliant Livelihood Activities in Armenia\n\n\nprograms available in each municipality/communities in the targeted regions; (iv) referral protocols with\nservice providers; (v) an out-reach strategy or plan for social workers to identify and reach out to the poor\nand vulnerable households living in their communities; and (vi) training material to train social workers on\nthe social case management methodology. The training curricula will include the provision of socioemotional skills. The component will also support the implementation of the training. Training activities will\nbe conducted to train about 200 state and community social workers in order to serve the poor and\nvulnerable in the targeted regions, empower and address their social needs, refer them to the appropriate\nsocial service and programs and link them with job opportunities and business start-up capital (including the\nsmall business grants funded under component 2 of this JSDF grant). The salaries of social workers are paid\nby the government and respective NGOs, not by the JSDF grant. Local and international experts will be hired\nto develop outputs (i) to (iv) described above and a local firm will be hired to conduct the training.\n\n\n15.   Through their front-line position, direct contact with vulnerable households and home visits, social\nworkers have the potential to raise awareness on people’s rights, inform and refer to existing social services\navailable in the community and facilitate access to those services to poor and vulnerable households that are\ncurrently out of the system (as they did not register). Based on national household survey data the vast\nmajority of households in the poorest quintile (72.5%) never applied to the FBP because they think they\nwon’t be eligible under the current scoring formula. While the Social Inclusion TA is supporting MLSA to\nimprove the design of the formula, the proposed project will support the improved implementation of the\ntargeting system through social case management. The targeting strategy to reach out\n\nMention: \"national household survey data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Survey estimate of non-application to the cash transfer programme.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000117:19:2", "occurrence_id": "001f61f184ad01cb40cfe3e0", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  plots. At the sub-state level, FSCMD coordinates with District/Division Forest Offices\nand PA offices for sample plot measurements. At the state level, FSCMD coordinates with the State Forest Directorate\nthe data compilation across each state. FSCMD compiles all data at the jurisdictional and national levels and analyses\ndata for forest resource assessment, carbon accounting, and emission reductions by using sample plot data, satellite\nimages, emission factors. FSCMD also implements the National Forest Monitoring and Information System (NFMIS).\nFSCMD reports to NRC through FRTC. The NRC has a Carbon Accounting, Monitoring and Reporting Coordination\nSection (CAMRCS) which compiles MRV data and information reported from FRTC and prepares periodic reports\nrequired of the NRC. Upon approval from NRCC, NRC then reports to the FCPF CF, UNFCCC and/or other relevant\ninternational agencies.\n\n\n49. **Local communities will also be involved in measuring and monitoring activities, in collecting forest-level**\n**information, and socio-environmental baseline data for the Safeguard Information System (SIS).** Forest-level data\n\n\n10\n\n\n**Official Use**\n\nMention: \"sample plot data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"analyze\", \"summary\": \"Plot data combined with satellite imagery for carbon accounting.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000117:19:2", "occurrence_id": "6e580bc30218f190ce71fffe", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  plots. At the sub-state level, FSCMD coordinates with District/Division Forest Offices\nand PA offices for sample plot measurements. At the state level, FSCMD coordinates with the State Forest Directorate\nthe data compilation across each state. FSCMD compiles all data at the jurisdictional and national levels and analyses\ndata for forest resource assessment, carbon accounting, and emission reductions by using sample plot data, satellite\nimages, emission factors. FSCMD also implements the National Forest Monitoring and Information System (NFMIS).\nFSCMD reports to NRC through FRTC. The NRC has a Carbon Accounting, Monitoring and Reporting Coordination\nSection (CAMRCS) which compiles MRV data and information reported from FRTC and prepares periodic reports\nrequired of the NRC. Upon approval from NRCC, NRC then reports to the FCPF CF, UNFCCC and/or other relevant\ninternational agencies.\n\n\n49. **Local communities will also be involved in measuring and monitoring activities, in collecting forest-level**\n**information, and socio-environmental baseline data for the Safeguard Information System (SIS).** Forest-level data\n\n\n10\n\n\n**Official Use**\n\nMention: \"MRV data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"Compiled measurement, reporting and verification data for periodic reports.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}
+{"origin": "umar_pads", "source_key": "umar_pads:000117:19:2", "occurrence_id": "2022bee1a74347e63e793beb", "messages": [{"role": "system", "content": "Describe how the author uses the cited data mention, using only the passage. Output ONLY JSON with keys role, usage_action, summary. role is primary, supporting, or background; usage_action is source, analyze, curate, validate, or inform. summary is one concise sentence in the passage's language; do not invent details."}, {"role": "user", "content": "Text:  plots. At the sub-state level, FSCMD coordinates with District/Division Forest Offices\nand PA offices for sample plot measurements. At the state level, FSCMD coordinates with the State Forest Directorate\nthe data compilation across each state. FSCMD compiles all data at the jurisdictional and national levels and analyses\ndata for forest resource assessment, carbon accounting, and emission reductions by using sample plot data, satellite\nimages, emission factors. FSCMD also implements the National Forest Monitoring and Information System (NFMIS).\nFSCMD reports to NRC through FRTC. The NRC has a Carbon Accounting, Monitoring and Reporting Coordination\nSection (CAMRCS) which compiles MRV data and information reported from FRTC and prepares periodic reports\nrequired of the NRC. Upon approval from NRCC, NRC then reports to the FCPF CF, UNFCCC and/or other relevant\ninternational agencies.\n\n\n49. **Local communities will also be involved in measuring and monitoring activities, in collecting forest-level**\n**information, and socio-environmental baseline data for the Safeguard Information System (SIS).** Forest-level data\n\n\n10\n\n\n**Official Use**\n\nMention: \"Forest-level data\""}, {"role": "assistant", "content": "{\"role\": \"primary\", \"usage_action\": \"curate\", \"summary\": \"Forest-level information collected for the safeguard information system.\"}"}], "task": "usage", "is_synthetic": false, "origin_model": ""}